ED. 324==646 'ANTEOR XO.0.014-pice-: ;$40104).*; LE4 - 04,'.44040i=-*Vprir-4,'.0h**4-14:0 TITLE Pa.r:44;414.-J.RtObei4114CWO*44.444-Aid,etin4,-*?-tfe- _ -Na0.0nai Reading Conferencer Inc., liE. ***31., ::4!46,4.46-441001r:00.`gi* AVAlLABLE FRON rtg '#,PE ec**-0(P Eps PRICE pltw-Ro#.44e. *.-::ik,7,0434:4)*,##,EP13Pi Alult:L.49#Pb,4t.-4 W.diA0kAi.Ositeri.0*t,040.it0 DESCRIPTORS #a1.1,401S0et:. tau:001,04 -440.i**c.p2ticiel#4.PRO,Oilicia$,E;446.4,64 *444:0,.'WeiliOerad#,.*004147-,00100446iiV iTeActi114:: I7,10#74.0**rilk4e4VT-0-cW0e4t:-.:Ra4#0: .13.4e*On't'',ika41410:*ii;ffik44414.*titiin4, PraCtide-RelationahiP Literacy EV'enis IDENTIFIERS ABSTRACT Reflecting diversified,Ariews, thiS..:NatiOnal,'Reading Conference yearbook Containa -40, .artialeS'iii_-tWifiqd, at llitracy theory and, research. ArticleS::and:f;heir: .,4.4t.*-4,.4.ric.iucte*:',Itie',.xsitil. of Teaching" (J. V. :Boffman).; "ReadingitTir#04:',CP#9,0**;11.1;e: = 'Relations among. Three:'erspectives":..(T: -Noto#etu raj.te.F4q;y"' CM. H. .mc0.434r0 T.,-;14.0-0.d.t:***Itee:4t#60 ,Persoii4 uni**4rid4.-pg:ok14*00#0-1.4:;*,:-:**0_, r,Joiiiing#e-D03#0:.:Iteseafchers.-arid, (13:1-ShannOn);, "The 1RC learbooks P.At,04a*'=(R AO,343.44.#:*14= gtitti,aillai.,*.pOrieith's_.4,47=.RWFiestAradepS in ot4ees);; Program's Designed-to ,prolliOte.SUCcessr Developing InSights, about the, Nse of :yhildren1:0,,p#4406, for A. Teacher of Lan4tia46 and Literacy qr.Otbr Literacy: Novice lihrile'LangUage Teacher* in-,!conizefitiOnal aq4e ,Effects Instructional Environments"IN., pactax*n.,4* of Structural Faotor,s::of 'EFpoSitOrt Texte*,',,TergOera, ,Judgments of Writpig NfaiM"^ (p.E,-00)04_,"&j.;0114#U0#4:41.417:;9f40,00yice -Tea,oher,St :xxiOw344e.,--q4311#ROS.r) :004,111##4.70.4 rsri.4,0*AtOo.,, RW4114 in. ah '?4,4.1rPt4q--40r,k-.,fPF -4744 c/-144.;*4".14111m.asuA atil:,othet4p '"Effect Of tarlY lteracy Interventidk .Oft-lindergarten l*F.044cins of hil.ipSand .0.40i70,4 441.4tezi±, ne*Warieti; t#414?,Reiie*mq..#-.14'4.11.n*tr..44*0-_ 14-044.dr* -P,00,r4,L _*1,11-11.'*,1,-01#r*Yr.-74,04:4P0:7441# Aorop..os,p. TeAch#:;,:*-6_ of'3goly,syllapic:Words,:Folgv- and AphiOenientir , . Siiithdrade :Disaabit-Readera . , e).; "The. Inxuence Phoi4c*Intuptuwok0341,44,-PF*9,0" ft,,,Oielq*A'"004.0irg 4.4401,1',Pq001.1441CP4#e#00 0.-:'#01#44ii'tP-00,444 . 1:40.*f#Fking, co********* --111#0-r_ 04.ty stu4*.e,;41-#4,tk- ,Co100.4,PnsiOn atiei.:(divri3o4tibn: 'Spe:ikeri Jr. and to itioeRt443.-) ; synth4s Of ,iipspara* OsUes" IR. B. "The Construction of ,iiaeroiileo by Good and =4" -.m.407/i_f5, - ,,--- W,bS - & 0.i.44r` --'iiiiiiT"---- --Oftle 7_141.**-4, it, -A44; fi*iii-0.§4i00 #14-,-, flt-' ..2.1-, tle if*#:*itbiiiii.A:'4T'j'g0.* '01- :,:* gil 11,1*-401:ir -Good Readers ndWrie otp f_L,,Oof,:: 4..or:f A r 92,:Orwi-Thoe4iii4i0:etc-.*:q.:'-***iXt #4100.44:0-4,g, fie.: 0 , :45i#4,. _,-, .. . .--!'00.1*-:-'='' r.. -64WIP..444-iti" *ittAtib- 44-044**0.:404#01e**1: 4,4INk. Sir*tiiiiii_r. -4' ,,-.44..fieprts#671 No-oppg, , 0.1,4,, 0 :+. :Pait#,-44.. p*ritto:?,(#14-41its*,44- 404#204. *014t$4#43,17.ilitai.0410 iiii*--4: -0444 -0,61.1' 40..3 -'-7X.9-e9" (!3*.. 1,P1._ .ki#4,4*.-,-,p:f '*4410.iik ,4#444,,,*riiO4& ifiO4 a:,., f: -,,,,,.... P li,-,:,,- sf.,-:f0..,:i , -#0..i*Y* 110Y ... * * . _ . . -. . Retsrpluotobs suppii . -,: ' 4451f.-6iii,-,iii tit:,. -1,ek 'Etia'e ,t ,'daiiiid mid-e f;o the origina3. -dosument-: ,',r- ,-f:, _, , '. 4 : s `': _= ,- -* "- * ,_ W.; 4.X.2 "PERMISSION TO REPRODUCE THIS MATERIAL iN MICROFICHE ONLY S BEEN GRANTED EY 1.3 S DEPARTMENT OF EDUCATION E.;,,cat eoai Posed"" a .c `- ^e,as ;,..C,c.0 as -anqes ao. s "oo.e s TO THE EDUCATIONAL RESOURCES INFORMATION CENTER (ERIC)." ! :',Overert ES.,CA'CNAL. ,,ESCL;PCES 7.4rOP4ATIC4 CENTER E C Of vew szatea r 'r,SCOCJ trent oo not onessan, reG,esent oqcal GERI pos.pon o DO"CY _ LIThRACY MORY ALySFS FROM MUL DRESEARTh: AitAbidiks Thirty-ninth Yearbook of The National Reading Confereuee JERRY ZUTELL SANDRA MCCORMICK Ohio State University Ohio State University With the editorial assistance of PATRICIA O'KEEPE MONA CONNOLLY Ohio State University NRC Headquarters Published by The National Reading Conference, Inc. 1990 " NRC YEARBOOK is published annually by the National Reading sConfere_nee,:Inc.,;:;;, 11 East Hubbard, Suite 200, Chicago, IL 60611, (312) 329,2512._ POSTMASTER: Send address changes to: NRC Yearbook, 11 East Hubbard, Suite. 200, Chicago, IL 60611. , -, SUBSCRIPTIONS: Institutions: $45.00 domestic or $55.00 foreign, lier year,.er, as, part of a combination subscription witn NRC's Journal of Reading Behiriipr (totit_,-1 price $80.00 domestic or $95.00 foreign). Available to professior4incliVid,41,;!'4*,.::::1 ily, or student members who attend the National-Conference as.part ef membership in the National Reading Conference, Inc. Members who de not attend the Confetenee ?.4 may purchase the Yearbook for $25.00. Also available for use in university/cellege/ courses. Write for information. Copyright, 1990, National Reading Conference, Inc. Microfiche copy is available, from ERIC Reproduction Service, 3900 Wheeler Avenue, Alexandria, VA 22304, The articles in the NRC Yearbook are indexed in Psychological Abstracti, Iliff410": Social Sciences & Humanities Proceedings, Annual Summary of InvestigationtRe,,, lated to Reading, and Educational Research Information Clearing PlOuse. BecauSe2 the NRC Yearbook serves as an open forum, readers should not construe the OubliShing of the contents as implying advocacy or endorsement by the National Reading.Conference, its officers, or its members. NRC is a not for profit, membership c9rPPIOon: All business correspondence, including changes of address (includelhe old addreas):: and applications for membership, should be sent to: NRC, 11 East Hubbard, Suite; 200, Chicago, IL 60611. Manuscripts accepted for publication in the NRC Yearbook undergo an impartial review conducted by the Eda:-wial Advisory Board and the Co-Editors. Manuscripts) must be original works which have been presented at the Annual Meeting of the National Reading Conference ard have not been published elsewhere or submitted, simultaneously to other publication outlets. 4 a - oFwgits,ANplg*frO7440:04§, QF THE- NATIONAL RIEAD'71110-CONS*404: ; : INC. GFAY-p. 6UFFIF,Iiresident MiChigun.State University ROBERT tIERNEY, President-Elect Ohio State UniverSity DONNA ALVERMANN, Vice President Univeisity of-Georgia TAFFY RAPHAEL, Treasurer Michigan State UniVersity MARY W. OLSON, Secretary University of North Carolina at Greensboro JAMES V. HOFFMAN, Past President University of Texas, Austin RICHARD ALLINGTON, Board Member SUNY at Albany JAMES FLOOD, Board Member San Diego State University JUDY NICHOLS MITCHELL, Boar4 Member University of Arizona DIANE SCHALLERT, Board Member University of Texa.S, Austin JERRY NILES, Publications Chair Virginia Tech JUDITH C. BURNISON, Executive Director , . "Peter- Emmy:University, N..1. York ThSI denesco Jaiç 'Iiiiiigti'afi,iikfOulgt1 at Hilo' Dpp' Pa,144.!6= of Itt#'17,11 s. 1,-'"! 'P.t0epu.,tY,9rMasaqIust-tts Schools )1VIllians Brow .,.,, . -Barberaielniow,, , , Eiii*lyri. cAiginiliiiversity 'at .140eict,--;"- .1)81,444.416.010 , *eati UniVersity ..:... MarquetidUnivemity Illariiiy,ennard University of Connecticut DianèDEod ,Ohlo-,StiteVniversity betki4Oilan' Pm,clUeljnivirsity University 'Janke:mai disSigia, Michael Maids UnikeiiitY of Utah . . ,,. Public Schticila, diaries ,.:,, hilielaii:Jeius Dreher , 17.10.'.i.r40 of Mrayland PaitrieisiEdwards ,...'.:4 IdiehiOthfStiitOMniversity Geortia,Saktrii:UniVeisity Ciediis Esinkii Yetk.q4veisity ,:. t.. 1,...,,,. thifi**.if,.0ArtY,NnFiSor! 4, .,'Ei.ariiiiup Ititernatienal, Reading Association ,I. SN' 'Et,i.slTi4ii Stitt University Jai* fli,tod A " RUileWunliersitY , San Diego'Stiae University .. ,,,,,,, Ohio-State University at Newark Die Otindirioii -,-; Piligi*t!**ity *.`1011,Harlier = 1.11.*Oi.6(*.Y1ctolia ...i.:;',:- Slasouri NPdek 16Atii*i.Pkithan ,... iusburgh '. lre4r.q,*0*56iti iPthn- s> C.'":f 2 '' Aiiitheasisii4fittis:siana University *iiiiiii,,ii',iitiik ,--",4'. --iiiii" ;,''...,,-- Stalo University . - -., :;'''",°.'"'". , Divid Uniiiersiti of Hritish Columbia }-, - Illinois it Evelyn D. Freeman f"::, 't":,:' 4,1 Kneet4iithers :koholatu r oificiio;:.-f:-tkt,*tcid at c.16.aFi4ke iiiii',StiiL?",.UniVeisity ..41Y1011iiid , ; , , 1 Schools, Piun.r-' 54: a Zinii lyarme., :MyId -4- Uixivrrsi "? ' lofi'Mraticaliddkl.3100 's ' UMyeTit , , A:queTRI Uvé WçflhYnL,veltyt , 'StatOtdveigty kiwe titiveisitY , Stiodri,StotskY!.-; 'Haiverd-Utuversity. , . Suianie:EAVade' .; NAv,ersity,- V:0-,otinftte0-s.at Austin Peter mversiyof Kentielcy U4iVCialftit'filOisiott, ;;Il=ryf. = Univeraity-o Cleveland Stile Uriveisity 41arj.0 University of Kentucky , , , .,,. . - 1,,, Z' Thil#10.0 tVES".0* irOffereact. " , - tiiiiersitk4f-toilth-Flarirlii :?..fe40i,s'spitevitiveriity jrJBru *;;;k catiii,646;, ,4:140.04.7,4a0.)34l?ersitY OPPck *V4ivelsitY AVOn:06inreIndiana UOVersityPuidue University at Fort Waynp', ,thOef,.410-006s*,ardoitinA* r7.7 Lindá-JoDeGoff UniVerSity "nrGiOrgia Wuona §1,4*-1.;!4i0"0.0- Mee0eiVitz kAti*Paigiigi4rt lo1i-6.Okeyer LynnC Smith 6f,toledo Colnifibia University Barbitrn puzietti SonthethJllUnjversi Aa State University Sinid0,4oBintalorth UniveriitiY otcalitornia at Berkeley MlcbádiCazñlj Ohio 'state University , lesnO#tgpicie, UniversirY of Nkiein , 0.10/f144 2 OW4*.k,CAUSErAWARDTIRNGSTON AWARD it4010ki4d4 AbfiR0s: Myth of teriiiiing. 1411-6011. HOfkian ANNUAL REVIEW OF RESEARCH' 1,?..eiiding-Writihe:coiwctioris;:The relations among.three perspectiyesnniothy-Shainallan Robert J. llerney INVITED ADDRESS Motivated literacy Mary M. McCaslin STVDENT RESEARCH AWARD The effect of reader stance on students' persona' wulerstanding of literaure Joyce E. Many LITERACY HISTORY Joining the debate: Researchers and reading education curriculum Patrick Shannon The NRC Yearbooks database R. Sc9tt Baldwin John g. Reaclince Jeanne Shay Schumtn John P. Konopak TEACHERFACTORS AND CLASSROOM PRACTICES The Curricular experiences of at-risk first graders in programs .designed tO,prainnte suecess elinaa tint* TeOeSe_ir jie*4kieinf114,f:*iOftkCi**':cigtdfiief-i_1*ahri!fix -142itgi_ic'ei,4AfIts#00iitrakic __ . , _ Becoming:010410T oflitOpey:Iflinice:whigeiangtaageleachers cpfive01401h4trisctOaid tkilianMen4' , . 'y , hnii!em.en04-#70PreitiOtr*ti9n: GOOng lirindPies lar:OFTICHIF17- 4evelOtknt }C.ink4e1.4ttY . Jennifer Goodman victoria L RiskoCharles K. Kinzer Nancy-N/5re Debbie'Rowe Janice Carson Effects of grouping and difficulty of materials on reading achieventent Michaetl- Kamil .4 ,' =,12t W. Christine Rauscher Reading instruction in selenee at the transitional grades: Beliefs versus -practice &Male Konokak HancliCiithern Wen Japiiiole WO, ?ARCM!! Ttita bean 1.4nnia--Holomon Mary Everett Nancy Weems Lealie Aroeneaux The effects of structural factors,of erpository teas on tearitee Judgments of writing quality Beverly E. Cox A longitudinalitudy of priservke teachers' knowledge smictures Beth Ann Herrmann A comparative study of the teaching If-olive/ten of itifluential and nonieuendal teacheri and readirig-coMprehension,deVelopment Roi*it B. Ruddell: :Marilyn,E.'briheirn ..ltidith Barnes Best ideas: Sources and liftmen* Sara zAnn Beach 45' -153- - 8:01deitikto kapt-frOmte*PrOteryke conteittlitichersi view 0:theirrole Armes the*indos0 of their student-prOfe:ssor 0Jague Jaurnals Thomas Jan Zulich 'EAkLY,LETERACY.INTERVENTIONS1 . Thi iRfluence of literacy-erirkhed play settings on preschoolere-engagement 1 with written language Susan B. Neuman KathyRoskos Shared book reading in an Early Start program for at-risk children Jana K.'Mason '179 189 Bonnie M.., Keir ouistine McCormick Effect of early literacy Adervention on kindergarten achievement Linda Al Phillips 199. Stephen 14:11orris Jana M. Mason Bonnie M. Kerr Parents' perceptions of children's reading and writing developmen: in a whole-languag kindergarten program Beverly J. Bruncau Timothy V. Rasinski Richard P. Ambrose WORD KNOWLEDGE Word Length and first word recognition Luci Lipscomb Philip B. Gough A syllabic-unit approach to teaching decoding of polysyllabic words to fourthand sixth-grade disabled readers John Snelfelbine Incidental learrdng of word meanings by kindergarten and first-grade children through repeated read aloud events Cynthia B. 'Leung John J. Pikulski The influence of phonics instruction on spelling progress Laurie Nelson ii - 223 231 241 04/tf:41rWKY-OtOit Y.0:10_1pw***(44 4 *ie.( ooti Yof ii.p. ... siWeijii. '.:170.* : Woloshyh ' ,.. . . Differeneei ,thelling..behaviciti:andlheir relation.to .,.. iii`siorY , ., , , , n reading . . ... ,.._. . -''0#0.011eliS iP44#40.60,:kkaOS tune-iit4iiiiiii#j"-: AlieSSing:ChikifenWiferOiOng-sitategies, §i1011344074iiii* Developing lawrpe;Yorming, fourth-grade, inner-city students' a:bitiiy_ to comprehenCrriOatke JaniesH. Moknthai Lexical cohesion in cbmprehension and composition: A synthesis of researCh issues Richard B. Speaker, Jr \John U. Barnitz Joan P. Gipe The construction of narratives by good and poor readers Rose-Marie Weber READING AND STUDYING EXPOSITORY TEXT Students' level of commitment to their naive conceptions and their conceptual change learning from texts Janice A. Dole Dale S. Niederhauser 3 Interactive teaching and learning: Facilitating learning disabled students' transition from novice to expert Margaret A. Gallego Grace Z. Duran David J. Scanlon The effect of PLAE upon students' test pvformance and metacognitive awareness Sherrie L. Nist Michele L. Simpson Correcting misconceptions: Effect of type of text Katherina Maria Joanne Mons Johnson 320: Transfer effects of prior knowledge and use of graphic organizers on college developmental readers' summarization and comprehension of expository 'oxt 3 Ernest Balajthy Renee Weisberg TO elo. iff-10K4 Rtirse aCkeYelsent? :nvi.,sti8gtion ifv.040001.0 e. ,--,t_ -mnqnoruc Ididrei4ability;to,:utt ._ _. , - ,. ntethod:*editcdtional" - keyword _ .....,.,_ ice#PA.Wilhh!fekrAtgifegle.elasSfOoir,s 2 3W 140.0,10Y0ci.fni' 1W-0*.dit.'gtOtS;ANP titgliA& ,scriPti* analySis of-good:reader.? and Writers' cOncePtS ofauthtuaV irciAis one, three'aiid fOe *ob,01-kldist1er _bdstruclingroriverieition:'Peer responses to student writing Safah .1:"MiCarthey RACY MATERIALS &zry literacy strategies: Activities represented in current basal readers LesleY Mandel Morrow Rachel Parse Types of writings included in basal reading programs, kindergarten through second grade: An investigation of changes from 1983 to 1989 James Flood Diane Lapp 381 395 ADULT LITERACY Reading strategies of marginally literate workers: A case study Verna Haskins Denny 403 SECOND LANGUAGE READING PROCESSES What 12 readers remember: Is it related to their awareness of text structure? Sally A. Hague Steve Olejnik 409 Factors effecting second language text comprehension Keiko Koda 419 -NRC PROGRAM, AUSTIN, TEXAS, 1989 429 PREFACE As was true last year, we are highly pleased with the content of the 39th NRC When we put the final touches on the Table of Contents and werd able to see as a whole the scope of this year's articles,_ several things _were evident. First, a glance down through the titles shows such diversity that it appeari there should be something here for almost everyone. Second, an examination of the anthori of these articles discloses an excellent mix of work by the most respected individuals in our field, along with perspectives from many promising new contributors. Third, the titles, as well as the content of the articles, reveal analyses from varied philosophical and methodological paradigmsas reflected in the title we have chosen for the 1990 Yearbook. edition of the Yearbook. These diversified views are a strength of the National Reading Conference, and we oelieve, of this publication. The Yearbook is truly the work of the National Reading Conference membership. This year 77 members served on our editorial advisory review board and an additional 27 served as guest reviewers. Reviewers maintained high standards in recommending manuscripts for acceptance and provided concrete, thorough suggestions for revision. Of the 104 general papers reviewd, 41 are published in this 39th Yearbook; there also are 3 special papersthe NRC Annual Review of Research, one invited address, and the student award paper. These articles, along with the Presidential Address, represent the work of 99 NRC members. The special papers except for the Presidential Address, as. with the general papers, all received outside, blind reviews. The overall acceptance rate for this NRC Yearbook was 41%. We thank the authors for their high quality research, reviews, and position papers, feeling certain that these add important knowledge to our literature base in the literacy field. Our deep appreciation also goes to the reviewers for their conscientious attenfion to their tasks. We wish, also, to give special recognition to Mona Connolly, our graduate student and editorial assistant for the Yearbook, for her intelligent and careful attention to all manner of details and to Pat O'Keefe, our liaison with NRC Headquarters, for her patience and knowledge. Our job has been made easier by the contribu- tions of so many competent professionals. Jerry Zutell Sandra McCormick 14 1, I Csc:- es, 4. . Irs a Crs r - . 4: , 't7 `" le ' .4, r. 4 , :047 A 4 ; .6 A I OSCAR S. CAUSEY A WARD The Oscar S. Causey Award is given each y..ar for outstanding contributions to reading research. Dr. Causey was the foander of the National Reading Conference and served many years as Chair of its Executive Committee. This year's recipient o: he Oscar S. Causey Award, P. David Pearson, is enormously well known across our field. First, his interests cross many people's interests; he has an abiding interest in research, anenthuiug interest in practice, and fur unfaltering belief in the benefits of research to practice and practice to research. Second, he writes well and wntes a lot; he speaks well and speaks a lot. Third, and by f4r xst important, what he has to say is so thougntfuly considered. I believe it is his o..pth of knowledge and his intellectual honesty that satisfies. Several yeais ago, David and I were both scheduled to present in 4 "what rcseara has to say to instruction" session at a regional 1RA meeting. We v, -e to ynthesize research findings about comprehension and vocabulary and to point to those finCns that had a strong enough research base to allow us to suggest with confieente that they Ix implemented in classrooms. David synthesized th e. findings intelligently Emd interestingly. I particulrfly noticed haw careful he was about staying with what the research said, and not offeri^g things he belizved were probably true as being true. He listed five findings, and then a sixth that he believed would turn out to be true, but stated clearly that it could not be supported with research. From that time on I have seen David, time and time again, clearly distinguish between what rarch suggests and his own opinions Depth and breadta of knowledge and honesty are hallmarks of David's work This is important because David is so influential among practitioners. Of course, it is not only teachers whom David informs. Certainly his students, other scholars, and researchers have benefited from his work. I focus on teachers becitu._ many are not as aware of the research as researchers are, and therefore can be more susceptible to accepting the opinions of authorities as grounded in research. An incident that happened to me speaks directly to the enormous influence David has with teachers. I was being introduced for a talk I was to give at the local IRA in a southern state by a teacher who did not know of my work st hand. The individual had called me for a vitae, and from it had written a formal and rather lengthy account of a variety of my endeavors. Finally, with a flourish, she waved her notes aside, looked directly at the audience and said, "Look, Dr. Beck must be pretty good, because even P. David cites her." I believe David Pearson has made enormously useful contributions to our field I think we -..re lucky to have him. And so, it is truly a privilege to present the 1989 Oscar S. Causey Award to P. David Pearson. Presented by Isabel L. Beck December 1989 1e 01111, S * .. _ zk 4 0 t 1 ALBERT J. KINGSTON AWARD The annual Albert J. Kingston Award for Service to the National Reading Conference was established in 1985 to !tenor Professor Albert Kingston foryears ofeledicated service to NRC. Professor Kingston, an educational psychologist and reading specialist, was President of NRC in 1965-66. The 1989 recipient of the prestigious Albert J. Kingston Award is Michael L. Kamil. Michael Kamil received his Ph.D. from the University of Wisconsin ill 1969. Currently he is Associate Professor of Education at The Ohio State University. During the past 20 years, he has authored, co-authored, or edited over 40 books, book chapters, and journal articles for the literacy field; has received research grants from the National Science Foundation, the National Conference on Research in English, the USOE Right to Read Program, and other organizations; and has presented papers at over 60 national and international meetings of scholars. In 1983 he received the Milton Jacobson Readability Research Award. One of the great benefit; of receiving the Albert J. Kingston Award is that you have the privilege of helping to choose the recipient for the next year. Everyone who receives the award in an organization this large must have the same humbling feelinb---"Why me, when there are so many who are equally deserving?" NRC is, after all, run by the membership. Its success and its vitei:iy depend on a membership willing to serve its purpose. The big bonus of the Albert Kingston Service Award is that you get the chance to help correct any possible error of the previous year by picking your own candidate for the coming year. I cannot say, of course, how much discussion has gone into previous choices, but this year the choice was so clear there was almost no discussion. Everyone involved in the decision agreed that this year we have to give the Albert Kingston award to Michael Kamil! I'm sure that a vote of the entire membership would confirm our choice. When I tried to list all Michael has done for NRC, it seemed easier to try to think of something he has not done. That proved even more impossible He has been truly a Man for All Seasons to us. Michael Kamil has served on the Editorial Advisory Review Board of the Yearbook; served as editor or coeditor of three NRC Yearbooks; served on the Publications Committee , of which he is currently the chairperson; served as a member of the Board of Directors; chaired the Subcommittee on Standards and Ethics; served on the Editorial Advisory Board of the Journal of Reading Behavior; served as editor of the Journal of Reading Behavior; served as a reviewer of the National Reading Conference program proposals; and been a frequent contributor to the conference program and to the NRC Yearbook. I believe I am safe in saying that Michael is loved by everyone in this organization who knows him, not just for all that he gives to us but for the way that he makes us feel when we are around him. We love you, Michael, and even if it doesn't say so on the plaque you are receiving, we wanted you to know it and to consider the Kingston Award as our way of saying "Thank you." 18 Presented by Thomas H. Estes December 1989 THE MYTH OF TEACHING' James V. Hoffman University of Texas, Austin The story is told about a femme astronomer giving a lecture to a crowded hall, of students and parents over such basic concepts as the rotation of the earththi' orbiting of the planets, the place of the solar system in our galaxy, and so on: At'the end of the talk, an elderly woman in the back of the room stood and shotirecl: of what you say is nibbish. Everyone knows that the earth is flat and is behig,heicl up on the back of a giant turtle." The scientist paused briefly ancLthânt:respiiiided, "Well, if that is the case madam, then what is the giant turtle standing'bn?"-The woman shot back, "Very cute, but sorry, sonny, it's turtles all the way down!" As in this anecdote, as in the world at large, myt), and science are typically portrayed as antethetical to one another. Science is truth; myth is falsehood. Science is serious; myth is humorous. Such a portrayal is misleading. Science is the way in which the verifiable knowledge of the world is represented by mankind. 'Scientific knowledge is advanced through the kind of systematic inquiry that we refer to as research. Myth, in contrast, is the way in which mankind has attempted to explain and understand that which is not readily verified. Although scientific knowledge has advanced enormously over the millenia, it has barely begun to address, let alone answer, the questions that give birth to myth. Many find the myths of ancient cultures as trivial, or as revealing of ignorance, but this is because myths are locked, in terms of expression, in the science of that day, not because they reveal anything less than fundamental human wondering. Many think of myth as only something of the past, not of the present. Not so. Myth is as alive and important today as it has ever been. As humans we continue to struggle to understand our cosmos, our origins, our purpose, and our transcendency. Within each of us lives a personal mythology that is constructed in a fashion compatible with our scientific knowledge. Our personal myths govern our interautions with the world; our shared myths govern our sozial institutions from family, to state, to church, and all of these find expression as we communicate with one another through language and the arts. Within this mythology, we find the assumptions, values, and beliefs that are the powerful, driving forces in everyday living. The roots of teaching are to be found more in mythology than in science. The evolution of teaching from labor to a professional status has come as a result of shifts in societal values, changing perceptions of schooling, and an expanded economic IP esidentia; Address, The National Reading Conference, Wednesday, November 29, 1989 Literaq '11165rY and Rilealic_ reality, not as a result of scientific breakthroughs. We sometimes forget tharthe: historical antecedents for our profession are not particularly Inspiiing,(powell;198O)`1 . It was not antil the 1950s that teaching achieved,,in sonri.'sniall'sense;,a:prOki* sional status. A positive mythology about teaching- WaSrez-l'ergirig: Teac,hers`,-*Cte; bright, knowledgeable, and conunitted individuals (mostly wonien) :deseryingi3Of,,t spect and appreciation. Teachers had the power to mach any'Sthcleht,:,94::,itl*? profound difference in that studenes life. Schools Were.iinOrtant;1fisfitueiOni:W significantlearning took place. The teacher Was at the center ofthe"educatiOnal et.tterprise, not the principal, not the district, and certainly not the itate:Ilielpay\i/aa',Iowi,', the hours long, and the conditions not always ideal, but thii wis,all part Of themythi that inspired those both inside and outsHe of the profession. It was a Normai2itc444painting. It was, perhaps, too good to be true. Out of the Cold War and Sputnik, a crisi in confidence in America exploded. Schools were attacked for bt"..ing ineffective and; unresponsive. Commission after commission was formed to study and.makeiecornmendations on correcting the ills of schools. The fundamental question being:Laskeil was: How do we judge the effectiveness of what is going on in schools? Ili effect;. the myth that had been building for decades was being challenged. To)thoSe outside of the profession it became obvious that there was no scientific knowledge'base for teaching and schooling to draw on in defense of teaching practices. There Avere. assumptions, beliefs, and valaes informed by personal experience and supplemented: by craft knowledge that gave shape to a mythology, but no scientific knowledge base: informed through research. Our efforts to develop a science of teaching emerged out of this crisis. One has only to look at the editions of the Handbook of Research in Teaching, published over, the past two decades, to see the roots of and the fruits of this work (Travers, 1973; Wittrock, 1986) Within the field of literacy education, wc can point with considerable pride in the emergence of research communities like the National Reading Conference and scholarly publications like the Journal of Reading Behavior, the Yearbook of the National Reading Conference, and the Reading Research Quarterly. I am concerned, though, that in our efforts to construct a science ol teaching (of teaching literacy in particular) we have ignored the power of myth in teaching. As.a research community interested in language and social institutions, I fear we have not given over sufficient attention to the important relationship between science and mythology I become more convinced daily as I work with teachers in preservice and inservice settings that research is intruding on rather than informing to good teaching: My thesis is a simple one: Thom is far more harm being done in classrooms today in the ^ame of research than thei is good. The obvious victims are the chilthen. Less obvious victims, but no less itu,ncrtant, are the teachers and the researchers themselves The teaching community and the research community are becoming more and -1 more alienated, at just the time when we need each other most. Not one of us can take pride in this situation, nor can ive simply sit on the sidelines and take no action. These are the areas I would like to explore with you through this essay. It is toward a positive intersection of myth and science in teaching that I will direct my thoughts Further, I will reflect on our responsibilities as a research community to address this problem. I logize, in advance, for taking a somewhat parochial view :1 . 20 The Myth of Teaching on this topic. I will base my argument around areas of activity that I have observed and studied first hand. Later, I will attempt to broaden this treatment to include a more national perspective. I begin with a consideration of the field of research in teaching, a field that emerged in the 1960s and gained a national identity during the 1970s. Research in teaching was dominated through the decade of the 1970s by the process-pthduct paradigm (see Dunkin & Biddle, 1974). Working within this paradigm, researehers wer ._ to document not only that teaching can make a difference in,students' learrung, but also that certain teaching processes are associated with positivelearning outcomes. The correlates of effective teaching were chronicled in unending list after unending list. By the beginning of the 1980s, it became clear in the research in teaching community that studies in this process-product tradition had taken us just about as far as they could toward understanding effe Ave teaching. Critics pointed out this research did little to advance a theory of teaching and learning, was unidimensional in its focus on teachers' behavior affecting student behaviors, failed to capture the multidimensionality of classroom life, and suffered from its tight focus on the teaching ari learning of so called "basic-skills." As Lei\ Sherman (1986) has noted, it was a paradigm that satisfied the questions that gave birth to it in the first place. Having done so, it died. Although the research community may have abandoned the paradigm, administrators and policy makers have not let it rest. With the increasing pressure for reform in education in general and for teacher accountability in particular, teat.her evaluation systems drawing on the findings from the process-product literature began to appear across the country (Defino & Hoffman, 1984). These instruments took the so-called correlates of effective teac ..tg and transformed them into checklists organized around a dualistic conception of a teacher as manager and instructor. As a manager, the teacher is responsible for establishing a classroom work system that maximizes academic time and engaged time As an in- structor, the teacher is responsible for engaging the students in direct instruction focused on specific, measurable learning outcomes. Through the use of these instruments, hopes were raised amoni, educationally minded reformers for weeding out the incompetent from the profession, for educating the mediocre toward more effective practices, and for rewarding those who are truly outstanding (Hoffman & Defino, 1985). There was some resistance among experienced, professional educators to the use of these instruments to evaluate teaching. The initial breakthrough t.ame when these systems were introduced into the evaluation of beginning teachersa particularly vulnerable group in a political sense The beginning teacher is neither fish nor fowl and, given the notion that gatekeeping was viewed as an important step toward reform, this is where the systems were hrst ptAt into use Their use was established in this one particular setting and gradually expanded to become a part of the evaluation for all teachers and used as the basis for merit considerations, career advancement, and even dismissal. The fact is, these instruments have failed miserably in realizing their intended outcomes. Although many states across the country, including Texas, mandate the use of evaluative instruments th-wing on the process-product research base, there are few examples of any teachers ever failing to pass them (Hoffman, Edwards, O'Neal, 4 Literacy Theory and Reicair,h: Barnes, & Paulisson, 1986). Indeed, the ojpical pattern is for teachers to "top out". on such assessments. To assert that these instruments have not achieved their intended purpose is net to say that they have not had an effect on teaching. Clearly, they have. To characterivr this effect, I would like to diverge for a moment to describe very briefly:and ill-*eryi simple terms some notions regarding teaching, and learning that draWcheavily. On the, work of Walter Doyle, Thomas Green, and somn of my own studiesofichissrocim.., practices. Doyle (1983) has proposed that perhaps a Imre powerful way to examine teacbing and learning in classmoms than the process-pmduct tradition is to be achieved!by, focusing not on what the teacher is doing or saying, but rather on what the strident's: . are doing and saying and, by inference, thinking. He describes the "academic Work" that students are engaged in in terms of a 'task model. From the student's perspectivei, the classroom is seen as a place where work is to be completed and products generated for some kind of evaluation by the teacher. The kinds of products generated J114, range in scope from a simple worksheet on the letter B completed in the, first.gtade to a complex literary response assignment in a senior level honors English course. We can think of these tasks in terms of many dimensions. Two of the most important are risk and ambiguity. Risk refers to the likelihood that a task can be accomplished successfully by students. The basic measure of risk is how well the students might dr, on a particular task if they were given no instruction at all. Completing a page of problems in mathematics that involves some previously learned algorithm would be a low-risk task since the students could likely succeed on thcir own. Ambiguity refers to the clarity of the task in terms of the product to be generated. Completion of the problems on a mathematics prace.ce page is low in ambiguity characteristics. It is quite easy for the teacher to express, and for the students to understand, just what is expected. In contrast, consider the example of a Junior-levet teacher trying to teach students how to compose a persuasive essay. Here, there is greater inherent ambiguity in the task because ?lie teacher may have some difficulty in describing precisely what constitutes a high quality persuasive essay. In tura, it may be difficult for the students to understand clearly what the teacher's expectations are. Ambiguity and risk, as task characteristics, can operate independently of one another Onc can, for example, increase the risk characteristics of a task without affecting ambiguity A teacher might En interested in developing a student's automaticity in decoding through repeated practice with a story. The teacher requires that a particular section of the text be read at a minimum rate before allowing the student to move one The teacher has increased the risk in terms of the likelihood of success but is still dealing with an unambiguous task. Although the dimensions of risk and ambiguity may have some independence from one another, they are not independent of the kinds of learning one might be interested in fostering. To illustrate this point, consider a model with risk identified on one axis and ambiguity on the other as a heuristic for considering task to learning relationships (see Figure 1). The model is divided into four quadrants and there is a. diagonal line radiating out from the origin at a 45 degree angle with a positive slope. The diagonal line represents a continuum of learning distinguished by the amount ill .."COGNITIV,B , RISK AMBIGUITY Figure 1. A heuristic model for considering teaching-learning-task relationships. ,of.cognitive activity requfred on-the part of thostudent. Tbe.continuMn rangtia firun the 'simple, stimulus-response type-learning to the , level, of ciitic4thiirkinikri-ein."' volVes, the, application of both et:incepts rind-Values..eVerattYpes of 'leaChing7;,arealso represented on the diagonal Mat are aspciatect with the Various leyelS cif:coltniti*e, :activ4 tonditioning is a -type of teaching one associates, with- Sthilitlus-`,reitO*' learning. Training is a type of teaching associated -with, the chaining-,,Ofstiinulu.4, response type learning into complex performance algoriflunS.,InitruCtiOn is ,ru-tyPe7Of teaching -associated with concept development, and ifidoctrination,a type of teaching' associated With the development of criticalleascin-ing-(Green;,-`071): Now -consider the four quadrants lathe incidet In Cluadrazt I, (low4iski rlo* ambiguity); you find memory level typo tasks that .rcquire ooly.kote;leietlitiini0 such as a student being required to learn some toUnd tosyinbol correipondenee Thonics exercise. In quadrant II, (high risk, .low runbiguity), you hod xoutinelevet type tasks such as the one described earlier C.esigneit to foster greater 00;114164,c In .deCoding skills. In quadrantIll,,(high risk and highambiguity)iyou-find highéi *et :cognitii,e, level finks requiring the development artd,airlicah'on of neW.concepts'auCh as the task of composing a persuasive 'essay.,,And. ttrirly,, in quadrant ,kpOW,iiii and-high- ambignity), you find tasks that foster 'appreciation, and-the develotiment of Values.'liere, for example, you might consider the- task' Of a teachei readiai iii§tIdtto -students from good literature while encouraging divergent responsee *int the stit4eats2, Applying this, framework to the- study Of teaching-and. learrung:An:classroams Might iriVolve examining the naturoof the.tas4presealted tostud-entkin *MS of therisk and- aifibiguity. characteristics. What kinds Of:04.5'06 the teacher.Sslecq,tioW- z 6 Literacy Theory and itesearch.-: are 'these tasksintroduced to students? What is the teacher's role in supporting task completion? What kinds of "in flight" decisions does the teacher make to adjust the task during instruction? What meanings do the students construct around this task? What kinds of learning are associated with participation in these tasks? One of the interesting findings from task research relates to the dynamic inteinction between teachers and students during instruction: It has been discovere6hit4 students, in general, tend to resist tasks that are .high in ambiguity,and risk. the introduction of suth taslcs into a classroom immediately sets into motion i pia** negotiation between the teacher and the students. 'Students will, work: to mince,* ambiguity and risk features. For example, the teachei introduces the.iask. of wrifing . a persuasive essay. She tells the students that in 2 weeks they will belequireci to film in a completed product. That essay will be graded and count for 40y% of theiptarrn: grade. She assures the students that she will teach them during inis 2-week period how to write a good essay. It is here that the negotiations begin. For example, the students might ask for spe9c parameters for the task. How many pages? They might' , ask the teacher to provide them with a model of an excellent paper. They might ask for a chance to turn in a first draft for feedback that they can revise before: the final submission. In each case, the students are attempting to reduce the arnbiguit/ of the task. The teacher may resist these attempts to negotiate for fear of reducing the. potential for learning The teacher understands that to hand out models might lead to mimicking which is a low level memory type learning outcome. It is truly a negotiated process. The teacher is not simply free to hold the standards as high as she might like without some encountering some risks. The students always have their trump card to play in the negotiations, and that is cooperation. The teaching- learning contract is based on principles of trust and cooperation. Pushed too far, students may become uncooperative and teaching stops and learning stops. The work system disintegrates. If the teacher has the goal of a smooth running class in the sense of minimum disruptions and "noise" in the system, then several options are open to him. The first is to readily negotiate with students by reducing e ambiguity and risk characteristics of tasks to a rote memory level. The other option is to simply avoid the teaching of difficult content altogether Here we find the teacher who, for example, skips thc. unit -.a electricity because it is too complicated for his students and substitutes instead a unit on the water cycle. Or, the teacher who moves a mil from one reader down to another because the vocabulary and concept load are too challenging. The choices are fairly clear. On the one hand, the teacher who wishes to challenge students intellectually, to push the frontiers of learning, must be willing to tolerate some uncertainty in management and order in the classroom. Higher level learning requires action and interaction. Some students may resist initially. The teacher must be skilled in motivating and instructing and not succumb to the pressure to abandon higher level thinking goals On the other hand, the teacher who is primarily concerned with the conditions of order and cooperation in the classroom may find the easy road is filling the classroom day with tasks low in ambiguity and risk, thus reducing the opportunity to learnthe expectation to learn. The first example is consistent with a professional myth of teachingthe belie( that a goad teacher is one who holds incredibly high learning standards and is knowl- 24 Iheuvth qt-Tebehing, nle,.skilled,,flexible, and creative in helping all-children, bending theurunotk !..rated and "at-risk" learners to realize these roali. The-second cx.miple of the clais400rn filled with trivial content probably comes closest to.what the science:of teaching (as.teVeged through prOceia-prodnat research) har identiticiliS."affectlYe." Teacheri subjeCted to eValuan.,In derived from the Proeeispiuddet traditions learn One simple lesSon froth- the raPpraiSit pro:earn-The easiest: way 1.0 score Well on An(1=i, -appraisal is in teach only content thatis at therrioripleveL Wittv the asset:1*U -ww resk,and-low tonbiguity taskcharactenstics, good management niprietictilly, assured. ..EUrther, this _content, iS both easily niolded into seven' Step; fotinultii. teaching 'training) and readily meattired in terms of learning outcomes; .Altheugh seine teachers claim to haVe 'canned" lessons tO pull outit node° for an unscheduled observation and.thatthey return to mid teacidtig after thC, 'observation is over, the fact is these kinds of appraisal insturnents'brive,created norm in many states for what effective teaching looks like. What ldnd of mythology of teathing is compatible with this scientific viewl It it, surely one that minimizes teaching to a technical-skillnot even-a craftand eer, tainly is not a professional view that emphasizes responsible decition-inalting. The research in teaching and accountability moiement is Pt one exaMple of . where research findings are being used in a way that intrudes on the lives of teachers, and students. Consider a second area. prescriptions for practice deriv,Pyi froin -the "effective sehools" movement. Paralleling, but disdnct from, the development of the research in teaching literature during the decade of the seventies, one finds the growth of the "school effectiveness" literature. Beginning with the work of Weber (1971) and followed along by many others, researchers began to identify schools that were succeeding with students in terms of academic achievement where other factors (e.g., socioeconomic statiis) would overwhelmingly predict failure. Out of this wort, the correlates of effective schools have been identified. Like the effective teaching correlates, there are any number of lists of school level factors that are associated with success. There is wide agreement in this research community on the importance of such factors as clear mission, instructional focus (with an emphasis on basics), the principal as instructional leader, and frequent assessment of student learning (Hoffman & Rutherford, 1984) Them are any number of staff de. ..'Ipment programs under way in school districts and in states across the country that draw on this research base. Let me examine just a couple of these areas pith you to explor _le effects on teaching. We begin with the notion of "instructional focus." This is translated operationally to mean that all teachers should be teaching the same thing at the same time. At the campus level we find principals, n their role as instructional leaden ,i..quiring teachus to move through the same required curriculum materials at exactly the same pace Ai the state level we find similar efforts. In Texas, for example, the state has attemprd tc a.,hleve instructional focus by identifying a set of essential 0e znts. These essential e-....4nents serve to define, at a minimum, what must be taught by every teacher to every child in every grade level in every major cunicular area. The state has implenr ,ed an annual minimum skills testing program that is tied directly to the learning areas targeted in the essential elements. All of the state's requests for instructional materials are tied to the essential elements and the associated assessment instruments. The 44 S titeriey ThediY:additdetard, publishers have responded in the designing of their materials, not just for Texas,:, for the, world at- large. And what are the effects of promoting this conception of insfructional"(007*, have created a trivial curriculum around easily measined learnirig,OtcOMes.alid'Opee: ' again intruded on the teacher's prerogatives and resporiffiititits related1O initriiCtide- decision-inakingthe hallthark of the:professimialteadia'4,Myth, I have a _friend -who is a 'classroom teacher kindeit4ten level) who,,haSz. immersed fotthe past 2 years-in one Veition of effeetiVe ichOpis trainlng.:st*o to me a recent inservice activity in which all of the'teacheis'iri hei:_seiloal were, in small groups to consider and discuss solutions tr partieular all of the problems related to instructional focus. Om of the ceses:slepiCted*._40. grade teacher working in a school located in a very poor commuhitY. Each.yeat this : teacher would spend 3 weeks in the spring teaching a poetry unit that slie hadeàoped. It was her belief that all children needed to be exposed at an early age to-thdbeautiful language and expression one finds in poetry. She felt it particularly nedeSSail to teach this unit in this setting because the children in her class were riot Oftmi exposed to rich models of language. One day the principal visited her classroom- for an observation and found her teaching this unit. In a follow-up meeting with-the, teacher that afternoon, he suggested that the teacher's time and the students' timewould be better spent on the basics and that she should abandon her work with thd . unit. The question for discussion in the group was: What should the teacher do? My friend did an excellent job of convincing her small group th.: the teacher should stick, to the unit and instruct the principal on how the "basics" (and much more) can be; taught through poetry. When the groups came back together to share their soludons, my friend's group was the only one recommending that the teacher stick to the unit. The other groups all concluded that the teacher should drop the un,. and teach the basics in order to achieve an instructional focus consistent with that of the entire school. The principal, who was directing the inservice and following the programmed materials for establishing an effective school, affirmed the position of dropping the unit She cut short the discussion of the merits of the alternatives with the statement: "It's not important whether you believe the effective schools' principles. All that's important is that you do it." What kind of teaching myth can survive the fury and folly of instructional focus when that is translated to mean sameness? Woe to those who venture out of quadrant I on the modti: I := the lines surrounding quadrant I are quickly becoming the boundaries for schooling. Although the examples of research intruding on the lives of teachers I have discussed are few in number, their presence is so overpowering that I am amazedwhen I walk into classrooms and find exciting, creative teaching going on. It is a. testimony to the commitment of classroom teachers in Texas that the myth that led them to teach survives the onslaught of educational "reform." Although the examples I have reflected on are focused on the geographical area. that I have continual interaction with, I know the rest of the country is not immune. In preparing for this essay, I wrote to several colleagues across the country soliciting- 26 _ , ------ - _ _ `.7 lhi'Milh:of-Tcaching ; :ekatirples -from their experience where_.they found the findings of research being op- lolied.in an intrusiye way. The return :ate on mr"not-to-snientific" qUestinunaire kis over-9S% and there wat no shortage of exaniples.' The lapels often differed ftnm . one-area'to another, but their-responses suggeited to me that-the Underlying Mentality ---..04..inoVementscare the same across the country:2 -Although:the exiniples I -have given,relate to areas that I have sorneli:searciiexperience in, there ate other areas of activity that are,equalliithportaat`tha4:004:: *ow on-such as the recent legislation in thig sue that reiiiiries altftrSt-giacte,i4dellita'4 tio be assessed for dyslexia and-those found to have it to be treated4ith a .11,PrOYen'; -*gram of remediation. . v . - And-finally, although the examples I have reflected on represent, 'in inyeatinia:: lion, misapplications of research findings by policy makers, there are numerous other examples of intrusions into the lives of teachers in the name of research that- havea questionable research base, as in the case of the learning styles movement and-the left-brain/right-brain literature. What responsibilities fall on us as literacy researchers deal with these abuses? We can ignore the situation altogetherdig our heads in the sand, or worse yet, build walls around universities only venturing out into the real world to gather some data now and again. We can rinclaim science as innocent, value free. We can dal.: that some of our best friends are teachers, forgetting that for every teacher we-know there are thousands more who know us and our work only through our interpreters. The problem with the "I'm innocent, I'm a scientist" approach is that the suffering is too severe to be ignored. The abuses are toc rooted in the system to go away if we ignore them. The problem is that we are part of the problem and, therefore, must be part of the solution. We can begin by becoming proactive as ._ iividuals in polic: 4 programmatic those among us, for example, who initiatives to try to make things better. There have become act iely involved in trying to improve State Assessment and National Assessment processes. There are those among us who have become directly involved in the development of programs (commercial and noncommercial) that build on current ref earch. There are those who have become active in the whole language movement, a movement sweeping the country because it has a solid basis in theory and because it stresses teacher empowerment, the power of myth. Individuals who make these efforts do so at some risk to their status in the research community. Surely there are philosophical and ethical issues involved here, but I trust we can find resolution to these concerns in a way that does not separate researchers from practitioners but builds bridges. I applaud all of these efforts, even though I might not be comfortable personally with some of the out:omes. I believe the more involved researchers are in the w orld of practice, the more we will insure that practice and science are in tune with one another. Acting out individually can make a difference, but it is not enough. We can take steps in our own research to adopt methods or combinations of 21 would like to extend my appreciation to the following individuals for sharing with me their insights on such issues. Richwi Allington, Donna Alvermann, Kathryn Au, Robert Calfee, Diane DeFord, Jan Dole, %her Doyle, Gerald Duffy, James Flood, Lany Fricdman, Yetta Goodman, Jerome Harste, Elfrieda Hicbert, Peter Johnston, Michael Kamil, P David Pearson, Virginia Richardson, Robert Ruddell, Patrick Shannon, Steven Stahl, and Paul Wilson. Methods- that explore the personal, constructs of: teacherkaheut;teaching, the'liersenal meanings that students censtraet,aS parthfleariijng,(seeErickson,,19 dO0hC0',,,t41i6rk-fP1**140'646.*** of classroons liii 4 0.610.0le with both hi*PretivP 444 '40,*14,00 tr:-.10t*L04,4911.,111,Vi -research We car): come -"closer to eaptUring;-teachely,intititionS,anfterhattse, u#(10',1414:how. teraOcts 80* ONet).4tIgiY,Oti regardu exCellence: in ,teaching.that, coMe mit of Sueh research efforts'.Will.oic4*.gik-Olfilc , kookgc.Eurther,,these:findings.-mithe,:lesa'sedUctir for ,quick fixes apd,, therefore; leig vulnerableto the,kinds_of,ahttseStasSeeiatect. precessiitoduct iresearch. Adapting, our research methods is something that can help hithe Jen& this is not a sufficient response in that it does little to address the.eeireat,'Oroblemk we face. We can speak out as individuals against the abuses surrounding us. le Bre (1988), one of the leaders in the process-product tradition, writes: Research on teaching and research on teacher effects in particular, has 'a grestt:cleal; to offer by contributing to the development of a knowledge base thinfoith professional; practice. However, it is a misuse of such research to use it as a bisis for diveloPing, simple-minded and rigid guidelines of the "behavior X correlates with the !trident achievement gain, so teachers should always use behavior X variety." (p. 20) Surely, we must speak out as individuals. But it is not always clear that a single. voice, however renowned, will be heard above the noise of a stampede. Speakingeut . as an individual is not enough. We can assume a voice as a total research community and in unison "just say, no" to the absurdities that surround us. This is, in fact, what I believe we must do if, we are to break out of the horrible cycle we have become locked into. Where might we find such a collective "voice"? I do not believe that the NRC is the appropriate platform. The NRC has a singular, focus and that has been and should continue to be as a forum for sharing origiail: research. The NRC is dedicated to the advancement of a science of literacY.thrOugh, research. It is the goal of science that binds us together. This is not to say .that always agree with one another. Anyone who has spent time at NRC in sessions.er, eavesdropping in "vital issues", recognizes the tremendous diversity in our memher-, ship. Our diversity is rooted in the fact that we do not always share the same mythology. And that is as it should be. It is perhaps the differences in our mythology that make ts interesting, amusing, challenging, motivating, and even aggravating to one, another Occasionally we fall victim to the temptation of trying to use research,to prove our particular mythology to be better than someone else's. But we recognize this cannnot ever be done and come back together year after year to share research I would not want to threaten this focus or this diversity by asking the NR,C ro assume a new role. I do believe, though, that it is perfectly appropriate for the NRC to assume 4 leadership role in encouraging action by the literacy research community. I am askink, that the NRC consider sponsoring a meeting of the leadership of such organization's, as the National Council of Research in English, The American Educational Reseateh 1I *yth of Teachirig sociation, The National Council of Teachers of English, and the International Readmg.Association. The purpose of such a meeting wouldobe to explore possible ways wwhich we, as a profession, might be able to monitor the application, misapplication, And ignorance of literacy research in policy and commercial' initiative:3 Ithas been observed that for every complex problem there i& a simple solu;tion,and it is usually wrong. What I am proposing is neither simple no-r'is Solution to all of the problems facing teaching today. It is simplya:kfarting,pOirit. I: '-.believe such an effort can make a difference in the long run. ThkpeOpre fosticintthe ;-; ,...A.inds of abuses I have cited are well-hitentioned individuals, but tyPicalb, nnsnformed Or uninformed, not just about research findings but about whatieseareh y.aock-.40; ..can be expected from it We can challenge the popular perception that good ieaOhitig can be mandated through policy initiatives whether those initiatives comefrOfirdie central office in a local school, the central office in a district, or the staie agency for education. We can, perhaps, begin to turn the tide toward a return to a 'trusting, empowering view of individual teachers. I am not, by the way, waxing nostalgic for the good old days of the 1950s. Myth without science is ignorance. We must continue to work to create a science of literacy learning and teaching, but that science in the hands of teachers must live with and take life from myth. A single science, perhaps, but not a single myth. The diversity that makes us interesting to one another as researchers is the stuff of which exciting teaching and schools are made. Several years ago I attended a reception honoring 10 outstanding classroom teach- ers. Each teacher was given a few minutes to describe what brought them to and sustained Clem in teaching. One after another they related moving testimony to their personal commitmer.ts to help, to serve, to enrich the lives of children. The final honoree began by saying that she felt a bit guilty listening to all of the other winners She confessed selfishness as her prime motivator. She described herself as addicted to learning and that teaching was the only place she could find to satisfy her habit That 'confession' and the comments of the other teachers reminded me of the power of myth in professional life. It reminded me, too, of the words of Joseph Campbell (1988) when he wrote: People say that we're al! seeking meaning in life. I don't think that's what vo.'re really seeking. I think what we're seeking is the experience of being alive, so alive that our life experiences on the purely physical plane will have resonances with our own innermost being ad reality, so that we actually feel the rapture of being alive That's what it's all about, and myths are the clues to the spiritual potentialities of the human life. (p. 5) Can we create a science of teaching literacy that supports the experiencing of a personal mythology in teaching? Car we bring about a renaissance of thc. art of teaching which is nothing more or Las than the creative unleashing and expression of a personal mythology? Can we demonstrate to teachers that science is their ally and not their enemy? Can we be as tolerant in cur acceptance, indeed encouragement, of multiple myths of teaching as we are in learning to accept and encourage multiple myths of research? I believe we i..an and we must do all of these. It is our professional obligation. 'On December 3, 1990, the NRC Board of Directors approved a motion to sponsor such a meeting jorsj, Lia In the end, it may not be turtles all the way down, but-as:Gerty Duffy (19,8ile calce noted, it may seem at times to teachers that their ieet are surrounded b3),anitotoov-;:: At such times myth is not a luxury but a necessity to persevere and perhaps:eVeti excel. REFERENCES , Brophy, J. (1988). Research on teacher effects: Uses and abuats. Elementary School Journak8920:. Catnpbell,J. (1988). The per of myth. New York: Doubleday. Defino, M., & Hoffman, J. V. (1984). A status report and content analysis of state mandated,reacir, induction programs. (Tech. Rep. No. 9057). Austin, TX: The University of Texas at Aun, and Development Center for Teacher Education. A Doyle, W, (1983). Acalemic work. Review of Educational Research, 53, 159-199. Duffy, G. G (1982). Fighting off the alligators: What research in real classrooms has to say about reading' instruction. Journal of Reading Behavior, 14, 357-374. Dunkin, M. J., & Biddle, B. J. (1974). The study of teaching. New York: Holt, Reinhart, & Winston. Erickson, F. (1986). Qualitative methods in research on teaching. In M. Wittock (Ed.), Handbook of , research on teaching (3rd ed.) (pp. 119-161). New York: Macmillan. Green, T. F. (1971). The activities of teaching. New York: McGraw-Hill. Hoffman, J. V fi Defino, M. (1985). State and school district intentions and the implementation of tie2wi, teacher programs. (Tech. Rep. No. 9067). Austin, TX: The University of Texas at Austin, Researchl.,s?41 and Development Center for Teacher Education. , , Hoffman, I V , & Rutherford, W L. (1984). Effective reading programs. A critical review of outlier4 studies. Reading Research Quarterly, 20, 79-82. Hoffman, J. V., Edwards, S. A., O'Neal, S., Barnes, S., & Paulisson, M. (1986). A study of state= mandated beginning teacher programs. Journal of Teacher Education, 37, 16-21. Powell, A. 0, (1980). The uncertain profession. Cambridge, MA: Harvard University Press. Shulman, L S (1986) Paradigms and research programs in the study of teaching: A contemporary perspeet,,, tive In M C Wittrock (Ed.), Handbook of research on teaching (3rd ed.) (pp. 3-36). New York:, Macmillan. Travers, R M (1973). Second handbook of research in teaching. Chicago: Rand McNally. Weber, G (1971) Inner city children can be taught to read. Four successful schools (CGE Occasional Papers #18) Washington, DC Council for Basic Education. (ERIC Document Reproduction Service ,, No. ED 182 465) Wittrock, M C (1986) Handbook of research on teaching (3rd ed.). New York. Macmillan. 30 4., - 4- - z. = k 1,TIADINGWRITING eONNEC.PQNS: !rligAELATION'AMO 'Mgt #ERS,Pgq174S' : - . , 144j7,4 - r litob0;t3:'TIOney Ohid.pate ilnivirshy Over the past decade there have been a ntimberOttheereticatanalyseS:a4,0b-;.ished-,summaries,ef reading-writing relations (Aulls, 1985; Atiten,,..1983;7Beiinger, Rid*, prianinri;.'1,9132kHantill Sc,119NOtt;490(1; .405$4;Ahlin.a4a4i 19_80, 1.00.;:!iaiSio, 1982, Stötsky, 19$2, :19:03;?Tos,- 1984;:Tierney'8C Natio ;4083; Wilson ,..49T.-,7116, osplisko i4a49014taiio.s: as dlaisexist,,howevr,-:kaye usually tteafe.4 #4420,oarne;=friim singic .iethrticat Perspietive: RecentlY,y0'.00p),002,0 -0.4* 771ipoi:iesyiew chac is to be.published in tiA Ifondhook of Reading Research ThtasJeCiryea, yctunpublrshed review considers' ,studies conducted from all'thrje opw usedtoexaMine ihe COnnectionSbetween readiag. and viriting::Thikoper will attoieto bui1d-upon and'extenci'tliat Chapter. 4-1.ereW,e will, review some of the theoretical 'and' einpirical.work,:that has:26:0C , 1 P4c4 .Of 0i64.0 .0ersijegtives, 91T.1.9 ofthe more tYee* , ,affeits .and We will attempt to analyze Some of the underlying asStinnitiniiSsutolting, theworkineacharea . Ourpuiposeinthisistoattempttodiscoverwbetherthe " ininii.ate assumptions are due to simple inattention to the otherperi '1-, 'veaiiriihether. they resulefroM serious disagreeinents abatit the nattne'ind:Meariing of hieraêy I it . former, that is if the differencei are ,More "sodiologibat, or Jiiitprieat, , #003cheinent of the.varioustraditions would kern-to be in order, and'swkieeining .-"Ogether of t...eories might serve to extend the power of reading-writing4elatiOnsiip This analysis will be limited for the most Tart tO Work completed; siiiCe 140: iidiist,studies conducted before this were single nonprogranimatic efforts fSitch studies va v4Thre1Y, were based on theoretical positions.that.Went initch.beyond,,, the yariableabeing ,:exiitilined or the manipulations being made. Since,198kreliance On d*Or'y OPO. extensive, research'illis become increasiniiy, pingelnynaii6,- ninnbtra niyariiiRes --N:ConsiderOd:have increaspi, as has the complexitY: and 'sophistication of the me:as:tire!. L*nt and data analYtic techniques. V. Litericy Theory and keie4.41 : SHARED KNOWLEDGE-SHARED PROCESS Studies from the first perspective, shared knowledge-shared process, linen** identify communalities that exist in reading and writing, and how 'wen:light-0**#, , reading and writing achievement simultanemisly by.taking advantage Of,*Se C00,,,,#, nalities. These studies proceed from a belief that reading and, writing eingoy;jr,o,oplt,-4 mon knowledge base, and that within reading and writing iimilar,Mental'ope#01ii. are used to process the information. ., ., ..,- ., tt The theoretical perspectives underlying the shared knowledge-Sbored!ipt: 'sa::.."t, studies are reading theory, cognitiVe psychology, and linguiStics. The largeStiiettene:: 4 age of these studies have been conducted by those whose primary acidentic co., : '- t., merit is to reading education or reading psychology, Although these stpaio**- focused on children as young as 3 years old (Galda, Pellegrini, & Cox, 1989)Ihrritigby,4 adults (Sheen & Heerman, 1985) most of the studies have focused on elementiOii'l .. ,,, school children or older persons with limited reading abilities. f. Knowledge Relations . What kinds of connections between reading and writing have been identepiA Analyses of children's compositions and test performances in reading have reyealeCi significant positive correlations in the area of vocabulary (Maloney, 1968, ShanOluin,',f,';',1 1984; Vairo, 1976); syntax (Evans, 1979; Evanechko, 011ila, & Armstrong, ,i914;:it Heil, 1976; O'Hare, 1973; Shanahan, 1984) narrative and expository text orga.nizitoo4,4 (Braun & Gordon, 19R4; Cox, Shanahan, & Tinzman, in press; Gordon & 1982; Hiebert, Englert & Brennan, 1983; Shanahan, 1984; Stein, 1978); spelling anclq word recognition ability (Hammill & McNutt, 1980; Juel, Griffith, & Gough, 190;-,:,q Moni: & Perney, 1984; Shanahan, :984); phenemic awareness (Ferroli 1987; Juel, Griffith, & Gough, 1986), writing mehanies including mpitO1ization',..1 punctuation, and correct grammar usage and reading achievement (Hammill Nutt, 1980); spelling accuracy and reading fluency (Zutell & Rasinsld, 1985); use cohesion and reading achievement (Cox, Shanahan, & Sulzby, 1990; Cox, Shanahan,-, & Tinzman, in pr-cs); appropriateness of voice in composition and reading achieve.:l ment (Beach, 198', Cox, Shanahan, & Tinzman, in press); motivation (Bruning, Shell, & Colvin-Murphy, 1987); development of maid idea in the writing of narratives', and reading achievement (Beach, 1984); patterns of response to literature (Simon,,,,, 1980); development of background setting in narration and reading achievement' ,f,2. (Beach, 1984); creativity in writing and reading comprehension (Fishco, 1966); reading comprehension and writing productivity or fluency (Bippus, 1977), and sense genre (Langer, 1985). A common property of these studies is that they are based on static measurements, of the characteristics of texts that children have written and outcomes of produtt.,,? measures of reading ability Many of these measures come from a skills-based instruct.,:, tional tradition (i e., spelling, syntax), whereas others are drown from theories ot: linguistics (i e , cohesive harmony, story grammar components), or motivation (i.e.,,;4 attributions). :-.Reading-Writing Connections Process Relations One particularly important innovation of the 1980s has been the attempt to measure reading-writing relations through the analysis of reasoning, problem solving, and information processing strategies and activities. Think alouds, pause analyses, or :retrospective accounts (Birnbaum, 1982; Langer, 1986; Martin, 1986,, 1987;-Ryan,, 1983) have been used to determine the processing patterns in reading and,Writing. These studies have identified relations across reading and writing between Cntities such as idea generation, metacognition, structuring, evaluating, revising, roOriitoring, qtrestioning, and hypothesizing. Up to now, such studies hF focused On rathéi -small groups of students, with little effort to use common nomenclature or to identify processing similarities across the studies. Nevertheless, they have generallY fouad 'moderate levels of relationship across reading and writing, and from this it aPpears ;that readers and writers approach their tasks in similar, or at least in parallel, ways. Generally, in all of these knowledge-process studies, the correlations between reading and writing have been rather small, and usually of no more than moderate levels, say about .60 (Belanger, 1987; Shanahan, 1984). in other words, it has rarely been found that specific reading and writing measures explain more than 30-40% of the variance in each other. These figures might be improved, of course, through the use of more reliable measures, or the construction of more specific reading measures (as has usually been the case with the writing measures), the usual outcome of interest has all too often been a global, standardized reading comprehension test. However, such improvements would likely have a limited impact on the size of these correlations as evidenced by the results when tests have been reliable (Shanahan, 1984), and when more specific measures are evident (Langer, 1986). Even attempts to measure several of these relations simultaneously (Langer, 1986; Shanahan & Lomax, 1986, 1988) have only resulted in increasing the median amount of variance explanation by a modest 10-15%, still a moderate level. One might extrapolate from the magnitude of these correlations that reading and writing represent somewhat overlapping modes of learning, but that they are not sufficiently overlapping for one to suffice for the other. Ins. r words, it is doubtful that reading activities alone would be sufficient to lead to learning to write, or that writing actil Sties alone would be sufficient for developing reading ability (at least with regard to the types of instruction suggested by the aforementioned medsures). Similarly dubious would be attempts to teach some underlying component of literacy knowledge or process without opportunities to use and develop this component in both reading and writing. The possibilities and difficulties of using reading to influence writing ability, and writing to influence reading ability, is borne out in the experimental studies. It has been found that reding activity and instruction can influence or enhance writing ability (Eckhoff, 1983, Felland, 1980, McConnell, 19831. -../1:1 that writing instruction can enhance reading ability Bereiter & Scardamalia, 1984, Eiucational Testing Ser- vice, 1984, Kelley, 1984, Straw & Schreiner, 1982). However, the transfer between reading dna wriig occurs with varying degrees of success (Belanger & Martin, 1984, Campbell, 1976. Crowhurst, 1987; Ferris & Snyder, 1986; Michener, 1985; Nielsen, 1980; Raphael, Englert, & Kirschner, 1986). (Also, see earlier reviews by Shanahan [1986] and Stotsky [19831.) We estimate that significant cross-lite:acy transfer occurs 16 in only about 30% of the experimental studies; a figurepossibly inflated by ..the ten=..; dency of journals to include only positive results. However, we do not wantto that in rich contexts where reading-writing connectiOns, are ongoing anct..eitenSi*: (unlike the contexts described in most of the exPerimentalitudies)transfer-Might:iip; more likely. It is also possible that these experimental-PrOgrarrasjMYO*4if.fti*:,;, impact on students. From shared knowledge-shared process studiesit seems apparent that writing hive only partial communality. TWO alternatiyeexplanaticinfOr..._ the intercorrelation of reading and Writing have' been -Put foryi.#0. suggested that reading and writing are so functionally different that it would(bc sible to measure reading and writing in an, entirely-coinparable-manner, Shanahan and Lomax (1988) have argued that a Closer relatinnship-rnight-beevi in an instructional context that encouraged children to unite reading1andwo#pg, and to be aware of this union. There is some provocative evidence that would' suggeS( thg., readers and writers are more likely to make such connections when.theelassroOrn.:, literacy environment provides an abundance of reading and writing activitieS (Stansell? & MOSS, 1984). The relations of reading and writing are not simple. Some children to'be able to use particular aspcts of knowledge or process well in reading while.not able to do so in writing; others evidence an opposite pattern. A number of studies.' have identified children who were good readers but poor writers, or poor readers while being good writers (Belanger, 1987; Loban, 1963; Martin, 1976; Tierney, 1983).. many as one out of five children appear to fit into these unexpected categories (Be,,'anger, 1987) Thus, for 80% of children these correlations might suggest provocative, instructional possibilities, whereas for these others there rmght be factors that Ate,. associated with the exten't to which reading and wraing are interrelated. With ..feW:, exceptions, the studies of reading and writing relations from a knowledgelprocesS sharing perspective have disregarded children's .nstructional ....stones, thus, most .of these studies provide us with a picture of the relationships under unspecified instruc, tional and social influences, and we cannot know whether instruction specifically, or, the literacy environment in general, can be shaped to influence the magnitude or nature . of the relations or whether they can alter the size of these poor/govd cohorts for wht.,.11 the ccrrelations would be minimal or negative. Nor have such students been, trailed lorg;tudinally to identify whether these patterns change over time with maturation orlearning. Basic Assumptions of the Perspective Perhaps the correlations between reading ., writing are even further confounded. For example, one particularly vexing issue with regard to reading-wnting relations oncerns the role of a general ability or intelligence factor in the correlations. A common reliance of reading and writing upon intelligence could be sufficient to ex-, plain the amounts of relationship that have been reported. In other words, the readingwriting reiationship would not be a functional entity that could be taken advantage of' instructionally, but would be instead a simple artifact due to an underdeveloped thedr retical model This is a difficult criticism to refute as most studies have not partialled; . iv-stag-Writing-Connections 17 out intelligence or general ability. One exception is &study by Galda, Pellqrini, and ;Cox (1989) which partialled out the entire relationship with an IQ measure. However, 'the only measure of high reliability in this study was the IQ test, and thus, the result might Lr an artifact. The counter arguments to this indicate that the patterns of relationship,.between 'literacy and intelligence or general ability are not consistent; they typically have been ;found to be limited in the early levels of literacy, development and to incLease as, literacy development advances (Singer, 1978). Conversely, studies of the rtacling7, lwriting relationship have shown levels of relationship to be relatively stable.aer64 reading levels (Shanahan, 1984; Langer, 1986). This stability Would not beexpected, if the relationship was entirely due to general abilities or intelligence. And; it the; relationship was just an artifact of general ability, intelligence or some simaar faCtor, then we would not expect instruction in reading to have an impact upon writing,;or vice-versa, since general abilities are hypothesized to be relatively impervious to. instructional interventions. As has been indicated, studies of the influence °treading instruction upon writing and writing instruction upon reading have often had significant positive results. Finally, analyses of the relationship with reading, wtiting r I intelligence indicate that reading and writing load on different factors (verbal,comprehension and verbal fluency, respectively) of IQ (Sincoff & Sternberg, 1987), though this might be due more to how reading and writing are measured than to the nature of reading and writing ability. The shared knowledge-shared process studies to a great extent proceul on the basis of idealistic or absolute vk.ws of reading and writing. We term these as idealistic in the sense that they define reading and writing pmficiency in absi.liute terms, separate from any notions of social, communicative, or functional effectiveneQs (this is the:, views, good view that has traditionally dominated school curricula). According to writing is lexically rich, has complex sentences, uses hierarchical or story x,..nmar organizations, is mechanically standL.1, and uses zrthesive links. Good reading results in exact pronunciation, speedy word recognition, and high amounts of recall from simple texts. Although it seems probable that these variables have been selected because of their 'mplication in some, usually unmentioned, aspect of functional effectiveness, because the tasks are generally separated from real uses of reading and writing, the measures have become ends in temselves. Similar proble1.-.:. are inherent in the process studies, although here the ideal view shifts to the actions of the reader-writer and away from the text properties. Whereas those studies which have focused on products have been rtevoid of consi4erations of function and process, the process studies have temied to disregard products. In other words, there is a tendency for good readers and wr;ters to be viewed as those who take a global approach to text, and who are planful and self-aware of their own performances, no matter what the outcome of these actions. The impact of such restricted views of reading and writing is quite pervasive. Growth, in accordance with these views, tends to become largely a hum issue of ...mply learning to do more of a particular activity or to perform an acOvity with greater fluency. For instance, it would generally be assumed with a shared knowledge view that readers and writers become more proficient with word recopition and spelling as they get older, or that they could read or write more complex texts with 18 Literacy Theory and Reseaicia, practice. Some studies have complicated this view, though only slightly, by showing that there are discontinuities in the nature of the reading-writing relationship*ith, development (Birnbaum, 1982; Shanahan, 1984). That is, although the magaitUde of: re."ng-writing relations seems relatively constant across gde levels,..the,jpeCifie asi i reading and writing that are interconnected t:eeminglythariges With Matura tion. In accordance with this view and befitting the measturs employed, for:)!Ounger children, the relationship is largely due to word rxogaition-word productiOn fact*, whereas with more mature readers issues like vocabany, ideation, and teitt ; tion become more important. This suggests, therefore, that development maystill,be: linear, but linear within a hierarchical organization. The problem which arises from such a .zw is that development is not separabl,e from learning and curriculum. Development results from learning, and the reason for the identifiable discontinuities may have more io do with die complexity of what learned in reading and writing than with the individual's cognitive or linguistic abilities Of course, if as a result of these definitions and approaches development emerges . as being tied so closely to what is taught, then it is not surprising that the role of' instruction, particularly direct instruction, is often seen as a critical feature within these models (Stotsky, 1983). According to these views, development is controlled' by learning, but learning is controlled to a great extent by instruction. (Examine Table 1 for a graphic comparison of these assurnptiona with those underlying the communications and collaborative approaches.) COMMUNICATIONS E.TUDIES A second perspective for examining reading-writing relations is communications. Studies fro,- a communications perspective analyze how writers anticipate the needs Table I Assumptions Inherent in Three Perspectives on Reading-Writing Relations Shared KnowledgeShared Process Communications Competency Idealistic, Alnolute Social, Conditional Conditional (Accepts ideal or social) Development Discontinuities of Information Biologicalcognitive Steady State (Possibly biological) Learning Instructionalexperiential Instructionalmaturationalexperiential Instructionalexperiential Separability Yes No Yes Context Role None Social Social & Functional a6 Collaboration 19 Reading-Writing Connections fiotential readers, and how readers use their thinking about authors to enhance their reading comprehension. These studies emphaize comMunications by treating the _reading=writing zelatic aship as a negotiation hetv*n.readers and writera::ithetMical Concerns have long been an issue in Compositiort theory and researer,-(Bakhtin,j973; Iiiioffett; 1058; Nystrand, 1979) and at various times,the-rdleofauthor's'imerifiOns f- his: been a consideration in reading- and' reader-resportse thearies-as*ell_ (Riehl*, -1029; Rosenblatt, 1978). The condatt of this latter, wait haS :the fact that reading theorists have generally:taken cognitiVe-i*obleiii-Solving;,a0=,p-roaches ;nstead of social ones, and literary, theorists:have:been enthialleif,koF.inust: , of the pr.st 50 years, by the so-called "New Criticism," wiaitrrejectic* dtfauthifs. Intentions as a pertinent construct old with some of the newer dectinstrnetiOniit approaches. Current thinking about the nature of reader-author negotiations represents an amalgamation emanating from pragmatics, schema theoretic notions of reading and writing, reader response theory, and a resurgence of interest in social aspects ofEteracy. In other words, current communications studies rely on theoretics mderpinnings drawn from various sources including rhetoric, reader-response theory, conithunieations, speech act theory, literary theory, sociolinguistics, composition theory and Soc:.1 cognition. It should be noted that ;lot all theorists support the idea that thinking of authors and audiences are useful constructs. Elbow (1987) has suggested that writers should ignore audience, as he cautiors that too much thinking about audience, particularly during writing, will serve to cognitively overLad a writer, diminishing writing quality. Stanley Fish (1980) goes so far as to claim that a text can mean anAbing that an interpretive community wants it to with no regard for author intentions. Social theories of literacy do not permit such unbridled, and seemingly endless excursions away from a text, as they recognize the rights of authors as well as readers. It has been claimed, for example, that during read.r.g and writing a social contract exists between readers and writers (Tierney & LaZansky, 1980). Interpretation does occur, and should occur, but the reader has a responsibility to conduct this interpreta- tion in some accord with their beliefs about author's intent. There must be a good faith effort by both readers and writers to communicate, and meaning-making must take place within these parameters. Writers Thinking About Readers D. Rubin (1981) found a small but significant connection between the ability to consider the needs of others and writing quality. Kroll (1978), proceeding from a Piagetian framework, found that younger children were unable to consider audience needs in writing, whereas they could do so in oral !arouse. Older children could d need in both writing and recognize the types of information that an audience v speaking. He concluded that children were able to decel.ier during oral explanations because of the dialogic nature of the communication, while they were mom egocentric in the situation that on the surface appeared to be monologic rather than dialogic. In a later analysis of the persuasive writings of 9-year-olds, Kroll (1984) found that students did make a serious, if not always effective, effort to meet the needs of their 20 LiteracY Theory and Resear:c11: audiences. "Few of the letters manifested either gross egocentrism or a blatant disregard for the reader's needs," (Kroll, 1984, p. 425). More persuasively, perhaps, studies have demonstrated that writers often do ap;: tempt to show sensitivity to the needs of their.audiences by making adaptations itt,;., their writing. Typically such studies have manipidate4 writing conditionshy indicithig; that students should snite a composition Multiple times ,fc-, altemativeltudiencel;., usually varying the audiences by age (Beach & Anson, 1988; Plasse,,1982;?ientice; 1980; D Rubin, 1981; Strange, 1986) or intimacy (Cruwhurst taine, 1984; Richardson, 1980; Rubin & Fiche, 1979). Sach stddies haVe.shoWn that students alter vocabulary choices, use of slang, complexity of apitaX, .ainoUnt off: words, amount of deference, how they establish relationship, cohesidn and types and' amounts of information included in their compositions as a result of the andience variations. Such adaptations do not appear in the research findings consistently, hoWever (Plasse, 1982; Prentice, 1980; Smith & Swan, 197P\ More important than the inconsistency of results is that-these efforts can be criticized for artificially increasing the possibility of finding audience-related yaria, tions, as by their design they encourage students to make such adjustmenti..(When is the last time that you wrote a message fa two or more diffetent pretend audiencesthat really weren't going to read the compositions anyway?) Generally, these studies' find that older students are able to vary their writing to a greater extent than could younger students (usually 9-12 years old). Thus, Fontaine's (1984) conclusion, that "nine-year olds seemed to be trapped, having neither a real nor a representational, image of the audience, but only an ill-defined sense of the 'other' " (p. 20). It has been demonstrated that the most effective writers tend to think about audience more during revision, during the reading portion of the writing process, than during planning or composirion (Raforth, 1989). Most studies of audience awareness have neglected the long term effects of audience awareness training and focused just on the material at hand. A notable exception to this was a brief experiment conducted by Greenlee, Hiebert, Bridge, and Winograd (1986) that showed the influence of instruction on adaptation for audienue. This study examined the effects of writing letters for real audiences versus writing letters as a classroom exercise It reported that the genuine-audience instructional condition led to improvements in grammatical sophistication, handwriting quality, and length of letters Students were found to increase the range of discourse functions that they used in their writings too What characterized these writings as being genuine is that they were actually read and responded to by an audience beyond the classtoom. Similarly, the writing process-confe.ence approach widely popularized by Donald Graves and his colleagues has been found to bc effective in improving children's writing (Hillocks, 1986). including their view of themselves as authors having certain concerns for their audience (Tierney & Rogers, 1989). The hallmark of this rather complex instructional intervention is the social inter.tction of student authors and pel. audiencza. More formal attempts to teach students to anticipate reader needs have been effective as well. Schriver (1986) used "readers' prototols," chat is, readers' responses to existing texts, to sensitize her students to readers' needs and problems. The protocols demonstrate the types of problems that readers have with texts. The analysis of these protocols has been found to have a significant impact on college 38 - -,eading-Writing Connections 21 -students' abilities to predict potential readere' comprehension problems with their own 'Writings. A very different approach has students analyzing the needs of a specific audience .C.31ack, 1989). This study is notable because it askal volunteers to write a perinaiive :essay for a genuine audience that they were interested'iminfluineinglin:thiS.case a student government council). Students were first asked to analyze' the audierkee,*rite an essay, and provide some kind of self-analysis of the persuasiveneSS of their *ay. k :On the basis of their audience awareness students were thin groUped;tinditalf.Of them were taught how to analyze their audience (through a series Of questiOns:abont andi- ance knowledge, values, attitudes, and goals), and rewrote their persuasive pieces. The others simply reworked their pieces without the audience awareness-questiOns. It was found that the intervention increased the persuasiveness of the final papers, as judged, not in ideal terms, but by the genuine audience that had been intended by the writers. No differences in the final performances were due to the initial differences in audience awareness that were found to exist. At least for persuasive writing with specific audiences, the use of an audience analysis questionnaire appears te be a potentially effective way of improving writing. Readers Thinking About Writers There is much less evidence available ...at would indicate that readers think about authors, though the little bit that does exist would suggest that they do think about authors, at least when they are having trouble understanding a text. Think alouds during reading with high school and adult level readers have indicated that when they are having difficulty understanding a text they tend to makc comments regarding author's intentions (Flower, 1987, S. Martin, 1987). In fact, Flower (1987) found that approximately 60% of the difficulties incurred by mature readers wt.-, resolved by resorting to a consideration of authorship. Tierney, I 27ar1sky, Raphael, and Cohen (1987) wported similar findings when fourth gradezs encountered and dealt wi.11 incon- sistent information inserted in text. Suk..h statements are substantiilly less common during the reading of relatively easier materials. Descriptive analyses of the rt. ciing development of children in environments that stress authorship and the social connections of literacy, through peer conferencing and other 3imi;ar activities, have suggested that children adopt a more transactional stance with regard to text (Short, 1986) and that, as a result, children increase their ability o interpret texts in effect;ve ways (Rowe, 1989). Studies of the influence of metadis- course in text (Beauvais, 1989)that is, discourse that an author places in text to make sure that the reader is aware of author perception and intentionhas been found to improve reading performance under scme c....1ditions Finally, studies of the incidence of authorship activities, particularly discussions of author's intentions during peer conferencing and discussions of authors during the read:ng of bawls and children's literature selections, have been correlatei significantly to differences in chi!. dren's ability to re:ognize errors aci :Ziscrepanch.., in text (Shanahan, 1989) and to students' sense of themselves as reaaers and their views of text (Tierney & Rogers, 1989). Again, there is very little evidence that inducing students to think about authors Will tiihance :their reading Terfomnaricebut what islavaliable4S--iii-agree:*at:4 ,F4 this general, proposition. Holever,,ktudies Of author awarencsa'have betP.:riither,yektie;:: with regard .to the theoretical difaensions of thia-constructaines kich,as .theicilt:Ofz.4 reaciet'l,Rtgpos6s; ,interactions or- cliscrepancie,r ia.;.reader',s,-anci the telative-role of it as an interpretive-versus a.MOlivatiOual 4s*;'40(1,4,*.k.414e,,I. autlior awareness for different types of texts have not been riddrhy--.anyrit studies. 444 _ Basic Asswnptions of the Perspective The definition of competency in reading and writing is quite different froin,tha evident in the shared knowledge-process perspe,live. The communicatigas perapectiVeil maintains a view of competency that is more co Iditional. It is nqan atisoluteAet_iitg standards, but is instead dependent on readers' and writets' goals, intentionS,.--indr...,::,e.; circumstances. Accolding to communications views of reading and. writing-liteiacy,;-:'. use is goal-directcd. Writers set out to entertain, inform, persuade their readers, whereas readers attempt to enjoy and learn from their traniklions.",...1-1.2. text The true measure of success in such social endeavors is whether the cbinintiiii44.-,.<24 tion took place effectively with regard to the goals of the participants. There:Can* no ideal text, only text that is effective under some set of social chrumstances:oi-74 conditions. (This social dimension being discussed here is not a "context,".per'se; is not simply background. Though "social context" for learning certainly exists, heTe., we are discussing the social dimension of literacy; an integral part of literacy Jtse1f.):-_,;;:4 Unfortunately, on the basis of the empirical procedures adopted in many of thisi studies, it would be simple to conclude that, according to communications-views (if literacy, good readers and writers are more adaptive of their texts no matter what the,. communicative outcomes. These studies, for the most part, have shown that students can think about authors and audiences and make adaptations in texts on the basis of; categorical information abou' pretend readers and writers. They have not shown, with, one exception, that these a.laptations actually make texts more communicative, an4 in some cases have gone so far as staying with absolute judgments of quality (such; as using holistic ratings rather than primary trait ones). The choices of measures. have generally been based on ideal views of what is communicative rather than ofi... communicons, and in this regard these studies are not very different from those labeled as shared knowledge-shared process. And what is the nature of development within the communications approach? To a greater extent than was evident in the shared knowledge-process approach, biological development, as opposed to learning, is a major issue. Theoretical constructs of Piaget, Kelly, and Werner have been used in this regard. Kroll (1978), for example, has attempted to tie writing development to PiageLan stage theories of egocentrism and decentration, whereas others, such as Piche and Roen (1978), have drawn views of literacy development from Kelly's personal construct theory and Werner's comparative-organismic theory. Within these theories, development is both quantitative (more features of audiences can be considered) as well as qualitative (the complexity and abstractness of the features change with experience). Social development in reading and writing is characterized by categorical changes .4 *`;4Reading-Writing Connections 4.- [1: 23 in the types of information used, the subtlety of this information, and the ability to rise this information differentially on the basis of a variety of potential psycholoelal states. Because of these developmental assurfiptions, studies in this area usually have considered age or maturation differences in subjects, while neglecting reading and writing knowledge. The relative contributions of instruction, experience, and maturadon to communicafion development is still an open question. In any event the reliance on social psychological theories has led to the adoption of a very different view of the role of instruction in literacy development. To a much greater extent than was true of the shared knowledge-process studies, these studies seem to emphasize the importance of genuine practice (the value of practice as a source of experience rather than as a simple repetition of something already learned). In this perspective, the major part of learning appears to be situated in the actual social interactions that taxe place through literacy. Practice is not isoleed from context nor is it rote in nature, as flexibility and resourcefulness rather than habitual response are what is pursued. According to this view, practice is not just useful, it is essential; and the authenticity of this practice is a critical dimension of its potential influence on learning to read and write. A major assumption of shared knowledge-process studies was that reading and writing could be learned from each other, or that both could be positively improved by insLrucuonal emphasis on their joint components. Similar claims are evident in the work on reader-writer commanications. It has been suggested that learning to think about authors during reading will have an impact on writing ability, or that thinking about audiences during wiiting will make one a better reader. Unfortunately, to date, we have more affidavits than studies addressing this suggestion. There is a dearth of studies showing that learning to think in a comniunicative manner i"--om one side of the text would influence in any way the social thinking on the other side of it. Perhaps the reason such avenues of study have been neglected in communications approaches while being emphasized in shared knnwledge-process ones has to do with then basic views of the separability of reading and writing. According to communications views, the wnter always works in anticipation of the reader, aria :he reader must consider the writer s message in terms of the communicative context as well Reading and writing in this sense are not separable. COLLABORATIVE USES OF READING AND WRITING The final set of reading-writing studies are examinations of the collaborative or joint uses of reading and writing. That is, these studies consider the impact of using reading and writing together to accomplish various tasks These studies 1-ave a variety of origins. One idea behind such studies is that reading and writing activities entail vanous types of thinking or reasoning, and if these activities were combined effectively, tasks could be completed better than would be possible if only reading or wnting were used, or if they were used separately. Another basis for these studies is tied to the fact that in the "real world" reading and writing are used interactively rather than separately. If we want to maximize the benefits -c *heir use, then we should explore their combined potential. 24 Literacy Theory and Research' : There are several collaborative studies that have considered school or academic:: tasks such as learning from text, critical analysis of text, discourse synthesis papers,, composition revision, composing with computers, and writing across the curriculum; In addition, a number of studies have analyzedthe combinarion ofseadiagand.writing in the workplace or towards the accomplishment of a "lumber of other social goalai.::: Sticht (1980), for example, found that military personnel in the wnrkplace rarelrjUsS, read, but instead used reading together with other activities including writing. An& Blake and Snyder (1988) found reading comprehension to be a prerequisite to seccer.1.'v ful workplace writing. Collaborative studies gain much of their impetus from theories drawn from cogni7 rive psychology, and the research in this area tends to be done by psychologists and: educators with backgmunds in either reading or writing. There are two major attas of stu in this perspective: those that look at learning and those that consk Ir discourse synthesis. The learning studies have been rather prescriptive in nature, that is, they usually induce students to use reading or writing in a particular manner, whereas the discourse synthesis efforts have been more descriptive. McGinley and Tierney (1989) have challenged the prescriptive approaches as being too constricted to reveal the time power of reading-writing combinations. As they stated, ". . . if we wish to understand more fully the roles that reading and writing play in learning and thinking critically, we must continue to explore students' dynamic use of a fluid set of recursive reading and writing engagements as opposed to examining a rather static set of prescribed reading and writing juxtapositions" (p. 263). However, it is an understandable approach to take since reading and writing do not have to be used together for learning from text; it is possible to learn from reading alone or from reading combined with other processes Text synthesis studies have been more observational and descriptive, rrobably because text synthesis by definition requires a combination of reading and writing, although it doesn't necessarily require a particular type of combination. Most of these studies have tended to have a very practical orientation, by the nature of the tasks that have been examined. >. Reading and Writing far Learning A popular recent approach to the use of reading-writing to learn from text has been the matching of read; tg with various types of writing tasks (note taking, written responses to factual questions, personal responses, formal analyses of texts, and so on) SP.:dies by Colvin-Murphy (1986), Marshall (1987), Newell (1984, 1989) and Newell, Suszynski, and Weingart (1989) have found that the more extensive responbes have led to more learning (possibly due to more thinking time) and to a greater amount of sensitivity to author's craft or to closer integation of prior knowledge of text content On the basis of think alouds of students doing some of these types of writing, it was found that the more extensive responses led to more concern with structure and relationships, w.lereas the less involving forms of writing led to the least concern with these Lang, r and Applebee (1987) concluded that general classes of writing assignments are likAy to lead to very specific thought processes, and consequently to different amounts and types of learning. However, the combination of tasks is probably not this simple. Konopak and 42 -:Iteading-Writing Connections 25 Konopak (1989) in a similar comparison of' note taking, study questions, and essay writing found that in eighth graders (subjects younger, and presumably less literacy proficient, than those used in the other studies) there were no learning differences. Penrose (1989) even found that, on some measures of comprehension, writing tasks actually led to lower scores than other types uf study tasks, probably because students did the writing differently than the teachers had intended; there-was nu itbmatic cognitive outcome. Despite having the same assignments, their interpretat..' ts of the ,tasks had led them to adopt different goals, and these goals led them to carry ont,the reading-writing tasks in different ways and consequently to gain different amounts and types of learning. Tierney, Soter, O'Flahavan, and McGinley (1989), iound that the combination of -eading ai writing, under certain conditions, contributed to increases in understanding and led students to alter their positions on controversial issues, probably because the combinations of reading and writing contributed to dialectical or critical thinking about the issue being examined more than was evident when reading or writing were used alone. These results did not occur under all conditions or for all combinations. They concluded, "Data from the present study suggest that thought processes change over time, and that to assess the reasoning operations engaged during different tasks, researchers must consider time and other contextual features and view the processes more dynamically" (Tierney, Soter, O'Flahavan, & McGinley, 1989, p. 168). In other word., simply assigning an essay to be written after reading will not necessarily ic?fl students to learn. This effect is likely to be mediated by a number of variables including prior knowledge, attitude, purpose, literate ability, ability to return to a textto reread or revise at various times in the process, availability of discussion or other sources of information, and contextual variables that have been strangely absent om most collaborative studies of learning from text. While Langer and Applebee (1987) attempted to nredict the thought processes enabled by specific writing activities, Tierney and his colleagues took a much less prescriptive orientation and argued that it may be problematic to ascribe to any specific kind of writing a certain set of thinking operations or outcomes. The nature of thinking likely to occur with a particular type of writing, such as note taking, is apt to shift over time and circumstances. The learning outcomes and thinking nperations ascribed to types of writing will vary with the wnter's engagement. Tierney and his colleagues proceed from the notion that a domain can be "cnss-crossed" in a variety of ways and from a variety of perspectives so that a greater amount of knowledge flexibility can be derl-d Reading and writing, rereading and rewriting, and meving back and forth between reading and writing are just some of the ways that they speculate this criss-crossing can take place They do not propose that certain types of mat ;14 and writing will necessarily result iv a predetermined set of learning outcomes. Discourse Synthesis Discomse synthesis or "reading-tu-write" studies have usually approached these issues with inore concern for individual mit-rences (Spi-ey, 1984) In this type of study, students are usually provided with two or more texts on a particular tcpic that 26 Literacy Theory and Reseal*: contain overlapping and nonoverlapping information, and sometimes even discrepant,. information about the topic. Students are then asked to write a report using these source materials. These studies have generally found that better readers wrote better, syntheses, in part because of their ability to select out more information with crosaf, textual importance (Spivey, 1984), that better readers were able to develop more. complex triting plans probably because of a greater sensitivity to the structural propert., ties of texts that had been read (Spivey & King, 1989), that students-with-MOre3. extensive prior knowledge were able to elaborate on the synthesized materialin more:, specific and evaluative ways (Ackerman, 1989). As with the learning studies,,hoir one approaches a task has an impact on the outcomes. It has been found that students . select a variety of ways of combining reading and writing on the basis of goals, prior academic experiences, interpretations of teacher-given directions, and their own estimation of the zrtual value of the effort. Synthesis tasks appear to have two typical purposes when used in schools: to create a written object and teaming or knowledge transformation. Studies indicate that the creation of an effective written object through discourse synthesis will be mediated to a great extent by factors such as who the potential audience is, the amount of knowledge the learner has, how the text will be used, and the nature of the task and text. (Newell, & Winograd, 1989; Penrose, 1989). If students simply tabulate information or organize it superficially, knowledge transformation will not necessarily occur (Kantz, 1987). Such studies do not result in neat prescriptions for how reading and writing should be put together However, they do arrive at a number of suggestions about the clarification of goals and the exp'oratinn of alternatives with readers and writers. In Penrose's (1989) words, "When we choose to assign writing as a learn;ng activity, we need to let our students know not just the kind of writing we want them to do ;.:ut the kind of learning we want them to engage in" (p. 16). Basic Assumptions of the Perspective With the shared process-product and communications perspectives, It was relatively easy to discern assumptions We are somewhat less ctrtain with the collaborative studies These studies are very specific to particular literacy routines, contexts, and goals, and the researchers who are studying a particular type of collaboration might not be attending to the decisions made by those studying a different type of collaboration. The competen7 assumption in these types of studies is most basically that good taders and writers are able to use text together flexibly and effectively. Effectiveness, of course, defined in task specific or context-specific terms for a particular goal. collaborative studies reading and writing are not inert tasks, but they are goaldirected ones, and competency is a conditional property dependent on the nature of the goal and the context While in the shared process-product studies It would be possible to assume that good comprehenders are good readers, for instance, in the coi:aborative approach comprehension would not be a sign of competency, though it might be required in a particular use of reading. Unfortunately, the theories of competence for most of these studies are at a level that far outstrips our traditional measurement techniques, and although the definition 44 27 Reading-Writing Connections of competence here is clearly more situational than is true of more traditional views of reading and writing, these studies generally have not done a better job of specifying contextual conditions, and, remarkably, have often done a poorer one. These studies have been most praiseworthy when they have used multiple indicators, especirily online measures such as think alouds or debriefings in combination with product measures. However, studies on discourse synthesis and learning from text have often theoretically defined effectiveness in terms of creativity criticality, integration of information, personal involvement, or knowledge transformations, while using measures of t-unit length, word counts, tabulations of idea units, multiple-choice recall tests, or nonspecific holistic evaluations of writing quality. The reason that these ir Jasures match up so well across perspectives is that these researchers appear to assume that reading and writing alone are necessary preconditions for using reading and writing together competently, and they have consequently, 27.11 probably unwittingly, accepted the competency positions of the earlier discussed perspectives. No clear developmental positions have been specified with regard to collaborative views. These studies, by their use of older students, have given a "steady-state" impression of development as if such changes or reorganizations do not happen or are irrelevant. However, this is probably not what these researchers intend. For examp:e, Beal (1987, 1990) in her studies of how students use reading to revise their compositions, found cleat developmental differences in children's abilitks to recognize textual problems, although similar developmental problems were not evident for the correction tion of these texts. In other words, maturational limits on reading ability were consequently limiting writing. The role and nature of teaching in these collaborative studies is an area of clear division and argument. Then. Are those who believe that clearly defined literacy routines cen be taught and lean.. ..nd that students wit; become more competent in their joint ases of reading and writing (Langer & Applebee, 1987), and those who consider the powerful properties of reading-writing combination to be too complex to be learned effectively in such simple ways (McGinley & Tierney, 1989) Although direct instruction in literacy routines may be found to be effective in the accomplishment of various goals, it is also evident from a number of studies or conditions that a more guided expenential approach, similar in some ways to apprenticeship, is piobably more iP line with the true goals that we hold. In this perspective, reading and writing are sepaiee and separable entities Reading and wnting are not the same tasks, or combining them would make no difference The claim here is not that reading and writing are totally different, just that they accomplish difLrent, though overlapping, ..,gnitive goals, and that under various circumstances one might be superior to the other. In fact, Tierney and McGinley (1987) go so far as to suppose that the processing similarities of reading and writing are what make them so effective for co% ering or criss crossing a domain of knowledge in different ways that are learning supportive. A FINAL COMPARISON OF ASSUMPTIONS Each perspective has its own purposes and its own traditions All of them offer us a greater understanding of worthwhile issues that can contribute to our grasp of the _451 _ development, nature, and meaning of literacy. That these have been largely separate traditions is apparent not only in their different purposes, and in the differences that, exist in their variables and measures, but also in their reference lists. Those whose task it has been to till the fields within any of these perspectives have been loathe to attend to the plow marks in the next field. During the 1980s the literacy research community has witnessed expanding visions; it has been a period of newly opened doors and increased breadth. This expansion of vision is well documented in the programs of NRC over this decade, and hi the expansion of reading-writing research (from virtually none in 1980, to the point where "Resources in Education" has given it its own index entry and boasts more than 530 articles and research papers). This expansion of vision has been a healthy one, but it has in some ways led to a kind of fractionalization of the purposes, approaches, and directions of the field. We have at times been guilty of a too-shallow consumerism, rather than recipients of a deeper understanding of literacy. A greater amount of interaction among these three perspectives could contribute to an increased depth of understanding in the 1990s. The recent coming together of reading and writing could be a harbinger of this vision for the final decade of the 20th century. We have much to learn from and to share with each other. The work on shared knowledge-process has suggested useful instructional innovations, contributed to our understanding of the generalization or transfer of literacy learning, and to a better understanding of important measurement issues 'in reading and writing (such as revealing some of the limitations of using writing responses as an evaluation of reading ability) The work on social dimensions and contexts of literacy learning and use have demonstrated the necessity of understanding the role of communication in our other measurements It has also encouraged a wonderful opening up of classrooms to a number of socially rich experiences. Collaborative approaches similarly highlight the importance of functional or goal-related contexts on literacy development and use, and serve to identify important practical goals for reading-writing education and have provided impoitart insights into what needs to be learned and how we might guide that learning. There are differences in the basic assumptions of these perspectives, but these seem to be due to the partial specifications of reseal_ problems that have been done (Mosenthal, 1983) and to historical accident. For example, is it that researchers who work on collaboration have jus examined samples in which developmehtal change is unimportant, or do they reject deNelopmental chan:;e zs an important construct? Is it that shared knowledge-process measures of competency are based on deeply held beliefs about the nature of literacy, or are they intended as proxy variables for predicting what texts or performances might accomplish the most in social and functional sertings? (A belief not entirely unjustified by the results of various social or collaborative studies that have employed some of these measures.) The suggestion here is not that we all should study the same things in the same ways, only that we have a responsibility for making sure that our research adds up to sumething A good deal of this adding up might come from a greater attention to and use of research findings across these perspectives. We need to make greater use of the issues, measures, and findings of each other's efforts. There have been a number of exciting examples of this type that might point to a more productive future. Dyson's 46 29 Reading-Writing Connections (1989) insightful inquiries into the social func& s and learning of text structures in reading and writing; Raphael and her colleagues' (1986) attempts to combine instruction of text lunowledge into a socially complex reading-writing context and to study the joint and separate effects of these, Penrose's (1989) analysis of the social context of collaborative performance, Spivey and King's (1989) attempts to consider the role of reading-writing knowledge in a collaborative situation, and several others. We are not suggesting that these studies are exemplary in all regards, or that these are the only appropriate examples. However, we are suggesting that studies such as these that use constructs and insights drawn from across two or more of the perspectives have a greater potential for increasing our knowledge and understanding at this time. If we follow such leads and stress cognitive, communicative, and activity-based apprc:hes, not separately, but togetherand use these to consider a whole range of cognitive, affective, linguistic, social, and functional performances across clearly specified social and functional contextsand if we recognize and attempt to understand the joint influence exercised by instruction, experience, and maturation on learning in these contexts, the 1990s will be a time of a deeper and truer vision of what it means to be literate and how we can help our children to get there. REFERENCES Ackerman, J. (1989, Apnl). Reathng. writing, and knowing. The role o; disciplinary knowledge in c "mpre hension and composing. Papa presentai at the meeting of the Amencan Educational Research esssoci ation. San Francisco. CA. Aulls, M. (1985). Understandmg the relationship between reading and wnting Educational Horizons 64 39-44. Auten, A. (1983). Reading and wnong. A mutual s pport system Joarnal of Reading, 26. 366-369 Bakhtin, M. M. (1973). Marxism and the philosophy of language L Matejka & I R Titunik. Trans ) New York: Seminar Press. Beach, R. (1984. Apnl). The effect of reading ability c c wenth graders narrative writing Paper presented at the meeting of the Amencan Educational Research Association, New Orleans. LA Beach. R., & Anson. C. M (1988). The pragmatics of memo writing Written Communication. 5. 157- 1P3 Beal. C. R (1990). The development of text evaluation and revision skills Child Development. 61, 247-258. Beal. C. R. (1987). Repairing he message. Children's mondonng and res....in skills Child Development. 58. 401-408. Beauvais. P. (1989). A speech act theory of metarhscourse. Written Communication. 6. 3-10 Belanger. J. F. (1987. Mayi. Reading achievement arta writing proficiency A critical review of research Paper presented at the meet.n8 of the Canadian Cuusts.11 of Teachers of Engiish, Winnipeg. Manitoba. Canada. (ERIC Document Reproduction Service No. EI) 282 180) Belanger.). F., & Martin. R G. (1984). The influence of improved reading skill on o^ting sL Alberta Journal of Educational Research, 30, 194-212 Barnet. C . & Scandamalia. M (1984). Learning about wnting from reading Written Communication I 163-188. Bippus. A. C. t1977). Inc relationship of p ably of students' written language. productivity of writing, and reading i.omprehension in grades four and six Dissertation Abstracts international 38, 3993A (University Microfilms No 77-28. 639) Birnbaum, I C. (1982). The reading and composing behavior of selected fourth and seventh grad: stu dents. Research in the Teaching of English. 16. 241-260. Black. K. (1989). Audience analysis and persuasive wnting at the k.ollege !eye: Resew. h in the Teaching of English. 23. 231-253. Literacy Theory and Reearch Blake. M , & Snyder, 1 (1988). Occupational literacy: Compking the communication cycle in the work-, place. Nei England Reading A:soar:Idol Journal, 24, 14-18. Braun, C,, & Gordon, C. (1984). Writing inst. uction as metatextual aid to story schema applications. lid J Niles & L. Harris (Eds.), Changing perluectives on research in readingllanguage processing 01 instruction (pp. 61-65). Rochester, NY: National Reading Conference. Bruning, R , Shell, D. F., & Colvin-Murphy, C. (1987, December). Development of self-efficacy outcome eroectancy fa- reading and writing: A regression and causal modeling approach. Paper: presented at the N 11 Reading Conference, St. Petersburg, FL. Campbell, D. B (197o, i study of two methods of teaching English in .1 community college setting. Dissertation Abstracts International. 37, 4808A. (University Microfilins No. 77-2761) ColvinMurphy, C (1986, December). Enhancing critical comprehension of literary te,as through writing.. Paper presented at the National Reading Cunferenoa, Austin, TX. Cox. B E.. Shanahan, T., & Sulzby, E. (1990). Good and poor readers' use of cohesion in writing. Reading Research Quarterly, 25, 47-65. Cox, B E , Shanahan, T Tinzman, M. (In press). Children's knowledge of organization, cohesion, and voice in written exposition. Research in the Teaching of English. Crismore, A (1982) Cognitive processes in the composition and comprehension cf wntien texts. Evaluation in Education (Vol. 5. pp. 231-246). Toronto: Pergamon Press. Crowhurst, M (1987. April) The effect qf reading instruction and writing instruction on reading and writing persuasion Paper presented at the meeting of the American Educational Research Association, Washington. DC. Crowhurst. M & Piche, G L (1979) Audience and mode of discourse effects on syntactic complexity in writing at two grade levels Research in the Teaching of English. 13, 101-109. Dyson. A H (1989) "Once upon a time" reconsidered. The developmental dialect between funcuon and form. Written Communication, 6. 436-462. Eckhoff, B (1983) How reading affects children's writing. Language Arts, 60, 607-616. Educational Testing Service (1984) The El's evaluation of Writing to Read. Princeton. ETS. Elbow, p (1Q9- ,^,-Ising my eyes as I speak. An argument for ignoring audience. College English, 49, 50-69. Evanechko P , Oltila, L & Armstrong, R (1974) An investigation of the relationships between children s performance in written language and their re<iding ability Research in the Teaching of English. 8, 315-326. Evans R V (1979) The relationship between reading and writing of syntactic structures. Research in the Teaching of English, 13. 129-135. Felland N (1980) A national survey A the level of composition achievement (supenon average) of twelfth grade composition students and selected personal charactenstu.s. envtronmentul factors. Dissertation Abstracts International, 41, 3037A. (Universiry Microfilms No. 81-01, 428) Ferroli 1 be Shanahan, T (1987) Kindergarten spelling. Explaining its relationship to first-gnsde reading. In I E Readence & R S Baldwin (Eds ), Research in literacy. Merging perspectives (pp. 93-100). Rochester, NY: National Reading Conference. Ferns, I , & Snyder. G (1986) Writing as an influence on reading. Journal of Reading, 13. 751-756. Fish. S (1980) Is there a text in this class? Cambridge, MA. Harvard University Press. Fishco D T (1966) A study of the relationship between creativity in writing and comprehension in reading of selected seventh grade students (Doctoral dissertation, Lehigh University). Dissertation Abstracts International, 27. 3220A-322IA. ,wer L (1987) Interpretive acts Cognition and construction of discourse. (Occasional Paper No. 1). Berkeley. CA: University of California, Center for the Study of Writing. Fontaine, S I (1984) Evidence of audience awareness in the writing and writing process of nine- and eighteen year olds (Terh Rep No 143) (EPIC Document Reproduction Service No. ED 258 183) Galda, L , Pellegrini, A D . & Cox. S (1989) A short-term longitudinal study of pr=hoolers' emergent literacy. Research in the Teaching of English. 23. 292-309. Gordon, C I , & Braun. C (1982) Story schemata. Metatextual aid to reading and writing. In I. A. Niles & L A Harr4 (Eds ). New inquiries in reading research and instruction (pp. 261-268). Rochester, NY: National Reading Conference. , .:keading=Writing Connections 31 Greenlee, M. E., Hiebert, E. H., Bridge, C. A., & rimograd, P. N. (1986). The effects of slifferent audiences on young writers' hitter writing. In S. Mies & R. Lalik (Eds.), Solving problems in literacy: Learners, teachers, and researchers (pp. 281-289). Rochester, NY: National ReaCog Conference. HammilL C., & McNutt, G. (1980). Language abilities and reading: A review of the literature of their relationship. Elementary School Journal. 80, 269-277. Harste, I., Woodward, V., & Burke, C. (1984). Language stories and literacy kssons. Portsmouth, NH: Heinemann. Heil, H. (1976). The development of selected language variables in two modes of writing and their relationship to reading comprehension in the primary grades. Dissertation Abstracts International, 38, 137-A. (University Microfilms No. 77-13, 727) Hierbert, E. H., Englert, C. S., & Brennan, S. (19831r. Awareness of text structure in tecognidon and production of expository discourse. Journal of Reading Behavior, 15, 63-79. Hillocks, G. (1986). Research on written composition. Urbana, IL National Conference on Research in English. Juel, C., Griffith, P. L., & Gough, P. B. (1986). Acquisition of literacy: A longitudinal study of child= in first and second grade. Journal of Educational Psychology, 78, 243-255. Kanre, M. (1987). Promises of coherence, weak tomer:4 and strong organization (Tech, Rep. No. 3). Berkeley, CA. University of California-Camegie Mellon University, Center for the Study of Writing. Kelley, K. R. (1984). The effect of writing instruction on reading comprehension and Orry writhig ability Dissertation Abstracts International, 45, 1703A. (University Microfilms No. DA 84-21, 346) Konopak, S. P., & Kocopak, B. C. (1989, April). Differential effects of writing tasks or eighth graders' learning from text. Paper presented at the meeting of the American Educational Research Association, San Francisco, CA. Kroll, B. (1978). Cognitive egocentrism and the problem of audience awareness in written discourse Research in the Teaching of English, 12, 269-281. Kroll, B. (1984). Audience adaptanon an children's persuasive letters Wrinen Commwlication. I. 467-4717 Kucer, S. (1985). The making of meaning. Reading and writing as processes Written Communication. 2. 317-336. Ablex Lange:, S. A. (1986). Children reading and writing. Structures and strategies. Norwood, Langer, J. A. (1985). Children's sense of genre. Written Comnumication, 2, 157-187. L.anger. I. A., & Applebee, A. (1987). How writing shape rhinking. A study of teaching and learning Champaign, IL: National Council of Teachers of English. Lehr, F. (1981). ERIC/RCS. Integrating reading and writing instniction. Reading Teach.r, 34, 958-961 Loban, W. D. (1963). The language of elementary school children. Champaign, IL Nadonal Council of Teachers of English. Maloney, H. B. (1968). An idenuncatieli excellence in expository composition perrorinance in a selected 9A population with an analysis of reasons for superior performance Dissertation Abstracts huernational. 28. 3564-A. (University Microfilms No. 68-2432) Marshall, J. (1987). The effects of writing on students' understanding of literary *tat Research in the Teaching of English. 21, 31-63. Martin, P. (1976). A comparative analysis of rearFng and writing skill Six case studies EnglIsk Australia. 14, 17-21. reading and writing Paper Martin, S. H. (1986, December). A companson o f cognitive processe .. p-rsented at the meeting of the National Reading Conference, Austin, TX. Martin, S. H. (1987, December). Thr meaning.making strategies reported by providetu readers and write-s Paper presented at the meeting of the National Reading Conference, St. Petersburg, FL McCoonell, M. A. (1983). The effect of literature exposure and writing practice on the original narrative writing of second grade children. Dissertation Abstracts International, 43, 26I9A (University Microfilms No. DA 83.00, 306) McGinley, , W., & Tierney, R. (1989). Traversing the topical landscape. Reading and writing as ways of knowing. Written Communication, 6, 243-269. Michener, D. M. (1985). A quasi-expenmental study of the effects of reading aloud to third grade students as reflected in their written composition skills. Dissertation Abstracts International. 46, I545A (University Microfilms No. DA 85-18, 078) ' 32 Moffett, J. (1968). Teaching the universe of diSCOUrse, DOM 1: Houghton Mifflin. . Morris, D, & Perney, J. (1984). Developmental spelliag as a p.xlictor of first-grade reading achievement.. Ekmeruary School Journal, 84, 441-457. Mosenthal, P. (1983). On defining writing and classrocaa writing coropeuzce. In P. Mosenthal,-L Tatnoc dk: S. Walmsley (Eds.), Research on writing, princkqes and method:1'pp, 26-74). New York: LOnginfari.,:: Newell, G. (1984). Learning from writing in two content areas: A case studylprotocolanalials4tesearch7: in the Teaching of English, 18, -205-287; Newell, G.. & Winograd, P. (1989). The effects of writing on learning from expository text. Frgt#1!>.., Commwikation, 8, 196-217. Newell, G., Susrynski, K., & Weingart, R. (1989). The effects of writing in a reader-based anittextaaatiV' mode on students' understanding of two short stories. Journal of Reading Behavar, 2/,`37-18.,. ,;;;, Nielsen, B. F. (1980). Effects of reading on children's narrative writing. Dissertation Abstracisiritiii07Ilona!, 41, 99A. (University Microfilms No. 80-16, 081) Nystrand, M. (Ed.). (1979), What writers know:The tanguage, process, andstructure of written discourse. New York: Academic Press. Nystrand, M. (1986). The structure of written communication. New York. Academu. Press. O'Hare, F (1973). Sentence combining improving student writing without formal granunar Instruction (NCTE Committee on Research, Report Series, No. 15). Urbana, IL National Council of Teachers of English. Penrose, A M (1989) Strategic differences in composing. Consequences for learning through writing. Unpublished manuscript. Piche. G , & Roen, D. (1987). Social cognition and writing: Intapersonal cognitive complexity abstractness and the quality of student's persuasive writing. Written Commwikation, 4, 63-89. *_,,sd ' Plasse, L A (1982) Thz influence of audience on the assessment of student writing. DLuertation Abstracts . International, 42, 3940-A. (University Microfilms No. DA 82-03, 053) Prentice, W. C I :80), The effects of intended audience and feedback on the writings of middle grade pupils Dissertation Abstracts huernational, "1, 934A-935A. (University Microfilms No. 80-20, 379) Raforth, B E 3989) Audience and information. Research in the Teaching of English, 23, 273-291. Raphael, T E , Englert, C S , & Kirschner, B. W. (1986). The impact of tea strucmre instrucrion and social r ontext on stude.us' comprehension and production of evository text (Research Series No. 161) East Lansing, MI: Michigan State University, Institute for Research on Teaching. Richards, I. A. (1929). Practical criticism. New York: Han.uurt Brace. Richardson, E M (1980) The quality of essays written for distant and intimate audiences by high and low apprehensive two-year college freshmen Dusertation Abstracts huernational, 41, 971-A. (University Micrrrilms No. 80-21, 064) Rosenblatt, L (1978) The reader, the text, the poem. Carbondale, IL Southern Illinois University Press. Rowe D %V (1989) Author/audience interaction in the reschool. The role of social interaction m literacy learning. Journal of Reading Behavior, 21, 311-350. Rubin, A , & Hansen, J (1984) Rearling and writing. How are the first two s" related? (Reading Education Report No 51). Champaign. University of Illinois, Center for the Study of Readiag. Rubin. D L (1981) Adapting syntax in writing to varying audiences as a function of age and social cognitive ability. Journal of Child Language, 9, 497-510. Rubin, D L . & Piche, G (1979) Development in syntactic and strategic aspects ot audience adaptation skills in written persuasive communication. Research in the Teaching of English, 13, 293-316. Ryan. S N ( '983) An examination of reading and writing strategies of selected fifth grade children. (Doctoral oissertation, Georgia State L'niversity). Dissertation Abstracts International, 44, 2105A. Schriver, K A (1986, April) Teaching writers to predict readers' comprehension probkms with text. Paper presented at the meeting of the American Education Research Association, San Franaisco, CA. Shanahan, T (1980) The impact of writing instruction on learning to read. Reading World, 19, 357-368. Shanahan, T (1984) Nature of the ruding-writing relation. An exploratory multivanste analysts. Journal of Educational Psychology. 78, :16-123. Shanahan, T (1986) Tb- reading writing relationship. Myths and realities. Wisconsin State Reading Ass&ct- ation Journal, 30, 9-18. Shanahan, T (1988) The reading-writing relationship. Seven instructional principles. The Reading Teacher. 41, 636-647. . 'Reading-Writing' Cônnections Shanahan, T. (1989, December). Authorship and critical reading. Paper presented at the meeting of the National Reading Conference, Austin, TX. Shanahsa, T., & Lomax, R. (1986). An analysis and comparison of theoretical models of the reading-writing relationship. Journal of Educational Psychology, 78, 116-123. Shanahan, T., & Lomax, R. (1988). A developmental comparison of three theotetical models of tbe readingwriting telationship. Research in the Teaching of English, 22, 196-212. Shanklin. N. K. L. (1982). Related reading and writing: Developing a transactional theory of the writing process (Monograph in Language and Reading Studies). Bloomington, IN: Indiana University. Sheen. S. V., & Heennan, C. (1985). Intercorrelations of measures of reading and writing: Implications for college research on reading and writing. Perceptual and Motor Skills, 60, 677-678. Short. K. G. (1986). Literacy as a collaborative experience (Doctoral dissertation, Indiana University). Dissertation Abstracts International, 47. I674A. Simon. P. E. (1980). An exploratory study of the relationship between re..V.ing response and natactic maturity. Dissertation Abstracts International. 41, 2533A-2534A. (University Microfilms No. BO25. 960) Sincoff, J., & Sternberg. R. (1987). Two faces of verbal ability. . Intelligence, I I, 263-276. Singer. H. (1978). Research in reading that should make a difference in classroom instruction. In S. Samuels (Ed.), What research has to say about reading instruction (pr . 57-71). Newark, DE: International Reading Association. Smith, W . L., & Swan, M. B. (1978). Adjusting syntactic stnictures to varieti levels a emeience. Journal of Experimental Education, 46(4), 29-34. Spivey, N. N. (1984). Discourse synthesis. Constructing texts In reading it& writing (Outstanding DissertatMn Monograph Series). Newark, DE: International Reading Association. Spivey, N. N., & King. 3. R. (1989). Readers as writers composing from sources. Reading Research Quarterly. 24, 7-26. Stansell, J. C.. & Moss. R. K. (1984). Ouldren's perceptions of reading and writing: Making the connections. In 3. Niles & L. Hams (Eds.). Changing per4reczives on research in readingilanguage processing and instruction (pp. 283-287). Rochester, NY: Natireal Reading Conference Stem. N. (1978). How children understand stories. A developmenud analysis (Tech Rep No 69) Urbana University of Illinois, Center for the Study of Reading. Sucht. T. G. (1980). Literacy and human resources developlent at .vrk. AtexandeA. VA. H sources Research Organization. Stotsky, S. (1982). The role of wnung in developmental leading Journa of Reading. 25, 330- u0 Stotsky. S. (1983). Research on reading.' writing relationships A synthesis and suggested direct. guage Arts, 60, 627-642. Smange, R. L. (1986). An investigation of the ability of sixth graders to write with sense of audience (Doctoral dissertation. Indiana University). Dissertation Abstracts International 47. I638A Straw. S., & Schreiner. R. (1982). The effect of sentence manipulation on subsequert measures of reading and listerrng comprehension. Reading Research Quartezly. 17, 339-352. Tierney, R. 3. (1983, December). Analyzing composing behavior. Planning, aligning, revising Paper presented at the meeting of the rational Reading Conference, Austin, 7X. Tierney, , 1. J., & LaZansky, , J. (1980). The nghts and responsibilities of readers and writers A contractual agreement. Language Arts. 57. 606-613. Tierney. R. I., LaZansky, 3.. Raphael, T., & Cohen. P. (1987). Author's intentions and readers' interpretation. In R. 3. Tierney, P. L. Anders & 3. N. Mitchell (Eds.), Understanding readers' understanding (pp. 205-224). Hillsdale, NJ: Erlbaum. Tiemey, R. 3.. & Leys, M. (1984). What is the value of connecting reading and writing' (Reading Education Report No. 55). Urbana. University of Illinois. Center for the Study of Reading Tierney, R. I., & McGinley. W. (1937). Reading and writing as ways of knowing In Language and learning. Proceedings of the 13th annuu: Australian Reading Conference (pp 19-31) Gosford. NSW: Ashton-Scholastic. Tierney. R. I., & Pearson. P. D. (1983). Toward a composing model of reading Langaage Arts. 60, 568-580. Tierney, R. 3., & Rogers, T. (1989). Exploting the cognitive consequences of variations in the social fabric of classroom literacy events. In D. Bloome (Ed.), Learning to use literacy in educational settings pteraiy ant **I:and eognyve (90,s2:56465); Notwood, Ticipoyi-R, SO'tcr, A O'Fidaiwi,,..L`Fi, (iSkTkiff*ii.of Opoii,thitd4ag criticidli. Reading Re-Warr-A Nartssiyi:26., 141,473:: reldias "sal wrking V1,40.'EALiI.f. (074 '11!.# 4106.4114 k4:4)reli *TY- ift.414*Iiii,iald athinveeent In *it 37!,i10594I. :0iiiitrikt444-094:* 0°;:-.7649 -Wilson.;M.1.(19,31). A riiiovi of'recentetaiest.0 Ca ih,e'Oefr.0943,9t-e".4118,4*-40.1* iti#14 Tiacher,.4i0190-001. - , WO, T. V. (1985:-Dicenter) : SPelltig'ob#Ity and Orsiffig AN0014:1,9±f#fi relatelfar OfrdindRth graders? Paiiiiiiiii4todtd the peOd!ig,Ot,themoilya St. Ottersburg; FL. `,9; 32 1111.1.211/6L-- MOTIVATED LITERACY' Mary M. McCaslIn Bryn Mawr College I awoke tine morning a couple weas ago with the knowledge that I was "ready to write" this paper on "motivated literacy My preparations accompanied the sthell of a second, necessary, pot of freshly brev g coffee. I conducted the search for the scotch tape and white-ow, ..Z.ocked that t: sapler was loaded, the cartridge in my pen full, pencils sharpened, and clean nturow-ruled white printers' paper at hand. Coffee and tools now ready, I noted the dictionary on the side table, retrieved my outline, and turned on the conputer. My writing routineshere recounted minus the angstam, I know, of not much interest, but I dwell on them in part because I suspect thT, are commonplace and because I think they have much to say about the acquisition and maintenance of motivated literacy. I note that how I go about the process of writing in 1989 is mnre an elaboration than a replacement of earlier scripts. I do not need my fountain pen and, frankly, the only time I use a stapler anymore is to fix a torn hem. But, in another sense, I do need these tools that are now incidental to the actual production process. Perhaps because they were essential components of an earlier learning, the learning of what it meant to be "literate," they remain necessary but are no !onger sufficient I now cannot write without a computer, it has been this way for some time, although I cannot recall precisely when ! changed from typing my paper on the computer to writing my paper with it. I only knov, that the computer has transfouned my thinking and my writing ir fundamental ways. It has become part of my "tool kit." We cast all recount our journey into the world of literate behavior. discovering how to look at the Sunday funnies, keeping hack of time until recess, counting out "one potato two pota . to see who is "it" for a game of tag, telling ghost stories at sleep-overs, writing the obligatory thank.you notes for birthday greetings, adding items to the grocery list at some point knowing that "eggs" go in the same column with "milk." As a child I spent many hours sming customers charge receipts for my father's business while my mother posted the hooks. It occurs to me I should look up "posted" m the dictionary. I do not know if it is a "reel" wcrd, only that it meant a lot to me. Ruler, pencil, and, most importantly, eraser in hand, I would settle in at the kitchen table and add the figures in each stack, only to have them redone on the Plenary address pmented s the annual meeting of the Nanonal Reading Confereace. Austin. Texas. December 1. 1989. The author wishes to acknowledge and thank Thomas L. Good for his helpful comments The research program described In this paper was funded in part by the Junior Leave Program of Bryn Mawr College and a Spencer Postdoctoral Fellowship. 35 Sta. adding machine by my mother. But sometimes my sum was solely correct. she had misentered a figureand that's what "psted" meant. It was my expression of an 8year-old's smugness and said with the cadence of my mother's more frustrated expletive "posted" was idiosyncratichighly personal and saturated with meaning. Each of us owns such "special" words, whose connotation is rich and private. I go on because wish to make the point that we need net adopt the role of "outsider" to disco ver the pattern in the phenomenato explore the metaphors of apprenticeship, guided participation, shared knowledge, and sociall1 situated learning (cf Lave, 1989; Rogoff, 1989)in our understah I iug of motivated literacy. Nor does. the pursuit of motivated literacy seem to profit from a narrowing of attention to` differences in knowledge, power, status, or lvlongingness embodied in linear-change constructs like "expert" and "novice," or "newcomer" and "oldfiraer." Ineted, reflections upon our own jcarneys into knowing what it means to be literate construe a web of persons, situations, tasks, and activities, evolving roles, developing uncktstandings At this stage of our learning about motivated literacy, a broadening of perspective seems a more tenable strategy. So, we each have a history of literacy passage in the world of childhood z.:id into the w mid of grown-ups that involves the everydayness of observation and modeling, and direct engagement of tasks and activitiesmany of them simply routinethat are verbal and nonverbal, and that, although unique in its particulars, I maintain is global in its essence. Through our multiple social, instructional worlds we. have learned what it means to be literate, what was once interpersonal knowledge has become intrapersonal knowledge I maintain that mothated literacy is like any other meaningful learning It involves both incidental and intentional processes within the learner and within the sociarinstructional s%tting, hence, to understand the dynamics of an individual learner, one must attend to the changing social contexts withm which the developing learner emerges. Motivated literacy, then, is learnedboth as it is acquired and as it is maintainedand here understood NV i th i n the dynamics of a Vygotskian framework of social and individual "emergent" interaction (Wertsch & Stone, 1985). First I will elaborate briefly upon historical Vygotskian theory, highlighting the role of socialization processes and the development of functioual language, in the facilitztion of what I call Adaptive learning. I will describe a program of research on the processes adaptive learning and I will argue that adapthe learning is fundamental to the construct of "moth ated literacy." Second, the potential of this peispective, which builds upon Vygotskian theory,, for exploring the dynamics of motivated literacy is examined by considering what it means to be literate in a cross section of our society. current popular culture, the academy,, and our public schools. The focus here is upon the messages our students likely receive from parents, teachers, and "everydayness" about what it means to be literate Third, conceptions of motivation and ability are examine -! as they may inform a construction of mot...iied literacy. Emphasis here is on the multiple and potentially ..ompeting definitions of mothated learning held by schools, teachers, parents, and peers. Finally,, some implications for the integration of these conceptions as they may inform instructional practice are Lonsidered with special attention to current reform efforts in the teaching of readit% . aammosslawskriarmiwalMarl 37 Motivated Literacy HISTORICAL VYGOTSKIAN THEORY There are three interdependent facets of a historic_'. Vygotskian perspective that are especially relevant to a tho _y of adaptive learning and its implication ?. for motivated literacy: (a) the multiple functions of language, (b) internaliwoioa processes and the nature of change, and (c) the unit of psychological analysis (for mOre comprete discussion, see McCaslin Rohrkemper, 1989). Each process informs the indiViduars mediation of experience, an experience that is at once cultural (it repreSenis SOeiallY structured tasks and tools) and historical (it reflects the "storehouse" of what -Way we call "semantic knowledge," "learning to learn strategies," "comprehension monitoring," and "metacognitive awareness"). I will briefly discuss each construct. F Theory of Language Vygotsky was an avowed Marxist and his theory of language extends Engels' position .hat communicative, social language evolved from and within human le and was uniquely human. Similarly, Pavlov (1927) provided a critical distill.. between the "first" signal system, perception, and the "second" signal system, language. Pavlov hypothesized that language, the second signal system, was the cause of the differences between human and animal learning. He argied that "it is precisely speech which has made us human" (as quoted in Slobin, 1966, p. 112). Thus for Pavlov, as for Engels, speech was peculiar to humans, and, in interaction with perception, the first signal system, allowed one to master the environment as opposed to being controlled by its stimulus properties. Language, then, is responsible for the human ability to direct and mediate behavior. The mediational and self-directive role of languageof the second signal systembecame the cornerstone of Vygotsky's research and theorizing. His interest was in the dual functions of language, communication and self-direction, and how these evolved. In particular, he focused upon the dynamics of childh....'s transition from exposure to word meanings by others in their social, instructional environments to children's emergent ability to expose themselves to word meanings and thereby direct their own behavior as well as communicate with others. The developmental sequence of the two functions of language is from social or interpersonal to self-directive or intrapersonal. The implications of this progression are critical. Not only does language acquire two distinct functions, but the soureP of self-directive inner speech Ls the social environment the cultural and lu ,torical language environment. The structure and function of each type of speech -external communicative and intemal self-dhectivediffer. Inner speech branches off communicative (external) speech. Vygotsky argued that ultimately, , "the speech structures mastered by the child become thc basic structures to his thinking" (Vygotsky, 1962, p. 51). Inner speech, then, is the opposite of external speech. External speech involves turning thought into words, whereas inner speech involves turning words into thought (1962, p 131) Inner speech is thinking in pure meanings and is the link between the second signal system of the social world and the thought of the individual. 38 Literacy Theory and Researe Processes of Internalization The sequence of language development, from interpersonal and c,ommunicative with others to intrapersonal and self-directive, squarely locates the emergent capacity for "self" direction in the interpersonal realm; the role of the social environment preeminent. Mind is the product of social life, it is a "form of activity which was., earlier shared by two people (originated in communication), and which only later, as a result of mental development, becomes a form of behavior in one person" (Luria;, 1969, p. 143). The psychology of the individual is a multiplicative product of his social encounters. "Emergent interaction" has been coined by Wertsch mid Stone (1985) to capture the dynamics of this internalization process that integrates the social/instructional, environments in the child's experiencethe interpsychological, cultural worldwith child's natural developmental processes. Internalization, then, is inherently social and interactional, and at its core is the mastery of signalslanguage. This corception of internalization embeds the individual within her culture; it blurs the disfnction between self and other. The individual is intricately a part of the perceived sock2 world; thus self-knowledge is not independent of knowledge of others One could argue that reports about self are not interpretable without a context of "percept a of other" within which to analyze them. Indeed, one finds many empirical examples to suggest that terms like "self-perceptions of ability" vary with context and/or comparison group (McCaslin, 1989, Midgley,, F;Adlaufer, & Eccles, 1989; Reuman, 1989). Unit of Psychological Analysis Vygotsky voiced concern about the false dichotomy that characterized much of psychology at the turn of the century and that continues. He anticipated present-thy attempts to integrate "will" with "skill" when he wrote (Vygotsky, 1962): We have in mind the relation between intellect and affect. Their separation as subjects of study is a major weakness of traditional psychology since it makes the thought process appear as an autonomous flow of "thoughts thinking themselves segregated from the fullness of life, from the personal needs and interests, the inclinauons and impulses, of the thinker . . . . (p. 8) Thus, Vygotsky argued that the basic unit of psychological concern was the integration of the afcective with the intellectual and their emergent interactional ongms with the socialinstructional environment. He used word meaning as the basic unit of .nalysis for exploring this integration. Current Soviet psychologists have challenged this last position and offer instead the construct of "activity" that embodies toolmediated, goal-directed action as the appropriate basic unit with which to examine the integration of the affective with the intellectual (Wertsch, 1985, Zinchenko, 1985). FROM THEORY TO RESEARCH These three tenets of the historical Vygotskian theory of the social ongins of psychological activity inform the structure of a program of researa on adaptive learn- 56 39 Motivated Literacy ing to which I now turn. This structure also organizes subsequent sections concerned with the implications of this general perspective for a theory of motivated literacy Ultimately we will explore cultural definitions of literacy, the potentially competing constructions 4: motivation that emerge from multiple social/instructional environments, and the potential informativeness of the integration of these cultural messazes in approaches to reading instruction. Adaptive Learning: Towards a Definition My primary interest is in the development and enhancement of what I call "ada2tive learning" (see also, Rohrkemper & Corno, 1988). By adaptive learning I refer to that ability to take charge of frustration and maintain the intention to learn while enacting effective task strategies in the face of uncertainty--taking charge of one's motivation, emotion, and thinking. Adaptive learning allows one to initiate and to transform tasks: It enables proactive behavior in either case. Adaptive learning is not yet another euphemism fc high ability. Indeed, my research provides some support for the hypothesis that given the structure o our classrooms and the diagnostic insensitivity of much class- Jm work, moderate ability learners are provided more opportunities to become, and to know themselves as, adaptive learners. This is because high ability learners too often headily succeed at tasks that lack challenge. Hence, these students have little need to be adaptive learnersto reread, redo, or seek assistance. Too readily earned success on tasks that are not difficult enough, and thus are understimulating, is not educative Uniaformative success, or success that occurs without increased understanding, does not further learning, not about the subject matter at hand nor about oneself as an agent in the face of uncertainty. . Thus, high-ability :,....ners are less apt to experience thos situations of hard learning in which we learn about our frustration thresholdsin which we learn how to swallow the lump in our throat and go back and try a different way In short, those situations in which we learn how to lAve with stress and we learn that we are more than our achievement. Experiences that facilitate adaptive learning that high ability students do engage in are likely those that occur when they transform a classroom task to make .t more challenging (e.g., compete against earlier times to completion, construct an essay with internal rhyme) or to occur in nonschool settings (e g , with parents, community groups, sports). Similarly, low-ability learners also are inure apt than their moderate ability peer to suffer from diagnostic errors in task Jifficulty. These students chronically fall short on tasks that are too difficult, and thus overwhelm, or succeed on tasks that are so prescnptive and SU furiously designed for success that the learner is rendered numb attainment. In either case, mindless failure or un:nformaand ultimateiy p&ssive , tive sucess, the low-ability learner is less apt to experience a range of task challenge that results in strategic self- and task transformation behavior Rather, it is the "moderate" ability learner who is more apt to experience a balance between more and leSs easily "learned" tasks and "earned" outcomes. Thus, the moderate ability learner has more experience W ith a variety of learning outcomes. This 'likely affords a distinction between process and outcome in the first place, and secondly, , allows for an understand.ng uf the malleability of the learning 4? 40 Literacy Theory and Research process Such understanding allows elaboration and fine-tuniug of a range of learning and motivational strategiesan expanding and specialized tool kit. Hence, it is the moderate ability learner who is most apt to display what I call adaptive learning: to recognize that tasks, and the strategies we bring to them, are malleable. And it is the moderate ability learner who is more apt to enact this knowledge rather than become undone by a too difficult task or defeated by one that is too easy. These learners know that they can recovei from initial failure or incomplete learning and yet sfill be capable of success. They do not overdwell on either outcome. Moderate ability learners.main- tain a relatively flat emotional profile; they hold generally positive, but focused; expectations unless evidenc, does not allow them. When moderate ability learners do fail, they tend to not engage in self-evaluation and instead remain neutral and keep their thoughts about themselves close to the task (McCaslin Rohrkemper, 19890. They are more rilient. Adaptive learning is not ai ther term for high ability, then, it is acquiredthrough experience with a range of tasks within multiple, supportive social/instructional environments. Adaptive learning is not isomorphic with self-regulation. I term the facility to transform and initiate tasks and self "adaptive learning" rather than self-regulation because i want to stress inter- rather than intraindividual states. It is essential to stress that a Vygotskian perspective highlights the role of the social/instructional environment in the development of adaptive learning By social/instructional ehvironment I refer to institutions, parents, teachers, peers, tasks, and activities that students influence and are influenced by as they engage in learning, be it about themselves, their community, or the imagery in Jane Eyre. From this point of view, then, adaptive learning underlies motivated literacy. The research program described here attends to reported inner speech as a function of task difficulty, type of socAL'instructional environment, and individual differences among learners It can be con.-,idered an elaboration of a Vygotskian perspective that incorporates insii,hts from attribution theory (e.g., Weiner, 1985), information processing theory (e g,, Simon, 1969), social learning theory (e.g., Bandura, 1977), and socialization research (e.g., Baumrind, 1971). Conception of Change One way to conceptualize "emergent interaction" that involves .chool-aged children is to consider ihe co-occurre ice of aevelopmental processes with a change in socialization Thus, children experience an increase in the number of social/instructional environments in their lives at about the same time that they become capable of eing in control of themselves lather than Lontro.:ed by the stimulus properties of the social 'instructional env ironment Children's sociaLinstrutional worlds expand considerably at about the same time they develop an increasing facility with the second signal system ar.d emergent capacity for self-direction. This is especially the case for children of working mothers (Scarr, Phillips, & McCartney, 1989). Exposure to an increased number of sociaL'instructional environments requires adaptive learning Como (1989) for example, discusses just this point in her analysis of classroom lite-acy being able to read 'assrooms as textwhat she defines as the "proce,, of coming to know the commonly acknowledged stnictures and functions of 58 . 41 Motivated Literacy classrooras and of being able to use this knowledge productivity [sic] in the social and academic roles that classrooms define" (1989, p. 30). Some sociallstnictional environments are more supportive, informative, and appropriately challenging and, thus, facilitate adaptive learning better than others. And gaps between the social/ instructional environments of hoitie and ;chool can demand adaptive learning of some students more than others. For example, the basic classroom social frame cif turn taking, that rests upon a master-subordinate physical frame, is congruent with early adult-child interaction in middle-class homes (Corno, 1989, Heath, 1982). Middle-class parents, like teachers, also are more apt than working class parents to ask their children questions for which thr: parent already knows the answer (Heath, 1982). And striking inconsistencies between home and school can occur in the norms surrounding things like helping behavior, usually valued at home, yet often considered cheating in school (Good, 1988). Interviews with parents, teachers, and students foceled on the enhancement of adaptive learning and the treatment of mistakes also provide some insight into the associations some .Audents have wi h adults repeating themselves "slower and louder" (Mc Cash'', in preparation). A teach rdr answering a student question by restating more slowly and loudly usually is assuming that the student needs repetition and more time to "stay with her," or was unable to hear, or that understanding and listening are correlated. Parents' slower and louder speech, on the other hand, is often associated with children's misbehavior, or thoughtless behavior, and thu., has a ring of culpability, it is a warning and often a promise of punishment. Hence, some students are confused, but convinced, that the teacher yells at them when they ask a question. They tacitly learn that not understanding is akin to misbehaving, one does not make mistakes, one behaves badly. Thus, home and school language and social interaction patterns can differ in important ways. As Corno (1989) argues, not to address the differences between home and ctassroom social, instructional environments is to place some children in a "catch as catch can" position toward their adjustment to the classroom. We know that familiarity Instills perceptions of self control and that perceptions of self-control promote effortful behavior (Bandura, 1971). Currently, our classrooms are more familiar social/ instructional env ttunments for some students then they are for others. And Is is so often the case, students frt.a higher socioeconomic homes are moi:. likely to receive i. farniLs language and a familiar set of expectancies in school th a.".. an, their less advantaged peers. Classroom socialization also homogenizes speech and other social behavior, and so defines what It means to be litcrate even at it teaches literacy (Graff, 1979 in Corno, 1989). Thus, classroom sociaL instructional environments not only make demands on students, they are simultanouusly sources of empowerment as students internalize and mediate their expenences (Halpenn, 1976, Rohrkemper, 1984, 1985). As students acquire more expenence .1 classrooms, their language to describe classrooms and their pexcepUons of themselves within mem becomes more school like than homelike. Hence, parents' relative lack of facility to describe their children in classrooms becomes more pronounced as their child nes through elementary school (McCaslin, in preparation). Indeed, students' sense of themselves as learners maps onto teachers' 9 s, 77 42 Literacy Theoty and Research.. constructions so that students' self-descriptions and reported inner speech in the face of effortful learning is congruent with teacher description and instructional language (McCaslin Rohrkemper, 1989a). One hypothesis to emerge from this scenario con-, cerns the extent to which the capacity to integrate the home and school social/instructional environments is an important determinant of enhanced perceptions of oneself as a learner, the development of functional inner speech, and, hence, adaptive learning., The Functions of Inner Speech in Adaptive Learning Inner speech guides thought and action in nonautomatic "effortful" (Posner, 1979) cognition. In this research program, two types of inner speech were identifiect that reflect concern with the integration of the affective and the intellectual (McCaslin Rohrkemper, 1989a; McCastin Rohrkemper, 1989h; Rohrkemper, 1986; Rohrkemper & Bershon, 1984; Rohrkemper, Slavin, & McCauley, 1983). Self-involved inner speech reflects control over the self through enhancing motivational and affective statements (e.g. "don't get mad [because] you cart io it, just hang in there"). Taskirvolved inner speech reflects control ove; the task through problem solving, strategic instructional statements afforded by the task, and modification of the task if necessary and possible ("ok, start over on a new piece of paper with a completely new way [rereads directions, changes algorithm]"). Together, self-involved and task-involved inner speech enable adaptive learning by allowing students to modify the task or the self, and by. empowering them to initiate and transform tasks or self. Results indicate that students differ in the fluidity of their reported inner speech, the sophistication of the taskinvolved strategies that they can employ, and in the types of affective and motivational configurations that enable them to persevereto overcome self-doubt and maintain the intention to learn, and to quitto recognize when perseveration is overdetermined and not facilitative. It seems reasonable to hypothesize that, even within developmental and task differences, the sources of taskinvolved inner speech are more readily identified and homogenous, and tied to specific school learning or, if found lacking, to sttatt...tt ability level and/or quality of prior experience In contrast, sources of self involved inner spezch are likely more varied, reflecting multiple influences from home, school, and peers. An example may help clarify this distinction. The following were excerpted from interviews with two sixth- grade girls discussing how they handle the "hard stuff" in math. Their reports ze typical for thi it age group wher mporting inner speech associated with difficult tasks. It should be kept in mind, however, that these students are discussing their approaches to coping with learning stress in general. Inner speech Involves turmng words into thought; here we have compounded the process by requesting that the pathway be made prototypical and then communicative for others. Thus, the density and structural differences that are theorized to characterize inner speech have been stereotyped and diffused in the translation The reports are, nonetheless, informative in that they provide clues about the functions of inner speech. A lot of times I get sick of things so I just want to stop. And I do . . . I always, whenever I'm working and I just get sick of working and I just stop because I can't stand it anymore, I think of things that are, I like to do. Like in school, I'm going 60 lfutivated literacy 43 to play with my friends. I think, "Um, all the things that are fun that we do, and stuff. But I have to get this done and right before I can go and do that." Compare this student's self-involved strategic use of fantasy, combined with real- world contingencies, to keep her on task with her classmate's reported strategies described below. The first student's reported inner speech indicates tht learning (or perhaps, more correctly, successful performance) is a means to the goal (fun time with :friends). Her classmate's reported inner speech indicates motivational and emotional supports that are enabling means to the goal of meaningful learning. Well, I think I'm going to get them all wrong. And I kind of feel like I have to get up and walk around and think about it. I feel like I have to stop and work on somettung else for a little bit. I might get up and work on spelling for a minute 'cause that's pretty easy and I don't have to think about it, 'cause spelling I just know the answers and they're right there. I can think about the math and what I'm going to do . . [It's time for a break) when I get pretty frustrated and think to myself you can't do this and I start tearing, I start biting my pencil then I know I have to get up and do something clse. I just I get so frustrated with it I can't think . . I start to ficille with my hands, go like that. I know I have to do somcthing else. 'Cause I really get mad. I don't take a real long [I mak] nme, maybe just ten minutes. Then I come back to work again. Just to get it out of my mind for a minuo. Both students conclude with similar procedural or algorithmic task-hvolved strategies to reach solution. Their self-Involved paths to that solution illustrate the range and complexity of self-directive inner speech and its dynamic interplay with one's general comprehension of oneself as a learner. The examples underscore the question of internalization, and in sc doing, move us away from locating the "psychological" solely within the individuai. We look instead to the nature of the multiple social/ mstructional environments that, through emergent interaction with the individual, result in unique 'Lamer cunstructions and re constructions of self-direction Thus, understanding how a studcnt copes with present learning frustration involves some understanding of how pnur and ongoing socialization influen:es of home and school have been internalized. We look as well to the specific events that allow this development, to the types of tas&b that stimulate inner speech. As stated earlier, tasks that do not require striving ' do not challengi and therecjre do not directly pic .ide the opportunity for the development of adaptive inner .peech. Similarly, tasks that are too prescriptive do not allow students to learn abuu , themselves as learners znd therefore do not enhance the development of self-directive Inner speech (see also, Rohrkemper & Como, 1988) One implication uf this interpla) betv,een task demands and the development of adaptive, functional Inner speech wak..er:._ how to design :asks that will enhance the integration of self-Involved and task-Involved Inner speech so that each is mutually supportive In sum, students differ in their affective and intellectual strategies for coping with differing tasks. A Vygotsloan orientation is distinctive in its interest in the emergent interaction between the developing individual and the changing contexts of his or her ; multiple social, instructional env Iruiients. This internalization process, in interaction with tasks that are challenging and Informative, results in unique constructions of self s, and fluidity of functional inner speech, and hence, adaptive learning z 1 ; : i 1a Literacy Theory and Research CULTURAL DEFINITIONS OF LITERACY We now consider some implications of students' adaptive learning, and the general Vygotskian perspective, for a theory of motivated literacy and informed classroom practice. We first turn to current debates about literacy and how they differ in the, popular culture and the academy. Popular Culture American citizens have become worried of late about our relative standing in the world Certainly concern with our economic competitiveness has been with us through: much of the decade; indeed, our worries about Japan can now be expanded to include, the i'uropean common market. Our national identity, that sense of what it means to be an American, at some points reduced to "Engiish language only" conceptions, also is likely to undergo increased deliberation as other countries no longer provider, the easy conn.ast points between "us and them" that makes us comfortable with the "us " So, I think it is fair to say that American citizens have become worriedly self-; conscious. And worry has fueled debates over what it means to be literate, and who..; gets how much, when, and where. The popular press is replete with international_ achievement test comparisons. We read that our children's ranking in math andscience and literature and geography and facility with languages is not any more competitive. than are the cars produced in Flint, Michigan. We learn that our country's willingness to invest ii. die production of knowledge (in research, development, and dissemina(ion). other than defense weaponry, is fallingbehind the investmeats of other societies. , We read of the "closing of the American mind" and student editorials in the Newt York Times (1989) claiming that, at one of our most prestigious colleges, students are A not allowed, let alone encouraged, to learn Instead, students are pushed to quickly, produce It is "extra good" if you can complete a double major. Parent concerns that their infants ultimately get into these prestigious schools are well known, profited from, and parodied in movies like Baby Boom. We laugh. There is emerging evidence that we damage our children in our self-promotion of their success, yet we still. indulge Popular culture portrays both an insecurity and a concomitant assertion that there is a cannon, there is a standard of knowledge that is requisite to being literate. And this Christmas, the flyer in New Yorker tells me, you can purchase J.D. Hirsch's (1989) A First Dictionary of Cultieal Literacy. What Our Children Need to Know at your local Doubleday Book Shop. What a Christmas. I dwell perhaps overdwell -on popular culture conceptions of literacy, and the context of those conceptions, because of the marked influence that everydayness has on individual thinking and behavior. Jacobs and Eccles (1985) for example, have documentcd a ..hange a decrease in raothers* assessment of their daughters' mathematics ability as a function of popular peports on the research program on gender and mathematics giftedness conducte ' Johns Hopkins University. The Academy Even as the spin it racks at the supermarkets tell us there is a body of knowledge, that if retained, will mak as educated, the academy is rife with debate about what it 62 , 1 45 drfotivated Literacy .means to be literate and the eraics of differential access to that state. My college is not unique, I am sure, in out discussions about the relation between tradition and "knowledge." Academics debate the stance that some texts ate worthier than other 'texts, that there is some authority concerning what is to be known, that knowledge is 'historical memory not mere "data" or information in the ever-present. It is a lively debate, to be sure, and the source of some wonderful constructs like "cultural autolobotomy" (Ozick, 1989, p. 124). is that faculty and students around the country are angling to define My 1. , what it means to be literate and for whom. How do we think about itmerican minorities outside of the Sociology Department? Women outside of Women's Studies programs and child development research? International and nonwestem cultures outside of political science? The scenes at Stanford University this past year were a vivid but not unique example. At the same time that curriculum debate and revision are thc struggle of the academy, however, the popular culture is quite confident that it has the hold on the means to and what it means to be literate. Our changes in curricula clash with popular conceptions. Perhaps is this nowhere mori., evident than in the Whole Language movement toward reading instruction (McCaslin, 1989) CULTURAL DEFINITIONS OF MOTIVATION Multiple, and at times ,ompeting, definitions of what it means to be motivated and able are embraced in our culture. It is useful to consider the Ntential array of these t..onstruaions from parents, principals, teachers, and students, as part of the context of motivated literaq. To illustrate, let us examine four sources of "authority" on what it means to be motivatedfo,ir distinct social instructional environments,.nat overlap in the life of one sixth ,:ade female, Nora, a studerit of moderate ability.. Home SociallInstructional Environment Nura's family t,onsists of several 'mothers and wwking parents. Each is expected to "du the best you t. an." Effort is highly valued and effortful learning is emphasized more than ready learning associated w ith ability. So much so, that Nora's mothe* is concerned that Nora does nut "earn" her way heause "she learns easily and doesn't need much study." There is no premium on high native ability. Effortful performance is distinct frum effortful learning in Nora's family Effortful performane essentially concerns acting responsibly. The home approach to responsi ble behav lor means that -ertain mistakes are "OK" (e.g., those that occur in spite of sustained effort or that are due to "legitimate lack of awareness) whereas others are not (e.g., due to lack of sustained effort or "irresponsible" behavior). Given the high va/ue that home plaes on effortful learning, one hypothesis that follows that sanctioued mistakes followed by effort may we:: be the most valued behavior in Nora's farm.) morality is intertwined with effort, self awareness, self reliance, and the golden rule. 63 46 Literacy Theory and Researc School Social/Instructional Environment Nora's school district's motto is "one year's growth for every child" as defined ky scores on the swadardized achievement tests admin:stered each spring. Nora's school expects to meet, if not exceed, district level goals. Academic expectations are paramo mt fro.a the principal's perspective, especially-standardized test performance School-level expectations for student test performance and the premium on realized high abii.ty infiltrate the classroom though several': channels, ranging from who is selected to read the morning PA announcements, tir-,1 whose achievemeti. is celebrated in those announcements, to who leaves the room foi "gifted" classes At the school level high ability combined with effort that results in high achievement test performance receives the highest acclaim. Classroom Social/Instructional Environment Nora's teacher's (Mrs Smith's) conception of student learninp differs from both the effortful learning and performance that may or may not be successful, valued by Nora's mother, and the effortful ability, resulting in high standardized test scores, valued by the principal Ars Smith values successful, effortful, achievement. She seemingly equates effortful behavior with high classroom achievement, classroora tasks are believed appropriately targeted in difficulty level and sequenced by subskill' so that students will perform successfully if they try. With effortful cognition, students w ill learn "Learning" means successful achievement. Thus, student effort is defined' by successful achievement. In this classroom. like Nora's home, effort is always expected. An important difference, however, is the linking of effort to outcome. At home, effort per se, independent of Jutcome, is required. In the classroom, effort is defined in large part 1-1 the outcome Thus, at home one can evaluate one's effort by the value and intention . that underlie the process, in this classroom, for this teacher, the outcome of effort must be knew n to determine its value. And the value of effort espoused by mother and teacher is at cross purposes to the value on ability cmnveyed by the pnnemal and, as we will see, constructed by the students. Peer Social/Instructional Environment These are sixth graders Teachers observe and interact with groups of students at the same age levels and, as a result, have a broader conception of age-related behavior, motiv ation, and cognition than do most parents. Hence, what Mrs. Smith understands as developmental, "stage related," patterns in peer relationships and judgment style pical for ,sixth grade students, Nora's mother attributes to Nora's tnegattve) personal- ity traits that mother "wants changed." Both women are aware of the peel and friendship pressures students this age are experiencing. Neither Nora's mother nor her teacher appear aware of the premium sixth-grade students place on ability, however. And Jr these sixth-grade students, ability is defined by rate As Stipek (1984) and others have discussed (Ames, 1988, Nichols, 1984), by sixth grade students are well ensconced in a compensatory perception of ability and effort such tha; morc expended effort indicates less expendable ability. Effort sakes time Amount of time spent on task ;s a public Index of effort readily 64 'Motivated Literacy 47 available to students as well as classroom researchers Only sixth graders are not apt . JO equate "time on task" with motivation or opportunity to learn, rather, they are ,more likely to infer level of ability. Nora and her classu:tes readily discuss how fast they can do their work relative to others, how people feel when they are slower than other, and, importantly, how to compete on rate without being too obnoxious so as not to jeopardize friendships, ; Hence, the fourth effort/ability message in Nora's social/instnictional worlds emerges: effort is inversely related to ability; one's personal worth is defined in large part by one's ability; the student who visibly and continually tries has limited ability. The msultant cliques organized around status in the "gifted" programs and other school-level perks are not surprising. Albeit by a differing route, the students arrive at a hierarchy similar to their principal's. Mrs. Smith's, and especially Nora's mother's definitions of motivation, based as they at; in effort, the controllable aspects of learning, define the "also rans." We see, then, another constellation of influences on the developmcnt, enhancement, and enactment of inner speech. MOTIVATED LITERACY AND WHOLE LANGUAGE INSTRUCIION Within a general Vygotskian perspective, Nora's motivation to learn is interpret . able within the framework of emergent interaction, the integration of the social/instruc tional environments of her experience her rruerpsychological cultural worldwith her natural developmental processes. Nora's conceptions of literacy are also interpret . able within this framework. The emergent interaction of her developing capacity to "know," her changing understanding of literate behavior of what it means to know, and her negotiation of the multiple, and as wc have noted, potentially contradictory notions A literacy held by home, "everydayness," and school (an ' even here potentially defined differently by principal, teacher, and peers) is a useful tool for under . standing motivated literacy. Nora's mother may well look poorly on an assigned essay that comes home with errors in punctuation and spelling, yet boasts a "good thiraing" message from the teacher. Nora is apt to get a "mom talk," as she calls her mother's sermons, and made to do it over until it is "right." Nora's teacher's credibility is likely on the line, certainly by Nora's mother, and now perhaps by Nora as well. Both Nora and her mother are getting messages from the popular culture that the form and physical attractiveness of the messages are what is important not its accuracy or level of cntical comment. I suspect that this is not an unusual scenario for teachers who are trying tu teach dnguage arts as sumethtng other than, or in addition to, grammar and spelling ruics, and who have children write their own stories for reading material Certainly these differing con, ptions of what it means to be literate between home and school have some role in Nora's emergent motivate..1, literacy. What these examples allude to, of course, are the curricular changes in language arts instruction that are associated under a pretty broad umbrella termed "whole language." As I understand the whole language perspective, advocates share with Vygotsky (1978) the belief that it is not possible to have a direct influence on and produce clange in another and look instead to educative ..;portunities that are socially 48 Literacy Theory and Research, I situated Indeed, whole-language advocates are sympathetic with T. S. Eliot's lament- that "we had the experience but missed the meaning" (Eliot, 1943). Hence, they, reject any involvement in "direct instruction" or the presentation of teachers and texts as authoritarian transmitters of knowledge to the awaiting, if passive, learner. This portrayal of "educator" is not likely the one envisioned by Nora's mother, or anyone else's, for that matter. In their zeal to reject reading instruction that ay.., tends to direct instruction in everything but comprehension monitoring, some wholelanguage advocates seem to have equated lark of instruction in comprehension with the futility of instruction in omprehension. I doubt that this is seen as educative by the popular culture criteria or in many homes. Indeed, the failure to instruct does not iitorm the efficacy of the instruction if it were to occur. The Vygotskian portrayal of the active, constructive, and reconstructive learner who does not prefi fiviu transmitted information in simple, direct ways, has the potential to empower the whole language perspective, not merely envelop it (Mc Caslin, 1989) A Vygotskian perspective of emergent interaction, self-directive inner speech, and adaptive learning does much to frame research questions on motivated literacy that inform practice and enhance theory developmentand thereby enables self-criticism and the accumulation and interpretation of evidence. .; MOTIVATED LITERACY AS SOCIALLY-SITUATED LEARMVG In closing, the construct of motivated literacy is a socially situated um. It is in dynamic flux Moth atee literacy exists in the emergent interaction of the imeipersonal culture that, as we have seen, consists of multiple, potentially competing, social/ instructional ens ironments, with the intr4personal, natural developmental processes of the individual What is motivateo literacy in .989, by definition, will not be so defined in 1999 Our task as educators, as I see it, is to provide learners the opportunities and the supportive stxiarinstruaional environments within whii.1 to become Japtive learners who transform even as they are transformed by motivated literacy. REFERENCES 'Me% C (19149 espr! 4. /ue.ement goals and student learning suwegies Paper preserned at the meeung of the American Educational Research Association, New Orleans. Bandura, A (1977) Social learning theory. Englewood Cliffs. NJ. Prentice Hall. Baumnnd D (1911) Current patterns of parental authority velopmerual Psychology Monographs, 4, (No. 1, Pan 2), Corno L (1989) What it means to bc literate about classrooms In D. Bloome (Ed ). Learning to use literacy in educational settings (pp. 29-52). New York: Ablex. Ehot. T S (1943). Four quartets. New York: Harcourt, Brace. & Co. Good T (1986) What is learned in zlementary schoo In T.Tomhnson & H Walberg ,Eds.), Acadenuc work and educational excellence Raising siactent productivity (pp. 873-114). Berkeley. CA. McCutchen. Halperin M (19"6) first grade tezchers goals and k.hildren's dcseloping perceptions of school. Journal of Educational Psychology. 68. 636-648. Heath S B (1982) Questioning at hc.i d at school A comparative Audy. In G. Spindler (Ed.), Dotng the ethnography of schooling. New York. Holt. Ftiriehart, & Winston. 66 ..caudr 49 -Motivated literacy Hirsch, J. D. (1989). A first dictionary of cultural literacy. What our children need to know New York. Houghton Mifflin. Jacobs, 1. E., & Eccles, 1. S. ,1985). Gender diffmences m math abtlity. The impact of media reports on parents. Educational Researcher. 14. 20-25. Lave, J. (1988). Cognition in practice. Mind. mathemancs. and culture New Yatic. Cambridge University Press. Luria, A. R. (1969). Speech development and the formation of mental Foetuses. In M Cole and 1. Maltzman (Eds.), A handbook of contemporary Soviet psychology (pp. 121-162) New York: Basic Books. McCaslin, M. (1989). The crnte.as of social comparison. Unpublished marnoctipt. .McCaslin, M. (1988). Whole language. Theory, instniction, and future implementation Elementary School Journal. 90, 223-229. McCaslin, M. (in preparation). Adaptive learning study. Midglay, C.. Feldlaufer, H., & Eccles, J. (1989). Change m teacher efficacy and student self. and taskrelated beliefs in mathematics during the transition to junior high school Journal of Educational Psychology, 81. 247-258. Nichols, J. (1984). Conceptions of ability. In R. Ames & C. Amcs (Eds.), Research on motivation In education: Student motivation (vol. I, pp. 39-73). Orlando, FL Academic Press. (hick, C. (1989). A cntic at large. T. S. Eliot at 101. New Yorker. Nov. 20, 119-114, 149-154. Pavlov, I. (1927). CondKonal reflexes. London: Oxford University Press. Posner, M. (1979). Cognition An introduction. Glenview, IL.: Scott, Foresman. Reurnan, D. (1989). How social comparison mediates the relation between ability-grouping practices and students achievement expectancies .n mathemaucs. Journal of Educational Psychology. 81. 178-189 Rogoff. B. (1989, Febmary). Social Interaction as apprenticeship to thinking Paper presented at the Conference on Socially Shared Cognition, Pittsburgh. Rohrkemper. M. (1984 The influence of teacher socially. ion style on students' social cognition and reported interpersonal classroom behavior. Elemeruary School Journal. 85, 245-275 Rohrkemper. M. (1985). Indirtdual differences in students' perceptions of mune classroom behavior Journal of Educational Psychology. 77. 29-44. Rohrkemper. M. (1986i. The functions of inner speech in elementary students' problem solving behavior American Educational Research Journal, 23. 303-313. Rohrkemper. M.. & Bershon, B. (1984 The quality of student task engagement Elementary school students reports of the tauses and effects of problem difficulty Elementary School Journal. 85. 127-147. Robrkemper. M.. & Como. L. k IY88). Success and failure co classroom tasks Adaptive leami.ig and classroom teaching. Elementary School Journal. R8, 299-312. Rohrkemper, M. McCaslin. k l989e, April). So, raho, non. intrapersonal awareness, and the developmru of self-directive inner spec .h in elementary school students Paper presented et the meeting of the American Educational Research Association, San Francisco. Aohrkemper, M McCaslin t1989bi. Development of self iegulated learning A Vygotskian perspective In B. Zimmerman & c. Schunk (Eds.), Self-regulated learning and academic achievement Theory research, and practice (pp. 143-167). New York: Springer-Verlag. Rohrkemper. M.. Slavin, R., & McCauley. K (1983. April) Investigating students' perceptions of cogn. nce straiegie.s as learning fouls. Paper presented at the meeting of the Ameman Educational Research Association, Monneal. anney, K Vv taking mothers and ...tit families American Psychologist 44 Scarr, S., Phil*, D.. & 1402-1409. Simon. H. (1969). The sciences of the artificial. Cambridge, MA. MIT Press Slobin, D. lWkl Soviet psytholinguishts In N O'Connor (Ed Present day Russian psychology (pp 109-151). London: Pergamon. Stipck, D. (1984 The development of achievement moivation In R Ames & C Ames (Eds Re search on motivation in education. Student motivation (vol 1, pp 145 174) Orlando, FL Academic Press. Vygotsky, L. (!962). Thought and language. Cambridge. MA. MIT Press. Vygotsky, L. 0978). niind in society, "he development of higher psychological processes Cambridge. MA: Harvard University Press. 6 '7 50 Literavinior; and :.:-75-" Weiner, B. (1985). An attributional theory of motivation and emotion. New Yrs& Springer-Verlag. Wertsch, J. (Ed.). (1985). Culture, communication, and cognition: Vygotskian persrclives. New X.00,4;:Cambridge University Press. Wertsch, J., & Stone, C. (1985). The concept of internalization in Vygotsky's account of the genest.lof-, higher mental ftmctions. In J. Wertsch (Ed.), Culture. emu* nication, and cognition: Vygo41:44_,Il;;; perspectives (pp. 162-179). New York: Cambridge University Press. Zinchenko, V. P (1985). Vygotsky's ideas about units for the analysis of mind. In Wertsc4s-(E). Culture, communication, and cognition. Vygotskian perspectives (pp. 94-118). New YOriC University Press. 68 ' . THE Et.FECT OF READER STANCE ON STUDENTS' PERSP'AL UNDERSTANDING OF LITERATURE Joyce E. Many Texas A & M University In her transactional theory of reader response, 12 isenblatt (19-78; 1985) describes the text as serving as a pattern for the reader, guiding the reader as he or she creates ;:a personal version of the literary work. This uniquely individual literary experience that each reader creates and the factors which influence that personal meaning making are important to researchers examining students' response to literature. Reader-response theory's emphasis on the role of the reader has resulted in a valuing if individual interpretations. Literature is seen as events to be lived through, offering opportunities for self-knowledge and for understanding others (Cooper, 1985). Consequently, new ways of describing the varying interpretations reached by different readers have evolved which acknowledge the validity of personal understanding (Cox & Many, 1989; Lehr, 1988). One factor which has been hypothesized to affect a reader's understanding of a work is the reader's stance, or focus of attention (Rosenblatt, 1978, 1985) An efferent stance indicates the reader's attention is focused on the information which is to be taken away from the reading and can result in an analysis of or study of the text (Cox & Many, 1989). When assuming the aesthetic stance, on the other hand, the reader's focus is upon the lived-through literary experience and the thoughts, feelings, images, and associations which are evoked. Although Rosenblatt and other reader response theorists, researchers, and teachers have focused on the aesthetic stance and personal understandings of literature as a point of discussion or as an underlying assumption in their works (Corcoran, 1987; Evans, 1987, Probst, 1988, Rosenblatt, 1938, 1978, 1985, 1986), little research has been conducted examining Rot.enblatt's concept of stance in response to literature or how stance is related to other factors in students responses. Only one study (Cox & Many, 1989) has investigated the relationship between a reader's stance and level of understanding of literary works. As part of a larger study, Cox and Many examined the free responses of 38 above-level fifth-grade students to four novels The purposes of their study were ia) to develop data-driven instruments to describe the stances taken in a response and tL le s. el of personal understanding reached, and (b) to examine the possible relath.,..ship betv...en stame, level of understanding, and story preference Although prior to Cox and Many's study Rosenblatt's use of stance had primarily been used to refer to the focus of attention during the actual leading event, the results significant indicated stance plays a role in affecting expressed responses as well .51 52 Lit:racy Theory and Researc positive relationship was found between the mean stance for all four novels and trigii mean level of understanding reached, (r= .36, p<.0001). The purpose of this study was to further explore the variations in stances takerk,i in expressed responses by investigating an older population. Although junior high:1 students' responses to literature have been previously investigatt,. terms of thesi objective or subjective formulation of the response (Applebee, 1978), the relationshi between reader expectation, comprehension, evaluation, and preference (Cullinan,, Harwood, & Galda, 1983), and the elements of the response (Golden, 1979; Purves,,,,1 1973, 1981, Rogers, 1988), no research has examined the stances taken in junior high4 school students' res?onses or the relationship of aesthetic and efferent stances t.& personal understanding of a literary work. Furthermore, Cox and Many's study examined only the relationship between the mean stance and mean level of understanding for all four novels read, leaving unan-., swered the question of whether the relationship between stance and level of under-standing is influenced by text Although much research has documented the effect of text on students response to literature (Purves, 1973, 1981), recent research intacates . response strategies can be consistent across texts (Beach, 1987). Therefore, this study examined the relationship of stance to level of understanding for three individual short stories to provide information as to whether that relationship is text specific. Specifically, the purposes of this study were (a) to describe the stances taken in eighth-grade subjects' responses to literature, (b) to analyze he relationship between the reader's stance in a response and the level of understanding reached in the reponse, and (:) to analyze whether the relationship between reader stance and level of understanding is consistent across individual texts. METHOD Subjects Subjects for the study were 5' eighth-grade students (26 males and 25 females) in two intas.t classrooms involved in a larger research project (Many, 1989). Two participating schools were chosen, one serving students from a low-socioeconomic level and the other sell irg students from a middle to upper socioeconomic level. One class w as randomly selected from the available eighth-grade Engiish classes at each school. Materials Three realistic short stories were chosen thruugh a piiot study which used six possible selections Reseirch indicates realistic stones are preferred reading in the uppei elementary and middle school grades (Golden, 1979, Purves & Beach, 1972) and stone, needed to be short enough to allow students to complete the reading in one sitting Therefore, cntena for the six initial story sdections was based on probable f nterest , appropriate readability, , and story length. Using an adapted version of Sword's (1985) "Criteria for Ev aluating Picture &or) Books," a panel of leading experts rated all Nix stories as above averugc on elements of plot unification, plot believability, u Reader Stance 53 Imaginative plot, main character portrayal, believability of main character, use of vivid imagery, and establishment of mood. The stories used in this study were those selected for thz larger research project based on overall preference by all students participating in the pilot study The ratings of the three selected stories (1high, 5low) by the eighth-grade students in the pilot were "The Dollar's Worth" (Werner, 1979)-2.5, "The Secret of the Aztec Idol" (Bonham, 1976)-3.1 (hereafter referred to as "The Aztec Idol"), "The Runaway" (Holman, 1976)-3.3. Procedure Pilot study. Two classes at the eighth-grade level from a university lab school participated in the pilot. The st xlents were drawn from the same population as the subjects in thc actual study itself. Students were asked to read and respond in writing to one of the six storie i. and then to rate the story on a 1-5 scale. Results from the pilot study were used to determine the stories to be used in the actual study and to refine data collection procedures. Data collection. For each of the short stories, subjects were asked to read the selection a.nd then to respond to the prompt, "Write anything you want about the story you just read." Data were collected in three separate episodes over a 9-week period. The order of the stories was randomized from subject to subject to account for possible influence of story sequence on response. Data analysts. Data were analyzed to determine the primary stance of the response as a whole and the level of understanding reached. The instruments used to code the responses are described below. The reader's stance when responding to the literary work was examined using Cox and Many's (1989) Instrument for Measuring Reader Stance on an Efferent to Aesthetk Contuitium. This instrument was based on Rosenblatt's description (1978, 1985, 1986) of the aesthetic and efferent poles of the reader stance continuum and Corcoran's desuiption of the types of mental activities involved in an aesthctic reading (Corcoran, 1987). Like earlier data-driven methods of classifying response to literature (Applebee, 1978, Galda, 1982, Purves & Rippere, 1968), this rating system emerged from the data analysis of subject responses. Responses rated on one end of the 5-point continui m indicate a primarily efferent stance, while scores on the opposing end indicate a more aesthetk. stance. Table 1 gives a brief descrption of each level on the instrument. The responses were also classified according to the level of personal understanding reached using An Instrument for Rating a Reader' s Level of Personal Understanding (Cox & Many, 1989). The Instrument evolved out of Cox and Many's research and is based on Applebee's (1978) levels of meaning and Ricouer's (1976) interpretation theory. The level of understanding rating incli-,atft the degree to which the response is tied to story events and the level of abstract generalization reached in the response Table 2 gives a brief description of each level of personal understanding. It Is important to note that the instruments allow for responses demonstrating fron, ,w to high levels of understanding at both the efferent and aesthetic poles of 1 54 - Literacy Theory and Research Table 1 Levels of Reader Stance on an Efferent to Aesthetic Continuwn ...EMU( Levels Description 1. Most Efferent Response 2, Primarily Efferent Response 3. Elements of Aesthetic and Efferent 4 Primarily Aesthetic Response 5 Most Aesthetic Response Analysis of elements according to outside structure (literary elements, realism, what was iearned) Retelling (concentration on relating the bory linn, narrating what the storj was about) Portions of both efferent analysis and aesthetic experience of work (equal emphasis on both, primary focu3 using a single stance indeterminable) Selection of story events or characters to elaborate preference, judgment, or description (I enjoyed it when . . . , I thought it was good/funny when . . .) Focus on the lived-through experience of the literary work (the world created while reading and the emotions or associations resulting from the experience). the continuum, as shown in the examples below. For instance, the following response would be scored at the mos efferent stance and would exemplify the highest level of understanding. Stance Rating ILevel of Understanding 4 It is a very unusual show It tells us that we can do anything we want to. It also tells grownups a thing or two One of the things it told grownups is. Before you step ahead make sure you've seen all the details. In contrast, the next exampk As also written from the most efferent stance but it would be scored at the lowest level of understanding. Stance Rating ILevel of Understanding I It was a pretty good story and I enjoyed it. The characters were designed pretty well but I didn't like the way the plot kept skipping time and not telling you what was happening. They picked a good setting for a plot like this one. At the aesthetic end of the continuum on the stance instrument, responses can also range in the level of understanding demonstrated. The next example shows a Table 2 Levels of Understanding Levels 1 2 1 Description Does not go beyond literal meaning of story Indicates some interpretation of story events Demonstrates understanding of specific story events through analogy to set or world 4 Reaches a generalized belief or understanding about life 72 55 Reader Stance response written from the most aesthetic stance which would be rated at the lowest level of understanding. Stance Rating 5Level of Understanding I I really enjoyed reading the book, it kept me curious throughcut. After I was finished I kept going back and thinking about the story. I could picture what was happening. Finally, as the next tesponse illustrates, aesthetic responses can also demonstrate the highest levels of personal understanding. Stance Rating 5Level of Understanding 4 I probably wouldn't have handled it as well as the family in the story did when she died, if my sister or daughter fell out of a tree and died when she was only eleven. The story really made you sit back and think about how unfair life can really be The coding of all the data was completed by the rese:Ircher. Independent raters, trained in the use of each !nstrument, coded a random sample of 20 percent of the data to check fot reliability. Interrater reliability was established using the Pearson Product Moment Correlation Coefficient. For the holistic rating of s*.mce the reliability was r = .79, and for the holistic rating of level of understanding, r = .81. RESULTS AND DISCUSSION Reader Stance Analyses of the stant.es subjects took in their responses revealed responses at all shown in Table 3, 38% of the total points on the efferent to aesthetic continuum responses were on the -Arent end of the ct....inuttm (ratings 1 and 2), with 9% of the total responses wntten from the most efferent stance (rating 1). The efferent response., comentrated for the most par a evaluating the literary elements or on the Responses at Each Point on the Stance Continuum Stories Stance 1 DW RUN Al 4 (8%) 6 3 (6%) (13%) 2 16 16 11 (32%) 9 (18%) (34%) 3 (21%) 9 (18%) 4 (19%) 6 (12%) 5 9 16 (32%) II9 119%) 13 (9%) 43 (29%) 27 (18%) 3 16 (6%) 24 (47%) (11%) 49 (33%) Note DW = "The Dollais Worth", AI= "The Aztec Idol", RUN= -The Runaway " 73 Total 56 Liveracy Theory and Research author's writing style. As shown in the example below and as contended by Rosenblatt (1982), many of the analyses of the literary works tended to be shallow responses. I didn't like the story at all. The story was too confusing. The story didn't tell anything much about the characters. The story didn't share the feelings of the characters The story was quite boring and I didn't like it. The story didn't explain anythingnever got to the point. When people write about stories they want to know a little about the characters and the story. Martha"The Aztec Idol" However, not all responses written from an efferent stance were superficial. Some students searchod the stories in an attempt to detemine the theme or what the author was trying to say In the following response to the story, "The Runaway," Victoria grapples with that very question. The Runaway was a story that I really didn't understand. I have my own conclu- sion about what it meaat b. t I don't tit the author's idea is the same as what I think. I think the author is trying to say (this is my opinion) that home is the best place to ;-le And that just because things may not go your way or your parents are pressuring you That your home is the best. Just because things lock good on the outside doesn't mean they're good on the inside. This girl Marcie, thought her friend was so lucky, and that her mother didn't always bug her, But once she went over and found out how it rzally was she knew that a family's true lovc is always best. So in this story it was probably made for someone who wants to run away. To try and warn them. Because someone else has always got family problems worse than yours. This may be what the author had in mind, but it is kind of confusing ur.less you really sit down and think about it for a while. I think someone who is experiencing problems would enjoy .sid understand better than I do. Rosenblatt (1978) has stressed that often readers fluctuate in their reading between an efferent and an aesthetic stance, such was also the case in 18% 4.,f the total responses in which no primary stance (rating 3) could be determined. In many cases, brevity made classification of a primary sti.nce difficult, while in others (as found in the example below) definite efferent and aesthetic elements were mingled in the response. This story was very good. The author (H .ma Werner) did a very good Job in giving the characters personality. Just by reading that story I hate Mr. Watts too. The only thing this story was lacking was a good description of this place. I pictured it out in the country where there isn't much traffic and there aren't many stores. But I don't know For all I know it could be in New York City. (Except for when it said Mr Watts went putting along the street at 10 mph.) Other than that, the story was excellent. Jim"The Dollar's Worth" In his response, Jim relates an effeient analysis of the character portrayal and description of !'ie setting in the story, "Tht, Dollar's Worth." He substantiates ills evaluative statements by g:ving us a glimpse of his evocation of the work, the Images and feelings which emerged as he pictured the story experience in his mind. Jim's response serves as a reminder that the stance in the itported response may or may not he consistent with the stance taken during the actual reading event. Jim seems to be 57 Readl.r Stance writing an efferent analysis of the story, based on a very aesthetic reeding of the lit,rary work. Of the total responses, 44% fell at the aesthetic end of the continuum (ratings 4 and 5), with 33% exhibiting the most aesthetic stance (rating 5). Although all of the most aesthetic responses focused on relating the lived-through experience of the stories and the emotions, images, ideas, and associations which were called to mind during the reading and reliving of the story event, the responses themselves were as varied and unique as the individual children who wrote them. Tne responses below illustrate some cf the elements which were found in the primarily aesthetic respnnses: imaging and picturing, relating associations and feelings evoked; and exteneing, hypothesizing, and retrospecting. Imaging and picturing. Aesthetic responses often include a de.cription of a visual image the revder pictured as reading or an account of how the reader imagines it ntify w'th characters a great would feel to be one of the characters. Some subjects deal, as found in the responses of one subject, Amy. In her responses t, II three stones, Amy chose to create her own literary works using the voice of t'.'e character in the stoy. The poem below was written in response to "The Dollar' Worth," a story of a young girl who encounters prejudice while working at a gas .Aation The penny-p:nching old man in the beat-up old car, Drives ;n, for his dollar's worth of gas. His is cold, and mean, And his mouth is set in it's downturned frown In disgust, I watch as his wrinkled old finger points, His eyes watching me, every step I take. I grew to hate him, And dread the days he'd come But I learned that he had no-one. Almost noone. And he insulted me, simply in a gesture of time gone by, and his forgotten past. He will be back tomorrow, In I,. beat-up old car, Watching me like a hawk. Pointing, frowning, Shooting Insults, And I will give him, his dollars worth of gas. Amy's response illustrates the possibilities for imaginative reading and responding when students foLus on living through the story experience. Her identification with the story charactei , Tnsh, resulted in her assuming Trish's role and feeling Trish's ieelings Other students identified with story characters also, but concentrated more on situations from their own lives which were similar to those experienced by the characters. Relating associations and feelings evoked. "The Runaway" tells the story of a young girl who feels smothered by her punts' love and runs away to a friend's house, only to find that grass is not always greener on the other side Forty-seven percent of the eighth-grade responses to this story were written from the most aesthetic 75 58 Literacy Theory and Research stance (as compared to 19% of the responses to "The Aztec Idol" and 32% of the responses to "The Dollar's Worth"). The subject matter of this story seemed to encourage what Cochran-Smith (1984) calls life-to-text connections, in that many of the subjects related their own problems, or their fantasies about or experiences with running away, as illustrated in the responses helow. I would like to be able to have a family who cares about me, and we could never be split up, &rid could talk out our family problems, and that I could always go somewhere called home. I feel sorry for Marcie that her and her mother couldn't work things out. Sometimes I think about ways I could runaway and how things would be for me like Marcie. David It was kinela of instering. I mean it was also stupid. I would have killed anybody if they bit me on the leg for no reason. And that chick is going to cry when she runs away! When I ran I was happy. Pluse going over to a friends house. NO WAY! Man ain't gonna help yo out none. Pluse She den't know what trouble is. I been living away from my house for 31/2 year. For running, ...earm, doing irttis, tresspassing, and -andalizing. She had no mason to go away from home. It was not realistic in some ways. Well there you have ii. The End. Stan For these students. the story hit close to home and they were able to relate definite associations which came to mind during reading. For David, the story offered a picture of a caring family life which he wished he could share. He saw the story as realistic, although difeerent from his own experience. Stan, while giving evidence of an aes- thetic experience by his willingness to relate the ideas and feelings which came to mind as he read, eventually rejected the teality of the btr y world because A uid not conform to his own experience. Hypothesizing, extending, and retrospecting. Some students, responding from the primarily aesthetic stance, extended the story line by hypothesizing background in f. nation or by continuing the stories to their own end. Others mused over what they were thinking as they read the story. In the following response to "The Dollar's Worth," Victoria explains a character's behavior by imagining what previous life experiences might have resulted in him being the way he was. This was a good story. It showed two different sides of reople. It showed .iow the man really felt and how he had to act. Some people don't underst..nd these two sides You don't really have to know someone, just look at how they act. If someone acts hyper or shows off, they may not be doing it just because they want to, but probably because they want someone's attention. This man in the story, Mr. Watts, probably wasn't really a mean man, he just wanted someone to talk to or he was just defending himself so people wouldn't feel sorry for him Maybe he didn't want anyone to know he was poor and lived with his sister because he might have felt ashamed. He also might not really have been prejudice against girls, but maybe something bad happened, that a girl to him. He might have even wanted to marry some girl and tii,., wouldn't. But you can't really be angry with people like that, because if you are a person you should know how people act. Victoria Victoria has taken advantage of what iser (1980) calls the gaps in the text, filling 76 59 Reader Stance in the unknown histoly of a character to rationalize his behavior In the next response to "The Aztec Idol," a story about two young boys who are conned by an old fisherman when they agree to buy a "secret" from him, Herbert goes beyond extending the story. Imagining himself in the boys' shoes, he reconstructs the personality and motivations of the old fisherman, remaking the character into one that he hi,hself would have believed in and liked. If the man was as much as a jerk as the man in this story ..nd if the secrets were so dumb I wouldn't have bought one. That man was so conceited and concerned about himself that I hated him. You could tell from the beginning by the way he talked about welfare, it was the way he said it and what he said about it that made you know he was a jerk. A friendly old guy who enjoyed kids might have had a different approach for selling a secret. He would have been nicer and lore interesting. Like .41 old man wilco loves to see kids steal peaches off his tree oecause he likes seeing the kids so joyful ard right when the kids got just one peach he runs out of his house -md shouts, "You rotten litde brats! I'll get you for this!" Even though he really doesn't mean it, he just likes to give the kids a good t ..ne .nd make them feel important. Like Amy who responded through poetry, Herbert !..icluded a creation of a new literary work, a vignette, as he constructed meaning from ' literary experience As well as creating original narrative forms, some students responding fro1n the aesthetic stance referred to stories they had previously read or viewed. I think one part of the story goes aiong with the movie with Julie Andrews in it. I also think Marcie should Rake] advantage of her parents "loving care " And get rid of that friend Hilda. I think that for an occirpation she ought to go into poetry If she would jur.t try and except ner parents "Losing Care" her life would be alot easier Arthur Arthur s reference to 4 Julie Andrews' movie gives an indication that he has made an intertextual connection as he r id, but because he doesn't elabon.te we P-e left wondenng how or if the conoct.tio. -fluenced the meaning he made from his evoca- tion of "The Runaway. ' Stance and Level of Personal Understanding To determine the relationship between stance and level of understanding, separate analyses of variances (ANO 'As) were conducted for each story. FIr the purpose of thc ANOVAs the variable stance was treated as an independent varihble and converted to a three-level rating. ( I ) mostly or primarily etferent, (2) element!, pf both efferent and aesthetic, and (3) mostly or primarily aesthetic. Separate analyses of variance revealed stance to significantly affect the level of understandinb reached for all three stones. Table 4 provides a summary of the ANOVA statistics and Tah!e 5 lists the means and standard errors for each story and the post-hoc analyses For all three stones, subjects fth.,u.ing on the aesthetic stance were significantly more likely to interpret story events, to apply story events tu life, and to draw generaliiations about the world. The relationship between stance and level of unde-standing proved to be fairly consistent across texts. For all stones, subjects who focused on the lived-througli expencnce of the story had a significantly :-.:gher mean level of understanding than '77 Literacy Theory and Research Table 4 Sununary of ANOVA Results for Each Story Source df Stance Error 2 47 Stance Error 2 44 Stance Error 2 48 MS "The Dollar's Worth" ( 4.418 so) .698 "The Aztec Idol" (n= 47) 4.450 1.184 "The Runaway" (n=51) 7.301 .866 Note. Differences in et across stories due to subject absenteeism. *p<.05. **p<.01. mp<.001. subjects who responded with no single primary stance. The aesthetic responses were also higher in level of understanding than the efferent responses and these differences were significant for two of the three stor;-;s. These findings indicate that the relationship between stance and level of understanding is not text specific. CONCLUSIONS The results of this study are significant in that they provide inrormation as to the range and complexity of stances found in eighth-graders' responses and investigate the relationship between stance and level of understanding using junior high school Table 5 Means and Standard Errors for Level of Understmding by Stance for Each Story Stance Level 1 2 3 it 9 22 2 3 16 . 15 -,1 3 "The Dollar's Worth" 2.116 19 22 9 1 M 1.56, 2.68b "The Aztec Idol" 1.68, 1.78, 2.63b "The Runaway" 2.00, 2.11, 9 27 3.11 b Note. Means with different subscripts differ significantly at r.:.05. 78 Std. Error .196 .278 .178 .232 .363 .272 .240 .310 .179 Reader Stance 61 subjects. While the results of suches analyzing written responses are limited since a subject's reported response (and the identifiable stance and level of understanding therein) may not reflect the extent of his or her reading experience, the following conclusions are suggested. Although the largest percentage of the total responses were written from the most aesthetic stance 33%), in view of the fact that the aesthetic stance is the focus deemed appropriate for the reading of literature (Rosenblatt, 1985) this seems regretfully low The aesthetic stance, focusing on the evocation of the literary work, was associated with imaginative and creative responses where students found the literacy experience meaningful and relevant. If teachers intend literature to offer unique experiences through which students can live, find pleasure, and reach understandings about themselves and the world, the aesthetic stance needs to be supported and encouraged. When students are asked to take an efferent approach to literature, for example in learning about literary elements such as plot, character development, and so forth, they should examine these elements in light of an original aesthetic experience of the literary work. Rosenblatt (1978) has underscored the importance cf the reader involved . . keep his sense of . . . [his personal evocation] as vividly in analyzing literature to and fully in nund as possible" (p. 174). This is substantiated by the shallowness and analytical distance found in many of the responses written from the efferent stance in this study. . In contrast, some of the responses at stance rating 3, which mingled an efferent analysis of the work with reports of the richness of the lived-through experience or even the lack of such an experience, were much more sophisticated and meaningful. When examining the relationship between stance and level of personal understand- ing, responses wntten from the aesthetic stance were associated with significantly higher levels of understanding. In terms of the classroom these findings underscore the importance of fostering the aesthetic stance when students respond to literature When teachers use ping-pong questioning techniques, where students parrot back responses to questions listed in the teacher's manual, students may assume the only appropriate focus when reading literature is to analyze the selection and retain important information. Although teachers may use such methods in an attempt to extend literal and inferential comprehension and to develop analyticJ thinking skills, inviting studenb to fully relive the hterar) experience could lead them to greater heights of understanding. That the results were significant across story selections indicates that stance is a tat.tur affecting response to literature regardless of literary text While individual texts may vary in their potential for encouraging the aesthetic stance with certain age groups (for example, the story 'The Runaway" elicited a large percentage of aesthetic iesponses focusing on the eighth-graders' associations with their life experienres), the occurrence of higher levels of personal understanding in responses written from the aesthetic stanLe was Lunsistent across all three realistic short :tones Consequently regardless of the literary works comprising the curriculum, teachers who want to encourage readers to find personal meaning in -rature should consider aesthetic indi 'Ina! evocations teaching strategies which promote and strengthen Such strategies would ideally (a) invite opt.. -.sponses, (b) give students time to respond, (c) provide opportunity to talk, (d) encourage personal and intertextual con- 62 Literacy ineory aad Research nections, and (c) recognize and encourage the focus ol . *eu:ion on the lived-through experience of the literary evocation. This research provides empirical support for use of the aesthetic stance, which has long been encouraged in both theoretical and practical essays. Additional research is needed not only on the stance taken in children's responses, but also on the reader's stance during the actual reading event, using techniques such as protocid,analysis. The stance children take in their response to informational texts as opposed to literary works is another area which needs investigation as well. By understanding howchildren focus their attention when reading and responding, we. can aid them not ouly in the productive reading of all texts, but also in discovery of how to live, through, relate to, and learn from the limitless supply of worlds found in literature. REFERENCES Art bee, A N (1978) The child s concept of story. Ages two to seventeen. Chicago. University of Chicago Press. Beach, R (1987. December) Applying life to literature. Reader's use of awob:ograplucal experiences to interpret texts Paper presented at the meetmg of the National Reading Conference, St. Ponsburg, FL. Bonham. F (1976) Secret of the aztec idol In A. Diven (Ed.). The Scnbner anthology for young people (pp. 116-124). New York: Charles Scribner's Sons. Cochran-Smith. M. (1984). The naking of a reader. Norwood.b11: Ablex. Cooper. C R (Ed ) (!985) Researching response to literature and the tcarlung of literature. Notwood. NJ: Ablex. Corcoran, B (1987) Teachers creating readers. In B Corcoran & E. Evans (Eds.), Readers, texts and teachers (pp. 41-74). Upper Montclair. NJ: Boynton/Cook. Cox. C & Many. J E (1989. March) Reader stance towards a literary work. Applying the transacnonal theory to children's responses Papet presented at the meeting of the Amencan Educational Research Association. San Francisco. CA. Cullinan. B E , Hardwood. K T & Galda. L (1983). The reader and the story. Comprehension and response Journal of Research and Development in Education, 16(3), 29-38. Evans, E (1087) Readers -rcreating texts In B Corcoran & E. Evans (Eds.). Readers, texts, and teachers (pp. 22-40). Upper Montclair. NJ: Bo:nton/Cook. Galda L. (1982) kssuming the spectator stance An examination of the responses of three young readers. Research in the Teaching yEnglish. 16, 1-20. Golden 3 M (l979) A schema for analyzing response to literature apphed to the responses of fifth and eighth grade readers to reAlistic and fantasy short stones (Doctoral dissertation. ObD State university. on) Dissertation Abstracts huernational, 3,, 5996A. (University Microfilms No. 7908149) Holman, F (1976) The runaway In A I-ven (Ed ). The Scnbner anthology for young people (17-21). New York: Charles Scribner's Sons. Iser W (1980) Inieraction between text and reader In S R. Suletman & I. Crusman (Eds.). The reader ut the text Essays on audience and interpretation (pp. 106-119). Princeton. NJ. Pnnceton University Press. Lehr. S (1988) The child's developing sense of theme as a repons t ,deraturt. Reading Research Quarterly. 23. 337-357. Many I E (1989) Age level differences in children's use of an aesthetic atance when responding to literature Unpublished doctoral dissertation. Louisiana State University. Baton Rout,..., LA. Probst, R E (1988) Response and analysis Teachmg literature in Junior and senior high school. Portsmouth, NH: Boynton/Cook. Purves, A C (1973) Literature education tn ten countr44. An empirical study. Stockholm. Almqvist & 8-0 63 Reader Stance --fve thwart, Puives, A. C. (1981). Reading and literature. American achievement in internatii IL: National Council of Teachers of English. alum, reading Purves, A. C.. & Beach, R. (1972). Literature and the reader. Research in respoase los mrests, anti the teaching of literature. Urbana. IL National Council of Teachers of English Purves. A. C.. & Rippere. V. (1968). Ekments of writing about a literary work. Urbana. IL; National Council of Teach= of English. %emu, P. (1976). Interpretation theury Discourse and the surplus of meaning Forth Worth. Texas Christian University Press. Rogers. T. (1988. April). High school students thematic interpretations of complex short stories Paper presented at the meeting of the American Educational Research Association. New Orleans. L.A. Rosenblatt, L. M. (1938). Literature as exploration. New York: Ap-o-son Century. fonal theory of the literary work. Rosenblatt, L. M. (1978). The reader, the text, the poem. The Carbondale, IL Southern Illinois University Press. Rosenblatt. L. M. (1982). The literary transaction. Evo..ation and response. Theor) into Practice. 21 268-277. Rosenblatt. L. M. (1985). The transactional theory of the literary work In C R Cocper (Ed ), Researching response to literature and the teaching of literature (pp. 33-53). Norwood. NI: Ablex Rosenblatt. L. M. (1986). The aesthetic transvuon. Journal of Aesthetic Education. 20, 122-128 Sword. %. (1985). Criteria fur evaluating picture story books (CEPSB) In W T Fagen, J M Jensen. & :. R. Cooper (Eds.), Measures for research and evaluation in the English language arts Vol 2 (pp 225-227). Urbana, IL: National Cauncil of Teachers of English. Werner. H. (1979). The dollar a worth. In J. Shapiro (Ed.,. Triple action short stories (pp 26-35) New York. Scholastic. 81 JOINING THE DEBATE: RESEARCHERS AND READING EDUCATION CURRICULUM Patrick Shannon Penn State University Literazy has long been considered a precondition of active citizenship in America. A literate populace svc,: id examine and discuss the issues of the day to make informed decisions concerning thuir future and that of this democracy. From Thomas Jefferson's Bill for the More General Diffiis'an of Knowledge (1787) to Joseph Mayer Rice's Survey of the Public School System of the United States (1893) to the recent spate of reports on the state of American schools (e.g., A Nation at Risk, 1983), literacy instruction has figured prominently in the curricula for public schools and has been the object of more study than any other school subject (Weintraub, 1982). Despite this attention, many argue that typical literacy lessons remain inadequate for preparing literate citizens who are able to meet the rapidly accelerating social, political, and industrial demands on an individual's literacy. Although most agree that some change is necessary, , there is much less agreement concerning what those changes should be and just whose conception of literacy and instruction should be validated at school. Many educational historians and philosophers (e.g., Cremin, 1961, Feinberg & Soltis, 1985; Kliebard, 1986; Prakash & Waks, 1985) agree that four interest groups continue to influence the American curriculum. :.anslated for literacy education, these groups include cultural literacy supporters, whole language proponents, teacher and school effectiveness enthusiasts, and critical literacy advocates. Each grt. , bases its definitions and claims for its position on Olfferent sets of epistemological and pedagogical assumptions, and each offers a differ- ent future for schools and society. Knowledge of these differing assumptions and histories equips the individual to articulate his or her position more clearly, to see connections between and among apparently disparate events and statements, and to contnbute more constructively to the debate by mediating between and among groups (Mitchell & Green, 19 4). In this paper, I attempt to delineate the positions, histories, and potential futures of each group to shed light on the debate over literacy education ....ading researchers' knowledge in American schools and to report on a stud, same. Cui.aral literacy supporters (e.g., Bennett, Bloom, and Hirsch) share assumptions with generalists often called "rationalists" (Prakash & Waks, 1985) or "humanists" (Kliebard, 1986). Within this position, literacy instruction should be organized according to the idea that schools' primary purpose is to develop students' intellect and their understanding of the best vf the past through initiation into the world of high culture and academic disciplines. It is a conserving, not necessarily a conservative, approach 65 66 Literacy Theory and Research to the literacy curriculum. In its strongest version (e.g., Peters, 1965), this group begins from the premise that "children start off in the position of barbarians outside the gates of the citadel of learning [school]," (p. 271) and the job of schooling is to make them rational human beings through the study of the classic literature of each discipline to develop students' "mind" in terms of factual knowledge and cLciplined vision ("to think historically, not just know some history," p. 278). Accordingly, rules of phonics and grammar are to be studied, not only for the linguistic benefits, but also becaase they help train students' analytic skills. Literary works of art should be studied to improve students' moral character and to initiate them into the traditions of Western Civilization. This position was the curriculum in American elementary and secondary schools until the early 20th century. Early challenges o its relevance were met with the Report of the Committee of Fifteen on Elementar: Education (Harris, 1895). During the 1950s, the position (e.g., Bester's [1953] Educational Wasteland) was used to attack progressive eduA, in American schools, and recently the position has enjoyed much atteri,../n through the work of Adler (1984) and Honig (1988). Whole language proponents (e.g., Goodman, Harste, and Graves) share most assumptions with "self-actualizationists" (Prakash & Waks, 1985) or "developmentalists" (Kliebard, 1986) who have at least a 100-year history of challenging the traditional cultural literacy position. Beginning from a more optimistic view of human nature popularized by Rousseau (1762) in Emile, the group reverses the cultural literacy basic assumptionsociety corrupts childhood innocence, moving children away from their basic goodnessrather than civilizing them as the cultural literacy advocates insist Consequently, literacy curricula should be organized around the child 5 natural development and struggle for self-knowledge and authenticity. Both teachers' and students' subjectivities become the criteria agains, which the cuinculum is developed and the success of the program is judged. That is, lessons to develop self-expression ar.d communicative use of language move language from being an object of study in elementary schools to becoming a tool for students' learning about themselves and what interests them. Started by G Stanley Hall (1883) and Francis Parker (1883), furthered by William Kilpatrick's Project Method (1918) and the Bureau for Educational Expenments, and formalized by pr^v incial and national departments of education in other English speaking countries, the whole language position enjoys a widespread revival from the child centered days of progressive education. Teacher and school effectiveness enthusiasts (e.g., Anderson, Pearson, and Beck) share assumptions with groups generally labeled "technical- (Prakash & Waks, 1985) or "social efficiency educators (Kliebard, 1986). Following E. L. Thorndike's lead, 'his group maintains that the literac) curriculum should be organized through scientific experimentation concerning which elements of the reading process are most important, in what order the elements should be taught, which teaching methods yield the greatest amounts of learning in the least amount of time, and what materials are necessary to coordinate the..,c- scientifically induccd parts of literacy lessons. Based on pnr:iples r,ci-r,!ifc management from industry and research summarized by Huey (1908) and Gray (10 i 9), th:3 position has dominated American education since the fourth report of the Committee on the Economy of Time in Education (Horn, 1919). 67 Joining the Debate Rejecting cultural literacy supporters' concern for the tradition and whcle language proponents' interest in the subjective, this group bases its work on positivists' assumptions about reality and human nature. Of primary concern are the effects of expenmentation and the standardization of instructional practice as expressed in the concluding statement of Becoming a Nation of Readers (Anderson, Hiebert, Scott, & Wilkerson, 1985). "America will become a nation of readers when verified practices of the best teachers in the best schools can be introduced throughout the vuntry" (p. 120). Cntical literacy advocates (e.g., Berthoff, Shor, and (iroux) share assumptions generally with gloaps suggesting schooling for "social responsibility" (Prakash & Waks, 1985) and "social meliorism" (Kliebard, 1986). This group promotes school and literacy curricula which would hep students to ide:ify, understand, and oppose unjust relations and conditions which originate, promote, or maintain inequality among citizens. Relying on L. F. Ward's (1883) rejection of social Darwmism as an unscientific justification of social oppression and his assumption that schrvsling and literacy were the only ways to redistribute cultural capital equally among cmzens, this group suggested that literacy and language become social tools with which studenv cone to understand themselv estheir history and culture, to dcc.ne *heir relationship with the larger social, pc:luta:al, and economic structure, and to act on their new knowledge to work toward social justice. Championed early by the members of the Educational Frontier, they reject the gu.se of political "neutrality [of the other positions] with respect to the gAeat 1 sues, which agitate society because it is practically tantamount to giving support to the most powerful force., engaged in the contest" (Count ,, 1932, p. 263). As with the whale language position, critical literacy advocates propose he inci- dent .1 study of one's language unless that language becomes a social marker for discrimination kElsasser & Ir.ine, 1987). Undei those circumstances then the discourse should be stu.-lic directiv to show its valre and logic. The cri:1,a1 literacy advocates scek a literacy curriculum that will enable students to use literacy as a tool for understanding then owil histones and culture. their connections to the current social structure, and their abilities to act based on their new knowledge. By exam.r..ag the histones and basic assumptions cf the four groups negotiating for control over literacy curricula in America, reading researchers can better understand the context in which teachers and they work. To determine whether reading researchers possess such knowledge, I sent a 16-item questionnaire to a random sample ct. reading researchers. The result, were vite surprising. Very few reading researchers displayed a grasp of either the history or the theoretical links within the four positions on literacy educa.ion. r.;ETHODS Subjects One hugalred and twenty reading -:searchers (20% of the total) were randomly selected frog., the oaectory of the 1988 National Reading Conference Prcgram 7s the 68 Literacy Theory and Research subjects for this study. Their work had been selected by their peers as meriting presentation at what is generally considered the preeminent reading research conference in the United States. The work that they preserted during the 1988 conference included experimental and naturalistic designs, reviews of research literature, and reports of on-going instructional projects. Each selected researcher studies or is employed at a university or college in the United States. Materials A 16-item questionnaire was mailed to each of the progenitors (e.g., William Torrey Harris, G. Stanleyresearchers. The items included Hall, E. L. Thomdike, and Lester Frank Ward), noted publications (e.g., Educational Wasteland, "Mc Project Method," The Psychology and Pedagogy of Muzding, and Dare the Schools Build a New Social Order), important asociations (e.g., Committee of 15 on Elementary Education, the Bureau of Educar,,,nal Experiments, the Committee on the Economy of Time in Education, and the Educationel Frontier), and modem advocates (e.g., E. D. Hirsch, Jerome Harste, Richard C. Anderson, and Ira Shor). The examples for the progenitor, document, and association sections of importance assigned them by historians (e.g., were selected based upon the level Cremin, 1961, anct Kliebard, 1986) and philosophers (e.g., Prakash & Waks, 1985). I selected the modern advocates based on the popularity of recent publications. An earlier version of the questionnaire included Marlette Johnson and Caroline Pratt in the progenitor category and Diane Ravich and Maxine Greene in the modem advocates category; however, none of the 10 curriculum professors could identify John..on or Pratt. I was told that to include %omen in any of the categories would lead to criticism that I used obscure items for the questionnaire. To vo:c1 such criticism, but acknowledging that many women played important roles in each of the four curriculum histories, I limited mv items to the above names. Procedures and Data Ar .lysis Questionnaires 7 -re mailed to 120 repsling researchers. Directions specified were: In a brief statement please identify each of the following items and indicate theoretical connections between and among them, if any exist." Furthermore respondents were asked not to conduct any research on the itzms, but to "work from memory." Conceming scoring the completed cp...stionnair.z, a I was assigned if the respondent gave any reasonably specific identifier (e.g., Animal lntell;gence, connectionism, or early experimentalist for Thorndike). Leaving ui :tem blank, providing mcc-rect response, or offering a nonspecific answcr was scored a zero. Scores for the historical identification were tallied for each group (from 0 to 4) and for knowledge of all four groups (from 0 to 16) To score the theoretical connections, a score of 3 was assigned if all four items in a grcup were connected, 2 if three were connected, I if two were connected, and 0 if no connections were made. these are people) responses were not accepted Incorrect or overly simplistic (e g., lack of theoreticai foundation. Highest possible as valid connections because of their total score was 12. Theoretical links were scored according to the descriptions offered by tr-emin and Kliebard for each group and ny assessment of modern advocates. 85 69 Joining the Debate RESULTS Sixty percent of the subjects (71 reading researchers) responded to the questionnaire. No one completed all 16 items, although each item had at least two respondents who identified it correctly. Total ss.ores ranged from 3 to 13 with a mean of 5.3 (SD 1 68). Respondents identified items from the teacher and school effectiveness expenmentalist position (scores ranged from 2 to 4 with a mean of 3 34) more often than they identified any of the others: cultural literacy (ranged from 1 to 4 with a mean of 1.6), whole language (ranged from 0 to 4 with a mean of 1.3), and critical literacy (ranged from 0 to 4 with a mean of 1.3). 1 ae modem advocates were the most often Identified and -,ften the only ones identified for a particular group. The most common response patt ;rn (19 respondents) identified only the modem advocates and E. L. Thorndike. Total scores for the connet tions ranged from 0 to 7 with a mean of 2.4 (SD 1.4). The most common response pattern (43 respondents) linked Richard Anderson with E. L. Thornke as psychologists interested in the scientific study of reading and Jerome Harste and The Project Method suggesting that Harste. advocates projects and theme lessons. Only the effectiveness:experimentalist position was conneet-d completely (by twc respondents only). Scores for the effectiveness!experimentalist position ranged from 0 to 3 with a mean of 1.7. Based almost totally on the connection of Harste with The Project Method, the whole language position scores ranged from 0 to 2 with a mean of 1.2. Only a few connections were made in the cultural literacy ''h a mean of A) and only two connections were position (scores ranged from 0 to made in the cntical literacy position (scores ranged from 0 to 1 with a mean of 1) In short, few respondents were able to link the present with the past theoretically DISCUSSION Although there are many limitations in this brief study (e.g., identification of items as an indit.ation of knowledge of history, , NRC membership as representative of the reading research community in general, links among historical item, as a meae of theoretical knowledge), the results of the _lady suggest :hat mading researdiers do not possess well-developed undeistandings the histories al philosophies Ala underlie the four et,sitins in the current dcbate zoncerning literacy education. Reading researchers responses lie in stark contrast to those of 10 curriculum specialists from universities Jurrounding my instiion whcse total scares ranged from 10 to 16 with a mean of 12.4 (SD .8) for the historical items and from 7 to 12 with a mean of 8.8 (SD 1.4) for the theoretical links. Reading researchers' lopsided performaaL, with its emphasis on the effectiveness:expelimentalist position suggests a limited rule within the current debate over literacy education, one more along the lines of Guthnes 987) policy driven research than Mitchell and Green's (1986) central role of mediator. Of coirse, reading researchers' preference for the erectivenesstexperimentalist position is understandable since It has been the dominant one in public schools and univers.ties d inng most of this century. However, not having familiarity with the 70 Literacy Theory and Research histories of reading education leaves reading researchers trapped in a present in which they must reinvent the rationales and philosophies for various counter proposals concerning what should or should not be happening during reading lessons and at schools. And they are easily fooled. For example, effectiveness/experimentalist reading researchers who supported Bennett's What Works (1985) did not recognize that his use of science furthered his cultural literacy philosophy rather than their own position (see Glass, 1987) That is, these researchers supported a document and pwition that works against their express d position. As a counterexample, note former Secretary of Education Bennett's lack of enthusiasm for the effectiveness/experimentalist's Becoming a Nation of Readers. The lack of historical and theoretical knowledge of the four alternative positions imp:des reading researchers' abilities to recognize similarities among apparently dis- parate studies, events, and rhetoric. A case in point may be reading researchers' abstract criticism of cultu: literacy as an attempt to homogenize American cultnie and literacy tastes while e perpetuate a type of cultural literacy among the members of the reading research comnwrity. Although reading researchers often claim diversity withir their field of study, their foremost journals publish almost exclusively research from one paradigm of educational science (Shannon, 1989), their state of the art descriptions are based on the same type of rtztarch (e.g., Becoming a Nation of Readers), and even both positions in their great debate argue from that research base (see, e g ('hall and Carbo in many issues of the 1989 Phi Delta Kapp, n). There appears to he an unacknowledged literary canon for literacy edncation among reading researchers, one comprised almost entirely upon the Effectiveness/experimentalist position If the results of this study are representative, this canon may be unacknowledged simply because it is unrecognized by a community with historical amnesia. If reading researchers seek to play a central role the largely political debate concerning literacy education in America, they might star.: with analyses of the histories of reading educatior, and the philosophical links betwetn the past and the present whi:ii lies outside the dominant position in the reading researcher community. To fail to co so may relegate reading researchers to a practal irrelevance in the eyes of policymakers, teachers, and the public. REFERENCES Adler. M. (1984) The Paidea program New York. Macmillan Anderson, R Hiebert. E . Scott. / . & Wilkerson. I. (1985) Becoming a nation of readers. Washington, DC. National Ins:lune e.:4 Education. Bennett. W (1985) What works. Washington. DC. National Institute of Education Bestor. A (1953) Educational wasteland. Urbana. IL. University of Illinois Press. Counts G (1932) Darc progressive education be progressive? Progressive Edm-anon, 9. 257-263. Cremin. L. (1961) The transformation of schools. New York: Vintage Els.. set. N . & Irvine, P (1987) English and creole in 1 Shor (Ed.). Fretre for the classroom (pp. 71-87). Portsmouth. NH: Boynton/Cook F-inberg. W . & Solos. J. (1985). School and society New York. Teachers College Press. Glass. G (1987) Wh4t svoRks. Politics and research. Educational Researcher, /6. -10. Gray W (1919) Principles of nrthod in teaching reading as derived from scientific invastigation. In I Horn (Ed ). Repei of the committee on the economy of time in /earning (pp. 106-130). Bloorrrngon, II Public School Publishing. 71 Joining the Debate Guthne, J. (1987). Policy development in reading education. In 2 Bloome (Ed ). Literacy and schooling (pp. 267-281). Norwood, NJ: Ab lex. Hall, G. S. (1883). Contents of children' s minds. Boston: Heath. Harris, W. T. (1895). Report of the committee of fifteen on elementary education. Boston. New England Publishing. Honig, B. (1988). The California reading initiative. New Advocate. 1, 235-241. Horn, E. (Ed.). (1919). Report of the committee on the economy of time in education. Bloomington, IL. Public School Publishing. Huey, E. (1908). The psychology and pedagogy of reading. Reprinted in 1968. Cambridge, MA. MIT Press. J..fferson. T. (1787 [1897]). A bill "or the more general diffusion of knowledge In P. Ford (Ed.), The Writing of Thomas Jefferson (p. 189). New York: Putnam. Kilpatrick. W. (1918). The project method. Teachers College Record. 19. 319-335. Kliebard, H. (1986). The struggle for the school cumculum. New York. Routledge & Kegan Paul. Mitchell. B., & Green, J. (1986). Of searchers, solons, and soldiers. How do eduzational research. pohcy. and practice relate. In J. Niles & R. Lalik (Eds.), Solving problems in literacy (pp. 234-243). Rochester, NY: National Reading Conference. A Nation at Risk. (1983). Washington, DC. National Commission on Excellence in Education. Parker, F. (1883). Talks qn teaching. New York: Kellog. Peters. R S. (1965). Education as initiation. In R. Archanbault tEd Philosopluca:Analysu and Education (pp. 269-291). New York: Humanities. & Waks. L. (1985). Four conceptions of excellence. Teachers College Record. 86. 79-101. Prakash, Rice. J. (1893). A survey of the public school system in the Umted States. New York. Century. Rousseau. J. J. (1963). Emile. (B. Schmidt, Trans.). New York. Scnbner. (Ongnal work publ:shed in 1762). Shannon. P. (1989). Paradigmauc cliversity in the reading research community Journal of Reading Belem- ior, 21. 97-107. Ward, L. F. (1883). Dynamic sociology (Vol 2). New York: Appleton. Weintraub. S. (Ed ). 1982). Annual summary of airesuganons relating to reading. Newark, DE. International Reading Association. 68 THE NRC YEARBOOKS DATABASE R. Scott Baldwin John E. Readence University of Miami Louisiana State University Jeanne Shay Schumm John P. Konopak University of Miami Louisiana State University The 1981 Delegates Assembly of the International Reading Asso-iation passed a resolution recognizing the importance of understanding the history of our profession and calling for an increase in historical research. Since then there have been a few summary publications that were designed to chronicle or critique the history of reading research or instruction (e.g., Moore, Readeace, & Rickehnan, 1983; Singer & Kingston, 194; Sissiman & Kaestle, 1987; Venezky, 1984). However, there is still much we at) not know about our own professional roots. One observation about readink., research that has been gleaned from retrospection is its sporadic nature (Venezky, 984). A topic will be investigated strenuously and then usurped by a more timely subject such that over time publications on specific topics are clustered together. For instance, the changing face of reading research wec reflected in the pool of manuscripts submitted to the 1986 NRC Yearbook and illustrates this clustering pattern. There were approximatel3r 20 manuscripts on writing and 20 manuscripts involving qualitative analyses. However, there was only one manu- script each on doze and vocabulary and none on readability or schema theory. 7t appears that many of the "hot" topics of a few years ago, as wen as many traditional subjects of reading research, simply were not in the manuscript pool. Is this topical bunching a superficial bandwagon effect the result of sampling error, or does it reflect an evolution of research priorities and demographic changes in the NRC membership? We do not know. What we do know is that as a profession we need more historical perspective than we currently have. In our experience it is not unusual for aspiring professionals to be largely ignorant of the literature prior to 197C They tend to assume that current articles with summarized, homogenized reviews of the past are sufficient "Why should I read these article. by Arthur Gates? Anything of consequence he had to say is common knowledge now, so why go wading through all that boring 1920s Fiuser (unidentified doctoral student). We believe that a knowledge of the past is fundamental to meeting current and future societal needs. Given the Yearbook's provocative and trend-predict:re nature, it occurred to us that the NRC membership would be well served by a permanent database of the NRC Yearbooks. Our objective, therefore, was to make the past of our organization more assible by creating a database designed to answer the following types of questions with respect to the National Reading Conference: (a) What major topics have emerged '73 in mading researrti in the past 37 years? (b) What is the publication stability of each topic ovrr time, for example, the doze procedure? (c) How have research 2riorities changed e: er the decades? (d) When did naturalistic analyses first appear in the NRC Yearbooks': Who was the author of the first one? (e) What manuscripts on the topic of text structure, for example, have been published in all of the Yearbooks? METHOD Materials PC-File: Version 5.0 (Button, 1990) was the database management system selected for this project. This software program operates with IBM vompatible personal computers and is capable of performing complex searches with large databases. The sources of data were all manuscripts published in all 37 volumes of the NRC Yearbook (1952-1988) with the exception of 1952 and 1953, which we have been unable to obtain even through interlibrary loan. Excluded from the database were introductions to symposia and abstracts published in lieu of full articles. Procedure The basic framework for topic descriptors evolved from an examination of: (a) the categorization scheme used in the Summary of ' tstigations Relating to Reading (Weintraub, 1989), and (b) descriptors used to he-v. editors identify reviewers' areas of expertise for Reading Research Quarterly and Journal of Reading Behavior. These topic descriptors were subcategorized under the following four major categories: Sociology: The primary focus of the article dealt with cultural influences, censorship, power of the mass media, political implications of literacy, and other issues of power and authority or societal change. 2 Psychology: The primary focus of the article was in the traditional educational psychology mode of theoretical and/or empirical explorations of issues in cognition and learning. 1 Physiology: The primary focus of the article was on biological, genetic, or other physical aspects of literacy, (e g., visual impairments, brain mapping, eye movement studies, etc.). 4 Pedagogy- The primary focus of the article was on program descriptions, explanations of classroom procedures, or studies with an actual instructional component. 1 Because the Yearbooks contain over 1,300 manuscripts, no one, all inclusive set of topic descriptors could be agreed upon in advance without forcing manuscripts into potentially inappropriate categories. We also wan..al to use the authors' terminology of the day (e g., the use of retarded readers in early literature versus our current use of remedial readers). For these reasons we began with a common set of about 100 topics and then adopted new topics as the review process progressed. A worksheet was developed which included the following: (a) volume and pages, (b) space for up to five authors, (c) the category, (d) space for up to three topics, (e) 90 NRC Yearbooks Database space for up to three sets of subjects, (f) space for up to three kinds of materials, (g) space for two kinds of analyses, (h) the date, awl (i) the title The worksheet information comprised the data. Training for the analysis of NRC Yearbook manuscripts for the four evaluators involvea 10 Yearbooks which spanned the years 1954 to IS 38. This was done deliberately to acquaint the evaluators with the terminology used in early Yearbooks as well as the expected diversity and potential complexity of manuscripts in more recznt volumes. The mean perce age of ageement between pairs of evaluators was 86%, and differences in judgments were resolved in conference. Each evaluator was then randomly assigned to one of the first four Yearbooks. That assignment order w maintained as the remaining 36 Yearbooks were then numerically assigned. Thus, each evaluator analy-zd 10 Yearbooks. The following general rules were used in the evaluation of manuscripts: 1. Categoriesonly one category could be selected, and one had to be selected. 2. Topicsthe author defined the topic, as stated previously; if the author failed to specify a topic, one from our list was used or, if necessary, created. 3. Subjectsprimary school was defined as K-3, middle school as 4-8, and high school as 9-12; if there were more than three sets of subjects, those that best characterind the study were used; if the article was simply a disctL,inn or description, subje:ts were not listed, but rather, this ir.formation was placed under topics. 4. Materials---on!y materials most pertinent ".o the study were listed; however, this grouping still proved the most unwieldy because of the idiosyncratic natthe of experimental materials used. 5. Analysesstandard current statistical design, and paradigm terminology were used 1n lieu of any author terms. Finally, It, avoid trivializing the procedure, topic descriptors were a.l.igned only when the evaldator believed them to be important in characterizing the manuscript Therefore, some manuscripts were assignc.d time topic descriptors while others were assigned two or one. The initial analysis resulted in a database of 557 unique topic descriptors. Some of the more unusual wpics included, cohesive harmony analysis, communjsm, heart rate, hypnosis. "..ernel distance theory, racial stereotypes, and syzygy. We also discovour efforts to standardize the review process each of the reviewers ered that in s; .,.,yncracies which caused artificia'. lustering of topics in certain voldemonstratet. umes. To make the database more consistent, we coapsed semantically similar categories and eliminated most of thc topic descriptors that were used fewer than three times. Each volume was then reassigned to a pair ef =viewers who reexamined each article and agreed upon its classification. RESULTS AND DISCUSSION The database consists of 1369 records, CU, for each article The following record, Harry Singer's first Yearbook article, is typical: 91 Literacy ?Theory ihei Riic 99 50 45 40 30 25 20 15 10 A A 1954 1959 A 1964 A 1969 A 1974 A 1979 A 1984 A 1988 ANOVA/ANCOVA Ethnographic/Qua11tative/Doscriptive Figure 1 Percentage of articles per volume year reporting analysis of variance procedures and percentage reporting qualitative techniques. volume: date: author 1: category: topic I: to;:c 2: topic 3: analysis 1: title: 14, 41-56 Record #247 1965 Singer, H. Psych ubstrata-factor theory theoretical models intelligence theoretical Substrata-factor patterns accompanying development in power of reading, elementary through collage level The percentage of articles that fell into the major categories were as follows. 59% psychology, 33% pedagogy, 6% physiology, and 2% sociology. There were 220 different topical descriptors (e.g., comprehension, gain scot-1, story grammar, col- Mid iearbooki Database 30 28 26 24 22 20 18 16 14 12 10 8 6 4 It A 2 \---- I 0 1959 1964 1, I 1 A A 1954 A A 01 I 1969 1974 . A 1979 A 1984 A 1988 Adult Reading En/argent Literacy/Beginning Readinq Figure 2. Pe.-centage of articles per volume year devoted to adult reading and percentage devoted to emergent literacy or beginning reading. lege reading) and 31 different analysis descriptors (e.g.. ethnography, historical, MANOVA, review of literature). There are tens of thousands of possible search combinations, and a comprehensive analysis of the Yearbooks is beyond the scope of this article. However, we can illus- trate the search capabilities of the database. Figure 1 shows (a) the percentage of articl ."s in each Yearbook using ANOVA or ANCOVA techniques, and (b) the Pereentage using qualitative methods. We chose to use percentages rather than acival minthers of articles because the number of manuscripts in '14 Yearbooks varies from alóW of 14 articles in 1958 to a high of 79 articles in 1970, one of three louble vOlume years. The first ar%or to report analysis of variance techniqUes in the Yeorbook Was MOon..orentz and Coker (1978) were the first: authora to report qualitative ald (10' awe 1 clearly indicates that analysis of variance techniques domintitecl technic 78 Literacy Theory and Research the Yearbooks from the mid 1979s to the mid 1980s and that the decline in reported ANOVAs since then corresponds with the emergence of qualitative methods. Figure 2 contrasts the topic adult reading with the combined topics of emergent literacy and beginning reading. Here the e-ita seem to demonstrate the organization's waning interest in adult reading and its waxing interest in emergent literacy beginning in the 1970s and continuing through the 1980s. With PC-File and the current database, users can conduct simple topical searches such as those illustrated in Figure 2. It is also poss.Lle to request bibliographic information based on any combinatiJn of variables. For example, the user could ask for a printout of all NRC Yearbook rtzticles written by a given author on topics X or Z between years 1959 and 1974 where aaelysis procedures included ANOVA. Our purpose in developing this databaze was to provide the NRC membership with a means of making the wisdom of the past more accessible to literacy researchers and students in the present It is cur intent to continue refining the database and adding to it with successive Yearbooks. la the meantime the database is available at no cost to any NRC member who is willing to send a self-addressed, stamped envelope and a formatted, blank, double sided, nouble density diskette to: National Reading Conference II E. Hubbard, Suite 200 Chicago, IL 60611 REFERENCES Button,1. (1990) PC.File Version 5.0. Bellevue, WA: Buttonware. Lorentz I L , & Coker, H (1978) Observed pattems of teacher-pupil classroom behavior as predictors of student growth in reading In P D Pearson & I Hansen (Eds.), Reading. Disciplinedinquiry in process and practice (pp 16-19). Clemson, SC. National Reading Conference. McDonald. A S (1Q57) A college reading program and academic performance. In 0. S. Causcy (Ed.), Techniques and procedures in college and adult reading programs (pp. 44-52). Fort Worth, TX. Texas Christian University Press. Moore, D W , Readence. I E . & Rickelman. R. 1 (1983). An historical exploration of content area reading instruction. Reading Research Quarterly. 18. 419-438. Singer, H . & Kingston. A 1 (1984) From the Southwest Reading Conference to the National Reading Conference A brief history from 1952-1984. In 1 A. Niles & L. A Harris (Eds.), Changing perspecthes on research in reading languoge ,L7rocessing and instruction (pp 1-4). Rochester, NY. National Reading Conference. Stedman, L C & Kaestle. C F (1987) Literacy and reading performance in the United States. from 1880 to the present. Reading Research Quarterly, 22, 8-46. Venezky, R L (1984) The history of reading research In P. D. Pearson (Ed.), Handbook of reading research (pp 3-38) New York. Longman. Weintraub, S (Ed ) (1989) Summary of investigations relating to reading. Newark. DE. International Reading A.Isociation 94 THE CURRICULAR EXPERIENCES OF AT-R1SK FIRST GRADERS IN PROGRAMS DESIGNED TO PROMOTE SUCCESS Melinda Lindsey Boise State University As defined by Slavin (1989), the term "at risk" refers to :Indents whose intelligence is within normal limits but who are failing to achieve the basic skills necessary for success in sthool and in life. The practical definition includes low-achieving students who are eligible for compensatory or special education services. Such students may be served in one of three major types of programs: compensatory or other remedhl programs such as Chapter 1, special education programs, and general education programs. Slavin- Madden, and Karweit (1989) have identified several themes common programs for at-risk students. First, the quality of prognIms that are among implemented matter more than the setting in which remedial or spcial adtication services are provided. Second, pullout programs are often poorly coordinated with the general education programs. Third, effective classroom and pullout programs for atrisk students accommodate indivilual needs while maximizing direct instruction. Fourth, teac5ing behaviors associated with outstanding achievemt gains for students pulled out for special instrudon tend to be the same as behaviors that are effective with all students. These behaviors invohe active instruction in which the teacher transmits the content of lessons, rather than relyinE nn worksheets, books, supplemental materials, and so forth (Crawford, 1989; Larrivee, 1989). The issue of program coordination merits elaboration. Consistency between general education and special program is often referred to as congruence. Walp and 'nstructional, and almsley (1989) identify three types of congruence: procede,.. philosophical. Instructional incongruence is frequently observed between general education and special programs. Allington and Johnston (1989) report that conflicts arise in several areas. Strategies used to control text difficulty may differ; natural language may govein one program and orthographic regularity the other The rssumed hierarchy of learnrg may conflict when comprehension of whole text is emphasized in one setting and decoding is emphazized in the other. Instructional strategies may vary, as when learner <tweeted instruction dominates one setting and teacher-direck. dominatc the other. These iifferences often stem from philosophical incongruence, reflecting differences in beliefs about the uaturc. of reading, reading disabili-y, and the roles of teachers and students in inst-uction. Few studies have criticqhy examined specific progams as implemented with atrisk, low achieving students. The purpose of this paper is to examine tilt. ..urricular 79 85 80 Literacy Theory and Reseatch experiences of the low achievers. Of interest is the quality of the programs they participated in and the congruence between their classroom and pullout programs. METHOD Subjects The study was conducted in a lower-middle class community located in the pacific Northwest during the spring of 1987. The target classroom was a first/second'grade: combination taught by Mrs. Starr (a pseudonym). This teacher was highly regarded. in her district as an effective teacher and an instructional leader. She was entlshisik about the program she implemented and committed to making it work. In Addition, she was particularly interested in its application with . ..deving children. She requested that at-risk students be assigned to her classroom and insisted on reducing the time spent by eligible students in pullout programs. For this investigation three first graders were nominated as low achievershy Mrs. Starr. John spent the entire instructional day in Mrs. Starr's classroom. Ginny was pulled ma of the classroom 15 minutes daily to receive remedial help in reading comprehension in the Chapter I program. Michael was pulled out of the classroom 25 minutes daily to receive help in reading comprehension and word identification strategies from the special education teacher in a resource room setting. Procedures The reading programs :mplemented for John, Ginny, and Michael were m.estigated primarily through classroom observation. All observations were made by the tesearcher, who assumed the role of unobtrusive nonparticipant. Informal conversations with teachers, summarized in journal entries, provided additional insight into students' experiences_ Children's written work was also examined. Notations thout the nature of the task and the quality of students' responses were made. Each of the three low achievers was observed on a different day. During the day's observation the researcher observed all instructional events that occurred during designated reading periods, in both Mrs. Starr's classroom and pullout programs. A protocol was generated on the basis of continuous observation of the actions of the student. Data Analysis The observation notes were reviewed after the completion of the stady. Fidelity of program implementation was described by comparing recommendations from prop= guides with notations relating to instructional events; these comparisons were triangulated with informal conversations with teachers. Student responses were described by comparing notations of students' actions with expected student responses, as outlined in program guides. These comparisons were triangulated with notations relating to examinations of student work. Time notations for teacher-student interactions were classified as lasting 15 seconds or less, 16 to 30 seconds, 31 seconds to 1 minute, 2 to 3 minutes, and continuous (sustained interactions lasting more than 3 minutes). , 06 ';S 81 Curricular Everiences Materials The Success in Reading and Writing (Success) program. The Success progam was the major basis of instruction in Mrs. Starr's classroom. Success is an integrated reading and language arts instruction program designed for =dents in kindergarten through sixth grades. Key components include use Of students' own language and everyday reading materials, no ability grouping, and daily composition and 'recreational reading. The instructional program in first grade (Adams, 1978) consists of five modules, each intended to last 30 minutes. Three are of interest in this investigation: *hat fellows is a summary of their intended purposes and expected student responses, along with a description of materials observed in use. The purposc of the "Phonics/Spelling" module is to teach students the "strategies necessary to decode with compreLension any word in the language they wish to read" (Adams, 1978, p. 21). The teacher begins by announcing a focus for the lesson (e.g., "words with sk, two syllables"). Students are expected to individually volunteer words and sentences containing the focal elements (e.g., "basket"; "Tammy went skating"), answer teacher questions about word identification or vocabulary, and orally read the resulting chart in unison. Students independently copy the chart and/ or write their own sentences containing the focal element. At this time the teacher circulates and asks individuals to orally read their texts and discusses word identification elements. Michael was pulled out of this component for special services; charts produced during observations of John and Ginny illustrate the typical text produced by the group: "Singing is fun," said The movie had a happy ending. Cindi Lauper is a singer. is mending her sock. The movie just ended. I am ready. My uncle is funny. is funny. You are running. People are digging under the dirt. "Clowns are running," said K______ On Slug lay I go to my friend's house and give his uncle underdogs on the swing. The purpose of the "Recreational Reading" module is "to establish student reading for enjoyment, as well as for information" (Adams, !.978, p. 47). Students are expected to select and silently read fiction and nonfiction library-type books, without interruption or restriction. During a representative observation of Ginny, students selected class books, comic books (e.g., "Yosemite Sam"), nasal readers, magazines (e.g., National Geographic), informational books (e.g., Life in Ponds and Streams), and story books (e.g., Stone Soup). The program also recommends that teachers schedule approximately six conferences per day with individual students, cycling through everyone in the class at least once a week. The focus of the conferences at the first-grade level is initially on wad identification strategies; later the conferences are to involve "sharing" of the books that the student and teacher are each reading. "Patterning," a module designed to last the first half of the school year, is intended primarily to emphasize identical letter combination patterns in various posi- t - 82 dons in words. The strategy for obtaining instructional material is the same as- in "Phonies/Spelling," with the exception that students generate only words, not-sentences. At the time of the study, this module bad been coinpleted and was replaced with an alternative reading activity. The time siot is intended by the program author to be used to extend the recreational reading module anotler 30 mintites. Pullout programs. Instructional materials and methods varied from thildjo child:, Ginny was expected to read words and sentences orally; think ..wOrksheet answers;, and answer the teacher's questions. Four Comprehension wolksheets (fOcusingAn:the skills of identifying rhyming words, locating- detailsi.making-inferenCeS;and:linsuat discriminadon of phrases) from the district's adopted basal series were ised:.-:MiChael was expected to orally blend consonant+ /a/ with final consonants, sound out hOnèti cally regular cvc /a/ words, pronounce sight words, and read passigei contaking words with Id/. Materials included worksheets and possages taken from phonetically controlled supplemental programs and teacher-made flashcards. RESULTS Fidelity of Implementation Mrs. Starr was faithful in her implementation of the "Phonics/Spelling" module. The two remaining components were substantially altered. Recreational Reading module. Mrs. Starr modified the module to eliminate recommended conferences because she believed that children's silent reading should not be interrupted by any "distraction," including interactions with the teacher. The single academic interaction observed during this module was initiated by the student. Patterning module. This component was not replaced by extending "Recreational Reading" another 30 minutes, as recommended. Instead, Mrs. Starr established small, homogeneous reading groups and placed them in the district adopted basal program for two reasons. First, the district had mandated that the Success teachers administer the basal unit tests as a measure of program effectiveness, and Mrs. Starr felt that students needed experience with basal instructional tasks to perform successfully on these tests. Second, she wanted stu:.nts to be familiar with basal instruction in case students were placed in classrooms using basal approaches rather than Success the following year. Groups worked under her direction in typical basal procedures. John, for example, worked with one other student; they took turns reading word lists and passages orally and answering literal and inferential questions. Because only 30 minutes was available for basal instruction, students did rot participate in teacher-directed groups on a daily basis. When not in groups, students were expected to read in a basal reader independently or to complete worksheet assignments. The only difference from traditional basal procedures was that suidents were free to select any of the basal readers available in the classroom, which included several grade levels from several different programs. Michael, for example, completed a Weeldy Reader written activity and a short vowel ditto, then read out of his own level of the district adopted basal reader. Ginny, who 28 Curricular Erperiences Was pulled out of this activity for 15 minutes, returned from Chapter I, selected readers from two different programs, and read them silently for the remainder of the period. Student Responses The three Iow performers generally tried to do what was expected of them, particularly in the basal replacement for "Patterning" and in their pullout programs. Panerns of deviations from expectations were observed in two major Success modules', however. Phonics/Spelling module. Deviations by the two students who participated in this module,. John and Ginny, could be found in two of the expected behaviors. Students were expected to volunteer examples of the focal element, though the level of expectation was not high. Mrs. Starr encouraged and praised contributions but did not hold individuals accountable if they chose not to volunteer. Ginny did not agepapt to contribute; John raised his hand to volunteer a word only once, in a tertative manner, but he quickly lowered his hand before the teacher noticed that it had been raised. During the independent seatwork task, Success students had the option of generating their own sentences rather than copying the chart. The level of expectation for choosing to generate novel sentences was not high; Mrs. Starr did not respond negatively to students who chose to copy. Neither Ginny nor John was observed to generate novel text. Recreational Reading module. Students were expected to silently read text. Only Ginny engaged in actual reading of text; the other two turned pages rapidly and appeared to look at pictures rather than process text. John and Michael had difficulty sustaining even this limited interaction with books over the entire period. Both were frequently engaged in other behaviors such as talking to other children and walkiag around the room. In one te'2,ing incident, John was observed to turn pages while his head was turned away from the book. Teacher-Student Interactions Success program. During the two Success modules, individual contacts that were for academic purposes tended to be both infrequent and brief in duration. John had three interactions. Of these, two involved reading sentences orally during "Phonics/Spelling"; one interaction lasted 15 seconds or less, the other lasted 1 minute. The teacher responded by announcing the next task in one instance and praising him in the other. A final contact, lasting 15 seconds or less, was initiated by John during "Recreational Reading." He showed the teacher a word containing a suffix in his library book; corrective feedback was provided. Ginny's three interactions occurred during "Phonics/Spelling." Each lasted 15 seconds or less and involved reading a sentence (rally. The teacher responded to each with praise. Michael was pulled out of "Phonics/Spelling" tor special services. Neither he nor Ginny had any academic interactions with the teacher during "Recreational Reading." ee sa LiteracyTheory and Beseiffir Basal group. John's experience in the teacher-directed small group activity lasteili17 minutes anti_ consisted of continuous interactions among the teacher and,the;tWp, students. There were high rates of successful student responding with iminediate: feedback. During the "free reading" activity of the basal period, Ginny sought and r*Iii teacher assistance in pronouncing a word on two occasions. Each interaction las 15 seconds or less. Michael had three academic interactions with the teacher during:the-1.Tc, ing" activity, all initiated by Mrs. Starr, .and all with the intent OtinStradingitather, than monitoring. In the briefest interaction, lasting 1 minute, the teachet..Xeyie* short i and short e vowel sounds and provided guided practice. The Osia. logo, ones (2-3 minutes) involved identifying words in his self-chosen basal'reader ing events in the story; in these instances the teacher worked simultanconslY witlikanother student who was reading the same book. She provided both input and coffee,. rive feedback. Pullout programs. Both Ginny and Michael worked with an adult on an individualbasis, Ginny for 10 minutes and Michael for 24 minutes. This time was observed to, involve intensive, continuous interactions between teacher and student, with- high, levels of student responding paired with teacher feedback or explanation. DISCUSSION Quality of the Classroom Curriculum Portions of the core classroom program are judged to be of poor quality in both student responses and teacher-student interactions. It is likely that this negatively affected the achievement of the at-risk students. Student responses. The student response that is most problematic in Success is generating examples of the focal element. The rationale for studert-generated text is that vocabulary derived from students' own oral language is more %leaningfirl. However, it is hard to argue that the resulting text ("My uncle is funny," etc.) is qualitatively superior to current basal reader passages. The sentences are not connected in meaning; as a result, they violate tacit expectations for text structure. Further, the task of generating examples of the focal element is both contrived and difficult. It is substantially more difficult thin either decoding the same words generated by another author or generating a message oc the individual's own choosing, without constraint. A second student response that may have negatively affected student achievement is copying. The common criticism of basal programs is their excessive and inappropriate rerance on repetitive, meaningless worksheet tasks. Yet copying requires even less cognitive effort than completing matching or fill-in-the-blank exercises on woricshcets. Copying neither guarantees that students think about the decoding and encoding elements of the text, nor entails purposeful, student-initiated communication with others. A third problematic response is siient reading. One characteristic of the differential instruction typically provided to low performers in basal programs is the emphasis 100 . funicular Experiences on oral reading at the expense of silent reading (Allington, 1983; Hiebert, 1983). Success appears to avoid this in allocating 30 minutes of daily silent reading. Further, Mrs. Starr went to considerable lengths to provide a variety of appealing materials, including an abundance of excellent children's literature. Yet two of the three low achievers had significant difficulty actually-processing text. One factor contributing to the low text processing is the practice of allowing students to choose whatever they wished to read. Although this may positiVely, affect student attitudes toward reading, it may also result in a poor mat& 'lietWiecri,the, demands of the text and the very limited decoding, skills oftheae beginnhig 'read*, A more important factor was Mrs. Starr's decision to eliminate indl.,yidtiateOnfer**S. Conferences were hr; sole opportunity to systematically guide stude* and comprehending connected text written by mature authors. Eliminating them reduced her access to information that would enable her to judge the match between student and text and to facilitate text processing. This is especially alarming because two of the three children received special services because oftheir problems in comprehension. "Recreational Reading" is an example of individualized instruction, in the sense that the use of different materials is thought to permit each student ro progress at a pace suited to his or her abilities and interests. This instructional approach, as actually implemented, has been criticized because the responsibility for teaching is shifted from the teacher to the student, and the responsibility for delivering content is shifted from the teacher to the instructional matelials (Good & Brophy, 1987). At-risk students, particularly at the first-grade level, may not be able to independently assume this responsibility. In sum, generating examples of the focal element and copying are qualitatively poor tasks. They fail to provide direct opportunities to engage in meaningful reading and writing. Silent reading is not an inherently poor task, but without teacher involvement, actual student responses may result in marginal experiences in literacy development. Teacher-student interactions. A major feature of Success is that teachers arc able to work with students on a one-to-one basis, following whole group instruction. The purpose is to enat. e the teacher to address individual needs through review and instruction. Interactions are intended to be more on the order of "mini-conferences" than the typical monitoring activities of teachers using basal programs. Yet, as implemented, teacher-student interactions that were for academic purposes were infrequent, brief, and entailed only incidental responses to a particular task at hand. Sustained interactions (2-3 minutes or longer) occurred in Mrs. Starr's classroom only in the context of the basal replacement, not in Success. The basal interactions contrast with Success both in length and in the teacher's interaction to demonstrate and to explain, es well as to monitor and provide feedback. It is unlikely that the individual student ...ontacts in Success modules were adequate for systematically tracking and accommodating the special needs of at-risk students. This suggests that a key area for continued investigation is the quantity and quality of individual teacherstudent interactions in all programs. These responses and interactions underscore two related points. First, as we inves... 86 tigate beginning reading instauction, we must critically evaluate the nature of student responsesboth expected and, especially, actual. Not ell that passes under the name; of a particular approach to instruction is necessarily et value. Second, we must look beyond the-mere presence of particular types of instructional materials. We mustexamine whether and how teachers assist students in using the materials in appropriate, meaningful ways. Incompatibility of Instruction The experiences of Ginny and Michael in their pullout settings appear tp be° qualitatively different from their Success classroom experiences. Text difficulty, was controlled by orthogiaphic features rather than by meaning; instruction was teacherdirected rather than learner-controlled; materials were teacher-selected rather than student-selected or student-generated; expected student responses differed; studentteacher interactions were sustained longer. Two points can be made about this incongruence. First, the core curricula of the pullout programs were not qualitatively inferior on every point of comparison. For example, although the instructional materials did not provide the same quality, of° narrative text that library books might prox ide, teacher guidance resulted in greater depth of processing than occurre in Success. The highly interactive instructien that is characteristic of effective practices for at-risk children seems a more critical factor than availability of instructional materials alone. Before pullout programs can be made to be consistent with classroom programs, it is necessary first to ensure that the classroom prActices are qualitatively justifiable. Second, the apparent incompatibility in methods and materials for Success low achievers was not limited to those who participated in pullout programs. Although he received no specia! services, John's experience in the classroom basal activity resembled Ginny's and Michael's experiences in their pullout programs more than it did his other experiences in the classroom. Thus any incompatibility between curricula was experienced within the Success classroom as well as between the Success classroom and the pullout programs. The decision to establish a basal reading activity on the surface appears defensible in terms of the pragmatics of running classrooms, the politics of innovative change, and the effecdve instrudional practice of overlapping curriculum with test content (Crawford, 1989). The irony is that Mrs. Starr had adopted Success because she rejected basal readers philosophically. In retrospect, it seems possible that Mrs. Starr implemented an activity that was incongruent with the rest of her program beause she had not yet achieved a coherent philosophy. If this speculation is verified, it suggests that achieving philosophical congruence is an issue for the individual classroom teacher as well as for the several teachers who must collaborate in serving children. A philosophy is the possession of the individuat rather than die program sheihe implements; it is not necessarily coherent, fixed, or well-articulated. Attention should be paid to teacher belief systems, for it is the teacher's interpretation of the curriculum rather than the curriculum itself that is presented to students (Brophy, 1982; Schmidt & Bu(mann, 1983). We should inquire into the teacher's philosophy of reading, or reading disability, and of the roles of teachers and students in learning. 102 0.4 , unicular Expirieneis 87: REFERENCE S Adams, A. H. (1978). Success in beginning reading and writing. Santa Monica, CA: Goodyear. Allington, R. L. (1983). The reading instruction provided readers of differing reading abilities. Th 'wry School Journal, 83, 548-559. ,Allington, R. L., it Johanon, P. (1989). Coordination, collaborntion,.and consilstency:-The iale*go. of compensatcny and sPecisl education interventions. In IL E Slavin N.,L.,itiii;0;: (Eds.);Effective programs for students at risk (pp. 320-34). ittistar: ZiropSY, J. E (1982). How teachers influence what is taught and learned in ciastiononis. Elontritary SChrial' kurnal, 83, 1-13. , Crawford, J. (1989). Instructional activities related to achievement gain in Chapter I claiies.:InR1-, B. Slavin, N. L. Kan/nit, & N. A. Madden (Eds.), Effective programs for students at risk'60. Boston: Allyn & Bacon. Good, T. L., & Brophy, J. E. (1987). Looking in classrooms (4th ed.). New York: Harper & Row. Hiebezt, E. H. (1983). An examination of ability grouping for reading instruction. Reading;Research Quarterly, 18, 231-255. Larrivee, B. (1989). Effective strategies for academically handicapped students in the regular classroom. In R. E. Slavin, N. L. Karweit, & N. A. Nsadden (Eds.), Effective programs for students di risk (pp. 291-319). Boston: Allyn & Bacon. Schmidt, W. H., & Buchinann, M. (1983). Six teachers' beliefs and attitudes and their curricular time allocations. The Elementary School Journal, 84, 162-171. Slavin, R. E. (1989). Stndents at risk of school failure: The problem and its dimensions. In R. E. Slavin, N. L. Karweit, & N. A. Madden (Eds.), Effective programs for students at rak (pp. 3-19). Boston: Allyn & Bacon. Slavin, R. E., Maddcn, N. A., & Karweit, N. L. (1989). Effective programs for students at risk: Conclusions for practice and policy. In R. E. Slavin, N. L. Karweit, & N. A. Madden (Eds.), Effective programs for students at risk (pp. 355-372). Boston: Allyn & Bacen. Waip, T. P., & Walmsley, S. A. (1989). Instructional and philosophical congruence: Neglected aspects of coordination. The Reading Teacher, 42, 364 368. - ". - 1 03 =; -OACHERvwyncTINapigo$1.$:-.80thrltii* VRENSI.LITERATIE1RE:tme.:EANGuAG.ANO:,4**.Cir , t.. . Mes V. Hoffman, Nancy ,L.Roser, 4Minffir i§attle,,C*4ateat, Univirsi6 tiine di Austin Both recent statewide literature initiatives /aid the pressure for,changesiii"fiteraCY. Iiirricula at local -levels Wm-served. to encourage.ihe- use ofehildret0 literatalein the teaehing of reading. FOr the inost part,fite.sWeeping riattife Of reforinagiak, been ,clue :to -a graksropts,thovement' among eduCatOra eager.40,ensure 'that- cpeu,,,;eaclypg, ,.420,4 upon quality, material s .that :supp 4it children's 'desires.:tti read`(CidiMaii.;',':psp) . However, as W ifit any inhovatton within ; a profession, practitioners command AVidelkdiVergent degreek: of 'knowledge, , interest, and cominitinent to the, httiOyatibn's pre- cepts Some May be s*ept along-by the reform movement, feelhig;litqcoMitfitMeht: ,-,td its tenets, as well as uncertainty as to how to impleinent chat,ge. The pitrPOSe àf this study was to examine such a group of-teachers' developing underitanditival 4eaching literacy while they were incorporating experiences With childien'sjiterattire into their classroom reading programs. BACKGROUND Data for this study were collected as part of a larger program effort titled Langitage to Literacy (LtL), a curricular project designed to enhance 'teacher's' 041.wai0 "Sy*, gles for incorporating Jiterature into their reading/language artS progMM (ROker,,Hciff; man, & Farest, 1990). Throughout.the project.(planned,cooperatiVelyby-sehpol dis trict and university personnel), volunteer teachers worked to make the content-Of their daily ste rytime more substantive through the use of plannedliterature units.' These instructional units provided for children's comparative views of themes; patterns; and relationships across books (Moss, 1984). In addition,.,the restructuring of storytime provided opportunities for children's responses tuliterature through discussion, ,art, drama, shared writing, and extensions across the curriculum. RATIONALE Three research perspectives supported the study's design: The first is drawn from the interpretive or qualitative research tradition in anthropology and, more recently, 90 education. In contrast to the positivist/behaviorist approach, the interpretive perspee-, tive emphasizes the meanings of actions as deri led from the actors' points.- Of: viewthe meanings participants themselves construct around a particular set of events', (Erickson, 1986). Although the focus for interpretive classroom stUdies -may* ori: either the teachers or the students, this investigation looked specificallpat teachers! insights and interpremions of the effects of currieular change. The second research perspective relates to the concept:of "ficademiework",as central to understanding classcoom teaching and learning. AdeOrding to ..-oYle (1983), the teacher's role in instruction consists of selecting (or creating),icademieworkter be introduced into the classroom, presoning this work to the.itudents lopgwitli information related t. the task staucture, sustaining students' eft:As to.cOniplett this work, and evaluating students' performance. In this study, the "acaderhic work" Of storytime was examined as it assumed a more central and critical position in the literacy curriculum. The third perspective draws from recent research in teacher cognition (i.e., Clark & Peterson, 1986). Available evidence suggests that teacher understanding is a key factor in curricular change (e.g., Lanier & Little, 1986), as are teacher concems.(Hall-, 1976). This study explored the nature of and changes in teachers' knowledge of learners, content, curriculum and pedagogy (Shulman, 1986), as well as documentedthe evolution of their concerns as they incorporated more children's literature into their classrooms. METHOD Participants The LtL project involved all kindergarten, first-, and second-grade teachers (N =78) from six elementary schools in a large Southwestern school district: 85% of the participating teachers were Hispanic, 96% were female, and all were volunteers. Over 90% of their students were Hispanic, the majority of whom were classified as Limited English Proficiency (LEP) and placed in bilingual classrooms. All participating schools ranked in the lowest quartile of Texas schools on the state-mandated minimum skills test, and two of the schools ranked in the lowest 5%. Data Collection and Analysis Questionnaire data were collected as part of the initial inservice for the LtL program The questionnaires gathered descriptions of teachers' beliefs about literacy instruction, as well as descriptions of their current practices. Other items explored teachers' Stages of Concern (Hall, 1976) about implementation of the project. In addition to the data collected during the initial inservice, two other sets of data were collected. The first related to teachers' success with implementing the project in their classrooms, as gauged by the Levels of Use (LoU) construct from the Concerns Based Adoption Model (Hall, 1976). The second set of data, and the one most relevant to the current report, consisted of a series of interviews conducted during the course of the project and at its conclu- , riachef4" ;104in ii4ighti . skin to both, Corroborate and elaborate the teachers':perceptions;, concerns,- and-de:pie- of unpleruenb' : ten, members of the research, Main:visited etas-stoop:is-op a,b1weeklibasis-"ancl met,bimonthly with giade4evel teams to discust effeetiVe pmeticet indresPond Oilcans. Two questions igitided -the collection and analyses or.these:data::(a)- How did: teachers,belieVereading/lingUage axis insttuctionviaachinging,intheirelasirOoMsl. Milat were teacher participants' insights into their oiin learning, as: *IL as *Or ,stiidents' learning as a rasa of using ;Iitarature,units'iii- stotitirrieTA4eieriptiOn Of the collection, analysis, and interpretation of each data-adored presented in itlitl to these questions. , RESULTS How Did ReadinglLanguage Arts Instruction Change? Self-mports, direct observations, and artifacts '(audiotapes, written records) collected prior to, during, and at the conclusion of the LtL project all suggested significant instructional changes. In one analysis to verify change, descriptive phrases Were ex- tracted from the pm- and post-qüestionnaires so that teachms' perceptions of their typical instruttion in reading/language rats could be compared. These phrases were grouped and categorized under relevant topicaL headings (see Table 1). To- Showchange,'each teacher's descriptive phrases were coded using the teachei itl,ndfYing number. By examining Table 1, it is possible to track individval teacher's Chinghig perceptions of inviruction, as well as to gain an impression of generalized:changm across all particip4nti. Certain activitier, appeared to be unaffected by the introducton of-the.Bterature sharing program (e.g., spelling instruction). Although some activities shoWed alight decreases in the number of "mentionings" by teachers (handwriting instruetion,:hasal reading, reading group instruction), others showed more markei deereas*(e.g., phonics instruction, vocabulary, workbooks, readiness skills, and ESL)Ancreairts were notable in those activities which were aspetts of the literature sharing innovation (story reading, book browsing, and listening stations). In addiden, pair actiVitidi appeared in the post-questionnaires that had not been a part of teachers' earlier deatiipdons of their classrooms. These included writing and drawing in response to literature and engaging in journal writing and writing workshops. Teacher change was viewed from at least four other vantage points: (a) compari- son of pm- and post-stated concerns regarding implementation of the project, (b) analysis of audiotapes of teachers' storytime, gathered on a biweekly basis, (c) inspection of individuil and collective teacher interview, and (d) observations of program implementation by a research project associate. Analysis of teachers' concerns about using literature at the start of the program revealed a primary emphasis on task concerns with some related personateoncems as well (e.g., How will participation affect my evaluations by the principal?). The predominant task concerns centered on finding, time and classroom space for the program. At the end of 18 months, personal concerns had diminished, but task con- , Table Reading Instruction preandeóst Ietlaniptentetuntkii* . . Aspects,ofInsixutitioii' Spelling Phonic; Handwriting P4Qiitifi-o-andr.e: 5, 6, 11,34 13, 14, 14, 24,25, 26, 32, 36; 40 :Post*csOcitinalie 1.2. .17,.11.9: 11, 19, _2O-24t #1:39. 41 1, 2, 6, 10, 21, 29, 14,-15, 16, 17. 23, 34 26, 29;32, 34, 35, 39, 41, 42 2, 19, 20, 26, 32, 3? Pre-Questionnaite 5, 6,8, 1413, 14, 1, 12, 32, 41 37, ESL 5, 20, 23, 29, 34, 36 30, 36, 4,0 3, 4,-6,:7,-8;?12, 14, 36, 39:21.-24 2324; 24-26, * 30, 35, Language Experience Voeabwary Basal Reading 3 2, 5, 10, 13, 14, 15, :7, 21, 37 3, 4, 6, 7, 0, 17, 19, 23 42- 3, 26 5, 3, 14 Science EiplUration- 1.4. 13, 26, 29, 32 Stoi* Reuling- 21 3, 11,45; 70, 100_1,34.*4° 40,41_ tad1118 /36g-Questroi 0'7%i i; 20,, 43;15, .39, 42 804,-19;2y, Jr; 23, - _ ',19.0.-g.41,"---1§L..-_ 713;24:2526 10.39'40'14V- 7-12=13-A4 I =19 -211/4:f. : .- - 410 4, $; 36;--41,402F kbëts , ,10'il3 '14-34 -42'; 849,-,13,04,15; , ,, ":" '-,,-..:'..:- -', 43'34', `f--..: .!:-. -------. . , z..,. . ....2 111;3941- Paintj4:40r.4** -2.9: PloviePi:F4istrips 16, 34 Drann 6-, 943; 24, '34,-3 AO** Tell Lñguagc plo 19, 21i,".'41; 2; 8, 23, 24, 34, 42-: . . , . . .... . ._ -39--- 41r ', - 2.!--8*?..--.!. - -- 4'- - '- , , -- ... , _-..---.- , -.- .- --:- l'..z -?- 2' 25 34' 8 AlVD,ReAPoelts.0,-, :4 A6,19 2.' o5 ".. 2; 34, ,6, 79, 34: --'7.- .-si-715,-,ii*n6,,amm. tNIIIncriCO data ins titia table reflect ieacber - .:8; -42. _ :-;,:, , ... 6,:8, 19,25; 41? , , cerns were still in evidence Again, the primary task concern was one related to time. Some concerns about impact were also represented in these follow-up responses. Teachers' concerns about finding Vme for sharing literature are interesting when considered along with the data on success of program implementation. Frorn analyses_ of a sample of audiotaped storytimes, it was determined that the actual time spent yritli literature nearly doubled over the course of the project. Initially, storytime averaged 15 minutes per day. By the end of the project, books were shared and tespOndedifron, an average of 29 minutes per day. (For a more detailed discussion of these_dati,-soe Hoffman, Roser, & Farest, 1988.) In addition, interviews with teachers (structyred around the LoU framework), as well as their self-reports collected at the_encLof the project, confirmed successful implementation of the critictl features of the literature sharing component by over 95% of the participants. On a scale of 1 to 5, with a 1 indicating no success with implementation and a 5 indicating a high degree of success, the mean rating for success with "setting up a book center" was 4.58, wIrseas the perceived success rating for reading daily from books in the unit was 4.32, indicating that teachers felt they had successfully implemented the twk, program features they had described as their greatest concerns (space and time). These self-reports of successful implementation were confirmed on the basis of first-hand observations of classrooms during the regularly scheduled site visits of the research team. In addition, classroom visits validated the characterization of teacher change. For example, in contrast with the findings from initial "walk-throughs" of all targeted classrooms, in which few tradebooks were in evidence, later visits revealed the significant collections, displays, and artifacts of a literature-enhanced program. These multiple sources of data, taken as a whole, suggest that participating teachers were successful in making changes in their instructional programs. Specifically, they seemed to be successful in restructuring the task of storytime in their classrooms. What Were Teachers' Insights Into Their Own Learning and Thea. Students' Learning? As part of a debriefing interview at the end of the LtL project, teachers were asked to rtspond in writing to the questions. What do y3u feel you learned and what do you feel your students learned as a result of sharing literature in units? The teachers' responses were subjected to a content analysis. Each response was analyzed for its constituent propositions. A total of 93 propositions were identified related to teach: learning, and 191 propositions related to student learning. The average number of propositions per teacher was 2.8. These propositions were grouped based on similar foci, using the constant comparison method described by Goetz and LeCompte (1984). Each grouping was then assigned a topical heading or theme based on the content of the responses No preexisting categories were used. To establish interrater reliability, 12 graduate students and professors of reading were asked to categorize a random sample of these propositions into the category structures that emerged from the data. Reliability was calculated at over 80% for both category structures xteacher-learning and stude Topic.il categories were then rank-ordered from those receiving the most attention to those receiving the least. Teachers' insights into their own learning. Five major categories of teacher learning emerged from the teachers' perceptions of their own growth. ka, Program/Curricu- 109 -c; re Perceptions of Their Own Lca:ntlng _ airowro..... taity with thildien's teranire' ikeniation Eit04,14, ***igi' 000 ' Suategies and Techruques for Sharing Books Development ofnew stratagi,es P:r14,i0fiirg *ills in Children's Learning/Children as Learners Effects of literature on students' learning and insights into,children as learners Teachers' feelings about books and the opportunity to share them Personal Response to Literature Sharing slitthire -0#1? "Iti**ed-iitit4 tohtarathThey oin thelitilestdetaili "I've loath' i'or: lite** thatfiejeciO inn& Was sueh.a RICasaa,t*pe,7 rienee." lar Insights, (b) Familiarity with Children's Literature>, (c).Strategies andtechniqttes, for Sharing Books, (d) Awareness of Children's Learning,..0.40) Persenal'ige to Literature. (See Table 2 for definitions and exainples otteachers",con:Uhents.) Teachers' insights into the legitimacy of read'aitind tiine0 report gram/curricular insights) stand Out in their reports. In addition; Childienrs iiterat*),. learning a great deal about boOks for children (familiarity With as well as ideas for sharing fiterature. Their, insights into childiens; re,SPonSiS 'and capacity for dealing with complex thought (childien leanting/chiOth suggested that the feedback they received from their students was Pogitive and snbatifitial. Finally, they reported enjoyment from working_ in'the program- 'and recognition' that their enjoyment was contagious (p _sontil,Tesponse to literatUre sharing). Teachers' insights into student learning. Four Major categories of teacher aware, ness of student learning emerged from the analYsis of c4;PrOpOsitions: (a) ltesponso tO Literature, (b) Literacy Acquisition, (c) Oral Langtiage:peyelOpmeit, and (d),Ap, .preciation and Enjoyment of Books. (See Tabie 3 for explandioni Teachers reported such diverse student !earning as improved listening *ills, ii le . . r , ,erceptwns :Sudo. 4.- wit.* "....,......,..., .. thsights. -.. .. *304-4! 0er:it* '4taraak 40000h z `',!inailYStonetiillathOrVand; ' Oral Language Developt..ent 7: tt Appreciation and Enjoyment Attitudes toward voluntary or free reading time "*AW-4111ti:': 4-7449 = rtudetitS-7get.this-,4xeitetp: aboOrb-,°°!*"... JOIN. creased motivation, improved comprehension, broadened concepts of readingfibeliefili in themselves as readers, and a more interactive classrootk. At the end of the 18-month iMplonentation- perjorl,..partiCiPatipg Jea,eherS kwere videotaped in group interviews the purposn of the* intervieWs.3*.to:,00,11,*,,,,:,: record of teachers' observations, narratives, anclaneedotal,reSOOnsesio:the-&,0$A,5!: components (including the use of literature units in storYtitne);..frotrt,Ithe,21.491**,4 posed during these interviews, responses to relevant questions (those inviting teachcr reflection on their oc ''-arning,and students' leantin0Were,sUhjeCted.:.* content analysis procedures as Were the written-seipOntes. The: initial Category.; Strue-,7, ture served these observations as_Well: A random sainPle othese çcopositions, 0.t:407-rized by professors of reading education, established interrater reliability at ;.90'after, points of discrepancy were negetiated. In addiiion,th. the verifiCationof:_eategOrielf;, themes of teacher and student learningthat emerged from the Videotaped the tapes also provided rich, elaborated accounts of the effects oi:literattire ugage,int4;: the classroom. 7or example: We'r even buying books for our classrooms. I think we're unit-wise now. Before, we were just walldng in and getting inythiiig We coUld find. And now you think, "let me make this unit," . . . "let me mike . . it's just reallrgreat when you put, a neW Oak out and they, wail; in thereom, Just to see the excitement on their faces of "WOw! New bnOks!" And' hOw they, go to thern and look at thent They just 'really want to go ano get those books even it they.haven't read the storiesSieprkreally'acitingl ,i'etichers' Developing Insights I think probably the biggest difference that I've noticed'from last year to this year is the kids picking out patterns, and looking more for al tors and illustrators, and noticing things that are similar in the books and things that are different between the units. CONCLUSIONS Sharing literature in classrooms and encouraging children's personal responseshelped to provide these teachers with insights about their children's ,literary, Understandings as well as with insights into their own learning. Specifically, *se-4400p reported they learned about the legitimate place for sharing literature in the cIas:sionm, . gamed appropriate strategies for shxing that literature with children, and better trader srood.the instructional environment (and the "tools") that promote student-learnir All of these areas of knowledge---of the learner, of subject matter, of pedagogy anf: of curriculumconform to Shulman's (1986) proposed categories of te-cher leaning At a time in which lifelong teacher education and teacher "empowernret" are predominant factors in the professional model, it seems imperative that teachers refleet on the tasks of classrooms from a variety of vantage points and across time asking cogent questions about the effects of teacning practices on student learning. In tilis study, the teachers' refrain was repetitive: Children's language growth and reading habits are affected by good stories shared in related groJpings. Their reflections on student growth simultaneously revealed their 3wn power as learners. Ultimately, although change was evident, these teachers already possesred far more tacit knowledge than they themselves had aclhowledged. Perhaps, feeling entrapped in prescribed goals, curriculum, texts, and routines, the;f had lost sight of tor failed to find time for) so:ne of their own viable tenets. In evidence throughout their written and verbal responses to the project were their "new" ideas, but more important, evidence that their old ideas were being reexamined, juggled, and reformulated. In some ways, these teachers seemed to feel freed to test their assumptions, to dr If neglected practices, and to try again that which had always seemed reasonable. Yet, although validated in some beliefs, they seemed challenged in others. Whether labeled is leancers or "reflective practitioners," these professionals assumed responsibility for their own learning as they w orked to gain a better foothold on shaping their children's literacy education. REFERENCES Clark, C. M & Petersen, P. L. (1986). Teachers' thought processes. In M C. Watrock (Ed ), Handbook of research on teaching (pp. 255-296). New York: Macmillan. Cullman, B. E. k1989). Latching on to literature. Reading initiatives takt hold School Library Journal. 35, 27-31. Doyle, W. (1983). Academic work. Review of Educational Research. 53. 159-199. Erickson, F. (1986). Qualitative methods ir. research on teaching. In M. C Wittrock (Ed ), Handbook of research on teaching (pp. 119-161). New York: Macmillan. Goetz, J. P., & LeCompte, M. D. (1984) Ethnography and qualitative design in educational research Orlando: Academic Press. kin; o:fg: (1416).-TkestuOy of individual teacher arid profeasor concerns about irmOvations. , (19-.).1anituage. to, literack Proje.41,,Jattuariz19.#7=-,Ittne zui4_,**t'kAtls*':rei9ii;=T6*-E*c,ittOpi ttiTilcY f; le;Faiest, 6Liter±*44atipiltrat:04.1,*4.01114:4151W::; 4'4 ituticn? ,fr6ia,_ ic-...ii 674.1iily..4lia4zviiita0d an4.44***Ott-d*,1*irOlib*,4nt'B:2.1.t,:g , eiiie ez::x'. s: B:laldivin(Edi), pialOgges *ft lite/40 re jetorn.(pii,_431,43:311)... 'Ria4iiiii'P?iiiince., ' - , ,, , , . -.:-.:'-': -- -' Lanier,..3: E:,,A'lLiitle,:riudith`*:,,(19,6). Arsorch:,0 iilicoiy, M. N echiciOtiG.., 4.1!14: O.' Wittnifac:,,-,,, ; IlandbOOk c'of rilertriir on ieachhii '(pp.. 527=569). ,New;:rOt*:hfacMitian. MOis, 3 :T:"(1984).,6Cils- Units hi 'keratin: _A:hand,boOkfoi:eletpettiaify,..;:chOol teacskiii.,:pr4* Waiiofiat, ekiiricil ofTeacher::-Oi English. . ..., ... ,.., . Roser,-N. L,Ikii-fipin, i. V., it ,Earest, C. L (1999): Language, literature, and ",atlisk!=!;chil.40n,.:-.4:11t4.,', Readiag Teacher,-43, 554-559. Shulman, L. S. (1986). Those who understand: Knowledge growth in teaching. Educatiohal lielegicherr;'.41:, 15, 4-14. Ye; f?':4 Z:7p BECOMING A TEACHER OF LITERACY: NOVICE WHOLE LANGUAGE TEACHEE4:1N.C9NYgNPANAI.,.4 INTRP clIPN4 ENVIRONMENTS' , . .,...,. -; , z ,., 1J. Nancy D. Padal(singOlin G. Nelson , Kau itate' University, Although field experiences are generally considered essential:ferteac,;:;prsiart tion, critica charge that they often result in learning ,about error, rather than careful, thought and scholdfship Little, 1986). Alvernrann.(in,presS)Observes that:the 'trial=and-ati*Of support craft" conception of teacher education, although She calls for additional stUdy of preservice reading teacher -e4.000;,,./ particilarly context-sensitive examinations of prospective teachers' development esti-401*e** titioners. Such study should explore prospective teachers' roblems and concerns .which; given the complexity of the teoching-learning process, are likely tO change Oyer time..., Previous research has described these concerns prior to theinduction year: 'fronrIpersonal concerns, such as feelings of adequacy, and acceptanceon coneerna.r.elate4Joi children's needs and the impact of instruction (Blankenshii& Cunningham, 1979, Buitink & Kemme, 1986; Caruso, 1977; Fuller, 1969). ,Indirction-yeafteachers.typi: cally cite problems related to such teaching issues as diSciOline, and managetneit, school routines, and student motivation (Gray & Gray, 1985; Veenman, Recent paradigm shifts in literacy teaching ahd learning,..may cause additional problems and concerns for prospective teachers; they may learntheories aral,Practiees in college that they do not see hnplemented in the field, Allen, Kiernan, 04 140* (1988) found that prospective teachers' confidence increased when classrooin practiees. were congruent with their own college coursework. It is likely that the reverse may also be truethat attempting to implement whole language principles and practices in conventional classrooms may decrease confidence and generate concerns. Thus, descriptions of whole language, prospective_teachers' problems and concerns related to literacy instruction may inform efforts to support their growth as teachers. Case studies (e.g., Merriam, 1988; Yin, 1984) of two prospective .whole language teachers were developed to catalog their problems and concerns over an academic year and to uncover aspects of the classroom environment that affected their development as teaches of literacy. Two questions provided a focus: What concerns do prospective whole language teachers express as they begin to teach reading. and writing? How do they attempt to solve problems encountered during literacy instruction? s. ' 100 Literacy Theory and Reseafcli INFORMANTS AND SETTING Sheila and Karen were informants for the study. Both were traditionallycagedi undergraduate elementary education majors. Each had Acpuotr*,:d 0410th iotq reading .and writing Methoda courses cato,4,; beginning, of the rstudy. These courses :fOcnied on,relatiOnaliins,ainong;Oedrigegical. and; literacy-learning theory and practice and On the teacher as informed dcision , maker within the frainework of whOle language/lenginimeikperience.principleS:-Iie; informants and researchers were well-acquairited, 'and- the .researchera;had:sockfic-} knowledge aoout the content and intent of their College coursework inliteracy learning. During the year of the study, Sheila and Karen completed font 8-weekfield' experiences as part of a collaborative teacher education project coordinated jointly, by personnel from a large, urban school system and faculty from a nearby, University.. The first two experiences (3 full days per week) involved observing, assisting,cliss- room teachers, and taking responsibility for some aspects of literacy instruction, Concurrently, they completed a third reading/writing methods course, which: included weekly seminars designed to support their growth as informed decision-Makers and reflective practitioners. The second two field experiences involved full-time student teaching. Literacy practices in tl school system where Sheila and Karen taught were in the beginning stages of transition from conventional, skills-dominated instruction to more holistic instruction. Curriculum documents and district-mandated recordkeeping procedures in reading, for example, focused upon a traditional scope and sequence of reading skills. Inservice programs for teachers, on the other band, focused on aspects of whole language learning. The nature of classroom literacy instruction within the district, coupled with Sheila's and Karen's understanding of whole language principles and practices, suggested that their development as teachers of literacy might be particularly interesting to study. DATA SOURCES AND ANALYSIS Data for the study were gathered from written journals, interviews at the end of each semester, and interviews immediately following observed literacy instruction. In the written journals, kept during the first semester of the study, Sheila and Karen described and evaluated their own performance as literacy instructors, including their concerns or problems and attempts to or ideas about resolving them. End-of-thesemester interviews focused on teachers' evaluation of their own changes and learning. Sheila and Karen were observed eight times throughout the year. Field notes from observations provided a framework for follow-up interviews that focused on instructional evaluation and interactive decision-making. All interviews were audiotaped and later transcribed. The researchers collaborated in locating data pertinent to the study and reducing the data for analysis. The process first involved identifying data from all sources congruent with the research goals (Guba & Lincoln, 1981). Data sets or case records . Becoming a Teacher of Literacy (Patton, 1980) were then created for each field experience for each teacher. These were combed for patterns and regularities (Goetz & LeCompte, 1984; Lincoln -k Gubai 1985). Inductive analysis uncovered categories of ,eonceras which were then applied to the literacy learning events which comprise4.:4w.case records. Three processes were used to establish the trustworthiness and'platpibility :of indepenfindings. The researchers completed each stage of data reductiOn Tiiangulatic dently and then collaborated to share insights and reselve of the data (Mathison, 1988) revealed that categories Were reflected itiz,the.Affiltiit. data sources, although differences in depth e' explanation Were noted (Le.,:intOieWs following observations yielded more detailed evaluations of instruction'thaniviitieji. journal entries). Finally, member checks (Merriam, 1988), conducted Periodkalry throughout the study, established that emerging findings and interpretations were plausible from informants' perspectives. FINDINGS The inductive analysis of case records yielded eight general categories which are displayed and defined in Table 1. Most literacy learning events included concerns or problems from more than one category, and each teacher's concerns or problems spanned all categories. Both had relatively few concerns as the year began, which might be expected, since their field activities involved observing instruction and assisting in minor ways. As their instructiontil iesponsibilities increased, so did their concerns and problems. Differences between the two teachers emerged from this preliminary analysis. For example, Sheila armulated problems and concerns related to students more often than Karen did, Karen seemed more concerned about her own knowledge. Further analysis of findings revealed interactions among concerns and several key contextual differences. Becoming a Teacher of Literacy: Sheila Sheila's first and fourth field placements were in the same fifth-grade classroom. Her second and third placements were in a third grade and a kindergartn Her concerns and problems during all ph cements clustered in four categories. beiiefs and practice_, students, implementation and personal. Closer examination of the situations that prompted these concerns revea1.d a relationship among these categories and a pattern in Sheila's development as a teacher of literacy. Sheila's instructional responsibilities in reading and writing during her first two placements typically involved teaching kssons from teacher's manuals, as directed by her cooperating teacher. Her first experience, for example, "consisted of a workbook page that was to be used in conjunction with a story that they had already read." In evaluating this lesson, Sheila commented that the students "didn't learn much. I have a feeling that those who al- 'y knew how to use an index were able to answer the questions. The ones who . . . diein't know low to use an index were not able to learn Literacy.theory and fietioik Table 1 'CiiiegorieS of Concerns and Problems -1300 and P.Ortices: conc:erns or problems related to (a),,p,utting Orie's,beliefs into practice, , 'or () 'being asked.to do something that conflictetkwithOimet that I like hoIe ianzuage . . . and fUlly stipPort uniyie of eYerithing) shnr' be doing to be most effective? "I`thought quitte few, thinks.that 1,Warisked,rtodo: were dumb." . Materials concem.t.or problems related to using or Selecting thaterials-Exarriple::"I,Ia4t woUld do it so I forged ahead . . . . The manual assumedrhat,fiteae thifd gri510-$10,094,91,40,' know, the days of the week. Of course they do . . . ." Planning; Concerns or problems.related to planning literacy instruction. Example:."1 had'to think hard to come Up with approaches to help them understand?' Implementation: concerns or problems that arose as lessons were implemented. Example: "one person said 'sad!, and then everybody said that Students: concerns or problems related to the impact of instruction on children. Examples: "they weren't getting my point." "I wonder if the students are learning." Personal: concerns or problems related to the teacher's attitudes or feelings. Examples: "a nervous wreck," "scared to death." Knowledge: concerns or problems related to one's own lack of knowledge. Examples: "couldn't understand why they didn't just say the two rules and figure it out." "You don't know what to expect out of that age level of kids." Managemeni: concerns or problems related to managing instruction. Example: "I didn't like how they had their days set up . . . . Everything was chunks of time . . . . And that was hard for me to work with." from the worksheet . . . I certainly wouldn't use that STUPID worksheet. We would use a real index . . ." Another lesson, again presented to her by the teacher, involved , poetry: "It was a big flop. The children were so bored I didn't know what to do with them. But the teacher had given me the material from the teacher's tuanuai . . . ." Sheila's concerns about these lessons centered on instructional Focedures, the impact of instruction on children, and her own frustration and embarrassment. A description of an experience midway through the second placement shows the interaction among these problems and concerns: "My teacher asked me to [Introduce a story by following] the manual . . . it assumed these children were dumb as rocks . . . . But I said I would do it, so I forged ahead . . . . The first thing the manual wanted the students to do was read the words . . . . They thought I was crazy. . . . . Then I read [a] paragraph . . . Each time the children heard one of the words they were to raise their hands. Dumb . . . . Then I was supposed to read the paragraph again and point to the words (written on the board) each time I said one. Dumber yet! . . . I think the children should have been insulted . . . . All of this was meaningless to the children . . . . I was embarrassed to be trying to teach this lesson." Toward the end of both of the first two placements, Sheila decided to alter manual suggestions somewhat. Her evaluations of these lessons were more positive, but she still questioned if children were learning: "They thought it was fun. I don't know if anybody learned anything or not . . . . The ones that already knew how to spell the 103 Becoming a Teacher of Literacy words were the ones who volunteered . . . [If I had the choice] I think I would not use it at all. But the kids thought it was fun." Sheila cleated her own lessons twice during the first semester of the study. One "teachable moment" began when a child shared a story shes had written at home. The other was a poetry lesson that involved brainstorming, drafting, editing, and, preparing final copies for a bulletin board. Her concerns about this lesson, which spanyled several days, were related to unexpected responses from children: "In fifth grade, they hadh't encountered it [brainstorming] yet . . . they were not used to being handed;sornithing back to change . . . . They didn't even know what the term `rough draft' meant." She solved each of these problems by ask'mg children what they thought- an&hy explaining tne concepts. She was satisfied with the results: "The kids were so excited . . . because they wanted to do it . . . . Each child experienced'great success . . . . They were proud to see their work on the bulletin board, and each wanted to read their friends' poems and have their friends read theirs." Sheila's instructional experiences during the student teaching semester reflected these same three patterns. Early in her third placement, she was asked to "teach schwa . . . . I ended up doing it, and I floundered and was exasperated. I hated it." Eventually, she decided to take mote direct action "So then pretty soon I got bold and I said, 'Can I take my reading group to the auditorium?' And we did some role playing and we acted out the story . . . and we drew things on the chalkboard, and we wrote things, and we did some language experience. And that got better as time went on." This incident became a turning point for Sheila, she took responsibility for "her reading group" during the rest of that placement. At the beginning of her fourth placement, "I walked in and I said, 'We're going to do this and we're going to do this.' And everything was [fine]." Her problems and concerns still centered on beliefs and practices, but in a different way: she appeare4 concerned that her practices were best reflecting her beliefs. For example, she had ccocerns about active involvement and questioning strategies. In addition, she evidenced concern about helping children take responsibility for their own learning. "Ryan came up to we and asked, `Is this long enough?* And I said, `Did you write everything you wanted to say? Do you have any more ideas you want to put on this paper?' " Overall, Sheila progressed through three general stages as she became a teacher of literacy. At first, she did what the teacher asked her to do. Since she was required to use teacher's manuals, many of her concerns related to the conflict between her emerging beliefs about literacy learning and the practices she had been asked to follow. She also expressed dissatisfaction and frustration about the impact of her instruction on children. She sensed that they were not interested or learning, but seemed unable to solve this problem, since she did not feel free to devise her own lessons. "When she tells you, you have to do this. You know that you could do it a different or better way, or not at all. You still have to do it." During the second stage, she still followed the teacher's instructions for the focus of lessons, but made her own decisions about implementation. Shc evaluated these lessons more positively, but questions persisted about whether children were learning, as opposed to practicing things they already knew or simply having fun. The final stage, when she became "bold," was characterized by independence in all aspects of 104 Literacy Thoory and Rescarch planning and implementing instruction. She took this responsibility rather than waiting for it to be given to her. Although management concerns increased at this point, she expressed satisfaction and confidence in her own ability and in the impact of her. teaching on students: "when I got bold and started saying, 'This is how I want to do this,' that was when I started." Becoming a Teacher e iteracy: Karen Throughout the year of the study, Karen taught in a fourth-grade:classroom, a kindergarten classroom, a sixth-grade classroom, and a second-grade elassrodru..She frequently expressed concerns and problems relaux1 to beliefs and practices and tO her own knowledge. Again, closer examination of instructional situations sheds some light on Karen's development as a teacher of literacy. Unlike Sheila, Karen began participating in literacy activities almost immediately: "Since the second day of class. I have been reading to the students daily:" During this same placement, Karen asked to implement journal writing in the class, because "I thought that this would be an effective way of putting the theoty, 'The mom you write, the better a writer you become' into pracesce." The teacher agrted, and Karen began. She thought that the experience went well; her concerns reflected differences in her beliefs and those of the cooperating teacher: "My cooperating teacher [insists that students write] in cursive . . . . I wouldn't want some students to hold back due to not wanting to write in cursive." "They are so used to bein structured and spoonfed . . . . They still ask me questions about spelling and topic selectioa . . . them are usually a lot of questions at the beginning . . . but once they begin writing, they seem to move on rapidly and enjoy writing." During her first placement and most of her second, Karen implemented some of her own ideas, but only on "my time": "She does not give them the opportunity to write in their journals evety day or encourage them . . . [to read or write] in their spare time. I, on the other hand, suggest this quite often. I really push reading a good book or writing in their journals whenever they are finisho: with . . . worksheets. This power of suggestion is starting to take root." If Karen felt strongly about situations, she tended to ta2k with her cooperating teachers about her own views: "I was brave enough to suggest this to [cooperat;ng teacher] and she agreed." "So I took the chance and asked about writing a book about our class . . . ." Sometimes Karen simply proceeded according to her own beliefc as when she tutored Sam, who was having difficulty learning cursive handwrit- ing: "[cooperating teacher) has given me a 'mile high' pile of worksheets . . I realize Sam needs practice, but . . . I think he shou I practice his writing in some situation that is meaningful to him . . . I think I will be sneaky and use my own approach . . . ." Karen's knowledge concerns had two sources. Some reflected insecurities ubout her knowledge; during her sixth-grade placement. she also worried that students might "know more than I do." Most of her knowledge-based concerns, however, were related to finding the "best way" to put her beliefs into practice: "I like . . . whole language . and fully support it, but am unsure of everything I shoukl and shouldn't be doing to be most effective . . . I want to do the best job possible." These concerns 119 105 'Becoming a Teacher of Literacy escalated during her kindergarten placement: "I walked in on the students going through the alphabet in this fashion: B-B-B-Ba-Ba-Ba-Banana, C-C-C-Ca-Ca-Ca-Cat. Needless to say, I cringed." Sometime later, Karen and the cooperating teacher'"got into a big discussion about teaching language arts. She asked me how twouldleachit and I said I would use whole language [and] explained the theoiy behind it She said that she has been slowly moving in this direction . . . but that she doesn't have enough information." The teachez then asked Karen for infornaation,:whobe-. do something . came concerned about her knowledge: "I'm afraid . . . I want this to work out so she is convinced of the power of whole language . . pleace give me any advice . . . ." During her student teaching semester, Karen was again able to put,h,4 own ideas about literacy learning into practice. As a result, she was frequently concerned:about issues related to beliefs, practirPs, and implementation. As she planned lessons, for example, she questioned such ..-Aes as how to encourage involvement without taking conlrol of the conversation away from children. Later in the semester, she:noted that her instructional focus differed from the teacher's: "She'll focus on syllabi.: or something . . . . I'm not focusing on a skill . . . I'm focusing on just reading." By the time she entered her fourth placement, her confidence increased and she readily evaluated situations and resolved her own concerns: "She preached that shewas whole language and I . . found that she didn't know what whole language was . . . from my experiemm . . . she was doing things like journal writing . . and it wasn't really writing, it was fill in the blanks. So I . . . decided that I was going to do journals and let them writ,. what they were going to write." Despite differences in their teaching experiences, both Sheila and Karen completed the year feeling successful. Sheila commented, "By the end I was so excited . . . . I just think reading and writing encompass all the other aspects of learning . . . and you can teach anything as long as you make it relative to what's going on." Karen's self-evaluation focused on confidence: "It is hard to know if you are doing the right thing or not . . . [but] I felt like knew what I was doing . . . . I believed in myself much more . . . toward the end I saw things I didn't like, but I tried to . . . implement them in my own way." CONCLUSIONS For Sheila and Karen, the process of becoming teachers of literacy involved weighing the value and logic of what they had learned about literacy instruction against their cooperating teachers beliefs and practices. Concerns and problems arose as they attempted to mediate differences between what they believed or had learned and what they observed or had been asked to do. Although they completed the mediation process and resolved their concerns and problems in different ways, several aspects of classroom contexts seemed to affect their development as teachers. The most pervasive contextual factor was the extent to which classroom praefices- and the literacy-learning beliefs underlying them corresponded to the prospective teachers' own emerging beliefs. Until she "got bold," Sheila typically followed the teachers' routine, the message she received was, "Do it my way." In Karen's case, Literacy Theory and Research teachers allowed her to implement her owr..ileas, but expressed satisfaction with ther own ways of teaching; the message she received was, "Do what you want, but my way works well." Although ur 2.'.astandable from the cooperating teachers' points Of view, both these stances created concerns, since classroom practices conflicted with the prospective teachers' literacy-learning beliefs. The mtture of materials and activities available for literacy instruction was another contextual factor that affected concerns, since these also often conflicted with prospective teachers' beliefs. Both expressed concerns about whether and how to use crnVentional materials provided for thorn. Bntli also attempted to use materials in Mannerscongruent with their own beliefs but were not satisfied with the results, particularly in regard to children's learning. Communication between cooperating and prospective teacher was et another contextual factor that appeared to influence these prospective teachers significantly. The prospective teacher? perceptions of cooperating teachers' expectations seemed particularly important. Neither expecting total adherence to the established routine nor expecting to learn new ways of teaching from the prospective tt .tcher was helpful; in fact, these expectations caused a majority of the expressed concerns. This study described the process of becoming a teacher of literacy by examining how two prospective whole languagc o...szhers reacted to conventionsl instructional environments Although an exploratory study, findings suggest some contextual influ- ences on prospective teacher? development that warrant further examination. Of these, the congruence between prospective and cooperating teachers' beliefs and practices, the types of literacy instructional materials available, and communication between prospective and cooperating teachers seemed most critical. In recent years, we have become increasingly aware of the importance of viewing literacy instruction from a sociocommunicative perspective. The results of this study suggest that viewing prospective teacher's development from this perspective may reveal key aspects in the process of becoming a teacher of literacy. These, in turn, may inform efforts to suppon prospecti ve teachers* development. Allen V , Freeman. S . & Lehman, B. (1988. December). A literacy educational model for preservtce teachers Tieuulating observation and rell.cdon into exemplary practice. Paper presented a the meeting of the National Reading Conference, Tucson, AZ. Alverinann, D E (in press) Reading teacher education. In . rt. Houston, M. Haberman. & J. P. Sikula (Eds ). Handbook for research on teacher education. New York. Mwmillan. Blankenshio. I W . & Cunningham C H ('979). Classification of elementary school student ..achers expressed needs. College Student Journal. 3, 374-378. Buitink. I , & Kenzie. S (1986) Granges in student teacher winking European Journal of Teacher Education, 9, 75-84. Caruso. J. J. (1977). Phases in student teaching. Young Children. 33. 57-63. Fuller, F F (1969) Concerns of teachers Dev elopmental concepfion. American Educational Research Journal. 6. 207-226. GoetL I P. & LeCompte, M D (1984) Ethnography and qualitative design in educational research. Orlando. FL: Academic Press. Guba. E G , & Lincoln. Y S. (1981). Effective evai ,don. San Francisco. Jossey-Bass. .Aq A A`i - 't. :mitt OtENTiko 0000 ii400.1*****NO MitiCOLF*FOkcUIWCAW4ipst.fr Kim McLarty; Je0ifef GOodoiSin,N Peabody, c611iii This paper will describe the design principlesuf autirricalinn der °hance student's literacy.developmentand acquisition of knoWledge.,,wci ,an -overview of our work in video-based'insinietional,conteits :Iwo; Risko, Rowe, & Vye, 1989). In our overview, we- 40.$40,4*-0:i4t#104A,..* provide a brief description of OUT research project ex-a-mining, theeffeC; strtictipn on studenie 'learning. We then identify instrUctional deciskm our curriculum development. Our curriculum, based on what we have called "anchored instructiou;" designed to create a rich .source of information within a shared learning enyirdarpent that generates interest and enables students to identify and define prOblems Whiiethey explore curriculum content from many different perspectiv..!a (BranSforil; Vye,"Kinier, Risko, 1990). We have reported previously results of investigatiOnS,..that '1,tave demonstrated the effectiveness of anchored ins! !ction for enhanci 3tudentW:VbeabU7 lazy, reading comprehension, writing, and Anowledge of sciciar- itudiebbliCeing (Kinzer, Risko, Vye, & Sherwood, 1988; Risko, Kinzer, doodinan, -MaLaity,, Dupree, & Martin, 1989). The design principles discussed here are gniunded in research (referenced above, partly supported by OEM grant G008710018) and can be applied by others who wish to implement an anchored instructional approach, especially using a vidwdisc-based anchor. OVERVIEW We have been exploring the value of providing anchored instruction to enhance student's development of ilteracy knowledge and general knowledge across the curriculum. As suggested by theorists such as Dewey (1933) and Hanson (1970), we believe that students' acquisition and use of knowledge is greatly enhanced when -they are immersed in opportunities to examine phenomena and to experience change in perceiving and understanding important concepts. Introducing students to pew eonaipts and principles within problem solving situationssituations in,w/* teacheriguide Stu, dents to use information as tools for solving problernsallows them to exiierience the changes in their own thinking and an opportunity to elaborate On andevaluate the new information (Bransford, Vye, Kinzer, & Risko, 1990). Unfortunately, for too 109., . Si no Literaey tilenrjrandit:' many students, the introduction of concepts and theories seems like the mere introduc:-',. tion of new facts to be memorized. When instruction does not encourage StudentS examine ideas from different perspectives or to use information for sOrving PrObierns; it is unlikely that these students will experience liOW the newinforinatinn,:ciii,# their thinking in new, but relevant contexts. ,FOi*:these stuclentsikticquit4faCtS,::)? remain inert knowledge (Whitehead, 1929) beeause.students kail information when it is needed for solving relevant prOblems. Anchored instruction provides rich and Cohesive inionnational contextslhat \ep able students (a) to identify and define problems;,(b) to specittreasoits4#0yeit solution; (c) to generate, under minimal teacher guidance, StrategieS forsolying identified problems; and (d) to observe quickly the result ot their attempted'sOlOn4,tuf.dents are then introduced to additional information that is relevai# to the anChor, and are able to use their, problem-solving strategies to solve related problem. The.rnajor,t goal of anchored instruction is to enhance learning retention and use of,co Icelits by, letting students experience the changes in their perception and understanding of the. anchor as they examine information from multiple points of Niiew. Central to our implementation of anchored instruction is the use of videodiscs tO create rich problem-solving environments that serve as shared contexts for explorationand discussion. Ideally, anchored instruction (a) involves a problem-oriented approack to instruction; (b) involves sustained thinking, often in groups, about problems; (C), permits students to integrate skills and knowledge that in normal curricula remain.. disconnected; and (d) does not presuppose extensive background knowledge. We used the film The Young Sherlock Holmes (Spielberg, 1985) as our primary anchor for relating content across subject areas. This film is available commercially and was purchased in its videodisc version for a cost of approximately 830.00. The; content of the film is sufficiently rich to support sustained thinking about problems and concepts embedded within our curriculum. For example, students were immersed in the exploration of the film to study plot and subplot information and to identify characteristics of a well-developed story. Students were encouraged to define story attributes and to monitor their perceptions of how story elements relate to each other (e.g., how character motives lead to goal statements) and to their comprehension of plot development (e.g., the resolution of goal statement). Students used this information to compare the "Young Sherlock" story to other stories, including written mysteries (including but not limited to other Sherlock Holmes mysteries) and when writing their own stories. Second, the anchor was rich with the information that was needed to comprehend relat-n text and to facilitate class discussions. Often, students have comprehension problems or have limited access to class discussions because their prior knowledge is insufficient or not accessed appropriately. Video-based anchors extend instruction beyond the typical lecture, text and workbook approach to delivering information because they contain much richer and more easily-processed sources of information than are avaiiat' in verbal, or even wrhten, description. With videotape, we can combine dynamic, visual and auditory information (e.g., facial expressions, affective states, gestures, voice inflections, scenes of towns) with oral language and print. With the introduction of random access videodiscs, however, we have additional ability to . .1.24 lementing ii;:etioredinstruction return almost instantly to any segment and re-view events. This rapid access to rich information can have powerful effects on students' learning and cognitiVe development because it increases opportunities for finding relevant issues that-arc ettibedded in the in,pres4. video (Brans!ord, Sherwood, Hasselbring, Kinzer, & Third, the content of the film, set in turn-of-the-century VictoriapEngland, permitted students to httegrate skills and knowledge that ,in traditional eunict4a reniain disconnected. Students acquired useful inforrnatiooahout v.-that it *as during this time in history as they explored the filin. Their ppoittiOns anci, ttrickrStanding of the Victorian era were expanded ,as they studied' the,filth and "related;,t_Otito identify historical, political, geographical, and technological issues relevant i9t1oi, time period. Fourth, using the film as an anchor pmvided a shared learning context forinecliation. Students who view video in the absence of a mediator may be entertained, brit they may miss most of the opportunities for learning that the video provides. Traditional instruction is limited, also, because teachers and students usually do oot have common background experiences to which new learning can be connected. In such situations, teachers usually tell students answers instead of providing problem-solving environments in which students acquire methods to frame problems and use information independently. Conversely, teachers and students in our curriculum shared the anchored experience, and our teachers mediated students' learning by arranging instructional conditions and providing instructional feedback that helped students to recreate and examine mutually familiar information so that discussion and learning was facilitated. Our teachers provided organizers, such as character analysis sheets, which were completed as a group and used as a reference for later analyses of content, such as the comparison of a character's behavior across several episodes. We believe there are several important differences between anchored instruction and traditional curricula. First, our curriculum goes far beyond the loosely connected thematic instruction that is provided by many commercially prepared materials. Our instruction provides explicit information about the anchor and explicit links to the multiple ideas that are related to the anchor. Second, information in our curriculum was integrated around specific problems to be solved (e.g., students were erPouraged to view entire episodes of the film to identify characters` motives contributing to characters' actions and how the setting information, such as social conditions of that time period, may allow or inhibit such actions). In contrast, traditional curricula often introduce different ideas in different contexts. In most basal readers, for example, students are introduced to different components of story grammars. Each component is often introduced in a different story rather than in a single story. Similarly in many attempts to teach social studies infol:nafion, different components (e.g., geography, social conditions, history) are taught in the context of different examples. Traditional curricula are organized so that science, mathematics, reading, writing and so forth all tend tc be compartmentalized and taught in different contexts rather than integrated into single contexts. Third, we believe that the use of vit1-1 provides students who are aehind their peers in reading development access to information that forms the basis of class discussions. Further, videos contain much richer information than is available in some printed media and in computer programs with limited graphics. , Gestures, affmtive states, scenes of towns, music, etc. accompany the dialogue..This. rich source of information allows for the possibility of finding relevant informatitin. that is embedded in the video. DESCRIPTION OF RESEARCH PROJECT Our research on anchored instruction was conducted oyer, a three,Yearpeii two fifth-grade classrooms. The main goal of our project *as to deiielbp,..aAit based, anchored cuniculm and to assess its effect on students! criticaLthinOng,_ independent use of information. Second, we planned to develop a cutiitim4at: could be-implemented by classroom teachers. During the first 2 years of this project, students participated in either the extteri=,, mental, anchored program or the control proram. Studenta in both programa,' viewed . the film and were introduced to the same basic content (e.g., vocabulary words, ittirY elements, social studies concepts). The major difference between the eXperiMentat and control programs involved the use of the video-based anchor. In the eitperimOtat,, program, each lesson was tied to the anchor. In the control program, lessonifolloWed' the more typical format of focusing on target concepts within contexta tftat were, unrelated to each other. There "2S no long-term anchor for the control' program Across the 2 years, analyses of data collected from multiple sources (stildents' pre- and posttests, field notes and analysis of video- and audiotapes, and student and, teacher interviews) reveal large differences between the experimental and 'control students on several measures. Experimental students, across ability levels, were better able to describe character's feelings and corresponding motives, and they wrote storlei` that were more causally coherent, generating plots that linked character actions and events to goal statements and goal resolution (Risko et al., 1989). Recall of vocabulary related to story information and spontaneous use of vocabulary in novel contexts were significantly higher for the experimental groups (Kinzer, Risko, Vye, & Sherwood, 1988; Risko, et al., 1989). INSTRUCTIONAL PRINCIPLES FOR CURRICULUM DESIGN A careful examination of our lesson plans, extensive field notes recorded during classroom observations, and interview data helped us to identify seven major decisions that guided curriculum development. We believe that these decisions have implications tbr instruction beyond the scope of our project; they are general and need to be considered when developing any anchored curriculum. The seven key decision points which guided the development and implementation of our anchored instruction relate to.. (a) choosing an appropriate anchor, (b) developing shared expertise around the anchor, (c) expanding the anchor, (d) using knowledge as tools for problem solving, (e) teaching with the anchor, (f) merging the anchor with literacy experiences, and (g) allowing student exploration. (0kliiniiitinkAtiehatid Instrietian .Guiding Principle 1: Choosing an Appropriate Anchor Choosing an anchor depends largely on the curricular goals that are targeted. That is, an anchor is not equally appropriate for all teaching situations. We:began by analyzing the fifth-grade curricultun'to clearly speCifxhistructiortilgoals.10rice these ;goals...were established, we examh.ed several possible yideo.suiciiOrtfyit were.identi- fled as,:being appropriate for the age group, coasidering such thiags as language, vialence, interest, content, and so on. State. guidelines mandated curricular goals relating to plot, characterizatioir,:iro-, cabulary and comprehension development in reading. In sociarstudies included a unit of American and world history, whereas the science Ctirrieriluri iiicluded units on inventions and simple machines. These beeime the,core arc4,6f our curriculum. We thus searched for an anchor that included a strong plot Ilne,, with' clearly defined character development, definite causal structures thafcatild"be related to story grammar elements, and verbal and visual clues pertaining to a clearly definable period of history. Some potential video anchors were rejected because they were too general, thus requiring too much external, teacher-imposed structure before curricular goals could be related to the anchor, or because they were so narrow that setting information could not be generalized to a historical time periodthat is, the sating might not have included "markers" that could be targeted as indicative of a particular historical Period. Thus, the most important decision in choosing an anchor came before a range of anchors was evaluated. The critical element was a clearly articulated set of instructional goals. The choice of anchors depends on the relative match between these goals and the video anchor. Guiding Principle 2. Developing Shared Expertise Around the Anchor The second decision for curriculum development relates to facilitating students' expertise around the anchor. We began with the belief that a well-developed base of information could provide numerous, rich examples of target concepts that would enable the teacher and students to make links from the movie to contexts across curriculum areas. However, after evaluation and observation we realized that one .,howing of the film was insufficient to build an awareness of the complexity of information present in the anchor. It is not necessary to show the entire film repeatedly to address this issue. After the film was viewed once, teachers and students together identified meaningful units within Young Sherlock Holmes. Using the random search and access capabilities of the videodisc technology, the teacher quickly found and replayed these scenes Class discussion of each scene centered around the characters and actions within the respective scene, as well as how the scene related to the rest of the film. Students noted the frame numbers for the scene under discussion and also brainstormed to find a label, representing the main idea, for each scene. These labels became mnemonic reference .nts in later class discussion. Other activities included the preparing of a storyboard for the film where the important segments were listed, a defining picture leas drawn, 1:2 7 in_ 444 714*as poatgtiri:i*C. refereneeloigiie for Studeingifixt,teaCher,Wli'en*shint.to:finit , .. 0._,..)..-t9,'14e az; example or to make a point. - ,..ti*Set-ved trim Se Sitickati.to, haVe clearly in mind 'Mc ;gerieraLPOOOKtife =814M4401.1*:' 4SSilte.tit'in,Oeising:#49000/;*** 'Icrienitedge of the aqçhor Iincreased An obser'YediiikonieoUthik. .Stildent4eOeratea digcOSskin,*11:04's;u1044**C4 - 0006i4: ill*:*e0 *01, 4.i.lcu00.s:' , t. 7 Derrina;:c"Nylij 46061 -tlie`ihit-theity in That'itliiser Mai; Gokrbniiii:,-"Whi*Oultirkt.theihaVe hit,bim in the lead?" Loids:-"Beeinie it Might be likebecitiiie YOU inighi not feel itas David: "It may be a blood vein, itha-.YOu'Iiii-lein aid.thatlWoUld'aendAt circulatiori.;Anithei Ming If It bits him In the head, ittio it 11# She*oqi So oOt*t WooM *kn." Paula:* "It it had hithiinlithe head; it WOUlti probablY "batlike oft Mrs. Goodman; "WhY wOnid it boince off, Paula?" Paiila: "All ihe hair." (David is mumbling in the background that it wouldn't bounce off. He deep% bnY Peres idea.) Gwen: "I have this body chart, and it has a vein right here [points to neckland-it-is the inain.blood vein that travels tip your neck.' Mrs. Goodman: "What is the name Of that vein?" Gwen: "It's the main vein that comes tip fromsour arm." David: [on your head] "you've got hair, then the skin, then the skull. You've.got to go through just to get to the brain. On the neck there's no hair, and the-skin won't stop it." Louis: "You've got clothes." Mrs. Goodman: "Do you have bones in your aeck?" Andy: "In herilth we are studying the skeleton." (Lindsey gets up and goes over to the chalkboard to look at a chart titled-"All-Kinds of Skeletons." Mrs. Goodman suggeSts that they could ask their health 'teacher to; answer their question. Dean says his Morn works at the Health Department. Mrs.Goodman suggests they ask her also.) Note that discussion moved from the video anchor to a body chart used for. their- health class and a student's explanation of how a dart in a neck- veiii-couldlCiii person. Students also were able to answer each others' questionsOnce the leacher . and students had developed expertise abOut the anchor, Unica to other subject iireas, and to their prior experiences became a common occurrence within ihe classrooni:- The transition from teacher to student responsibility appears to be a natural OutgrOWili of developing a shared expertise. Guiding Principle 3: Expanding the Anchor We began with an anchor that we felt was sufficiently rich to provide a foundation for our curriculum. However, we found that one anchor, even one that has a wealth 1 . st. .ming-AricKareOnstructiaii. o0aried conterat;Atild not theet all the needs of our students and lurricdum.,Instead,, 1,11140r.9046is ex.4ted k'irsti,Saniegatis:.,ininforrnatiOn-provided by the anchor required us,..to introduce additiOnal, related miteriais: We - Chose another. movie,. Oliver TiviSt:(4.ail,:194),, :Aairt a**, th4plves of.WillouglibyChw (Aiken) ,1903),;iliat,Wereilso set Inthe 'Vktor*4,Eia,.`The introduction * Oliver Twist providednur,studentsviith,infOrtfaiiotr, to .cOtrastc and extendliformation. For exianole,.:theliving,conMoOs'ni the rich, As .Sherlock, were contrasted With liVing Conditions* the paOr aS sten in qliver Additional information, used for aChieving a OlOre balanced tit,eser (ntion,.was.,Seiected; ,,the basis of how well it related to curriculum goals '. nrat, the primary anchor. '.-Comparisons and contrasts were explicitly made betWeen new infonnatiOnqind::the. anchor. Second, students' interest began to dwindle as they became very fam iliar With the anchor. This-problem disappeared when additional Content sets wereadded. This was seen most explicitly in students' increased understanding of conctitS such as orphanages and child labor laws. Students in the experirnental:gioup' had a intick, _ richer understanding of these situations in Oliver Twist as a result Of .eadint abont these aspects in The Wolves of Willoughby Chase, which was introduced toe?itend the upper-class concepts presented in Young Sherlock Holmes. When We expanded . the anchor by introducing the additional content sets, students' intereSt rose as they used knowledge from the anchor to explore and understand new information. Guiding Principle 4: Using Knowledge as Tools for Problem Solving Our next decision related to building students' knowledge in ways that allowed them to use this information to solve problems or to relate information across content areas. As expected, the anchor provided a meaningful shared context from which students acquired new information. However, their use of information to identify or solve other problems didn't occur spontaneously. aal curricular decisions stood out as being most effective in helping students retrieve information when it was needed for problem solving. First, students were given a purpose for learning new information that they could accept as a valuable reason. For example, exploring the film to identify events 12nding to goal resolution helped students to develop a cohesive plot structure when writing their own stories. Second, students were provided with organizational tools that would help them recall information when it was needed. For example, students were shown how to use a story grammar outline to help them organize and remembtz plot information. Third, students were provided with opportunities to use this knowledge in real problem solving activities. For example, stuaents assuming the role of an American Congressional subcommittee explored multiple texts before reporting on medical advances of the Victorian era. With mediation and practice, problem solving began to occur more frequent,. Teaching infranation well and in an interesting fashion did not insure that the students would use it in other settings. To help our students learn characteristics of situations requiring use of information, it was necessary to provide opportunities for students to find relationships among ideas within and across contexts. For example, Lib:6041*n and plots and subplots were studied in their entirety so that students could trace eVenta-, contributing to character motives and goals and evants.leadifig *gold-,attainiAenii Instead of asking our students to recall 'facts aboUtthe=eharaeters,:nt;eventaiiiiti* story; students received problem-oriented instruction reqUiring theinIo:aziinparciiir.! acter motives and plot events acrosi the anchor andWrittO mstructi provided a way for students to practimapplying,their inawiedge.andsee.4val*in solving problems. We believe that thietype of- biStrtietioninCreasesiitadentai.*4 to understand the conditions and constraints'of knowledge So that transfer to new contexts is more likely to occur (Anderson, 1983, .1987). Guiding Principle 5: Teaching with the Anchor We anticipated that use of video in teaching would come naturally to teachers: once the film had been shown and discussed in class. Just as good teachers:encourage. their students to reread parts of a text to find specific information or,tn-correct derstanlings, we expected our teachers to revisit specific scenei to enhance.; eta* discussions. Although this occurred to some extent, We found that teashersjnidallYi... verbally referred to scenes to remind students of the film content. TheY did dot *Olt return and re-view specific scenes to illustrate concepts within, tje'cQntetoftii class discussion. Our teachers required encouragementto conacionsly,prepare 'plans incorporating specific references to scenes that would tie to their instructional goalS...,, They also needed practice with equipment before they would go,to the scene during,. s instruction. We believe it is important to actually revisit scenes because there isa difference between saying "Remember the library scene? Remember 'how-pudley-acted?", and actually using the anchor by saying "Let's !aok at the library scene,and notice how Dudley acts." In the library scene, there is dialogua between two charmters, Sherlock and Dudley. Sherlock outsmarts a very proud Dudley when he explains that Dudley's "antique" watch is a fraud and not an antique at all. When students. view the scene, they can notice Dudley's tone of voice, the manner in which he holds his head, and his gestures, all of which can produce the inference that Dudley is pompous In a similar manner, students can be asked to review this scene to detect a character trait of Sherlock (perceptive ) or to determine how Dudley may feel in this exchange with Sherlock. Instead of just mentioning the scene, the actual reviewing of th u! scents allowed students to notice more specific information on particular aspects of the anchor. As we found with our students receiving anchored imtruction, teachers' acquired expertise with the anchor enhanced their spontaneous use of information presented on the film As teachers became experts on the anchor and felt more comfortable using the equipment, they were more flexible in acce., .g and using scenes that explicated imtructional goals. Guiding Principle 6: Merging the Anchor with Literacy Experiences Since vt,t believe that it is possible to strengthen reading and writing skills through the use of video-based anchors, we wanted our curriculum to provide cohesive links between visual and more traditional literacy experiences, both oral and written. Although it is important to link the video anchor with literacy experiences, how they 13 0 --'1111,.E , !nPlenierUitig AUcluiredinStruction i7 ute linkedis a crificaLissue. Literacy experiences needed to be. strongly related to thf,., -+Unchor. Asking students to explore the anchor to identify conditions Contributing +to :chitracterS' !nouns was a precursor to story writing tha targetedcauses forchuracters' aetiOns ,and'goal statements. The anchor proVided many opportunities to read, write, and use oraP4iguage. A:slasSroom neWsletter was written, ainied at sharing inforination.abOut,;t*Oeber .+voli 'as information students had,explored+becpuse theirinteieSt Aval+PitfUed:by,,a+ ,speciac aspect of the anchor. Group -Writing activities,:including author circles, 01- lowedstudents to critique each other's workand'refine writing thit cod:F*40100: .or,could,appear in the newsletter. Oral reports and +skits based oirseinas*4r,,t* anchor were presented by groups to the class. Library research tinie, provided:so: that, students could pursue interests arising from studying the anchor, facilitated+develciPrnent of research skills. In short, linking the video anchor with other reading, writing, and Oral language activities allowed students to become more active in their dWilearning. They-were able to expand what they were learning from the anchor into other literacy experiencer Students' stories began to reflect what they were learning about story structure-from our anchor. Students' reports on the living conditions of the Victorian era represented an analysis of societal and economic issues of that time in history.+Students developed expertise, based on library research, on the women's movement, child labor, crime and punishment, health and living conditions, and education during the Victorian era. This was then related to the eu1rr4nt status of these issues. Additionally, we began to see links between the anchor and what our students were choosii,g to read in their free time. Guiding Principle 7: Allowing Student Exploration In order for students to develop a sense of expertise, opportunities were provided to allow students to explore the anchor as well as to explore areas of interest generated by the anchor. T3 facilitate this exploration, the videodisc was made available to studentsit was not only under teacher control. Students were able to access the anchor and to find information, using the videodisc, during group project time or during independent study time. For example, when exploring Victorian architecture or setting information, students were able to find relevant scenes without asking for the teacher's help or permission. This resulted in further er.ploration through library research, with the findings used to check the film's accuracy. To effectively use the anchor for independent exploration, however, it was neces- sary that students had available to them an outline of scenes and frame numbers. Purposeful exploration required that students know where to look in the anchor to find relevant Information. This was provided by the outline, developed jointly by teacher and students, listing scenes and frame numbers (see Principle 1). The other major aspect of exploration was sharing. After students explored topics of interest either individually or within group projects, time was provided for class sharing of information. Sharing was accomplished either through the newsletter, mentioned earlier, or through formal skits and/or oral or written presentations. This resulted in several benefits, including more open-ended but relevant discussion that was often initiated and controlled by the students. Allowing time to explore the anchor, sparked interest in relevant topics that we had not anticipated. SUMMARY In summary, there are seven principles that can be used to guide the,selectitiii;: and construction of anchors and the selection of teaching activitie,v4o)aCc'orripanyi, them. The principles are interrelate& For example, un4 students develotire:x0Orti.ie.. around the anchor (Principle 3), they cannot u. their knoWledge for Piiiiipsefitl.explO-, ration (Principle 7), or to most effectively link the video anchor to (Principle 6). However, anchored instruction is not presented as a stand-alone ori new puce= dure to enhance learning. There is much known about what makes instruction more and less effective. Some of this knowledge includes the importance of backgroopti: knowledge (Anderson & Pearson, 1984), the importance of shared knowledge between teacher and student (Chapman, 1978), the benefit of cooperative learning (Slavin, 1983), the importance of motivation (Wigfield & Asher, 1984), the facilitatiVe OffeOt of problem generation (Bransford, Franks, Vye, & Sherwood, 1989) and so oneUnforr tunately, although teachers generally acknowledge the importance of these issues, too often instruction does not incorporate this knowleuge. Durkin (1984), for example, hr.-sfound that teachers are aware of the importance of prediscussion to address background knowledge, yet found only a very small percentage of teachers using this procedure, even when advocated in a teachers' guide. Teachers often state that constraints relating to time, social, and linguistic factors in the classroom often preclude implementing' what research has shown to be effective. Based on what we know about effective teaching, anchored instruction was developed to provide opportunities for teachers and students to work cooperativcly in shared experiences The use of anchors encourages students to view anchors from their own perspective, to link learning across subjxt areas, to find information to explore problems that they identify, and to experience changes in their comprehension as they are introduced to new ideas from the teacher, from the texts, and from their peers. Providing an anchor is close to what has been advocated in other professions for some time. For example, it is closely parallel to case-based instruction in law and medicine. Case-based instruction, as described by Christensen (1987), Learned (1987), and Rasinski (1989), is a process-oriented appmach that encourages problem formulation and problem solving within thematically tied unit, of instruction. Yet the success of students' learning dopends on the richness of data presented in the cases (Gragg, 1940; Learned, 1987) and on supporting instruction that encourages students to examine cases from multiple points of view. A goal that is common to both anchored instruction and theme-based case analysis instruction is to provide rich contexts for learning so that the content can be studied for different reasons. Such cross-examination enhances problem-solving and cohesive learning (J. Pichert, personal communication, April 3, 1989). The principles presented in this paper are intended to be guides for those wishing 13 2 *kni**106904.1#stIRctiodViirOitc4.,A4thotig4i4eFi' illuch.00049.ke pripOipitt will oftect cA*. instiuctional 'se** 04 mied it** how Oese atichOisi Wo aie bstitiolog to see the "pOteatitil tfrot4ti4g, heir anchors in additional curriculum **is. pripCiile`4:60§d'OovO,44::lio OontiniiailkioAir4:***4jps#4004,4,! at- 1teped1ñ OoniaiPs; REFtkENCES ,Alicen,I. (1963). Wolves of willoughby chase.:New York; bonbleday. .AnderSon, I. R. (1983). The architecture of cognition. Cambridge, Mk IlaiVird Puivi*SitYttcsa_x. Anderson F R. (1987). Carnegie-Mellon Universiti, skill aequisidom Corunilition of Weak.:*dind !Cm solutions. Pry_chologiced Revkw, 94,-192-210: Andersen, R. C., & Pearson, P. p. (1984): A :chtheoretic view of proceiles reading corripiebepi* In P. D. Pearson, R. Barr, id. L. Kano, '& P. Moscrib.11 (Eds.). Handi?Pok of rfakti ke:r*Tiqi (pp. 255-291). New ,Yorkt Longman. Rhko, V., Rowe,. -Bransford; & Yr, N. (1989). Designing lari,iatiorii th thinking:' Some initial thiughts. In S. McComick & J. iutell itcrs.),,Calidth4 difd'iaekl perspèéth'ifor literacy research and instniction (pp. 35-54). Chicago: National Reading Conkririe. Bransford, J. D., Franks, J. J., Vye, N. J., & Sherwood, R. D. (1989). New. iiipiOaehei th-pistiuction: Because wisdom can't be told. In S. Vosniadou & A. Ortony (Eds.), Similarity c.nd aiinlogkal reasoning (pp. 470-497). New York: Cambridge University Press. Pransford, J. D., Sherwood, R. S., Hasselbring, T. S., Kinzer, C. K., & Williams, S. M. (in-press). Anchored instniction: Why we need it and how techntilogy can help. In D. Nix & R. Spina (Eds.). Advances in compsit--r-video technology. Hillsdale. NJ: Lawrence Edbaum Associates. Bransford, J., Vye, N., Kinzer, C., Risko, V. (1990). Teaching thinking and content knowkdge: 'Toward an integrated approach. In B. F. Jones & L Idol (Eds.), Dimensions of thinking and cognitive instruction (pp. 381-413). Hillsdale, NJ: Edbaum. Chapman, R. S. (1978), Comprehension strategies in children. In J. Kavanagh & W. Strange (Eds.), Speech and language in the laboratory school, and clinic .(pp. 308-327). Cambridge, Mk MIT Press. Christensen, C. R. (1987). Teaching with cases at the Harvard Business School, In C. R. Christensen (Ed.), Teaching and the case method (pp. 264-270). Boston: Harvard 1.31,inets School. Dewey, J. (1933). How we think. Restatemens of the relation of reflective thinking to the e&scative process. Boston: Heath. Durkin, D. (1984). Is there a match between what elementary teachers do and what basal reader manuals recommend? The Reading Teacher, 37, 734-744. Gregg, C. I. (1940). Teachers aiso must !earn. Hantard Educational Review, 10, 30-47. Hanson, N. R. (1970). A picture theory of theory meaning. In R. G. Colodny (Ed.), The nature and function of scientific theories (pp. 233-274). Pittsburgh: Univasity of Pittsburgh Press. Kinzer, C., Risko, V., Vye, N., & Sherwood, R. (1988, April). Macrocotuests for enhancing vocabulary acquisition. Pape: presented at the meeting of the American Educational Research Association, New Orleans, LA. Lean, D. (Director). (1948). Oliver Twist (Videodisc). Los Angeles: Pararrount. Learned, E. P. (1987). Reflections of a case method teacher. In C. R. Cr seri (EQ.), Teaching and the case method (pp. 9-15). Bostro: Harvard Business School. Rasinsld, T. (1989). The ease method approach in reading education. Reading Horizons, 30, 5-14. Risko, V. J., Kinzer, C. K., Goodman, 1., McLarty, K., Dupree, A., & Martin, H. (1989, April). Effects of macrocontexts en reading comprehension, composition of stories, and vocabulary devekpment. Paper presented at the meeting of the American Educations! Research Association, Srn Francisco, CA. T *--'0**W(1955):1/4#.."4. $04--asid inoti *apt(o4.4_, e-'40pitit* mos.00**, , _ - , - , tiVtg:cTS OF OROUth40 ANtrOlOiOLTt tot SillA7E144:',01: REPKO-04**W iti1040:14 .1414 OhöS 0 til'044**111.s,,er Nopervilleillinots Schookspurnct No. 203 " ,, ,Reading instruction is organized frequently.*tindsinall,JuiniogenConsi*ps '0( s414nts. Thc `01:00 .Y041 of-ti#iprq00,tir, PiO'0841,4'',11#94-4000S4984:1984; Hiebert,j1983).,444-(00540:fiki, number of investigatnrs 1,18.geSt41' 0.4 =itP4e44.4,14i-iiii: groi1P,.FPOW6.ii400.s..104.1**** -t,00b4cic whcn colPPAr.ed,to th6 st4dOts in111? 4h4PilitYriti.,15.0Ps',:q0:retki4,?4##2,11.(198:4) conelude, that homogeneous grouping hit few desirable COnsequencei fOr low- ability students. Eldredge andBUtterEeld (1984)idernOnstrated that;snidentsishowed no -differences in achievement when instructed in beterogineons,grOups: tither:I* traditional batal ability groups. Slavin's (1987) review of the literature found no,qualifying studies that directly cdinparedinstzuction insinall:abilitigrouptwithin classes with students receiving whole-class instruction hi nariabiEty:grOUPed dates. In a comprehensive review of classr -irn organization, Barr (1989) sugg*d.that although there is differential treatment of students in reading grotgps,.thereis,nOt necessarily a consistently increasing difference in achievement as studenti ,progress through the grades. Rather, there scetne4o be a constant difference between student. grade level and placement in materials after' first grade. An important variable that has been widely ,neglected, in these stud:is the discrepancy between students' placements in materials and theiractialoadinglevels. This placement relies on determining a, student-_"instructional levet" -ThenOtion unnalying instuctional level is that there is an optimal leyel of difficulty.öf Materiels for instruction to be effective: If there is 'any validity-to the instruCtinn-alleVel" concept, placement could be a critical variable.,If materials are MO-easy, 'learning might-be imperied.-11 materials are too difficult,.learning woiddproceed top sle*IY, if atall.. (See Shanahan, 1983 for a discussion,of,the concept of instru,etional 164). Instructional level should interact with- abilitY grotipintWithin.claSsroon*when tuction 'That ist stUdents-ln,,0446.0ilitythan is _only a -tingle text used .for groups:should have smaller discrepancies betweeir their texPitleYela end ,thelevel, of difficulty ofthe materials. Students in lower ability gronps:shOulrl Ond the materials, relatively moredifficult than other sMdents.,Shitlenti in the high groups shouldhaYe :very easy Materials relative to their readinglevels. :ft , 122 Literacy Theoiy and Resemr: Barr and Dreeben (1983) suggest that instructional pacing may be a dominant: factor in achievement. Pacing should, of course, be related to the difficulty of the:: materials used for instruction. With faster pacing, materials would effectively beco*_ more difficult. There is conflicting opinion on whether the raateriali used for instriic7,-,x:., tion should be relatively easy or difficult. However, none of the research studie.i: grouping have systematiCally exanined this variable. In the present study, the difficulty variable was manipulated byassuiningjhaI materials that were too easy would produce less leaming-than-materia1s4aLlwere': more difficult. Since there is an ethicalproblem in asking students to s`perid a yegmaterials that might not produce learning, below grade level materials werez.:**cl: inappropriate. Consequently it was decided to manipulate this variable bybsingiiiaterit als that were on or above assigned gade level. Although this solution does not:allow for all possible comparisons, it produces a range of discrepancies across ability groups, since high group students are using materials that are relatively much "easier" than the materials that low group students are using. The effects, if present, should show up as an interaction. In summary, the following study examined the effects of two variables: small, ability group instruction versus whole-group instruction and difficulty of instructional materials compared to measured reading ability for students in the third, fourth and fifth grades. PROCEDURE Thirty seven classrooms were solicited on a volunteer basis in a single suburban school district The dominant instructional pattern in the district was tracliticcally ability grouping Teachers were assigned to one of two conditions. Traditional groups or whole-class instruction. Within the whole-class instruction variable, teachers were either assigned to on-level materials or to above-level materials, the matting materials used by the teachers were either the basai readers for the grade level or for one grade level above. There were 8 third-grade classes, 12 fourth-grade classes and 17 fifth-grade classes El-wen classes were traditional instruction, 11 were whole-group on-level instruction, and 14 classes were whole-group above level. (One fourth-grade class was allowed to use traditional instruction with above-level materials. These students are not included in the statistical analyses. Their scores are included in Table 1 as a matter of interest.) A total of 869 students participated in the study. All students were given an informal reading inventory (Burns & Roe, 1985) during the first and second weeks of school. These were administered by paid assistants Students were also given the reading portion of the Iowa Test of Basic Skills (ITBS) (1986) which was scored for Fall administration. Students were also given the ITBS at the normal Spring administration. All ITBS scores are reported in percentiles. (In addition, students were asked to keep daily logs of reading they did outside of school Teachers were asked to keep logs of the amount of material they covered daily. These data are not reported in the following discussion.) Because of presentation formats, the amount of time required for primary instruc- 136 1 mg:anditeadmg Nchievement ;Table 1 iiticablilmy and Comprehension Gains by Grade, Instructional Fonnati,and Materials_ 1." Ity N 69 0 75 74 Total 218 4 95 27 115 Total 70 307 5 107 0 87 150 Total 344 Grand Total 869 VobuIai. Class Type Mat. Diff. Pre Pos3 Trad Trad Whole Whole On Above On Above 66.30 7140 5.:10 59.00 60.00 63.40 57.30 71.60 7.30 8.20 6.87 55`.311 6-1.0 674- Trad Trad Whole Whole On Above On Above 65.60 71.70 61.40 60.70 70.00 74.70 64.60 62.60 4.40 3.00 3.20 64.10 62.90 57.30 59.60 68.50 4:40 68:30 5.40 64.10 6.80 On 72.70 69.90 -2.80 67.10 68.30 l'.20 70.10 69.00 66.90 66.80 -3.20 65.40 72.30 6.90 -2.20 63.50 67.50 4.00 4.03 -2.73 Trad Trad Whole Whole Above On Above 1.90 3.13 68367'19s" 65.00 - 540 5.50 tion in the whole-group conditions was less tha-- the full time required in the traditional conditions. (Since teachers have to present only one lesson instead of three, there is a great deal of difference in the instructional "overhead" time.) In irckT to attempt to equate for the amount of time spent in reading and reading-related instruction, a literature program was included with the whole-group classes and was also used to supplement the basal reading materials. No formal measures of the effect of this program addition were taken. RESULTS Each student was assigned a reading level for word recognitivn and one for comprehension based on the results of the informal reading inventory The placement discrepancy for each student was calculated by subtracting the grade level from the IRI results. Table 1 shows the vocabulary and comprehension gains as a function of treatment conditions. grade, instrustional format (small groups or whole classes) and materials difficulty (on- or above-grade level). Table 2 shows the breakdown of scores as a function of ability (high, medium, or low), determined by a three-way split of each f.3 'NA.' -'11t. - '', , ,:',.. k:' ". '. ' '' ''''L ,,,,,,, .'f': ,;,°,1";%, 4VO4,41,1?-e.90,, :, : '' '',.,;'T,/,'. ains as crFunction of Ability on Vocabulary, InspViOnat tormat. L 3 4 4 19 21 25 21 23 24 29 34 32 37 38 Whole Whole Trad On Above On H, M, L H M L H Whole On Above H M 5 Trad On H M -4.40 -0.20 Above 19.60 L H L H H M -14.27 - 9.60 L -5.20 -1.13 38.10 4.74 12.67 2.11 -2.50. 240 13.90 -1.69 9.50 4.00 2.40 3.00 2.60 10.10 -6.20 -6.20 -8.80 -4.90 L Whole _ -8.50 -8.20 M Averages: On Vocab Trad 6'.40,Y .1.66';- -10.00 -8.00 M Whole ,..., ,' ,-,-,:s '33.50- L On c '.:tvoi0;21,..,' 13.50 Whole Whole ,, -8.00 110 L 35 33 , 1,. ..84_, =7.40, M 19 5 1-444 18:00 24 27 39 29 29 50 42 58 21.902 L H 4 29 :3:01Y--,:' M 40 5 11,1i9V1, 5.50 -, 0 '''- ,;:-1- ,, ::'',4 7.-60 , . : :-,'- 0.70 1.20 9.10 ,. Comp Trad 1.86 6.40 8.00 5.42 Whole 3.43 4.87 11.47 6.59 Above Vocab Whole -7.73 -6.77 19.53 1.68 Comp 0.67 2.47 13.50 5.54 Overall Vocab -9.07 -5.83 25.43 Comp 1.99 4.58 10.99 A 1 . ;0 group on the basis of the vocabulary pretest. Table 3 presents a similar breakdown on the basis of tho comprehension pretests. Two analyses sf covariances were performed for the vocabulary 'scores and-lhe, comprehension scores. gach analysis use4 the.posttest scom as the depencleatyarialge and removed the effects of the pretest scor:es and dm, discrepancy leyel*thoseading, material with grade placement. (AlthOu-gko ain IMOros am:repotted In thet table,:theY, r _ r, Rhteenz - 614 leading Gains.aS a Fano., ,n of Abilityon Compreheniion, InstruCtiaiiar_kaimaf, z , Akaterials Difficulty MIONNIONNEM tClass 11 26 .'",, type Mid Dift On 16 23 24 Whole On Whole Above 17 30 28 22 23 : . 4 33 Trad On 4 28 30 34 Whole On Ach. Yee-***.Y,'.,,C41.1!" . It ' M L H M L H M L H M L 40 29 33 25 25 28 49 Trad Whole On On Above 36 60 Averages: On Vocab Trad -10.40 - 1.70 18.00 '5:80: 2.80.- 0.74 L. 6.60 15.80 H - 6.10 M L H M 0.70 2.00 - 1.30 11.21 4.50 16.50 -2.30 -6.90 3.64 8.30 -1.20 16.43 H M - 2.46 - 3.60 - 3.80 - 3.50 L 0.10 11.80 L Whole 6.90 '4.40 4.60 11.10 0.10 11.20 16.70 10.30 5.30 ::-'-.:A..:, ,, - 5.70 44 L Above -..: M M Whole 110 H 33 19 4,4::,40,C.!:-: C-901:*-- 2.10 1.90 8.00 3.50 H 23 26 . `L 5.40 3.30 Above Vocab Whole Comp Trad Whole Whole. Comp 0.00 0.57 7.80 2.79 -5.40 H M - 1,47 3.38 - 6.03 - 2.13 10.27 - 1.87 3.77 3.60 L 5.03 4.61 5.55 2.35 19.50 5,75 6.85 : 3.28 18.q7 Overall Vocab 0.64 2.99 6.13 Comp - 3.57 3.57 19.02 5.42 were not used in the analyses to avoid the problems of compounding unreliabilities when using gain scores.) The independent variables were treatment group (traditional, whole group/on level, and whole group/above level), grade (3, 4, and 5) and either comprehension or vocabulary abilty (high, r.edium, or lor). For the comprehension scores, both covariates were significant, but none of the main-effects wr-e significant. The two-way_interqction of Grade x Comprehension fe. -.. ,A Ability was significant, F(4, 813)=4.56, p<.001. For .the Vocabulaty,scores,, covatiates were again significant. The main effect for Vocabulary,ahility Was, significant, F(2, 813),= 7.627, p<.001. The two-way interactirnifof TreatMeifyie yoCabn:,,, lary Ability was also significant, F(4, 813)=2.922. p.5.02: 'No iitlicteitectS Wtre:, statistically sifiniOcant in either analysis. A regression analysis was-f,conducted with poittest comprehenSiopseoteS dependentvariable. Pretest scores for compiehension,:,yield ati-A2=44:20 **fi. 011: tered 'alone in a stepwise, regressiOn;-Wthen clisCrepanaYlit;cOmprehensioh;10,'materials is entered, R2= 45089. The other Variables did hot sign amounts of variance. When grade, treatment condition and cotripiehefiSinii,IS were entered, R2= .45189. A similar analysis for vocabulary posttest scores yielded an R2= .35980;When the vocabulary pretest was entered; when discrepancy level was added to the:equa? tion, R2= .37937. Grade also significantly increased the variance accounted R2= .38310. DISCUSSION The results suggest that there is no osential difference between instruction patterned on a whole-group model compared to the traditional homogeneous abilitY groups when the outcome measures are standardized achievement tests. Although there is an interaction between treatmetA and vccabulary ability an the outcome measures, this effect is due to a nonlinear differeace in only one cell and is not interpreted further. The analyses also suggest that the discrepancy of the materials from the measured reading ability levels of the students accounts for a significant amount of variance in the scores. This suggests that a great deal of attention be given to the placement of students in materials, regardless of their ability or the instructionai organization of the class. The variables accounting for the most variance in the posttest measures were the pretest measures. For vocabulary, discrepancy and grade level also accounted for significant variance. The simplest explanation for this is that there is too much variance within classrooms for the grouping pattern to have much of an impact. That is, in the comprehension analysis other variables contribute more variance and obscure potential effects of grouping and materials difficulty. Note, however, that this is not true for the vocabulary analyses, v. hPre materials difficulty does contribute significantly to the variance. Milky does present the strongest effects in the analyses. That is, poor readers gain mote than good readers. This can be explained, in part, by ceiling and floor effects, as well as regression to the mean; all of the students can be expected to score closer to the mean on the posttest. This translates into gains for the poor and losses for the good readers Althctugh these gains rnay be explainable as artifacts, they should not be ignored. On the basis of this study, two conclusions seem warranted. Instruction in small groups compared to whole classes was not superior for the students in this study. pin* thi4Reopig Achievement consequently, any potential social benefits of whole-group instruction would suggest that -whole-group instruction should be preferred to small ability- groups. -Difficult Materials seem to have produced higher scores. The instrictional implication is that , when choices are to be made for instruction, more difficult materials shOulii be seleeted #ther,than less difficult.materials. Several cautions should be observed in generalizing these conclusions. Firit;_these subjects are not beginning readers. Students in primary grades tint or secondi:Msy -. have different needs and may respond differently to instruetion and, ortinkPatterns. The present study did not address this issue; the conclusions fror*hiS stuy must be limited to upper elementary students. Second, these findings are'liiiite4 by the outcome measures used in the analyses. The exclusive use of standiirriliedleit scores severely restricts the generality of the results. Finally, the type of teadiiiig was not closely controlled. There may have been a great deal of "hybrid" teaching occurring. The teachers were not exclusively trained nor closely monitored. More research is needed to control these possibilities. REFERENCES Allington, R. (1983). The reading instruction provided readers of differing reading ability Elementary School Journal, 83, 548-559. Allington, R. (1984). Content coverage and contextual reading in reading groups Journal of Reading Behavior, 16, 85-96. Barr, R. (1989), The social organization of literacy. In S. McCormick & I. Zutell (Eds ), Cognitive and social perspectives for literacy research and instruction (pp. 19-33). Chicago National Reading Conference. Barr, R., & Dreeben, R. (1983). H. -hoots work. Chicago: University of Chicago Press. Burns, ?., & Roe, B. (1985). Informal Reading Inventory (2nd ed.). Boston: Houghton Mifflin. Eldredge, I. L., & Butterfield, D. (1984). Sacred cows make good hamburger. A report on a reading research project titled Testing the Sacred Cows Ili Reading." Provo, UT Brigham Young University (ERIC Document Reproduction Service No. ED 255 861) Good, T. L., & Marshall, S. (1934). Do students learn more in heterogeneous or homogeneous groups/ In P. L. Peterson, L. C. Wilkinson, & M. Hallinan (Eds ), The social context of instruction (pp 15-37). New York: Academic Press. Hiebert, E. H. (1983). An examination of ability grouping for r....aling insouction Reading Research Quarterly, 18, 231-255. Iowa Test of Basic Skills (1986). Chicago: Riverside. Shanahan, T. (1983). A cnnque of P. A. Killgallon's stui. y. A audy of relationships among certain pupii adjustments in reading situations. In L. Genale, M. Kama!, & I Blanchnd (Eds ), Reading research revisited (pp. 577-582). OH: Merrill. Slavin, R. E. k1987). Ability grouping. A best-evidence synthesis Review of Educational Research 57, 293-336. o'r AIN§ITIONAL READING INSTRUCPQN IN-SCOCE GRADES: BguEt z*Ots:p:AcTIO - Ronnie Konopak 'NAST; C011:61!. Louisiana Stare Univers 5ty Ellen Jampole SUNY Cortland Indiana-Purdue, UnfveriitYrit Port Wayne, ,Mar,* hëII CentenaryColiege Rita Dean, Ii.ennie Bolomen, Mary Eve!rett, Nancy Weems, and -Leslie -ArCentaux Louisiana State University According to Chall (1983), the transition from elementary to middle:school 04* a developmental break between learning to read (acquisition and flueney cis, leading, skills) and reading to learn (use of sIdlls as tools for learning content) .cAlthnngh'the' assumption had been that students would automatically ttansfer their neWly aciPthed skills to expository text trading, research (e.g., Calfee & Curley; 190) haS,^shoWn that bask: word recognition and comprehension skills are not sufficient for:stndents tn_ cope with more complex materials and reading tasks. As a contotience,:r.ontiar teachers are urged to consider both the reading difficulty of their textsnfid the reading sophistication of the.- students, and to provide guidance where the need is eVident (Vacca & Vacca, 1989). However, although research attention to content reading has increased in thepast 10 years, its goals and instructional practices have yet tc win wide acceptance by content teachers (O'Brien, 1988). Observational studies at the elementary le'vel'(buikin, 1978-79, 1984) and at th4 secondary level (Hinchmann, 1987; Ratekin, Simpson, Alvermann, & Dishner, 1985; Smith & Feathers, 1983b) have shown an initiuctozial emphasis on unguided, isolated practice, with text reading used as 'nuns to "cover" the subject content. Further, survey/interview studies at the secondary level (Meek, 1977; Smith & Feathers, 1983a) have found that neither teachers nor students involved in such activities view reading as an important learning tool. The present study attempted to extend previous research findings by investigating two areas which had previously been unexamined. First, the researchers were intereste(' in addressing the transition from elementary to middle school learning. The researchers were particularly interested in this transition because of differences by grade levet in teachers and program emphases. That is, elementary schooLteachers are usually generalists and instruct in self-contained classrooms, while middle school teachers are content specialists and instruct in departmentaliped prograths. In thit study, the researchers chose to focus on two grade levels--4fth and sixththat repro- 130 Literacy Theory and Researc, sent the transition from upper elementary to middle school in the region where the' study was conducted. Second, the researchers chose science as a content area to examine due to stu-, dents' reliance on text materials for learning but difficulties in comprehension and learning (Lloyd & Mitchell, 1989). Further, according to the 1986 1,1AEP Science Report Card, "distressingly low" :;chievement scores were found for 3rd-, 7th-, and: llth-grade students who had been tested. In addition, students at all three grade leyels reported not taking a science class that year, and that for many who did take.a.class;.. instructional time appeared to be limited to and dominated by teacherleetures ,andtextbook reading. Based on these findings, the researchers were interested in examin- . ing 1.Av reading was considered by teachers in science at the transitional grade levels. In addition to grade level and content subject, the prestnt study also differed from previous studies in its theoretical framework. Although earlier studies (e.g., Durkin, 1984; Ratekin, Simpson, Alvermann, & Dishner, 1985) had compared research-based instructional recommendations with actual classroom practices, this study also exnlored the relationship between teachers' beliefs about reading for content learning and how those beliefs were realized during iostruction. This framework, adapted from Clark and Petteson (1986), involves two related domains. (a) teachers' though!. processes, including theories and beliefs and their relation to teacher decision making; and (b) teachers' actRms and effects, including both teacher and stndent behaviors, as well as student achievement. The premise underlying this framework is that there is a reciprocal relationship between the two domains, that is, teachers' actions are largely the result of their thought processes, which in turn affect teachers' subsequent actions. In this study, the focus was on the relations among beliefs about reading for science learning, lesson planning, and subsequero instructional activities. Specifically, then, the purpose of this study was to examine how teacher beliefs and practices would vary according to th elementary or middle school roles in a reportedly difficult subject area The general question guiding this investigation was, "How is reading considered and used by teachers of science at the transitional grades"" In particular, five important components of instructional practice were examined within this broad question. Adapted from Ratekin et al.'s (1985) study, these' components included (a) lesson purpose, (b) class organization, (c) instructional materials, (d) instructional activities, and (e) evaluation. Based on instructional recommendations made by content researchers (e.g., Readencc, Bean, & Baldwin, 1989), these components became a basis for observation to the Ratekin et al. study. In the present investigation these components not only guided observations but also were a hasis for post-observation interviews with teachen and rtatdomly selected students. METHOD Participants Participants were two fifth-grade teachers, each with one seif-containeu elass, and two sixth-grade science teachers, each with one randomly selected departmentalized class One fifth-grade teacher and the two sixth-grade teachers taught at different ,143 - _ , 'atm-Ft:on . Tablel enat Eadcground of Teachers by Grade-and Cias's Yfchig Ttxkti Pba-` -ye** it. Prese-nr Grade 1:00. 28 4 26 16 1 Highest Ikgre-- Held MoSt Reicnt Rtading Ed.S. B.A. M.S. Metlyridi Course Menibership in Professional Organizations Subscription to Professional Journals Cuniculum Determinant Work with a Reading Specialist 1978 1978 1978 1963 yes yes Yes YPS' Yes Yes .: yes Yes school state state State (self) no yes yeS public schools, whereas the second fifth-grade teacher taught at a university taboratcry school. The lab school was deliberately chosen by the researchers asits',prinurt teaching/learning environment; the Public schooli.weie. purpose is to provide a randomly selected from elementary and middle schools in the, sante The four classes were similar in that they were heterogeneously grouped and Coniiitedl of 25-35 students. Prior to observations, each teacher was asked to provide professicmal dembiraPhie: information, as well as her beliefs and practices on reading and learning in the kience classroom. Table 1 presents the professional demographic informatiorit'Table4:presents the beliefs and practices on reading for science learning. As can.be Seen, the:: four teachers were similar in their interest in professional development but differed in their educational backgrounds and teaching experiences. In addition, -although ;all believed in a variety of experiences during the learning process, they varied in their beliefs about the purpose and role of reading in the science classroom. _ Materials and Procedure Two researchers per class attended science lessons for 5 consecutive teaching days that constituted a complete science unit or subunit. Before the observations toOk place, each research team held interviews with the classroom teacher on her professional background and beliefs and practices. In addition, tharesearchers and teachers discussed the objectives and content of the upcoming lessons. During the observations, the researchers Wrote field notes, andiotaped the lessons, and examined pertinent materials. Generally, these materials included lesson plans. (with both teacher and school/state componentsj, textbooks-and Worksheets, SupPOrting aids, student reports and projects, tests and quizzes, and any manipulative§ uted Table 2 Beliefs and Practices Pertaining to Reading.itr Science by Grads and Class Ret4ing;Cpmponents Role of Readhig in Science Learning Determine Student Reading Levels Determine Text Level Difficulty Accommodate Student Differences Prepare to Study New Unit Prepare to Read New Unit Monitor Students' Learning Follow-up Students' Learning Evaluation of Students' Learning Flexibility in Decisionmaking *Oil* -5tlia"ba, develops'icitowleilge'of content and relations achievement, other tests 6P01.1?ficl' gives.informatiiin givesiüferinaiion oral reading, tiists reudini0eacher atittiulateaf ideas expoienee Fry Readability Formula student grcupings access prior knowledge, SQ3R study vocabulary in context observe, quiz 6th!PP:b2 gives infiiiiitation student failure to comprehend different materials/ methods new vocabulary, projects reading teacher, informal tests expedenee "comes leveled" variety of experiences oral readings, short readings study guide study vocabulmy in context quiz, study guides motivate, fun tasks relate to students partictpation, homework questions, quiz variety; writing variety; hands-on variety; skill building variety; labs variety vrxiety variety variety some some little/none enough 145 'tor lab activities. The researchers did not participate in ckiss lessons and collected their data as-nnebtiiitively as-possible. Follo,wing all observations, each research; tc. am conducted' scparati-: inteiviews with ihe' teacher and three randothly seletted-ritUdcnts%on,,theirp'e-icePtiona:OP4he immediatelY- preceding unit The interviewa--Were striicturod4rotind,*fiVe:InStfU07 ,tional'eoirt7onents (i.e.lesson purpose, claSS'Organizatioti,-inateritkaCtiVities,::and evaluation) and how they were actualized' in,- class. IiLaddition.,:-#*.teactieisziiikii students were asked to respond to what they felt were conitraints an teaching/learning: in the classroom. Data collection resulted in 15 hours of observations andliitcOieWS per class. Data Analysis Data sources included (a) pre-observation teacher interviews on profe-ssional background, beliefs and practices, and proposed lessons; (b) observations of 5 consecutive teaching days, including field notes and audiotapes; (c) materials used in conjunction with instruction; and (d) post-observation interviews with teachers and students. All data were analyzed using Miles and Huberman's (1984) concurrent flows of analysis: data reduction, data display, and conclusion drawing and verification. In this process the data were examined recursively and displayed in unordered and ordered matrices that were driven by 0.e research question. Data sources were triangulated to validate an occurrence and to control for biases from _other sources. Fmal-interpretation was achieved following searches for meaningful patterns across, between, and within classes, involving the multiple perspectives of the whole research team. RESULTS AND DISCUSSION From the analysis, stonger patterns emerged across and within grade levels, rathcr than between grade levels. Generally, althoubh reading was deemed important, planned, and used for study, reading instruction was not necessarily included. What instrucOon was observed usually dealt more with isolated skills than with internalization of subject content. Further, when examining the five instructional components, the following patterns appeared. (a) lesson purposes were both explicit and implicit; (b) organization was primarily whole class, (c) materials included textbooks, worksheets, and, for the fifth-grade classes, manipulatives, (d) activities included lecture with recitation disch individual seatwork, and (e) evaluation varied from worksheets for one fifth- (public.) and une sixth- (public-2) grade class, to worksheets, observations, and projects for the other fifth- (lab) and sixth- (public-1) grade classes. In addition, when ordered by site, the teachers were ranked from 1 (highest) to 4 (lowest) on effecti.e instruction. This included attention to the five instructional components, as well as to the students' learning. The fifth-grade lab teacher was ranked highest, followed by a sixth-grade (public-1), fifth-grade (public), and sixthgrade (public-2) teacher, respectively. As described in Table 3, the fifth-grade teacher who was ranked highest used a variety of settings and methods, addressed readiness and independence, and relied on multiple evaluations. The lowest ranked sixth-grade 10. 131 dion Pradices of Teachers kinkedli Highl 4isk4loW): ces Varietyin .E.i.ring/MethodS: 74i'lier415,011*1 gre***0: whale ki#Itgiouo dithuiskies effective US-a ofA/V tinatinfals Addressed Readiness/ Independence Multiple Evaluations soar:via-dotting strategies encouraged readiness! independence: included supplementary readings shared independent projects Itrffic44, *Of reqiii041111*/**4 000* e*,` Puitictie.. q**-1,0#64** or 'Mx' activities OOMpleted as whc4 provided live specimens no tii*t90.4 of.-,in0.1vi4a! compared field notes little organizilion- in prepared organizational charts used variety of evaluations; limited Use of reading and. resources used limited variety of evaluations: worksheets tests observation questioning worksheets art independent projects tests Pro0r4 plan.ning teacher, on the other hand, demonstrated little variety in the five components and neither prepr.red for nor monitored students' learning. Prior to observations, interview data indicated that three teachers focused on thinking and reading skills in science, such as classifying, observing, and inferencing. The fourth teacher (sixth/public-2), however, felt constrained by state guidelines to teaching process skills in isolation. In the post-observation interviews, all four teachers stated that they focused on concepts, process skills, and development of ideas. They felt that their most effective activities were concrete experiences such as labs and creative projects, as well as games and worksheets. All reported that these activities were appropriate in helping students think about and learn new information. in addition, class organization was limited to whole-group leo= and independent work, with class discussion as the focal activity. Finally, three of the four teachms felt that their instruction was most adversely affected by students with limited skills, tne fourth teacher (fifth/lab) was most affected by outside visitors. Students' post-observation interview responses both agreed and contrasted with the teachers' responses. All students stated that they had understood the purpose of " Reading Instruction in Science 135 the week's discussion, although few were able to articulate the specifics of the lesson. All thought that they had learned the material well, even though several had not performed well on cad-of-unit assessment tasks. In addition, content and activities consideted important by these students included information relevant to their own lives. This included a fifth-grade (lab) unit on classifying animals which the students were able to relate to their own pets, and a sixth-grade (public-1) unit on drug abuse which the students related to their own family and friends. Material not considered relevant included a fifth-grade (public) unit on cell division and a sixth-grade (public2) unit on optical illusions. Finally, student misbehavior was cited as the major catiSa of interference to their classroom learning. Given these findings, the four teachers appealed to vary in their consistency between what they stated as their beliefs and ideal practices and what their plans and insttuction actually revealed. The fifth-grade lab school teacher showed a more consistent relationship than the other three teachers. As described in Table 2, she defined reading as a method of developing contcnt and relationship knowledge ana implemented materials and activities that generally supported this belief. Given her extensive teaching background and educational specialist status, this result is not surprising. Of interest was the sixth-grade (public-1) teacher ranked second; she also bellowed in reading as a learning tool and that guidance was necessary to facilitate student learning. Her instruction was constrained, however, by her strict adherence to school/state objectives and content, aptly demonstrated by her inititling and dating her curriculum guide when completing requited ob; :tives. The fifth-grade (public) teacher ranked third had a narrow view of reading for science learning, as well as the role of reading instruction, and limited teaching expenence. however, she attempted to facilitate learning through group work and hands-on experiences. The sixth-grade (public-2) teachei ranked fourth also had a narrow v lew of reading and reading instruction in thc science classroom. Although preand post -observauon interv le% data indicated that she felt her instructional decisions.' activities wen.. facilitative and appropnate, her actual practice did not support her statements. This investigation indicates that, to some extent, teachers thoughts and beliefs arc critical and important aspects of their instructional effectiveness, thus supporting previous researth (Duffy & Ball, 1986). Specifically, teachers plan and implement their decisions based on aleir view of the role of reading for science lean_ 1g and how reading instruction should be carned out thioughout the learning process. As pointed out by Clark and Peterson (1986). teachers' thoughts and actions can be profoundly affected by the task demands, as well as by the teachers' perceptions of these demands. Further, teachers may have greater or lesser opportunity to decide on and implement their beliefs, tor example, teachers may be allowed less flexibility in planning due to camcula decisions made by othet sources (school:district), as evidenced by one sixth grade (public 1) teacher. Overall, as demonstra.ed in this study, teachers' thoughts on instruction, learning and reflective analysis following a lesson are critical when examining the classroom environment (Readence, McGee, & Konopak, 1989) Although this research is limited by its brev ..y of observational period and the use only four teachers and classrooms, the resalts generally confirm earlier reseiach f...dings on the lack of reading instruction in a content subject. However, when ordered LiteracY Theory and it citaith'3 1 by site,_grenter distinctions are defined; again, the fifth-grade lab school teacher dein- onstrated more effective reading instruction than did thalticgett-rahlrecLsbithlride (public-2) teacher. Recommendations for future research include extending the nbsert vation period,,probing for mom information concerning'.beiiefs,and:idea1',practieeS and examining other content areas for Comparison. In.addition,--aldinngh'ilr,presentstudy Focused on research-based recommendations from reading- speeialistv, Tut*: research could address recommendations from specialists in the content field: REFERENCES Calfee. R. C. . & Cur' R. (1984). Stnicture o reading comprehension (pp. 161-180). Ne rat in content areas. In J. Flood (Ed.), Understai.ding rk, DE: International Reading Association. Chall, I. (1983). Stages ojrng developmen. v York: McGraw-Hill. Clark, C M . & Peterson, P L (1986). Teachei. thought processes. In M. C. Wittroek (Ed.), Handbook of research on teaching (pp. 255-295). New York: Macmillan. Duffy, 0 0 , & Ball, D L. (1986). Instructional decision-making and reading teacher effectiveness. In V Hoffman (Ed.), EffectiPe teaching of reading research and practice (pp. 163-180). Newark, DE: International Reading Association. Durkin, D (1978-79). What classroom observations reveal about rerting comprehension. Reading Research Quarterly. 4, 482-530. Durkin. D (1984) Is there a match between what elementary teachers do and what basal reader manuals resommend? The Reading Teacher. 37. 734-744. Hinchmann. K (1987) The textbook and three content area teachers. Reading Research and Instruction. 26. 247-263. Uoyd. C V & Mitchel. 3 N (1989) Coping with too many concepts in science texts. Journal of Reading. 32. 542-545. Miles. M B & Haberman. A M (1984). Qualitative data analysis. Beverly Hills, CA. Sage. NAEP (1986) Science report card Denver, CO. National Assessment of Educational Progress. O'Brien, D (1988) Secondary preservice teachers' resistance to content reading instrucuon. A proposal for a btuadtr perspective In I E Readence & R. S. Baldwin (Eds.), Dialogues in literacy research (pp. 237-244). Chicago, IL: Nation.). Reading Conference. Ratekin, N . Simpson. M , Alvermann. D.. & Dishnex. E. (1985). Why teachers resist content reading instruction. Journal of Reading, 28. 432-437, Readence. Bean. T . & Baldwin. R S (1989) Content area reading (3rd ed.). Dubuque, IA. Kendallr Hunt. Readence, I . McGee. L . & Konopak, 13 (1989, April). Teachers' beliefs and instructional practices in literacy Paper pi -rued at the Ninth Conference on Readh g Research (CORR9). The Center for the Study of Reading. New Orleans. LA. Rieck 13 3 (1977) How content teachers telegraph messages against reading. Journal of Reading. 20, 646-648. Smith F , & Feathers. K (1983a) Teacher and student Kreepoons of content area reading. Journal of Reading. 26. 348-354. Smith F . & Feathers. K (1983b) The role of reading in content classrooms. Assumptions vs. reality. Journal of Reading. 27. 262-267. VaCCa, R T , & Vices. I A (1989) Content area reading (3rd ed.). Glenview. IL Scott. Foreman. 149 THE EFiek.CTS OF STRUCTURAL FACTORS OF EXPOSITORY TEXTS ON TEACHERS' JUDGMENTS OF WRITING QUALITY' Beverly E. Cox Purdue University There is strong linguistic support for the notion that hierarchical organization and cohesive harmony represent stable variables that partly differentiatz well from poorly written text in English (Cox, Shanahan, & Sulzby, 1990; Grimes, 1972; Halliday, 1985; Hasan, 1984; Langer, 1986; Meyer, 1975). In exposition (i.e., imibripative text), expert writers usually organize their ideas hierarchically or logically under a top level or overall organizing structure (often called a rhetorical predicate, Langer, 1986; Meyer, 1975). Children, however, tend to elevate lexical (sentence level) predicates, especially descriptive ones, to the top level as their organifmg frames for exposition (Langer, 1986). Whatever the organizational structure, it is developed, elaboratPAI, ane tied together with cohesion devices that can express text redundancies in a particularly powerful manner through cohesive harmony interactiois (Hasan, 1984). The power of cohesive harmony interactions comes from their multiple lar rs of repetitions. Specifically, the same or similar noun, verb, and functional (i.e., implicit case grammur role) information is repeated across multiple sentences in a text to fa m cohesive harmony interactions (see Hasan, 1984 for a complete description). This study examines if and how teachers respond to these researc a-based structural aspects of text (hierarchical organization and cohesive harmony) yhPa they judge writing quality. It is important to investigate this relationship for several reasons. First, knowing how to use cohesion and organizatiou in their own writing is significantly related to children's greater reading achievement and writing quality (Cox et al., 1990). Consequently, reason suggests that young writers should be developing greater expertise with the..e structural aspects and teachers who conference with and evaluate young readers;writers should probably focus some attention on these meaning-making structures. However, little is known about whether teachers note or react to these important structural aspects in evaluating a child's writing. If they do not, it seems unlikely that th., w ill attend to them either during reading or writing instruction time. Second, this study is important because a child's attempts to develop a more sophisticated ,Aganizational structure may result in diminished cohesion (Cox & Stewart, 1989), possibly because the organizational structure :eqi.:res more of the young writer's attention (e.g., LaBerge & Samuels, 1974). On the other hand, greater atten'I gratefully acknowledge the assistance of Dons Caro, Carolyn Gray, Susan Glanzos, Beverly Or% and Lana Smith for their help in the holistk scorings. 138 Literacy Theory and Research tion to text cohesiveness might result in less expert overall organization or the two might vary togeter Though it is not yet clear how developing knowledge of organization and cohesion is sequentially interrelated in a child's development of literacy, it is important to understand h^w teachers respond to such variances as they evaluate children's texts. Third, because of discourse analysis's power to identify and evaluate cohesive and organizational aspects of text, it is intuitively appealing to suggest that elementary teachers should be given more technical linguistic and discourse analysis training. The validity of this suggestion, however, should be carefully investigated. It is important to understand if teachers who have had such training respond differently to texts that vary in organization and cohesion from those who have not. Finally, both the technical training and time requirements suggest discourse analy- sis, per se, is impractical for use by classroom teachers. Typically, analyzing a text for its hierarchical organization and cohesive harmony requires two trained readers to independently read, analyze, and graph the text and then come together to compare their -nalyses and resolve differences An important question remains, how Wo' d greater specific instruction or experience with these linguistic aspects difference in the way teachers respond to children's writing? As education iitoves toward more holistic assessment in language (Resaick, 1989), understanding and documenting what is being evaluated by teachers will help others understand and accept teachers' holistically based professional decisions. Specifically, this study asks. (a) Are differences in technical linguistic training and discourse analysis experience related to teachers. Aigments of writing quality? (b) Do teachers' judgments of writing quality align with those suggested by discourse analysis" That is, do teachers' judgments of quality appear related to differences in cohesion and organization? (c) How dc teachers respond, in terms of assigning a writing quality rank, to variances in overall organizing structure (e.g., a lexical versus a rhetorical predicate) and to variances in microstructure (e.g., more or less cohesive harmony)" Spec;6^ally, do variances in these aspects relate to differences in writing quality decisions for either technically experienced or inexpenenced readers? To help answer these questions, this paper compares technically (i.e., linguistically) expenenced and inexperienced teachers' rankings of writing qt,ity for elementary children's expository texts with differences in hierarchical organization and cohesive harmony identified through discourse analysis. METHOD Subjects The subjects were 6 teachers with classroom experience rang:ng from 3 to 15 native speakers, all were familiar with written English and all professed to read extensively for their own recreation and coursework. Three of these teat-hers were just beginning graduate studies after having taught for several years. Each of them had had only one undergraduate linguistics (..ourse several years prior. They had never engaged in any linguistic analyses. They were considered the technically rars 4 inexperienced group. 151 139' sStruenire and *riling Quality Judgments The other three were former teachers who were graduate students at the end bf their doctoral work. All had some recent experience in linguistics courses and linguistic analysis and so might be expected to be more sensitive to and knowledgeabTe about issues such as cohesion and hierarchical structure. They were considered the mere technically experienced group. Materials The materials were 96 expository texts. To interpret the holistic scorings that are the focus of this paper, it is helpful to understand from whom and how these texts were obtained and how they differed in terms of cohesion and organization. The texts were written by 48 children, 24 third g.,.-aders and 24 fifth graders. The children were evenly divided between good and poor readers at each.grade level on A their normal curve equivalent (NCE) scores for reading comprehension the ba. on the Iowa 'rest of Basic Skills (Hieronymous, Lindquist, & Hoover, 1989). They were randomly selected from a median split subject pool with the NCE scores for good readers ranging from the 68th to 92nd percentiles and for poor readers from the 8th to the 32nd pera tiles. The children were seen in two separate sessions at which an expository article (one on ants, one on big cities) was read and discussed along with other information the children knew. The children were asked to use any of this information to write reports for others about their same age who were interested in gathering and sharing informatien about these two topics. A discourse analysis of the 96 texts confirmed that distinct differences in their hierarchical organization and in their cohesive harmony existed To determine this, all texts were analyzed for hierarchical organization by (a) assigning all independent clauses (modified T units) (Pappas, 1981) to a position in a tree structure, (b) identifying the type of predicate structures used (rhetorical or lexical) (Langer, 1986), and (c) calculating the proportion uf well to poorly organized T-units both across each composition and within particular tiredicates. All texts were also analyzed for cohesive harmony by (a) constructing chains of identity or semantically related cohesion de vices, including nouns and verbs (Halliday, 1985, Hasan, 1984), (b) identifying cohesive harmony chain interactions (Hasan, 1984), (c) calculating the proportion of each text's total words (tokens) which were involved in cohesive harmony interactions (Hasan [1984] calls this proportion a cohestve harmony index [CHID, and (d) calculat- ing the CHI of well-organized portions of each text and for particular predicates Proportions were calculated for both cohesive harmony and organization to prevent bias data from the well recognized fact that better readers and writers generally write longer texts (e.g., Loban, 1963, 1976). The discourse analysis revealed that these children used 189 description lexical predicates of which 62 were poorly cohesive (average cohesive harmony index [CHI] = .30) and 127 wcre strongly cohesive (average CHI = .73). They also used 32 rhetorical predicates. Twenty-five were very cohesi...y developed (average CHI= .84) and seven were weakly developed with cohesive harmony (average CHI= .22). All analyses were conducted Independently by two trained scorers. Interrater agreement ranged between 85% and 95%. All disagreements were resolved through discussion. Complete details of these analyses are available from the author 152 Literacy Theory and Researe' Repeated measure ANOVAs showed grade, F(1, 44)=7.552, p<.05 and reading. level, F(1, 44)=17.760, p<.001 were significant main effects for differnces in hierarchical organization in the children's expository texts (Cox, Shanahan, & Tinz- mann, in press). There was also a significant main effect of reading level, F(1,. 44)=4.006, p<.05 for greater use of cohesive harmony in exposition.(Cox et al. 1990; Cox, Shanahan, & Tinzmann, in press); though counter to developMental 'expectations, no significant main effect of grade was found. However, it was assured'thaC the texts varied significantly in hierarchical organization and cohesive harmony prior to holistic scoring. Prior to holistic scoring, the texts were typed and mechanical errors were corrected so that the readers would not focus on them rather than the meaning-making aspects of the texts. Then, the three technically experienced teachers met and discussed Myers (1980) holistic ranking procedures. Following hir recommendations, they selected anchor (exemplary) texts for each rank and decid d on four general critr ia to guide their evaluations. These criteria were: (a) Was ra. writing clearly necognizable as exposition? (b) Was the piece clearly and appro.T.ately organized for exposition? (c) Did the information flow smoothly? and (d) Were connectives used appropriately? The technically experienced group independently ranked all 96 expository texts holistically using a writing quality scale of I (poorest) to 4 (best). Interscorer agreement was 91 All disagreements were resolved by appealing to the third ranker or, if needed, through discussion. Later, the three technically inexperienced teachers met and were asked to independently grade all 96 texts on i writing quality scale of 1 (D or F poortst) to 4 (A or best) No training in holistic ranking or discussion of Myer's procedures, criteria, or anchor texts was provided The teachers were simply asked to grade the papers according to how well written they considered them to be as reports, much as they would grade reports written in their own classrooms. Interscorer agreement was .86. As before, all disagreements were resolved by appealing to the third ranker or, if needed, through discussion. Statistical Analysis Pearson product moment correlations were calculated (a) between the teci.nically experienced and inexperienced groups holistic rankings of the texts, and (b) between each group's rankings and the cohesion and organization indices established by the discourse analysis for each text. Significance was r = .28, p<.05 for 48 cases. RESULTS AND DISCUSSION All three questions v ere clearly answered. First, differences in technical linguistic training and discourse analysis experience do not seem related to teachers' Judgments of writing quality The technicall; experienced and inexperienced groups ranked all texts similarly (r= + .83, p<.(1). Where differences occurred between the two groups, they never exceeded one i-ank With these two groups of teachers, technical 15 3 141 Structure and Writing Quality Judgments training and experience did not alter the way they evaluate children's expository writing Second, both groups of classroom teachers' judgments of writing quality did align with those suggested by discourse analysis. Both the experienced and inexperienced groups' rankings were significantly correlated with a text's being more well-organized (r= + .61, p<.01 and r = i- .64, p<.01 respectively) and using more cohesive harmony (r= + .39, p<.01 and r= + .37, p<.01 respectively). Third, both groups of teachers tended to respond to differences in structural soph.stication in similar ways. For example, both groups of teachers tended to rank texts organized around lexical description predicates below average in quality (average rank, 2). Average to lower levels of cohesive harmony (average CI-II= .53) in description lexical predicates generally earned below average quality rankings (1 or 2) from both groups. However, ooth groups tended to rank descripe Ay organized texts above average (3 or 4) when the texts were developed with considerably more cohesive harmony (average CHI= .73). The simpler and less appropriate overall organizing structure (viz., a lexical description predicate) was ranked above average in writing quality, largely when the text was developed in a highly cohesive manner. In contrast, when a child attempted a more sophisticated organizing structure (viz., a rhetorical predicate) whether or not it was developed with a high level of cohesive harmony, both the technically experienced and inexperienced readers tended to rank It above average in qiality (average rank, 3). For example, seven texts that attempted a response rhetoncal predicate and had an average cohesive harmony index of only .17 still tended to be ranked aboe average (3). Obviously, readers respond to other aspects of the text beyond cohesion or organization. However, the response to cohesion within organizational structure is particularly interesting and deserving of further research. One interesting question is why these readers consistently ranked attempts at more sophisticated rhetorical predicate structures so highly, , even though they were poorly developed cohesively One possible explanation for this observation is that both groups of readers intuAively and consistently responded to variations in structure that suggested developing knowledge about wntten language in young authors. On the one hand, they recognized when a more sophisticated and appropi...e organizing structure was attempted and rewarded this in these young writers with a higher score regardless of the text's cohesive devalopment On the other hand, the less appropnate lexical description predicate was only rewarded with higher marks when it was developed with extensive cohesive harmony Another explanation for these readers' high rankings of rhetorical predicates with little regard for their cohesive development is that they were more familiar with overall organization and so saw it as more important. It is likely that overall organizing structures receive mure attention in school instruction than do cohesive patterns, espe cially in the sense of cohesive harmony. Consequently, both groups of teachers were more senvitive to the mere sophisticated predicates than to differences in cohesive harmony, , and so, ranked the rhetoncal predicate texts higher regardless of their cohesive development. This could be interpreted as an unfortunate lack of breadth in their knowledge of important text variables related to literacy development. However, both groups' responses to the use of cohesive harm, in the description predicate suggests the first interpretation is more likely. Correlational data must always be treated cautiously, especially with only a few ir -, 142 Literacy Theory and Research teachers having ranked the texts, but there is a clear suggestion that teauriers who are native speakers, familiar with written English, do refer to a text's organizational and cohesive structure in appropriate ways when they evaluate children's writing. It is important for teacher educa rs to understand how generalizable this finding is and how this teacher-know/edge develops. Was it these teachers' common background as native and literate speakers of written English that enabled them to evaluate children's expositoty texts in such similar ways? Was it also this common knowledge base that allowed both groups to recognize and respond so similarly to a child's divergence from exper' models, possibly treating these differences quite supportively as developing knowledge of English text structures? The implication that seems clearest at this point, is that greater experience with wriven English text rather than technical structural linguistic training may be important in teacher education. Although hierarchical organization, predicate structures, and cobsive harmony may be complex and technical structural aspects of written language, they may also be like vygotsky's (1962) spontaneous or everyday concepts for teachers who are familiar with written English. Specifically, these complex aspects may be learned through experience in reading ,..nd writing text to construct meaning. Future research needs to investigate whether the findings and interpretations reported in this study hold with greater numbers of teachers, across different ethnic and cultural groups, and with preservice teachers, because what is found w ith these diverse groups should suggest important directions for teacher education. REFERENCES Cox, B E , Shanahan, T . & Sulzby. E (1990) Readers use of cohesion in writing. Reading Research Quarterly. 25, 47-65. Cox B E , Shanahan, T . & Tinzmann, M (in press) Children's knowledge of organization, cohesion, and voice in written exposition. Research in the Teaching of English. Cox, B E , & Stewart. !.1 (1989. April) Young readers' use of coherive cues in writing expository text structures Paper presented at the meeting of the American Educatior.al Research Association, San Francisco. CA. Grimes. I E (1972) The thread of discourse. ithaca. NY. Cornell University. Halliday, M A r. (1985) Ar. introduction to functional grammar. Baltimore. MD. Edward Arnold. Halliday. M A K , & Hasal.. '.. ;1976) Cohesion in English. London: Longman. Hasan 1 (1984) Coherence airl cohesive halmony In J. Flood (E.,J.), Understanding reading comprehension (pp 181-219). Newark, DE: International Reading Association. Hieronyrnous. A N . Lindquist. E F . & Hoover. H. D (1979). Iowa Test of Basic Skills. Iowa City. IA. Houghton Mifflin LaBerge, D , & Samuels, S 1 (1914) Toward a theory of automatic information processing in reading. Cognitive Psychology. 6. 293-323 Langer, I (1986) Children reading and writing. Structures and strategies. Norwood, NJ. Ablex. Loban, W (1963) Language of elementary school children. A study of language and relations among speaking, reading, writing, and listening (Tech. Rep No 1). Urbana, IL. National Council of Teachers of English Loban, W (1976) Language development, kindergarten through grade twelve (Tech. Rep. No. 18). Urbana. II: National Council of Teachers of 'lliglish. Meyer, B .1 F (1975) The organization of prose dnd its effects on memory. Amsterdam. North-Holland. Myers, M (1980) A procedure for writing assessment and holistic scoring. Urbana, IL. National Council for the Teaching of English. 155 ' , 1*-rakr41.'c of tviddl-Synerghtic! z devCloPment of nanntive cii langUagu,det!elPinnent: An exaininanon if iolsisive,,i!armony of ',FfOriri;-prOuc.14 fin)getee Ott;-.. dktation, and,writing: thipublisbeil doc;niel 44101461];* OfIo State P;i44...ri*,- 'st "'Columbus: lt*.ick,;;F:,(1989..Pe2oimbr), YterdeilP school and out.-Pa#7.Ptese*teil.,t th! POIng of tiii.$000 ; R.444 Oinlerence,,Austiii. TX .09,64 7.ughi Ond language. (E. Hanfmann'smi 0 Vakai, cankidgii M.4." nd.NewYork 143.t. Pritt aUcl lohn _ 156 1 A LONGITUDINAL STUDY OF PRESERVICE TEACHERS' KNOWLEDGE STRUCTURES Beth Ann Hemnann University of South Carolina Our image of the effective teacher has changed considerably. Today the effective teacher is viewed as a "thoughtful professional" (Peterson, 1988), adaptive to contextual and situational classroom demands (Duffy, 1989), able to confront cultural realities of schooling and engage in critical self-examination, thoughtful discourse, conceptually based decision-makinz (Clark & Peterson, 19S6) and reflection (Schon, 1983). The knowledge structures of "thoughtful professionall" are believed to be organized, coherent and highly integrated, which facilitates adaptive, reflective thinking (Herrmann & Duffy, 1989). Teacher educators generally assume that preservice teachers' knowledge attictures become like those of thoughtful professionals as they complete a teacher education program, but little is known about the longitudinti development of preservice teachers' knowledge structres. Based on recent cognitive psychology research (Fredericksen , 1984; Schuell, 1986), expert-novice work (Chi, Glasser, & Rees, 1982) and recent research on teachers' knowledge structures (Herrmann, 1988, 1989; Roehler et al., 1987), this study describes the longitudinal development of the knowledge structures of three preservice teachers (Students A, C, and G) as they move through three phases of a teacher education program. Phase I includes a teacher effectiveness course and a practicum course, Phase II includes various methods courses, and Phase 111 includes 15 weeks of student teaching. Two research questions were posed: (a) Do preservice teachrs' knowledge structures bec. ie more extensive and coherent as knowledge is acquired in teacher education courses? (b) Do preservice teachers' knowledge structures become more integrated as knowledge is acquired across various teacher education courses? This paper describes the knowledge structure development of the preservice teachers across the teacher effectiveness course, the practicum course (Phase I), a reading methods course (Phase II), and student teaching (Phase III). METHOD Subjects Subjects were 1 female Elementary Education major and 2 female Early Child hood EduLauon majors beginning Phase I of a teacher education program in a large 145 157 146 Literacy Theory and Research southeastern university. These students were randomly selected from among 20 juniors enrolled in the teacher effectiveness course taught by the researcher. At the beginning of the study all 3 students were 22 years old, they all had a grade point average of 3.2 or better and they had all completed the same basic required courses (e.g., English, math, history, and foreign language courses). The students also had similar backgrounds; they all three attended elementary and secondary schools in the southeast and had extensive babysitting experiences. All 3 subjects came from families of noneducato rs. Context D ita reported here were collected within the context of foui 15-week courses: a teacher effectiveness course, a practicum course, a reading methods course, and student teaching. The researcher taught the teacher effectiveness course in which all 3 subjects were enrolled. The course focused on effective classroom instruction, educational settings, and student learning. Sample topics included teacher instructional behaviors and student outcomes, teacher thinking and decision-making, and school effectiveness. Students participated in field-based school observations. The practicum course was taught by Elementary Education and Early Childhood faculty. Student A was enrolled in the Elementary Education section of the course; students C and G were enrolled in the Early Childhood Education section. Both sections of the course focused on gc.neral classroom methodology, materials, and technology. Sample topics included learning environments, child behavior, and instructional planning. Students participated in guided classroom observations and developed lesson plans and learning activities The researcher taught the reading methods course in which all 3 subjects were enrolled The course focused on the reading process and methods for developing effective reading Sample topics included literacy, metacognition, and the literate environment Students conducted approximately 12 small-group tutorial sessions. During the 15-week student teaching experience the students participated in approximately 85 clock hours of supervised classroom teaching as well as weekly seminars taught by Elementary Education and Early Childhood Education graduate students. Materials Subjects constructed an ordered tree (Naveh-Benjamin, McKeachie, Li & Tucker, 1986) at the beginning of each course, at the semester midpoint, and at the end of each course except for student teaching when pm- and post-ordered trees were constructed This technique allowed subjects to display how concepts Included in their knowledge strueures were tied together into a network uf relationships. Subjects also constructed ordered trees in two earlier phases of the study isee, Herrmann, 1988, 1989). Procedures for Constructing Ordered frees In the present phase of the study,, as previously, subjects listed, categorized, and labeled words and phrares about effective teaching, arranging categories to thaw relationships among groups After construction of each ordered tree, subjects recorded 15 siii,ireTetiehérk KnOwledge..§ruCtures Effective Terching of Reading (Chunk 1) Wad Lan Page Est:alio= (Chunk 2) ming aperientu mania wsdasusjog owes using wins room alumni (Chunk 4) typia wire sa taxis wads (atak 6) relokoships aging sight Ccaprehension Idesuillarion (Chunk 3) =Wass writing (Chunk 9) (006k 7) puma Ewing post head Wet oil TIMICITs ws:fng wads brageeoi Mae cxloi w/ tisoctipity ussiysis Ntunerical Measure &ram Concepts Chunks Avenge Numarof Coocepu Pa Chunk Horizontal Lads of Cbunb (3) and Chunks at the Widest Pobt (4) Crverall Numaical Saxe - /11410S 27 9 3 3 7 3 2.5 8.5 4.3 (f Rating Score) Cbarenee Measure Coherau chuftts Cchetra Venially &taxied Sequoias Bred= )111108 8 out of 9 (89%) 16 cut of 17 (94%) 8.9 9.4 Overall Coherence Score .9.2 ( Rating Score) Fig; re /. A sample ordered tree about effective teaching of reading. or audiotape anti; wrote a description of relationships among concepts and how specific aspects of the teacher education program influenced thei,. thinking. Figure 1 is a sample of one subject's ordered tree. Procedures for Analyzing Ordered Trees Extensiveness and coherence. A numerical measure and a coherence measure adapted from Naveh-Benjamin et al. (1986) and validated by Roehler et al. (1987) were used to judge extensiveness and coherence of the ordered trees. For the numerical measure, the following were counted. (a) concepts, (b) chunks (clusters of concepts), (c) the average number of conpts per chunk, and (d) a combination of horizontal levels of chunks and chunks at the widest point. These criteria were used because earlier studies of experts' and novice learners' ordered trees showed that novices' ordered trees varied greatly for each of these categories, whereas experts' ordered trees wete structurally similar. Figure 2, published previously in Herrmann (1989), shows a :tandard 10-point rating scale based on experts' ordered trees used to convert the numerical scores to ratings which were averaged to obtain an overall numerical score. To illustrate how the numerical measure was up:xl, consider the sample ordered IDA 148 Literacy Theory and Research, RATING CONCEPTS 1 2 so 11111111 3 4445,6 o7 I 11111 I 010i0 .30 :40 i O 19 10 IFO $O ICO . CIWNKS 3 AVERAGE{ COWEN'S Pea (MUNK 9 * 0 2.4 Figure 2 01 11j 1111. 4.6 4:4 COMSINED DEMI ANG 3 41 ' 41 4.11 2 3 96 34 12 41 3.'4 44 3.'2 16'.11 09 oR o 1i0 1 i0 I 40 40 tia zç 110 I 2.,3 3?) 2.4 2.6 1 14 111111111 7 o6 o514 o 3 0241 o 1111111111 :it 44101 19.2 21..6 24 21.4 22 2i° 26.4 MS 312 12 33'.6 1/11 ea t '4 tli 11 .. /1 34 354 40:3 it 1 11 Rating scale for determining ratings assigned to numerical categories. tree shown in Figure 1 This ordered tree contains: (a) 27 concepts, (b) 9 chunks, (c) an average number of 3 concepts per chunk, and (d) a combination of horizontal levels of chunks and chunks at the widest point of 7. An overall numerical score was determined by averaging ratings assigned to each of these categories. The coherence measure was used to determine the extent to which relationships between concepts were logical. This measure was employed because in earlier sWdies novices' ordered trees varied greatly hi terms of logical relationships established among concepts, whereas experts' ordered trees did not. For this measure, relationships within individual chunks and across vertically extended sequences of concepts were scored following a two-step procedure. First, based on students' descriptions, chunk coherence was determined by examining relationships between concepts included in each individual chunk. For each chunk in which relationships depicted among concepts were logical, one point was awarded. Percentage of coherent chunks was computed and converted to a rating (e.g., 94% = 9.4). Second, coherence across vertically extended sequences of concepts was determined by examining relationships among concepts included in each vertically extended sequence, one point was awarded for each sequence containing logical relationships among concepts. Percentage of coherent vertically extended sequences was computed and converted to a rating. An nverall coherence score was obtained by averaging the two coherence ratings. To illustrate how the coherence measure was used, consider again the sample ordered tree shown in Figure 1. The figure shows that 89% of the individual chunks and 94% of the vertically extended sequences (e.g., effective teaching of reading 0. language experience using experiences) were judged to be coherent. Chunk 6 was judged to be incoherent because the grouped concepts included in the chunk (strategies, using sight words, using context and using structural analysis) are illogically connected since using sight words was not considered to be a strategy. The percentages and ratings assigned to each coherence cater and overall coherence score for the sample ordered tree also are shown in Figurc . Two graduate students conducted the coherence rating. First, following conventions established during training, each rater independently scored each tree. Second, raters discussed 4.screpant scores until 100% agreement was established. Integration Each student's final ordered tree: from each of the four courses were 60 ANIMILIAN 149 Preservice Teachers' Knowledge L'tructures compared to deermine the extent to which knowledge acquired from the courses was integrated. The ordertd trees were examined to determine the extent to which two knowledge integration criteria were met: (a) three or more chunks repeated across the final ordered trees, and (b) a combination of newly acquired concepts and concepts acquired in earlier courses included in each repeated chunk. RESULTS For this paper, 27 ordered trees were analyzed. From phase 1, final teacher effectiveness course ordered trees, which served I. a base line measure of each student's knowledge structure, and the practicum course ordered trees were analyzed. From Phase 11, reading methods course ordered trees were analyzed, and from Phase III, student teaching ordered trees were analyzed. Changes in Extensiveness ard Coherence Overall numerical and coherence scores were studied to identify patterns of ;hange in each student's knowledge structure (Table 1). An erratic pattern emerged across Student A's numerical scores, indicating inconsistent change in the extensiveness of her knowledge structure. The numerical scores of Students C and G are consistently low, indicating little change in the extensiveness of their knowledge structures. Although coherence scores for all three students tended to be higher than their numerical scores, an erratic pattern emerged across all three students' coherence scores, indicating inconsistent change in the organization and coherence of their knowledge structures. For example, Student A's coherence scores dropped off during the practicum course and did not pick up until the end of the reading methods course On the other hand, Student C's coherence scores were fairly high during the practicum course and quite high during the reading methods course, but her coherence score was considerably lower at the end of student teaching. Student 0.'s coherence scores were higher during the praah_um and student teaching than they were during the reading methods course. Evidence of Knowledge Integration The students did not repeatedly use three or more chunks w hen constructing findl ordered trees. Instead, eaLh student's final ordered tree contained only concepts acquired in the o arse for which the ordered tree was constructed. DISCUSSION Findings from this study generally do not support the hypothesis that preservice teachers' knowledge structures become more organized, coherent and integrated as they complete a teacher education program. Rather, these results suggest that as preservice teachers move through a teacher education program, they develop rather hnuted, Lourse-specific knowledge structures, fairly well organized and coherent dur- 161 on themes and size back tation teacher product of Most of man & is Lack effort courses in establish coherent Most of course be made passivity, (Clark explicit teacher & Rickert, and which to Emphasis teacher the on Attention teacher-student findings connections a to little between knowledge new in content still concepts to curriculum in with in study teachers with 9.0 6.5 7.1 taught into by preservice of 10.0 cohesion and teacher in 7.7 to Practicum 8.9 7.5 11 3.0 2.8 Course OT#3 10.0 6.3 2.6 2.9 6.8 10.0 6.7 3.6 2.5 3.0 OT#1 Scores 8.2 Metlods OT#2 Reading education integration structures Course 8.2 several 10.0 9.7 4.0 2.4 5.8 OT#3 acmss education program (Wilson, program. 10.0 9.0 8.9 3.1 1.8 3.9 6.8 3.5 1.8 2.3 OT#1 organized effort concepts. and 3.6 2.4 7.6 isolated, to Numerical OT#1 program process- 2.0 1.8 4.5 an lack entire integrated change. current 7.0 9.0 9.3 knowledge Aan research experimen- across faculty experienced conceptual themes preser- Teacher Effectiveness knowledge from tended little Course teacners' various traditional learned (OT) 2.4 5.0 7.8 below. programs develop rathei the these OT#2 ts an knowledge with Scores Tree teacher enrolled with associated individual to be an most contemporary Scores Coherence difficulty course creativity, Numerical Ordered 9.8 7.2 8.7 most teacher previously tend preservice pedagogical ere Professional" 1979) part, Coherence nonintegrated eftL faculty each reflection, this by Change and behaviors the taught in overall 1988). grounded Overall teachers Conceptual experienced but study Overall but why education Integration teaching, relative teacher adaptiveness, Brophy, Student Courses and this "Thoughul associated are and they to help emphasis on For Cohesion coordinate specific strive decision-making & Studeat CA OT#3 about Evertson, Student courses, reasons why teachers Peterson, Overall education programs are teacher preservice Teacher (Anderson, and Student preservice Becoming c.impliance, Little, difficult educators pedagogical structures interaction, Student G C A possible explored education based 1986; The one reason with Peterson, Studett they 1986). develop teacher Most did 1987) conceptually & ing way teacher to knowledge of (Lanier 1988) content research it making may teaching Lack man, which deveoped empha- The Schul- courses and integration. ErlaS, course-by- (Good- departments, curricula the Three 10.0 10.0 Teachipg OT#2 Student Preservice Teachers' Knowledge Structures vice teachers in this sway were provided with information and experiences giounded in "process-product" criteria for effective . tang such as drill and practice procedures, mastering specific teaching behaviors ane candard operating protedures, rather than criteria for becoming a thoughtful professionaLesuch as, understandings of poniplex classroom social systems and conceptually based adaptive decidion-makingi:This may explain why the preservice teachers' knowledge structlien *ended to be Mote Promdurnl, such as the knoevledge structures of teaching teChnicians, rather than organiied, coherent and highly integrated, such as the knowledge stmctures of thoughtful profes- sionals (Roehler et aL, 1987). SUMMARY AND CONCLUSIONS This study represents an initial exploration of the lengitudinal development of the knowledge structures of three preservice teachers as they completed a three-phase teacher education program. While the results of this study may not be reflective of the general population of students enrolled in Colleges of Education, findings suggest that the preservice teachers developed procedural, nonintegrated, course-specific knowledge structures rather than organized, coherent and highly integrated knowledge structures. A general lack of cohesion across the teacher education program in which these students were enrolled, as well as a lack of emphasis on teacher conceptual change and the teacher as thoughful professional may have heavily contributed to the technician-like development of the preservice teachers' knowledge structures. These results have implkations for teacher educators, teacher editeation research and teacher education reform. First, teacher educators need to break the faculty tradition of not engaging in collaborative discourse and coordinate individual faculty efforts into an integrated program of study that emphasizes overall teaeher conceptual zhange and cohesive themes of conceptually based erofeesionalism. Second teacher education research needs to explore the effect of in _gated teacher education programs on the development of preservice teachers knowledge structures. Finally, teacher education reform efforts need to shift from focusing on suiface-leiel academic issues such as degrees, majors, lert,i of programs, grade point averages, test scores and programmatic issues to focusing on deeper issues such as the quality of instruction proviled across teacher education programs. REFERENCES Anderson, K . Evertsun, C.. & Brophy.J. (1979) An experimental study of effectivi. teacluni, in first grade reading groups. Elementary School Journal, 72. 193-223. Chi, M.. Glasser, R., & Rees, E. (1982). Expertise in problem solving. I n R Sternberg (Ed ). Advances in the psychology of human intelligence (pp. 7-75). Hilsdale, NJ: Erlbautn. Clark, C. M., & Peter.m. P. L. (1986). Teachers' thi..ight processes. In M Wittrock (Ed ), Handbook of research on teaming (pp. 255-2.`). New York: Macmillan. Duffy, G. 0989). 4 sewing teacher edscation reform from the trenches. The need for on line study of fracher education practices. Unpublished manuscnpt, Michigan State University. East Lansing 163 -.11111:11101 152 Literacy Theory and Research Fredericksen, N (1984) Implications of cognitive theory for tnstruction in problem solving. Review of Educational Research, 54. 363-407. Goodman, J (1988) University culture and the problem of reforming field experiences in teacher educahon. Journal of Teacher Education, 39, 45-53. Herrmann, B A (1988) An exploratory study of preservice teacher's knowledge structures. In J. Readence & R Baldwin (Eds.), Dialogues in literacy research (pp. 347-353). Chicago: National Reading Conference. Herrmann, B. A (1989) Th evolution of preservice teachers' knowledge structures. In S. McCormick & .1 Zutell (Eds ), Cognitive and social perspectives for literacy research andinstruction (pp. 511-519). Chicago: National Reading Conference. Herrmann, B A , & Duffy, G. (1989, March). Relationships between teachers' conceptual understandings, teacher responsiveness and student outcomes. Two erploratory studies in teacher rducasion sentngs. Paper presented at the meeting of the American Educational Research Association, San Francisco, CA. Lanier, J , & Little. J (1986). Research on teacher education. In M. C. Wittrock (Ed.), Handbook of research on teaching (3rd ed.) (pp. 527-569). New Yoit: Macmillan. Naveh-Benjamin, M , McKeatchie, W., Lin, Y., & Tucker, D. (1986). Inferring students' cognitive structures and their development using ths. "ordered tree technique." Journal of Educational Psychol- ogy, 78, 130-140. Peterson, P (1988) Teachers' and students' cognition knowledge for classroom teaching and learning. Educational Researcher, 17, 5-14. Roehler, L R . Herrmann, B A . & Reinken, B. (1989). A manual for constructing and scoring ordered trees. Unpublished manuscript, Michigan State University, East Lansing. Roehler. L R., Duffy, G G_, Conley, M., Herrmann, B. A., Johnson, J., & Michelsen, S. (1987, April) Exploring preservice teachers knowledge structures. Paper presented at the meeung of 111 American Educational Research Association. Washington, DC. Schon, D A (1983) The reflective practitioner How professionals think in action. New York. Basic Books Schueil T J (1986) Cognitive conceptionb of learning Review of Educational Research, 56, 411-436. Wilson, S Schulman, L & Richert. Pi. (1987) "150 different ways of knowing". Representations of knowledge in teaching In J Calderhead (Ed.), Exploring teacher thinking (pp. 104-124). Sussex: Holt, Rinehart & Winston. 164 A COMPARATIVE STUDY OF THE 'TEACHING EFFECTIVENESS OF INFLL___ ITIAL AND NONINFLUENTIAL TEACHERS AND READING COMPREHENSION DEVELOPMENT1 Robert B. Ruddell Marilyn E. Draheim University of California, Berkeley University of the Pacific .1 Judith Barnes University of California, Berkeley Our understanding of the relationship between teaching effectiveness and comprehension development is of central importance to the improvement of literacy skills for students at all levels of education. The centerpiece for this understanding is found in the role of the teacher in directing instruction (Duffy, 1982; Durkin, 1978-79; Ruddell & Harris, 1989); the current study focuses on teaching effectiveness in developing reading comprehension. Specifically, the study examines instruction implemented by influential and noninfluential teachers. An influential teacher is that special person whom we recall in a vivid and positive way from our academic years (Ruddell & Haggard, 1982; Ruddell, 1983). We can often remember the name of influential teachers, their grade level, and even their personal characteristics and teaching style. Most of us have been fortunate enough to have had one or more such teachers in our academic experience. ' . fact, research (Ruddell, 1983, Ruddell & Kern, 1986) suggests that high achievers average three, and low achievers average one and one-half such teachers from the 30 to 40 teachers they encounter between kindergarten and Grade 12. An influential teacher is defined as a teacher identified by a former student as having had a major impact on the academic achievement and/or personal life of that student (Ruddell & Haggard, 1982). Previous research (Ruddell, 1983) on influential teachers, based on responses from former students, revealed five distinguishing characteristics. uses motivating and effective teaching strategies (45%), helps with personal problems (21%), creates a feeling of excitement about the subject matter content or skill area (15%), reflects a sense of personal caring about the student (14%); and demonstrates adjustment of instruction to learner need (5%) The responses of high and low achievers revealed nearly identical patterns across the fve areas, indicating that regardless of achw vement level, these students perceived their influential teadiers in a very simiLii way. Most educators would perhaps suggest on an intuitive basis 'This paper wia presented at Symposium 106, Reading and Learning from Text in the Classroom The Teacher Vanable. This symposium was dedicated to the memory of the late Protsor Harry Singer. University of California, Riverside. 153 165 154 Literacy Theory and Research that these student responses reflect characteristics of highly effective teaching and that such teaching should, in turn, produce higher achievement levels. We wondered if influential teachers would actually teach differently from noninfluential teachers and if, in fact, they would produce higher reading achievement levels. Three purposes thus guided the design of the study. These were: first, to idenhfy instructional characteristics of influential teachers in contrast to noninfluential teachen during reading comprehersion instruction; second, to examine the way in which these teachers develop instructional goals, instrucdonal strategies, and monitor student responses during comprehension instruction; and third, to compare primary-grade student reading achievement in influential teacher classrooms with that in noninfluential teacher classrooms. The rationale for the study is based on the Interactive Instructional Model of the reading process (Ruddell & Kern, 1986; Ruddell & Speaker, 1985) which is highly compatible with the Singer interactive model of learning from text (Singer, 1987). In both models the teacher is conceptualized as the critical decision maker, prior to and &ring the instructional episode (Dreher & Singer, 1989). Our model (Ruddell & Kern, 1986) proposes that the effective teacher creates a highly motivating Instructional Environment that actively engages the student through language use, instructional strategies, and the use of text-based meaning cues. This en% 1Tonment activates the student's Knowledge Controla sort of mental switchboardthat enables the student to set a clear goal direction (affective state) and instructional p.an (cognitive state) for the lesson, and also helps the student monitor whether or not the goal is being reached qnd the plan is working (metacognitive state). The student's goal and plan, in turn, directs the student's memory search for knowledge and experience-based schemata using Prior Knon ledge and Beliefs relevant to the content being read or dmeussed. Finally, the interac#:an of these processes leads to the Instructional Product, ranging from comprehension to new knowledge. It was our expectation, based on the Interactive Instructional Model, that influential teachers would demonstrate more effective use of these mstructional components than would noninfluential teachers We thus hypothesized. first, that the influential teachers would exhibit significantly higher levels of performance on the Classroom Interactions Rating Scale (Ruddell & Haggard, 1982), accounting for these model g-omponents, than would the noninfluential teachers, .econd, that the detailed analysis of video recorded lessons would reveal a more precise use of the Knowledge Control component, in the for-- of instructional goals, plans, and munitoring, by the influential teachers; and third, that reading achievement for the primary-grade students in the influential teacher classrooms would be significantly higher than reading achievement for students in noninfluential teacher clas:,rooms. METHOD Our studen and teacher sample for the study was based on an extensive data ba.se collected 8 years before the present study . A research project studying the impact of inservice training on teacher and student perf,,rmance had been directed by the first 166 155 Influential and Noninfluential Teachers author of this study, over a year-long period. in a primary-grade school in a West Coast metropolitan city. This research project used extensive videotaping of the 24 teachers in the school, under carefully controlled conditions, and standardized reading achievement tests had been administered to the 522 students, le.;idergarten through Grade 3. Eight years after the development of this data base student records were examined in the school district in the aaempt to identify the former students in the year study. Our search revealed that 132 students were currently enrolled. A questionnaire was designed to facilitate the identification of teachers who were influential in the academic careers of these students. Each student was asked to identify the teacher or teachers who had significantly influenced the academic and/or personal life of the student in the present and in previous years of schooling, assuming such a teacher existed. Our analysis of student responses revealed 79 influential teachers from the pnmary grades through the junior high and high school levels, with many teachers identified by more than one student. All students who identified influential teachers specified the teacher's name, grade level, content area taught, characteristics of the teacher, and the academic or personal influence of that teacher. From the questionnaire 4 of the 24 teachers in the original teacher population were identlfied by two or more students as influential teachers. These four teachers thus constituted our sample of Influential teachers. From the 20 teachers remaining in our onginal teacher population, 4 were selected at random to constitute our noninfluennal teacher sample. The instrucunal samples selected for analysis consisted of a video recording which had been developed in all c:assrooms in September of the school year. Each teacher, kindergarten through Grade 3, had been asked to use the children's literature selection, Alexander md the Windup Mouse, by Leo Lionni, as they would normal!) e it in their classroom to develop comprehension skills A group of c: students had been selected at random from each classroom for the presentation, thus providing for a range of achievement levels. The lesson for the 8 teachers averaged 20 minutes in length with a range from 15 to 26 minutes. Two types of analysis were used in our study of the instructional episodes The first analysis, related to the first hypothesis of the study, , used the CI-55room Interaction Patterns Scale. This scale w& based on L A extensive research review of teaching effectiveness (Ruddell & Haggard, 1982, Ruddell, 1983), and examined five instructional components each with a scaled rating from 7 (highly descriptive) to 1 (not descriptive,. These components were. (a) Classroom Comm, lication, defined by tcacher-student interaction patterns, receiving-caring, clarifi.ation-resolution, and awareness of student expectations-performance, (b) View of so: defined by sense of seif, enthusiasm. locus of control, and control sharing, (c) Management Style, defined by sense of purPose and goal orientation, moth ation and cooperation, timing-pacing, and flexibility, ') Problem Solution Approach to Learning, defined by intellectual cunosity, attitude toward learning, sens:Ivity to teachable moments, and questioning strategies and response paneins, and (e) Teaching Effectheness, defined as the mean of the four factors above. Observations of t:.e inflaential and noninfluential teachers were completed independently by two trained rateis with an interrater reliability of 95 The Wik.o.on Two-Sample test, a nonparametric test, .vas used to test for statistical 167 156 Literacy Theory and Research significance (Marascuilo & McSweeney, 1977). The analysis was one-tailed and assumptions for -%is test of independence of observations within and between samples were met. Our second analysis of the instructional episodes was descriptive in nature and focused on die second hypothesis of the study concerned with effective use of Knowledge Control c 3mpnneuts in comprehension development. The components of the Ruddell a.id Kern model of interactive teaching 0986) consisted of Goal (i.e., teaching objective, motivation), Plan (i.e., instructional stategy, comprehension levels used), and Monitoring (i.e., discourse strategy providing student feedback and encouraging self-monitoring) (Ruddell & Harris, 1989). The content of these categories was inferred from the enacted data observed in the videotapes based on E.,.kson's microethnographic approach to the analysis of classroom teaching encounters (Erickson, 1982) Descriptors for comprehension levels useu were derived from the Ruddell taxonomy of comprehension/thinking levels (1978) and Rosenblatt's theory of literary transaction (1985). These levels are. factual (i.e., literal recall of information); interpretive (i e , manipulation of information to develop new meaning), applicative (i.e., transfer of information and meaning to a new situation), and transactive (i.e., encouraging the reader to become one with the meaning of the text, or experience). This analysis was conducted by two trained observers with an mterater reliability of .90, and comprised the major descriptive part of the study. Student achievement test data, related to the third hypothesis of the study, were collected in September, at the beginning of the school year, and in May, at the conclusion of the yea- The Metropolitan Readiness Test, Forms A and B, was administered to students in kindergarten and grade one, the ETS Co-Operative Primary Battery consisting of Reading Comprehension and Listening Comprehension (Forms I2B and 23B) was used with students at Grades 2 and 3. Gain scores %sae calculated - students in the influential and noninfluential classrooms. Because the classrooms had combined Grades 2 and 3 for cross-grade gi.uping instruction, the test data for students in these grades were combined. By combining the test scores the separate effects of uncontrolled variables that could interact unpredictably were minimized (lssac & Michael, 1981) 'f he gain scores were computed and contrasted for statistical significance using one-way ANOVA with each analysis tested at the .05 level. The assumptions of the independent observations and approximately equal vanance were met. RESULTS Instructional Characteristics of Teac*Iers Our findings revealed distinct and s.buificant differences favoring the influential teachers on each of the five variables analyzed using the Classroom Interaction terns Scale As noted in Figure i, significant differences were found for classrocm communication ( p< 01), view of self (p<.02), management style (p<.03), problem solution to learning (p<.02), and overall teaching effectiveness (p<.02). These findings provide strong indication that the teaching effectiveness of the 16,5 157 Influential and Noninfluential TeacIlers NEInfluential Teachers Non-Influential Teachers 52 4.9 47 A Classroom Commonlca llon * p < 05 Mini of Self MilmiComollt StYla * Problem Solution * Approach to Learnt:1g TaacNho Ettedveneas Classroom Interaction Patterns Figure 1. Teacher effectiveness. Ratings of influential and noninfluential teachers. influential teachers was higher than was the eff :veness of the noninfluential teachers. Knowledge Control Effectiveness Our descnpti ve analysis of Knowledge Control effectiveness revealed that the two groups of teachers developed instructional goals, plans, and moraaring in distinctly different ways. Instructional guas. The ability to develop instructional goals operationalized in the fomi of objectives and aiotivational strategies was markedly different TL influent:al tcachers, regardless of individual teaching styles, appeared to hoe a clear understand mg of how to identify and select objectivc geared toward de veloping an in-e-pth understanding of the reading sele;tion. Story mot., ation used by these teachers relied heav ily on intenial dnve motivations midi as intellectual curios;ty , self understanding, aesthetic appreciation and problem resolution in the story develcpment, and only rarely used the external motivation of teacher epectation The use ef internal motivation, self-understanding, is illustrated in the question, "Would you rather be a real 'Vinyl? Why?" mouse (like Alexander) or a windup mouse v es were often confined to develBy contrast, the nouinfluential teachers oping literal tia ,arstanding of the story using external motivation, that of pleasing the teacher, by providing a teacher preselected and text based literal response, for exam ple, "OK, so which mouse here always gets screamed at?", "Why were they (the toys) in the box?" 169 158 Literacy Theory and Research Instructional plans. A second difference was found in the degree to which the lesson's goals were related to and implemented in the plan of story development. All of the teachers, influential and noninfluenfial, read the story to the students and asked questions during and after the reading. The comprehension levels stimulated by the questions, however, differed noticeably between the two groups of teachers. The influential teachers used interpretative and applicative type questions predominantly; for exanTle, "Why do you think Alexander (the real mouse) changed his mind about being just like Willy (the toy mouse about to be throw away)?", "Do you think tbe little girl really loved that windup mouse? Why?", "What do you think the author was trying to tell people?" Factual questions occurred primarily for clarification of meaning when needed. The noniniluential teachers, on the other hand, cmicentrated neavily on literal questions designed to verify students' ability to provide accurate factual recall of selected parts of the story. Althoug,, these teac.....:s attempted, on occasion, to arouse the students' curiosity or develop aesthetic qualities in the story, they rarely pursued a line of questioning to develop these features. For example, one teacher asked, "What was Alexander's big problem?" A student respozded, "He wanted to be loved " The teacher then nodded approval and proceeded to an unrelated question concerned with the nature of the lizard's magic pebble, rather than pt.-suing the student's response to consider ways we feel loved ahd express love, the central theme of the story. Instructional monitoring and discourse strategies. The third area of difference between the influential and noninfluential teachers was found in monitoring and discourse strategies used to guide the development of the students' comprehension. The influential teachers appeared to have established an instractionul goal based on the central stury theme. They then gently, but firmly, used planned discourse strategies to guide the interaciion to reach this goal. These teachers provided opportunity for the students to expre.s; their ideas com,,letely and listened attentively to studt.,t responses. Student responses were thus validated as these teachers listened with obvious interest to what their students had to say and monitored further through clarifying and extending type questions Teacher responses were used, at times, to guide students back to the central topic of discussion. Although the influential teaci-ers had estabiisFed definite comprehension goals ;mplementation plans, they also responded readily to teachable muments to further the lesson objective This is illustrated in the following ..terchange which occurred at the conclusion of the story as the snidents were examining the collage-type story illtrtration and a child asks, "Which one is Willy?" T Ci. T. Cl. T. C2. T. C3. T. Can't you tell? No. I don't knov . It's hard to tell. How could you tell them apart? Because he's a windup mouse. Anything eke about them that was different? Yes, he had a key. Yes, anything else? Roundwheels. Yes, maybe. 170 Influential and Noninfluential Teachers r4 i59 Kind of like an egg. T. Sort of. C4. His ears were like two drops of tears. T. Well, that's a good descriptioncan you think of anything else about the way M. Lionni chose to make the mice? Here's Alexander. Here's Willy (shows picture of each). C3. One's rounder. C2. One of ihem is smooth and the other one's rough? C3. Because one's a toy. T. 1h hich one would that be, the smooth one or the round one? C2. The smooth one. T. That's probably the one I would choose-because I would think of a toy-(intarupted). C4. Bccause a real mouse would have fur. T. And so h wouldn't be very smooth would he? C3. No, he would be rough with hair sticking out. This interaction also illustrates the skillful use of clarifying, -xtending and ra.aing type question, in the discourse to develop interpretIve and applicative levels com-rehe n s ion . By contrast, the noninfluential teachers used discourse strategies which were more controlling in nature, more text-driven, and less responsive to teachable moments. These questions frequently focused the students' attent-m on specific aspects of the story prompting a literal recall response. Students were guided to limit their responses to short phrases which appeared to be des_bned to fill in the detal anticipated by the teacher's inquiry. These teachers also used a controlling strategy by directing the factual recall question to a particular child. These features are illustrated in the following interaction. T. What did you like about the story? CI. I liked the part where he found the pebble. T. You like where he found the pebble. Where did he find it, Timmy? By a box. T. Wh., CI. CI. By a box. T. By a box. What were some of the things that were in the box? C2. Dolls-(interruption) T. There were old toys in that box. Why had they been placed there? C3. Because they were old and couldn't work T. And they couldn't work. What did they plan to do with them, Henry? Henry, what did they plan to do with the old toys? This type of interactior. reflects a very limited repertoire of monitoring and discourse straL6ies available for use in guiding the students thinking process to higher compre hension levels. Readint Achievement Differences Our analysis of reading comprehension achievement gains for students in the influential and noninfluential classrooms, as presenteA in Table 1, reveal statistically significant differences favoring the influential teacher students at Grades 2 and 3 , Literacy Theory and Research Table 1 Reading Comprehension Gain Scores (and Standard Deviations) for Students in Influential Teachers' and Noninfiuential Teachers' Classrooms IPIMMIMIMMo Influential Teachers Mean Achievement Measure Metropolitan Ret.ding Readiness Test Grades K & 1 ETS Listening ComprehensionGrades 2&3 ETS Reading ComprehensionGrades 2&3 28 54 51 N 17.96 (16.72) 5.83 (5.69) 36 15.29 (12.43) 34 34 iiothafluential Teachers Mean 14.31 (8.81) 3.14 (5.08) 8.14 (15.29) 1.32 n.s. 5.24 .05 5.19 .05 These differe nces were found for both listening t.omprehension (p<.05) and reading comprehension (p<.05) for the two grade levels. Although positive trend differences %el-. resent for students at kindergarten and Grade 1 for tl.z. influential teacher classrooms, these differences dia not reach statistical significance A clear testing limitation present at kindergarten and first giade because of ceiling effects and the low sensitivity of read: less tests in measuring reading comprehension growth This rmiy account for the absence of statistical significance differences. DISCUSSION Oki findings, both qualitative and quantitative, suggest that the influential teachers in the study were more effective in developing comprehension procesbes with their students than were the nouinfluential teachers. Teaching effecti....ness differences, related to our first hypothesis, favored the influential teachers on each of the five teaching variables of the Classroom Interaction Patterns Scale at statistically siviificant levels Three of the scale variables, Classroom Communication, Management Style, and Problem Solution Approach to Learning, are intimately related to the development of skilled comprehension processing (Ruddell & Harris, 1989, Ruddell & Kern, 1986). The C:assroom Communication v ariable reflects the teacher's ability to interact effectively with students by receiving .nd clarifying responses, reaching resolution in discussions, and in demonstrating sensitivity to student expectations and performance Management Style reveals a clear instructional goal orientation and the effective use of motivation, a iming and pacing in the lesson. Problem Solution Approach tt, .timing indicate. - ability to stimulate intellectual cunosity,, build a positive attitude toward the topic under discussion, and effectively monitor responses and use questioning strategies which foster higher level thinking. The staestical significance findings obtained for the interaction pattern scale variables were strongly wpported by our descriptive analysis of the classroom instruction for the two groups of teachers. The influential teachers were highly skilled in their 174 161 Influential and Noninfluential Teachers use of Knowledge Control, as posited in our second hypothesis. This was revealed through their use of clearly formulated goals and objectives which were closely linked to and served to guide their instruction. Student motivation was, for the most part, internal in nature ranging from intellectual curiosity to story problem reselution. By contrast, the goals and objectives of the noninfluential teachers were often vague in nature and frequently nonfunctional in directing the instruction. Student motivation was most often external and designed to provide factual recall, text-based responses to fit the teacher's preseleeted answer. Although both groups of teachers used an instructional plan of oraA story presentation followed by questions and discussion, the comprehension levels emphasized were markedly different. The influential teachIrs' plan included the predo iinant use of higher level inteteretative and applicative type questions and used factual questions for claiification. Their questions sought to i:wite the students to become involved in the story and to perceive story events from a number of view'. -tie., fur example, Alexander, Willy, and the little girl. The instructional plan used by the noninfluential teachers, however, focusexl on factual level questions related to specific story details hrtnictional monitoring and discourse strategies, in the fo:- _f teacher-student interactions and questioning strategies, were distinctly different between the two groups of teachers. The influential teachers used clarifying, extendinz and raising type questions in their monitoring type interactior with students, promotino, higher level ctAnprehension processes which followed the clear development of the story theme Walt time was us,x1 frequently to provide opportunity for comprehension processing, enabling the students to express their 7Jeas as comple.ely as possible Student feedback and respunse validation was in evidence as the teachers 'istened to students with keen interest and followed with observations and clarifying, tAtending, or raisin, type question,. The noninfluential teachers, by contrast, relied heavily on a corsolling factual recall level. strategy with focusing type questions at We hypothesized that these KnJwleuge Coiurol components iRuclittll & Kern, 1986) should be directly related to and influence student achievement gains The data related tu our third hypothe, re ealed statistically ignificant achievement growth for both reading comprehension and listening cnprehension at Grades 2 and 3 Although positive trend gain differences were present for kindergarten and Grade 1, these differences did not reach sta: stical significance. In conclusion, our findings support the observations of students that their past while influential primary-grade teachers were highly effective teachers. The exploratory in nature, provides insight into the relationship between teachinE, effectiveness and comprehension development. Further, our findings support th, close connec non between effective use of the Knowledge Control components in teaching and reading comprehension development, particularly at upper primary-grade levels REFERENCES Dreher, M. J., & Singer, H 0989) The teacher s role in student success The Reading Teacher, 42, 612-617 Duify, G. G. (1982). Response tu Borko. Shavelson. and Stem There's more to mstrucuonal decision making in reading than the empty classroom." Reading Rt4earch Quarterly. 17. 295-300 1 72 162 Durkin, D (1978-79) What classroom observations revcal elanut reading comprehension instruction. Reading Research Quarterb , 14, 481-533. Erickson, F (1982) Taught cognitive learning in its immediate learning environment. Education and Anthropology Quarterly, 13, 148-180. Isaac, S. & Michael, W V. (ISM) Handbook in research and evaluation, (2nd San Diego, CA: Edits Publisher. Marascuilo, L A & McSweeney, M (1977). Nonparametric and distribution free methods for the social sciences. Belmont, CA: Wadsworth. Rosenblatt, L. M (1985). The transactional theory of the literary work: Implications for research. In C, R Cooper (Ed.), Researching resi ..Itse to literature and the teaching of literature: Points of departure (pp. 33-53). Norwood, NJ: Ablex. Ruddell, R B (1978) Developing comprehension abilities. Implications from research foran instructional framework. In S J Samuels (Ed ), What research has to say about reading instruction (pp. 109-120). Newark, DE: International Reading Association. Ruddell, R 13 (1983). A study of teacher effectiveness variables of influential teachers. In M. P. Douglas (Ed ), Reading, the process of creating meaning for senses stimuli (pp. 57-70). Claremont, CA: Claremont Graduate School Yearbook. Ruddell, R B & Haggard, M R (1982) Influential teachers. Characteristics and classroom performance. ln ; A Niles & L A Harris (Eds.) New inquiries in reading research and instruction (pp. 227-231). Rochester, NY: National Reading Conference. Ruddell, R B & Harris, P (1989 A study of the relationship between influential teachers prior knowledge and 1-eliefs and teaching effectiveness. Developing higher order thinking in content areas. In S. McCormick & I Zutell (Eds ), Cognitive and social perspectives for literary research and instruction (pp. 461-472). Chicago: National Reading Conference. Ruddell, R B & Kern, R B (1986) The development of belief systcms and teaching effectiveness of influential teachers In M P Douglas (Ed.). Reading. The quest for meaning (pp. 133-150). Claremont, CA: Claremont Graduate School Yearbook. Ruddell, R B & Speaker, R B (1985) The interactive reading procc,s. A model. In H. Singer & R B Ruddell (Eds ), Theoretical models and processes of reading (pp. 751-793). Newark, DE: International Reading Association. Singer, H (1987) An instructional model for reading ans; kaning from text in a classroom betting. Journal of Reading Education. 13, 8-24. 174 ; BEST IDEAS: SOURCES AND INFLUENCES Sara Ann Beach University of California, Riverside Where do "best ideas" come from? Are the "best ideas ' research-based or basically a product of conventional wisdom and strongly held beliefs? Have "best ideas" changed over time as knowledge has accrued about both pedagogy and reading? The answers to these questions are neither sirnple nor straightforward. The choice of a set of "best ideas" is a valu, .aden decision, influenced by a wide range of social, cultural, educational, and economic factors. Research can produce "best ideas" or it can produce "worst ideas that are nevertheless implemented and valued. The purpose of this paper is to describe how "best ideas" become valued by reading professionals, describe some of the sources of "best ideas" for reading teachers, and to examine if best ideas have changed over the last century. "Best ideas" are defined for this discuss.on as both the beliefs about reading and instruction and the teaching strategies valued, encouraged, and utilized by reading professionals. HOW ARE BEST IDEAS CHOSEN? In order for ideas to become valued, they must first become known. Clifford (1973) states that educational ideas and research are thrown into and filtered through the general marketplace of ideas. They become known and accepted )y ailtural diffusion, "an obscure, ambiguous, often involuntary transaction syster,i whereby innovations and ideas arc spread widely throughout some extendeu subsociety or the whole culture" (p. 25). If the research and ideas substantiate existing personal and group opinions, they are accepted. If they contradict these opinions, they get neither nodce nor acceptance. Choosing and valuing an idea or practice, then, is dependent on the existing opinions of the receivers of the knowledge. Other influenLes Impact the 4,.ceptance of an idea as having use and value. Rogers affect the adoption of an idea into P. value system. (1983) lists four attributes First, the ideas must be perceived as being relatively advantageoi,:s t) the :ndividual or organization. If the idea is perceived as being much better than what is already believed, then it has more probability of becoming valued. lf, however, it is perceived as being only as good or worse than what is already bel. 4, the idea will probably be discarded. A second attribute is the compatibility of the id with existing values, the idea is with existing past experiences, and present needs. The more values and beliefs, the higher the probability of acceptance of the idea as a "best" idea. The less compatible it is with ex.sting opinion, the less probability it has of 163 164 Literacy Theory and Research adoption The third attribute is the complexity of the idea. The easier the idea is to understand and implement, the higher the probability that the idea becomes valued. The more difficult the idea is to understand, the less probability it has of adoption as a best idea The fourth attribute is the degre- to which an idea may be observed or experimented with If an iuea can be easily implemented or if its Implementation can be observed, the potential user of the idea has the opportunity to evaluate it before making a decision about its effectiveness. Best ideas, then, are chosen because they correspond with and confirm existing beliefs, are perceived as advantageous, are relatively uncomplicated, and can be exper- imented with Best Ideas for reading instruction would be chosen, consequently, by reading professionals on the basis of which ideas correspond with their existing theoretical orientation to the reading process and their beliefs ahmit the type of instruction that should be delivered (Kinzer & Camck, 1986). Harste and Burke (1977) defined the th, mical orientation of a reading teacher as the particular knowledge and belief sy stem a teacher holds toward the reading process. This theoretical orientation would include whether the teacher believed reading was a bottom-up, top-down, or interactive proccss as well as whether reading instruction should be content-centered or child-centered (Rupley & Logan, 1985). Those professionals who believe reading is primarily a bott -n up process may tend tu 'glue ideas which emphasize the goals of word recognition and literal c.,..ipr-hension of the author's meaning. Text-based or content centered instructional ideas focusing on mastery of sici;ls would be valued more highly Professionals whc. believe reading is primarily a top-down process may tend to value ideas w hich emphasize the construct.on of meaning using the reader's prior knowledge and resources. Instructional ideas which were reader-based and stuclent centered emphasizing the holistic nature of talguage would be valued more highly Professiona:s who believe that reading is an interactive process may tend to value ideas which emphasize the interaction bc....Neen the reader the text that ieads tu the construction of meaning. Both text basedcontent-centered and reader-based student centered instructional ideas en.?1.nalzing both comprehension of content and increasing reader resources would be highly valued. An individual's theoretical orientation toward reading and reading instruction is not the only influence on the choice of best ideas. Another important influence is the context of the teaching situation, the implicit and explicit social system of which the teacher is part and which mediates Whavior (Barr & Duffy, 1978, Buike & Duffy, 19-70) It includt:s tie nature of the students in the classroom, the commercial reading program adopted by the school, the grade and ability level(s) being taught, the community surrounding the school, and the experiences of the teacher. Therefore, the choice of best ideas that a particular reading teacher chooses is balanc- d by the constraints placed on him or her by the community,, school, and classroom in which he or she teaches If the community and school advocate a bottom up, content-centered cumcu!um emphasizing the acquoition of basic skills, the best idec chosen by an individual teacher may be more consistent Ne th those demands than with a more student-centered, top down orientation, regardless of the beliefs of 'h.. teacher. If the administration of the st.hut ! requires the teacher to use only the commercial reading program adopted h., the school district, he or she will have to adapt his or her beliefs tespecially if they do n, .Itch the assumptions of the textbook) to the constraints imposed on him or 176 165 Rest Ideas her. If the students in the classroom are perceived to be low ability, tea. .s will often foci% on text-based, content-centered instruction. On the other hand, I they arr perceived as being of high ability, the instruction often becomes very reader-based and student-centered. Pnmary grade teachers often have a tendency to focus on "zxtbased instruction (word recognition slc.:1s) whereas intermediate grade teachers focus more on comprehension. How do these theoretical onentations develop? Can they change? Stansell and Robeck (1979) studied the development of theoretical orier Ions among preservice .riculum in a university teachers. Students in various phases of a teacher preparata were assessed throughout their preparation period using tht Lireoretical Orientation to Reading Profile (TORP). They concluded that theoretical orientations of the students developed as result of the classes they were required to take and could change as a result of the theoretical onentanon of the professors under whom they studied. Moore (1981) agreed that wade formal training piograms may influence the development of a theoretical oncntation to reading dunng preserv ice training, this influence often did not carry over into classroom practice dunng the first year of teaching. Reesag (1984) found, however, that advanced level courses in reading did influence what teachers did in the classroom. Bean (1980) explored the degree to which a graduate course in reading would update teachers existing beliefs and practices. The students in the graduate class completed a self rating scale on their teaching of reading at the begmning of the course and again at the end. The co Ube, a blend of psycholinguistic theory and practical teaching strategies, was instrumental in updatmg teacher beliefs and practices. A teacher's beliefs and theoretical onentation, then, arc formed as a result of the early courses taken dunng preserv ice instructiu...s well as dunng professional courses .an change if a teacher is given taken alter certification. They are not static, eaching situations. Both preser guidance and shown how to apply the beliefs to v ice and staff development training should focus not only on prov iding incirmation to teachers but also prov ide follow up guidance in the applicant'', of the information to actual teaching practice a WHAT ARE SOME SOURCES OF BEST IDEAS? Best ideas come from several sources. One of the sources consulted most frequently dre a teacher's peers and the reading specialist in the school (Ruddell & Sperling, 19E4 Sawyer (1976) surveyed 200 pnmary -grade teachers chosen at ran dum (torn the membership of the International Reading Association to find out what they used as sources for ideas and assistance At the top of the list were tecchers in the same school. Llgan and Erickson k 1979) surveyed elementary teachers and found that a one-to ..ine conference with the reading specialist was a preferred way to get ideas to refine instruction. R. Robinson (1979) found that classr:n teachers get ideas fro... local reading specialists who share new information 4. research findings Armed at solving common pro' 'ems. Another source of ideas arc university classes, methods textbooks, inservice pre sentations, and "experts in the field. Sawyer's survey (1976) revealed that profes- 177 166 Literacy Theory and Research sional textbooks were the second most frequently used source. Stansell anti Robeck's study of preservice teachers (1979) showed that both the prufessor and the reading course influenced the choice of best ideas. Moore (19'...) obseved that both formal training programs and professional books inauenced reading instruction, but that formal training programs had less influence than other sources. Logan and Erickson (1979) discoverd that inservice presentations provided ways for teachers to refine instructional methods. The most preferred inservice workshops were training sessions with consultants on specific problem areas, classes or workshops for college credit, and presentations by reading specialists. Bean (1980) concluded that graduate classes that were a blend of theory and practical teaching strategies were instrumental in changing teaching practices. Pearce (1984) observed that busy teachers got their ideas from inservice workshops as well as coursework and the textbooks accompanying them E -en the "experts" are influenced by textbooks. A column in The Journal of Reading entitled "Rea 'digs That Made a Difference" chronicled the influence of specific books on those reading prof-ssionals who are now considered experts (see, e.g., H. Robinson, 1980; Harris, 179). A third source of best ideas are journal articles, professional meetings and research (Pearce, 1984; Ruddell & Sperling, 1988). Sawyer's (1976) survey showed that the most preferred features of journal articles were activities for classroom use, enrichment activities, and suggestions of materials for classroom use. Ngandu (1978) analyzed the content of the International Reading Association National Conventions from 1962-1977 She discovered that the convention became more and more geared to the needs of practitioners One recent book of best idtzs, Becoming a Nation of Readers (Anderson, Hiebert, Scott, & Wilkinson, 1985) is based on the most current research on reading. A fourth source of ideas is cc mercial reading materials used in the classroom (Moore, 1981) Shannon (1982) stated that teachers rely on the ideas in commercial reading materials because they belieNe that the materials have scientific validity. Additio,-,ally, they feel pressure from the school administration to use them. HOW HAVE BEST IDEAS CH kNGED? Best ideas in reading, in fact in all of educational practice over the last century, have been influenced by a variety of economic, social, cultural, and educational factors These factors lude the increased time spent in school by children, the increasing flow of immig.ants into the country, the increased need for literate workers, the diminist.ing and then resurgent influence of religion, and we changing concepts of child development The schools did not initiate new practiccs because of these changes They began to reflect the changing practices and beliefs about children as well as changes in society. Table 1 summarizes some best ideas over the century. Before the 1920a, the best ideas in reading instruction were the almost exclusive use of oral reading and of drill The learning of decoding skills was emphasized. It generally was believed that comprehension would occur if word recognition was mastered Reading methods texts of the period emphasized oral reading and synthetic phonics (A. Hoffman, 1983; J. Hoffman & Segel, 1983). 4. 78 Best Ideas 167 -4, Table 1 Best Ideas Across the Century* 1920s 1930s 1950s Emphasis on read- Increase breadth and efficiency in insnuction. Acquire greater independence in readhig. Reading part of unified Provision of adequate experiential background. Development in reading related to other language arts. Definite instruction in reading rather than just tesr exercises. Directed toward specific valid ends such as rich experiences, broad Intermts, enjoyment, and growth in fundamental reading ability. ing as a thought-getting process. Recognition of importance of wide experience to good inte-pretation. Increase in amount, vari- ety, quality of reading materials. Systematic development and independent use of reading habits. New classroom organizat:on. Use of informal Program- Development of basic reading habits. Intriguing materials used. Base instruction on capacities, interests, needs of students. 19603 Interrelationship of learning to read and role of reading in personal and social development at all levels. Make explict the close relationship of reading and the other language arts. Continuous program, each level building on the previous one. 1980s Parental involvement in early readi.ag. Stinvilating classrooms. Well-designed phonics lessons. Interesting materials for instruction. More time for comprehension instruction, independent reading and writing. More comprehensive assessment of reading and writing. tests to discover needs. *From NSSE Yearbooks (1925. 1937. 1949. 1961). During the 1920s emphasis shifted to silent reading. Research began to make itself felt, although the influence of the research derived more from the opinions of the researcher than the quality or results of the research (Venezky, 1984). Thorndike's view of "reading as reasoning" combined with the Herbartian emphasis on meaning and content moved the view of reading to more than just word calling. Reading methods texts began to emphasize the "look-and-say" method of teaching reading as well as silent reading. A dichotomy arose betwotn those who felt reading should be taught as sequnntial skills and those who felt it should oe based on the pm-poses of the child (Paris, Wixson, & Palincsar, 1986). Dunng the 1930s the use of oral reading returned to the classroom (J. Hoffman & Segel, 1983), although the emphasis on silent reading continued. Reading methods textbooks emphasized the systematic teaching of reading skills using the controlled vocabulary of the basal readers (A. Hoffman, 1983). During the late 1940s and the 1950s, basal readers, thc Directed Reading Activity, and round-robin oral reading became entrenched in the schools (I. Hoffman & Segel, 1983). The skills approach was favcesi, with stress placed on phonics instruction. Ly the early 1961s, re9eag methods texlooks were expanding to include discussions of individualization, the Language experience approach, programmed instrucon, and linguistic rcadnrs. e4 t ; 168 literacy Theory arid Research In 1985, Anderson et al. revie wed the research on reading for the Office of Educational Research and Improvement. The result was Becoming a Nation of Readers, which could be classified as the best ideas for the 1980s. Have values really changed over the last century? Not really. Those ideas valued in the 1920s (importance of background knowledge, use of a wide variety of reading materials, emphasis on reading for meaning, encouraging permanent interst in reading, integration of reading with other subject areas) are still valued today. What has changed over the years is how best to achieve these valued aims. What, then, has the increased research in readf7g contributed toward the formulation of best ideas or changing values of reading teachers? It has served to illuminate the many aspects of the reading process as well as highlight the different paths leading to the same goals. It has brought diversification to the field of reading. As Dewey said in 1929, "Corn:nand of the subject matter and the scientific method . . . libe:ates the individual; it makes for diversification rather than set uniformity" (cited in aifford, 1973, p.3). This diversification has led and will continue to lead to more d. hanging methods to implement best ideas which will hopefully lead to better reading instruction and a more literate society. REFERENCES Anderson. R C . Hiebsn. E Ii . Scott. J. A.. & Wilkmson. I. A (1985). Becoming a nanon of readers. Champaign. IL.: Center for the Study of Reading. Barr, R . & Duffy. G (1978. March) Teacher concepnons of reading. The evolanon of a ri search study. Paper presented at the meeting of the Amencan Educational Research Association. Toron a Canada. Bean T W (1980) Can we update expenenced tebichers' beliefs and practices in reading? Reading Horizons. 20. 183-187. Buike, S . & Duffy. 0 (1979. Do teacher Lon..epuons of reading influence innnictional pracnce? Pal.-- presented at t z =hag of the Amcncan EducaLanal Reward-. Association. .ian Francisco. CA. Clifford, G J (1913) A history .if the impact of research on teaching. In R. M. W. Travers (Ed.). Second handbook of research on teaching (pp. 1-46). Chicsgo: Rand McNally. Hams. A I (197f" Readings that made a difference. Those who taught me about reading. Journal of Reading, 23. 6-8. Harste I & Burke, C (197') A new hypothesis for reading teacher research. Bah tetehing and learning of reading are theoret;cally based in P D Pearson & J. Hansen (Eds.), Reading. Theory. resec-ch. and practice (pp. 32-40). New York: National Reading Conference. Hofhaan. A R (1983. May) The reading meshods iesthook. Changes and trends. Paper presented at the meeting of the International Reading Associatioa. Anaheim. CA. Hoffman. J . & Segel. K (1983. May). Oral reading instruclon. A century of controver=. (1880-L..0). Paper presentai at the meeting of the International Readii g Associauon. Ant vim. CA. Kinzer C . & Camck, D (1986, Teacher beliefs as untnutiono,. influences. In J. Niles & R. Lalik (Eds.). Solving prcblerns in literacy Learner, teachers, and researchers (pp. 127-134). Chicago. National Readin3 CoM7erence Logan I , & Encksoh, L (i979) Elementa, teachers reading inservice preferences. Reading Teacher. 33. 330-334. Moore 1 (1981, October) What sufluer.ces a teasher in providing reading instrucron A research view. Pape: presentl at the me..ting ot the College Reading Asswation. Louisville. KY. Ngandu. K 778) Issues and trends IRA rAtional conventions. 1962-1977. Reading Horizons. 19. 123-127 .180 Paris. S. G.. Wixson. K. K.. & Pa Wieser. A. S. ;1986). Instructional approaches to reading comprehension In E. Rothiopf (Ed.). Review of reseorch in etkcation /3 (pp. 91-127). Wt.shington. DC. Americae Educational Research Association. Pearce. J. (1984. November). Sharing the word. Reading for those who haven't time. Paper presented at the meeting of the California Reading Association. Oakland. CA. Reese& C. (1984). A quantitative study of the effect of required reading courses on the teaching of reading skills. Unpublished master's thesis. Kean College of New Jersey. Union. NJ. Robinson. H. A. (1980). Readings thst made a difference. The book that made me flexible Journal of Reading. 23. 296-299. Robinson. R. D. (1979). Reading research. What difterenca does it make? Readin ariZOns. 19. 267-271 Rogers. E. (1983). De-talon of innovations. New York: Free Press. Ruckll. R. B.. & Sperling. M. (1988). Factors influencing the use of literacy research by the classroom teacher. Reserach review and new directions. In J. E. Readence & R. S. Baldwin (Eds ). Dialogues in literacy research (pp. 319-330). Chicago: Plational Reading Conference. Repley. W. & Logan. J. (1985). Elementary ter '-er's beliefs ...exit reading and knowledge of reading content. Relationships to decisions about re ng outcomes. Reading Psychology. 6. 145-156. Sawyer. J. M. (1976). A study of selected prunary teacher 5 preferences in professional publications abota reading. Unpublished doctoral dissertation. Duke Univ !sity. (ERIC Reproduction Service No ED 126 486) Shannon. P. (1982). Some subjective reasons for teachers' reliance on commeicial reading materials, Reading Teacher. 35. 884-889. Stansell. J., & Robeck. C. (1979. November). The deleloprnent of a theoretical orientation to reading among preservice teachers. Paper presented at the meeting of the National Reading Conference. San Antonia. TX. Venezky. R. L. (r984). The history of reading research. II) P. D. Pearson. R. Barr. M. L Kamil. P Mosenthal (Eds.). Handbook of reading research (pp. 3-38). New fork: Longmmi. TEACHING STUDENTS TO LEARN FROM TEXT: PRESERVICE CONTENI TEACHERS' CHANGING VIEW OF THEM ROLE THROUGH THE WINDOW OF STUDENT-PROFESSOR DIALOGUE JOURNALS' Thomas V. Bean and Jan Zulich University of Hawaii at Hilo The process of enculturating future content teachers in the role of guiding students' learning from text occurs during required content area reading me".....YYJs classes. More than 31 states currently require some form of content area reading course for preservice teachers (Bean & Readence, 1989). Most of these courses introduce students to an interactive model of learning from text that emphasizes the role of the teacher as a guide who carefully selects, modifies, or enhances the text (Singer, 1987). The teacher is viewed as one who instnicts and learns in a reciprocal relationship with students, making inctructional decisions prior (Dreher & Smger, 1989; Schon, 1987). to and amidst the ongoing activities of a For example, a teacher may develop three-level study guides and orchestrate small group discussion of the guide questions. The role of the teacher as a faciliWor who helps students cope with often unfriendly texts in mathematics, the sciences, and other fields is the dominant philosophy expressed in many content area reading methods textbooks (Readence, Bear., & Baldwin, 1989; Singer & Donlan, 1989). Despite this national commitment to content reading, O'Brien (1988) argued that "content reading professors wage a frustrating battle to convince preservice teachers ::om a multitude of subject area disciplines of the value of content reading instruction" (p. 237). In O'Brien's skeptical vL.w, these preservice teachers reject a content readi:sg philosophy because of various competing subcultures. First, a student's major tscipline, whether English, agriculture, or .1129- mimics, is a strong force in shaping p.rcep- tions about teaching. For example, textbooks may be shunned by the discipline in favor of practical "hands on" learning. Second, field-tmsed practica experiences in schools exert a great influence on preservice teachers' development. Preservice teachers may hear about t& value of guided teaching, three-level study Odes, and cooperative learning in their university content reading course, but In their school-based 'This paper was part of a National Reads Conference Symposium or "Reading and Learning from Text in the Classroom. The Teacher Variable,' in honor of Professor Harry Sing e. and lus many contributrons to research and NRC. Hwy was a good friend and collaborator over the years. His =wry informed the work of this group. 171 172 Literacy Theory end Research pm-#::1= they often experience classrooms where an assessment-dnven curnculum places a premium on text explicit comprehension. This dichotomy of university and school value systems has been a perennial stumbling block in preservice programs. University faculty operate in a subculture that awards a questioning stance while teachers are often rewarded for developing orderly and productive classroom routines (Cher land, 1989). Field experiznces function at the level of an api.-enticeship where preservice teachers try to adapt to existing practices through imitation. The experienced teacher functions as a skilled technician, delivering instruction. In contrast, the education -ofessor attempts to create a professional educator who reads journals, attends ...,,nferemes, engages in long-range staff developraent, and continues taking university classes. Cher land's proposed solution to this dichotomy is to adopt a model of reflective practice that rewards both preservice and inservice teachers for functoning as professional educators. A third and important fa. Jr interfering with preservice students' acceptance of content area reading was overlooked by both O'Brien (1988) and Cher land (1989). A preservice teacher's individual development plays a pronounced iole in the value placed on the content area reading course. Individual de: elopment in this case refers to the novice teacher's "ihstitutional biography" (Dritzman, 1987, p. 221), or the memory traces of all the individual's school experiences. Some impressionistic data collected by S.cwart and O'Brien (1989) suggests that individual students v.. -y in their perceptions 3f the purposes of the content reading course. These misconceptions undoubtedly stem from the student's individual experiences in education and influence their willingness to integrate guided teaching an.! ..ontent. In the Stewart and O'Bnen study preservice students were asked why, they felt a content area reading course was renvired Based on a qualitathe analysis of students' tesponses using constant comparison to generate cau pries, four dominant categories were constructed. (a) on target'b' diagnosis and remediation, (c) self-remediation, and (d) no idea. These research.as found that 23% were on target, 34% saw the course as a way to learn diagnostic and remediation skills, 39% viewed this as al opvvrtunity for self-remediation; and 4°4 had no idea why the course was required. The present study explored preser ice content area teachers' changing views of a required course thr_ ugh the window of student-professor dialogue journals (Bean & Zulich, 1989) Students and the professor wrote for 10 minutes at the beginning of class and then exchanged their journals for mutual response prior to the next class meeting Typically,, six journals were collected and one student took the professor's journal to r ad and respon,' to. The content area reading course involved a fieldbased observation ai..1 participation experience in juniar and senior high schools in a multicultural community. The case study method was adopted for this study (Merriam, 198i We conducted an intensive content analysis of three student-professor dialogue journals representing the following disciplines. (a) English, (b) agriculture, and (c) mathematics. Our ra ionale for selecting these particular content fields centered oh the view that we might expect a prospc-tivt English teacher to be more receptive to a ontent an..a philosophy while studeMs in the lisciplines of agriculture and mathematics nuE,ht be less reetpt've, given the 'ess text-bound nature of their fields. 183 Teaching Students to Learn from Text 173 METHOD Subjects Three students who kept dialogue journals over the span of the content area reading course were selected for analysis. Pseudonyms were created as follows: Sarah in English, Lee in math, and Karen in agriculture. Sarah was a mature returning student in her senior year with teenage children of her own. Karen and Lee were both seniors in their twenties and single. Sarah and Lee were of Caucasian ancestry and Kaim was part-Hawaiian. Materials and Procedures These 3 students averaged eight dialogue journal entries during lie semester of content area reading. Each was transcribed for analysis. We used 3 x 5 notecards for initial analysis and interpretation. In the initial stages of data analysis, both researchers, as a form of niangulation, independently read through the 3 students' journal entries from initial to final entry, teking notes on the 3 x 5 cards. Following Merriam's (1988) case K.rdy procedures, we jotted down notes that reflected general impressions during our reading and analysis. For example, a student entry that revealed some trepidation about engaging in the first day of observation/participatioh in a school caused us to jot down "student worried plea going into a school. These initial notes were then organized into categories. Ten categories emerged that encompassed students' comments. (a) acknowledging the value of content area reading, (b) attitude toward reading, (c) specific content area reading strategies, (d) concerns typical of preservice teachers, (e) evidence of discipline-based subculture membership, (f) uncerte;-ity about teaching and one's own ability, (g) immersion in the field experience, (h) change in perception of content area reading, (i) interest in political and professional matters, and (j) general comments to the professor. Constant reinspection of the student entries resulted in the final development of 4 categories that subsumed many of the elements of the original 10. The final four categories were: (a) value of content area reading, which encompassed references to specific stiategies and changes in perception of content area reading, (b) preservice thoughts, which includ uncertainty about one's own tz4ching ability, ((..) professional immersion, which included evidence of subculture mem-aersnip and interest in political and protessional issues, and (d) attivide toward reading. RESULTS A frequency count of the number of times Sarah, Karen, and Lee made comments in the four categories v, as revealing. Actual student remarks at the sentence level in the journals comprised the data for this frequency count. Table 1 displays the frequency of student comments in each of the four categories. These data reveal the diversity in students reflections about ccntent area reading. However, the data require further discussion to fully appreciate these differences. 174 Literacy Theory and Research Table 1 Frequency of Student Comments in Each Category C4,..agory Value of Content Reading Preservice Teacher Thoughts Professional Immersion Attitude Toward Reading Sarah Karen 10 7 7 2 5 3 6 4 Lee 9 English Major Sarah, as an avid reader, had 10 positive comments on the value of cotrtent area reading. For example, she commented, "I'm going to be spending spare time with Chaucer. I'd like to incorporate text previews with this unit and vocabulary preview." She knew and was able to discuss specific teaching strategies she consideresi incorporating in the teaching of English literature. Throughout her journal she graOled with how best to integrate content area guidance and the time constraints of teactieig. In the category ref preservice-teacher thoughts, seven entries displayed anxiety about acceptance in the school. Sarah wrote, "So now I know I'll be out at Pahoa High, School for my O.P. and hope I'll have a peat English teacher to model after. Part of me just wants to jump in and get involved with a real class and real students (enuf already of the theories etc.). The other side of me anticipates with nervousness the idea of being up in front of the class-30 pair of eyes critically scnitinizing the new teacher!" Within the category of professional immersion, Sarah had five mentions. She wholeheartedly embraced her placement in a rural 12th grade English class, immediately rereading Wuthering Heights, the initial text, and later plunging into her unit on Chaucer. In addition she occasionally commented on the education courses, with some displeasure at the amount of theory introduced. Sarah had a markedly positive attit.ue toward reading. She alluded to her own positive view of pleasure reading six separate times. She commented in detail on the fantastic old copy of Wuthering Heights she found with wxxlblock prints. She argued that we all need pleasurable escape reading, demonstrating 'coal an English major's perspective and a view espoused in the content reading course. Agriculture Major Karen presented a distinctly different picture across the four categories. She mentioned the value of content area reading seven times but her comments seemed more like platitudes than philosophical anchors for her teaching. rar example, she said she would use the many strategies learned when she is a teacher. She also related these comments to her own reading ability. Karen displayed a good deal of insecurity in readkg throughout her journal comments. Karen's preservice-teacher thoughts consisted of only two mentions. She wrote, "I only hope that I'll be able to provide a mwarding educational expence to all Teaching Students to Learn from Text 175 students." Similarly, in the area of professional immersion, ;he had only thnx comments. Karen's attitude toward reading consisted of four negative comments. She men- tioned a strong aversion to reading required texts but also a lack of time for leisure reading. "Never too late," she told herself. This personal struggle with reading was a strong undercurrent throughout all her writing and thinking. Mathematics Major In contrast to Sarah's keen interest in many d_aensions of content area reading and teaching in general, and Karen's focus on herself as a learner and reader, Lee, our mathematics teacher, came to us with a schema for teaching and ;earning firmly rooted in the mathematician's subculture. In the category encompassing how students valued content area reading, Lee made only one passing comment, which occurred in the first journal entry, and displayed resistance. In this particular entry, Lee referred only once to the nature of the class, calling it a "reading course." In this same journal entry Lee provided the matnematician's subculture perspective on content reading, commenting that, "students don't read math texts except to copy an example." Claiming to have reached the level of "math maturity" where one sees the value in reading mathematical materials, Lee admitted, "Even now, I don't get overly -xcited with reading." He saw little productive value in having students write, co_ .aing it to be "the English teacher's job." His journal continued to demonstrate an avoidance of content area reading. Lee broached the topic of preservice-teacher thoughts nine times. His initial concerns about placement in an observation/participation assignment echoed those of his peers in English and agriculture. He worried about being liked by the students while tiyin; to maintain a separateness he viewed as crucial for a teacher. He quickly embraced the role of an apprentice to the host math teacher, imitating the teacher's approach. Lee mentioned six professional immersion statements or questions that displayed the st...ing pull of subculture membership in mathematics and the culture of the school. His entries were analytical. He often referred to teaching and learning in cause-effect terms, as in, "They're working hard and it is paying off." However, like other novice teachers, Lee valued teaching successes from a predominantly egocentric stance. Studies of preservice teachers suggest that they are often focused solely on personal needs like comp:aing the planned lesson and being liked by students (Fuller & Brown, 1975). In terms of attitude toward reading, Lee ignored this dime.lsion. Re had no entries that referred to his own reading abilities or interests apart from mathemaics, in marked contrast to his peers in agriculture and English. DISCUSSION One of the most striking features of this case study analysis is the variability in role perceptions acrel.s these three students. Although they share many of the 176 Literacy Theory and Research chafacteristics of preservice teachers, there i less commonality on this dimension n we expected. Our interpretation of this variability relies on three important factors that seemed to shape these students' views of their roles as content teachers in relation tO the required content area reading course. First, the particular subculture of the discipline does indeed, as O'Brien (1988) found, weigh heavily in how each of these individuals perceive their teaching role. Second, the disparity between the culture of the university content are :eading course and the school site interacts with the particular content discipline to induce an apprenticeship stance rather than that of a reflective change agent (Cher land, 1989). Finally, a student's individual development, consisting of past experiences in school situations, plays a heretofore uncharted role in shaping how a person will manage the multiple and often disparate cultures of the discipline, the school, and the content reading course. We discuss each of our three students in relation to these three competing factors. Sarah, our English teacher, represents a discipline with strong links to the value system of guided teaching. She readily embraced text previews and vocabulary strategfrs that would illuminate the obscure language of Chaucer for her high school studentr She displayed concerns typical of preservice teachers in terms of preparation and acceptance (Livingston & Borko, 1989). Yet she was a mature, cnficient individual with an stitutional biography of largely positive experieoces who departed from the typical apprentize stance charted by O'Brien (1988) and Cherland (1989). Indeed, this factor of individual develepment needs further study in relation to subctdture discipline and the culture of the school. "Lifelines" (Jeweler, 1989) are student created visual antobiographies of past experiences. These student-constructed drawings provide a means of accomplishing Britzman's (1987) recommendation that teacher educators 'rip novice students reconcile their own educational ext .ences in relation to th-ir dtvelopin? role as a teacher. In addition, as a research tool, they may shed light on a novice teacher's individual development when combined with dialogue journal entries We are only beginning to explore this additional source a:biographical information. Karen, the agriculture major, displayed concerns about her own reading skill. These concerns seemed to override subculture and school-based issues. Stewart and O'Brien's (199^' study found that 39% of their content area reading students saw the course as a to self-remediate difficulties they might be having. Karen comtr ented that ..ad "enormous amounts of reach..." in all her course,,, difficulty remembering what she reads, ar -1 requests for suggestions on how to solve these problems It is difficult to speculate on how Karen's struggle with reading might carry over into teaching where modeling enthusiasm for reading influencer students' attitudes it' 3 study of 1,000 preservice teachers' reading autobiographies, Manna and Misheff (1987) found that 28% of the students mentioned that reading was not a priority in the homes where they were raised, and th:s f41tor caused them to lack confidence in reading throughout thei: lives. Karen mentioned she read romance novels at one time and now feels guilty about not taking time to engage in leisure reading. In her field placement she worried abuut agriculture whex few textbooks are used As the course progressed, Karen mentioned valuing the class, yet she mentioned few specific strategies apart from an emphasi. on technical vocabulary. Unlike Sarah, Karen is a more traditional college student without the extensive 187 .$ 177 Teaching Students to Learn from Text life expenences Sarah brought to the class. This developmental factor, coupled with the subculture of agriculture where "hands on" learning is favored over texts, woule normally result in some tensioi. zNer the philosophy of content reading. However, because Karen focused on her own attitude toward reading and her own difficulties with learning from text, thi.. ;actor reduced the strong subculture alliances O'Brien (1988) observed in his course. Finally, Lee, our mathematics major, was most strongly influenced by the subculture of his discipline and the culture of the school. The content area reading course was simply a hurrile to be overcome along the way to becoming _ math teacher through imitation of an expert. Lee viewed himself as an apprentice to a mentor teacher who held the truth for one in math. Lee's indifference toward reading in general and content area reading in particular may have bP-en partially related to his individual de.elopment. Like Karen, Lee was a younger student, caught up in problems of dating and breaking up, purchasing an old car, and finding his way But his journal reveals a genuine alliance with the discipline. This alliance, in our vie , overshadows the issue of individual development in Lee's case. Thus, we would argue that the issue of how preservice teachers go about inte grating content area reading concepts with their individual schemata for teaching is highly variable. Based on nese case studies, we believe that the competing forces of d.scip1ine and schoel-based cultures, as well as individual development, play uncharted roles in a preservice teacher's efforts to function successfully within the profession. We are in the process of conductint, a long range study of our students' dialogue journals at various stages in these nov ice teavhers' professional development to investigate these factors in more detail. REFERENCES 1.32)-1, T. W & Readence, J E. (1989). Content area tradmg. Current state of the art In I Flood & D. Lapp (Eds.), Content area reading and learning. Instructional strategies (p 14-23) Englewood Cliffs, NJ: Prentice-Hall. Bean, T. W., & Zuhch, J. (1989). Us.ng dialogue journals to foster reflective practice with pre-service content area teachers. Teacher Education Quarterly, 16, 33-40. Bntzman, D. (19871. Cultural myths in the making of a teacher Biography and 3ocial structure in teacher education. In Okazawa-Rey, M., Anderson, J., & Traver, T. (Ed..), Teachers. teaching, & teacher education (pp. 220-233). Cambridge, MA: Harvard Educational Review. Cherland, M. R. (1989). The teacher educator and the teacher. When theory and _awe conflict Journal of Reading, 32, 409-411 ,). The teacher's role in students' success The Reading Teacher, 42, 612-617. Fuller, F. F., & Brown, O. H. (1975). Becoming a teacher In K Ryan (FA ), Teacher Education Seventy jourth Y earbook of the National Society for the Study of Education Chicago University of Chicago Dreher, J., & Singer, H. (I Press. Jeweler, J. (1989). Lifelines and the freshman expenence Unpu'..:.bhed seminar Hilo, HI The University of Hawaii at Hilo Freshman Experience Program. Livingston, C., & Borko, H. 0989). Expert-novice differences in teaching A cognitive analysis and implications for teacher education. Journal of Teacher Education, 40, 36-42 Manna, A. L., & Misheff, S. (1987). What teachers say about their own reading develc.ment Journal of Reading, 31, 160-169 178 Literacy Theory and 11%seareh, Merriam, S. B. (1988). Case study research in education: A qualitative approach. San Francisco: Soucy- Bus. O'Brien, D. G. (1988). Secondary preservice teachers' resistance to content re*ding imtraction: A proposal for a broader rationale. It J. E. Readence & R. S. Baldwin (Eds.), Dialogues in literacy research (pp. 237-243). Chicago: National Reuling Conference. Readence, J. E., Bean, T. W., & Baldwin, R. S. (1989). Content area recv"Ig: An integrated approach (3r1 ed.). Dubuque, IA: Kendall/Hunt. Schon, D. A. (1987). Educating the reflective practitioaer. San Francisco: Jossey-Bau. Singer, H. (1987). An instrucdonal Model for reading and lerning from text in a dustpan setting, Patrol af Reading Education, 13, 8-24. Singer, H., & Donlan, D. (1989). Reading and learninu front text (2nd ed.). Hillsdale, NJ: Erlbattm:, Stewart, R. A., & O'Brien, D G. (1989). Resistance to content area reading: A focus on preservice teachers. Journal of Reading, 31, 396-400. 189 THE L1FLUENCE OF LITERACY-ENRICHED PLAY SETTINGS ON PRESCHOOLERS' ENGAGEMENT WITH WRITTEN LANGUAGE &man &Neuman Temple University Kathy Roskas John Cannil University Of the range of activities that may engage young children in literacy, none is perhaps more self-directed than play. As children discover and invent literacy through play, they develop important generalizations about written language as a meaningful activity (Harste, Woodward & Burke, 1984). Indeed, recent research suggests that preschoolers' active involvement with literacy through play is an important developmental stage in becoming literate (Goelman, 1984; Y. Goodman, 1984). In this capac- ity, play offers enormous potential as a resource for literacy learning in the early years. There has been a tendency, however, to overlook play as a curricular tool in emergent literacy. Although Vygotsky (1978) hypothesized a central role for play in literacy development, the traditional notion that children are not "ready to read or write" before schooling has steered educators away from promoting literacy behaviors in play environments. Only recently, in fact, has play even been seriously considered as a rich contextual setting for observing emerging literacS, behaviors (Ga Ida, Pelle- grini, & Cox, 1989; Jacob, 1984; Pellegrini, 1985; Rowe, 1989; Schrader, 1989; Yawkey, 1983). Overlooking play as a context for written language has had at least two unfortunate consequences. With the exception of the book corner, print has not typically flowed through in-school play environments as it may in many homes and community setlings (Leichter, 1984). Since it has been documented that children are spending an increasing amount of time in these early childhood settings (Kagan, 1989), fewer opportuni- ties may be available for them to become involved in naturally occurring literacy routines, such as going to the grocery store, the Post Office, and the library. A second consequence closely follows. By not using play as a context to foster literacy, professionals are missing out on opportunities to promote children's emerging conceptions of reading and writing from a developmental perspective. In postponing written language turning to be taught more 'formally" outside of a play setting, we run the Ask of having literacy become less contextualized and less functional, and, therefize, less meaningful from the child's point of view, making literacy learning at some later stage seem unnecessarily difficult and irrelevant (K. Goodman, 1986). In contrast to these practices, we propose that play can serve not only as a 179 190 180 Literacy Theory ai.d Researc cunicular u)ol in support of literacy development, but also one that influences it. If afforded more opportunity to engage in literacy-related play, children may reveal and share with one another their preferences and competencies, and in so doing, creat the conditions necessary for socially mediated literacy learning. Indeed, play may serve as an important resource for childre:, explore their developing conceptions of the functions and features of priat in the preschool and primary school years. To explore this hypothesis, we designed a study to examine the influence of literacy-enriched play centers on preschoolers conceptions of prl-u. Specifically, tht study addressed the following questions: (a) Do literacy-enrichk , play centers influence the fnequency of literacy demonstrations in the spontaneous play of preschoolers? (b) Do these play centers enhance preschoolers' concepts about print? (c) In what ways might physical design changes in the play environment influence the name of children's print activities in play? METHOD Subjects and Setting TUT- seven children (25 boys, 12 girls), ages 4 and 5, from two urban preschool classes (N = 20; N = 17), participated in the study. The preschool servta families from diverse ethnic backgrounds (83% Caucasian, 15% Black, 2% Asian; and socioeconomic status levels Both classrooms were in close proximity to each other, were similar in spatial arrangement, and included identical play areas. housekeeping, blocks, small manipulatives, book and art corners. Few print r, .terials, aside from books in the book corner, w^re includod in these areas in either classroom. The huetvention Design To examine the effects of literacy-enriched play centers on children's literacy demonstrations, the physical en% ironments of both classrooms were redesigned. Since design changes may effect children's play behaviors ma., broae'v (Johnson, Christie, & Yawkey, 1987; Morrow, 1989), it was important that these ra ai.ications take into account existing spatial arrangements as well as common functions of reading and writing among pieschoolers as evidenced in our previous research (Neuman & Roskos, 1989) In this respect, an understanding of :he environmental setting ;Is well as information from the child's point of view were considered in the intervention design. With these con-lderations in mind, five basic design hanges were made in the classrooms. I All play areas were more dramatically carved away from one anothei .ad clearly marked a.zing semi fixed features, such as cupboards, screens, tables and hanging mobiles. 2 The labei:inc, of items in the physical environment was inaeased. For example, ctorage bins for blocks and art materials were identified by illustrated and printed 3 signs. Four distinct play centers were created. Post Office, Library,, Oface and Kitchen. These centers, resembling activities familiar to children, might Le easily linked to - Jit4tacy:tiCtivitiei:tintside theTteSchool, and thuslmight. help-to faciliMte.:ivritted_ 41,104654**Y3liztP4.)014iiF 04.1**- PhYSical uPace, Vas : reairatiked,lo alloW'for:MOVeMent,ibetWOn,th9 `.iitc*Y-0.0.00 centers Basedn previous #telfi#0 -imity of i&cifiC play Centersjppeirod,O'foSter ---,1410.740,,t490.-w101000;ciint`91 4.0 04 tgit.egi.j.144 lioYyOUng children)authintic- gy '(a ...real-,it604#- the genital 'envitaninent),, and atilitY.(itiefUltiesiiii:141400., their "itnitativelittrioy attaMpts). Figurel illustrates the tiesign of the play environments, along with an.akhreViatL& list of literacy props. AWORR.grMS AREA qT TAUS ART asPIT.E. ;sample* of literacy prone in centers: Nitchen plev Center Telephone books A telephone Emergency number decals Cockbooks Food coupons Grocery store ads Post Office Cenlet Stationery and envelopeo Hail boz Computer cnd address labels Posters and signs about zip codes Tote bug for mail Qiiisamiismssnisz Calendars Appointeent bc A Signs Magazines for waiting TOCU Assorted ferns labrory Pley Center Library book return curds Children's books Stamps for borrowing Book narks Sign in/sign out sheet Figure 1. The design of the play environmenis with literacy props. .alaamarAaosmaammanffigAgtlealls. 182 Literacy Theory and Research Procedure Prior to the intervention phase of the study, three measures of literacy behavior in play were obtained over a 2-week period. First, using an observational procedure developed by Singer and Singer (1980), each child's actions and language (verbatin) were recorded during their spontaneous free play time for a 10-minute period on fonr separate occasions by two trained observe/a who were graduate students in language-, arts. Using videotapes of preschoolers literacy in play from our preyiolia4Mciy_ (Neuman & Roskos, 1989), observers were trained not to interpret behaitior; biit=10: record what actually occurred during observational periods. A total of 40 Minutes of observation was recorded for each child, yielding 148 play protocols. Seetnid, activity in four different 8-Tas (housekeeping, book corner, art table, ManipulativesT board games) was videotaped for 30 minutes, four different Vines, for a total of.2 hours ner play area. Third, each child was individually administered the "Sands" booklet of the Concepts about Print (CAP) test (Clay, 1979). Following these procedures, the physical play environments of each classroom were enriched during nonschool hours with literacy-related materials. Over the next 4-week period, no formal observations took place as children became accuStonied. to these design changes. During these free play periods, teachers and aides were encour aged not to intervene or restrict any areas, but to allow children to freely move through all the play centers. Using the same observational and videotaping procedures, children's play was then systematically observed once again during a 2-wmk period. "Stones," another form of the CAP assessment, was administered to each child. Analysis Play protocols were analyzed for evidence of literacy demonstrations, defined as instances of reading or wrfting-like behaviors, Such examples included scribbling, marking on paper, pretending to read, book-handling, or attending to print in some manner. Boundaries segmenting each play behavior that included literacy demonstrations were established to allow for the coding of each demonstration. Two indicatcrs were used to establish boundaries: shifts in the focus of the play acti vity (e.g., switching from playing in the library to the kitchen), and shifts in interaction between the players (e.g., a player initiates play/talk with someone else on a new topic). Coders counted the number of literacy demonstrations for each child during the 40 minutes of observation prior to and following the intervention period. Two research assistants independently coded a sample of 20 protocols to determine the reliability of the coding procedures; intercoder reliability indicated .98 ageement. Differences of means tests were used to analyze pre- and post-intervention changes in literacy demonstrations and children's concepts about print. Videotaped play activy was qualitatively analyzed using the ethnographic procedure of typological analysis (Goetz & LeCompte, 1984). Play framesplay bound by a location and a particular focus or interaction (Bateson, 1955; Sutton-Smith, 1979)were established and 34 literacy-related framer were isolated for further analysis. Successive viewings of a sample of 10 randomly selected frames, prior to and following the intervention, were examined to analyze in what ways the literacy en- 19,3 r; , s-- - 183 Literacy-Enriched Play Settings richment may have influenced preschoolers' literacy demonstrations and play behaviors. RESULTS Since there was no attempt at this stage in our work on the literacy enrichment of play environments to obtain a matched sample as a control group, the following results are suggestive and should be interpreted with caution. Fmdings reported in Table I indicate that the average number of literacy demonstration in play rose sharply over a 2-month period, with young children spontaneously using almost twice as much print for play purposes than prior to our intervention. Children's CAP scores, as well, rose significantly during this period Although it is impossible in this study to suggest a causal relationship (i.e., literacy demonstration "caused" higher CAP scores), these results do suggest that environments "littered with print" can promote children's imeractions with-literacy. Though these quantitative ?flanges are important, the qualitative analysis revealed a number of more subtle and complex changes in literacy demonstrations cl ithin a literacy-enriched play environment. These changes were characterized by r...1 trends reported in Table 2. One was the striking increase in the duration of literacy demonstrations. Whereas demonstrations prior to the enrichment tended to be quite brief, those following the physical design changes were far more sustained. For example, the average duration of literacy-related play frames was about 1.7 minuws prier to our intervention; following the enrichment, these frames lasted approximately 5.18 minutes. Related to this trend, was a marked charge in the density of literac-y demonstrations, that is, the number of demonstrations coded within individual play frames. Following the intervendon, these demonstrations seemed to interlink, forming chains of related literacy behaviors. For example, before enrichment there were few connected demonstrations, averaging approximately 1.57 within a literacy-related play frame; after enrichment, however, the literacy-related play frames contained on the average of 9.0 literacy demonstrations. In fact, at times, the literacy demonstrations in these chains were so closely connected that they became almost indistinguishable from one another. Table Effect; of a Literacy-Enriched Play Environmetu: Mean Score Differences Prior to Enrichment Literacy Demonstrations Conventions About Print (Clay, 1979) *p<.01. **p<.001. Following Enrichment SD M SD 1.51 1.95 9.16 3.90 2.83* 11.51** 2.90 3.77 Variable 'Table-2. Prior.M'Enrichment Variable: 'polio-Wing - 1.71 1.57 Density - .42 .49 M`eisured in minutes. p<X11. **p.001. A brief comparison of two play frames illustrate these two trends. For #ample;,, prior to efiriehnient: _ Michair ilplaying house with-Scott in the housekeeping, area._te -basic:046f drawing paper and a,bOx of markeia. Afterthe boya Sit downit the.tahleand confer briefly_Oer thz-paPer,,tbekii4in to play with dfiTereikt color ou the paper. Thek then Aurit-to sorting a,-; stacking=pOil and:r44*-§#1".*: - cupboard in preparitiOn for cooking. All total; in this play frame,,there was one literacy derocitistratiew:Whiek4aStest,...,;;: approkimately 30 seconds. As indicated .here, writing aPPeared,to..belhelodus:Of exploration, -rather.than used in the service of the play therebreadly.-In i:ontrast, literacy-related play themes in the .enriehed play2centersrappeared--: more instruMental to the play experience, andtherefore, Seemed to set Off'a-chain literacy-related demonstrations Oflonger duration. For exaMple: Michael and Scott are in the Office play Center. TheY atepliying "sign-UpTheYi want peoPIe "to sign-up" for the homeless. Scott.has a'sraall clipbOard'and,'peiteil. lie circulates throughout the claisroom, asking differeatteleheriaO`Cbildlitip'sik*, their names on his Clipboani.Michael rernains hi the Offiee."Writhe Periodically he-looks up and directs Scott to aslc so-mane:else. FMStio;ti,iftUriiii; with a list of signatures. Beth boys-pretend io "enter" :the- list iitio*cOinpoter. Scott points to names on the list and Michael tines. When dOne,.SiOtt reMeires,the paper from the clipboard and is sent out again to gather More This literacy-related play frame, lasting approximately 15 minutes,. Suggests:MIreading and writing activities ,became more integral and, useful to the,,aot4414tOVP-9.C: the play itself. In fact, in this instance the print activities are the action 4kich, bound the play into a coherent theme. , Upon closer examination, the typological analysis of these literacy-relatedplaY..; frames revealed five characteristics of change in the literacy demonstrations inclicati'i.e,' of the aforementiOned trends. Literacy demonstrations in the enriched play centerS became more, Usefid; t purposeful, and more unified. Cnildren used literacy-to obtain and,convey informationvital to their play schemes such as to "sign up" for activities and t6 "read'-''recipes for meal preparation. Rather than incidentil to the ;play, literacy was:PreSsedl#:.;,., serrice for Some larger goalthe realizatial of play-purpose; Demonstrations became more .situOred.11.--q,licitplay ormtexts,:,such as the.Offiee: and the Post Office,,prOvideddistinetjfi,00s of-referent:es and the litericy,props:in each center served as Concrete cues for *racy Uses. OonteXtnal . _ Literacy-Enrichid-Play Settings supportud and helped to situate play themes strengthening the fabric of play arfl -guiding its course. For example: Hilary and Dana are sitting at the table in the Post Office play-area. Before them are envelopes, seals, stamps, and a mailbox. They are writing letters, then inserting them in the mailbox at the corner of the table. They have repeated a "write a lettei , put in an envelope, seal, stamp, and address it" procedure two times. They than their' letter!, retrieve them and then pretend to read the messages. _ Suddenly, Hilary begins to scribble rapidly on her paper. She shows Danaher scribbles and Dana says, "Baby, you're bad!" They both giggle and continuernakink and sending letters to one another. Further, in contrast to the literacy demonstrations prior to our intervention that seemed frequently isolated and randomlike, these literaey demonstrations became increasingly more interactive as children used literacy as a medium for social exchange. Kent and Ricky have been trying to get Dana's attention for some time during the play period, but she has not responded. They retreat to the post office and decide to write to her. Huddlii.4 together, Kent dictates, "Dear Dana," and Ricky writes. Finally, the note is finished and inserted in an envelope. The boys walktoward Dana at the office play center and pretend to knock. They approach Dana. "Dana! Dana! Here!" They hand over the envelope giggling and covering their faces with their hands. She smiles, scans the envelope, opens it, pretends to read the note, and returns it very carefully to the envelope. Kent leans toward her and says, "You wanna come over to our house?' Then they both run back to the post office and repeat the entire note writing routine. Along with the greater interaction, came a dramatic change in role-taking as children became postal workers, office managers, advocates of social issues, and librarians. In the following example, children are demonstrating their knowledge of library routines and authenticating role definitions. David and Scott are playing m the Library Center. David runs over to the teacher and says: "Mrs. G., want to come and get a book?" She comes to the center. Both boys help her look for a book. Mrs. G. selects Where the Wild Things Are. Scott then "marks" a card with a stamp. David records the same information on a larger paper. Scott tells her that the check mark refers to the day she must bring the book back. In sum, rather than isolated instances of children scribbling or coloring, literacy demonstratio within the enriched play environment tended to become more functional for these young children and more embedded in their play activity. . As a result, the play itself appearecl to lend mater coherence and meaning to literacy, supporting the children's exploration of its multifunctional nature. In short, as these two activities interacted in the enriched centers, a new and more dynamic relationship between them seemed to emerge. CONCLUSIONS The results of our study suggest that literacy-enriched play centers have the poten- tial to influence young children's literacy activities in early childhood settings. The contextualized settings and ready availability of numerous literacy-related props 186 Literacy Theory and Research neared to influence the frequency of children's engagement in reading and writing activities through play. We would argue, however, that even more important than frequency, these enriched play centers fostered more sustained and involved literacy interactions. In doing so, the literacy demonstrations began to resemble what Heath (1982) defines as literacy; events, that is, "occasions in which written language is integral to the nature of the participants interactions and their interpretive processes and ,strategiesi'' (v59).:A.S. these demonstrations appeared to gain event-like status, they began to influenee not: only children's participation in reading and writing, but the play itself. ióughchil dren's playful elaboration and uses of written language, play as a tool to make sense: of and learn more about their world was enhanced as well. Our study, of course, cannot provide evidence that enriched play centers necessarily result in increased literacy learning. More controlled studies across varying gtouns of preschoolers in different settings need to be conducted before srch weighty claims. can be made. However, our study does suggest that literacy-enriChed play centers, can indeed, inalce a difference in children's literacy behaviors through play. With well-planned design changes in the physical play environment, play may became an increasingly important context for the discovery and exploration of written language. 1LiFERENCES Bateson, G. (1955). A theory of play and fantasy. Psychiatric Research Report, 2, 39-51. Clay, M (1979) The early detection of reading difficulties. A diagnostic survey with recovery procedures. Portsmo.1, NH: Heinemann. Galda, L , P dlegrini, A. D., & Cox, S. (1989, March). A short-term longitudinal study of preschoolers' emergent literacy. Paper presented at the American Educational Resealed Association Conference, San Francisco, CA. Goelman, H. (1984). The discussion: What was said. In H. Goelman, A. Oberg, & F. Smith (Eds.), Awakening to literacy (pp. 201-213). Portsmouth, NH: Heinemann. Goetz, J., & LeCompte, M. D. (1984). Ethnography and qualitative design in educational research. Orlando, FL: Academic Press. Goodman, K. (1986). What's whole in whole language. Portsmouth, NH: Heinemann. Doodman, Y. (1)84). The development of initial literacy. In H. Goelman, A. Oberg & F. Smith (Eds.), Awakening .o literacy, (pp. 102-109). Portsmouth, NH: Heinemann. Harste, J , Woodward, C., & Burke, V. (1984). Language stories and literacy lessons. Portsmouth, NH: Heinemann. Heath, S. B. (1982). What no bedtime story means. Language and Society, 2, 49-76. Jacob, E (1984) Learning literacy through play: Puerto Rican kindergarten children. In H. Goelman, A. Oberg & F. Smith (Eds.). Awakening to literacy (pp. 73-83). Portsmouth, NH: Heinemann. Johnson, J F Christie, J. F., & Yawkey, T. D. (1987). Play and early childhood development. Glenview, IL: Scott, Foresman. Kagan,S L (1989). Early care and education: Reflecting on options and opportunities. Kappan, 71, 104-107. Leichter, H. J. (1984) Familiar as environments for literacy. In H. Goelman, A. Oberg, & F. Smith (Eds.), Awakening to literacy (pp. 38-50). Portsmouth, NH: Heinemann. Morrow, L. M. (1989). literacy development in the early years. Englewood Cliffs, NJ: Prentice-Hall. Neuman, S B. & R%kos, K. (1989). Preschoolers' conceptions of literacy as reflected in their spontaneous play In S McCormick & J. Zutell (Eds.), Cognitive and social perspectives for literacy research and instruction (pp. 87-94). Chicago: National Reading Conference. 197 a ^ eYra,s*. PYLY.11,1411t,51., te behavioc A re** 11,0 cOgia Of, ,EdualtioisalRe*edick4.7,-107:121.., Rc40.4';K.',09,PY,'PiteaFt at v'faiic idge ine.ltrategica in Sett-seJeCteit itiMve; D. W. (1489):- Prischoeleti', use bi rueSictignitiVan'o41 eVentsAn S. IACCOinsitiA J. :*.ilel!:T4114cagrqsfre.,fu,4 sockdperspecsivcsfor litar-c7, ,,:c,..:,(14119). Yritten" kitguagc use within the context of ytrug s Maw, Pd Rfi*e.firch aufnu.n,y; 4. 2257214 inPrachoolers: A Sfiiii; D. (190): Teleilslitnyi0eirig and irwaiiiiitiVe .. ,dplay . , iP.PPr,t 6.'0* Pill*e.C.F.0.040:01. (PP..0);"*C4**''''' *.u.t0. *nth, 4 Ada in thclitY-ClanliS40.: Hsr.ard University Pines. 'ijawkey, T. (1980.-:Preienti play and loin,* growth In irioe chIldiewli,avotslikParic aiit State University. (ERIC Document Riiiroducties Serviceifo.'Eb 231`,55* , 198 fonneylva= silAtiEp;BooKgs4p:pip.; sTAAT pjtopRA44Fp 'sA Jana M.. MOO; Bonnie Ma Kerr, tioct.S4Ohits Univeriity of Christine -MaCprinick Eastern Illinois University There is increasing evidence thas children from socioculturally ave.* hornkin,: this country:areat risk for school :fidinre ;(Anderstm; 4 *ofccs, 1984, Teale;;186), in part becatise our schools,place a high value, On thedoniinant *Ole= class ,apProaclr to: beecnning literate. Teachers ;often emi*It ;01,:iscizOA1-41*074. able te participatem book reading diseussions iz a sinadar mann:er and to 40040= eficeit;siniil*Biefacy.'eveots and pradfices .with their Pare* liaret;beelias:efa4ilia'' do not engage inidentical literacy practioes or interact With their children in,thesaine Ayays children:0nm to school with varying knowledge about literacy ,and:!aryirig intereatin.itOecinisition. Flew might, our public schools best meet the needi of children with diVerSe,baCk. grounds = and-disPositions? One , hypothesis is that an early and Intenake, literacy will lead ta,greater awareness of and interest iniotAiiig and wife*: 'Fici*yef; nOwn about how to-identify children,who:Mightlienefit from 40.tylito*. exPeriences- in a school setting; hOw suCh a:program Might tO be Organi*,;:Pi what might be possible, long-ternthenefits. This study Was.set up to ckterinine the effeefiVe,ness.of early, and intensivenexpoSureto materialsrdesigoed to proniateeineitent leadPreseloOlerS. ing, hi a languageand literacy-focusedprograrn: fOr The study eVolved from the series of oidi6s.(masop., 1986.; McCormick &Mason, 1989a) int which.the use cifeasyfto-reciteAtti,41, has been shown to rnatch young childrenia interrat hi, print and impact on the, early:reading skills of Children ,V.,,ho tyPical/Y dci`,not prosPer tinder the aystematic basal reading instruction _in schotil girnple, short storiai-were cOnsiniated to provide obvious connections between, Spoken, and printed, Words s4F.i ,thatA,:a44-$:, year-old children could readily learn to recitetilehookS. Theiernaterialayveredefel= oped :a the context ,of Mason's (1980) developinental'iriodei of early reading..#1,:this, model Maion Proposes a first level of readini,'deVeltnentla Which:ehildreir reef-44: WO:print ,hy using ,themtent of a message lyithin 4conteltt thiallinechildren'hegin to"recagnize and naMe ierterShut ad,ofi.,:,1,46,404040090, °te learn,or remeinher words. Aecording tO yeneikY,(197S);kariwledge of letter nanies, faCilitates/the ;process of reading 17.y, pi4iilgstljelatters°imrofiatey,foniliar, which ta aseobiató (1984 aigops -that letter names ,gie ideritifialile,-refererits _ , 190 Literacy Theory and Research phonemes. This initial level of understanding is followed by a second level of reading' development in which children become aware that letters signal particular sounds and, that these phonetic sounds, usually beginning with initial consonants, can be heard in words arrl used as cues for word recognition. Children who have not experienced informal literacy activities that are compatible with the first level of reading may be at risk for failure if they receive the typical reading instruction in kindergarten and first grade which emphasizes adtivities that match the second level of development. Walsh, Price, and Gillinghanr (i9fi8),.fOurtai letter naming knowledge (Level I knowledge in Mason's hientrchy),varied:Widely In the middle of kindergarten and letter naming speed was strongly related to later progress in reading. The Little Books are materials to be used in activities apprppriate for children at the first level of early reading in that they oiler a meaningful, Contextsupported introduction to print which allows all children successful opportunities to view and appreciate print and to behave like a reader (Mason & McCormick, 1981). In the McCormick and Mason (1989a) study, a Head Start program in a small midwestern city was supplemented with a Little Book Program. Half of the groups read and discussed six Little Books in school. These books were then mailed to the children at home, and another set of six little books were mailed during the kindergarten year. The remaining groups received a similar amount of small group discussion time and an equivalent amount of mail. Results at the end of the Head Start year mowed that the children receiving Little Books readily learned to recite the text and that these children often "read" the Little Books at home, frequently involving their families in their use. These children also showed Fester intgest at home in telling and hearing stories, trying to print, and trying to read than did children who did not receive the books. Follow-up on the children's progress in kindergarten,showed that the children receiving the Little Books were better at approximating the text with a written-language-like story for both familiar and new Little Books and that they were able to identify significantly more letter sounds than the control group. A parental assessment of their chEd's interest in literacy activities at the end of kindergarten was also significantly higher for the group of children receiving the Little Books. Although the positive impact of the Little Books was fairly dramatic, especially the finding that these materials used in shared reading at school appeared to generalize the acquisition of letter-sound knr ledge, a serious limitation was that the number of children in the kindergarten follow-up was quite small. Thus a large-scale demonstration was needed to substantiate these findings. Although much research has appeared regarding the Winds of early reading skills many children bring to kiailergarten and first grade (e.g., Mason, 1989; Teale & Sulzby, 1986), little systmatic research has examined ways to break cycles of school failure. Encouraging suggestions, however, appear in an edited book by Allen and Mason (1989). Common themes include familiarizing by preschool teachers with the tenets of emergent literacy and a wide array of reading and writing activities, and encouraging them to become aware of the mappings of spoken language to written language. Building on those themes, then, our question is 'whether a Little Books Program, which allows children to discuss story themes and recite the printed texts, provides a unique opportunity for emergent literacy progress. 00 Shared Book Reading 191 METHOD Research Setting: The Early Start Program The Early Start program is a developmental program aimed at individualizing instruction and socialization for 4-year-old children deemed at risk for school failure in the state of Illinois. The program uses several screening measures for entry. One measure, the Chicago Early Assessment (Early Assessment and Remcdiation Laboratory, 1984) is a test of visual and auditory discimination, fine and gross moto, ..yelopment, and overall language abilities. This formal screening measure is used in coinbination with family and social factors acquired from home visits and interviews. Should a child score below a prespecified sone on any of the subtests, or come from a family setting in which it is felt directed school activities would be beneficial to the child, the child can be enrolled in the program free of charge. The research was carried out in two Early Start schools that were located in a midsized urban setting. A teacher, full-time aide, and half-time helper wcalced together in each classroom. The half-day program of instruction included whole class time, free tirne, small group time, snack and recess. The teachers were committed to enhancing overall language and concept development during whole class time when they read trade books to the children, did calendar work, shared currcnt events, and engaged the children in music and body movement. During free choice time, which the teachers called "Discovery Time," children chose from centers around the room, principally, blocks, writing, fine motor (which included puzzles and game manipulatives), science, dramatic play, library corner, art area, a sand table (which was often converted to other textures such as water, corn, and colored rir.e), and quiet or private space The children participated in a number of thrse areas during each day, and informally interacted with each other and adults while doing so. During small group work the children, were grouped according to similar needs or strengths and participated in teacher-directed actiuities. Throughout the day the children received individualized attention in whatever activity they were participating. In addition, the program was set up to involve parents in their children's education. The school held conferences three times a year where progress evaluations were discussed with parents. The teachers conducted home visits and had "Parent/Child Days" in school when only children accompanied by a parent could come to school They also provided parent workshops on parenting and school issues. Participants There were seven teachers and aides and three teacher helpers involved in the study. Each teacher taught two classes of children, except for the head teacher who taught one class in the afternoon and whose aide served as the teacher in that classroom during the morning. Each class had no more than 21 students and in all, 240 children from 12 classes participated in the study. Complete data were available for 232 chil- dren, and all analyses are based on that smaller number. All of the children were identified as at risk for school failure. The majority were from low socioeconomic status families, and an approximately equal number represented white and black cul- 192 Literacy Theory and Research- tirral groups. Boys slightly outnumbered girls. There were 52 girls and 63 boys in the neatrnent group and #*7 girls and 60 boys in the control group. Fewer than 10 children spoke a language Ither than English at home. Materials The Little Books (McCormick & Mason, 1989b) are books designed for promotirg beginning literacy development. The books consist of 6-9 pages with one simple line drawing per page and words or phrases that closely match each illustration. Tbe books are storks, as defined by Prince (1973), in which an event culminatWoe changes or the initial theme finishes with an enjoyable twist. For example, the "stoty Snowman depicts the building of a snowman. Each page adds a feature to the ilhistralion until he is completed. The pages read, "One big snowball. Two big snowballs. Eyes and nose Great big smile. Hello Frosty!" The books are writt4n about familiar topics for young children and feature high frequency content wirds. These characteristics combine to make the books simple, predictable means for engaging young children in discussions that emphasize meaning and print awareness and acquisition of new knowledge about written language features. It is important to emphasize that the Little Books were developed to complement, rather than to replace, language and literacy activities or trade book reading. The Little Books highlight print and meaning at a level where young children can begin to make connections between the spoken and written word by developing independence in print awareness and the act of reading. Procedures In May of the preceding school year the Little Books were introduced to school personnel and proctlures for their use discussed. In September another workshop for teachers was tido and follow-up visits were made with each teacher -..then she began using the Little Boeks to insure fidelity to the treatment. Background data on families were collected through a September home visit by the teacher. The 12 classes were gmuped into morning and afternoon classes and then three from each group were randomly assigned to either the treatment or control conditions. Teacher influences wem controlled by having each teacher serve as her own control. Each teacher taught one class including the Little Books as a small group activity, and one class without using them. For example, one teacher used the same following small group activities for both classes, sink/float chart and materials, making necklaces that demonstrated the concept of three, using tools, and puzzles. She added Little Book reading for her Little Book class by shortening the small group activities. The intervention began in mid-September, and continued throughout the year for all weeks longer than 3 days, with a klok per week shared, resulting Li 28 books being read and discussed On Mondays, the teacher introduced the book with her enlarged copy to small groups of students. She showed the cover, requested predictions or discussed the illustration and title, and then read it aloud to the group. The children were encouraged to join in the reading when they felt comfortable. After one or two readings to the group, the teacher encouraged the children to read it with her, first as a group, and then individually. Sometimes children took turns reading each page; other times, they were encouraged to read the whole book "by themselves" o n t./ in subgroups (e.g., the boys read to the girls or vice,versa). Mistakes were gradually corrected through nereadings and by teacherdirecting children to the words, such as pointing while reading. Harsh, immediate corrections were avoided, especially when the meaning used by a child was the same as that conveyed by the text; On Wednesday the books were reintroduced and reread with small groups. thiring these sessiors discussion of the book ;opic and individual reading attempts wae made by the children. On Fridays, books were read as a groupAuld the teacheradesignNIbobk-related tollowup Activities. Some of these activitieS Wire print-related,- sirchis writing a class story similar to that of the Little Book, whereas others were teatt-,:but net pririt-related, such as making a class snowman mural tc haag in the hillwaY.Wheo the hook was about building a snowman. At the end of the week each child ifi the Little Book classes received an individual copy of the book to take home and share with family members. This extensive practice with each book was a critical feature of the program. All children were individually assessed on two measures: The Test of Early Language Development fIELD) (Hresko, Reid, & Hammill, 1981) and an emergent literacy criterion measure, drawing on the Beginning Educational Assessme3t (BEA) (Mason & Stewart, 1990), which assessed print concept development, lent: knowledge development, and reading and writing developmeat. The TELD was chosai to measure overall language development and consists of measuring form and content of language in both expressive and receptive modes. The emergent literacy mess= was devekwd to pinpoint changes in letter, word, and book concepts. Reading and writing development subtests were added to the emergent literacy measure for the spring testing. Tests were given in September or October, and readministeredin April. After the Little Books program was initiated monthly observations of all classes were held to account for literacy activities other than those surrounding the Little Books as well as to account for how the Little Books were being used. A second parent questionnaire was collected in the spring to acquire information on home literacy mcluding children's interest in reading and writing. The Little Books program ended in the middle of May at the end of the preschool program for that year. RESULTS The first question we asked was how 4.Iuldren progressed in literacy development over the course of the year. Table 1 presents the descriptive statistics over the year for the three dependent measures that were given at the beginning and end of the school year for the treitment and control groups combined. At the begianing of the year all groups were comparable, and over the year substantial growth occurred for overall language development (TELD), print concepts, and letter knowledge. Writing and reading abilities, measured in the spring, showed that the children were also emerging as readers and writers. The principal question was, however, whether the treatmen *. influenced emergent literacy development. End of year means for language ability, print concept knowledge, and letter knowledge indicated that children in the treatment classrooms had significantly higher posttest scores over control classrooms on letter naming, F(1, _ .. Stofsiivi 81.4agg Combhvg eo#Cepts, l',etapnt* ancl, àetr _ "Cl 44:criiii; ...,..,,,,,, ,t,;. i,4*.,is ..,, * ,,;i:ftc, kt.- , ,"'1:',,, ,, ,.;,4,:,.. --:::: , 2 f,.. :. 1, - - 10.79 18;17 f566 _ 14' 4::;:t4ri 7.84 34I 3.35 5.07 25.45 12.16 64 21.37 66 Writing posttest 2.54 .83 7 Reading Pothest 1.84 .87 4 10)1ii Aix lie iiiCiitst liatest t 5.59' 11.59 'P4=0 unit: '-..4. : Lt.-...t- 150i -44ePte 'ireiat- ,. ...- --: 4- - *,igsci}.'- 20; zo 08 - - 'Nom* = 2.32 , , 224) = 13.70, p<.01. 'nun were insignificant wroup differences in pnnt c*,ncOpinand no differenc,....i in language abilities. as .Measured,by the TELD To ,test for treatmmt effects i is -nlio ,important40',Coniidettl!s: group of factors beeatise literacy coneepts.cannet_t)c:ConipteteiiiSol:ated. fromh other. They interaCt and influenee each other in is yet 'Unitnown .Wayi Before-i4cialat7k, ing a nniltivariate ahnlysii of liariande, ,Virkables mere-tit uncovcrai remove,,dverlapping constructs. 'The correintionS presented in-Jable.Z:,.shOw. strongest correlations disted for pretest and pOStiest VersiOnSof'theisanietei*Aernsi*,,_ measure correlations -were within the:low to, niodeT.ate range;indfcatint6**044,311,4 structs being invasured were :fairty distinet- and: could be model. A multivariate analysis of variance44NOYA) tonapare4-*Ocients ',On.; a dependent variables: language abilitY;,-print: eOneept8; le4ii:k0a*le4kei 4,04:041: reading. After seveml iterations; the ,beSt-fitting.Modet.Aised,tile..folig*int independent variables:. children7s pretesflangnage'and liter.41:).40,10540?'iitert4t* 'ilterier(based on beginning-of:1par: hoMe- intaivieWS);,:pc9blen*diiingAekt::: (such is refusing to answer verbaliy),!geekier, and interVentiOn treatinent or no). This Model Was signiteant with WillceS',14tinbdiMulti*iiiiF(49,..., 961)= 38.08:p. 001'. 'Moreover,. the ;'independent AfariOleS,AS a sctc2nqibuted significant #ffeL74 feir Pachbf PIO144.** and 111P.TOY. OrOpti Multivanate 1f(8,,, 44)-testi*C0 significant beyond the c.c)9.11Vial f9k!i,401c1.10*10g6=4,1'6-8-7,- t. laniu* abihty=481 49; print ConCePtSi= 457.0, -.WritinO 852:36,, .and...,read,:-: . ardlit,e01-4 35, id; 68,01 iht 480,0Oa dOendalt fatiaN0i=dthitiStiOOs de:Ot Naii.#1**otO 00414i4e0 00.14104,4i. 03140, :Thai W*4,1014 PietOt ***it4tter 46*10.4K 044 *404110,,t0:catiOi*-.aastinisaki optai _knOilkdgOi tittle Books affectiJ beginning print 4**.e4 iii#uen* literacy 4,07/000e0ti. TN* it*il*Ottf.944.***41.104,-, ks.:#e PTirs.44:-44.4**04iiittoio* Sit6-017;3* tii...#4:0FONOti#E44liok**,-***4040, by _ltfp,":=4 ba0-a-4itctetiOtis-effrO'tafilio*co4depol_ itfuscd "0"ft- -..*Ijii00 to "ls,...i0i.,***0 ,for littok:!Oowleogei #44 writing, with gi440iii 00 0OtpriAfgly,- a Oirdintei0.14, in 10.0.4-in4 NOM_ at home influenced Ove0 Idnitu4e q*e.11,0.. 4,00 ,NY,h litOOOY,COno4W011_ *us 24ty at boufie did bptttriai ri,Caga4iagia0dioii4 letters, liandialg poott. and ift04.:-Of cOuisei it is -italigsgbld:ta: if itii*St,i300x044,..tibility, ,aboy:prOpioted-intefest, but a relationihip- ..., -CONCIMIOM, 'Pe* mi.144-00.n00:** th-14, an mftmal -!,410,0,106.k xe*** 4;0 Uttic B. enhantie cirtalsr aspects of earl}, ,litets0 1ef Oariate F Tests with E(1,224) for LangUage anOiteracy,Conaegt$ 0'!"7" 1400 ate, itioseest Letter Kno:i*ledge Piètest 1.27 103-' il; 8S4" 48; 14:0' ... ..13 '.42or J....thiiifra-g-p-Ability, Pretest . -. 33.57** 4.80* 428* 4.20 9.31" Cbild interest 3.05 8.397" Gender 1.70 2.11 8.52" 9.83" 8.73" .30 .33 1.07 2037" 29.66** SI Print Concepts Prenest- Testing Problem Constant 7:12": '. -.4'.,,:.- ;70' 1.72 4.0* :5; -_ 20.0C l'k.05.**p<.01. young ,children has a significant relationship with subsequent 4rig(in 1984; Mason, 1980; Walsh, Price & Gillingham, -1988).'The fietthafthe Littleppo intervention contributed to the criticaLeady reading-rAzip indicate that the Majority of theie preschool childreiL were at thejnitial kifct childieu*ere Context-bound and,using,thernessages Within thesontext pf the, bo to gain aeceSs..to the individual:letters, (Mason, '1980), Thns;f:the./406.*** helping children to build foundations for More con*eritionallitereey..abititle4i that letter knowledge can be enhaatiedAnaugh-sharin&the-,LittOoOks,.reSPeefalik', when-few, if any, direet attempts were. Anatle to teachletter,kkoAlige,,:StipPorti.,* hypothesis of levels of develoPmentand'the role.of tonteXt-SupPOrtedreadingAs one of the earliest forms of print awareness- (Mason,1980). The,Littlei0oci14.-,prograny. helps to develop beginning print awareness in a,rezaningful,' sUppOrtiVectinteitt. The . simplicity of the books makes it .possible for the childrenio connectinformilly* graphic symbols and letter nimei. The Little Books provide an important, .pethaps sential balance' iq turiseboO literacy prOgrain. They balance the importance, of,the,print and o'O., story, :fewirade... books offer this balance because their ,rich story lineg and lOng textstihengit fostering, language andlistening comPrehensconk,are not intended tO makePrint 4.:On:Cepts:aepee, sible. Sniolkin,,,:oolori, and yaden :(1908) have begun to, lo*at the, leaining:.* occurs fro* print-Saliert bopks,-AOLf0401024*114), Print i$ .6.4.14r*-4 frog! the text, Such as in dialogue hubbies, kiSInore. readiiy noticed by5, yOung children , OfitiP1P. Pc Oks can Olow *eful c941CF**,!?-qw,-PCPPr.T..., v40.1. A tiritg' ail4"40illar 4:00 line t° supply i'llie-110440010-#t pnly.a fey, werds on each,litie, and strOng,pietn-le support on each tiage -ziChildttn leatn- to -tead '..i46:,tiat'and; to oe *taw* the te*t on sithsetittent teadhigii that h#ei but tisentiai peried*hen resuitch-indii:ateS that eonte4t!soppoiteilleattigki-,,NChih.0# on mmtiyt... Iead ' *904 use of tin44.40404- b9girs'i can P1.014 Start,Ot, Other- kpoztjangua'ge4ta,ithe4 ptechcol frogratna REFERENCES Allen, J., & Mason, J. (Eds.). (:989). Risk makers, risk takers, ruk breakers: Reducing 4:* . Nem" learner's rortsmouth, blki:'17reneiiisan. . Anderson,. A,-& Stokes,-S. (1984). SOcia:_ind inaritutional influences on ths r!'NetliOrpeat . , literacy. In H. Goelman;.A Oberg, & F. Smith (Eds.), Awrikening t nteracy mouth, NH: Heinemann. -Early Assessment and Remediation Laboratory (1984). Chicago EARLY Assessment. thicage: Eiricatjri of die City of Chicage. Elin, L. C. (1984). koworthoaraphy alters spoken language-competencies in :hildren learning tóèd aka: Language awareness and lent ing to read OP 119-147)spell. In J. Do ...ging le R. Valtin New York: Springer-Verlag. -Heath, S. B.-0983). Ways with words. Language, life, and work in communities and classrooms. Ne*, York: Cambridge University Press. Hresko, W. P., Reid, D. K., & Hammill, D. D. (1981). The test of evrly language developtnent. Ausiin, TX: Pro-Ed. Mason,,J. (1980). When do children begin to read? Reading Research Quanerly, 15, 203-227. Mason, J. (Ed.). (1989). Reading and writing connections. boston: Allyn & Bacon. Mason, J., & McCormick, C. (1981). An investigation of prereading instruction from,a developmental perspective: Foundations for literacy (Tech. Rep. No. 224). Urbana: UniverSity of Illinois, Centei'f for the Studs of Reading. Mason, J., McCormick, C., & . -nagri, N. (1986). Lesson negotiation paw= a teacher and preschoel children. In D. Yaden & W. S. Templeton (Eds.), Mendinguistic awareness and beginitIng.literaiy: Conceptualizing what it means to read and write (pp. 159-17?). Portsurouth, Niff Heineinann. Mason, J. M., & Stewart, J. P. (1990). Beginning Educational Assessment. Iowa City, IA:- Artierican Testronics. McCormick, C., & Mason, J. (1986). Intervention procedures for increasing preschool children's interest in and knowledge about reading. In W. Teale & E. Sulzby (Eds.), Emergent literacy; Writing and reading (pp. 90-115). Norwood, NJ: Ablex. J. (1989a). Fostering reading for Head Start children with Little Books. InMcCormick, C., & Mz J. Allen & J. Mason (Eds.), Risk maker-, risk takers, risk breakers: ::educing the risks for yaatg literacy le.rners (pp. 154-177). Portsmouth, NH: Heinemann. McCormick, C., & Mason, J. M. (198%). Little Books, Glenview, IL: Goodyear Press. Prince, G. (1973). A grammar of stories: An huroduction. Paris: Mouton. Smolldn, L. B Conlon, A., & Yaden, D. B. (1988).,Print-salient illuitrations in children's picture books: The erirrgence of written language awareness. In J. E. Readence & R. S. Baldwin (Eds.), Dialogues in literacy research (pp.19-68). Chicago: National Reading Conference. Tcale, W. H. (1986). Home background and young,children's literacy dev4opment. In W. H. Teale & Sulzby (Eds.), Emergent literacy: Writing and reading,-(Pp. 173-206). Norwood, NJ:. Ablex. = _ yok.'<Vele't "r", s , - - . ..-- At g,FFEcr OF EARLYLITEICACY INTERYPITION:, ON: IONDEGARTEN' HIEVEMENT oe, MiffOilalql*friftx, Jai* ett-r ofIgnO4411*41.47010iden , -A subStantial ntimber of concepts ahait'langnage. ate '*.qhfjOi CO* .19157,, These Oticci* -II* *PI: t#00,101110:14,`-il -booii; %Oat on the:pago;is_print, that ptint carries meaning, etc:)-,4 t , . 10:011 01,Ortiint 19, 04_4:1404:110 (POTe o *te**.kip 901,; .of print in theltOme, howeVer, it not stifftelent fp!. to'fierpiire these:00,40i, 'diey Matt alto take patt in interactive family eiPeriendes-Withprint-4:04.0.0,1-4,.., ifiatterv(barkin, 4982). One of the most important Of theteAnteraetiOns is reading: KoWevir, there tends to be a low incidence of *l c,. l irk: *fillies, (Heath, 1983; Teale, 1986), In Contrast, thid4le-914sy.,040*, 0*(§F,:Oro,* -4miliat With letters and words, and able to ileatt *king at hooka and likening to ttories. Since rficiat school;bak prostanta: _are gexe toWard middle=ciass children, a mismatch-between the literaeacquitediat theliteracy requited to.participate in school is virtually assitred.fetilOw=4106*: chil- ow dine dren With -the dun of lessening the risk of schocil failure fot suCh:Odteti,We, extended ibe home intervention study of Mecormick andl Mason- (1.986) .0tR, kindergarten7year Little Books Program aimed at encouraging patent-el* 00 leacher-childbook reading. Thelittle:Bookb (McCormick & Maton,.1990) are designed to.capitilize:UPon what children knoW. 'Specific features of the bOokainClutle.theinliowittev*gley; are, thematic and contain familiar topics to increase theelaiWe. eXpectation *at text.sbould make sense, (b) they are' writton, using everyday'higb-trequency-content'Wordtlto ,, ;- facilitate links between spoken and written:language, (c) thetecia a stiOng:fit;betWeen -illustrations and:text to make clear inat both .teXtand:pictUre,fratne'dte-nteaning, (d), titey are Written using phrases and-simplsenienees'to promote oemptoiiension,at the ,tniniinal- discourse level, (e) the story:ends **a' culminating ided,:te.,crege a sense , of-intrigue or amusement and provide text closure, and (f) a tgiticiaci, PaiticiPatiOn' 'This research was supported by grants from the Andrew W. Mellon Fuuldation and the Office of the Vice-President, Memorial University of NeWhiunfilaud, clOfb,:f0:86),:underlies,the preientatiOn,and:piotitx, oiois:t4Ogic,s.:45,?; fOSte The denee: jeCtive, of the Present study'Was, to ciet0;#109***rtal?*.001_41d0. , ikt9ffitY:4009p*Ptv. 0900es ,ffsPA 'kirideriar:t0 children, and iv4e44t i40.0".0,44krOti4Pife.:41.5.000418.10 nient type,:cimminnity statirS, and SehooltYpe., METHOD Sample The study was conducted in Newfoundland, Canada. Access 0.:01111,04-0.,* proximately 40 schooli from three school boards was granted.:We- -groupedthe schos into rural village.(drawing students froM only one small- cOrMminitY),;:inratcolie0Or.,, (thawing students 'from a number of amaltcoMmuuitiek in 44,g94);:04:ffibin:-N. communities were:not large by inany standards. BOtAtietOriicalkAtt:ysoi, hoapital or a medieal clinic that provides einergency or shOrit7terni*Oli;:., least one bank, alibrary, a Shopping-center, ,and'ogicT WIth urk-ui Center. The nrral communities typically have one. or Awn:Sinall4rOCeryi.s, gen,-1 Stores and a post office. From each of these.groupings, fottiChttels:Were. randomly selected. From each group of four, schools 1,:tre asaigned'iantleinlY to.oue:, Of the treatment groups (home, home and school, rznool, and cantrolj. Of the 12 schools, 6 had one_ kindergarten class, and 6 (4 of the _urban SchOols;. and 2 of the rural collector schools) had tWo kindergarten classes lint the, tame !,,r4,10,fk When a teacher had two classes, both received the same treatment-1m schools, class size averaged 17 children; in the rural colleetor schttols, classsi averaged 18; and in the rural village schools, class size averaged 20. The.oVeraltelass size averaged 18' children. All children attended school for ono-half day. Tà each: school, all kindergarten ehildren participated in the project. In all, 18 classesAMO2 schools and 325 kindergarten students participated. Complete pretest and posttestdatii were obtained for 309 children. Instruments Three tests were used: The Metropolitan Reading Readiness Test (MEI) (Nurss & McGauvran, 1987), the Circus Listen to the Story Test (Circus) (Educational:Testing Service, 1976), and the Emergent Literacy Concepts Test (ELC). The ME:T and CirmiS, are widely used group tests, and the ELC is an individual test designed for the study; to acquire more detailed information on emergent literacy than proVided iv the grOUP: tests. The MET measures auditory memory, letter recognition, and language and listen- ing; the Circus assesses recall and interpretation of oral language;,and the 4c cletpr., mines whether children have acquired such basic concepts ttbotit print as identifying: the front of a book, giving meanings for words (for exarnple, bird), ay.i4 clissifyp*, (e.g "What are some foods?"). The group tests have two levels of difficulty: MET1 and Circus-A were used asTretests; MET-2 and Circus-B as posttests. The ELC has two forms, the difference being that ELC-2 includes two sets of items not on the TLC I: reading words from the Little Books in and out of context. Procedures. treatment. Twenty-four weeks of Little Books intervention, uz,ing one Little, pock per week, compIemented the provincially prescribed Language Devetopment Fit)grain. -There were fourf,treatnient In Treatment 1,, a control grônp received no Little'Books. In Treatthent.2 (use of Little Books at home only), a new book was given to each child at the start of each week -for the child to take home', and react:v/44400s. Patents' cooperation was solicited beforehand and a demonstiation,, isingroiideo'of a parent and child working with books, and an explanation of the Inaterinis was given. Also, a set of guidelines adapted from McCormick and Mason (1986) Was prepared and given to parents. The guidelines provided the following infoithritiOn: overview of the project; description of Little Books packet; general pointers; suggestions for reading the Little Books (make a cozy arrangement, talk about the Main idea, read book aloud, have child help you read, and encourage child to read often); suggestions for improving parent-child interaction; suggestions for use of the color, count, and opposites books; suggestions for use of ABC books; and recommendations for making books with children (tell a simple story, give stories a snappy ending, choose words and phrases carefully, and organize the pages of the book). In Treatment 3 (use of Litue Books both in school and at home), a different book was introduced by teachers each week. Prior to the first week, teachers attended a workshop on the project and were given a set of guidelines drawn from McCormick and Mason (1986) that included the following information: introduction to the early literacy intervention program; how to prepare for the lessons; procedures for using the Little Books (opening, modelling, tryouts, and closing); how to follow-up after the lesson; and general pointers. Teachers were asked to follow this routine: on Monday, introduce the Little Book for that week by reading it to the whole class like any other story during reading time, on Tuesday to Thursday, work with smaller groups of children one group at a time and assist each to read the book; oo Friday, ask each child to read the Little Book, and send the Little Book for that week home with each child. Approximately 10-15 minutes each day was devoted to the materials. Parents were instructed as in Treatment 2. In Treatment 4 (use of Little Books in school only), teachtrs proceeded as in Treatment 3, but did not send the books home. Pretest data collection. Pretest data was collected from mid-September to early October, 1988. MET-1 was administered to each whole class over 7 sittings, each lasting from 10 to 15 minutes. Circus-A was administered to groups of 5 or 6 children and took about 25 minutes. The ELC- I was administered to as many randomly selected students from each class as time permitted. Testing time for ELC-1 was approximately 30 minutes per student. Posttest data collection. MET-2, Circus-B, and ELC-2 tests were alministered in late May to ea.ly June, 1989, following a testing schedule similiar to the pretest schedu lc , 214 Aa, Literaey Theory and ReSerirei RESULTS Means, standard deviations, and sample sizes for MET and Circus pretests ,and:, posttests by treatment and site are presented in Table 1. We eate four conelniiciris from these data. First, sample retention was >89% for all cells. Seerd,,alltPeans were >1 SD below the U.S. norm. Canadian norms are unuvailablesin ,t4e* pleasures, but there is little reason to believe that they Avould,be substantiallY..4ifterent,froin those of the U.S. Thus, on average, these entering Newfoundland kindergaihiert:scepr at risk of school failure. Third, on average, urban kindergartners score0,1jigheF,011Al measures, and village and collector students scored about the saine,:Fonith, differ7 ences in posttest means are not relatal ady to treatment. A planned multivariate analysis of covariance on these data was not possible because there was a significant Treatment X MET-1 interaction in the two,way. . ANOVA with treatment and MET-1 as factors and MET-2 as the dependent variable. Instead, a multivariate linear regression was performed with MET-1, Circus-A, treatment, and site regressed on MET-2 and Circus-B. Results are presented in Table 2. The strongest predictor of MET-2 and Circus-B was MET-1. Moreover, both pretests were significant predictors of both posttests in the univariate and multivariate testi. Neither treatment nor site had significant effects on either posttest. However, there was a significant MET-1 x Treatment interaction in the univariate case with MET-2 as the dependent variable. The interaction occurred as follows: in the home and school and school only treatments, students scoring lowest on MET-1 performed worse on MET-2 than the control group, while students scoring highest on MET-1 performed better on MET-2 than the control group; for the home only treatment, students scoring lowest on MET-1 scored higher on ME1-2 than students in the control group, while students scoring highest on MET-1 performed the same as students in the control group. There was a significant MET-1 x Site interaction in the multivariate case. Urban students scored highest on all tests. The village students scored higher than the collector students on both pretests but lower on both posttests. The MET is not overall an emergent literacy test, but a measure of ability to achieve in school. Thus, it could be argued that use of the Little Books should not affect MET scores. However, performance on some subtests, such as Beginning Consonants and Letter Recognition on MET-1, and Beginning Consonants and Soundletter Correspondence on MET-2, ought to be improved by the Little Books since the books expose children to meaningful print. An ANOVA using treatment and the sum of the aforementioned NtEr-i subtest scores as factors, and the sum of the aforementioned MET-2 subtests s Ole dependent variable, showed no treatment by MET-1 subtesl interaction. A subsu Trent ANCOVA controlling for the MET-1 subtests showed significant treatment (p<.001) and MET-1 subtest effects (p<.(101) on the MET-2 subttsts Post hoc comparisons using Tukey's HSD test showed that ad- justed posttest means for all three treatments (home only = 14.1; home and school = 16.2; school only = 14.6) were significantly higher than the adjusted posttest mean for the control group (12.5), and the home and school adjusted posttest mean was significantly higher than the adjusted posttest means for the home only and the school only treatments. Table 3 presents the mean., standard deviations, and sample sizes for the ELC- 212 .-!. q.,1*-10-,'4WT,,-414rife* ea* i*06.44 *40*, an4 Sample , 4:04:tinelf; _IlreitiStfp- iiiiiit lit! :Site - --- poet:to'', .4147 140T-r 14:2; Vfllage Urban 1.8`.... 1,5i.t 20 43.5 - _ 14Y-, -. .;1267 4.3P.* Collector Village Urban 44 35.7 '4. Collector 12.2 43'9= ,.... ,42,.: fj:i:,---, 'I.:''', ' 35.6 13.9 14.7 14.7' 15. 16 13 45.8 15.0 42.6 31 31 47.8 13.9 14.9 4.3 30 42.0 15.3 30 t9 38.5 31 12.1 14.1 17 3.4 r, ' .38 °114'.9 13.1 - 22:5. 24: 40. 3,6. 14.2 17 17 43.1 13.9 28 36.3 14.5 22 14.8 46.4 26:t 5.7 , 22.98.3: 20 Village 42.2 14.5 3.8 22 Urban 12.9 22 38.5 18.2 35 13.6 27 34.2 14.9 22 41.5 11.3 Urban School . 5.7 36 31 Village , ". 18.2 36 14.5 Home/School ,. i. 7 13:0. 41- orna 37:2 - ''-145S: 44', Collector 1and-ELC-2 tests by treatment. Perforitances' orr 28 12.8 4.9 22 12.8 4.7 35 n 37.2 15.8 31 16 `, . 27. 216 8.0: 22 21.4: 1.8 31 miinately: equal: Aowever; On 'ELC-2 tlier:control,grodp:sconed. to*ei. vhick in turn scOred approkiniately the same. ANOVA:re:culla shOivid rietreatiiietit by shoyert ,significapt trcat,O..ent:, ELC-1 mte1ction An rAtICOYA controlling for .*.007)-E44::j34-d-1, P:f*ts,(0.60)?,:f0:-hgeOPilit00.44sil*'70.4y71}1.4P all Attatirienti grduis, tot showed that the agitopz.v posttçst means 6.111.10 r 77.1111reP75,- **ft:3,g- *0*s* AfglAt: '9A23 0396 004: ..4-0,032 '''ct$4, '','!: 64 =-A 00 '4.:760, -0:10 -4.14 -0.5-0'. -0613, qi#.11f4 Ch9.°D. -2.227 l(Scboel) .sit*,,_ 7_ -(VINI) (colkOr) -3.404 3.416 . ii.4.0-1Pi-Trc.atinents :(by,tici*) (by porde/Schonl) (04S.*901) ME:14 biSItaiic 0.113 0.093 qiiYil4ic 0.110 -0.162 (by C911-tatn) Chins bY Tie-fitment . --0.154 0.308 (by, }{6m0 (by HoMe/ School) (by School) - 0.129 'Circ.* by Site (by !linage) (by Collector) 4836 '0.219, 0053: ;-. 0.367 0.274 0.289 0.241 0.129 -0.221 - 0,190 -0.135' 0.7,44 0.130 5.266 0.674 . -0.053 -0.096 R-squared ,06i ., ,,..- 0.074 -0.472 -0.161 Constane': 3. 0.49' .9.217 -0.100 -0.073 0.258 7.247 _ .0.015 4.2*: 0.114 '''.0.40:: 0:229: ,_. -h 0.611 ap<.05 for MET-2. bp<.05 for Circus. cpC.05 for multivsriste test. Table 3 Means, Standard Deviations, and Sample Sizes for ELC-1 and ELC-2 ELC-1 Treatment 61.14°1 142n..) %Ss,- I,* Hthir/S0001 Selmol ELCI, SD 80.0 85.7 93.2 87.3 23.9 25.4 30.5 _274' Treatment 18 19 . 25 128,1 168.7 172.0 164:3 34.8 43:0 . 48.6 , ..only*-17 I :4; lid** Schäol =16219; ichOol'OnlyF 166.6) **significantly. Idglier. k:iiiiiillió,:OkistOir.P6s4c-st iIr-F110'*6 cOtrcif8010-037.51.,'.1440.0*.i: 5:.0mr: 0004. . , significant ' -,:l'of test;44,.,hyPOtheiis' that the Significant eifeet; vist...anjitifiet---04-4PoStteit -:,t044ur:O:ilie:Pi.0,04 40siiiiCd',6-°*,`Iii*A14,..*0 'i*L:10kli9041:*iiii.00.1:**1' to- the El:%...+2"..-A . , J.:,Little-1.. Books , removed... I., , ,,,repeated*ith . ....... . the-itenis-directlylelated 4esS-Significantlieatthent effeet (0.,:irli))*tt'sifotioit.. k , IF. . DISCUSSION ' ,bveran Perfortnance = On average, the Newfoundland kindergartners studied are at risk of Seficielfaiftire: Th:ej, perform like Some minority, gionpst about I SD heo* the 40.:4, 100 with the reSultS of the ScutiiiiM News Survey Solutharn NeWipaper Oritiiip;19 in *hich Ne*fotindland was rePoried to haVe.the highest rato-of baiic,and finktiOnal **ye-inept on-the Canitilin'Tekt illiteracy in Canada (approximately 44%), . cf. pitie Skills= (King, 1981) (administered nationally at:grad:0 4; ,8; ,andt,12),14 .whieh Newfoundland Children score consistently-below the Canadian: national ii6f* (P0artment of Education Newfoundlancland ,Labrador; -.1989): 'School childien in NeWfclundland "experience persistent disproportionate school laiiiire"' (Ogbtr. MatuterBilinchi, 1986,.p. '73) in a manner similar to-some etimic Minoritieiht,the United States and Canada. However, given,that Newfoundlanders do -tv..4t helot* to ,an ethnic minority in Canada, the explanations of their school failtire,foree tis'iOlook beyond the visible features of ethnic and cultural minorities to the Underlying beliefé, attitudes, and exPectations that Newfoundlanders hold about literacy. Only the briefest sununary of some of the social; palitical,:and econonió facters that may influence contemporary literacy levels in Newfoundland are poisible induis paper. Placing value on literacy is a recent phenomenon in-Newfoundland, wherecompulsory schooling did not take effect until 1942 and was not 'enforced ttittil :the. mid-1950s. Most early settlers were unlettered fisher-folk from .Englancluand, Irilh immigrants from peasant stock. The ruling British fishing, admirals were-interested' _ , primarily in their own wealth and not the security and independence of-the early settlers. Settlers were forced to exchange their yearly catch of fish for food' End supplies, creating a subsistence form of living. Despite attempts by early-missionaries to cstabhsh schools, it was not until the,middle half of this century that the perceived need for literacy became widespread.-Before that time, most employment,Was fishing, there was no established context for literacy because the mrly settlerS did not need Or perceive a Aced for it, there were too few cchools, and access to,education wasgmited because of a small population in a large number of isolated.comnuinid# scattered along 5,000 miles of coastline. As a conseqUente, most:of the parents-of the children in this study belong to only the first or second generatiorynf'Ne*foundlanderS, t6 experience compulsory schooling. 4ven so, many didnot complete school, and defer to the schools in the job of literacy4evelopment. It is in such a context that the low literacy achievement in Newfoundland=must be understood. . 1.-t Literacy Theory and Research- = Treatment Effects The lack of treatment effects on either the MET-2 or ..:ircus-B were at: disappointing. Face validity judgments indicated that the tasks required by:theteits were of the type that beginning readers should be able to do, *It* that the Litt* Books should lead to higher performance on them. It seems,'howeYer,ihikthp..gfrc of the Little Books is more specific. The ELC-2, which was designed tir Mr*** directly the knowledge the Little Books were intended to foster, v as Oita ri;S**,:e to the treatment. Even when the items specifically related to the !..ittleBooltrviteLremoved, a significant though somewhatdimim th&I effeet was found, iriskatitig that: the result generalizes beyond the Little Books themselves. This led us, tohypothesize, that there would be an effect of treatment on those subtests of the MET:thi4.releted:to emergent literacy. The results confirm this hypothesis. However, the fact-000ns that the Little Books arc intended in the final analysis to improve children's reltdingi taken as the construction Ili' leaning, and reading in this sense was not measured by. any of the instruments. Thus, comparisons between treatment levels at the end.tif subsequent grades, when measures of meaning construction can be used appropriately, are quite important. Interaction Effects The MET-I x Treatment interaction, showing that the lowest achieving students profited most from the L.;ttle Books when they were used at home only and the highest achieving students when used in school only, helps support the hypothesis that the home has a crucial role to play in literacy development. It can be assumed that the students achieving the highest upon entering school alreadI have had a rich home literacy experience This experience typically promotes knowledge of a metalanguage about literacy (words such as "title," "story," "word," "printed") that forms the basis of thc language of instruction that teachers presuppose children know (Tem- pleton, 1986). So, those children from homes that encourage such a metalanguage are at an immediate advantage, because they can undt._:and teachers' talk. Children who come from backgrounds that do not promote this metalanguage are highly likely to be at a disadvantage, given that the schools will assume they have it. However, those low achieving children using the Little Books in their homes, where knowledge of a metalanguage of instruction would likely not be assumed, probably had for the first time an opportunity to learn this metalanguage and consequently were able to profit more from school instruction. Without this home intervention, the schools might have continued to presuppose a knowledge of language the children did not possess. This metalanguage hypothesis also would explain why higher achieving students were able to profit more than lower achieving students from the Little Books used in school only. Although differences between urban and rural students come as no surprise, it is a challenge to understand the MET-I x Site interaction, showing that students from collector schools scored higher than students from village schools on hfiXT-2, having scored lower on MET-1. Some researth suggests that a school presence in a community is instrumental in supporting and transmitting the significance of education (Spindler, 1987) Some children in the collector schools are from communities that have tdrN; ,J01.11iis fact tnay be reflected hi theinaretest acbre:s being lewer: .thin children schOols; On the other hand, nnai 51-acheol, children 'in:*,illage -schods, hot e!xickFd, to .80 'PliOly tele*** OuckOs '-fe*O-0,0041411100 as ChilOrn 9011,7_06r-40004- M.4.1*. Or' diversity of tko- 4utiNtsck991', e*Fic.40s1 of ti* dellecteor gentip' bait:beneficial inhknee'itivn end-of-jeitiTtrfonnince: In the .final apalysil:,,- the yo411: of die Little BPolcs mu$Irt PirloWqrT15ffer" -On i'esdi;nt.'Ci.ti:t4ciP-1*,:diffclet)43,11 teadink 6ffccti..it-Oce e40:prPioef-:1-rzAncl,3. :fsivering, the:44e- Becks, treitnient? We kike fo be Ole_ to report ph COde I °effeets' 'V° " REFERENCES Clay, M. (1973). What did I write? Auckland, New Zealand: Ectr,Ational Books. Department of Education Newfoundland and Lahradce. (1989). Educatian statistics. St. John's, Newfoundland: Governmerx of Newfoundland and Labrador, Division of Eviltition and Research. Durkin. D. (1982). A study of poor black children who are successful r-eaderi (Reader Edneatioa Report No. 33), Champaign, R.: University of Illinois, Centex for the SturlY 'Of Readlai. Educational Testing Setvice. (1976). Circus listen w the Stoty. MontereY; CA: W.IMcGraw Ferreiro, E., & Teberosky, A. 0982). literacy before schooling. band= HeineMann. Heath, S. B. (1983). Wais with words: Language. life and work in communities and aostroonts. Neo Yo& Cambridge University Press. King, E. M. (Ed.). (1981). Canadian Test of Basic Skills. Don Mills, Ontario: Thomas Nelson. McCormick, C., &Mason, J. (1986). Intervention procedures for Increasing preschool children's interest in and knowledge about reading. In W. Teale & E. Sulzby (Eds.). Emergent literacy: Writing and reading (pp. 90-115). Norwood, NJ: Ablm. McCormick, C., & Mason, J. (1990). Link books. Glenview, IL Scou, Foresman. Nan, J. R., & McGauvran, M. E. (1987). Metropolitan Readiness Test. San Antonio: The Psychological Corporation, Harcourt Brace Jovanovieh. Ogou, J. U., & Matute-Biarclu, M. E. (1986). Understanding socks-cultural factors. Knowledge, identity, and school adjustment. Los Angeles: California State University. Evaluation, Disaemination and Assessment Center. Rogoff, B. (1986). Adult assistance of :tildren's learning. In T. E. Raphael (Fel), The coraws of schoolbased literacy (pp. 27-40). New York: Longman. Southern Newspaper Group. (1987). Literacy in CanadaA research report. Ottawa. Author. Spindler, G. D. (1987). Education and culnuo' process. Prospect Heights, 11.: Waveland Press. Teale. W. (1986). Home background and yt g chIldren's literacy rt.velopment. In W. Teale, & Sulzby, E. (Eds.), Emergetu literacy; Writing and reading (pp. 173-206). Norwood, NJ: Ablex. Templeton. S. (1986). Literacy readiness, and bands. Reading Teacher. 39. 403-40. , PARtNt$ FikiittrtIONS:clE0ifitiiitt$!iitAiitg6,-,i,S0 wg1006;pwt'OMPimpli-.**1100.1 !qcrOggpmqvctikcX4*** 04 . - . ri' Beverly J. Bruneau, Timothy V. Rasitishi; Kent State University Many educators would agree that whole language is an idea oéàjób Dissaiiifled with the syStematie, analytic, and abatfaCt apoilja0,4t!i?- 'found in a traditional basal apProach, nu4iy teachers have nP5e4;f00:001.0,14,0§.,.. teacher-Student negotiated Whole Iangnago orientation:Zoodmin aria Giiiittiiiii(i981): have described A whole-language approach asone.that focuses on the consiuctson of meaning, on the processes of prediction, seIectioo, coofumpoN.ao4 self-cFrpitym, and' on tho functional uses of reading and writing: The ible Olt* teaeher language programs consists of creating stiniulating learning enVirOnMentS,.PrOiding , ñgei . r,. relevant Materials, and being a source of encouragernent and-support for and writert. The role of parents, too, is increasingly recognized as crucialio successfutschool experiences. In her review of research on parental involVement in schopTaand team, mg, Henderson (1988) reports that the role parents play in facilitating stiideat ilriPrOVe ment is undeniably important From the hdine environment to school ,n1VOlyeinent, when parents are actively involved in positive ways, imprciVethenfin Student aehievement follows. Moreover, Williams .and Stallworth (1983-84)=found- that Parenta do seek active and substantive roles in their children's education. In reading, the substantive role that parents can play is being increasingly reeognized (Anderson, Hiebert, Scott, & Wilkinson, 1085). Rasiniki and Fredericki,(1088) reviewed the literature on parental roles in literacy instruction and have suggeSted a set of principles to guide that involvement which include: making literacy an.tivities a regular part of family life involving real reading and writing activities, creating and supporting an informal home literacy environment, and capitalizing on and fostering children's internal interest in reading. Although they acknowledge that parental involvement is critical, whole language advocates have not explored adequately the full range of issues that are related to parents. Parental perceptions and support of curriculum changes toward a whole language orientation is an area of particular concern. Parents are not always aware of innovaflons in curriculum and may resist efforts to change an existing curriculum they have viewed as wholly satisfactory. The purpose of the present study was to determine parents' perceptions of a kindergarten program in the face of change toward a holistic language/literacy,eunieuhim. Do the beliefs patents have about literacy learning influence their acceptance of 4,14 - - . . 'Onuidulat change? itseinia-_itia;triticaF to doeitinent-.. beY*-4,016*11:4:c4,1344 9r r019491t9E-4491 9nt.';- '90999.*OPic*P4ts *EN ..:****koct4nt 0.00T5,4*! 04.P Intoie toWard-graat#'191fela, of rchicáons itanthia at9di-(a).What belie* do, .100,099t-- of *Ong *4°:it1119.40A- lang liaga açroath matches their OWIF_OPOOP11101* writing ,PrOgrant,'Whit$1444-40:Parentrt-ftiV,919F 1..tatt#14?'() 7*-,hat:cOtre.rn0 0;00k* 4.1** ab 40-49iiineoir . METHOD , - . Partzapatus _ 71:4,0,i141##PO4.in t.1*Pte.frOnt.4ea0Ptive 0,t11.0.-YOF;..M*04,1#15.-kIK enroll 00400:1*4 4t a university-based child develop ntce*rflice a.erves'aPPrOxittlateik-;17-41lldrett,f1Pra it.daaCY-ti#94git, kintkitatte9:4,.. Ptitntally;E-.901'1114t*JO .40P0,44d.k' base-4,-60: the 1510Cirs.0, of "d0*ol*h*IY:#P10**.-P,FfOj-Cr,tf.r:Y.9t?t #1,4*.;"i' as 9:!'44E:k;.;-,,, by the .114#q44-44.004.101. of Ed*49#.0fx944g, C444#0,M.0.40,100;;;C 1987): .ClasSes, axe Organiaed to alloW: children. maitiniMn: choice of activities, nd ,instructi9P-i$'04142.0' 40, In litrItitm Aar& the 1447g4iten-P241gtigled a '"letter Outh.0.1-e1(741604i0 progam, tha,*thie.01:9.. -0.1;11044* r A**14t0r" 1988, the nm*hltad 4,169nergartan tOach'er -4,496 400E9 Meant to a Wh91e4angnagc.APPPOaChi 4-ehaage:in c'1*,r.iet:dt99 04,140. cepted by-the center direetor. Parents Were infer,medOr,t,415 91-712(11*,C9ang919t9..91.4k a fall newsletter, and-the program .c.vas forMally.f4plairied td .0**,4:at annual Parent Orientation Meeting-held in:October. 'Id 8440,, many with the kindergarten teacher during aniferenees and the:teicherrepolteg t5104,4 about literacy learning during these individual meetings frequentiyarid thrdu year. Data Colkction and Analysis In May a letter was sent to parents of the 25 children aUgthciñdeare asking for parental perceptiona of their child'S reading atad,writing development dnng the kindergarten year. Nine parents, all Mothers,:vdluntaered tdheinterview May or June. Foiir mothers had' older Children vhd` learned to "traditional 'approach:" Three of the radthers*prOleacheraant 5.!Other*_ pleted or -achieved °roe college etincall9n. Tho int4r*Ws three PriaciPid researchers who. Were' afhliatatk- wjtk.the. nniveraitY:-. centcvOiien-cqed: inteniew,91.PstiP,Pts.,-v!qe 40, 'IL eata' 'own 'worda.'0Pradi0+- 19,7PY!-, crah.19.-4 interviewS, were Midintapedancl*OscriWfdr,analSisii: Ockt .t4eInt.otyi6vt,P0060.)1110,-A fRarenrintemew"Tailr, about yOur , , , tasding andwriting clef 1. iloiv do you think children je4rn tO ***MOS of litenicy:reiatedACtivities do mai) at borne with your cbiId?- ; ,. years,. fee& 4 -Has the way your tisilci is iliOgiiKesiiing at'tomc? '-kc***14,ko,Wdeserib' e the`aPPeatteh take te the *citing of child's *hit 'kinds-Of conoerns or questions do you have regardingyOur , Each interView was analyzed through categorical analysts, 0 --:.,4:-100)-- h.. which deMains were constructed and organiied into faxiMoinies. AO, , Y14.414,ce ' was constructed tor each parent based on the categOrical arialYsis. F..*Ii.Of the,nme eases wai then searched for themes relevant to the research'ilitestions. . ., --, RESULTS Beliefs About How Children Learn to Read and Write .-, Each of the parents began by describing early home factors they believed were important for literacy development. All 9 mothers stated it was important-to fiave books available in the home_ and Alt believed it, was important_ to , read ,regularly to their young children. For example, Mrs. H said that both her children learned,to read through "exposure and-being read to" (Transcript 5). All parents reportedihat they read nightly with their children at home. Five parents mentioned that not only did they read to their children, but that they also believed it was important to encourage their childreeto try reading during these regular reading sessions. Mrs. N described a pattern of "sharing" the readirigiof a book with her child, "I would read it first, and then she would read, Theushe'd,teaii it over and over again" (Transcript 7). Mrs. D described a similar pattern? "Usually, _ ._ .. i- :,...,_ -.A he'll want to read a book and then he'll want me to read a book" (Trait** 4). -=2. ..:' . Four parents stated they believed an imPortant component of childrerea learning to read invo:yed parents modeling reading behavior. Mrs. A summarized; ",My hitsband and I model reading a lot so that she would grow to think reading is the natural thing to do" (Transcript 1). Mrs. D stated, "Both my'husband and myself read a lot" (Transcript 4). The role of phonics in early reading was mentioned by 5 of the parents; although their opinions varied. IVI.s. S. believed most strongly in the impurranceof a'iyatematic phonics program" as a way ef preventing ,tailure in reading: She,stated-that her daughter knew the letters and sounds because she had taught her priortO'kindergarten. Mrs. B believed that children learned,differently, "Some learn, by.siiht:and, Some by a systematic phonics approach" (Transcript-2): Mrs. ivi thought,phanics had no plaCe .., A 212 LiteraeyiheorY'ai41414Ch in initial reading experience, "I'm not big on phonics, even if the schciol:hadL, it, I Would not use it at home" (Transcript 6). Two parents discussed the importancoof children developing knowledimpfliter= acy through readingenvironment4Print as a beginning Ore:ailing.- For,eXaMple,,, A reported"We'd oe driving ancl she'd, see, grocery store. She's read 'Coke' on cans and whcnzslieAVaSl,*4'Ahe,,yi.o.gAiiit _ signs oii4V"(TranStript4). Several parents disenised,the importance offiuictional,we ofWri3O Five parents. Mentioned their children were involved-MA*04g; note family members. 'Often the children printed these note*theinielies;*1 providing the correct-spelling. Mrs. C said she often- WrOte,he'r dau placed them in her lunc: box. She believed that attempting to reachheseznOtes,,, important in motivating her child to read. _ Changing attitudes. Mrs. K indicited that her belief,in chi1drenleamed read and write had changed as a mult of her son's kindergartekexperie*. commented on how she was now approaching her youngest chaaiteMptat,,,_ "I am less compulsive about how children spell and I am leaS1att to cO*0 read" (Transcript 9). Mrs. A felt more confident that what sheintuitivelydid at hoMe . was helpful to her daughter's reading development. "I thought I was deingit (yalnin pretend reading) to make myself feel better. I knew there were a couple Of were really reading and so I thought I should work with her in a more systeinatie manner. I thought maybe I need to sit down with her and finger each word'on a pieee of paper. It (the year's experienc4) reassured me that what I was doing was okay'',! (Transcript 1). Did the Whole Language Program Match Parent Expectations? Of the 9 parents, 7 believed there was a close match between their expectations of a kindergarten literacy program and the kind of experience their child encountered daily in the kindergarten program. However, this was not entirely true for 2 of the parents. Mismatches with expectations. Mrs. K was initially concerned that her son would not be properly prepared for first gade. During the course of the school year, she: changed her thoughts, aligning her beliefs with the kindergarten program. In deseriti4 ing her change in feelings, Mrs.K emphasized the importance she at first attributed to phonies instruction. Her three older children had had a phonics-based kindergarten program, and she was concerned about the work her youngest son v as dein: in kindergarten. She was worried bechuse she did not receive "feedback about her son's, performance" in a familiar manner (i.e., workbook pages). The lack of a systematic phonics program did not match with her expectations of a kindergarten prograin Which "prepared" children to be successful in first grade. Even at the year's end, some doubt remained, "I am uncertain. The kindergarten program in our district used latter People' and emphasized sounding out and my son is not good at sounding out" (Transcript 9). However, by the end of the year Mrs. K appeared to balance her worry about her 2,2J. n's lack of ability to "sound out" words with his excitement about books and roAciing; which was greater than that of her other children: Specifically, she mentioned toad 't:rittitude toward_ reading, and feelings_ about_ himSelf, "He Wain't ;0" [Transcript 9]. Mrs. K spoke very positively of the 44,frie enii,iroatnent. classroomlrad "lots of praise, no criticism." Sholil$4.0*:0,ty-the alfe c4ilciren SupPorted 'each other. Mrs. K appeged to,belie)e.tha,t-:eyen , son,. might not have the same phonic knowledge at, Other,ahriclie ,disa4vantaged. "I'm concerned', that sOme -of the neighhOthoOd` _1 and are:familiar with some sight word's thatile . okay. if he gets the right teacher and the right etiviroirMent"%, Mrs. S, I remedial reading teacher, on the other hand, ':_re that the new program was poor because it did not teak: wntC be' 46, hei, belief cs in -a_;,tistamatic- fashion. She attributed The failure of her junior highstudenta to a:lack of feitMclation in phonics. Turthermore, Mrs. S was concerned about the problem of poor self-esteem Which she observed concurrently with lack 3f reading Progress. "ArrythingyOu ,erm do to prevent a child from feeling negative about reading, you have to do" (T0000 8). According to Mrs. S, a systematic phonies program in which childreiriprogress through clearly defined skills is the best way to prevent failure In readlOg. MrS, S's two older children had learned to read through a "letter of the Week" pin ...grani and from her perspective not only learned letters well, but enjoyed it. "The children eat food associated with the letter and do a lot of positive things like singing and daneing" (Transcript 8). In contrast, Mrs. S believed that her daughter's whole-language experi- ences were not enjoyable, describing her child as being extreinely frustrated with mted spelling. She felt it was unfair to ask children to spell before they had learned their letters. Mrs. S. believed this experience endangemd her daughter's self-esteem. She considered Nithdrawing her child from the program, but decided to have her remain because movement to a new classroom might be distressing. To counteract the "poor instruction," Mrs. S implemented a systematic reading and writing program with her daughter for 30 to 45 minutes each evening. Matches with expectations. The remaining 7 parents believed the program matched well with their expectations of an appropriate approach to beginning reading. All felt that the match was due to the support and encouragement their children received. Mrs. D summarized the importance attributed to the affective climate, saying e, "The main thing is my son is with adults who care about him" (Transcript 4). The role of the teacher seemed to be especially important to the patents in describing why the program was well-matched with their expectations. Labeling objects around the room was, in particular, one teacher practice that many parents viewed as positive. Mrs. M confessed dux she, too, began to label objects throughout her home. The mothers described the teacher's enthusiasm for books which they saw transmitted to their children. They commented on being impressed with the amount and quality of children's literature as well as the interest their children were developing in authors and illustrators. All positively described the manner in which the teacher interacted with children and modeled acceptance of individual differences. Mrs. D noted, "There was a real valuing and encouragement of any effort" (Transcript 4). Other Concerns with the Whole-Language Curriculum Two of the parents had no concerns with the kind of literacy program offered:t their children during the year. Mrs. C, in fact, described the program as2,a.aktutalt extension of her own interactions with her daughter, stressirigthat°,4fifort*Icty..",oti:1_ instzuction, functional use u print, and involvement of the teacher,Were components she believed in. This parent stated she believed an iipportant;reaSonSof her daughter's good progress was the compatibility ketween hinne,and D also had no concerns, indicating that her son was "adaptable"' and.cOuldi,easitY4'i2, adjust to a variety of situations. Five parents were concerned about the process of invented.spelling.'144:44, pressed the greatest reservations, indicating that her daughter did nrit 1*e the-back,. ground, although interestingly, she was aware that her daughter;clitlineW;allOf her: letters and sounds because she had taught them to her prior to icindeigarteiMja:i75,, repeatedly stressed that when children are required to do something in. Whicit,theYi have not had adequate trainiiig, they will not be able to do the task,:and fea ; frustrated. Mrs. S's concern reflected her belief that invented spelling was not the -:7-41 way to develop knowledge of writing in young children. Although the other four parents also described feelings of frustration their children; experienced, they were more concerned with how to help their child adjust to invented: spelling. These children apparently had done much early writing at home, and:the parents had helped through supplying the correct spelling of words. Mrs. A noted her._ daughter wrote frequently prior to kindergarten. Much of this early writing included letter writing to relatives and friends and responding to children's magazines. "She would dictate, we would spell and she would write the words" (Transcript 1). However, Mrs. A observed that as invented spelling was encouraget in the school her daughter, who had been a "writing maniac," became frustrated. She reported that her daughter would say, "I don't like writing because the teacher won't help me. She won't tell me how to spell the words" (Transcript 1). Mrs. A described her own , frustration during this period, "I thought now we have really set her up. We were telling her how to spell words. Now she is stuck. She often knows what she is doing is incorrect. She knows it doesn't look right" (Transcript 1). Mrs. A noted that during this time she was able to talk over her problems with the teacher, and as the teacher worked with her child, her daughter became less frustrated with the process. Mrs. N described her daughter's frustration in not being able to reread her in- -. vented spellings. "It was very hard in the beginning of the year because she would bring things home an 1 say, 'Mommy you read it' and I would have absolutely no idea of what it was meant to say. She would be very upset, and we would have to sit down with her and say, you can read it. That was a terribly difficult few months" (Transcript 7). The second set of concerns reflected the parents' worries about first-grade placement Mrs. S believed that without a strong phonics program her daughter would not have the skills to be placed in a top reading group and would come to tee herself as a poor reader. Despite her changing expectations, Mrs. K continued to wort), that her son would be handicapped by his lack of phonics knowledge. On the other hand, three parents were concerned about the different classroom atmospuere their child g23 .;,Might,ericounter With a skills-haied,first graile.;Mrs:-MWorriedibetit.how*dati vi,..040,Tea0t. to a Struettired ehissiOom.,Mrs.;13; hived the neW Se . , . 'W MO:;).cbelief*'llet 00. *0114 40.41" Two parents also expressed cOPOrn, . i*t'Aii*:- a $111i:.!cici' OPOick,towaticl 010_ caned;her.dmighter "anrintahle"landthe Other poteq,ihifner IMPLICATIONS This exploratory study opened a; small, window whiel!.00**,49 ,underatand the.perceptioni3Of a particular group Of:Parenta'whOse childrex gaged' in a deNieloping, whole language :program: One, theme **prising,- is . ationgly- supported bY the . findings of this atinly,:, isthow senouslyi .gi9up of middle sgspaints viewed their role in helpingtheir. ogleank., Write. All reported they' believed' literacy development heginS; at litinie; nhz their welts they laVe illtratiated hoW Serietisly theY have 00-r4k04144.00040. through regularly interaating with their children In readinL'andWriting#Peile: It seems that one reason for high degree of acceptance aOlie.WholeAgintge -approach-is-that it validates and eitends literacy activity that had: beeir.oeCtikiki the home. The informal caring, low risk environnient, and, the foeuS on-the °affective dimensions of reading are some important characteristics that were .foundt)0:4i-g)e home and in the kindergarten curriculum. Ho's+, other parents (such as .Mr_s..S) : vô hold an analytic orientation toward literacy instruction and engage in those,a.etiintiea at home would respond to a whole language curriculum is a question Wortirptirs:tling: For this group of parents acceptance of a whole language aPPmachteWarde0y reakling activities was not difficult. Parents responded enthusiastically to` their chil . dren's growing interest in reading books..Developing an enthusiasm and a-lbve for stories was a goal shared by both parents and the teacher. Several parehtS, also rePOrted that during the course of the year they were able to share responsibility for feading with their children. The parents seemed more comfortable in helPing their.childrea in these early reading attempts. Mrs. H stated that if her daughter doesn't knoW a Weird, "I tell her what it is" (Transcript 5). Mrs. K reported that she had "learned" tribe more tolerant of her son's errors in reading. She noticed they did not interfere with his understanding of the story, and his enthusiasm for reading appeared greater than that of his brothers and sisters. Interestingly, the parents who expressed satisfaction with the whole language kindergarten seemed to feel that the emphasis on informal social interactions and development of positive feelings in children about themseIve. as readers led their children to reading. The one dissatisfied mother felt that the approach taken to reading inst-tiction actually resulted in diminished self-esteem. In both cases parents were concerned about their children's affect, although their perceived outcomes *ere considerably different. It appears that these parents are seriously concerned-about their children's affective disposition toward reading and how they view themselves as beginning readers. 216 Literacy The Ory arid i6eiii,gb Geaeral acceptance not withstanding, some parents Voicetttrong concermahoit'one component of the program, helping children With Writing spell ing. Parents were concerned that their childrea, did, not-haVe 'appropriate:skills O. successful at this task. For some, encouragiag i0efited spelling was an'apProaob,;: was contrary .to their previous pattern oliaPplying spelling How we can better help parents prepare their child for writingis ati iinpartatit,,atea:' further investigation. Parents also expressed concern about what would'hapj)en to their chinnfiist grade. This demonstrates how aware parents, are Of thel ,rectaf,.:inajar.cutricalit* changes on their child's cognitive development and on self-0904er*. It should not be surprising that many parents Will-haVe Major rsseriltiOns:abatit. whole-language approaches in their entirety. Such curricula are:foreign:to theiv,olvm, educational experiences and notions about proper approaehetio Magadan. need to be sensitive to the concerns of these parents. While realizing' that.theY,:may never convince some parents as to the merits of a whole-language classroOM Mieritar., tion, frequent communications, explanations, and opportinities to observe inthe class.;, mom will assuage many parental concerns. Moreover, teacher& eaii_PrOvideenough. 0 evidence from first-hand experience or published reports to demonitrate pareras, that whole-language approaches have been successful in promoting word rev:ignitionproficiency, mature spelling, and continued success in school. Finally, the findings underscore the importance of parents and teachers working, together. Several parents who were highly supportive noted that when they had questions they could go to the teacher who would listen to theh concerns and at those times provide them with information. The overall feeling that the teacher was working' for her child appeared to be a strong factor in the parent believing this kindergarten experience was worthwhile. REFERENCES Anderson, R C , Hiebert, E. H., Scott, J . & Wilkinson, 1. A. G. (1985). Becoming a nation of readers: The report of the Commission on Reading. Washington, DC. National Institute of Education, U.S. Department of Education. Bredekamp, S (Ed ) (1987). Developmentally appropriate p:actice in early childhood programs serving children from birth through age 8. Washington, DC. National Association for the Education of Young Children. Goodman, K S & Goodman, Y M (1981). Whole-language comprehension-centered reading program (Position Paper No 1) Tuscon: University of Arizona, Arizona Center for Research and Development. Henderson, A. T (1988). Parents are a school's best friends. Phi Delta Kappan, 70, 148-153. Rnsinski, T V , & Fredericks, A. D (1988). Sharing literacy. Guiding principles and practices for parent involvement. The Reading Teacher, 41, 508-512. Spradley, J. (1979). The ethn, graphic interview. New Yosk: Holt, Rinehart & Winston. Williams, D . & Stallworth, J (1983-84). Parent involvement in education project. Austin, TX. Southwest Development Laboratoiy. 225 - WORD LENGTH AND FIRST woRpltgcop_NWON. Luci Lipscomb'and PhIlip:R.:Gough, University of Texas at Auitin 40940' How do children learn to read words encountered for the first time?' acknowledged that a child will use a variety of techniques to help 4160fy and reinernber printed words. If useful, these techniques- Wilt be retained and'iii0=4,0friture word identification. The child has discovered a helpful strategy for aPprOtching iurfa;. Some strategies employed by children prove to be inadequate for diSCrinrifitning, between new words. For example, total reliance on context or only attendinktiithe initial letter may soon fail to provide the child with the precise word: If children talk on such methods, learning to read can come slowly and with many "inactiiiticieS '(Gates, 1949). It is necessary for the child to recognize which techniques are unproductive and replace them with successful ones. Samuels (1976) suggests that whzn learning to recognize a word, children associ- ate the written form of the word with the spoken form At til:s nage, this written word is an arbitrary collection of written symbols for the child, inia 'the Connection between the printed and verbal word forms can be arbkrary as well. Thia paired- associate learning can be successful or unsuccessful for the child depending upon how unique that association is. Gough and Juel (1989) and Gough and Hillinger (19801 have shown this association to be selective in that the child selects a specific cue used to associate the verbal pronunciation and the written stimulus, distinguishing it from other stimuli. The cue selected may be any characteristic of the word, such as the configuration or even the font, which may be used to discriminate the word from others. Gates and Boeker (1923) found that children did not look at the entire word for recognhion but picked a distinctive cue to remember the word. Gough (1985) conducted a study in which a thumbprint appeared on a card with a novel word. Although the children learned to identify this card faster than words on other cards, when shown the same word without the thumbprint, few could read it. In addition, when the thumbprint was shown on a bard with a different, previously knowa word, the word originally associated with the thumbprint was given. The origiaal word was given even when the thumbprint was shown on an otherwise blank catd with no word on it. Gough's study indicates that the cue selected by children may even be an extraneous cue not directly connected with the word at all and that attending to an extraneous cue (e.g., a thumbprint) may cause the children to ignore salient cues. A variety of other cues from which a child might pick are suggested by Samuels 217 , 26 (1976). He writes, "The cue upon which the learner focuses his attention may be,i, letter, letter group, word shape, in fact, any characteristic which helps to sat MIS word' apart from others" (p. 271). Another study of first word ilearning%br GOugh,(ins preparation) supports Samuels' view that the' cue choieninightlre cOmponert(of:,thc,:, word, such Pc the first or last letter. Children were .taught trE-readleur 4y6tdir,-.0a_ individual cards by standard paired-associate learning titoOedikes. ANT ,tnaatetiat these words, tize children were shown cards with the'same sYords,,exe*Ott half;ort the cards, the first half of the iVonl was hidden. On- the-rentainder Oldie csrds,tt last half of the word was hidden. The children Avere shoiiiilne5caftla and identify the words. The results showed that the children tended PI teeekelze when shown either the first half of the woni or the last half, but not bOila.-It.Seenis', they selected a cue from one part of the word, they ignored the Othet Part of the wotd. These studies showed that children select specific cues to recoanize a neW wOrt4, The present study was conducted to determine if children might choose a tifo-Petty Of the word, such as length, for identifying a word if the length of the word distinguishes, it from the other words learned. METHOD Subjects Forty-eight 4- and 5-year-olds served as subjects in the study. All of the children were enrolled as students of a private daycare center in an upper-middle class neighborhood in Austin. The group consisted of 26 girls and 22 boys, all Caucasian. The students came from the classrooms of several different teachers. The main criterion for inclusion in the study was the inability to react the lbw target words used. The daycare ce.lter did not provide any direct reading instruction to the children. Alphabet recognition was only taught as it was brought up naturally during the activities of the day. When pretested on alphabet knowledge, no significant differences were found between children in different groups in the study. Materials Eight words were selectee to be used. Four of the words were four-letter animal names The remaining four words were six-letter animal names. Each fot.1-letter name was carefully yoked to a six-letter word for similar:des m vowels, configuration, and familiarity (e.g., bear-beaver) to reduce the prolAbility that these features might be identified as distinguishing cues. The words were each printed on a separate 3 x 5 index card. The words used were &ear, yoked with the word beaver; pony yoked with monkey; duck yoked with turtle; and fish yoked with lizard. Groups Two groups of children were presented words which were all the same length except one Twelve children were taught three 4-letter animal names and one 6-letter animal name (the 4446 Condition) whereas twelve others received three 6-letter words 22/7 s. , - 'Firit Oord Ric-Oration Ainci. one 4,TletteWord (the 6664 Condition). The word of distinctive length was substi- luted; for thc,yOlccd word of uniform length each time. For example, bear and beaver were iubstitutedlor each other since they were,!helqke4.p14.these'sroups,the word Of' distifictiYeJength was alternatnly selected In from the 1... options solhat each of the optitnis.ap-peated art equat *fiber Therefom three of,the children within the 4440 C-9,04icta we***Intk4 t!';`*5-Cds k.ar,,, pony, duck, ahd lizard, with lizard hling.the,six4etter, wind, subitititied:for fiik, nice mote children in the same condition receiyed the WOrds and fish, ,where'turde Was the six-letter word,,aubstituted for 414 Mor#0 and beaVer were also rotatedis,the six-letterwards ivithin the 4446.cOntlition.'ThisWas,dOne to lessen the chance that one word might be unique iu ways other than 4gth andihetebY provide additiOnal cues for recognition-. In the 6664 Candition, the four-letter, anintal name was rotated in the same manner among_thefoutalternatlyesTwenty-feur other preschoolers served in the remaining two groups. In these goups, four words of uniform length were learned. Twelve childrenvere taught the four-letter animal names (the 4444 Condition) while the last group of twelyi-were shown the six-letter words (the 6666 Condition). The words contained minimal consonant and vowel rr,petitions, and words within the same groups did not begin or end with the same sound. These were the same words used in the 4446 Condition and the 6664 Condition. Identical procedures were followed to present these words as were used with the other groups. Procedures A pretest trial was conducted individually with each child. The words were presented on 3 x 5 index cards one at a time. The child was asked if he could read the word. Of the 50 students tested, only 2 could read any of the words and these students weir eliminated from the study. When a child responded that he did not know the word, the test administrator told him what the word said. Therefore, after every presentation the word was correctly pronounced while it was still in view. The words were presented in the same manner for subsequent trials. Pairedassociate learning procedures with anticipation and correction were used. The child was shown a card and asked to read the word. Whether the word was read correctly or not, the test administrator repeated the correct word after each trial. The responses were written on a record sheet. The words were shown in random order for each trial to avoid having the children simply learn the sequence of presentation. The mastery criterion was set at correct identification of all four words in a group for two consecutive trials. A maximum of 15 trz p!! was ptesented. Overall, 79% of the 48 children reacl.ed the criterion within 15 trials, although there were distinctive differences between the groups as will be discussed later. RESULTS The learning curves werc plotted to show the average number of correct responses per mai for the 12 children within each group. For the two groups where the majority 220. Literacy- Iheotiand iteseirel; of the words were four letters in length (the 4446 Condidowand the 4444 Condition),_, the learning curves showed that on the average students in the 4446 Conclition did: reach the criterion of correctly identifying all foUrof the:words fOrstWo COnsecutive, trials raster than-students in the 4444"condition. The differeficebetWeen the learning, rates of lese two groups can be attributed to the Word,Of clistinctiwietigth:(thkai letter word) in the 4446 Condition. The six-letter name waalearned*itii fewer triala. than the four-letler names required. The learning curves of the children in the groups where thc nisti(*tii; of were six lettera in length (the 6664 Condition and the 6666COnditiOn)sshoweit sirailatt_ results. The percentage of correct responses for the studenti were' sweraged plot for each of the fifteen trials. The Oildren in the 6664 tOnditiOrisori ayeragereae the criterion level faster than the childrenin -the- 6666 tonditien, Whe,06:**4-:. were uniform in length. Once again, the difference in the nuMber Of iriala reqafted: to reach the criterion level was due to the word distinctive in lengthi the foni-letter word. The six-letter word did not require as many trials to learn correctly as did,ihe four-letter words. The data were analyzed by a 2 x 2 x 15 analysis of variaote with the first factor being the majority word length, the second factor being the length of'. the-mins-Ay, word, and the last factor being the repeated measure of trials. No significant-difference was found for the main effects of majority or minority wotd length. The childrenin the 4446 Condition and 4444 Condition did not learn the words significantly faster or slower than the children in the 6664 Condition or the 6666 Condition. In addition, it was not statistically significant whether the minority word was four 1Nters or six letters in length. A statistically significant difference was found for the interaction between the majority and minority word length. ft did not matter whether the majority of the words contained four or six letters, or whether the minority word contained four or six letters; the difference in the rate of learning occurred when the minority word was different in length from the majority of the words in the. group. When one word was distinctive in length, as in the 4446 Condition and the 6664 Condition, the children learned that group of words significantly faster (p<.05) than the children who were given words of uniform length, as in the 4444 Condition and the 6666 Condition. The chart in Figure I compares the learning curves of all four of the groups. DISCUSSION The results support the hypothesis that the children used the property of length as a cue for reading a novel word. The children did learn the word distinctive in length significantly faster than the words of uniform 1 th in both the 4-446 Condition and the 6664 Condition. The chart in Figure 2 shows t the four-letter animal name was even learned faster when it was included in the 6664 Condition compared to the other four-letter names in either the 4446 Condition or the 4444 Condition. This was true even though the same four-letter word appeared in each group. In other words, the word fish was learned faster when it was in the 6664 Condition than when the same word was in the 4446 Condition or the 4444 Condition. ,2 9 3 1 4 Ad 1 - Figure I. Learning curves orail foni conditicifir,; six-ktter:aniniaLname:was learned Taster -When,it Ws.S -distinctive;length in tise 4446:coildiiiiiii-as.ceiNiared!to the oth6r:six4ette'r àithr t4e 6004'CO:0-ditiOroF,PP 6,60.6004160".. A Ot.On4-` hyPcIttr-St of int5equ **Pad 'thS Ff4Ft Of *01-121',100,* 01* word learning. WOuld'Avords of longer length OroVide MOre cues froM` and actually be easier for the children to lelint-The reSultS incliettkd: 11.4 70 (668)4 4- 444(e) 41- 4444 SO 10 1 1 3 4 II S 1 S 10 11 1111 Trials Figure 2. Learning,curves of onlylhe four-letter words from the 6664 Condition, On 4446 Condition, and the:4444 tvidgretips containing primarily six-letter words tended to be Iearned faster, the differ. once wa not SignifiCant. It is important to_ note ; ttiat:the -percentagei rea0in the cr,ite!kon level'was:Elifite'v,1144-Pi4Y-<-5,9* Pt -COndition reached the criteridn level while pco? Of the Students in the Met. OtgiOn, Th,6604'Co9ditiol- not PAIY',PiDA04,00'1,0v041IP distinetive lengllr, btit the six-letter we* seemed td -proAde:aiifficient eves fdef_ the children _to learn all, of the wordi. The:other. two grotipe- fell bet*pen.theac: t*OWith 75%_reaching criterion in the 4446 Condition, itid:9,3*:achionktho ,Criterion level in die,6666 COnditiOn. It sato postiblethat the longer ,Wards PrOidedagreake number of cues from which to choose, thus supporting the idea dr:elective aSioeiation in initial word learning. Implications As shown by this study, if length distinguishes a word from other words:being learned a beginning reader will make use of this property to aid in_word identifieation. If students have not learned to recognize the letter-sound correspondences,, they, inay use such cues as length to build up an initial lvtnk of sight words. At thiS,point,,any feature which helps a child differentiate one word from another might be benefteial: Knowing even a few words in this manner may allow children to read theirfirst text, which can then help them discover more discriminating cues within words. Recogniz ing a property of a word, such as length, also signifies that the child is at least attending to the word form. This in itself indicates initial steps toward reading. REFERENCES Gates. A I (1949). The improvement of reading. A program of diagnosuc and remedial method: (3ni ed.). New York: Macmillan. Gates. A 1 & Baker. E. (1923). A study initial stages in reading by pre-school children. Teacher's College Record. 24. 469-488. Gough. P B. (1985). Fust word learning by selective association. Unpublished manuscript. Gough. P W & Joel. C. (1989). Les yemitres etapets de la ret -maissance des mots (The first stages of word recognition) In L. Rieben & C. Perfetti L 'apprenti lecteurRecherches empiriques et implirstions pedagogiques. Neuchatel et Paris: Delachaux et Niestle. Gough. P B & Hillinger, M L. (1980). Learning to read: An unnattual act. Bulletin of The Orton Society. 30. 179-196. Samuels. S 1 (1976) Modes of.word recopition. In H. Singer & R. B. Ruddell (Eds.). Theoretical models and processes of reading (2nd ed.). Newark, DE. International Reading Assoctation. 2 31 _ - 44,,syLiApiONIT APPROACH:TblEAeHiNd.:inetiikkG'Oi -70,M)111/14,AND:g4TIVORAPP.PDXSYLLABIC --1MSABIRW-RFADERS-L-- John Shefeibine University of Texas . _ Skilled, readers identify words by breaking them down-into syl*10: .Studies.'recently reviewed by Adams (1990) indicate that for sleillecLz#detS (a) is automatic and instantaneous, (b) occurs during visual's:Canning*, aqh'elanite time that individual letters are being identified, and (c) is uied as well as real words (see Mewhort & Campbell, 1981).,PoardecOcreis, &thóàthi hand, have a difficult time reading polysyllabic words even when tho ;144 Pir4ilfif* syllable words (Just & Carpenter,. 1987; Samuels, LaBerge; & AreMer091.8)i_ ThiS inability to figure out "long" words can be particularly crippling in .the niiddje ixid upper grades where vocabulaty is not controlled and where large numher's of poIy-SY1labic words occur in reading materials across all subject areas. Helping studenti with limited syllabication ability is a pressing need. Unfortunately, conclusions regarding effectiveness of syllabication instrOtiOn have been mixed in the relatively few studies that have been conducted (CanneY & Schreiner, 1977; Cunningham, Cunningham, & Rystrom, 1981; Groff, 1971; Johason & Baumann, 1984). Researchers have attributed these digappoindng results tO theOriti-cal and practical problems associated with rules for dividing words into syllalilea,:. Common concerns include a lack of theoretical support for dictionary rides for Syllabication; the numlr, complexity, and low utility of some rules; and instsuctionsitptfoce- dures that "resuppose" pronunciationstudents are required to prom-nee Wads before dividing them. The present study developed and tested an approach to syllabication instruction that greatly de-emphasizes the importance of rules for dividing words and that, instead, stresses (a) syllable automaticitythe ability to identify individual syllables effortlessly and automatically or at sight, and (b) syllable pattern identificationthe kientification, in a flexible manner, of possible patterns of units in a polysyllabic woid. In a simplified version of how these cOMptinents interact when reading' polysyllabic words, students identify possible units within a word, relying Mostly -upon:familiar syllabi% but figuring out unknown units when necessary. They identify, -0,iondunce, and blend the units in a flexible manner until a match is made with a word in their oral vocabulary. s Literaey Theory and Rese THEORETICAL Ett CKGROUND Single-Syllable Units Adams (1990) maintains that skilled readers' ability to parse and identify long. words is based on their knowledge of spelling patterns. This does not nletilett. students simply need to learn a limited number of syllables. Such an inturpretrition.is incorreat not Only because there are km many syllables (over 5,006,SOnidhig., Adams), but also because a set of letters that constitutes a syllable in- one mord inay, not be a syllable in another word. What kinds of spelling patterns, then, might be emphasized in a program o instruction? Although syllabication is more than identifying a set of memorized Syllables, teaching certnin syllable units should still be helpful. Affixes that function as syllables are worth considering because they are limited in number, occur frequently, and, especially in the case of suffixes, are reasonably consistent across words. Support for another, more general category of syllable unit is found among linguists who agree that only certain combinations of vowels and consonants are permitted in English syllables, and who generally accept the existence and utility of two types of syllablesopen or free-vowel syllables (e.g., mo) and closed or checked- vowel syllables (e.g., om) (Groff, 1971). The ability to identify open and closed syllables in print not only enables students to pronounce frequently occurring kinds of units, but also should help them perceive likely units without the aid of division rules. Additional evidence in favor of teaching affixes and open and closed syllables is present in a study by Shefelbine, Lipscomb, and Hem (1989). They found a significant relationship between students sight knowledge of these syllable units and their ability to read real, polysyllabic words; they further noted that some students could not accurately pronounce common :yllables in isolation even when given plenty of time. Clearly, an important part of identifying syllables in a word is being able to "read" each syllable unit Some students do this poorly and hence P:v .:nable to read words even when the words are divided for them. Programmatically, this study developed syllable automaticity first by teaching students vowel generalizations for open and closed syllables ani then by providing much practice in reading a wide variety of such units. As recommended by Adams (1990), Durkin (1989), and others, students were encouraged to pay attention to all letters in a syllable A strategy for remembering irregular or less easily decoded affixes involved spelling them by letter name. Syllable Pattern Identification This variable involves identifying and combining likely syllable units when they are strung together in a polysy Ilabic word. Although some syllable patterns entail entire units that frequently occur in words (e.g., affixes), others depend on the frequency with which certain strings of letters occur (Adams, 1990). Since more frequent patterns tend to occur within syllable boundaries lather than across syllable boundaries, less common interletter associations automatically "pull" syllables apart. Fox example, since d is 40 times less likely to be followed by an n than an r, skilled g4 3 !.,. *.144,00c-04 ksAbla 1,*114rY betv-ieP (1,4114-11*Aid#0,1*P9t, between d and r in chddren.. Ac0444109 A441Ps.iiiie''Olitiiire`itie4101°44*eri,qt*.,;A,p,00# ,pattern associations determines *oak tp.ttp.ks;* pulled tQgether . *hicii 1*'04.110 aRO.T-Or 00144, tst, not inkait,ly .1:40 paid*. Whether 4, string of letie,r,s,4a ,q.PATIPO lot*: iliatomuna them Thus, klentitying **as_ of syllableiuires ,deVelePed,and comPiet knOwledge,of letter 444;sPelling pathtknQw ge needed,for reading single syllableS in kolation., A.S.lar,;,,$)Rorilatie.,11A00Fice,;; Adams, (1990).suggesta 'that teachers encourage ,stildelts 'Sequent:es ,that occur in syllables, words, and hlends aiiddigräpli METHOD Participants The 51 students in this study all had difficulty with syllabication according to criteria described below. They were selected from four sixth-grade classes iOajüni9r high school and four fourth-grade classes in two elementary schools that 'ler the juniorbigh. The schools were part of a large urban district and Served a heterogi4e6us. lower- and middle-class population. There were 28 fourth graders (14-syl1abicAriiit instruction and 14 control) and 23 sixth graders (15 syllabic-unit and 8 control). In accordance with teachers' wishes, students were assigned to a condition by classiither than randomly. All of the students within a class who met the criteria:participated. Students in this study correctly read fewer than 13 of 22 polysyllabic psendOora on a measure that followed a 2 by 2 design: two-syllable versus three-syllable Pseudowords and morphemic versus nonsense units, for example, admem, no.101bUs, Sasvop, and fudlempo. Their mean percentile seores (and standard deviations) in word identification, as measured by the word identification subtest of the Woodcock Reading Mastery Tests (Woodcock, 1987), were 40.6 (17.1) for the fourth graders and 17.1 (13) for the fifth graders. Vocabulary knowledge was assessed vie the vocabulary sub:est of the Stclord Achievernont Test (Gardner, Rudman, Karlsen, & Merwin, 1982). This particular measure was used because items are read to students and hence are not affected by poor decoding ability. Vocabulary was included because students with higher vocabulary knowledge should be more successful in matching approximate pronunciations with real words in their oral vocabulary. Mean vocabulary percentile scores were 43.7 (SD= 25.0) Li the fourth-grade group and 27.4 (SD= 21.7) in the sixth. Instructional Program for Experimental Group Over a 6-week period, the students in the syllabic-unit instruction group were taken out of their lancuage arts classes and taught 30 10-minute lessons, one lesson a day. The 14 fourth gr tders were taught as a group by a graduate research assistant. The 15 sixth graders were taught in another group by the principal investigator. Both' teachers followed identical lesson plans that included detailed descriptions of teaching t 226 4 "cz procedures and actual content. The instruction was teacher-directed and fast-paced, requiring students to respond frequently and as a group (i.e., chorally). Four teaching routines formed the core of the syllabication program: transformations, sight syllable practice, practice with real words, and division practice. Initial lessons included only the first two activities. When students became more comfortable reading single-syllable units, practice with real words was added. Di on practice was incorporated still later on in the latter part of the instructional progiarn. Transformations. This routine developed students' ability to read open and dosed syllable units (e.g., om and mo). After an introductory lesson on two hasic.vowel, generalizations (one vowel at the end is long; one vowel not at the end is short); students rmlarly were asked to read 15 or so open or closed syllables. The syllables were prestated in sets of three that all contained the same vowel, for example, ogmogmo. First, the teacher wrote og on a chalkboard. The students read it. Then the teacher changed or "transformed" og to mog by adding an m. Students now read the new syllable. Next the teacher erased the g and the students read mo. This,process was repeated with other sets of syllables such as fifimim and abrabra. When students made vowel errors, the teacher quickly prompted them to apply the appro- priate generalization. (Where is the vowel? Is it going to be long or short? What sound? What syllable?) Sight syllable practice. This routine developed students' ability to identify at sight over 50 acixes and an assortment of Latin roots. Fifteen to 20 syllables were presented during each lesson. Five were new, 5 had been introduced the previous day, and the remaining 5 to 10 were from more distant lessons. Each syllable was practiced in at least four separate lessons. During a lesson, the teacher wrote each syllable on the chalkboard and asked students to read it. Syllables were reread in random order two more times. Speed was not encouraged since it was important for students to take time to figure out syllables they did not know at sight. Difficult irregular syllables wm spelled out by letter name when they were missed or forgotten. Practice with real words In this activity, the teacher wrote teal polysyllabic words on the board, syllable by syllable but with no spaces or marks between the syllables After writing one syllable, the teacher paused and the students read it; then the teacher wrote the next and the students read that one. and so on. Students then read the entire word as a whole. When appropriate, the teacher pointed out how accents and schwas cause individual syllables to change. Tea to 15 words were pre- sented and read twice during each lesson. These words, mostly three or more syllables in length, were not seleetecl from any :larticular source but were meant to be challenging Students particularly enjoyed this component of the iessans because they felt, they were not dealing with "baby" words. For example, the ;011owing words were used in Lesson 16 radar, notation, vitamin, dece'tful, unchangeable, confidence, preventive, disagreeable, disgraceful, experience, nitrogen, tobacco, cumbersome, reconstruction, and adventure. Division practice. During &is routine, .:tudent: were again presented with unfamiliar polysyllabic words. This time, rather than being assisted with each syllable, 140100pproach 4hey'were encouraged tofigure out possible units on their own: ?beteg* deyelOPed: AlicieknoWiedge of possible strategies': by, piMiTig,L419,010,010,0f4 .04'1*!* , POPP,0401deilts"Yaleil theY hick 0000115',',$t,040*..**, PO, o loo for p0.00,,,affiiCa'firit. They thekwerelolOcatelikely oick y4ihiaii**3#4i.p044, pam 0.00), attention tO open and'cleSed sYllabies: 'Snitlenti,were:afack tiOili.000:9 tonerozations -to tieltf,diem ida4tirybpeii foifciosc4 syllable patterns (4fOr.,i4OW,e'r, tWo or more consonantS, try a short sound, ,iii14, 09 tOrli#3;*0 folowe ,by a Single consonant, first trY a lpi* sound 04 tiien, *kik** of, flekibilitylot,ating alternative Uniti,ff the twit- cineS;Clo **fspap,tilr*, itieSSed thrOu-gh iiitiCieling and-prompts and through ,Wordslhat AtAnot common patterns. This latter approach is similar to the `Ifree.-wheeline svllcblcation snategy advocated by Groff (1971). Instructional Program for Control Group Students in the control group stayed in their regular language arts classes and received no special instruction. Although the lack of a treatinent for the enntrtitgrouti, is commonly critized in studies of this design, the choice here was deliberate;i* the weight of the evidence against the efficacy of traditional instruction: Tniit*, more directly, neither the investigator nor the principal and teachers Could.jnitifYtaking valuable class time to teach syllable division rules and provide syllable diVision practice, activities that repeatedly have been ihown to have little effect upon syllabication ability. Analysis The dependent measure of students' syllabication ability ries the total number of polysyllabic words correctly read on the Woodcock and a second graded word list, the San Diego Quick Assessment (La Pray & Ross, 1969). The two tests were administered twice, first as a pretest and 7 to 9 weeks later as a posttest. The average number of weeks between testing was 8. Three-step, fixed entry multiple regression analyses were used to analyze the effectiveness of the instructional inte:vention. Using posttest scores in polysyllabic word identification on the Woodcock and San Diego as the dependent measure, pretest scores were entered on the first step, folluwed by vocabulary, followed by treatment condition. RESULTS Stuticnts receiving the syllabic-unit instruction made significantly greater progress in their ability to identify polysyllabic words than did those receiving no special instruction. This was true for students in Grade 4, F(3, 24) = 5.4, p< .05; students in Grade 6, F(3, 19) = 13.7, p<.01, and students in Grades 4 ;Ind 6 combined, F(3, 47) = 14.5, p<.001. The interaction of grade with treatment was not significant. Nor was there an interaction of treatment with vocabulary. Pretest and posttest means (and standard deviations) of correct polysyllabic words on the Woodcock and the San Diego *,- 228 Literacy Theory and lieserirek' were as follows: Fourth grade syllabic unit-64.1 (9.0) and 74.6 (11.0); fourth grade control-60.4 (12.5) and 64.6 (12.7); sixth grade syllabic unit-63.1 (13.5) and 72.6 (16.4); sixth gtude control-73.4 (13.6) and 75.6 (12.1); fourth and sixth grade syllabic unit-63.6 (11.4) and 73.6 (13.8); and fourth and sixth grade control-65.1 (14.1) and 68.6 (13.3). DISCUSSION Directly teaching students how to pronounce and identify syllable units and then showing them how such units "work" in polysyllabic words appears to be a Worth-. while component of syllabication insttuction and should help reduce or retnediate this source of reading difficulty among intermediate students. It is noteworthy that the few successful syllabication instruction studies reviewed by Johnson and Bautnann (1984) also included attention to syllabic units. Cunningham (1975, 1979) taught students to use familiar words to identify unfamiliar one- and two-syllable words. Gleitman and Rozin (1973) taught a shnple syllabary to kindergartners. The results of this study are encouraging when one considers that (a) the students only received a total of 5 hours of instruction and (b) the dJperAent measuzes involved real words that were neither ditectly taught nor patterned after words that were taught. Some might argue that improved comprehension while reading connected text is a more appropriate measure of syllabication instruction program. We feel such a standard may be misleading because many students will continue to have problems with comprehension because of limited fluency, background knowledge, and/or comprehension strategies. Howe% aver time, students' increased ability to figure out polysylleric words independently should positively affect other problem areas and comprehension in general, particularly if students read frequently and widely. There appears to be a definite need for at least some syllabication instruction at the middle-school level Among the eght fourth- and sixth-grade classes we assessed, 15% to 20% of the studentz typically were having difficulty reading polysyllabic words In one low-level sixth-grade class, the proportion was close to 50%. Because of its initial emphasis on single-syllable units and v.,wels, the syllabic-unit approach in this study seemed appropriate even for students vho were having difficulty reading ;ngle-syllable words substantial proportion of our students had major difficulty with vowel sounds, so much so that we brieay reviewed vowel digraphs and the final e generalization ) The constant use of syllable units, which are unfamiliar and often nonsense like, discouraged students from relying on sight words and forced them to learn and apply new strategies that focused on spelling patterns. As mentioned earliei, including real but difficult polysyllabic words kept students from thinking they were being recycled through first-grade phonics and noticeably improved their motivation. There are differing views of the role of vowels in word identification and syllabication Some reading textbooks state that vowels are relatively unimportant, citing, as an exampk, our ability to read sentences such as "_ l_k_ t_ r _d." Adams (1990) suggests that vowels act as spacers and that replacihg them with an asterisk still allows words to bz recognized (*mp*ss*bl* = impossible). This may be less true for less common words (*phr*d*s**c and *pp*n*g*). Our own position is that skilled 23 vailariba ., I.. , . , . - , eai OP this _Oecaus.c of .rl,r,w44,v,jcve,d ti !919w1c4se 411.4-Pu teit,14*-9rds 'in .00:ttiVlar De0 . 0040-00,40-i*ii:e41.146 04. , ,L**48L9.T10,iii;444: '9.-iirPOkt*OP:itt:0*'/44ithSr 4104, if o 01-0;...1REIger'Oe-91f-001,, lijr,teaCher*,,Olei4s070:.Thii eikuttils-40-,devOopkolit-:4-*, mui1cumcilum -Ot 4994U-0444 li-4:.s.fli;*c.$0.P. 0,36 beYPPd our Iual, 0 l44sons because , %Otait,ina!t 04in 5 hours of instruction will iirOt*Ok* 0,e. 44 perfosinance ., ,4;z ::,24 REFE41,3NCES- , - Mims, M. J. (1990). Beginning to read: Thinking and learning about print Cambridge, MA. MlTPteas Canney 0 & Schreiner, R. (1977). A.study Of the iffeativeness of seleeted syllabication suleaaad , phonograrn patterns for word attack. Reading Research Quarterly, 12, 1024124.. Cunningbani, P. M: (1979). A compare/contrast theory of mediated Word ideariti*Ca. ReaifitiiTeacliere 32, 774-778. Cunningham, P. M. (1975). Investigating a synthesized theory of mediated word identification.'ltiodirig. Research Quarterly, 11, 127-143. Cunningham, P. M., Cunningham, J. W., & Rystrom, R. C. (1981). A new syllable strategy and reading achievement. Reading World, 20, 208-214. Durkin, D. (1989). Teaching them to read (5th ed.). Boston: Allyn & Bacon. :j. Gardner, E. F., Rudman, H. C., Karim, B., & Merwin, J. C. (1982). Staiford achievement test. San Antonio, TX: Psychological Corporation. Gleitman, L. FL , & Rozin, P. (1973). Teaching reading as a syllabary. Reading Research Quarterly, 8, 447-483. Groff, P. (1971). The syllable. le. nature and pedagogical usefidness. Portland, OR. Northwest Regional Ecluearional Laboratory. Johnson, D. D., & Baumann, J. F. (1984). Word idendfication. In P. D. Pearson (Ed.), Handbook of reading reseatrh (pp. 583-608). New York: Longman. Just, M. A., & Carpenter, P. A. (1987). The psychology of reading and language comprehension. Boston: Allyn & Bacon. La Pray, M., & Ross, R. (1969). The graded word list: Quick gauge of reading ability. Journal of Reading, 12, 305-307. Mewhort, D. J. K., & Campbell, A. J. (1981). Toward a model of skilled reading: An analysis of performance in tachistoscopic tasks. In G. E. MacKinnor & T. G Waller (Eds.), Reading research: Advances in theory and practice, Volume 3 (pp. 39-118). New Yam: Academic Press. Samuels, S. J., LaBerge, D., & Bremer, C. D. (1978). Units of word recognition: Evidence for developmental changes. Journal of Verbal Learing and Verbal Behavior, 17, 715-720. Shefelbine, J., Lipscomb, L., & Hem, A. (1989; Variables associated with second-, fourth-, and siztngrade students ability to identify polysyllabic words. In S. McCormick & J. "%tell (Eds.), Cognitive and social perspectives for literacy research and instruction (pp. 145-154). Chicago. National Reading Conferen.t. Woodcock, R. W. (1987). Woodcock reading mastery tests (revised). Circle Pines, MN. American Guidance Setvice. _ JL, INCIDENTAL LEABNIWPF WORI:i*EANLISOW KINDERGARTEN,,ANDORgTIGRADE-CIKDONI*ROLO REPEATEWREAD7ALWAYENT$' Cynthia B. Leung apii John J. iqiuiski University of Delaware Researchers continue to seek an explanation for the remarkable growth in children's word knowledge. A fairly reasonable estimate of vocabulary growth' bythe average child in the elementary grades is about 2,000 to 3,000 words:a, year (Nagy 8. Anderson, 1984; Nagy & Herman, 1987). Systematic, aggressiVevrOgiarna of direct vocabulary instruction, however, appear to be able VS teach only abont 250 to 350 words per year (Beck, Perfetti, & McKeown, 1982). Indeed, therois nuiv f*ly good agreement that the vast majority of vocabulary learning takes place incidentalbi, and that this incidental learning occurs when learners encounter unknown worth in independent reading (Nagy, Anderson, & Herman, 1987; Nagy, Herman, & Anderson, 1985). Two recent, well-designed studies suggcst that reading aloud to children is also a rich potential source of vocabulary growth (Eller, Pappas, & Brown, 1988; Eiley, 1989). Eiler et al. (1988) studied the effects ot reading aloud two Brian Wildstnith picture books to kindergarten children. Ten words in each of the books were targeted as unlikely to be known by subjects. The 20 c Aren, none of whom were judged to be reading in the conventional sense, were individually read one of the books and then asked to "read" the book back to the examiner using die illustrations as prompt& Each child's use, failure to use, or misuse of these 20 words in stoty retellings was the focus of the study. On each of the next 2 consecutive days the book was tweed to the child, followed by the child's "rereading" of the book. No special attention was paid to the targeted words; there was no instruction with these words, nor eny discussion of them. Approximately 3 weeks later the se= procedures were repeated with the second picture book. The analyses of the kindergartners' reading reenactments showed a significant increase in the appearance and contextual, appropriate use of the targeted words rith each retelling of the stories. The results suggested that vocabulary growth in young children can take place through exposure to the oral reading of written text, Although these results are encouraging, the study, which was ran d a larger investigation, did have a number of limitations, most of which were acknowledged by its authors. Most notably it did not include a control group of children who were not read the stories, nor did it include any pre- or posttest measures of the children's knowledge of the meaning of the targeted words. Elley (1989) also studied the effects of reading aloud to children on their vocabu231 gt514_, 232 lary growth. In one study, which included 157 seven-year-old children, classroom teachers read the same story aloud three times to the children as a group. There was no instruction or discussion of the targeted words. Analyses of the differences between the pre- and posttesting of 20 difficult vocabulary words, using a multiple-choice vocabulary test, showed a mean increase of over 15%. This study too had limitations, primarily the iack of a control group. In a second study, using 127 eight-year-old children, Miry (1989) used a welldefined control group to look at the effects of reading two different books aloud thrpe. times. He also added an experimental condition wherein the teachers who reaphe books to the groups also offered some minimal explanation of the meaning of the,. difficult words. For or.e of the books the results were very similar to the results of the first study, showing pia increase of about 15% in posttest over pretest scores for the group that was read the book without explanations. When child:an were mad the book aloud and the teacher offered some minimal explanation of the difficult words, the gains, however, were almost 40%. The control group improved only about 2%. Vocabulary gains were much lesa impressive with the second book only a 4.4% increase as a result of three readings of the book, and an increase of about 17% when explanations of the difficult vocabulary were added. Thus, large text-specific effects appear to exist in vocabulary acquisition from read aloud events. The Elley (1989) article is also instructive in that it looked at the degree to which six variables (e.g., number of occurrences of the word in the text, the helpfulness of the language context clues) contributed to the probability of a child's learning that word. Thus, al"Tugh there is a somewhat limited literature to support the thesis that reading aloud to chi! is a rich source of vocabulary growth, the resegrch conducted to date is encouraging. The present study was an attempt to extend that literature through a partial replication and extension of the Eller et al. (1988) study, primarily through the use of a control group, through comparing the performance vi kindergarten and first grade children using the same texts that were part of the Eller et al. study, and through using a pretest-posttest design. METHOD Subjects The sub;ects were 48 children (24 kindergartners and 24 first graders) who attend1 a largely middle-dass suburban school in New Castle County, Delaware. Teachers from four participating kindergarten and four first-grade classrooms were asked to nominate all their students who had no previous history of high absences and whom they considered to be functioninr within an "average" range in language development and academic achievement. From those nominated students, 6 children (3 boys and 3 girls) were randomly assigned to a control group and 6 to an experimental group, with the restriction that an equa number of boys and girls be included in each group The age of the kindergarten subjects ranged from 4 years 9 months to 6 years 9 months, first-grade subjects ranged from 6 years 6 months to 7 years 11 24 Incidental Learning and Word Meanings 233 months. The mean age of the kindergarten cKtdren was 5 years 11 months, and 7 years 1 month for first graders. Procedures and Materials All subjects were given the Vocabulary Subtest from the Wechsler Preschool and Primary Scale of Iruelligence (WPPSI) (Wechsler, 1967) as a measure of general vocabulary development. This subtest was individually administered by specially trained examiners. All subtests were scored by the Second author, a certified school psychologist. The mean raw score of the WPPSI for the kindergarten contragroup was 26.58 and for the experimental group was 25.25. The mean raw score for the first-grade control group was 27.92, and for the experimental group 31.67. Two picture storybooks by Brian Wildsmith, The Owl and the Woodpecker (1971) and The Lazy Bear (1973) were used. They: were the same storybooks Eder et al. (1988) had used so that a comparison of the results of the two studies could be made. Likewise, the 20 target words or phrases used to study the incidental acquisinon of vocabulary were the same words used in the 1988 study. The ten target words/phrases from The Owl and the Woodpecker were: tapping, screeches, peaceIpeaceful, swooped, wise and clever, crotchety, crafty, gnaw, struggled, echoed; for The Lazy Bear they were. woodcutter, sniffed, enjoyed, speed, curious, marvelous, head over heels, shallow, bully, and glorious. A pretest of story vocabulary was constructed using the above list _ r target words/ phrases. The vocabulary Items from the WPPSI wen. alternated ,vith the target words in test administration to insure that the story vocabulary pretest list would not be so uniformly difficult that the children would find the task frustrating. The posttest consisted only of the 20 story vocabulary pretest items. All 48 subjects, both experimental and control, were individually administered the pretest by a team of seven trained examiners, including the second author of the paper. Examiners pronounced a test word and asked students to tell what it meant. Children were asked to tell the meu....6 of the words following a free recall format. If no response or an incomplete response was given, examiners asked the children to tell anything else they knew about the word. Subject responses were recordd through examiner notation. Approximately 1 week after pretest administration, each expenmental subject was Individually read one of the books by onc of three trained, adult readers, including the first author, the order in whmii the books were read was minterbalanced. When the adult .7eader 4.ornpleted the reading, subjects were shown a second version of the book, me with the wntten text covered so that only the illustrations could be used as retelling prompts. The wntten text was Loncealed so that the subjects who might have some beginning reading skills would not approach the task differently from those who could not read. The subjects were asked to "read" or "pretend read" the story to the adult, making their story as much as possible like the one that was read to them. The read aloud events were repeated in exactly the same form and sequence on 2 more days of that week. In most cases the three reading sessions took place on Monday, Wednesday, and Friday. The following week, the same procedures were followed with three reading sessions for the second book. No attention wa.> drawn to 1,4 ^-7:7 `nditf dielineahuhttyin the bOoks,;nOr *tti vccatuinry itoos0:1-.with tha Au**. BOA itie iduli.040dir,g4 nnd the children'..s ritteilints-vere audio record4,the *riitPir krin0i,ii.4rOsP of10y#0,, "1.kMii1y',..fol.itewed: the regniar kindergarten or first-grade schedule deringthe -coups-41 baring.the Weeictollb*intthe taiding ot.t6- seeonOl*Olci the toCY;fr.oiciihninii, Pc144. PP the 14.= 49,1t,*06#'3,vi* *J04*(1; to both 01P:- expérun nial *4-0i0 egtiir0 subjects u4ng C4e. On* Ot:t7c-0.4**4,s*c.Te OPOOed in Oia13,01S#, eriaminir gr-dririnistered"the preteSt and thi'postteSt ha aniParti 'The system 'for scoring the story yoCnhninripreteltind'Osttestliins: 0 pointNo knowledge of word inenning or-incorrect responie I pointPartial or. incomplete knOWledge of ivord meat', 2 pointsTargetword used in appioPriate, meaningful context 3 pointsSynonym or definition of target word All of the pretests and posttests were scored by the second author who, thrOughout the study, remained blind to the group membership of the subjecti. To establish, tire reliability of the scoring system for the pre- and posttests, both authors indepeodeutlY!., scored 10 randomly selected pre- afid posttests' administered to ldndergartnera and,,10 administered to firit graders. The rate of agreement for the scoring of eacitittrir-viat over 95%. Discu-skins were used to resolve differences in scoring, which were never more than a I-point differerce. Given this high rate of agreement, all other teas were scored only by the.second author. The coding system developed by Eller et al. (1988) was used to score the reading reenactmeut responses. The system consisted of 5 categories: CATEGORY ONE (NO/FAULTY KNOWLEDOE)-1 point target word not used, or nonsynonymous replacement used CATEGORY TWO (DEVELOPING KNOWLEDGE)-2 points mrget word used inappropriately or with syntactic error CAIEGORY THREE (SYNONYM)-3 points synonymous word or phrase used CATEGORY FOUR (ACCURATE KNOWLEDGE)-4 points accurate use of target word CATEGORY FIVE (GENERALIZED KNOWLEDGE)-5 points accurate use of target word in appropriate context and elsewhere in the text A total of 1,440 instances of word use (24 subjects x 20 target words x 3 reading') were analyzed according to tht. above system. Each target word received a rating of front 1 to 5 for each child. The total points for each category of word use were calculated. RESULTS Findings frum Retellings Table 1 showa the frequencies and percentages of the target words used by the combined kindergarten and first-grade children in the experimental group, according to the five category ratings The scores from the kindergarten and first-grade experimental 1-* ifOde:nia441*k._ .FreqifaCies,(andPereentagei) of Category Ratings forargetlexical Iteins-by *14* faiXindergartenandFirst.Oradecaribir4__ Cat4gmY .14,Q/FaitlOcwledge f pit. 1) Developing Knowledge 367 (76.46) 315 (65.63) 1 1 (tat, 2) ( .21) ( .21) Stnonytn (Cat. 3) 62 (12.92) 50 (10.42) 60 (12.50) 96 (20.00) Accurate Knowledge (Cat. 4) Generalized Knowledge (eat. 5) Totals 0 ( 8 .00) 480 P7' (61.88) ). 0 ( -00) 53 O1.0.4I. 120: (25.00) 10 ( 1.67) ( 2.08). 480 '480' subjects were combined because no significant differences between these two,groupi were found for the total scores, an unanticipated finding. Tie number of wOrds in Category I (no/faulty knowledge) and Category 3 (synonym) decreased across tile three retellings, whereas the number of words in Category 4 (accurate knowledge) and Category 5 (generalized knowledge) increased. Using the 5-point scoring system, each child wiL, given an individual the first retellings of the two books combined, the second retellings combined, and the third retellings combined. Scores per reading could range from a low of 211 to a high of 100. Analysis of variance yielded significant main effects for the rett ,gs: F(2, 46)=24.04, p<.0001. A multiple comparison of means (Tukey's HSD Test) indicated statistically sig- nificant differences (p.05) between the first and the second reteiling scores and between thc orst and the third retellings. There was no significant difference; however, between the second and the third retelling (Retelling 1: M=31.92, Retelling 2: M=38.33, Retelling 3: M=41.38). Table 2 shows means and standard deviations resulting from separate analyses of retelling scores per reading for subjects in kindergarten and first grade and for boys and girls. ANOVA by grade and gender was carded out, with difference scores between the first rctelling and the higher of the second or third retellings. No significant main effects were found for the gender of the subject or the grade; however, the difference in retelling scores between kindergartners and first graders approached significance, p<.065 (kindergarten: M=7.5, first grade: M=13.42). Analysis of Story Vocabulary Pre- and Posttests Table 3 shows the means and standard deviations for pretest, posttest, and gain scores by the grade and gender of subjects. No statistically significant difference was found between the pretest and posttest vocabulary scores with respect to treatment, 236 Literacy Theory and Riseiirell Table 2 Means (and Standard Deviations) of Retelling Scores by Grade and Sex Retelling 1 Kindergarten First Grade Boys Girls Total Group 29.83 33.08 32.58 30.33 31.46 Retelling 2 (6.78) (5.73) (6.96) (5.77) (6.36) 35.33 (11.31) 41.33 (13:28) 39.83 (13.88) 36.83 (11.24) 38.33 (12.44) Retelling 3, 36.58 (10.3) 4,5.58 (1I.,12) 41.33 (13.15) 40.83 (10.04) 41.08 (11.45) grade, or gender of the subject; nor were there any significant interactions. Differences by gender and treatment group, however, approached significance, with girls showing. larger gain scores than boys, and with the overall experimental group scoring higher than the control group. An analysis of covariance, using the WPPSI raw scores as a covariate, also did not show statistically significant differences. DISCUSSION In general the results support, to a limited extent, the earlier findings of Eller et al (1988), though the effects f reading aloud on vocabulary growth were not as strong in this study. Although there was a significant increase in subjects' appropriate use of difficult vocabulary from the first to the second oral reenactment of the books, the differences between the second and third reenactment, though in the expected directien, were not significant. One obvious difference between this study and the Eller et al study was the use in this study of a first-grade population in addition to the kindergarten population. However, the difference in the populations does not seem a reasonible explanation for the difference between the two studies since there were ro statistically significant differences between the performance of the first-grado and kindergarten subjects. Overall, our results are not terribly different from those of Eller e.t al.; all of the scores were in the expected direction. One unanticipated finding throughout this study was the general lack of significant differences between the performance of the kindergarten and flugrade children, although the differeace in retelling scores by grade did approach significance. This lack of differences was not restricted to the effects of the experimental treatment. There were likewise no sigrificant differences in the pretest of the story vocabulary items, nor in the WPPSI scores. The latter finding suggests that the kindergarten children refened to us may have been more advanced in language development for their age than were the first graders for their age. In particular, it seems that facility with oral language was not always taken ii. . account in the referral of kindergarten students since the range in subjects' ability to "read" stories varied greatly. Some kindergartners were still at the labelling stage while others supplied full and sophisticated stories. The second major finding in the study was that there were no significant differ- 21 4 'Mama Eeisining an.ti_Woiii Meanings Table 3 Means (rind kandard Deviations) for Pretest, Posttest,. and Gain Scores by 'Gtude-and Sex . , -At'FlerWIPn coptrol .ExOirimenta1 -FOt rrride : ariliol . , Experimental Boys Control Experimental Girls Control Experimental Total Group Control Experimental 17.92 (8.89) 15.42 (8.18) 21:08.(12.19) 3.17:(5:25)' 22.17 (9.88) 5.7 (5.31) 20.75 (10.70) 25.50 (9.30) 5.5&.(4.58) 25.00 (8.77) 31.00 (10.51) 6.00 (5.86) :"9.50 (9.88) 7 23.42 (10.99) 12 (8.85) 18 (13.07) 4.83 (4.49) 3.67 (5.38) 17.17 (9.49) 24.2) (12.4) 17.00 (7.12) 26.08 (8.90) 3.92 (5.58) 9.08 (4.23) 19.33 (9.73) 23.29 (10.84) 26.58 (10.95) 4.38 (4.38) 6.38 (5.48) 20.21 (9.63) 'Gain Xere3 were determined by finding ihe diffenmze between pm- and posoest scores for each subject In some cases the mean gain score tnay be larger than the difference between the means of the pre- and posttest scores. ences by treatment between the pre- and posttest measures of subjects' ability to provide verbal definitions for the 20 difficult words taken from the texts that- were read aloud. The research by Elley (1989) provided some basis for predicting significant differences in such a measure, but there were some important differences between this and Elley's study. Elley used s multiple-choice test, whereas we asked subjects to produce a verbal definitiona melt more demanding task. It may very well be that there is a contherum of difficulty of tasks which assess knowledge of vocabulary, so that being able to use the word in a contextually appropriate way in the retelling of a story, or choosing a picture from four choices to match a word spoken by an examiner may require less depth of vocabulary knowledge or perhaps less expressive language skills than prcaucing a verbal definition. Althou& we have not yet analyzed, except for the targeted vocabulary, the full conten! of the reading reenactment protocols, we did note several in which the children were able to use some words in a contextually appropriate way in the reading reenacumnt, but then failed to offer even an attempt at defining the word when asked to do so on thy pc.lricat. The present study was different from the Elley study also in that our subjects were younger. It may be that the impact of reading aloud is greater for older children. In addition, the books used in this study were different from those used by Elley 0989), and Elley did show substantial text specific effects on vocabulary acquisition. The results of the present study are enzouraging in that they partially replicate the results obtained by Eller et al. (1988) which suggest that reading aloud so children has an important, positive effect on vocabulary growth. However, the study also .7 points to a number of implications for future research dealing with the influence of reading aloud to young children. First, it ;s unfortunate that more is not known about the relationship of different approaches to assessing a child's knowledge of a word. There is an outstanding need for research in this area. Second, the procedure of reading books aloud to individual children was very time consuming. The recent work cf Elley (1989) suggests that reading to groups of children has a very sinificant effect on vocabulary growth; the differential effects on vocabulary development, if any, of reading aloud to individual chile:en as compared to reading aloud to groups need to be studied. El ley (1989) has begun some very important analyses ot the types of vocabulary items that children are likely to learn from read aloud events and has also found substantial text specific effects on vocabulary acquisition. This work, which eventually might provide teachers some useful guidelines for selecting read aloud books, also needs to be extended. Finally, the studies by Eller et al. (1988), El ley (1989), and the present study all used three readings of the same book assessing vocabulary growth. Certainly there is no good research basis for concluding that three rereadings is the optimal w.mber. Althoygh Eller et al. did find significant improvement in the use of the targeted vocabulary from the second to the third reading reenactment, we did not. Based on inspection of the reading reenactment protocols from our study, it appears for some rhildren, most likely those with more restricted language skills, having the text read aloud three times seemed very beneficial. In addition to in..reased appropriate use of the targeted words, many of these children produced increasingly longer, more elaborate reading reenactments protocols. However, some children, most likely those with very good language skills, seemed bored '.yr the third reading and retelling. The reading reenactment protocols of such children actually showed a reduction in length and detail with subsequent retellings The epitomc of this tendency was exemplified by one kindergarten child, who when told that he would be read The Owl and the Woodpecker for the third time, asked if it were the only book that we had, and then volunteered to bring some of his own books for us to read to him. Becoming a Nation of Readers (Anderson, Hiebert, Scott, & Wilkerson, 1985) concluded "The single most important activity of bulding the knowledge requited for eventual success in reading is reading aloud to children" (p. 23). The studies reviewed in this paper proNide added justification for that statement and suggest that ar, important way in w h...h reading aloud to children contributes to thei success is by broadening the language base on which successful reading exists. REFERENCES Anderson, R C . Hiebert. E H . Scott, J A. & Wilkerson, 1. A Becoming a natton of readers. The report of the comnusston on readmg Washington, DC The National Instinite of Educanon. Beck, I L , Perfetti, C A , & McKeown, M G. (1982). The effects of long term vocabulary msuucnon on lexical access and reading comprehension Journal of Educattonal Psychology. 74, 506-521. R G , Pappas, C C , & Brown. E (1988). The lexical developmetu of kindergartners. Learning from wriuen context Journal of Reading Behavior, 20. 5-24. - ,1 El:ley, W. B. (1989). Vocabulary acquisition from listening to stories. Reading Research Quarterly, 24, 174-187. Nagy, W. E., & Anderson, R. C. (1984). How many words are there in printed school Eagliah7 Reading Risearch Quarterly, 19, 304-330. tt.:5 Nag, W. p., Anderson, R. C., & 'Hendon, P. A. (1987). Lemuirgyordrneenhtp, f,om cotiat during:, nonnaIreading.-knerican Educatir ono; Research J0i177141, 24; 237470. Nagy, Eamon, P. A. (1987). Breadth ai3O depth of votiabtljati4loweidge:ImPPe#91,scck acquisidOn and instrisction. In M. G. Mckiown & M. E antis (Els.), The iliitpire of vacalrldtirY; : 19-36): Irdlsclale, NI: Eribanni. ,acqutsiiion -Nagy, W: E. Herman, P A & Anderson, R. C (1985). Learning wards from context. ReadfrigietilWk QUarterly, 20, i3J-253. Wechsler; D. (1957). kanuai for the Wechsler preschool and primary scale of intelligence. New l'o4: PsyChological Corporation. Wildsmith, B. (1971). The owl and the woodpecker. Oxford: Oxford University Press. Wilesmith, B. (1973). The fray bear. Oxford: Oxford University Pmts. . ' 247 THE INFLUENCE OF PHONICS INSTRUCTION ON SPELLING PROGRESS1 Laurie Nelson National College of Education Ir cent years, linguists, educational researchers, and classroom teachers have come to embrace a theory of early spelling development, first advanced by Charles Read in 1971, in v nich children's spelling is initially dominated by their phonetic intuitions, and onk, later gives way to more acquired, conventional forms. Developmental spelling thei y underwent subsequent refinement by Edmund Henderson and his colleagues at the University of Virginia (Beers & Henderson, 1977; Gentry, 1977; Henderson & Beers, 1980; Zutell, 1980); actual "stages" of development were posited and field tested in varied settings. Based on these theoretical refinements, most children pass through a succession of reasonably clear and identifiable stages en route to becoming proficient spellers. Although evidence of these stages has been collected time and again, research is still lacking that tests the durability of this sequence in sharply defined instructional settings. The Read/Henderson findings are undoubtedly important, explaining how most children acquire English spelling; but do all children necessarily learn to spell in the specified sequence? And if not, what deviations occur? A precedent exists for suggesting instruction can influence how children treat and store words. Barr (1972, 1974-1975), DeLawter (1970), Elder (1971), and others have demonstrated the diffemntial ,ontributions of phonics and/or word-based instructton to children's word learning in reading. There are logical grounds therefore for subpecting instructional influence on spelling development. In this paper, the possibility of instructional influence on the Read/Henderson developmental spelling squence is systematically explored. Developmental Spelling Stages According to Henderson (1981), children show the first sound in a word in their earliest systematic spelling attempts. Soon thereafter, the final sound is represented Feet spelled F or FT is common at this beginning stage (sometimes called the semiphonetic stage). Next, young children enter a phonetic stage of spelling in which the beginning, middle, and end of a word are represented (Henderson, 1980; Read, 1971) Henderson (1980) suggests that children spelling in the pLnetic stage use one letter per sound. Consonant boundaries (the first and last sounds in words) are recorded wuld like to acknowledge the insights of Rebecca Rarr, Darrell Morris, and Jerry Zutell in preparatton of this manuscript. Literacy Theory and Research fairly reliably because the sounds we hear in mcst consonants can be traced to their correct letter-names. Long vowels, because they "say" their own letter-name, are also generally rendered correct (FET for feet; RIS for rice). Having assigned the alphabet letters a, e, i, o and u the job of recording long vowel sounds, however, children at this stage could be at a loss when they go to record short vowel sounds. Read (1971) found that children actually seem not to experience any confusion; instead they naturally record the short vowel sounds with the phonetically nearest long-vowel neighbor until such time as the correct short vowel pairings are learned. Thus, dress is spelled JAS or JRAS, rich is spelled REJ or REH, and junk might be spelled JOK (the nearest long-vowel neighbor for short e is the alphabet 1:icor-name A; the nearest long-vowel neighbor for short i is the letter-name E, and the nearest lonl-vowel neighbor for short u is the letter-name 0 [Read, 19711). The third stage and final one of interest in this paper, the transitional stage, also represents an advance in how children are thinking about words. Key here is the child's movement away from a one letter per sound spelling strategy. The child begins to grasp the pattern principle in English spelling aenderson, 1980; Henderson & Templeton, 1986). Consequently, Fong vowels are .,arked, if incorrectly a. times: the phonetic rendering of feet [FET] may now go to ).-_fE; rice, once RIS may now be RIES At this stage, according to Henderson, the correct short vowel pairings are learned (possibly as a consequence of reading experience/instruction [Henderson, 19851), and JRAS improves to DRES. Influence of Ltstruction on Spelling Developmental spelling theory suggests that children in tht phonetic spelling stage may rely on knowledge of long vowel letter names to spell short vowels. Yet. conventional (correct) short vowels are often taught before long vowels in systematic phonics programs Such instruction supports the development of understanding in a sequence inconsistent with the proposed developmental scheme. Therefore the inclusion of a phonics-instructed classroom could make a useful tcst case in a study of instruction's influence on spelling. Regardless of instructional type, it was thought that it would not be until the phonetic stage that children's spellings would be likely to reveal instructional influence, since consonants make a limited lumber of sounds compared with vowels. METHOD The spelling data used in this study were collected as part uf a larger project c ontrasting possible instructional influence on several early reading phenomena. These spellings were drawn from a working class, semi-rural, phonics-instructed kindergar- ten, and a low SES, word-based instructed first grade. The spelling lists were given in one-to-one sessions with individual children, four times over the year (September, December, February, and May). The list itself was adapted from Morris (1981, see Table 1). 249 nce`On. an-d Category Placement IS 'YX* Xst 4RES*- 'as'YO00: GRAvo*s*. bks. DRAS STICK SK STC CK SEK.,, SW EK CIK* STEK SIC* crEK . LAMP RICE LP RS RC RE: BEG RICH BO BOG JUNK IC JEK JK RJ LAP RIS RIC BAG REJ REG ROS* RIES RISE RU* RIG', REH RIB* JOC 10K COMB CM KM C JUC* JUK* GOC GUC* GOK GUK* COM CUM* JUNC GUNK COME KOM KUM* *Indicates anomalous vowel treatments in words othetwise phonetically spelled. Anomalies In Spelling Three patterns were noted in the phonics kindergarten during data collection, raising the possibility that the developmedtal spelling sequence outline(' by Read and Henderson does not hold for all instructional settings. First, correct.short vowels (a transitional stage feature), co-occurred with featums Read and Henderson argued were benchmarks of the phonetic stage. Thus dress was smiled ES, and rich was spelled RIH. (At the phonetic stage the child tends to use a single letter to represent a consonant blend or digraph whose actual spelling must probably be learned through reading.) Second, some of the more able youngsters handled vowel spellings in unusual ways. One child who could orally segment each spelling word accurately, could not decide how to record the long or short vowel sounds. She recorded the beginning and ending consonant and left a space for the vowel: JR S for dress, ST CI for stick, R_S for rice, and so forth. Another child, at a porut in the year when all five short vowels had been instructed, "used up" the five vowel letters for the short vowel words on _ 244 Literacy Theory and Researcli - the spelling test. She spelled feet correctly and rapidly, then paused be;ore attempting two other long vowel words on the list. Finally she spelled rice. RIIS and comb: COOM. These spellings suggest she had to htirriedly come up with a way to represent long vowel sounds. Whr asked why she had doubled the vowel in these words she said she had used feet to figure it out. Third, several children spelled some long vowels with the same incorrect letter, time after time. Mail, instead of being spelled MAL, was spelled MEL; feet was; _ spelled FIT. These children seemed to be using "inverse substitutions," going to the' nerrest-short-vowel-neighbor to represent the long vowel sound, whereas,Read.had. discovered most young spellers going to the nearest-long-vowel-neighbor to represent the short vowel sound. Read (1971) and Beers and Henderson (1977), had also observed this reverse substitution phenomenon on a limited basis. Table 1 represents both the early developmental spelling stages and the anomalies set A in the phonicsinstructed kindergarten for the set of one-syllable words used in this study. Analysis of Spellings from P.tonetic Stage To investigate the spelling anomalies described above morc systematically, some ground rules were established. First, only those spelling protocols that adhered to the phonetic stage criteria as originally put forth by Henderson (Henderson, 1980; Henderson, 1981) were analyzed, with the exception that words were coded as phonetic even if the short vowel were correct. For example, JES was coded as a phonetic stage spelling since the consonantal treatment fit Henderson's "spelling by sound" criterion and only the vowel was conventionally rendered. Unmarked long vowels as in MAL and FET were easy to code phoneticas every letter fit Henderson's criteria. Even nearest-short-vowel substitution spellings of long vowel words such as MEL and FIT could be coded phonetic because no vowel marker was present. (Vowel markers are a hallmark of the more advaeced transitional stage.) Second, protocols were included only if 5 or more of the 10 words fit the phonetic criteria Further, if only 5 words fit the criteria, the remaining 5 words had to be split between semi-phonetic and transitional stage spellings. These conditions met, the entire list s ir:luded in the analysis. A provision was made for including words from the phonics-instrrted class. Not until each specific short vowel had been instructed were words containing those short sowels included in ..ge analysis. This same provision was not matte in the word based class, where long and S h o rt vowel instruction was less systematic. Third, in looking for 5 or more phonetically spelled words on a list, to.sly words with both boundary consornts and the vowel (CVC-patterns) were considered. Feet spelled FE was insufficient evidence to code a spelling as phonetic stage, evidence also had to exist showing that the child could hear the ending consonant based on his or her spellings of other list words. Once 5 words on a list were tagged phonetic, however, as a sixth word feet spelled FE was included in the analysis, providing additional data to analyze for the state of the child's vowel knowledge. Of the 74 lists analyzed from to= classes, 19 were included on the basis of 5 words fitting the criteria, and the other 55 lists were included because between 6 and 10 words fit the criteria. For both classes, 62 of the 74 lists analyzed were. hulk. Times 25.1 Instruction's Influence on Spelling 245 3 (February) and 4 (May); only then were most of the children in the phonetic spelling stage. Having determined criteria for selecting relevant lists, the data from the two classes were kept separate for subsequent analyses, to preserve the possibility of discerning different instructional influences. The analysis was straightforward. Each word was noted and tallied if it contained what could be called a developmentally precocious treatment of the vowel. For exam- ple, dress spelled JES or JRES was tallied as a precociously correct short vowel spelling because it appeared at the phonetic stage. Mail spelled MEL was noted as an example of a nearest-short-vowel-neighbor substitution for the long vowel a. The rationa/e for this was based on the observation by Read (1971) that children use similarities in articulation to represent short vowel spellings. In this case, the same intuition is in effect but followed in the opposite direeion. It is hard otherwise to explain why long a would be spelled with an E rather than an A. Referring back to Table 1, the asterisked spdlings exemplify precocious, correct short vowel spellings in otherwise phonetically rendered words and intuitively spelled nearest-SHORT-vowellong vowel sounds. These spellitgs were tallied as influenced neighboi renderings by instruction. RESULTS Figure 1 depicts the incidence of instnictional influence in both classes. There is a clear trend towards correct spelling of short vowels in the phonir- class when comparcd with the word-based class. Based on 209 words analyzed., phonks-iastructed children correctly spelled short vowels often (78%), whereas otherwise the criteria for list inclusion indicated that they were in the phonetic spelling stage This contrasts with children in the word based class who, based on 160 words analyzed, correctly spelled the short vowel 31% of the time while in the phonetic stage. A z.- 100 90 80 70 60 50 El Precocious Short Vowel 78% Reverf4 Substitution for Long Vowels 40 30 20 31% 17% 10 3% 0 Phonics Word-Based Figurt. I. The incidence of instructional influence on long and short vowel spellings by class. 252 246 Litenicy Theory and Research test fee comparing proportions from independent samples showed that the two percentPeps were significantly different at the p<.001 level (Ferguson, 1976). At the same time, phonics-instructed children substituted the nearest-short-vowelneighbor for long vowels 17% of all 170 worth analyzed, whereas word-based instrueted children made this substitution only 3% of the time on 128 words analyzed. A z-test comparing phonics versus word-based instructed children's use of nearestshort-vowel-substitutions again showed the difference to be sipificant at the p<.04 level. The instructional influence on short vowel words was significantly greater than on long volael words (p<.001) for each class. DISCUSSION The results of this study indicate a trend towards accelerated coma short vowel spelling with consequences for long vowel spellings on the past of phonics-instructed kindergartners. Given an instructional program in which the short vowel sounds are intensive; rehearsed (about 2 weeks of direct instruction per short vowel) and constrantly reviewed as new consonant sounds are tetsoduced, it makes sense that a deviation in a traditional developmental spelling sequence might occur. This is particularly so given that the developmental sequence was first identified in a preschool settins where no strong print-oriented instruction occurred (Read, 1971) and refined in a more eclectic first grade (Beers & Henderson, 1977). The significant results in the phonics class raise two issues. First, what was proposed as a "natural" developmental sequence (Henderson, 1980) may be deter- mined in part by how children typically learn to read, namely via word-based methods. Given intensive phonics-based methods, the developmental sequence may be altered. One implication of this finding is that in future spelling research instruction needs to be documented in order to gauge its influence on spelling development. Second, at the same time thdt some specific measureable instructional influence pertaining to vowels was documented in this study, I suggest that the basic tenets of Read and Henderson's work seem to hold in general. Read (1971) and Henderson (1980) each describe the phonetic spelling stage as one in which children spell one letter per Am& The results of this study confirm this notion. Further, a portion of the evidence suggested that kindergarten children in this study may have based their short vowel substitutions for long vowel sounds on articulatory gestures. This is similar to Read's (1971) finding concerning the spelling of short vowel sounds. In this study, the children turned to the instructed short vowels (instead of to the known long vowel names), perhaps because short vowel instruction had moved them away from letter-name intuitions. Either decision pivots on an articulatory base. Thus, although this study finds support for the influence of instruction on spelling, it also lends support to the existence of a developmental spelling progression. REFERENCES Barr, R (1972) The influence of instnictional conditions on word reognition errors. Reading Research Quarterly. 7, 509-529. 253 247 Instruction' s Influence on Spelling Barr, R. (1974-1975). The effect of instruction on pupil reading strategies. Reading Research Quarterly. ID, 556-582. Beers, J., & Haub:son, E. (19m. A study of developing orthographic oyncepts among first graders. Research in the Teaching of English, II, 13:2-148. DeLawter, J. (1970). Oral wading errors of second grade children exposed to two different reading approaches. Dissertation Abstracts International, 31, 3408A. (University Microfilms No. 7101, 097) Elder, R. (1971). OM reading achievement of Sattish and American children. Ekmeluary SchooLlournal. 71 . 216-230. Ferguson, G. (1976). Statistical analysis in psychology and education, New York, NY: McGraw MIL Gentry, J. (1977). A study of the orthographic strategies of beginning readers. Dissertation Abstracts International, 39, 4017A. (University Microfilms No. 79-01, 152) Henderson, E. t 1980). Developmental concepts of word. In E. Henderson & J. Bros (Eds.), Developmental and ovnitive aspects of learning to spell English. A reflection of word knowledge (pp. 1-14). Newark, DE: Lternational Reading Association. Henderson, E. (1981). Learning to read and spell. The child's knowledge of words. DeKalh, IL Northern Illinois University Press. Henderson, E. (1985). Teaching spelling. Boston: Houghton Mifflin. Henderson, E., & Beers, J. (Eds.). (1980). Developmental and cognitive aspects of learning to spell. A reflection of word knowledge. Newark, DE: International Reading Association. Hcnderson, E., & Templeton, S. (1986). A developmental perspective of formal spelling instruction through alphabet, pattem, and meaning. Elementary School Jonrnal. 86. 305-316. Moms, D. (1981). Young children's invented spellins. Illinois Reading Council Journal. 9. 17-22 Read, C. 0971). Pre-school children's knowledge of English phonology. Harvard Educanonal Review. 41. 1-34. Zutell, J. (1980) Thildren's spelling strategics and their cognitive development. In E. Hendersop & I. Beers (Eds.), Developmental and cognitive aspects of learning to spell. A reflection of word knowledge (pp. 52-73). Newark, DE: International Reading Association. 25 4 BEING, REALLY, REALLY CERTAIN -Y617 ICNOW THE MAIN IDEk DOESN'T-MEAN YOU:ipoi Michael Pressley University of Maryland Elizabeth Ghatala University of Houston Jennifer Pirie and Vera E. Woloshyn University of Western Ontario An important assumption in most models of skilled reading is that readers monitor their comprehension of text (e.g., Baker & Brown, 1984), that they are aware of when content is understood and when understanding is less than complete. Such awareness is presumed to play a critical role in regulating comprehension prodesses. Thus, if a reader believes prose is being encoded and interpreted az intended hr an author, there is no reason to modify processing of text. If text is being readqiiicklyt beginning at the first word of every paragraph and proceeding to the end Ouch paragraph, it is likely that rapid beginning-to-ending reading will continue. Alternatively, fer4ngs of miscomprehension can direct rereading of material alrtady coyered or alter ',Aocessing of subsequent content (e.g., cause the reader to read more slowly, ...ad carefully). In short, comprehension monitoring has been conceptualized as a critical executive process in skilled reading, regulating other processes that affect comprehension. Much of the early work on comprehension monitoring was done within the error detection paradigm. Students read text containing inconsistencies or errors (e.g., twq statements in a story about fish, one claiming they live where there is no light and the other that fish select their food by color). If readers noticed such problems, the argument was that they were monitoring their comprehension (e.g., Markman, 1977, 1979)they were detecting that their understanding of one part of a text conflicted with what they understood another part of it to mean. For the example, if the subject coded that fish lived in a completely dark enviromnent, the statement about selecdng food on the basis of color should be surprising and result in a report of text inconsistency. Unfortunately, there were alternative interpretations of failures to report errors 'This research was funded in part by an operating grant to the first author from the Natural Sciences and Engineenng Research Council of Canada and in part by funds provided by the Graduate Research Board of the Umversity of Maryland-College Park. Michael Preasley can be contacted about this paper at EDHD, Benjamin Bldg., University of Maryland, College Park MD 20742. 249 250 Literacy Theory and Researth in text, ones making obvious the inadequacy of the error detection approach as an index nf comprehension monitoring. Three of the possibilieies mentioned by Winograd and Johnston (192) were that failures to report errors could be due to lack of prior knowledge, reflect readers' general belief that printed texts do not contain errors, or result from rationalizations made by reauers to explain away inconsistencies. The interpretive difficulties with error detection stimulated the development of alternative methods for measuring comprehension monitoring. One teet lique developed in our laboratories (Pressley, Ghatala, Woloshyn, & Pirie, in press) has been to ask people main idea questions about text they have just completed reading (e.g., What is the author's purpose in this passage? What would be a good title for the passage?) and to elicit ratings of confidence in their responses. If a reader is monitoring comprehension, then confidence should be high when responses to main idea questions are adequate and low when responses are inadequate. Our most important finding to date, however, is that adult readers usually are moderately confident about the correct- ness of their answers to main idea questions, regardless of the adequacy of their responses (Pressley et al., in press, Experiment 2). Most striking, they are overconfident about poor answers, cw.sing them to bypass opportunities to restudy text (Pressley et al., in press, Experiment 1). In short, adults do not seem to monitor well their comprehension of main points in text, a type of monitoring failure that can undermine executive actions (e g , deciding to reread) that potentially could improve understanding of text. In the study reported here, we reexamined comprehension monitoring using the main idea-question paradigm The particular problem studied here was whether comprehension monitoring (and thus, responses to mthn idea questions) might be more adequate if students were induced to use an exceptionally stringent criterion, one more exacting than their usual standard Thus, students in a high-certainty condition read stories accompanied by main idea questions. They were instructed to continue reading and processing the passage and its accompanying question until they could respond to the question with a high degree of confidence. In contrast, one-reading subjects were asked simply to read the passage one time, to provide an answer to the passage question, and to rate the certainty of their answer. Based on Pressley et al. (in press, Experiment 2), the expectation was that one-reading subjects would rate both their correct and incorre;:t answers approximately equally and about 5 on a 7-point scale (i e confident, although not extremely confident). That is, their confidence followin, one reading was expected to be well below the 7-point ceiling of the scale, and thus, there would be room for the high-certainty instruction to increase confidence. One well established finding in the error detection paradigm is that instructions to shift criter4a affect performance, in general, any information provided to subjects about what constitutes an error increases accuracy in reporting errors consistent with the criteria specified in the instructions (c.g laker, 1985, Elhott-Faust & Pressley, 1986; Markman & Gorin, 1981). Thus, wt Airmised that asking subjects to use a different criterion than the one they normally would adopt (viz., one higher than their usual one) might increase critcal evaluation of their first responses. If so, they might review text addw -nally to det.rmine if their first attempt to summarize the text theme really produced an answer veridical with the meaning in the prose. If this manipulation 25C 251- eing"Certnin Yast *ha the Main Idea was successful in increasing the quality of answers provided to main idea.questions, it would sugbest a simple intervention for improving monitoring of main idea comprohensiOn. Readers cot:,..:1 be encouraged to adopt especially stringent criteria for decidIng they have understood the most important idea in a passage. METHOD Subject:: Forty undergraduates (22 females, IS maies; mean age =19.5 ps;age yiiige =18 to 25 yrs) who were enrolled in a first-year miversity course served tis subje!#s'in the experiment. Subjects were randomly assigned either to the high-cettainty condition or the one-reading condition. Materials Subjects read 10 passages (a different random other for each participant), each between 200 and 500 words in length. These were taken from SAT verbal subtests (e.g., College Entrance Examination Board, 1988) and covered literary, scientific,lind social scientific topics. The following example is typical of the length and difficulty of these readings: As soon as cable service was restored after the earthquake, Baron Okura replied to architect Frank Lloyd Wright's inquiry with a message ofsongandation: HOTEL STANDS UNDAMAGED AS MONUMENT OF YOUR GENIUS. HUNDREDS OF HOMELESS PROVIDED FOR BY PERFECTLY MAINTAINED SERVICE. CONGRATULATIONS, OKURA. Never one to display undue reticence in such matters, Wright .peedily convened a press conference at which he said nothing to dissuade reporters from drawing the inference that the Imperial Hotel was the only building in Tokyo that had remained standing through the disaster. In fact; however, hundreds of other solid masonry buildings in both Tokyo and Yokohama also withstood the quakemost notably those of British architect Josiah Condor, whose numerous structures suffered considerably less damage than Wright's. Nonetheless, the Imperial Hotel's thoroughly undeserved fame as the only building that had stood up through the great Tokyo quake was to prove far more unshakable than the edifice itself; and Wright's renown as the man who had designed and built it flourished accordingly. While by no means wholly responsible for the architectural revolution that was to revitalize the world's cities during the next four decades, the worldwide repute of Wright's Imperial Hotel was to facilitam and hasten its progress. Py the Orne tbis famous edifice was demolished in 167, le great earthquake had been instrumental in altering not only the appearance of Tokyo but also that of many of the other great cities in the world (College Entrance Examination Board, 191,A. p. 55). Each passage was accompanied by a short-answer question tapping the overall theme of the passage. Subjects were asked either the main idea of the passage, its primary purpose, what the author principally wanted to discuss, or for a title summarizing the passage content. Thus, for the example passage subjects were asked to com- plete the sentence, "The primary purpose of the passage is to . . . ." After completing the 10 passages and questions accompanying them, the subjects 252 Literacy Theory and Research took a complete 40-item SAT verbal section. This provided an estimate of individual differtnces in verbal competence. Procedure Before a passage was presented, the subject read the question accompanying it. High-certainty subjects then read the passage under an instruction to continue reading until they could provide an answer to the question with a very high degree of certainty (i.e., they could give an answer they were "very, very sure of." In contrast, onerear"ng subjects were presented the question before reading and were told to react it from beginning to end one time only and to generate an answer based on the single reading. After producing an answer, subjects in both conditions rated their confidence of correctness on a 1 (absolutely sure answer is incorrect) to 7 (absolutely certain answer is correct) scale, with the midpoint of 4 correponding to "50/50 chance the answer is correct " Following the rating, the subject proceeded to the next passage. RESULTS AND DISCUSSION Verbal Ability High-certainty subjects averaged 24.10 items ..orrect (SD = 4.42) out of < on the SAT verbal section, the corresponding figure for one-reading subjects wa., 23.05 (SD = 4 47) These means did not differ significantly, 438)=0.75, p>.50, suggesting approximately equal verbal ability in the two conditions. This was as expected since there was random assignment of participants to the two conditions, that it was so, however, makes less likely that other signi&ant difference% between conditions that occurred could be explained away as artifacts of differences in ability between the two conditions. Reading Time High-certainty subjects spent more time reading the passages than one-reading participants: the total reading time was 27.25 mins. (SD =12.88 mins.) in the highcertainty condition versus 18.83 mins. (SD = 2.72 mins.) in the one-reading condition, 1(38) = 3 03, p< 01 (Kirk, 1982, for this and all subsequent statistical references). Performance on the Main Idea Questions All answers that addressed the question and were consistent with the text were considered correct, with two raters achieving 95% ageement, disagreements were resolved by discussion For instance, for the question about the puspos of the example passage, any answer referring to how the c...rthquake was responsible for shifting world architecture in the ensuing years was accepted as correct. The extra time in the high certainty condition did not translate into significantly better performance on the -lain idea questions. High-certainty subjects averaged 5.70 correct out of 10 compared to 5.00 in the one-reading condition, 438)=1.02, 90(MSE =4 689). That is, even the students instructed to use a high cntenon 253 Being Cert. 'n You Know the Main Idea provided errant respoi.ses more than 40% of the time. The proportions of subjects answering a question correctly did not diff:r significantly for 9 of the 10 passages, greatest X2(1) = 1.62, p>.05 for these nine passage questions. For one passage, more high-certainty subjects (11 of 20) responded appropriately to the question than onereading participants (2 of 20), X2(1)=9.23, p<.01. Certainty Ratings Despite objective performance far below ceiling, high-certainty subjects were very certain of their answers, both when they were correct and whcn they were incorrect. The respective mean ratings were 6.43 (SD =0.22) and 6.24 (SD =0.19) out of 7. In fact, as is obvious in Figure la, no rating for any item by any highcertainty subject was lower than 6. The mean ratings of 5.09 (SD=0.82) and 4 43 (0.78) for correct and incorrect items respectively in the one-reading condition were lower than the corresponding means in the high-ceitahity cmdition, smaller t(38) = 7.06, p<.(191. Although the confidence ratings in the one-reading condition averaged on the high end of the scale, they spanned its entire range, both for correct and incorrect responses (see Figure lb). In both conditions, inspection of the distributions of ratings suggested slightly greater confidence in correct than in incorrect answers. In fact, the mean confidence ratings for correctly answered items were significantly greater than the mean ratings for incor ;et items in both conditions, smaller t(19)=3.03, p<.01. One easily replicated finding in the error detection literature is a relationship between reading ability and error detection (Baker & Brown, 1984) Good readers arc more likely than poor readers to ootice when text contains anomalies and inconsistencies. A parallel relationship was not obtained here. Correlations between verbal SAT performance and awareness of when main idea questions had been answered correctly versus incorrectly (defined as the difference between each subject's confidence ratings for correct versus incorrect items) were low in both conditions, larger Ir; = 20, p'> 20 Moreover, there wcre only nonsignificant .orr.1...ons between verA abiiity and confidence ratings for correct items, larger it = .29, p> .20, and lietween verbal ability and confidence ratings for incorrect items, larger Id = 15, p> 50 These failures to find significant currc'..itions between general ability and awarencss of performance are consistent with corresponding failures in Pressley et al. (in prcss) SUMMARY AND INTERPRETATION Nut getting the main idea of a passage is bad enough For a reader not to know that he or she did not get it is even worse. These two deficiencies sum to overconfi- dence, with this occumng for the full range of reading abilities in the un'versity samp:: studied here. Asking people to be really sure they answered a main idea question had twu negative effects. It slowed responding to the main idea question (i.e., .t increased effort expended) and increased confidence in incorrect interpretations of the main point. This pattern L. similar to one we obtained previously on a learning task in which college students stcdied two sets of sentences that differed in menrrabil g 9- I 2 3 4 5 Centkience Rating Cared Incorrect I 2 4 5 Wine= Rating 3 6 7 Figure I. Distributions of certainty ratings as a function of correctness of response for the (a) high-certainty and (b) one-reading conditions. ity (Hunter-Blanks, Ghats la, Levin, & Pressley, 1988). Subjects detected the memorability difference between sentence sets during study, reported expending more effort on the diffilult sentences than on the easy sentences and predicted they would have equally good or better recall of the difficult compared to the easy sentences. In fact, however, on a subsequent test over the sentences, students recalled many more of the easy than the difficult sentences. In both the Hunter Blanks et al. (1988) and the present study, it is as if subjects can monitor the effort expended on the task but not the memorial or comprehension consequences of that effort. Students in the Hunter-Blanks al. study did lower their assessment of how well they had learned the difficult sentences after they experiedced a test over the sentences, however. In contrast, subjects in the present etmly maintained high confidence that they had comprehended the main idea of the passages even after answering test-like questions ....111.6a1.1.11...M.11111,11.....A. Being Cenain You Know the Main Idea How can this latter finding be explained? Most of the incorrect answers contained elements of the passage, embellished with additional meaning by the maders. For instance, the following were incorrect statements of the primary purpose of the examJ1 of these statements rated as likely to be comet by students who ple passage, vx provided them (i.e., a confidence rating of 6 or 7 was provided for each one): "The historical analysis of how the architecture in Tokyo was established." "Point out the power of the press." "Say that quality can stand the test of time." "Why Wright was given so much attention for the building he built." The readers providing these responses constructed interpretations of the text, ones capturing some theme in the passage, although not the most important one. Thus, the example passage was revealing about how some of the architecture in Tokyo came about, it attested to the power of the press in enhancing a person's reputation, it told how Wright sent forth the message that his workmanship would stand challenging tests and it specified how Wright gained a lot of attention for a building he designed. Nor are these responses atypical of incorrect answers, most of the incorrect responses consisted of themes developed in the passages. One hypothesis suggested by these incorrect answers is that so long as a reader can construct an answer to a main idea question, one that can be defended in light of some of the passage content, he or she is at risk for believing the irterpretation maps well on to the main message in text. In presenting this hypothesis, we recognize that the texts used in this investigation and in our previous studies may be special cases. In all of our work on this problem to date, the passages have been challenging and inconsiderate (Armbruster, 1984) Of course, the main ideas of more considerate texts should be more obvious and thus, more likely to be comprehended. So might the main ideas of texts that are of greater interest to readers or more consistent with their expertise. The more critical question, however, from the perspecive of this investigation, is whether readers are aware of when they miss the point of such texts. A high priority should ben) determine whether the comprehension monitonng problem reported here is a more general one Even if it is not, howe, er, the deficiemy documented here is probably important, for readers dre often ,,onfronted with inwnsiderate texts covering content not related much to what they already know (Armbruster, 1984). Students often are required to extrau the main points from difficult texts. If the only instruction provided to them is to keep working until they are very sure the main idea has been identified, that bit of instrik,tion may kik, more harm than good by slowing reading, yet increasing confidentx in ith.orrea responses. Cntenon shifting alone is probably not enough to improve comprehension monitoring of inconsiderate texts. REFERENCES 9844 The problem of inconsiderate text In G G Duffy. L R Roehler. & .1 Mason tE...wi . Comprd.rnsion instruction Perspelmves and suggesnons (pp 202-217) New York Longman. Baker. L (1985) Pow du we know when we don t understand? Standar& for evaluating text zomprehen Arrnbruster, B B , sion In D L Forrest-Pressley, G E MacKinnon, & T. G Waller (Eds ).Meracognatom cognsnon, and human poformance, ol I. Theurencal perspecmes (pp 155-205) Orlando. FL Academic Press. 2 61 256 Literacy Theory and Research Baker, L., & Brown, A. L. (1984). Metacognitive skills and reading. In P. D. Pearson, M. Kamil, R. Barr, & P. Mosenthal (Eds ), Handbook of reading research (pp. 353-394). New York: Lonvnan. College Entrance Examination Board (1988). 10 SAT Exams (3rd ed.). Prinmton, NJ: College Entrance Examination Board. Elliott-Faust, D. J., & Pressley, M. (1986). How to teach comparison processing to increase children'., short- and long-term listening comprehension monitoring. Journal of Educational Psychology, 78, 27-33. Hunter-Blanks, P., Ghatala, E. S., Levin, J. R., & Pressley, M. (1988). Comparison of monitoring during . study and testing on a sentence learning task. Jmanal of Educational Psychology, 80, 279-283: Kirk, R. E (1982). Experimental design for the behavioral sciences (2nd ed.). Monterey, CA: Brooks/ Cole. Markman, E M. (1977) Realizing that you don't understand. A preliminary investigation. ChildDevelopment, 48, 986-992. Markman, E M. (1979) Realizing that you don't understand. Elementary school childrtn's awareness of inconsistencies. Child Development, 50, 643-655. Markman, E M , & Gorin, L (1981). Children's ability to adjust their standards for evaluating comprehension. Journal of Educational Psychology, 73, 320-325. Pressley, M , Ghatala, E. S., Woloshyn, V., & Piiie, J (in press). Sometinm adults miss the main ioeas and do not realize it Confidence in responses to short-answer and multiple-choice compohension questions. Reading Research Quarterly. Winograd, P , & Johnston, P (1982) Comprehension momtonng and the error detection paradigm. Journal of Reading Behavior, 14, 61-76, 262 DIFFERENCES IN STORY RETELLING BEHAVIORS AND THEIR RELATION TO READING COMPREHENSION IN SECOND GRADERS June E. Barnhart Northern Illinois University Important insights into ways in which literacy is learned and used in everyday activi6.es have come from researchers in emergent literacy. These investigators have focused on the speech and writing of preschool children and include studies that examine young children's developing awareness of oral/written language differences (e.g., Clay, 1979; Holdaway, 1979; Purcell-Gates, 1988; Sulzby, 1986), as well as features of particular forms of discourse, including sense of story (e.g., Applebee, 1977). Researchers in emergent literacy consistently report observations of 4- and 5-year-olds who are not yet reading and writing conventionally, yet display knowledge of the features of written language (e.g., Barnhart, 1986, 1938; Barnhart & Sulzby, 1984; Clay, 1975, Cook-Gumperz & Gumperz, 1981; Ferreiro & Teberosky, 1982; Sulzby, 1985a). In the process of becoming literate, c 'Wren encounter experiences with everyday, contextualized writinb ambling them to make inferences about functions of writing, as well a., relationships between visual symbols and speech sounds (Cochran-Smith, 1984; Harste, Woodward, & Burke, 1984). They also gain experience with another kind of written language where meanings are independent of dm situation and environment in which they occur. Gradually, they come to understand, through decontextualized writing, that meaning is made explicit within the resources r!C language itself, through syntactic and lexical features, as well as the conventions of various forms of discourse. For example, when young children take part in particular forms of discourse, such as reptated encounters with a favorite nursery rhyme or st -ybook, they are able tu abstract a conceptual perspective that steers their expectatkns and interpretations of similar forms in the future. For stories, they develop schemes, or a "sense of stury." According to Appkbee (1977), sense of story begins by 2'4 years of age or earlier, and appears to be well developed when children enter school (Mandler & Johnson, 1977; Stein & Glenn, 1979). The child's sense of story different:ally affects the reception and production of story. Research with beginning readers suggests that a well-developed set of semantic and syntactic expectations play a crucial rOe in successful reading. For example, research by Brown (1977) shows that the extent of a child's sense of story affects compreht....mun and ease in reading and listening to stor;es and affects ability to retell and crea stories. Children must draw on their linguistic resources at the syntactic 257 263 258 Literacy Theory and Research and semantic levels, as well as on their knowledge of story structure and convention, to form a tert appropriate and meaningful for listeners. A consiterable amount of work has focused on preschool children's growth in oral and written language, as well as story structure. The present study was an extension of this work, and examined the developmental nature of the liteeacy process beyond the, preschool yeess in a sample of second-grade children who were heterogendoiii With: regard to lsvel of reading comprehension. The research question whiekgiddeclithe study was: How can we describe story retelling behaviors in seeondliade children, with diverse reading levels, and what is the relationship between patterns of retelling behaviors and level of reading comprehension? METHOD Subjects The subjects were 24 children in a se, and-grade classroom from a public school in a northwestern suburb of Chicago. The children came from a lower-middle to middle SES area, and represented a range of ethnic backgrounds. Based on the results of the Gates-MacGinitie Reading Tests, Level B, Form 1 (Gates & MacGinitie, 1978) administered in October, the classmom teacher used Grade Equivalent Scores in Reading Comprehension to frs-m three grour for reading instructional purposes: (a) Above Grade Group (AGG), (b) Grade Level Group (GLG), and (c) Below Grade Group (BGG). There were 6 children in the AGG (5 girls, 1 boy), 9 children in the GLG (2 girls, 7 boys), and 9 children in the BGG (3 girls, 6 boys). The mean Grade Equiva- lent Reading Comprehension scores for the AGG, GLG, and BGG were 3.66 (range = 3.0-4.5), 1.78 (range = 1.7-2.0, and 1.06 (range =kindergarten-1.4), respectively An analysis of variance showed statistically significant differences among groups in the Comprehension raw scores (F(1, 22)=98.59, p<.0001). Post hoc t-tests showed all three paired comparisons to be statistically significant (p<.0001). The mean age for subjects was 7 years, 8 months (range =6,11 to 9,1) in September. Materials and Procedure Over a 4-week period, the children listened to their teacher read four unfamiliar folktales in a group setting (one reading of each story per week), the fourth story served as the stimulus story. The folktales were chosen from a collection titled, The ig Book of Classic Fairy Stories (Craik, 1987), and incl.ded the following, read in this order: (a) "The Frog-Prince," (b) "The Wolf and the Seven Young Goslings," (c) "The Prince With the Nose," and (d) "The Six Swans." Using techniques described in King (1989), King and kentel (1982), and King, Rentel, Pappas, Pettigrew, and Zthell (1981), each child par1icipat,4 in an individual retelling interview session which was tape recorded Immediately following the teacher's reading of the stimulus story, each child was asked to retell the story to a "naive" listener At the interoew site, the examiner said to each child: "Mrs. (teacher's name) just read you a story, but I wasn't there to hear it. I want you to tell me the whole story from beginning to end " To confirm the selection of the stimulus story as an unfamiliar folktale, at the Stoty Retelling and Reading Comprehension 259 end of each interview subjects were each asked ii they had ever heard the stimulus story before. None of the children responded positively to this question. Analyses All analyses were based on detailed transcriptions for each child, with length (measured in T-units) antl structure of children's retelling; independently scored by two trained raters. Initial agreement between raters was 92.1%, with consensus reached through discussion. The stimulus story co- tained 286 T-units. In a study of 100 folktales, Propp (1968) demonstrated that in traditional folktales there is an invari- ant sequence of 31 functions (or actions and reactions), although in any given tale sonic functions might be omitted. There were 16 of these in the stanulus folktale. Using this scheme, the number, type, and order of functions in children's retellings were analyzed in the present study. RESULTS Across all subjects, there was considerable variation in length and structure of children's retellings. Further, different patterns were observed among tht three groups that paralleled reading comprehens'in levels. Length of Retellings Although the length of retellings across ar Aren ranged from 3 to 98 T-units, there was no overlap ar_ng groups. An °vet. Kttakal-Wallace ANOVA showed statistically significant differences among groups (11(2) = 20.16, p<.0001). Retellings by chilchea in the Above Grade Group we- significantly longer (rd =86.5) than those by children in the Grade Level Group (M=34.0), U(6, 9) = 60, p<.0008) and the Below Grade Group (M = 10.6), U(6, 9) = 48, p<.001). Further, retellings by children in the Grade Level Group were 1_,;nificantly longer than those by children in the Below Grade Group (U(9, 9) = 80, p<.0003). Structure of Retellings There were also significant different-es among groups ir the nmiber of f_actions (H(2) = 20.16, p<.0001). Retellings by children in the Above Grade Group contained significantly more functions (M= 13.60) than those of children in the Grade Level Group (M =6.50), (U(6, 9) = 60, p<.0007) and the Llow Grade Group (M=3.00), (U(6, 9) = 48, p<.001). Frrther, retellings by children in the Gradt Level Group contained significantly more functions than those of children in the Below Grade lift% rent patterns among groups, four Group (U(9, 9) = 72, p<.002). T illus., children have been selected here for descoptive elaboration. Brand:, _ retelling is representative of children in the Above Grade Group, and excerpts are presented in Figure 1. All children in this group showed an awareness that their retellings had to follow a certain order. For example, Brandy (who recaded 13 functions) reold the story to the end of the first ,t.ven functions, then skipred to 260 1. 2. 3. 4. 5. 6. 7. 8. 15. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. Literacy Theory and Research The story's called "The Six Swans." Once upon a time a king was hunting in a great forest. He was lost in the forest. And he wanted to get back home. (pause) And there was this witch. And she said, "Do you want to go home?" And the king said, "Yes." "Can't you show me the way through the woods?"... And then the witch took the king to her, to the witch's house... And the wicked queen turned the six boys into swans. But the girl was up in her room so she was safe. And then she went to look for, and then one day, the next day the king went there to see his dear children. He came joyfully to visit his boys and girls. And ::.he girl told the king all about it. And the king took, told the daughter to go back to the palace with him. But she didn't want to go. And the daughter said, "Let me stay here one more night." And so he let her stay. (pause) And then right at night, at night time, at night time the girl left. 32. 33. 34. 35. 36. 45. 46. 47. 48. 54. 55. 56. 84. 85. 86. 87. 88. Se. 90. 91. 92. 93. 94. And when she couldn't stay anymore. She went to look for her brothers. And she was walking through the woods. And she saw this little house. And she went inside... And then she remembered that they were her brothers. And they could only take this, (pause) their feathers off for about (pause) an hour. And the bzothers said, "We can only shed our skins for a while." "Then we'll tutn back into geese." (pause)... They told her that she could not speak or laugh to anyone. And that she had to, (pause) and that she had to make these six shirts. And the little girl made a firm revol-- a firm resolvul--she made this promise. And just in time, then, and when the six years were over. Then they came. The six swans were flying over. The girl threw the six shirts on them. Then her heart leaped for joy. But one of the shirts didn't have a left arm. And when that happined, there wa,.. one of the brothers who did not have a left arm. But he had a left wing. And the witch got burned, burned to ashes, And that was the end. And they lived, and were very happy forever and ever. F4,411, 1 Excerpts firuin Brandy's retelling of "The Six Swans." (Above Grade Group) function ten She stopped, paused, and began again, this time sequencing the appropriate functions that had been omitted, and finally completing her retelling of the entire story in order Brandy and others in the AGG repeatedly self-corrected while retelling, using frequent pauses to reconstruct the original story. Retellings of children in the Grade Level Group were shorter, and contained fewer functions than those of chi!dreitc Above Grade Group. Figure 2 presents Steve's Story 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. Retelling and Reading Comprehension First there was a (pause) witch (pause). And tbe witch turned the queen's brothers into geese. And what happened was (pause) she found out one day. Then her father came. And was going there to see the, to see something. And acked, "Where are the brothers?" AnQ then said, "The witch turned them into swans." She went under (pause), sent in the swan's house. Let's,see, and she hide, hidded nnder the bed. And they took off their, um, cwans's feathers. (pause) It's called "Six Geese." And they, um, the, um, the swan, the girl came out from under the bed. X2. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. And recognized (pause) them from there. Because they were going, I mean, because they wele hez brothers. And the -text, and then she, um, the king fell in love with the other girl. And, um, and she burned. And she couldn't talk. If she talked it wouldn't work for them. And the witch burned. She ourned into pieces. (pause) And she turned into ashes. And they were all there. They watched. They watched her burn. (pause) That's it. Figure 2. Steve's retelling of "The Six Swans." (Grade Level Group) entire retelling, and is representative of those by children in this group. Steve recalled eight functions, and in the appropriate order. He made several self-corrections, as he tried to accurately recreate the events of the original story in a certain sequenct. Simi la to the AGG, retellings of children in the GLG also contained numerous pauses. Steve whispered, "Let's see." as if he was trying to "sort out" his remembrance of the sequence of the story. Jo the Below Grade Group recalled few of the functions in the stimulus Chi folktale. In addition, none of these children recalled the functinns in the appropriate sequence, and gave no indication of an awareness that their retellings had to follow a certain order. Sandy's retelling is representative of children in this group, and is shown in Figure 3. Her retelling was very brief and contained only 3 functions vA of the 16 in the stimulus folktale. There were no self-corrections among these chilciren, with some scattered instances of pauses. The pausing among these children, however, appeared to serve some global role as an aid for their general memory of the stimulus story, , rather than as a facilitator for organizing this corpus of functions into a sequence that matched the original story. The retellings across groups also differed in terms of oral'written language fea- tures and sense of story. Brandy's retelling contained oral language that was more like the language of print than Steve's or Sandy's. For example, Brandy used dialog, and embedded it within the frame of the story. Further, she consistently made clear who was speaking. Steve also included some dialog, but he omitted explicit reference to the speaker, requiring the listener to infer each speaker's identity (e.g., "And then said . . ."). Sandy included no dialog in her retelling. Brandy showed her knowledge of literary style by using some of the "literary '261 262 Literacy Thwry and Research 1. 2. 3. 4. 5. 6. 7. 8. It was about, um, a girl. And they turned into ducks. (pause) That'03 a, um, goose. (pause) Turned into ducks or gooses. (pausc) And she had turned all three boys into gooses. (pause) And they told her that she hdd to make, um, six things. And she turned then back. (pause) That's all it was about. Figure 3. Sandy's retelling of "The Six Swans." (Below Grade Group) vocabulary" and "literary qntax" found in the stimulus story. Some of the words and phrases that are typically used in writing appeamd in Brandy's retelling (e.g., "He came JOYFULLY to visit his children"; "The king went to see his DEAR children"). Brandy also used some word nrdzrs that are more :ypical of written than oral language (e.g., "Can't you show me the W.tY THROUCH THE WOODS?" "Then her HEART LEAPED FOR JOY"). Brandy's retelling was detailed and accurate, and her sense of story was welldeveloped. Past tense and a formal beginning were evident. She oegan by telling the listener the title of the story, add moved to a traditional stotT-opening line, "Once upon a time ." following the format in the stimulus story. "Story-like" plot structure was present, causal relationships were generally clear, and there was a distinct sense of an ending. In contrast, Steve and Sandy started out by relating an event in the story. It is not until line ll that Steve told the listener the title of the story, and Sandy never provided this information. Further, Brandy's etelling was more explicit than Steve's and Sandy's in several ways. She used more sr...cific nouns, verbs, and . . adjectives to make her retelling clear (e.g., "the WICKED queen"). Her use of reference conventions appropriate to written language can be seen by looking at the self rrections she made (e.g., And then the witch took the king to her.. . . to the witch's house."). Although Steve's retelling showed some of the linguistic means needed to produce decontextualized text, overall his text made greater inferential demands on the listener for understanding than Brandy's. He made extensive use ot pronouns, but lyt often omitted the referential nouns. For example, the listener was required to infer tha, the words "her" meant GIRL, "she" meant QUEEN, and. "them" meant BROTHERS. Steve's retelling also made greater inferential demands on the listenei by being less explicit than Brandy's. With omitted details and words (e.g., "And was going there to see something"), it was more difficult for the listener to cope with the logic of the story Sandy used pronouns in her retelling, but uniformly failed to provide the listener with any nouns needed for their interpretation. Her retelling was considerably more contextualized than Steve's, "iaking it very difficult for the listener to understand the text withow ldditional refere s. Although Brandy's xtethng showed her ability to assemble her language re- sources at the semantic and syntactic levels to reconstruct a text as written narrative, Steve's and Sandy's retellings were global, or "summative." The request from the examiner was "Tell me the whole story from beginning to end," but Steve's retelling sounded as if he had interpreted the request to be "Tell me ABOUT the whole story 268 Story 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. Retelling and Reading Comprehension 263 It's starting out. And the old lady is making some gooses. Some gooses from the boys. I see her! She's doing it now. (child looks at examiner and waves hands like magicians) How do you make gooses fram boys? Oh yuk, that sounds gross! And I don't even think anyone can REALLY do that. Kapowie? Now you're a goose! (child waves hands at examiner, like a magician) Squawk, squawk, squawk, squawk! (child waves arms up and dovn, like flapping wings) I'd like to be a goose. And thel're flying around. And then they're cave totalling houses. Woom, vroom! (child moves arms horizontai, to look like wings on airplane) And they're not afraid of anything. Just like me. (child laughs) Figure 4. Orson's retelling of "The Six Swans." (Below 43-acte Group) from beginning to end." In fact, Sandy began her retelling, "It was about a girl . . ." and ended it "That's all it was about." These examples by Brandy, Srcve, and Sandy are representative of retellings of children in the AGG, GLG, and BGG groups, respece.vely. However, a fourth example presented here is by another child in the Below Grade Group, and demonstrates the diversity that was present across the retellings of these 24 second-grade chileren. The re-..clling by Orson (age '/ years, 3 months) lacked detail and accuracy, and did not follow any logic of the story. His itnse of story was notably immature, and his retelling showed few features of story h...6uage, such as formal beginnir.g, dialog, and past tense. "Story-like" piot structure, causal relationships, and sense of an ending were not present. Perhaps most unique was Ornrr's inability to use the lair guage of the story. His entire retelling was recounted in the present tense, as if events in the story were happening at that very moment. His language sounded as though he was watching a movie and telling vhat he was seeing as he viewed things from moment to moment. Developmentally, the language in Orson's retelling was similar to that described by Sulzby (1985b) as "Following th:. Action" when kindergartners were asked tcr read a favorite storybook. Orson's atypical performance warlats some additional information for his case. Although Orson's grade equivalent ccore in comprehension was L I, he was not enrolled in any special education classes, and English was his primary language. The structure of chilch _a's retellings also differed across groups with regard to the type of folktale elements included. Propp (1968) states that the first seven functions may be regarded as the preparatory section in the folktale, and the folktale used in the present study contained six of these seven functions. Although children in the Above and Grade Level Groups included many of these first six functions in their retellingc, children in the Below Grade Group included few, if any, of these preliminary functions. These children most often began their retellings with Propp's eighth 264 Literacy Theory and Research function, in which the villain in the story family. ses harm or injury to one member of the 5 DISCUSSION Res. -.rch with oral 'wr:tten language differences indicates tit written language is more decontextualized than oral language (Chafe, 1982, Olson, 1977, Rubin, 1978, Tannen, 1982) Retellings by sor..e children in the present study were monologic and noninteractive, composed sithout the need for any paralinguistic cues. In contrast, other children created retellings that contained an oral conversational style that relied on interpersonal involscment between the storyteller and the listener. Chafe (1982) suggests that written language incorporates a richer and more varied vocabulary, and is more syntactically integrated than oral language. Certain children in the present study blended son.e of the literary vocabulary and syntax of the stimulus folktale Into their retellings, reflecting an implicit awareness that certain types of words, phrases, and word orderings are found in caorybooks. Importantly, the di" .cnces observed across the retellings of these second-grade children paralleled their lesel of reading comprehension. Those who were above grade lesel in reading comprehension showed a clear sense of the wntten-narrative register, and had a fairly mature sense of story. Retellings by children who wert reading at grade level showed a great deal of vascillation between oral and written registers, suggesting that this differentiation process continued to play a prominent role for them Finally,, children whose comprehension was belo... grade level demonstrated a rudimentary knowledge of "sense of story" and written language. Retelling stories has been used as an assessment tool in researd....b developmental patterns in story comprehension (e.g., Stein & Glenn, 1979). More recently, Morrow (1990) hus described the benefits of using story retelling as both an assessment tool mnd instrucConal strategy for story comprehension in early childhood. Beyond the preschool yeais, how es er, once children enter the formal instructional setting, comprehension is traditionally measured through a put _r and pencil test in which the child must read passages of text and choose the picture explicating the passage or answer a question about the passage This format allows only a single avenue for assessing comprehension, pros iding only one perspective on a child's comprehension of a story. Simc the child is required and allowed to give only discrete, literal responses, Morrow (1990) cautions that comprehension assessment is defined by the questions asked, and proposes that It should be the child's response that is the focal point of comprehension evaluation. Results from the present study suggest that story retelhng offers a valid source of information concerning the ability to reconstruct meaning by children beyond the preschool years Children enter school and are engulfed by formal reading instruction, howeser, the data reported here suggest that their knowledge and expectations about written language are still being explored, clarified, confirmed, and disconfirmed. These results show that, even among second-grade chindien who have been exposed to basal reading instruction, their underlying concepts about wntten language can be characterized as mistable dnd not yet-conventional. Asking children to retell stories 27t) Story Retelling and Reading Comprehension 265 offers us a means of gaining insight into the ongoing process of literacy development well into the primary grades. REFERENCES Applebee, A. N. (1977). A sense of story. Theory Into Practice. 16. 342-347. Bamhart, J. E. (1986). Written 'language concepts and cognitive development in kindergarten children. (Doctoral dissertation, Northwestern University). Dissertation Abstracss International. 47. 2096A. (University Microfilms No. 10396) Barnhart, J. E. (1988). The relationship between graphic forms :Ind the child's underlying conceptualization of writing. In J. E. Readence & R. S. Baldwin (Eds.), Dialogs in literacy research (pp. 297-306). Chicago: National Reading Conference, Barnhart, J. E., & Sulzby, E. 0984, December). Children's concepts of written language in emergent reading and writing. Paper presented at the National Reading Conference, Tampa, FL. Brown, G. H. (1977). Development of story in children's reading and writing. Theory Into Practice, 16, 357-362. Chafe, W. L. (1982). Integration and involvement in speaking, writing, and oral literature. In D Tannen (Ed.), Spoken and written language. Exploring ara.k, and literacy (pp. 35-53). Norwood, NJ. Ablex. Clay, M. (1975). What didI write? Auckland, New Zealand. Cook-Gumpetz, J., & Gumperz, J. (1981). From oral to written culture. The transition. In M. F. Whiteman (Ed.), Writing. The nature. developmem. and teaciung of written communication (pp. 89-109) Hillsdale, NJ: Erlbaum. Cochran-Smith, M. (1984). The making of a reader. Norwood, NJ: Ablex. Craik, D. M. M. (Ed.). (1987). The lug book of classic fairytales. New York. Portland House. Ferreiro, E., & Teberosky. A. (1982). Literacy before schooling. Exeter, NH. Heinemann New York. Teachers College Gates, A. I., & MacGinitie, W. H. (1978). Gates MacGinitie Reading Tests Press. Harste, J., Woodward, V. & Burke, C. (1984). Language stories and literacy lessons. Portsmouth, NH. Heinemann. Holdaway, D. (1979). The foundations of literacy. Auckland, New Zealand. Heinemann King, M. L (1989). Speech to wnting Children's growth in writing potential. In J. Mason (Ed ). Reading and writing connections (pp. 7-30). Boston: Allyn & Bacon. King, M., & Rentel, V. (Eds.) (1982). Transition to writing(Report No. 240-603) Columbus. OH. Ohio State University. (ERIC Document Reproduction Service No. E.1 210 752) King, M., Rentel. V., Pappas, C., Pettigrew. B., & &tell, J. (1981). How children write. A longitudinal study (Report No. 213-050) Columbus, OH. Ohio State University. (ERIC Document Reproduction Service No. ED 213 050) Mandler, 1 M., & Johnson, N. S. (1977). Remembrance of things parsed. Story structure and .ecall Cognitive Psychology, 9. 111-151. Marry..., L. M. k 990). Assessing hzldren s understanding of story through :heir ...onstrut.ton and recon strucnon of narrative. In L. M. Morrow & J. K. Smith (Eds.). Assessment for instruction in early childhood (pp. 110-134). Englewood Cliffs, NJ: Prentice Hall. Olson, D. R. (1977). From utte......,e to text. The bias of language in speech and w.iting. Harvard Educe, tional Review, 47. 257-281. Propp, V. (1968). Morphology of the jolktale. (L. Scott, Trans.). Austin. University .)f texas Press 11in-ell-Gates. V. (1988). linical and .syntat.tic knowledge of written narrative held by well read to. kinder gartners and second graders. Research in the Teaching of English. 22. 128-160. ra! and written language (Tech Rubin, A. D. (1 978). A theoretical taxonomy of the differences &Av. Rep. No. 35). Champaign, IL. University of Illinois, Center for the &tidy of Reading. in elementary school children Stem, N. L., & Glenn, C. G. (1979). An analysis of story con.,. In R. 0. Freedle (Ed.), New directions in discourse processing. Advances in discourse processing (Vol. 2, pp. 53-12(). Norwood, NJ: Ablex. _ : 266, Liviracy' 11**.and Sulzby, E. (1985a). Kindergartners as writers and readers. In M. Farr (Ed.), Advances in writing resew* Chlldren's early )viltIng devekprnini (Vol. 1, pp. 127-190): Norwood,.KI:Ablez. SWAY, EL (1985b): Children's emergent reading of favorite storybooks:A developmental study. Reading. Research Quarterly. 20, 458-481. Sulzby, B. (1986). Writing Ind reading: Signs of oral and written language ocgsnizationlit the Young child: In W. H. Teak & E. 'Sulzby (Eds.), Emergent huracy: Writing and reading 49).1loci;a:64i. NJ: Ablest. Tannen, D. (Ed.). (1982). ,Spoken and written language: Exploring orality and lltensey.,Norwoal; Ablez. ASSESSING CHILDREN'S INFERENCING STRATEGMS Susan B. Neuman Temple University It is widely accepted that the ability to draw inferences is critical for reading comprehension (Anderson & Pearson, 1984; Spiro, 1980). At the most general level, inferencing is a constructive thinking process, requiring the reader to elaborate upon the explicit information presented in a text. A large number of studies have demonstrated the integral role of infercncing it. the comprehension of and memory for text (Bransford & Johnson, 1972; Goetz, 1977; Kintsch, 1986). For many readers, however, inferential comprehension is more difficult than other comprehension processes (Hansen, 1981). Three explanations have been hypothesized to account for these difficulties. The first is that differences in prior knowledge may Influence children's ability to make inferences. Pearson, Hansen and Gordon (1979), for example, found that children with greater prior knowledge on a topic were able to draw more inferences than those with weakly developed schemata Thus, deficienClef in pnor knowledge may account for limited inferencing ;n certain situations Research by Paris and his colleagues offer a second explanation They suggest that young children tend not to apply their inferential strategies thoroughly, unless 1976; Paris & Upton, 1976). For specifically directed to do so (Paris & Lindau example, 7-year-old ...hildren failed to ...omprehei. nfcrences spontaneously and could not use Indirect cues to access memory. Neve! theless, when these childfen were directly encouraged to dramatize the sentences, they could use both implicit and explicit cues equally well. These results sugge r that wh:le developmentally capable, young children may not naturall, engage in strategies to "go beyond the text Evidence from Aassroom Instructional practices suggests a third alternative Studies report that students are not typically asked inferential questions in reoding (Hansen & Pearson, 1983). Further, teachers tend to teach their good and poor readers differently (Allington, 1983) resulting in poor readers receiving even less instruction in inferential thinking than good readers. Most of these studies, however, have examined inferencing at the point of retrieval or when an investigator imposc .. a task upon readers demanding such reasoning These types of nferemes may not be made routinely during the ongoing comprehension process. Further, as Frederiksen (1975) and Kintsch (1974) argue, inferencing may occur di 1: point when incoming dam is encoded into memory This suggests that studies measunng inferences at retrieval only may unden,:imate those that are made during the comprehension process itself. The present stuck, Jesigned to measure children's inferences, differs from th se previously cited in several specific features. First, it examined ;nfeicncing strategies 267 2 7t3 268 Literacy Theory and Research using a verbal recall technique as children are comprehending text. Second, unambiguous texts (two short mystery stories) were used. It was reasoned that well-constructed mystery stories might enhance ecological validity naturally encouraging children to predict and infer from text without direct probing. Thial, good and poor readen, were selected to analyze if differences occur in inferencing strategies. Fourth, the inferences strategies examined emerged from the subjects' reading of these texts, rather than a predetermined set of categories. With these considerations in mind, this study was designed to address the following questions: (a) Do young readers generate inferences as they read stories? (b) Which types of inferencing strategies do readers make during comprehension? (c) Do low- and high-proficiency students employ similar inferencing strategies? (d) Are there differences in children's ability to successfully apply inferencing strategies? Access to these comprehension strategies among adults and older children havt been obtained in rn_qiy cases through variations of verbal reporting techniques. Collins, Brown and Larkin (1980), f^r example, elicited verbal reports of skilled adult readers' thinking processes as they interpreted text. This technique, however, has not been regarded as most appropriate for young .or poor readers who may be less able to introspect about their cognitive knowledge (Brown, 1980). Introducing a modification of verbal reporting, Phillips (1988) used a limited-probe-when-necessary technique, where clarification questions were used after students read brief episodes of text. This approach helped to increase the completeness of reporting as well as to minimize the interval between processing and retrospection considered to be essential in obtaining reports of cognitive activity. Further, her appnich combined aspects of retelling and verbal reporting Students were first given opportunities to tell all they wished about a particular episode without probing, then, if or when necessary, clarification questions were asked Norris (in press), :7 a validation study, found that these verbal reports did not alter subjects' comprehension processes or performance. Consequently, a similar approach was adopted in this study. METHOD Subjects The subjects were 42 fifth-grade students from 11 classrooms in an urban school district in the Boston metropolitan area The sample, primarily from blue-collar families, was ethnicallyi. erse, including 70% Caucasian, 15% Asian, and 15% black. All students spoke English as their first language. None were identified as learning disabled High achieving students (N= 21), th_ .110 scored above the 85th percentile on the Metropolitan Achievement Test (Prescott, Balow, Hogan, & Farr, 1978) (M = 9f_' 81, SD= 4 37), and low achieving students (N= 21), those who scored below the 50th percentile (M = 32.67, SD = 8.85), were selected from each classroom. Materials Two stcries were used from the Bloodhound Gang mystery series (Children's Television Workshop, 1983): "The One-Ton Jewel," about a "white dwarf," a jewel 274.. 269 Assessing Inferencing Strategies supposedly from outer space ttu.... was to be auctioned for a great deal of money, and "The Blob," a story of a stolen ice sculpture. The stories were well-structured, involving female and male characters appropriate to tbe children's age and interest level. "The Blob" included 931 words, and "The One-Ton Jewel," 1,294 words. Both were written at the fourth-grade reading level according to the Fry Readability Formula. Stories were each divided into six episodes; each ended with the introduction of a new clue related to solving the case. Colored pieces of paper were in.r4ted booklets containing each story to indicate ends of episodes. General promptingq*dons were used if or when a clarification of children's inferences was required:Tor exaa,ple: ,a) Did you find any clues in your reading? (b) What do you thinfcsidll happen next? Why do you think so? (c) Does this give you any ideas? Procedure Students met individually with the researcher or a graduate assistant in reading and language in a private room for one session of approximately 30-50 minutes. Using a sample protocol, the researcher described the vetbal reporting procedures, emphasizing the open-endedness of the activity and assuring them that no corrections or grades would be given for responses. Students were then asked to read two stories, episode by episode. Story selection for both groups was counterbalanced. Students were told to request the pronunciation of any unfamiliar word, and to read at their own pace. After each episode, the re searcher asked each student what came to mind while reading the story, using the clarification questions only when the student was not clear or when he or she appeared to be hesitant to make inferences. Students were asked to verbally report on six episodes in each story for a total of 12 times. These sessions were audiotaped and transcribed verbatim. Data Analysis Verbal reports from each story were combined to form a protocol for each student. Each protocol was divided into tdea units, defined as a proposition containing at least one relational concept and one argument. Two judges examined idea units in a sample of 10 protocols to determine whether each represented a recall idea unit, one that was stated directly from the story, or an inference level idea unit, one that might be suggested but not stated in the text. Percent of agreement between judges was 98% The average frequency of recall and inference idea units was 36% and 64%, respectively. Protocols were then examined by three judges to determine which inferencing strategies children used in comprehending the stories. The term strategy was defined here as a plan or technique used by readers for interpreting materials. Similar to Phillips research (1988), it was consider-d independent of the correctness of the actual interpretation. Froni extensive discussions and alialysis of the protocols, a typology was developed following Trabnoes basic distinction of two types of inferences, text-based and slot-filling k1980). L ext-based inferences the individual finds semantic or togical 275 270 Literacy Theory and Research relations between propositions expressed in a story. In slot-filling inferences.the mdividual fills in missing information to make connections between events discussed in a text. A third category, referring to miscellaneous strategics including reitcreng and refraining from inferencing was also added. It was also clear that students, on occasion, attempted a particular inference strategy, but misconsamed information. For example, in ttying to bind togetherdiffT ent propositions, the child might draw an incorrect conclusion. Therefore,!-* trout-, categorizing protocols according to inference strategies attempted -and ftequ5n4 their use, an error rate was obtained, indicating the ratio of implaitsibloAriferencei.over attempts. Two judges then independently coded 10 protocols; interrater reliability on,the identification of strategies used was 85%. After establishing reliability of the coding system, each inference idea unit in all protocols was classified by strategy. RESULTS AND DISCUSSION The first set of analyses addressed inferencing strategizz :tile= use while reading stones, and analyzed whether there are differencen betwmn strategy use on the basis of children's reading proficiency levels. The second set of analys.% examined differences in the successful application of these strategies and cpuditatively analyzed the types of errors thu occurred when reading. Types and Frequency of Inferencing Strategy Eight inference strategies were used by the subjects. Three types appeared to he text-based. The first, binding, referred to an attempt to drawn conclusions on the basis of a number of statevl facts. ite logic followed something like, "if 'x' were there, and 'y' we e. there, theo they both must have been involved in the crime together." The second type, rebinding, similar to the strategy defined by Collins, Brown and Larkin (1980) and Phillips (1988), appeared when new information apparently led to a conflict in the ---Aent's understanding of the story. Here, the reader was forced.to either adjust new information to fit the past interpretation, or readjust their previous understanding with the new data. For example, following her decision that Smiling Jimmy stole the Blob, the student says, "Well, wait I think there probably never was a Blot " The third strategy, confirming, occurred when a new fact was used to explain a prior interpretation. This type of infereming appeared to model the procPss of instantiating slots within a selected schema (Andemon & Narson, 1984) as students attempted to provide a coherent overall representatiol of the "tory. For example, a student confirming that the rock or star was fake, added "because Vickie said that she could lift it up with one hand." Three types of slot-tilling inferences were identifed. Assigning default values occurred in the absence of specifically substantiating mformation in the text. In :his strategy, students constructed hypotheses about events of the story based on their background information, and/or their knowledge of story structure. For example, one student assurivd that lie "white dwarf," which ve.s a dead star from outer spas:, - ++ n - - "-e 271 Assessing Inferencing Strategies was actually a "dead rock and roll singer" from the band -White Dwarf." The second type, empathizing, involved a personal response. Here, children seemed to emotionally place themselves in the story, attributLig feelings to characters on the basis of their own beliefs and responses. For example, after reporting that Vickie thinks the "White Dwarf" is a fake, the student said, "Everybody's gonna be surprised when she picks it up." Proposing solutions, the third slot-filling strategy, referred to attempts to invent new solutions not related to information Nesented thus far in the text. For example, recalling that it was probably Smiling Jimmy who stole the Blob, the student proposed that he might "just dispose of the Frozen Blob in the river." Children used two other strategies. The first was simply reiterating a previously made inference without adding any new explanation or interpretation, such as "Yel, I think he did it." The final strategy inciAed refraining from responding, by sar:ng "I'm not sure," or "I don't know." lough reflecting a lack of knowledge, this strategy appeared at times to express children's .olerance for ambiguity or ability to remain open to multiple interpretations. The average frequency of shategies ust..4 and standard deviations are reported in Table 1. Clearly the slot-filling strategy of assigning default values was employed most often, accounting for approximately half of all reported inferences, Other strategies used frequently were text-based, incli.ding binding story elements together and confirming prior interpretations with aew information. Perhap_ lue to the task, there was little evidence of rebinding or empathizing with characters or character actions in e.ther group. Reiterating and refraining strategies were used with relative frequency, indicating a lack of knowledce or an unwillingness for various reasons to draw inferences. To examine differences in strategy i ,es among high and low readers, a multivari- Tai-.!e 1 Means and Standard Deviations for Frequency of Strategy Use for Good and Poor Readers Poor Readers Good Readers SD M 18.55 .95 10.75 1.99 6.98 12.2j 4.45 13.00 1.55 14.23 47.80 15.12 2.28 3.28 50.41 .95 12.93 1.45 Stra.egy Text-based Inferences Binding Rebinding Confirming Slot-filling Inferences Assuming Defaults Empathizing Proposing New Solutions Other Reiteratir3 Refraining 3.65 5.60 9.75 3.49 9.31 2 77 SD 2.11 6.00 4.77 L59 4.93 6.95 S.14 5.08 9.06 ate analysis of variance was performed with the frequencies of the uses of the eight strategies as dependent variables. No significant differences were reported between groups (F(1, 40)= Wilks Lambda, 1.25, p<.30). These results suggest that similar strategies appear to be used by good and poor readers when constructing meaning from text. Examining Children's Errors Though employing similar strategies, an analysis of mean frequencies of errors indicated striking differences in their successful awLeafion between good and pcor retders. 'With an average ef over 18 errors per protocol, poor readers clearly misconstrued information more fit, uently than good readers. A one-way analysis of variznee indicated that these differences were statistically significant (F(1, 40)=25.53, p<.0c1). These errors were qualitatively analyzed to examine the nature of these difficulties Three categories of difficulties appeared to account for students' incorrect responses: (a) overreliance on background knowledge. a reliance on intuition or prior knowledge of an idea or character trait in the face of conflicting textual informatk ; (b) overreliance on chon-term memory. a focus on decoding specific facts or words in a story while ignoring relations or meanings among these facts; and (c) inability to impose order on text poorly organized incoming textual information led to erroneous conclusions. Interrater reliability, established e,r error categorization, was .89. Once reliability was established, judges independently toded 554 inferencing errors. Table 2 describes number and percentage of errors by category for good and poor readers. Overreliance on background information to the detriment of co- idering all textual infonnation appeared to be the most common source of error. In qualitative terms, it also represented the most serious kind of distortion, often guided by schema contrary to the story's actual events and intended meaning. Given the poor match between schema selection and textual information, students had difficulty slotting incoming information, tending to rely on short-term memory rather than on forming a consistent interpretation. With inefficient stlema and wain- Table 2 Number and Percent of Errors by Category fur Good ana Poor Readers Error Category Overreliance on Background Knowledge Overreliarwe on Recal: Inability to Impose Order on Text Total Errors Number of Inferences Good Readers Number % Poor Readers Number % 64 43 7 5 41 148 4 113 11 16 406 1,065 39 934 156 137 15 13 273 Assessing Inferendng Strategies sistency among incoming facts, relationships among the parts and the whole were rather arbitrary and insufficient. These results, together with the previous analysis, suggest that while similar strategies appear to be employed by good and poor readers, poor readers are more apt to ignore the text in favor of their own intuitive responses, often rather arbitraiily recalling facts without evidence of consolidating them into a consistent and satisfying interpretation. CONCLUSIONS In contrast to previous assumptions about students' difficulties in maldng inferences, this stv.dy found that chile-en frequently engaged in a number of inferencing strategies. Inferencing occurred e.aing the encoding process, as children were interPre fing incoming data. This finding suggests that studies analyzing inferencing during retrieval alone may be seriously underestimating the frequency of infprences used by children in comprehending stories. Good and poor readers appeared to use a similar repertoire of inferencing strate- gies. This finding supports and extends research by Oakan, Wiener, and Cromer (1971), and Olshavsky (1976-1977), who found that strategies employed and frequency of their uses in comprehending materials die not significantly differ ankag high and low proficiency groups. However, poor readers clearly appeared to accept unconventional inwpretations of stories. For example, never questioning her assnmption, one student suggested that the meaning of putting a robber away or "in" ice as she remembered, meant that the robber was going to be "placed in an ice bucket." Sin:e ..the did not appear to comprehend the story to begin with, she was not able to detect when the meaning of the sentence had, iadeed, become anomalous. Quahtative analyses of errors suggests that, in conm t tc !ack of prior knowledge, many poor readers adopted inefficient schema, allowing them to accurately slot only a portion of text. When the text did not conform with their existing interpretation, it would tend to be either overlooked or "rewritten." Therefore, it was not lack of prior knowledge, as much as the wrong prior layow!^dge that students' brought to the text. This view supports Nichokon and Im lach's finding (1981) that children's prior knowledge often competes for priority in children's infereocing, with innsitive knowi edge at times interfering with the complex process of constructing meaning from text. In summary, rather than strategy training, teachers may well be advised to emphasize a number of direct instructional activities which help students focus on textual materials. Discussion of the topic to be read subsequent to comprehension might Ix one of the more efficient ways of enhancing children's understandh4 of stories. For example, as little as 10 minutes of general discussion prior to reading appears to si, nificantly affect children's comprehension (Neuman, 1988). Further, techniques that encourage children to attend to text and jvstify their responses on the basis of text information are important. For those readers with low expectations of print in partici.!ar, instruction ant' practice may have direct consequences on their inferencing performatice. REFERENCES Allington, R L (1983) The reading instruction provided readers of differing ability. Ekmentary School Journal, 83, 548-559. Anderson, R. C., & Pearson, P. D. (1984). A schema-theoretic view of basic processes in reading comprehension In P D. Pearson (Ed.), Handbook of reading research (pp. 255-291). New York: Longman. Bransford, J D., & Johnson, M. K. (1972). Contextual prerequisites for understanding. Some'nvestigations of comprehension and recall. Journal of Verbal Learning and Verbal Behavior, II, 717-726. Brown, A. L. (1980). Metacognitive development and reading. In R. J. Spiro, B. C. Bruce, & W. F. Brewer (Eds.), Theoretical iesues in reading comprehension (pp. 453-481). Hillsdale, NJ: Erlbaum. Children's Television Workshop (1983). The Bloodhound Gang Series. New York. Children's Television Workshop. Collins, A., Brown, J. S., & Larkin, K. M. (1980). Inference in text understanding. In R. J. Spiro, B. C. Bruce, & W F Brewer (Eds.), Theoretical issues in reading comprehensice. tpp. 385-407). Hillsdale, NJ: Eribaum. Frederiksen, C. H. (1975). Effects of content-induced processing operations o semantic information acquired from discourse. Cognitive Psychology, 5. 139-166. Goetz, E. T (1977), Inferences in the comprehension of and memory for text (Tech. Rep. No. 49). Champaign: University of Illinois, Center for the Study of Reading. Hansen, J (1981) The effects of inference training and practice on young children's reading comprehension. Reading Research Quarterly, 3, 1981. Hansen, I , & Karson, P. D. (1983). An instructional study. Improving the inferential comprehension of good and poor fourti.!-grade readers. Journal of Educational Psychology. 75, 821-829. Kintsch, W. (1974). The representation of meaning in memory. Hillsdale, NJ: Erlbaum. Kintsch, W (1986) On modeling comprehension. In S. de Castel!, A. Luke, & K. Egan (Eds.), Literacy, society, and schooling, (pp. 175-195). New York: Cambridge University Press. Neuman, S B (1988) Enhancing children's comprehension through previewing. In J. E. Readence & R S Baldwin (Eds.), Dialogues in literacy research, (pp. 219-224). Chicago: National Reading Conference. Nicholson, T , & Imlach, K (1981). Where do their answers come from? A study of the inferences which children make when answering questions about narrative stories. Journal of Reattng Behavior, 13, 111-129. Norris, S (in press) Effect of eliciting verbal reports of thinking on critical thinking test performance. Journal of Educational Measw-ement. Oakan. R , Wiener, M., & Cromer, W. (1971). Identification, organization, and reading comprehension for good and poor readers. Journal of Educational Psychology, 62, 71-78. Olshavsky E (1976-77) Reading as problem solving. An investigation of strategies. Reading Research Quarterly, 12, 654-675. Paris, S G , & Lindauer, B K. (1976). The role of inference in children's comprehension and memory for sentenses. Cognitive Psychology, 8, 217-227. Paris, S G . & Upton, L R (1976) Children's memory for inferential relationships in prose. Child Development. 47, 660-668. Nam- D , Hansen, I & '.,rdon, C. (1979). The effect of backgroand knowledge on young children's comprehension of explicit nc.1 implicit information. Journal o) Reading Behavior. 11, 201-210. L M (1988) Young readers' inference strategies in reading comprehension. Cognition and 1,struction, 5, 193-222. Prescott G , Balow, I , Hogan, T , & Farr, R. (1978). Metropolitan achievement tests. New York: The Psychological Corporation. Spiro, R J (1980) Constructive processes in prose recall. In R. J. Spiro, B. C. Bruce, & W. F. Brewer iEds ), Theoretical issues in reading comprehension (pp. 245-278). Hillsdale, NJ. Erlbaum. Trabasso, T (1920) On the making of inferences during reading and their assessment (Tech. Rep. No. 157). Urbana: University of Illinois, Center for the Study of Reading. 269 T:ZVELOPING LOW-PERFORMING, FOURTH-GRADE, INNERCITY STUDENTS' ABILITY TO COMPREHEND NARRATIVE' <4..j James H. Mosenthal Michigan State University For the past 15 years, the reading education community has been concerned with the nature of comprehension and the forms of comprehension instruction Building bridges from the former to the latter, from theory to practice, was provoked early on by Durkin's critique of classroom comprehension instruction and basal teathers' manuals (1978-79, 1981). There has been a large response to this challenge resulting in specific, theory-based strategies for providing "relativPly direct or explicit instruction in comprehension. especially for the low-achieving student (Brown, Palincstu., & Amibruster, 1984, p. 255). Representative of these strategies are Palincsar and Brown's reciprocal teaching strategy (' 986), Au's Experience-Text-Relationship strategy (1979), and Raphael's Question-A uswer-Relationship strategy (1982). All of these strategies follow principles of direct comprehension instruction as discussed by such researchers as Collins and Smith (1982) and Pearson and Gallagher (1983). But there is a second bridge that has been harder to buildthe one from tested comprehension practices to incorporation in ongoing classroom instruction. Wend ler, Samuels, and Moore (1989) found in their observations of teachers that the amount and quality of comprehension instruction was no different than that observed by Durkin (1978-79). They conclude that "teachers may be confused about the difference between assessment of comprehension and direct instruction of comprehension" and that "it is possible that either teachers do not know about these fecommendations because they are so new, or that teachers would like to use direct instruction but do not km w how to do it" (Wend ler, Samuels, & Moore, 1C89, p. 396). It is also possible that tha conditions of classroom iastruction mitigate against incorporation of direct comprehension instruction into ongcnug classroom practice Barr and Dreeben's work (1983) on the relationship between elasc-oom conditions novel or multiple forms of instruction in9osed and learning point to the constraints by these conditions. These constraints are especially restrictive when: (a) there is an extreme 1ange of ability within a classrocm, and a relatively large number of lowachieving students, tb) there is a lack of materials of various ".evels of difficulty within a grade level, (c) class size is large, and (d) the teacher has little experience (See Barr, 1982.) Mosenthal's (1987) attempt to replicate the effects of reciprocal teaching The author gratefully acknowledges the assistance of Mindy Boyer in the data analysis ar.d preparation cf the manuscript. .0 276 Literacy Theory and Research in the inner city, where these conditions were present, seems to bare out the difficulty of incorporating novel instruction, given restrictive classroom conditions. The r,urpose of this paper is to describe an instructional prcject worlemg with low-achieving fourth graders in a single classroom of an inner-city school. The signifrzance of the description is that it depicts the complexity of incorporating principles of direct comprehension instruction into ongoing classroom instruction. We attempt, to show how classroom conditions and principles of direct comprehension instruction interact in the development of feasible instruction that results in student growth. In the paper we address three issues: (a) res, inding to the conditions of clasSiootn teaching and learning, (b) providing explicit ..omprehension instruction, and (c) documenting growth. This work was carried out over a 12-week period as part of a longterm staff development project. CHANGING THE CONDITIONS OF INSTRUCTION The classroom was part of an elementary school of a public housing project in a large city The school district prohibited grouping for reading instruction and required that teachers teach all students from the grade level basal text. Classrooms in the school, on the average, had over 30 students and .esented a broad range of ability with a high proportion of low-achieving students. . students in the elementary grades worked from the grade level basal. In setting up the conditions for instruction the staff developer and the teacher took s'.-ps to change or respond to three conditions: (a) the large number of low ability students, (b) the nature of the materials, and (c) the exclusive use of whole-class instruction in a classroom of 32 students. Instead of grouping the whole class for instruction, and thus increase the likelihood of managerial and disciplinary problems, the decision was made to group the lowest achieving students in the classroom for a regular, 2-day comprehension lesson. The grouped students were the 7 lowest achieving fourth gtaders, as chosen by their teacher. Their average age at the end of third grade was 9.33. Their ITI3S Reading kchievement scores at the end of third grade averaged 2.43. Individual ages and scores are presented in Table 1. To match the reading ability of the students with materials of an appropriate difficulty, the staff developer brought in a set of second-grade basal readers (Eller & Hester, 1984a and b). The stories discussed in the lessons were taken from these reade-s The only criterion for selection was the teacher's and staff developer's intuition about the quality and interest of the stories. Most important in the instruction was the expertise and experience of the classroom teacher. The teacher had 12 years of teaching experience, though only in her second year in the city and school where the staff development took place. She 'Iad a strong rapport v. ith her students, and had excellent management skills and understanding of comprehension processes. In her previous practice she was used to grouping students for instruction She collaborated in the development of the comprehension lesson. 282 e: 277 Comprehending Narrative Table 1 Story Retelling Data The Tricky Troll Age ITBS R* C** 0*** 9 8 9 2.3 3.0 4 unk 5 10 6 '; 10 10 2.9 2.5 2.2 2.4 .00 .13 .22 .30 .00 0 0 unk .00 .75 .34 .58 .00 .62 S 2 3 Mil SD 9.3 1.7 .43 .39 .30 .17 .07 .13 .11 Sing Sack Sing 0 C R .75 .74 .64 1 - 0 0 0 0 .14 .38 .67 .88 .62 .72 .10 .43 .45 .45 - .29 .27 .23 .35 .10 1 2 1 0 0 1 .83 .75 R Benjie C .50 .76 .79 .54 .63 .55 .73 .64 .12 .42 .38 .62 .30 .48 .24 .38 .40 .12 0 1 1 2 2 1 2 2 1.57 .54 *R =Relevance or number of causally relevant clauses in retelling/total number of clauses in retelling. *C= completeness or number of causally relevant clauses in retelling/clauses in ideal retelling. ***0 = ()Income or parts of outcome includvi in retelling. DIRECT COMPREHENSION INSTRUCTION The comprehension lesson was designed to increase students' comprehension of narrative text and to be feasible in the classroom. The activity was carried out regularly over time (12 weeks). In addition, the activity provided the same kind of comprehen. sion instructicn that might be given to students performing at grade le I (see Brown, Palincs^r, Sc Armbruster, 1984). In other words, it focused on vading, writing, and discussion of text as structured by strategies developed by the teacher and staff developer. The use of the strategies followed principles tff direct comprehension instruction. The purpose of the two-part lesson and the strategies were explained and modeled for the. students. Over time the teacher guided the students in the use of the strategies. The students were engaged in a 2-day reading/writing comprehension lesson conducted once a week. During the first day of a lesson the teacher and students read following a modified Directed Reading-Thinking Activity or and discussed . DI TA (after ... ...Ler, 1969). At the end of the di-cussion the students wrote a retelling of the story. No assistance was given in the writing of the retelling. This was done in order that the students Individually attempt to work out the completeness or sensibleness of the story already constructed in the DR-TA. In other words, the DR-TA discussion worked out the completeness of a story as a group activity. As a result, all information relevant to a retelling was elicited in the discussion immediately prior to the written retelling. Before the second part of the comprehensiun lesson the tea, her and staff developer read the student.; retellings. The retellings were informally evaluated on whether they included relevant beginning, middle, and endi ag content. Based c i this evaluation the teacher and staff developer decided upon what introductory comments should be made at the beginning of the second part of the lessor to focus the group activity of writing 278 Literacy Theory and Research a retelling. For example, when the first retellings were brief and unconnected the students were encouraged to write more. When the retellings tended to Ye verbatim renderings of the story, breaking off abruptly when the student needed to finish, the students were encouraged to think of the retelling as a kind of sunpary which included information about the beginning, middle, and ending of a story. On the second day the teacher and students constructed their own written retelling of the story discussed the day before. The teacher followed a plan for generating a group written retelling developed by her and the staff developer. This included providing a "retelling-starter" such as, "This is a story about a girl who disobeyed her mother." Students were then prompted to discuss what information should come next in the retelling. Students wrote their own ideas of what should come next based on the group discussion. This continued through 3 to 5 cycles of asking "What should come next?" until the resolution of the story was reached and the students completed their group-aided retelling. The rationale of the 2-day lesson was to confront the students' own versson and concept of a retelling written on the first day of the lesson with a model retelling they helped construct on the second day. This awareness of a gap between their work as individuals aid their work as a group functioned as a stimulus to growth it, the discussion and writing of subsequent lessons. In terms of development, the intent of the lessons was to develop the students' ability to comprehend text as a whole as reflected in the completeness of their written retellings. The decision to use written retellings was made because of what it reveals about the students' comprehension. Retellings, at a very basic level, ask students to construct or represent their understanding of the whole of a text they have read. In one sense, the retelling reflects the extent to which th student has been able to construct a causal chain of the events of a story (see Trabasso, 1981). A causal chain can be likened to a target structure or coherent representation of events in a story that is a basic goal of reading narrative (see Collins, Brown, an..1 Larkin, 1980). This coherence or understanding is achieved through inferencing strategies that enable the reader to progressively refine the coherence of the causal representation of events. A retelling reflects a reader's causal understanding of a story as a whole arid the presence or lack of sophisticated inferencing strategies. Similar arguments IA oral retellings by young readers have been made by Gambrel], Pfeiffer, and Wilson (1985), and Morrow (1985). Another reason for using written retellings is that they are written. Writing has become a partner to reading in teaching fcr understandit.,-.. What wnting contributes that reading ane 4;scussion don't is the student's text- -a concrete manifvstation of the student's ur rstanding. Writing a retelling, like writing generally, requires that the student fix his or her understanding of a story in the form of a textin this case a retelling This is a different type of understanding than that reflected in a response to a question about a part of a story. It is a concrete reflection of what has been constructed about the story as a coherent, ca,asal whole. As such, the written retelltng forces a more exact and articulate understanding than required in oral comprehen5,on tasks. The written retelling, used to assess comprehension, is liability to the extent that it is influenced by the students' writing fluency. However, when treated as a 264 279 Comprehendiag Narrative comprehension activity or task, in other words as the object of instruction, the written retelling engages cognitive processing of te a in a manner and a degree which group discussion and oral retellings do not. Our argument, is that this processing of text, to the extent that it is individual and requires "fixing" one's understanding of the whole of a story, can be a main contributor to students' learning to comprehend as well as learning what it means to comprehend narrative: STUDENTS' GROWTH As noted above, the teacher and staff developer informally evaluated the nature of the students' unaided written retellings on a weeldy basis. As the students demonstrated a capacity to write art unaided written retelling which included relevant beginning: middle, 4ald ending content, this focus of the comprehension lesson was stopped. This occurred after the 12th lesson. A story summary and a student's unaided written retelling for the 3rd, 7th, and 10th lessons are given in Appendices A, B, and C. These lessons were based on stories titled The Farmer and the Troll, Sing Sack Sing, and Benjie, respectively. As is appar-ot from the contrast in the quality of the retellings, this student comes to understand the demands of the retelling task and is able to write a retelling including rdevant beginning, middle, and cnding content. His growth 1- representative of the growth of the other students. To provide a more formal analysis of change in the ^,uality of the students' retellings, an analysis of the causal cohesiveness of the retellings was carried out. The Analysis of the Students' Retellings In the formal analysis the students' unaided written retellings, the stories read during the 3rd, 7th, and 10th lessons wet.. analyzed. For each story discussed and retold, the staff developer and a research assistant wrote an ideal rettlling. It was ideal to the extent that only information necessary to reconstruct a causal chain for the story was included. These ideal retellings were develoP al based on an analysis of their causal relations following the method developed by frabasso and his colleagues (Trabasso, Secco, & van den Broek, 1984; Trabasso & Sperry, 1985; Trabasso, van den Froek & Suh, 1989). Two measures are cited as important by Trabasso and his colleagues; recall of information on the causal chain, and recall of important information within the causal chain. The causal chain is the chain of clauses which zre necessary to trace the develonment of the story from beginning to end. The ideal retellings are reconstructions of the causal chain of events for a story. Recall of important stor, information within the causal chain is content whose causal importance is high, relative to other content on the chain. These tr-,asures were adapted in the present study. Students' unaided written retelling: were co, causal cL parsed clausally. Clauses which matched clauses in the ideal retelling see how much of the students' retellings included information on the story, we calculated the proportion of the students' clauses which 285 280 Literacy Theory and Research could be matched to the ideal retelhng. This measure can be thought of as an indicator of the extent to which a student only includes causally relevant information. We will refer to this as the Relevance variable. As the amount of causally relevant information increases the value on this measure approaches 1.00. A score of 1.00 indicates that every clause in the student's retelling matched a different clause in the ideal retelling. We also calculated the proportion of the clauses in the ideal retelling which were matched in the students' retellings This measure indicates the extent to which, over .; time, more of the information v.: the causal chain is included by the students. We will refer to this as the Completeness variable. As the students include more causally important information values on this measure approach 1.00. A score of 1.00 indicates that a student's retelling matched the ideal welling. A third measure was created to get at the extent to which the student recalled the most causally important information in the ideal retelling. Because the ideal retellings were long (the average was 51 clauses per story) and complex when compared to the artificially constructed stories used by Trabasso and his colleagues, it turned out to be difficult to isolate a few, most in.;ertant clauses. Typically, sequences of clauses describing an action or a line of argument functioned as a single unit. In addition, many clauses shared the distinction of having the highest causal value in the causal chain represented by the ideal retelling. As a revalt, it was rarely that all of the causally most important clauses were included in the students' retellings. Therefore, it was decided that the most important information for the students to recal was information abo.it the outcome and resolution of the story. From the informal evaluations of the retellings it was evident that the students had the most difficulty including relevant ending information in their written retellings. To determine whether relevant information having to do with the outcome of the story was included, the following measure was developed Each story follows a general pattern. Each story first develops a problem or goal of thc protagonist (the farmer can't get rid of a troll who causes trouble, Rosita is kidnapped, Benjie is shy and his grandmother has lost her favorite earring) Then the story develops the action the protagonist takes to achieve the goaL'overcome the problem (the farmer poses a deceitful bargain; Rosita sings a song attributable to her, Benjie fights his shyness and tries to find the earring). Finally, the story develops a resolution (the troll accepts the bargain and never knows better; Rosita is freed by a friend and they arrange to have the kidnapper run out of town; Benjie finds the earring by overcomizg his shyness). Retellings were scored as to whether they included a complete depiction of the story outcome. We will refer to this variable as the Outcome variable. In scoring this measure, a "0" was given if the outcome was not explicitly mentioned A "1" was given if the outcome was given but was incomplete. For example, in The Tricky Troll, if a student said the troll accepted the bargain but did not say the troll was tricked in the process a scc e of "1" was given. In Sing Sack Sing, if a student said Rosita was freed by a friend but did not say that the kidnapper was run out of town, a score of "1" vas given. And in Benjie, if the student said Benjie found the earring but did nct say that he lost his shyness, then a score of "r was given If both parts of the outcomes described above were given a score of "2" w is assigned. 281 Comprehending Narrative RESULTS AND DISCUSSION Table presents descriptive data on *Jr, students' written retellinos. On the Relevance variable, which measures the proportion of the students' clauses that matched clauses in the ideal retelling, group mean sc.7,4; increased from .39 for The Tricky for Sing Sack Sing and Benjte, respectively. On the Completeness Troll, to .72 and variable, which measuns the proportion 3f clauses in the ideal retelling included in the students' retellings, scores increased from .13 to .35 and .40 from the 3rd to the 7th to the 10th story. Perhaps because of the small sample size, thesr -ifferences were not significantly differen1 in a one-way ANOVA. However, given tue teacher's and staff developer's informal evaluation of the retellings, the increase in the Relevance and Completeness variables appears to reflect the change in students' retellings observed during instruct.on. These seems also indicate that students' mtellings stabilized by the middle of the staff development (when Sing Sack Sing was read). In other words, with respect to the Relevance and Completeness variables, it is as if all that the students learned about retelling stories occurred by the seventh lesson. This conclusion, though, runs counter to the perceptions of the teacher and staff developer who continued the instruction for another 5 weeks. the Outcome variable, which measures the extent to which students Scores refold the outcome of a story, provide some support for the teacher's and :taff developer's perception that there was room for improvement in the students' retellings, even after the seventh lesson. Treated as a continuous variable, nit= scores on the Outcome variable increased from .14 to .83 and 1.57. These differences are statisti- cally significant (F(2, 17) = 11.24, p<.01). Using Tukey's HSD t I (Kirk. 1968), the difference between the mearr, for Benjie (Lesson 10) and The Tricky Troll (Lesson 3) is significant, p<.05. The mean scores for Benjie and The Tricky Troll are not significantly different from the score for Sing Sack Sing (Lesson 7) Thus, the difference indicates, with respect to tho inclusion of outcome information, that students did not understand the importance of this information in a retelling at the time of the seventh lesson. This means that students were still learning to understand what it meat_i to retell a story even after other retelling skills had stauilizet ' esson 7 when Sing Sack Sing was read. Impliit in the above argument is the notica that the improved retellings were the result of the instruction. It could be argued tha. .he retellings became more causally cohesive simply as a result of practice. It is impossible to untangle to wliat extent the students improvement .s due to the nature of the comprehension lessons or to the repetitive practice in writing retellings. The scores on the Outcome variable suggest that the students' growth was a function of practice and the content of th 'essons since, as noted, these scores did not stabilize by the middle of training as did the other measures. -in ether words, the iata from the Outcome variable suggests that the improved retelling were not simply a function of practice but of learning a conception of a retelling that included the zdequate depiction of the outcome of a story. The outcome measure provides some evidence that the lessons themselves played tant part in dire..ting students' thinling about what to include in a retelling. g8 ';xt 4_, -wor- This argument for growth over the weeks of the training is exemplified in the outcome information provided in the written retellings of Student 7 and given in Appendices A, B, and C. In his written retelling of The Farmer and the Troll, at the beginning of training, Student 7 does not state that the troll -accepts the bargain presented by the farmer, and, necessarily, does not state that in accepting die ttatgain; he is unaware of the trick that has been played on him. Thus for the outcome nieaSurei Student 7 was given a score of "0." In his retelling of Sing Sack Sing,,iin,the 7t1i. itieek of training, Student 7 does state that Rosita is rescued but does not mention that.* kidnapper was run out cf town thus eliminating the threat of danger to kokta.igefe, Student 7 was given a score of "1." :.ft his retelling of Benjie, in the 10flyyreek Of, training, Student 7 states that Benjie finds the earring and overcomes his shyness in doing so. For this retelling, Student 7 was given the max... nn sr .e of "2." Moreover, the set of causal measures suggest that as a of training and practice students develop a more sophisticated conception of the whole of a story. This conception is defined in terms of the representation of a storyls causal cohesion in a retelling. Over time, students included mostly relevant information in their retellings, more information on the causal chain of the story, and a more complete statement of the outcome of the story. It is important to note that this improvement occarthd in the context of regular, valid comprehension instruction, a condition of reading instruc- tion for low achieving students argued for by Brown, Palinesar, and Armbruster (1984). CONCLUSIONS In the previous section we have discussed changes attributable to the nature of the coniprehension instruction provided. In analyzing students' written retellings we Im e argued that, as a result of the instruction and practice, the students' comprehension improved. We have argued that the nature of the lesson and the quality of the retellings point to an understanding of story comprehension that is causal in nature. However, as important, if not more so, are the positive effects on students' learning provoked by restructuring the norms of classroom instruction. As discussed previously, these changed norms included grouping low ability students for direct comprehension instruction and using materials of appropriate difficulty. Evidence for the influence of the changed norms comes from an interview with 'e teacher at the cclusion of training. In that interview, the teacher discussed several advantages to the reading/writing lesson that are related to these concerns: I know at times in the beginning that they [the students in the reading/writing group] were elated that they were a part of a small group. I think the stories helped. They were stories they could read and they could enjoy. We combined silent, oral, and choral reading . . . . Most of the children are used to the round robin [reading] and I think they began te see that we don't always have to read orally. We can gain a lot of information from silent reading . . . . They like[d] the discussion. [At first] they weren't too particular about writing . . . . The minute you tell them to pick up a pencil . . . they become a little apprehensive and they're not crazy about it. But . . . d..), did enjoy [it] and I don't think it was a begrudging task [writing the 288 - 1-4 rr 910-r-q:epeitoAk "reitke retellings]. BasicallylMeirattitude toward thewriting changed] becaurse of the praise they received On Me ciithpletion of their writingtask [on Me first:day ofthe leSsim]. I iaw i-eriat.thiprovethent in their Writing skills, whichis soinething that we tea* don't do mr.dilen:becatise we just da not have the tithe. They began to develoP a very special sense about ... .théy kneWthe.organizatiOn of the prograin. They kneW tlxy wexe goinito cothe_backhOennd re* sidilif4uSS The next tithe everYbddy canie back with,pene*a&tpitiers and:. they write. So_they picked up on it Vely. well I thought ibat %Try-Oger to get tilek here and'disenss and.write,What Was tiling!, 7,7,bediiii;e_thei saw animproveinent in their Writing. Theisaiv impiOvethent and I Mink theifelt better aboth what they were doing. In these comments the teacher makes reference to a number of Oonditioha that had positive consequences that go beyond the effects of -the nature. Of tile fraining,, that is, the improved retellings. It is as if the consequences of changinithe conditions: of learning in the classroom enabled or iMeracted with the nature of the tra:Ining th stimulating student growth. Of great importance in the teacher's remarks are the effects of haiing.apprOPriate materials"stories they could read." Matched to the ability of the .studintS,-,:the stories set the students up for academic success and the perception Of glair oirti success. This basic alteration of classroom norms was necessary for the suceess of any kind of direct comprehension Instruction. As noted by the teacher, the proiiition of appropriate materials enabled the practice of such-basic reading activity as silent reading, and it enabled the students' full participation in discussion of the content of a story. Whereas the students' experience with reading was enhanzed by providing appro=: priate materials, the students' experience with wrieng was eahaneed by its mete inclusion as a regular activity connected to reading, and by its appropriatenesi given the abilities of the students. At the beginning of tinkling the expectationsJor the writing task were general. The teacher and staff developer looked for the inclusion nf beginning, middle, and ending content. The specific form these expectations toOk on evolved from what the students wrote in their retellings. No tangential exPeOtatiotia, such as a concern for transcription skills, intruded on the general expectations for the retellings and the articulation of those expectations in terms of what the student wrote. The effect on the students, according to the teacher was positive, again because of the students' perception of their own improvement. Perhaps because of the appropriateness of the materials and the tasks, the students experienced an enjoyment in doing well at academic tasks. This enjoyment was complemented by the praise/reinforcement of the teacher. But this enjoyment w& not immediate. It was a function of the regularity and predictability of the work and the students' growing perception that they could do the work and do it well. As the teacher states, "They knew the organization of the program." All of the effects described above help to explain why the students were elated at being part of a small group. It cannot be argued that simply being in a group caused the elation, though this certainly helped form an identity they hicl not had before. Their identity as a gaup ultimately was a function of their perception of the leedimacy of the work and their success at it. And this identity as learners is at the root Of theiipleasure. As the teachel concludes later in the interiew, "I was gratified that tlia chalren enjoyed it so much. They enjoyed corning into the group." REFERENCES ! Zr; Au, K. H. (1979). Using the experience-text-relatiomhip method with minority children. The' Reading: Teacher, 32, 677-679. Barr, R. (1982). Clusroom reading instruction from a sociological perspective. Journal Of _ ior, 14, 375-389. Barr, R., & Dreeben, R. ;1983). How =hods work. Chicago: University of Chkago Brown, A., Palincsar, A., & Armbruster, B. (1984). Instructing comprehenon-foStering'arliviiiiisII: interactive learning situations. In H. Mandl, N. Stein, . Trabasso (Eds.), Learning and comprehpi;. sion of ten (pp. 255-286). Itillsdale, NJ: Eribaum. , Collins, A., Brown, 1., & Larkin, K. (1980). Infer-axe in text understanding. In R., Spiro,,B, Bruce, -- . & W Brewer (Eds.), Theoretical issues in reading comprehension (pp. 385410)- 1* Erlbaum. Collins, A., & Surith, E. (1982). Teaching the process of reading comprehension."InD. K. Denernith R. 3 Sternberg (Eds.), How and how much can intelligence be increased (pp. 173-185): NorWoOd;- , NJ:Ablex. Durkin, D. (1978-79). What classroom observations reveal about readine comprehension inttruction; Reading Research Quarterly, 14, 481-533. --Durkin, D (1981) Reading comprehension instruction in five basal reader series. Reading kesearch Onarterly, 16, 515-544. Eller, W., & Hester, K. (1984). Tricky troll. The Laidlaw Reading Program, Level 7. River Forest, Laidlaw. Eller, W , & Hester, K (1984). Wide-eyed detectives. The Laidlaw Reading Program, Level 8. River Forest, IL: Laidlaw. Gambrel, L , Pfeiffer, W., & Wilson, R. (1985). The effects of retelling upon reading comprehension ann. recall of text information. Journal of Edacadonal Research, 78, 26-220. Kirk, R. E (1968). Experirnerual design. Procedures for the behavioral sciences. Belmont, CA: Brooks/ Cole. Morrow, L (1985) Retelling stories: A strategy for improving young children's comprehension, concept of story structure, and oral language complexity. The Elementary School Journal, 85, 647-6452. Mosenthal, J (1987). Learning from discussion: Requirements and constraints on classroom instruCOn in, reading comprehension strategies. In I. Readence & S. Baldwin (Eds.), Research in literacy: Meribte_ perspectives (pp. 169-.176). Rochester, NY- National Reading Conference. Palinmar, A S,, & Brown, A. L. (1986). Interactive teaching to promote independent learning from text. The Reading Teacher, 39. 770-776. Pearson, P. D., & Gallagher, M. C. (1983). The instniction of reading comprehension. Contemporary educational psychology, 8, 317-344. Raphael, T E. (1982). Question-answering strategies for children. Reading Teacher, 36. 186-190. Stauffe., R. (1969). Directing reading maturity as a cognitive process. New York: Harm and Row. Trabasso, T (1981). On the making of inferences during reading and their assessment. In 3. Guthrie (Ed.), Comprehension and teaching. Research reviews (pp. 56-76). Newark, DE: International Reading Association. Trabasso, T , & Sperry, L. (1985). Causal relatedness and irnprance of story tvents. Joarnal of Memory and Language, 24, 595-611. Trabasso, T , van den Brock, P., & Sub, S. (1989). Logical necessity and transitivity of causal relations in stories. Discourse Processes, 12, 1-25 Trabasso, T., Secco, T., & van den Brock, P. (1984). Causal cohesion and story coherence. In H. Mandl, N Stein, & T Trabasso (Eds.), Learning and comprehension of text (pp. 83-112). Hillsdaln, NI; Erlbaum. 29 v.: - _ Comprehending Narrative 285 Wend ler, D., Samuels, S. Jay, & Moore, V. K. (1989). The comprehension instruction of award-winning teachers, teachers with master's degrees, and other teaches. Reading Research Quarterly, 24, 382-401. APPENDIX A Story Summary of The Farmer and the Troll: A fanner and his wife live on their farm. One day things start to go wrong on the farm,,,Eggs are broken, the milk spoils, and so forth. The farmer and his wife figure out that a bad luck troll has come to live on their farm. They cannot get him to leave. In the spring the farmer, while getting ready to plant potatoes, discovers the troll underground. The troll says he, !Wes underground. The farmer, realizing he needs his field to plant potatoes, makes it bargain i'vith the troll. This year the troll can have what grows above ground and the farmer gets what grows below. Tne next year the farmer gets what grows above the ground and the troll below, and so on over the years. The troll agrees, not knowing he has been tricked by the farmer. From then on, each year the farmer planted a crop that grew its food under the ground or above depending on the bargain he had struck with the troll. Student 7's Unaided Written Retelling for The Farmer and the TrollWeek 3: 1. The farmer is tricking the troll to fall for his plane 2. so he can trape him in the hole 3. so that he cannot give any more troible 4. I think he's going to cacth him 5. so he Won't him any more bab Luke any more APPENDIX B Story Summary of Sing Sack Sing Rosita's mother tells her never to go to the river or she may be kidnapped, as others have. On a hot day, Rosita disobeys her mother and goes to the river. She is kidnapped by a strange man, put in a sack, and told to sing when ordered to or be beaten with a stick. Rosita sings a song recognizable by people who know her. A friend does recognize Rosita by her song and frees her. They put rocks and mud in the sack. The man is run out of town when he cannot make the sack sing. Student 7's Unaided Written Retelling for Sing Sack SingWeek 7: I. Roseitia disbeyed her mother 2. when She told her not to go to the river 3. And then She went to the river 4. and she Left her earings on the bank of the Lake 5. A man came around 6. got her earing off the bank of the village 7. he had sack 8. and Put her in it 9. and took her away from the bank of the there 10. .if!She didn't Sing Sack sigh U. 41 beatliou 12. for Saler ekririgwaS caugth 13;.4ir'd then.tyai tatight 14- her andter fond and come save her form 15- :t114Tu.t-Mu.4 Ocks.'111 the sack 16. aad as stion.as *dr* he say sack 17. or illbeat you:with Stick Benjie, a shy little boy, lives with his Granny in the city. Every Sunday, the two' of theta)* to church, and on the way home stop at the bakery for u small treat. One.Sunday, hoine, Granny diseovered that she has lost one of *earrings. Theysearchbtit'CiOpt,fin4; it liteirjie knows'hOw iinportant the earrings :dell?: GrannY.!He decides tOlook:440.fpk,06-4, earring and goes back to the bakery.- Overeoming his shYirtess,Benjie_eiplaina.f.tO lady the haportance of the earring. Together they find it. At home, Granny is hippy gird she doesn't have such a shy grandson anymore. Student 7's Unaided Written Retelling for BenjieWeek 10: I. This is a story about a boy 2. who was Shy Very shy to talk 3. every sunday they Walk to church Benjie and granny 4. Shy wore her very speical earrings to church 5. after church They went to the bakery 6. And when granny got home her earing west Lost 7. and then she told benjui he coild go outsid 8. bit benjin about tell his mother 9. he was goinig to tell her 10. he was aboit to go find her earinge 11. Went bake to the bakey to find her earing in the bakey 12. and the bakey lady told berijie to go Play 13. he tried to tell her that ganny eating wast ost 14. he went in the dark room to find granny 15. then he foid it in in the garbage can 16. then thke bakey said she will call the police on him 17. and then explode to her about etxy 18. and the he went home 19. and he had gave her the carings 20. and she said we lost the old bingie 21. she was 'mod of him 22. becaise he was not Shy any more 292 d %MY LEXICAL COHESION IN COMPREHENSION;AND COMPOSITION: A SYNTHESISsOlF,RESEAKVisgV0 Richard B. Speaker, Jr., john G. Barnitz, and Joan P. Gipe University of New Orleans Research in a variety of disciplines has explored the role of cohesion,(HallidaY & Hasan, 1976; Hasan, 1984) in reading comprehension and composition,.(Chapinan, 1987). Although such research has demonstrated the-function of 'colie4iVe;st2,**-5 such as pronoun reference and connectives in the Comprehension and, welRernid: ir,cess', of short texts, comparatively little research has examined:the role of le4:01',04s09. . on text production and comprehension or investigated the,role olle).441:400.slonger cohesive chain.; (see Chapman, 1987). Some studies.are now. enierging,Onfr the role of lexical cohesion and other cohesive devices in children's Writing (co:', Stinn4.-. !ran, & Sulzby, 1990; Cox & Tin7mAnn, 1938; McLin, 1987; Pappas, 1985); hóweVet, few researchers have examined children's sensitivity to and use of chains inVOIVing a 4.' ;'-- .,_.: ,., t'. , e- ,- r i--,-,, variety of lexical and other cohesive tokens embedded in naturally occuthat text or the relationship rf those chains to other text factors such as text pragmatics and se:rantic hierarchies (see Friedman & Sulzby, 1987). Meanwhile, a large body of research on vocabulary learning has emerged-Which can be related to lexical cohesion theory and research (cf, Elley, 1989; Freebody& Anderson, 1983; GiN, 1978-1979; Herman, Anderson, Pearson; & Nagy, 1987;. Jenkins, Matlock, & Slocum, 1989; McKeown, 1985; Nagy, Anderson, kHerinan, 1980). Vocabulary research should be particularly interesting 0 researchers who'inyeetigate cohesion for three reasons: first, historically, vocabulary knowledge 40:- kOn reported as highly orrelated with reading ability (e.g., Davis, 1944; Terinan, '1918); second, since the semantic structure of a text is encoded in its collection of cehesiVechains, salient vocabulary items are the impxtant lexica/ tokens in those chains; and th:al, the study of lexical tokens in cohesive chains should provide testable theories for vocabulary acquisition, composition, and comprehension research. The purposes of this paper are to shed light on the roles of lexical cohesion in comprehension and composition and to provide directions for future research unifying cohesion research and vocabulary acquisition research within the franiework of nn integrated wholistic literacy paradigm. This will lead to a better understanding of the effects of complex disc, use stectures in literacy learning and vocabulary acquisition. The paper will present the following issues: first; a shift in' the lexidal Corkionent of the cohesion model; second, the focus on reading/writing/speaking/listening connecdolls; and third, the nature of vocabulary acquisition from context. This paper is not intended to be a thorough research review; rather, it will highlight representative research studies relevant to the three issues while focusing on lexical cohesion; only 2-87 4.:-;:,..,... . 29 3, . - --;;-; ' , , Ii,'11 : 4.I .' 4:41 .......,....2,.._-.:-,`,.:,' :34 ..li ...--i t, Literacy Theory and Rescaeh one part of the larger cohesion system. (For discussion of instructional issues, see Baumann, 1986; Chapman, 1983; Irwin, 1986.) ISSUE 1. A SHIFT IN THE LEXICAL COMPONENT OF THE COHESION MODEL Halliday and Hasan (1976) produced the most influential cohesion model for English. Their cohesion taxonomy considered reference, substitution, conjtmstion, ellipsis and lexical cohesion as its major subsystems. The lexical cohesion subsyster consists of the vocabular; devices used in the text to cue a unified semantic.network to the reader. Readers derive unity by applying their knowledge of word Meanings which are connected by the text's semantic structure in chains through the. text; in essence, the cohesive chains of a text are sequences of words in the text which mark semantic relationships for the reader. Halliday and Hasan's lexical cohesion taxonomy consisted of r !iteration (repetition, synonymy, superordination, awl gener.1 word) and collocation. (See Halliday & Hasan, 1976, for details and examples). Stotsky-(1983), noting theoretical difficulties with this model, proposed a reorganization of lexical cohesion into two categories: seman ically related words and collocationally related words. Hasan (1984) also produced a major revision in the lexical component of the model; her analysis of lexical cohesion consisted of two categories: general (repetition, synonymy, antonymy, hyponymy, meronomy) and instantial (equivalence, naming, semblance). Of particular interest in Hasan's revision is the development of the concept of cohesive harmony and her extension of the analyses of the types of cohesive chains in text; cohesive harmony is a measure of the level of semantic chain intrAction in an extended text which accounts for the number of items (tokens) in the chains which are central to the text's structure and meaning, and therefore, directly related to a text's coherence (for details, see Hasan, 1984, or Cox, 1987). Hasan's revision of cohesion theory allows researchers: (a) to move from more simple quantitative measures of text cohesion such as the counting of ties to more qualitative analyses of different semantic chains containing lexical tokens, and (b) to perform a more complete analysis of cohesion as it extends through an entire text than was provided with the earlier rubric. Under the earlier rubric for lexical cohesion (Halliday & Hasan, 1976), some studies examined the full range of cohesive structures (see Chapman, 1987, for a review) Lexical cohesion was found to be more difficult to comprehend than reference, substitution and ellipsis (Moberly, 1978), but Monson (1982) found lexical structures easier for 7-year-old children than the other structures. Lexical cohesion, particularly the length of lexically cohesive chains, is closely related to the quality of a writer's work (Eiler, 1979; Neunet, 1987). Good writers produce essays with a higher density of cohesive chains per t-unit (Cherry & Cooper, 1980; Witte & Faigiey, 1981). Good adolescent readers also write essays with more cohesive properties, including lexical cohesion (McLin, 1987). Moreover, Stotsky (1986) found that not only did high rated essays by adolescent writers contaht more lexical chains than did -nv rated essays, they included longer semantic units 41 the creation of these chains. Under Hasan's rubric of lexical cohesion (Hasan, 1984), research has shifted 294 Lexical Cohesion 289 away from the study of short referential cohesive ties towards iaterest in longer, naturally occuning texts where lexical tokens appear in extended chains (see Pappas, 1985; Pappas & Brown, 1987). Cohesive variables alone do not influence text production and comprehension in isolation from other aspects of textuality (see DeBeaugrande & Dressler, 1981; Morgan & Sellner, 1980); indeed, the linguistic, cognitive and affecfive abilifies of the reader/writer play important parts in comPiehension of cohesive structures (for discussion, see Barnitz, 1986; Ruddell & Speaker; :1985)..For example, cohesive devices such as pronouns are not neceSsarilyslifficule forChildrea to comprehend in cohesive texts, especially when children apply aPpropriate 1rnoWl;; edge of content and discourse (Goodman & Gespass, 1983). Cox -and:Titkii!:!.04:, (1988) found differences in general cohesion knowledge among good and poor readers and writers. Pappas (1985) demonstrated that the structures of cohesive chainstheir co-referential relationshipsand the interacfioos of chains rather than the quantity of cohesive ties (see Mosenthal & Tierney, 1983) were important in developing a model of oral and written language performance. Because of this focus on cohesion in oral and written language, the nature of reading/wairmg/speaking/listening connections has become important in cohesion research. ISSUE 2: READING/WRITING/SPEAKING/LISTENING CONNECTIONS Current research and practice support the interrelationship of reading, writing, speaking and listening in the acquisition of literacy. The recent paradigm shift in literacy them is toward a more integrated wholistic them of literacy acquisition (Goodman, 1986). Research by Cox and others (e.g., Cox, 1987; Cox, Shanahan, & Sulzby, 1990; (--% & Sulzby, 1984; Cox & Tinzmann, 1987, 1988; Nuener, 1987) has demonstrateu a close relationship between cohesion variables and reading and writing performance at various ages. Learners' use of lexical cohesive chains in writing is related to reading ability; better readers use more and longer lexical cohesive chains in their writing (McLin, 1987). The cohesive harmony of the writing of good readers is higher than that of poor readers across grades and text genres (Cox, Shanahan, & Sulzby , 1990). These investigations and their theoretical constructs have produced a host of new terms and concepts related to longer chains which should be considered such as disto...re, coherer, precurser, intervxtion, and network (Neuner, 1987), cohesive harmony (Hasan, 1984) and cohesive density (Pappas, 1985). DeStefano and Kantor (1988) examined the cohesion variables in oral and written discourse of several ethnolinguistic populations (inner city black, Appalachian, and mainstream cultures). They compared children's reading material (basals and children's literature) to the oral patterns of cohesion in mother-child dialogues, finding many similarities between the mother-chilo dialogues among the ethnolinguistic groups and children's literature but a vast difference in basal materials. Ir particular, ellipsis and reiteration lexical structures in dialogues and children's literature were very similar, but reiteration was much more common in basal materials while ellipsis was rare. It should be noted that Tannen (1985), when considering the decontextualized nature of written texts, pointed out that a writer must .1exicalize suprasegmental, prosodic and. gestural information ,which would 'be present in, oral communication, Making lexieal -chains , an important, factor to ;consider ,in emminations of Oral and written diScOurSe. Furthermore, with ',the applieation!Of 'miles: ion a4lysitk*Sitiated discorriSe, resdarcheri have a tool Air identifying ,the, chains Of diSeOrtiSestrroUnding texts used in , the , sehoOls; ,thus, context* whieh; go beYondlhe te,xt,;e*!fe::*i4Yic4 1.Pd tiescr$60 in de*,#gler t4an iPft undefined, *Oiging aq.4;00,:010.#3`r* tcs:9l for naturalistic reSearch, in home :and, elasSioein, SettingS Or Or, eitperiMental ,reSeareb where the nature, of-paincular donteits is being n' inifeatigited:or:,rininipulat*; Th suggests tlie need- fiirreseaA into the nature of lexical tOkenS in chains ivhiehextend: through text and oral discourse. ISSUE 3: NATURE OF VOCABULARY ACQUISITION FROM CONTEXT dies of vocabulary and text comprehension suggest the importance,oflcxicàl cohesion to vocabulary acquisition. Vocabulary items can form the salient:tokens' in 7-- lexical chains or they can be trivially contextualized in a text, or, air:SinStruCtiOoal. procedure. Freebody and Anderson (1983) found vocabulary to be, the:Maio-4;j** influencing performance on compreheusion meaSures rather than nOnlexicat **On, measures. They also fOund both lexical difficulty, and topic familiarity ivelefaignificandy related to comprehension. Herman, Anderson, Pearion and NagY (1987)lotind . that students gained more vocabulary knowledge from well elaborated texts. These and other studies (e.g., Elley, 1989; Gipe, 1978-1979; Jenkins et al 1989; Nagy, Anderson, & Herman, 1986) suggest that context is crucial to vocabnlarY, learning and text comprehension. However, researchers have not yet reported adireet effect of vocabulary and lexical cohesive structures on both micrestrucMreandrnactO; structure comprehension, probably, because the newer rubrics (Hasan,,19,84;.Stets 1983), cohesive analyses of context structure, and the concept of cohesiveharrifonY have yet to be incorporated into research studies on vocabulary acquisition and t4c4ing. Furthermore, little attention has been,paid to the analysis ofcontexts surrorindin& vocabulary in text or in instruction which van be provided by the analyais occoheiiVe chains in text or in discourse. In esse*Ice, we posit that vocabulary acquisitiOn Only occurs in contexts where the oral and written chains containing a new lexicak token are extensive and important to the structure of the discourse. Therefore, *kali-wiry research and instruction must be reconsidered in terms of the construction of a set of, lexically harmonious, cohesive chains in a situational context where the new token is embedded in important oral and/or written language. This presents a new direction for the integration of lexical cohesion research with vocabulary and discourse structure research. SUMMARY, IMPLICATIONS AND DIRECTIONS FOR FUTURE RESEARCH Taken as a whole, this research synthesis has revealed the importance of cohesion theory to aspects of literacy learning. Although in recent years there has been a rapid growth in research on vocabulary's function in literacy learning, explanatory theories . Lexical Cohesior 291 still focus on the importance of context without necessarily developing explicit descrip- tions of the linguistic or instructional contexts involved. In the meantime, a slowly emerging body of research on lexical cohesion is providing explanations of the pOSSible functions of cohesive chains in comprehension and composition. This paper interrelated independent lines of research in order to understand text comprehension. The three issues discussed here lead to the following conclusions: (a) in recent years, the study of cohesion has moved to the examination of ;anger cobesive chains as opposed to a mere description and quantification of cohesive deVicand their potential effects on composing and comprehending, (b) karners' abilities:took cohesive devices develop in both oral and wr.len language processes, (c)lexical cohesion theory provides possible explanations for the effects of written at)4 situational contexts in the acquisition of vocabulaty. The synthesis implies that researchers in vocabulary and text comprehension reexamine lexical cohesion and cohesion theory. Likewise, cohesion researchers should discover the wealth of applications of cohesion theory to the understanding of important problems in literacy development. For the purpose of developing and testing theories of the functions of lexical chains in reading and writing, further research is needed in at least the following areas. (a) the proximity and density of lexical tokens in cohesive chains in extended text; (b) the roles of lexical structures in the interrelationships of comprehension and composition, including but not limited to vocabulary learning and vocabulaty choice; (c) the location of a chain's tokens in the semantic network of the text and the relationship of qualitatively different chains to the semantics represented in the text; (d) the effects of demonstrably different instructional and textual contexts on vocabulaty learning and vocabulary choice, (e) the effects of prior knowledge and lexical cohesion on vocabulary acquisiticn, text comprehension, and composition. Research must examine the effects of lexical chains on microstructural and macrostructural processing and production of text viewed from a transactional/interactive theory of reading/ writing in a variety of e .sbroom and sociocultural settings, with a variety of learners, both children and adults. REFERENCES Banutz, J. G. (1986). The anaphora Jigsaw puzzle m psycholmguistm and reading researeb In J W Irwin (Ed.), Understanding and teaching cohesion comprehension (pp. 45-55). Newark, DE, International Reading Association. Baumann, J. F. (1986). Teaching uurd grade students to comprehend anaphoric relationships The applia tion of a dilect instmction model. Reading Research Quarterly. 21. 70-80. Chapman, L..1. (1983). Reading development and cohesion. London, U.K.. Heinemann. Chapman, L. J. (1987). Reading. From 5-11 years. Milton Keynes, U.K.. Open University Press. Cherry, R., & Cooper, C. (1980). Cohesive ties and discourse structure. A study oj average and superior texts at four grade levels. Unpublished manuscrip., State University of New York, Buffalo, Department of Learning and Instruction. Cox, B. E. G. (1987). Cohesion and content orgamzation in the narranve and expository writing of children (Doctoral dissertauon, Northwestern Unmersity, 1986). Dissertation Abstracts In:emotional. 47, 2971A . Cox, B. E., Shanahan, T., & Sulzby, E. (1990) Good and poor elementary readers' use if cohesion in writing. Reading Research Quarterly. 25, 47-65. 292 Literacy The Ory a Roseesih Cox, B., & Su lzby, E (1984). Children's use of reference in told, dictated and handwritten stories. Research in the Ter A:rig of English. 18, 345-365. Cox, B. E., & Trumann, ,14. (1987; December). Children's knowkege of cohesive harmony in expositoty mt. Paper presented at the meeting of the NaCanal Reading Conference, St. Petersburg, FL. Cox, B. E., & Thumann, M. (1988, April). knowledge dffeeences aniong,children who.differ Lit rendiag and Wang proficiency. Paper presented at the meeting of the American Educational ResearchAssociadon , New Orleans, LA. Davis, F. B. (1944):-Fundamental factors of cotnprebension in rniding.,Psychorwerrika...9. 185-197.. De Beaugrande, R., & Dressler. R. (1981). Introdsiction to test linguistics. London: Longa*. DeStefano, J. S., & Kantor, R. (1988). Cohesion In spoken and_writteu dialogues: An ken:0g** nf cultural and textual constraints. Linguistics and Education,I, 105-124. Eller, M. A. (1979).,Meaning and choke In writing about literature: A study of cohecion iss the expository texts of ninth graders (Doctoral dissertation, Illinois Institute of Technology, 1979).. 'Dissertation Abstracts International. 40. 4571A. W B. (1989). Vocabulary acquisition from listening to stories. Reading Research Quarterly. 24, 174-187. Freobody, P.. & Anderson, R. C. (1983). Effects of vocabulary difficulty, ttxt cohesion. and schema availability on reading comprehension. Reading Research Quarterly. 18, 277-294. Friedman, L. B., & Sulzby, E. (1987). Cohesive harmony analysis: Issues of text pragmatics and macrostructure. In J. E. Readcnce, & R. S. Baldwin (Eds.), Research in iiierocy: Merging perspectives (pp. 297-305). Rochester, NY: National Reading Conference. Gipe, J P (1978-1979). Investigating techniques for teaching word memungs. Reading Research Quarterly, 14, 624-644. Goodman, K. S. (1986). What's whole in whole language? Portsmouth, NH: Heinemann. Goodman, K S., & Gespass, S (1983). Text features as they relate to miscues; Prawruns. (Tech. Rep. No. 7). Meson, AR: University of Arizona, Arizona Center for Research and Development. Halliday, M. A. K., & Hann, R. (1976). Cohesion in English. Londocu Longman. Masan. R (1984) Coherence and cohesive harmony. In J. Flood (Ed.). Understanding reading comprehension (pp. 181-263). Newark, DE: International Reading Association. Herman, P A., Anderson, It. C., Pearson, P. D., & Nagy, W . E. (1937). Incidental acquisition of word meaning from expo...SLons with varied text features. Reading Research Quarterly. 2?, 263-284. hwin, J W (1986). Unde standing and teaching cohesion comprehension. Newark, DE: International Reedit% Associati=. JeVrins, .1 R., Matlock, B.. & Slocum. T. A. (1989). Two approaches to vocabulary instruction: The teaching of individual word meanings and practice in deriving word meaning from context. Reading Research Quarterly. 24, 215-235. McKeown. M 0 (1983). The acquisition of word meaning frem context by children of high and low ability. Reading Research Quarterly. 20. 482-496. McLin, J P (1987) Coherence and cohesioa in the writing of eighth grade students ;Doctoral dissertation, University of New Orleans). Dissertation Abstracts International. 49,172IA. Moberly. P C (1978) Flirmentsry children's understanding of anaphoric relstiorships in connected discourse (Doctoral dissertation. Northwestern Univerrity). Dissertation Abstracts International. 39. 4787A. Monson, D L (1982. March) Effect of ope and direction on compreLension of anaphoric relatioruhip. Paper presented at the International Reading Association WORD Resexch Conference, Seattle, WA. Morgan, J L & Sam., M B. (1980). Discourt, and linguistic theory. In R. J. Spiro. B. Bruce, & W E Brewer (Eds ). Theoretical issues in reading comprehension (pp. 165-197). Hillsdale, NJ: Erlbaum. Mosenthal. I H , & Tierney, R. I (1983) Cohesion. Problems with talking about ten. A briefconunentaty (Tech Rep. No. 289). Champaign: University of Illinois, Center for the Study of Reading. Nagy, W E., Anderson, R. C. & Herman. P. A. (1986). The influence of word and text prop:rties on learning from contest (Tech Rep. No. 369). Champaign: University of Minas, Center for the Study of Reading. Neuner. I L (1987) Cohesive ties and chains in good and poor freshman essays. Research in the reaching of English. 21. 92-103. 29S - 'Lexical Cohesion Pappas; C. C. (1985). The cohesive harmony and cohesive deraity of children's onl and written stain. In 3:D. Benson & W. S. Oreaves (Eds.), Systernatic perspevives dircerse (Vol. 2, pp. 169-166). Nevi-Yce4 Abkx. Pappas; C. C. t Brown, E. (1987). Leatuing to read by madingt,Learning bow to extend the functional Poteattial of language. Research in the Teaching.of Englishi 21,160-177. ituddell; R. B.,.&-Speaker,:R:B., Jr. 41985). The Intenctive nadiag iwocc-ac-ivarkid. lea Singel R. B.- Ruddell (Eds.), Theoretical models and patties of reading (pp. 75I-793);Idada1é, Miura, 6totsky, S. (1983)...TYpes of lexkal cohesion in expoaitoty Writing: implications for deve4ophis thc **het!' -lazy of academie discoume. College Comporidon and Conyitunkistion, 34, 430-446. Stotiky,_ S. (1986). On learning to write about Ideas. eatesetorposision nnd Comoninicalink . 276-293. Tanym, D. (1985). Relative focus on iuvolvetnezt in oral and-written discourse. In D.:Ii..01sOitN. Torramx, & A. Hildyard (Eds.), 1,1teracy, language, and kansing: The nature and itivaigieniti reading and writing (pp. 124 -147). Cambddge: Cambridge University Press. Taman, L. M. (1918). Vocabulary tests as a measure of inmiligeres. Journal of Educadonal Psycholrgy, 9, 452-466. Wine, S. P., & Faigky, L. (1981). Coherence, cohesion and tvriting (panty. College ComposWors and Convnunkation. 32. 189-204. 7 -> -A" THE CONSTRUCTION'OYNARRATIVES BY GpOD AND Pook REmigssl- Rose-Marie- Weber University at Albany, SUNY This study, part of a larger study on linguistic differences between gOOd arid poor readers, focuses on children's abilities to construct narratives.' Ind.ripproahtaken here, constructing a narrative is viewed as-a process of selectiog;, interpretifig and encoding complex, causally linked inforniation into temporally ordered Cs: 0,%efien with components of orientation, evaluation, and resolution (e.g.,:i0slett,, 1986, fol- lowing Labov, 1972; Peterson & McCabe, 1983): Through t!**0.4.0, 4.4144i,s- , tic means that children 'use to transform the activities and 0441 relatiOns of piCtirred characters into narrative discourse are analyzed. Good and poor readers Are coriPared with respect to the ways that they solve the problem of encoding the eVerits so that the discourse coheres into a story structure and makes a point. This study of the productive abilities of children with differences in reading ability is intended to enlarge the details of linguistic knowledge ;..ad processing that normal and poor readers cesmmand so as to determine, with converging evidenee; the possible sources of poor readers' verbal weaknesses and lag in reading ability. Vellutind and Scanlon (e.g., 1987) have offered an explanation for reading disability that implicates linguistic coding deficits as the principal source. They elaborate their perstettive developmentally, pointing to deficiencies in encoding, storing, and retieving phonological and syntactic components of language in younger children that in turd lead to deficiencies in processing semantic components in older children. This study of good and poor readers' construction of narratives extends in a preliminary way Vellutino and Scanlon's work in progress on children's speaking abilities as they relate to reading abilities. Research on the construction of narratives by children offers a range of multifaceted views and findings on how children apply linguistic knowledge in complex ways to serve discourse. Studies of narrative abilities have delineated the intricate features that children develop and coordinate to present settings and characters, relate and advance events, and connect physical and mental states to activities. The contain and functions of clauses, the expression of causal connections, the manipulation of 'This study is based on data collected at the Child Research and Study Center at thc University at Albany through a grant from the National Institutes of Child Health and Human DevelopMect awarded to Frank R. Vellutino (Grant #ROI HD 096558-02A1). Thanks arc due to Frank Vellutino, Donna Scanlon, and their staff for subject selection, dam collection, and transcriptions, as well as to Lynne Ruggiero and Michael Green for subsequent malyses. 1114.10/4.111r 295, Li. 296 Literacy Theory and Research verb tenses, the use of syntactic and semantic cohesive devices to relate the given to the new, the choice of writlen language conventions have all been given attention. Further, the relation between the social setting On the quality of children's narratiyes, . such as speaking to a peer in contrast to speaking VI an adult, has been taken into account. To mention only several studies, Peterson and McCabe (1983) examined narratives spontaneously told to experimenters from three analytical points of view,:whereas Preece (1987) sketched the functions and forms that, children's narratiiTs,taktlii natural conversations. In other studies experimenters have used pictiire hOOksliiiii relieve the children of memory demands and allow comparability across.subjactS, but may introduce complications of interpretation. For instance, Berman (1988)cqricen trated on the development of 7.1.nguistic resources for relating tventsihrough time.as part of a long-term cross-linguistic study, whereas Stenning and Michell (1985) 'raced. children's development of the means for explaining the rehdions between states, and events, finding explanations earlier in children than theory would predict. Other studies concentrate on narratives in relation to reading. In this respect they extend the research tradition seeking relations between prod _dive linguistic abilities and reading ability (e.g., Loban, 1976) and complement the influential work grounded in story grammar that takes recall as a reflection of reading comprehension (e.g., Stein & Glenn, 1979). Geva and Olson (1983) examined first graders' spoken recall of stories told to them with respect to the status of linguistic features characteristic of story-telling in both spoken and written language and the mlationship between those features and their progress in reading. Differences emerged between these good and less skilled beginning readers, for instant., in the syntactic integration of content into complex sentences and, furthermore, in retelling a story more explicitly to a naive peer than to the experimenter who already knew the story. In a .s.ifferent vein, Roth and Spekman (1986) compared spontaneous stories told by normally achieving students with those told by learning disabled children. This group was classified as such because apparent "problems in reading, written expression and/or math" (p. 12), in spite of the children's average or above IQs and, it should be noted, normal linguistic abilities in phonology, morphology, and syntax. The stories told by the learning disabled children included fewer propositions, fewer complete episodes, fewer significant statements expressing explanatory relations among the states and events in the stories, and fewer elaborating details. This study was intended to compare good and poor fourth-grade readers' performance in constructing brief narratives from cartoon strips with respect tt, overall plot and to the structural elements that give a story coherence and value. Because this approach required that the children interpret the drawings as a basis for each narrative, it did not require either reading or recall. Thus the performances would presumably exemplify the children's abilities to select, organize, express, and relate the significant mental and physical states and events one to another so as to offer a story with a point. Examining children's construction of spoken narratives has implications for the use of retelling as a means of assessing the comprehension of written texts. Such assessment generally assumes that children are equally capable ot expressing their understanding of a story they have read and recallod (e.g., Kalmbach, 1986). So little ,. .7 fNarraIves tzy: " , , ela an-a-Ridding . e . , 11.01.717 1mm . . Gittupk- - l :.'4,-nt,, 1 :._ Z., , . :SO - - 1,1I;Oi . -SD Scale .: 7, , it =i-,:iiiit34-' 'i' --: -- SD '.-9.§2`: - ....- .- , . ...,i. 15 .07,":" . .,, - - ,- ::- 2 gb ke, sPore awe Equiv . SD . ' :-: 7::'1-7'.$:-:'74-41:''Z'';',' ,-`..-8 -iailmose Oral Reading fast, accuracy:sense: 4 ' :express* has been required of readerS when their comptehenSioO is assesedthroug auSwerS to quettions that the productive skills and cOorditiatiOn require a -6450, tOF. iraniO4. Yet t4Te'lllaY be .**, SplOng Children that need to be rectignizfd in* assess** ef comprehenazon,throi ,expi*Sed recall Aq'ionistofillog3) ha",s pointotiont,one colger, a good de what is .ietalleil,-hutiittie Erbinvhat is net tectdied: A coAqty...*.k*...:fititio, select an:appropriate scilema, may interpret the-tettint for recall aS?tequiti4tnity.ti_ci.#07: :response; or may suffer from dOcits in productive'. linguistic abilitY:, &cite:** tin study does not allow i!s to separate out these possibilities systeinaticaliY. METHOD ^ Participants in the study were 32 founkgraders from- advantsge&baskttonads..._ who scored 90 or,above.on the Perform** Scale of,the, WISP-It(Wechger,19.141) and had been judged free of grins disorders that. Might ile,irnpliCated;,in. icadiak, difficulties. Half were good, readerg scoring', at the, 50th, percentile 0,,a.bove.o*.die.. Gilmore. Oral Reading, Test (Gilmore gt, Gilinpre, 19'68); the odler..-baif;*0.001: readers scoring-at the 10th percentile or below. Table 1 gives the lO. andeading. scores for the twogioups. Theohildren,*cre asked to tell the story tor.each of 14.csrtean strips by Maitrice. Townsend th4tdepiiied characters, states, and eyeataintendafto`CRli*intti an ironic . E,1 ,1::',.1. . sttny. (Fez...instance,: in one-Sequence? ilia_ fust tranie.Showali:tnan.:*44. fya # ;!,414t- hand headint.for a turkey. behind.hith ate' twO iiitaisdAildiei: The secoid frame shows the sinan looking- at the children crying: is he holds . ttte#:oys,tlie turkey's frittic,siloWs.:the-manfRattsed..ne*-0 .tile dis.tcessetichil4r0 a.m4, ,Aci_444, ,p4i09.n9W, .94, t10.4*mp',-*_f,94#1;t:rii*OlOyig . head ..noW .lying-,on a Stump. *Elie , third llie n1P,'sini14*.dr.0, aqd` thPi tw.4394-14bla A16!?ing, 4, 04.1044,0.t..)' , , ....:Z.q -i 298 Literacy Theory and Reseirch, Eseh child was asked to construct a narrative by an adult experimenter who provided a model narrative in the simple present tense, describing actions and states and, expressing inferred causes. As tbe child told the story for each strip, the idnit also had the strip in view. Prompts were nondirecfive. Audio recordings of the children's responSes were independently transcribed by two researchers and differeneen resolved by a third. As a basis for comparing the reader ability groups, thc children's narratiVes Wore analyzed for number of viords per story, adjusted for false starts. They Weire:*eo examined holistically for the adequacy of the overall plot fine, the coheienee 0,thei story, tbe expression of the point, and placed in the following categories, as adaelzey from Stenning and Michell (1985): Interpretive: Provides coherent overall plot, showing understnnding of the point of the pictured events and explaiping the relationships among them and charactet.. motivations. For example, "See thisthe kids' father was gonna It was a um a Taanksgiving 'n they had a big turkey 'n then the kidsit was their pet like, ya lo,nw. 'N then like they were crying 'cause he was gonna get killed. 'N then the father looked, ya know, sad 'n then the turkey started crying. 'N then the father decided ta have the turkey for dinner. Ya know at the supper table." Descriptive. Gives frame by frame description of thc pictures; mus include small inaccuracies. For example, "Father's going to kill the turkey 'n he gets a chick [stic9.0] 'n then he puts on it the neck 'n the kids scream ra no. .nd then the father didn't do it. Then dr turkey and the two kids 'n the father had supper." Restricted. Incomplete, perhaps grossly inaccurate. For extunple, "The the man chopped down the tree 'n the girls are yelling at him anci they're crying and the rooster is yelling' at him and they get they have somethin' (to'll eat." Or, 'this girl's screamin' cause he had a ax in his hand and each both girls screamin' girl and boy screamin' cause he thought dey were chopping him and den on Thansgive he chop de turkey. . . . and der they ate him." Agreement on the classification by two independent raters was 81%, differences were resolved by a third rater. The narr-'ives were also analyzed for the incidence of clauses containing evaluations, that is, utterance: ".cir intensify, compare, correlate, and explain the complicating actions in a narrative and so contribute to expressing its point: "why it was told, what the narrative was getting at, or what to think about a person, place, thing, or event" (Peterson & McCabe, 1983, p. 33). Evaluations inchide exclamations; repetitions; hypotheses, predictions; intentions, purposes, hopes; negatives; explanations (Peterson & McCabe, 1983, p. 32). Further, linguistic features that contribute to the coherence of each narrative were analyzed, including (a) the perspective taken on temporality and its maintenance as expressed through verb tense and aspect, and (b) connectives other than and, and so, and additive so, including relative pronouns and complement that (cf. Stenning & Michell, 1985). RESULTS There were no significant differences between the good and poor readers in the length of their narratives (t(30) = 148, p<.23), good readers using a mean of 53.5 . ,construefion of Narratives -- 'Table 2 *eraderlibility Groups and Quality of Narratives Interpretive bescripthe Resincted (SD = 14.81) words and poor readers 46.7 (SD= 11.07), Significant, differeneeS fitr. pear, however, in t1.. distribution of the quality of the narratives as;:igerirretive, descriptive, or restricted in relation to the cartoon, strips ([2 N=4:48)7=; 193;92; p<.001)._ As indicated in Table, 2, the gcod readers _tald narratives, that w*41tiipreti v c in about half of the cases but restricted in about a tenth:Th:40r riaders ,z,YaVenarratives that fell in roughly the same proportions in each ,of the ihree,catsiOrtes. With respect to the incidence of clauses that conauted evaluationi *relation to the qt.ality of the narratives, a repeated matures ANOVA ivith ability facto? and narrative quality as the within factor, as given in lahle 3, Aeltleds, tity significant differences in the raean number of evaluation clauses between the edits (F(1, 30)=1.54, p = n.s.) However, it showed significant differences for narrative quality (F(2, 60) 22.02, p<.001) and a significant ability by narrative quality interaction (F(2, 60)=4.23, p<.02,1). , With respect to other linguistic elements contributing to the quality of the narrafives, good readers also differed from poor readers. In the choice of tense and aspect, the good readers favore:i the present over the past and in a few cases shifted PersPeefive once in the course of a narrative (present 66%; past 28%, shift 6%), whereas the poor readers chose the past and the present about equally often, shifted le in a, few cases, and also mixed the tenses in some (past 49%; present 42%; shift 6%; mixed ' 4%). With regard to connecuves other than and, and then, and additive so, the good readers chose 23 types of connectives over 175 tokens; the most frequent were so, "consequently" (e.g., "the kids didn't waq him to so they were crying and crying"); so, "so that" (e.g., "he wants to chop its head off so he can pluck it"); because; and but. The poor readers chose 14 types over 88 tokens, the three most frequent being the same as the good readers, but the fourth being when. Table 3 Mean Evaluative Clauses by Reader Ability Groups and Quality of Narratives :, A -,-....--.. Groups --. z, Good Poor Mean Interpretive Descriptive 20.4 12.0 5.6 6.5 16.2 6.1 . Restricted 1.5. 4.5 3.0 --,- DISCUSSION This study found differences between good and poor readers,in the waysthaUhcy applied and codrdinated their linguistic knowledge 40 construct narratiVesp.responap., to cartoon-strips. In derscribing,children't holiitfetifito*,PrOchiction,ffieanalysis pnavides inforination that inay,play ti,part in explainiag 0.:10041.11,, becoming fluent readers (Vellutinti &-Scanlon, 1987)., The Wealtitess: thrat, shoW in constructing narratives May have its source in theSkine lingaiStic deficiencies that limit their reading. Further, the difficulty in.ProdRetiVedlietiorSe'lp#, thèdiatc between reading comprehension and the expression of that coinprehenSiOn, tellings. Although the good and poor readers chose about the same number of tVo,t..tis;for, their responses, they differed with respect to expressing the-overall plot-anti OntcOme, of the depicted events. As a group, the good readers more oftenr-conitruetedailpar, and sufficiently elaborated narrative to track the shifts in pictared..states=and:,events coherently, providing an effective interpretation of the logicalrelatiOns amotigthent:. An equal proportion of good readers and poor readers siniply described the 4ctiVities_ in each frame adequately, a characteristic '..1und in younger children (Berman, 1:988),, but nonetheless regarded as sophisticated b.) Stenning & Michell (1985). The Ponr readers offered a much higher proportion of restricted, incomplete, even incoherent, narratives. The poor reade:s were not as skilled in marshalling their linguistic re= sources for the task in hand. The good and poor readers showed many indications of having control over varying narrative devices. Overall, they did not differ significantly in the ineidence of evaluations, those specific linguistic devices used to interpret, emphasize, and 4 . attribute cause to the actions in a narrative. Good readers elaborated their interpretative, narratives more broadly than the poor readers did; their evaluations dropped' ShaiplY, in their restricted narratives, suggesting that here the problem was not a linguistiethie but perhaps a matter of figuring out the picture or not having an appropriate schema. Poor readers, on the other hand, used evaluations as part of their narrative strategies, in a more sustained way across narratives of different quality, suggesting that for them the problem in their restricted narratives may have been more frequently a matter of coordinating linguistic expression. Two types of linguistic elements were examined to identify possible differences, between the reader groups in the ways that they solve the problem of transforming their interpretation of the picturet into language. In the case of the choice and maintenance of temporal perspective, the good readers favored the present over the past, whereas the poor readers chose both equally. In this respect, the good readers were not in accord with the performance of developmentally more mature children in their choice of tense in narratives (Berman, 1988; Stenning & Michell, 1985). It may be that in this situation the good readers followed the form of the adult experimenter, who told the sample narrative in the simple present tense. The poor readers provided a small proportion of narratives in which they failed to sustain the temporal perspective and shifted inappropriately more than once back and forth between pre tent and past. In the choice of connectives other than and, and then, and additive so the two groups of readers also showed significant differences. These elenitnts serve to bind *,, =e4 6` onstrichOii:4:Ndiiiiiiies 000**44y,proyidc orient.Opt, 440 tl* tp,,#.46,4o ori*ta:,ct0J ii4d',00.4.-.010:'afe,- 0*, eie**- t40s490P110. go-ookFto, in the .'*ariety.of sneh.-Conneetiiiet anci'didsnot nie the "tinea,thit,theY chose as frequnt1y as good tiaaoo( aiteiencts:th.ortil'ProduCtiaii-insinind-and poor readers are're .ewrent . interest in :aisessing ,:reading,eMiipreAtensiOn by iislting4tiltken,,, thy hatreiesd:.:4-1*IrMi?,`08./' has #944.51, 094 novice readers can benefit from c911ter0;104.** *** coOtitu,;ei a goO, .t4e,1100,1'13,4 kelleral** .91350.41g.aiRT.00, 010- ar.040144 64-inieir.4.* eiiiirOsilt *4404,...hai*OderstOod , , findings in thiS.,studY.remind,us- that- there arejtlifferences on the. eliseding,,Sio,-.of :retelling that cannot be simply attributed to poet. leading: REFERENCES Berman, R. (1988). On the ability to relate events in narrative. Discourse Processes, 11, 469-497. Geva, E., & Olson, D. (1983). Children's story-retelling. First Language, 4, 85-110. Gilmore, J. V t Gilmore, E. C. (1968). Gilmore Oral Reading Test. New York: Harcourt,.Brace World. Haslett, B. (1986). A developmental analysis of chikken's narratives. In S. G. Ellis & W, A. nonitue (Eds.), Contentporary issues in langvage and discourse processes (pp. 87-109). Hillsdale, NJ: Erlbaum. Johnston, P. H. (1983). Reading comprehension: A cognitive basis. Newark, DE: International Reading Association. Kalmbach, J. R. (1986). Getting at the point of retellings. Journal of Reading, 29, 326-333. Labov, W. (1972). Language in the inner city. Philadelphia: University of Pennsylvania Piess. Loban, W. D. (1976). Language development: Kindergarten through grade twelve (Researcli Rep. No. 18). Urbana, IL: National Council of Teachers of English. Morrow, L. M. (1988). Retelling stories as a diagnostic tool. In S. M Glazer, L. W. Sarfoss, & L. M. Gentile (Eds.), Reexamining reading diagnosis. Newark, DE: International Reading Association. Peterson, C., & McCrbe, A. (1983). Developmental psycholinguistics: Three ways of looking at a child's narrative. New York: Plenum. Preece, A. (1987). The range of narrative forms conversationally produced by young children. Journal of Child Language, 14, 353-373. Roth, F. P., & Spekman, N. J. (1986). Narrative discourse: Spontaneously generated stories of learningdisabled and normally achieving students. Journal of Speech and Hearing Disorders, 51, 8-23. Stein, N. L., & Glenn, C. G. (1979). An analysis of story comprehension in elementary school children. In R. 0. Freedle (ed.), New directions in discourse processing (pp. 53-120). Norwood, NJ: Ablex. Stan:ling, K., & Michell, L. (1985). Learning to tell a good story: The development of content and language in children's telling of one tale. Discourse Processes, 8, 261-279. Vellutino, F. R., & Scanlon, D. M. (1987). Linguistic coding and reading ability. In Rosenberg, S. (Ed.), Advances in applierApsycholinguistics (Vol. 2, pp. 1-69). New York: Cambridge University Press. Wechsler, D. (1974). Wechskr Intelligence Scale for ChildrenRevised. New York. Psychological Corporation. . ;:t 4.4: _" ":, `,,' : ,,,, ;,...7%, ",, q..,.:, ,..:',,,.." '','' .. :FftiPP2411*-1E*Ig:'9F., commn:Mw#10:0**-k: -ccoNtErrioNs:,_ ..:,, ':',.:CoptcgiryA.1 -,' GEt._,'''''''.%':,\ ,- . '',1,1174.-itTS:,:- '3 .. .±..2 ';'';'0'"'''i-.,"-'' '' .,?, Da1,441144.t.1*** -Univeisity..p.i,ciyik , , el .. .,.. ,._. , Recent_research in 'science edUcationlias_PonVidnd Ten4in.8,00-8..r°__ ......,Mw-1.T.:Pe-r '-,--- AndljOwerful tools to.exaininphoyi; snicienti,"!.e4kkopi textiS;;(*. inek'p*elitulkiot ijo: the. conatiuct`of "conceptual.Chinge,learnint:"' oncipikal.:41;iingelleitirniiiidt .,..;i:. '*i:-.Pfil:g.c::: -tiOi'..senin. of .;"6.44i4ur_44,-:Peir c1.0kOnt.; aiii*ledgP-O.i. 'g.c1S;i114t..;!:.1$4.4 0,00.444 ,voimconk infOrtaation obtained ibiint 4 'given plieniinefion--(H000ty*-He*,40.,. A9.84;Poiner, Striltei,E0stin,, 4,- ceozog, ,1982): We ,weidd!-,s# that conceptual. , -- A !.4 1 ;Plinio lethiting !iii.s nr-cn-fred Whan sIndents cni*O-04;f0a04 th#01-n.n0 make their Own;food, rather than obtaining food,frtiiii.the onyirennient; A linsPnreil 41 sniencn ncinnatinntas 0.5300v.PbCcinnw_nci104 *2._ ,c_00nrAnnt Ointnin". 'learning' is a veryclifficult prOcess. Science, eihicatOta,haVe-ptoVided,ConiPell.ing.;#4, dence that students ,are resistant to changing !. their iiuotieg,kpolegige4 Sometinies:_. called "naiVe conceptions," about nianyx 01,e.seietici.ge conoóptn tlieY,4F4r (90.0-_ ent -1981. Eaton Anderson & Smith 1984).°TeittbOoks fegifiar chiiiroolsr. instrua= tion and even good science ..instruction . often fall to:bruit about .conceptuak,changs, learning-in sciende,(AiiderSon-&:Sinith, 1987; Eatou,,-Anderion; &$niTith',;)!1#41 'Resea,icilers 'in rcictiri8"ndgCntinwhnvn :Pinv04'014.400'.03 ,06,itb..90.C. '9.1# stienceeddeators We .folind:,Suidents 0,/40;4e:4-y:py,appeafpliOld:00.their- I ,,.. -naiye;;but-intuitively aPpealifig, science-ConeePtiOns .40010_ reading texta,., that : -,;:,,..., 'inforpation that direetly contradicts, Studern! d000c004..(AiYOnninin..4-ii ...19,89; i Alvermann,, Smith, 8,c- Readence,. 1985; 47son,;1982). Yhellisfiookta.40, ie44:::te*. .., .that conflicts with :their -paiye.coneeptio041041114.,:***:4 infc*Inininii-:0,0' , their existing knoi_yledge structures (Alverniann*Eintki'.51Ri,OrivOinfil414*, pew inforination:'into separate- and diatinct-0041edge,:atioc4res:,..(Ap4S!*4_ 1977; i,-, 'lintili '1985)9F ifiPete the he* 44'9,1'1440A Titt0t,Ati47s, 911i ;.c.sii.thil?44.1' Re.,#entiYi'n*dinti.ini-nniOnzt 44-_iin keinn...4!'n4nninPOn 004**,0 ill* ':':i appear to prOiiibte'coikeiitual charige learning froth tea: -tonCeritical Variable APPears, .,to be the direct confrontation and- refutation of. students, ' naive. conceptions: 711,* ks,; '.:. . *b'en ; pa.* et a 104.4' d4ectlY,a0.4* 41.*',#::ii-r 4134 ,s.41400 rookkglyo.. oli* 'thet -direttlY' refute, those,haiiie. conceiitiop4;;40elt4::afe:10(0,14telTto cna*-ti.iniF: , " :.:,... Literacy Theory arid Research conceptions. Several studies (Alvermann & Hynd, 1989; Maria 1988, 1989; Roth, 1985) found that "refutation" texts appear to be effective catalyrts for conceptual change learning. A critical variable that has not been examined sufficiently in either the reading or science education research is the level of students' commitment .tolheiv naiVe conceptions and its effects on learning from textMarshall (1986),,Carne.c4sestle. examining this variable when she identified the "peraorial need to knOW" 10-0 imporn.. tant factor in students' willingness to change their naive conceptions. It Would:Make' sense intuitively that the higher the level Of commitment Students have to-theiLinitivej conceptions, the less likely they are to chang them through reading texts-a StIPPOSI, tion that research in social psychology supports (Bern, 1970; ROktach, -1970);,We would expect that high levels of commitment to one's naive Conceptions *OirlO lead one to assimilate new, incoming information so that it is consistent with what is. already known. Conversely, students who are unsure of their naive conceptions should be more likely to change them. Lower levels of commitment to one's naive conceptions may allow one to acconunodate the new information more easily. The purpose of this paper is to report data relevant to this hypothesis. In particular, this study addressed the following question: Are students who are sure of their naive conceptions about scientific phenomena less likely to change their conceptions through reading conflicting information in texts than students who are unsure of their naive conceptions? Because we were also interested in the effects of refutation text onstudents' naive conceptions, we asked an additional question: Are students who are sure of their naive conceptions more likely to change them if they read refutation text than if they read regular science text? This paper reports data obtained as part of a larger study that examined the influence of refutation text on conceptual change learning. METHOD Subjects Subjects in this study included 33 male and 57 female sixth-grade students drawn from classrooms in three comparable urban, public schools in a small westem city. We chose upper-middle class neighborhood schools so that a majority of students would ha%e little or no difficulty I :Ming the instructional materials we developed. Reading >cores on the Stanford A. hievement Test for our subjects ranged from a grade equivalent score of 2.3 to Post 7,igh School (PHS), with a mean of 8.8. Tests and instructional materials were developed for two science topics, cells and matter These topirs had been used in previous research (Dole & Smith, 1987, 1989) and the tests and materials for this sn.dy were developed and revised from these earlier materials. Prelposuests. Two tests wele created, one for the cells topic and one for the nts, sAniuniimeat to Naive Coheeptions .riipertOpic. R.aeh test was .used as both &Pretest' ancPposttest., contmOO' foul*. ,was Atiod:for- both tests. Thii iricluded a doyer iiagel,with,general qUestiOrit abditt.each trpie. T4c;coiei. page was, followed 4y. *ritte4 iastflOons',fot,Studenta-,f6. itSe what '-theyIcneY,,,about cells or Matter tci ansi),er the rem mtng questions Sticlenti' level of coniMitment to their paive.sceneeiitiOnSliiss :90509:44y, defined hOw sure they-were of their resPoiges;,.:After 00440*, followr ...stron, was asked: 'HOW- sure 7are You of yriiir:tirisker?'.''1:4#'::*-0:, followed by a request to.circle one of thefollowing:r.riot *ure;iPineWkal 3,14*r 00-0E; very stir 0. Questions for each topic were modeled after "akiliCation gnestions Often .04 '1?y ,..- science educators (cf. Anderson & Smith, 1983;-Eatoo,AndersonofkSni1th4984)., The items required that students apply wh.: theyknow about, the topics to real-werldphenomena. For example, one of the questions about cell growth_ was: If ths.grass on a lawn is mowed, it grows back in just a few days. llow does a stalk of grass grow? Try to explain what is going on inside the grass plant that makes it grow. Similar questions were used for the matter test. An item about the physical changes in matter subtopic was: When a volcano erupts, the magma flows out onto the surface of the earth and hardens. Explain how magma hardens. Scoring templates were created based on students' responses on the pretests. Responses were categorized into five levels. Level 0 was used for blank rr "onses or "don't knows." Level 1 was used for responses that reflected naive a,...ceptions. Level 2 was used for partially accurate responses; level 3 was used for mostly sophisticated responses. Level 4 included complete and scientifically sophisticated responses. For example, the scoring template for the magma question was: 0 = don't know 1 = the magma hardens, cools down, changes from a liquid to a solid 2 = molecules implied or stated directly, particles mentioned 3 = explicit mention of molecules plus 1 of 3 of the following: speed of molecules pmximity of molecules attraction of molecules 4 = explicit mention of molecules plus 2 of the 3 above The experimenters created the scoring templates together, resolved differences through discussion, revised the templates and created rules to follow fur ambiguous responses. Interrater reliability was established at 89%. Treatment materials . Three sets of instructional materials were developed for the two topics: traditional text (rn, considerate text (CT), and refutation text (R1'). The traditional text was taken from current fifth-grade science textbooks (Cohen, Del Giorno, Harlan, McCormack, & Stayer, 1986; Mallinson, Smallwood, & Valentino, 1984). To develop the CT we restructured the U to improve coherence, unity, and audience appropriateness (Anderson & Armbruster, 1984). To develop the RT we added refutation statements to the Cf. Refutation statements included direct confronta- 09 17, r.t," -pteriCy,Theery,ifiti R*0 tion -and refutation of student? naive conceptions: Theae itaternenti*re derived ripen studenteMostfree(itent naive responses On.die priteSt:.Fiiteaarnple,greteSt*iim*-, on the,Cells unit indicated ; that, many stirdents in otri sample '..thOtight, that; foOdiWas, used; orik,in their stOrnachs. Therefore, in develOeing .the:eella.refunitiOrt added theIefidittion:stateMent: . Sorne fieeiiIe think that-food is used only in peole'S:stonsachsi-blit thiS, _ nOt,t. Both-1he cella and matter units,,wemdivided intoseetions--,-kiectiOaszforithe:, cells unit and '4 for the matter unit Refutation statementi were added at the begitMing, and concluSion of each section. Procedure The study was conducted in intact classrooms at three schools. ,Thelwe eateeri-, menters met individually with each teacher to ensure that* topics ofcellt-and Matter had not and would not be covered during the 7-week, Oeriost,of the.study. Students were r9domly assigned to one of the threetexttreatments. Packets Were. prepared for students ;based on text treatment and topic- Order. To controllororder effects, half the students received the matter tests and materials- rtrst,,Sind lhe,Other half received the cells tests and materials first. Written scripts. Weit, premed, and_ read to subjects before all tests and treatments to control for the effect of-different, experimenters. Pretests were administered to all students at the same time of day over a 3-day period. Three weeks later, the experimenters returned to the schools and administered the text treatments over a 4-day consecutive period. Three weeks after the treatmerda had been administered, the experimenters returned to the schoolt and administered. the posttests. Posttests, like the pretests and treatments, were administered at the same time of day for all subjects in the study. Data Analysis Only items for which students gave naive responses on the pretest were used for data analysis (n =760). Two dependent measures were used in the analysis. The first measure examined students' level of sureness for their naive responses on the pretest. The "sureness" scale was originally designed to provide descriptive information used in the larger study. In this study we collapsed the 4-level scale into a dichotomous measure for statistical analysis. The second dependent measure used for data analysis examined conceptual change learning from pretest to posttest. Conceptual change learning was operationally defined in a conservative manner. Evidence of conceptual, change was scored only when students gave a naive response on the pretest and a scientifically sophisticated response on the posttest. Posttest responses scored at levels 3 and 4 were considered scientifically sophisticated and therefore recorded as "did change." Posttest responses scored at level I reflected naive conceptions and therefore were recorded as "did not change." Posttest responses of 2 were removed from the analysis because they did not reflect what we called "scientifically sophisticated responses," even though they may have reflected some increased understanding of the concepts. Posdest responses 3.1. A `- .'n-sdent? Commitment to Naive Conceptions of 0 indicating "don't know" were removed from the analysis because of their ambi,guity. ,For the first pkrt of the analysis we aggregated data acruss,41.three,teZttyPes and, both topics. We created four categories baSediOn,stibjects'.' sUreneis.of theii-naiVe conceptions on the pretest and whether they:Chinge-46 Sophisticated responses oi the posttest The.four categories mere: not sureldid chonit (4,Siti$C);./fOr #1;eldtit 'Change, (.4s6c), sureldid not --hange,(SIDNC),=anit'.3u,eldgefianie (S/DC).Wó -tfien calculated be.numbers- of 4ms-that fit into-40 categOry, andCOnVitle4*Senunibers,into percentages of total resPonseS.A chi-square anilYiis.was., determine if the numbers of items in the categories Were signifiCantlY dip:fent:POI' the second part of the analysis, we bzoke down each of the-fait; citteloriei into three groups based on text type. A chi-square analysis was employed On din Obiallied for each of the four categories to determine if there were significant differences among the three types of texts (TT, CT, and RT). RESULTS Findings for the first part of the analysis indicated that, overall, 85% of all naive pretest items were coded on the posttest as did not change (NS/DNC +S/DNC); whereas, only 15% reflected a change to more sophisticated conceptions (NS/DC+ S/DC). These data indicate that most students did not change their waive conceptions to scientifically sophisticated ones through reading the treatment materials. Further, students' level of sureness about their naive conceptions did not appear to be a factor in their change. In fact, naive responses about which students were unsure were more likely to be maintained than responses about which students were sure (p<.05). Of those naive items for which students were unsure, 46% did not change to scientifically sophisticated responses (NS/DNC). Of those naive items about which students were sure, 39% did not change to scientifically sophisticated responses (S/DNC). Furthermore, of those items about which students were unsure, only 8% reflected a change to scientifically sophisticated responses (NS/DC), and of those items about which students were suie, only 7% reflected a change (S/DC). These findingb indicate that students maintained their naive conceptions regardless of whether they were sure of them or not. In the second part of the analysis we examined students' level of sureness and their change on the posttest depending upon the tnree types of texts they read. Chisquare analysis revealed two categories where statistically significant differences were found (NS/DNC and S/DC). When students were unsure of their naive conceptions and they read the IT or the CT, they were more likely to maintain their naive conceptions than when they read the RT (p<.05). Of those items about which students were unsure and did not make changes to scientifically more sophisticated responses (NS/ DNC), 38% of the items came from students who read the U, compared to 35% from students who read the CT and 27% from students who read the RT. . Conversely, chi-square analysis revealed that of those naive items about which students were sure, they were more likely to change them if they read the RT than if they read the TT or the CT (p<.01). Of those items about which.students were sure 31 011MM -.TNrA.J'amcse OS-15 = _ :50; .tAi. 45, (..,,11_**. 50 .z. ,61 'ce zoa. 15 10 5 0 Traditional Considerate 'Reigtatian, 'TEXT Figure I. Percentage of NS/DNC and S/DC responses 'as.a function of teattyPe., and did 'make chinges-to scien4fically.morosophiaticatediesPOnsq,(SOf the itettts ettaile. 4'60 4udenti ca..me,fr* and stt1-4-ehei:'1.09,10,4 ;she CF lelatiOnship- between: the different, type:s of teSts atudchts-f reid;:,ttucie:Fita-: iurenesa about &it halve conceptions and:their As*, 40004 il-flOce,frfiv tire :1; Shows-that When Stitdenti *el* 4p0e0 About' theif fiiiya` nedoc6p0c0i.:0,0:* g0; Oore-l4t6-10.0n44-q.losaco*4404slif thiY'reAld.tha IT Oita: OrY ria4-!he: cror7tbe RT. AdditiOnalIY,.whin,Stildents Niere awe abbitt.th4rtai*CatrePitiot* they were more likeIyto change them if theY: read *1qt:hail-the Tr-Or the et._ DISCUSSION 'Students fit:,this study terided- ti ,inaintaht=theirr wave conceptions' re4c, Oc.ien044:100.1.04,t4ta qiive.09ceikias . Of w*her theY-*ItT , Su Commitment to Naive Conceptions 309 conceptual c:tange learning from texts; further ley do not support our original contention. It would appear that, at least for the two topics used in this study, one's level of sureness or commitment to one's ideas has little effect on one's ability to learn fmm texts. Why was students' level of commitment not a factor in their abilities to chair-. their naive conceptions? Perhaps the answer lies in the way we meaSuied,"leyel of commitment." It could be that our measure of "sureness" did not.capture,the IeveI of students' commitment to their naive conceptions. Perhaps additionaLinfauctions or a clearer definition of what is meant by "sure" would have been more ,effecpe in detertnining students' level of commitment. A more likely explanation may be that students' level of sureness or comniitinent is not important. Rather, some other affective variable(s) is. For example, a ntriable that may be more important is Marshall's (1986) "personal need to know." eedaaps the topics we chose were not relevant enough for students to care about, and therefore they had no reason to change their naive conceptions. Perhaps a more important factor is whethcr students care about the topics, rather than their level of sureness. For example, we certainly can imagine topicsevolution versus creatiodsmwhere one cares about the topic to a degite that is more likely to influence one's willingness to change one's conceptions. Results of the study do point to the importance of refutation text in conceptual change learning from texts, especially when students arc sure of their naive conceptions. When students were sure of their naive conceptions and they did change, they most likely read the refutation text. Figure 1 dramatically illustrates this point. Why does refutation text appear to be more successful in promoting conceptual cbange than IT or CT? Because refutation text directly states and then refutes students' likely naive conceptions, students may be more likely to notice the conflict between their ideas and the more scientifically sophisticated conceptions presented in the text. this may foster the cognitive conflict that some researchers feel is necessary for conceptual change (Hewson & Hewson, 1984; Posner et al., 1982). A broader, but certainly relevant, question is: Why did students hold onto their naive conceptions so strongly? Many educators would argue (and we would concur) that reading text is not sufficient to brizzu about conceptual eiange. Hinds-on experiences, guided reading experiences, discussion and experimentation would appear to be additional activities ti at may be necessar; igrhaps even sufficient, for conceptual change. The mle that text plays in this process needs to be defined more clearly. The affective dimension of conceptual change learning from texts requires further investigation. Results of this study suggest that sta-ents' level of sureness about their scientific conceptions may not be a critical variable in their conceptual change learning from science text. The resin% of this study also suggest that then may be no simple and clear relationship between affective variables and ..4gnitive outcomes. REITRENCES Alverrnann. D. E., & Hynd, C. 0987. December). Overconung misconceptions in science An on. tine study of prior knowledge actwaiwn. Paper presented et the =sung GI tin National Reading Conference, St Petersburg. FL. 0'; 310 LiteracY asi 14-54* Alvennann, D., & Hynd, C. (1989). Study strategies . cocrecting misconceptions in physics: An interven7 tion. In S: McCort 'nick & j. &tell (Eds.), Cognitive_cuid social perspectives for literacy research aivinstruction (pp; 353-361). ClaigM.Nitiaiiil Reading CarifereOce.Alvennain, D. E, Smith, 'L C, & Rei6esice J. E.- (1985) . Prioilitairwledge _and comprehension tit coMpatible and incompatible text. Reading Reseatchit24iFterlY4k4P-,-43&- Anderson, C.-W., & Smith & L (1983,, kill). eitticCen's c*ipeoiis: the eoncePt of unseen rays. Paper PO:Salted at the meeting of Me American :Sue.4,1104eolt ksociation;Ifontieal, Canada. Anderson, C..W., & Smith E L.. (1987). Teething :deice: In V.:Richards' on-Koehler (Ed.); ja;b7icakti, handbook: A research perspective (pp. 84-111)..New York: Lia*Min: Anderson, R. C (1977). The nodon of scheroata indthe educational-en** General disensiir* of the conference. In R. C. Anderson, R. J. Spire, & W. E. Montague (W.); Schooling iusdikeirciimisti6 of knowledge (pp. 415-431). Hillsdale, NJ: Erlbaum. Anderson, T. C., & Armbruster, B. B. (1984); Content arta textbooks. In R.-C. Anderion;i: Gabor:hi kr--R. J. Tianey (Eds.), Learning to read in American sdrools (pp. 193-226). Hillside* Erlheinti. Bern, D. J. (1970). Beliefs, attitudes, and human cffairs. Belmont. CA: *adswor-th. Clement,1. (1982). Students' preconcepdons in introductcay mechanics. American Journal of Physics, 50, 66-71. Cohen, M. R., Del Giallo, B. J., Harlan, J. D., McCormack, A. J., & Stayer, J. R. (1986). Science. Glenview, IL Scott, Foresman. Dole, J. A., & Smith, E. L. (1987, December). When prior knowledge is wrong: Reading and learn- 5- <-` ing from science text. Paper presented at the meeting of the National Reading Conference, St. Petasburg, FL. Dole, J. A., & Smith, E. L. (1989). Prior knowledge and learning from science text: An instructional study In S. McCormick, & J. 7.nrrIl (Eds.), Cognitive and social persoectives for literacy research' and instruction (pp. 345-352). Chicago: National Reading Confamce. Eaton, J. F., Anderson, C. W., & Smith, E. L. (1984). Stnelmus' misconceptions interfere with science learning: Case studies of fifth-grade studems. The Elementary School Journal, 84, 365-379. Hewson, P. W., Hewson, M. G. A. (1984). The role of conceptual conflict in conceptual change and the design of science instruction. instructional Science, 13, 1-13. Lipson, M. Y. (1982). Learning new information from text: The role of prior knowledge and reading ability. Journal of Reading Behavior, 14, 243-261. Mallinson, G., Smallwood. W., & Valentino, C. (1984). Science. Mcuristown, NI: Silver Burdett. Maria, K. (1988, December). Correcting misconceptions. H4.ing fifth graders learning with science tict. Paper presented at the meeting of the National Reading Conference, Tucson. Maria, K. (1989, December). Correcting misconceptions. Effect of tea type. Paper presented at the meeting of the National Reading Conference, Austin. Marshall, N. (1986, December). Prior knowledge: Facilitator or inhibitor of comprehension. Paper presented at the meeting of the National Reading Conference, Austin. Posner, G. J., Strike, K. A., Hewson, P. W., & Gaups, W. A. (1982). Accommodation of a sciailific conception: Toward a theory of conceptual change. Science Education, 66, 489-508. Rokeach, M. (1970). Beliefs, attitudes, and values. San Francisco: Jossey-Bass. Roth, K (1985, April) Conceptual change learning and student processing of science texts. Paper presented at the meeting oi the American Educational Research Associadon, Chicago. 3 14 . MititAeliVE.TEAOINGAi4LtAillIN:01Gi'fAciatATINO N4NO,DISABLEDiSrUPENTS!:tRASSIrrtwww*ywg' torElditit: , , - ," Grace Z. Dnran-ant7,0a$04. Scanlan Universityof Articria , They and research have sought to clarify the Copttihinions of y.iyinnsloffnsof intexaction on_ learninsi: Learniu 'has been Charactedied as, an. >NO* ,. requires cognitive and social collaboration (Anderson, 1984; ,,..Rninelharte,),1980;. yy= . ,gotsky, 1978) . On the personal level, the knoWledge an individual,acquireS,,accesSes,, ory, _and:aPplies to aid further acquisition all int..tract. The*netions.::cOntn:_teia-a: c9ntiniP39s. f*.ces., -1980)T kesea., cx144_,.ted ike#01144144, knowledge and idea relationship formulation (Reyes, Owego; paran,,:&: scindon 1989) imleamingind exprettion processes. Socially, interaction also hasbeen found crtb..,a to learning. rnowledge bases an.I proficient learning skills periorm,w4110,10OPPO#A,!!acilln:is tasks, beyond their performante when collahorating,with,peers(Sliyhk1980).' ever, the nature of these,independent tasks is limited.,*IWIedgekirliCatfOn'o, sort is restiicted to the near transfer type or to close-OnteXtp_si**J00*(Sa.01*-.. & Perkins, 1989). When taught to collaborate, student* of all skill,le'Velsteariv:_m*,,, and apply their knowledge more elaborately (i.e., using higher abitriction) AS-n* learning independently. Learningin -School is greatly influpiced hy the type of claSSroopyilitetaitiOn. encouraged in that environment. Appropriate 0*a:interaction and Out stinilates learning and is beneficial for *dents ofslyabilitios Interaction between, people is the frintipfe;nrechanispr by witichlenyning -etnit development occur (ygotsky, 1978). liygOtsky deseritied me social COMeitt as inStrumental in guiding cognitive development. He maintained that". . ..eyeryftip*On in the child's vognitive development apPears tWice,)Op two levels; firston,Me:sOCisl, between people as an interpsycliologiCal qitegory-and later on:th,e,persOial cojiiitivP level, inside the child, intrapsyChelogicallY . . ." 6.1ygatci, 1.978; p 31), During the imerpsychological phase, the indi,vidialinithicts- yAt4 external re'This reseateb is funded as part ol the IctuFtive Toching Project (0008630125) by the Office of Special E =lion and Rehabilitative Services, us. Depinnsent of education. c:5 312 Literacy Theory and Research sources to acquire new information and skills and collaborates directly with external sources, such as teachers, instructional materials and peer: (Reyes, Gallego, Duran, & Scanlon, 1989). In the intrapsychological phase, the student employs the information provided by these external resources to activate the students' prior knowlerige and, form relationships among what is already known and the new information (Rumelhart, . 1980). , Learning environments rich in external resources provide new information,thns-, activating students' prior knowledge. In these environments, learning is bestiacilitated when students are allowed to interact with each other (Au, 1980; Palinscar kBroWn,_..4",T, 1984; Slavin, 1980). Such learning interactions have been described as a partnetshiP between novices and experts (Vygotsky, 1978). A novice is an individual with limited knowledge who learns from a it -e informed expert. Through this alliance novices jointly experience activities with exp rnd gradually come to perform the same functions. First, an expert guides the novice through the activity, performing most of the cognitive work. Initially the novice participates as spectator and gradually assumes more responsibility. In time, both expert and novice come to share the cognitive work. Once competence grows, the novice is able to assume the expert role. Experts employ both cognitive and scnial skills for learning. Successful learning requires a double or split mental focus. Expert learners simultaneously focus on the content and monitor their mental operations used for productive learning (Locke, 1975). One important cognitive skill, self-interrogation, is used for identifying the relationship between ,Ine's prior knowledge and curricular content (Rumelhart, 1980). Expert learners apply and integrate this content with their prior knowledge. The process is triggered when an expectation about content has not been confirmed or when unfamiliar concepts are encountered frequently enough to impede full understanding (Smith, 1975). The expert reacts to the comprehension collapse by slowing down the rate of processing and allocating time and effort to achieve understanding. Experts negotiate meaning and expend their knowledge by using multiple resources in their environment, including 3ther persons. Student experts employ social skills to effectively interact with their teachers and peers in ways that draw upon others' knowledge. Learning in school contexts becomes problematic for learning disabled (LD) students because they often lack appropriate social interaction skills (Bryan, Wheeler, Rican, & Henek, 1976) and are often unaware of their comprehension failures (Torgeson & Licht 1983). Specifically, LD students lack awareness of: (a) their limitations as probleol solvers, (b) compensatory strategies to overcome such limitations, and (c) self-management techniques for monitoring and checking their own progress (Brown, 1982). These poor learner behaviors may lead to their characterization as school novices. Although LD students do not typically involve themselves in self-monitoring activities and other c:icient strategies when engaged in learning, research has documented that they are capable of activating these strategies when cued to do so (Bos & Filip, 1984; Wong, 1980). In this study, students were cued to overtly exercise the cognitive processes of knowledge assimilation and accommodation identified in Schema Theory (Anderson, 1984; Rumelhart, 1980). Students were enzouraged to activa:e prior knowledge by 3 1.6 ; . -trausitionfront Novice to Expert 313 , 1 sharing those ideas wah others and to form relationships among ideas. In this process, LD students progressed toward becoming experts. INTERACTIVE SEMANTIC MAPPING Our aim was to develop a context which facilitateciinteractive dialogue for:1mAthg social studies content. Crucial to this novice/eatpert partnerShip is th*, dialogue, a primary means by which support is providedival adjaSted: This,interactiVe., dialognie seitieS to challenge the novice and enables the student to participate,th$ learning process before becoming an expert. This study examined the expansion of students' knowledge in two ways: (a) the understanding of the procedures of participating in the semantic mapping strategy across time, and (b) the comprehension of content area concepts. Students in this study applied a metascript (Gallimore & Tharp, 1983) to, guide their verbal interaction. A metascript consists of intermediate level verbal Pr,iiMpts that are more universal than specific routines (e.g. "Does this 'mak selise?"),'but more powerful than general self-regulatory skills such as self-interrogation. Intermediate level prompts advocate responsive inter Ttions rather than specific level-presoribed formulaic statements. The intermediate level prompts used in this study included seven components of interactive teaching and learning (Bos & Anders, 1987; Gallego 1989; Gallego & Anders, 1988). Students were encouraged to: (a) activate prior knowledge by prompting each other to recall related past experiences, (b) tie new knowledge to old by connecting their related prior knowledge and experience to each other and to new information provided by their peers or by the text, (c) predict relationships by hypothe- sizing how their ideas and those of the text relate and prompt each other to identify alternatives, (d) use cooperative knowledge sharing by using others in their group as resources for information and consensus building, (e) teach concepts to their peers in relation to the organization of the semantic map or the passage, (f) justify relationships between and among concepts by explaining their responses, and (g) confirm understanding by questioning their understanding to resolve misconceptions. These interactive learning components were embedded in the instructional context of pre-, during, and post-readings of the text and in discussion for the purpose of understanding social studies concepts. The teacher initially modeled these components as instructional prompts to the student group. As students acquired the components, they employed them as learning promcts for each other. The interactive learning strategy assisted in obtaining two desired goals: (a) students' collaborative comprehension monitoring, and (b) students' comprehension of text concepts. METHOD Subjects Subjects were 6 bilingual 11- and 12-year-old girls identified as LD according to school district criteria. The school district criteria used for identifying students as '5:41 314 LD included (a) a severe discrepandy between intellectual fiinctioningtuntaeadernic aehievement and (b) one or more deficits in deter distx#papirevaluition teams Subjectswithastandardscoreof85orhigheronl , Wxhsler Intelligence Scale for `chiwoRevise4 (WISC-k,We'ehalet,-49,79,00 a disability in reading.** SeIected"for this.s;74y.:SUbjeCts':readirig, grade scores, an achievement,meaSure ranged froria_24 ta TheSe students were a-tepresetitathfc subsaMple Of subjeCts*liq patti larger intervention study -Rif the', iMprOVement of COntent area:,eti, teacher's role 'changed acrossfinit froni niadeling the Strategythi coaching the students on the use of the strategy. Procedure The students learned and rehearsed semantic mapping, an interaCtiiic ,itrategk, (Scanlon, Duran, Reyes, & Gallego, 1990; Stahl & VanciL,1986).UsitIgliyeniMetinsecutive related passages from a social studies text over 'a 5-week period A Oa chapter was studied over 3 to 4 days during each week. Students were engaged,iththes, activity for approxin.ately 40 minutes a day. Each week students partiCipidef lesson phases: (a) brainstorm, (b) clue list, (c) relationship map cokitia060;., ad (d) confirm understanding. Students began ,the lesson by brainstorming concepts relatedAo. a. contentarek, passage. The title of the chapter was presented as the central idedafor the 1;orainsterm.. As a guide for brainstorming, students were instructed to think of personal experiences, and previous knowledge related to the topic. Idess offered by students were,deperatect, for their relevancy by the group and accepted if appropriate. Secondly, studentsgenerr, ated a clue list by skimming the text to identify key vocabulary and cOneititSith,,thechapter. The sources for clues included pictures, titles, subheadings, and,Nfords in bold print. Thirdly, students made predictions regarding the relationshipS:amOng.the, concepts generated in the brainstorm and those presented in the text, andkpremPted each other to justify those predictions. Upon reaching consensus concerning,thcine-`, posed relationships, students organized and displayed their predicted relationships; among concepts by creating a semanv2.:, map. During the fourth phase, students.reed' the passage to confirm their hypothesized relationships, and then conferred with each other to review and possibly change their map accordingly. Data Sources Three data sources were collected and analyzed: (a) videotaped classroom interactions recorded during the implementation of the semantic mapping activity, (b) multiple-choice comprehension quizzes collected at the conclusion of weekly lessons, and (c) student-written summaries, also collected weekly. The 1st, 3rd, and 5th weeks of the intervention were videotaped. Three bilingual coders observed 15-minute segments of each of the four phases of each lesson. Each coder was assigned 2 students to observe. Coders recorded all verbal cmtributions made by their assigned students. Students' verbal contributions were coded according to interaction using four categories: (a) interactive, (h) noninteractive, (c) off task, and (d) procedural. During training sessions coders collect,..viy observed 2 target 3:404 Transition from Novice to Expert subjects at 2 separate times for 20 minutes each (5 minutes per lesson phase). After each lesson phase, coders compared their coding. When coding was not in agreement, coders discussed their decisions until concensus was reached. An interratez reliability 'of 86% was attained. Student uerances coded as interactive reflected statements associated Vrith the seven interactive components (Bos & Anders, 1987; Gallegoi :1989).',NoniOtekactivei utterances consisted of on-task statements that reflected a directive andi3OFnoticollabia= rative approach. Statements coded as off-task were utteranceathit'ifictooti** riv", the procedures nor to the lesson content. Procedural utterances refleeted,stateidenti which referred to the management and set up of the lesson. Two comprehension measures were used to document students' contentünderstanding. Multiple-choice comprehension quizzes were adniiiiisterici Weeify and scored for the percentage of correct responses. Test items included conceptiral iterna, requiring students to draw relationships among concepts, and vocabulary iterni, tequirr ing definitional knowledge. A similar but longer multiple-choice comPrehension test was administered at the conclusion of the 5th week and again as a follow-up measure 3 weeks later. A second comprehension measure was also collected each week. Students were asked to generate a written summary of the lessons' content. Students generated written summaries individually. Researchers encouraged students to write all they knew about the topic and assured them that their ideas, not writing mechaelcs, were of primary importance. The purpose of the writings was to identify student knowledge not identified by the multiple-choice quiz. The written summaries were scored as part of the larger study. Six coders were trained on the procedures of holistic sexing using a 10-point scale (Irwin & Mitchell, 1983). During training each coder scored the same 10 papers. When ratings were not in close agreement, coders discussed their reasons for assigning a specific score and reached consensus. Reliability was established at 92% agreement. RESULTS Student Contributions Across Time Contributions generated by the total group during week 1, week 3, and week 5 were examined. Trend analysis results indicated a significant increase in the amount of student contributions made across the 5-week period F(2, 14; 9.7, p<.001) (see Table 1). Tukey HSD post hoc tests were employed as pair-wise comparisons on the means to identify the significance of each time interval. Results indicated that there was a significant difference between mean scores collected during weeks 1 and 3 (p.05). Results for the mean difference for the verbal contributiens made during weeks 3 and 5 were not significant. Teacher Contributions Across Time Although teacher contributions varied slightly from week to week, trend analysis rt......dts document these fluctu..tions to be nonsignificant. Furthermore, the total amount 340-' -t "Table 1 TrindAneilysis for the Total Student Coritributiops,Aerps4eelti 13;.;and$: _ of teacher contributions across the weeks was not correlated to the written OunirnarieS. holistic ratings nor to the comprehension quiz scores. Written Summaries Across Time Holistic scores assigned to students' written summaries were examined. Trend: analysis results indicated a positive trend for summaries across theweekPgif54'(2i, 14; 3.17, p<.05) (see Table 2). No significant trend,was indicated betweentheholistic scores recorded for week 5 and those recorded fot the follow up 3 weeks later. Student Contributions and Comprehension Measures Correlational analysis irvealed a significant positive relationship between_total group student contributions and the holistic ratings (r= .68, p<.003):, Howeyer,Ahe. total amount of student interaction was not significantly correlated with the comprehension quiz scores (r= .06). Proportional data provided further information regarding the nature of theyoup interaction. In addition to students' incrtased participation in discussion mil-C*4 ip the number of contributions made over the 5-week period, the quality of these contributions also improved. The percentage of use of interactive, noninteractive, prbee-: dural, and off-task utterances revealed a consistent use of interactive statements and' an increased use of proceiural statements over the 5-week period. A decrease in offtask contributions was also documented across the intervention weeks (see Table 3). DISCUSSION We consider the intervention a success in aiding students' progression from novice status toward expert status for several reasons. Collectively, data revealed that students,. progressed in two ways. First, students' procedural and conditional knowledge (Crbss & Paris, 1988) for strategy implementation improved, that is, students demonstrated knowledge in how and when to use the strategy. The improventent of strategy:implementation was documented by the quality and quantity of the dialogue generated among members of the group. Increased quality was reflected by the students' task appropriate dialogue employing interactive utterances. The strategy was also successful in instilling a collaborauve environment in which students were free to share s3"2,0 _ .117: Transitiofrom ArOice to &peel : Table-2 -tiend Analysis for Holistic Ratings on Written SummariesAcroisFeeks 1, 3,. "14:5 golis4-14aigs" Intervention Week .5i '1 3 .76 5 Note. Percentage scale. , knowledge and use each other as information resources documented bYlOtlikiraietlquantity- of participation in the group discussions. Secondly, students increased their declarative knowledge (coo & the social studies content The holistic scores awarded to studenta''Written iUnimaries indicated a qualitative increase in their ability to express their tinderatinclin4 OfiteXt, concepts over time. Conceptual understanding was largely maifitainesl- thrO:Ugh'-the administrafion of the written summary follow-up measure (only one stu461V§:Perintmance indicated a decrease between posttest and follow-up). Through interaction with a supportive teacber and peers, the students were led to perform at an increasingly more mature cognitive and social level. These OPportunities enabled students to effectively respond to and challenge each other Hoy/ever, the teacher did not merely instruct the students and then leave them to work wigged. The teacher served to advise and model the apprOpriate strategies. -During Week 1 the teacher participated in interactions in which the students and teacher were mutually responsible for getting the task done During the following weeks the students adOpted more of the interaction initially unc1 ,Aaken by the teacher. The teaCher then acted less as a p...-4e1 and more like a coac'. by interjecting praise, encouragement, and some management. This study emphasized studei participation in the learning activities by providing opportunities to use the skills beivr instructed (Gallimore & Tharp, 1983), evaluating student performance, and providing information for the readjustment of instructional support (Palinscar & Brown, 1984). The interactive components encouraged the students to respond even if their level Table 3 Percentage of Use of Interact; Noninteractive, Procedural, and Off-task Contributions Across Weeks 1, 3, and 5 Weeks 1 3 5 Interaejve 44.5% 39.0% 34.0% Types of Interaction Procedural Noninteractive 0% 2.0% 0.1% 18.5% 34.0% 43.0% Off-Task Raw Total 36.0% 25.0% 19.0% 146 366 304 - 318 was not yet :diat of an exPert. Students' responses granted:the teacher oPpOrtintidea to:gaugi,Cnair.cOrnpetence and.render apkropriatefeedback.,:ik this 'way -4:titeracitiVe. :14irt*.g.PiTfict4i* ocOsi9n fer it**. *4.44,0v4i,1*if100'9f,:074.0cPi A-:leVal that in, traditiOnat instinction:ia :Oftertjauisked:by 4004' ;44,;:noencoot::,ici: reSpond until they 'IPprgarekf9`:*91:6*CP;:f0t6000:04' is 0109***PY0.*: percepbon1 of LI? students, according to-*44 ti!ejr:**01:01)0nOd them. (414e§usi4600,'91.s9P,"4(.0*. 1909). . The linduaLShiR, oUresPonSibility for Conditetinglhe'' les*, ,0001 teacher students Mar bnthe key to succesgukinternaliiatiOn ancigeneralizationnon, learning. The Opportimity :for verbal expression provided in this study, ernr studenta fry Validating student ideas its, viable and inaportantInter*etiye learning sidet in changing the students' self-perception from school nOvice to eiperts;;TheSefeattires may be esPeCially Worthwhile for LD students whom we fotind to befiefitlroM the collaboration and support of others. REFERENCES Anderson, R. C. (1984). Some reflections on acquisition of knowledge. EducationalResearcher, 13, 5-10; Au, K. H. (1980). Participation structures in reading lessons with Hawaiianchildren: Analysis of a cultural1y appropriate instructional event. Anthropology and Education Quarterly, 11, 91-115. Bos, C. S., & Anders, P. L. (1987). Semantic Feature Analysis: An interactive teaching strategy fOr fullitating learning from text. LD Focus, 3., 55-59. Bos, C. S., & Blip, D. (1984). Comprehension monitoring in leant' zg disable: and average students. Jounol of Learning Disabilities, 17, 229-233. Brown, A. (1982). Learning how to learn from reading. In J. Langer & T. Smith-Burke (Eds.),- Reader meets author, bridging the gap: A psycholinguistic and social linguistic perspective (pp. 26-54): Newark, DE: International Reading Association. Bryan, T , Wheeler, R., Felcan, I., & Henek, T. (1976). "Come on dummy": An obsavational study of . childrens' communications. Journal of Learning Disabilities. 0. 53-61. Cross, D R., & Paris, S. G. 0988). Developmental WA aoalysis of childrens metacognition and reading comprehension. Journal of EducatIonal Psychology, 87, 131-142. Gallego, M. A. (1989). Verbal interaction among teachers and elementary learning disabled students engaged ;tt directive and interactive pre-reading strategies. (Doctoral dissertation, University of Arizona, i989). Dissertation Abstracts International, SO, 2848A-2849A. Gallego, M A , & Anders, P. L (1988, November). Comparison of the quality and quantity of commurdeation during interactive and directive Instructional practices. Paper presented at the meeting of the National Reading Conference, Tucson, AZ. Gallimore, R., & Tharp, R. G. (1983). The regulatory functions of teacher questions: A microanalysis of reading comprehemion lessons (Tech. Rep. No. 109). Honolulu: The Kamehameha Schools, Kameharneha Educational Research Institute. Irwin, P & Mitchell, J. N (1983). A procedure for assessing the nwriess of retellings. Journal of Reading, 26, 391-396. Locke, E. Q. (1975). A guide to effective study. New Yotk: Springer. Moll, I. , & Diaz, S (1985) Ethnographic pedagogy: Promoting effective bilingual instruction. In E. Garcia R Padilla (Eds.), Advances in bilingual education research (pp. 127-149). Tucson: University of Arizona. Myklebust, H. R., Boshes, B., Olson, D., & Cole, C. (1969). Minimal brain damage in children (Final Report, USPHS Contract 108-65-142). Evanston, IL: Northwestern University Publications. Palinscar, A. S., & Brown, A. L. (1984). Reciprocal teaching of comprehension-fostering and comprehension-monitoring activities. Comition and Instruction, 1, 117-175. r_ `::. .,,TrcinsitionfroitilisfYic e to Expert , e E.°I.,,oallego, M. A., Duran, G. Z., & Scanlon; D. J. (1989).,Integration of internal coicepts aid ea:aerial factors: Extending the knowledge of learning dti#40.0c,les.cents; Ajiriiat of Eprly Adolisceice,:9,112-124. , E. (1900).' Schemata: The building Wake of eOgnition. In R J Spire, B._-____.e:Bruce, ljtipwerods.), Theoretical, issues in reading comprekensiarv.(pp. 331,8). Edbaum. Perkins, D. N. (198). Rocky mads to transfer Rethinking mechanisms Of i neglected. .Salomeni phenemenen. Edueational f.iychokgisk 24,, 113-..-,142. S ;Scanlon, D J, -Duran; G. Z., ReYeSEE L, & oallege,,M. A. (1990). Interactive smart* maititing Maniac* aubinitted for publication. .Slavin,:lt- (1980). Cooperifive learning. Review of Educatiorzl Reuarch, SO, 315=342. (1975); Compiehension and Leandng: A conceptual framework for teachers. New York:,Holt, Rinehart, & Vrmston. Stahl, S. A., & Vend!, S. J. (1986). Discussion is what makes semantic maps work in vocabulary instrucdon. The Reading Teacher, 40, 62-67. Torgeson, J. K., & Licht, B. G. (1983). The learning disabled child as an inactive learner Retrospect, and prospects. In J. D. McKinney & L Feagens (Eds.), Current topics in karning disabilities (pp. 3-31). Norwood, NJ: Ablex. Vygotsky, L. S. (1978). Mind in society. The development of higher psychological processes. Cambridge, MA: Harvard University Press. Wechsler, D. (1974). Wechsler Intelligence Scak for Children-Revised. New York: Psychological Corporation. Wong, B. Y. L. (1980). Activating the inactive learner Use of questions/prompts to enhance comprehennon and retention of unplied information in learning disabled children. Learning Disabilities Quar- terly, 3, 29-37. 323 2.- THE EFF}..... _ OF PLAE UPON STUDENTS' TEST PERFORMANCE AND METACOGNITIVE AWARENESS Sherrie L. Nist and Wgstele L. Simpson University of Georgia - Past research has provided considerable information about the differences between efficient and inefficient readersnnd the characteristics of effective learning ,strategies. However, these studies have not, fully addressed the more criticaLisSup Of training students to hyve executive control over these strategies in order to become independent learners (Weinstein, 1P88). To become autonomous learners, studentt must be able to plan, implement, monitor, evaluate, and if needed, modify a plan of adtibil 'with a variety of tasks and texts (Kluwe, 1987). Knowing that even College students Often lack the ability to monitor and control their learning (Anderson & Armbruster, .1984; Maid, & Berry, 1984; Pressley, Snyder, Levin, Murray, et Ghatala, 1987), we opemtionalized these executive control processes into a heuristic entitled PLAE [Preplan, List, Activate, and Evaluate] (Simpson & Nist, 1984). Based on tetrahedral models of learning (Bransford, 1979; Jenkins, 1979), PLAE focuses on five student-directed operations necessary for strategy control and regulation. Students must ,a) establish goals, allocate resources (i.e., select strategies, allot time), and make a plan of action that incorporates the appropriate strategies and distributes practice over time; (b) have a repertoire of strategies for the numerous tasks and texts they will encounter because there is no one superior or generic method of study; (c) select the most appropriate strategies based on the characteristics of text, task, and personal learning preferences; (d) activate and monitor a plan of action and make appropriate changes, when necessary; and (e) evaluate their plan's success or failure in order to plan for future situations. PLAE is a recursive model that involves students in four stages of test pmparation. In Stage 1, Preplanning, students find out information about the test and set performance goals by answering a series of questions. In Stage 2, Listing, students list the most appropriate strategiec and construct a task-specific study plan that outlines their specific goal for each study session, the amount of time they predict it will take tt. reach their goal. and where/when they will study. In Stage 3, Activating, students implement and monitor the plan's effectiveness and make adjustments if their plans are not working. Stage 4, Evaluation, occurs after students have received their test scores. Students evaluate their performance by diagnosing errors and looking for patterns of stengths and weaknesses. Thi information is then used as they plan for aubsequent exams. Two previous studies have been conducted to validate PLAE's effectiveness. In the first study, the planning variable, as operationalized by PLAE, was found to be 11,44.. .11.1maaga - more predictive of and accounted for a greater amount of the variance in test performance than did encoding, rehearsal, or wurcl knowledge (Nist, & Mealey, 1989). A second study focused solely on the possible tole PLAE Might have:. in improving both test performance and metacognitive abilities. In thiS'itudy,Wa foUnd, that over a 5-week period, students' test scores, as welt as.both on-line:and:glObal metacognitive abilities, improved (Nist & Simpson, 1989). HoWeVer;;Pastreirearch has yet LI compare students who were trained to use PLAE -With thoSe. whit Were, trained to use more traditional methods of management such as Scheduling nod tizing. Thus, the present study sought to anSwer tbe following qtreStions: (a): Woiricr: students trained to use PLAE perform agnificantly better on four.tontent imeitisns than an alternative group trained to use traditional time managernent skillsr(h) Would. students trained to use PLAE significantly improve their abilities to globally Predict test scores over an alternative group trained to use traditional time manageinent skills? (c) Would students trained to use PLAE significantly improve their abilities to engage in on-line predictions over an alternative group trained to use traditional timemanagement skills? METHOD Subjects The subjects were 56 at-risk students (45% male; 90% Caucasian) enrolled in four separate sections of an upper-level study strategies course at a major southern university. Students were mandatorily enrolled in this course as a prerequisite to taking regular core courses These students could decode words and comprehend brief passages as measured by their scores on a state mandated basic skill exam. However, they had difficulty understanding and remembering extended pieces of text as measured by a departmental exam over a college-level psychology chapter excerpt. High school grade point averages and SATV scores were equivalent for both groups (PLAE, hsgpa =2.54, SATV = 400; TM, hsgpa =2.56, SATV =407, p>.10 in both cases). In addition, they had a mean university predicted grade point average of 1.78 on a 4-point scale Two seciions were randomly assigned to the PLAE condition (PLAE, n= 26), and the other two sections served as the Time Management condition (TM, n = 29). Procedure In Phase I of the study, subjects in both groups recei ' intensive, direct instruction on a variety of study strategies. (Each of the two Tchers taught one PLAE and one TM group to control for teacher effects.) During this 5-week period, all subjects learned how to activate prior knowledge, survey and annotate text, and use a variety of recitation strategies. The overall training differed only in that the experimental groups received instruction on PLAE, and the alternative treatment groups received insmrction on more traditional time management techniques. During Phase nc5. II of the study, the 5-week data collection period which followed Phase I, all subjects constructed study plans or time management schedules as part of their preparation for each of four full-length content area chapter exams. PLAE group. The initial trailng took place over a 6-day period. (10 Day 1, the rationale for PLAE was discussed and procedures and examples of FLAP, were pro- vided. PLAE subjects were then assigned to conitnict-a plan for the first** and bring it to class the following day. On Day 2 the attributes or sin effeedyeilsin were discussed and students met in pairs with a checklist which desi:ribectp* itteniths.- and weaknesses. On Days 3 and 4, plan monitoring and fix-urstrategies **44cussed. Subjects took the exam on the fifth day. On Day_ 6 all exams *ereietitrned so that subjects could diagnostically evaluate their performance. With the exCeption of discussing the PLAE model in great detail, this same cycle was "ollowed for each of the remaining three exams. Time management group. The alternative treatment group also went through an initial 6-day training cycle that focused on time management principles. Fcir each exam they constructed a weeldy schedule and a daily "To Do" list. On Day 1, the rationale and steps for constructing schedules and lists were discussed. We distributed examples and asaigned students to create schedules and lists for the following day. On Day 2 the attibutes of effective schedules/lists were discussed and students met in pairs with a checklist to evaluate. As with PLAE subjects, the TM group spent days 3 and 4 di monitoring and fix-up strategies. On Day 5 they also took the exam. On Day 6, TM subjects were provided with the correct answer for each exam item and were permitted to ask questions on confusing items. Data Collection Four exams based on four full-length content area chapters from college-level texts (communications, political science, biology, and psychology) were constructed, each containing 40-45 objective items and a balance of memory and higher level questions. The reliabilities on the tests ranged from .68 to .87 and there were no statistically significant differences between mean item difficulties (.59, .64, .66, and .65 respectively, F= .871). Mean item difficulties were determined by computing the proportion of students getting each item correct and then averaging these proportions across each test. For each test, all subjects engaged in two key tasks as ways to measure both global and on-line metacognition. First, as they took the exam, they predicted whether they thought they got each objective item correct or incorrect. Students were instructed to put a "1" if they were sure that i: ,18.5 correct and a "2" if they had reasonable doubt about the correctness of their :nswer. From these responses, the mean proportion of correct predictions (i.e., they predicted that they got the item correct and it was correct, or they predicted that they got the item incorrect and it was incorrect) was computed for each of the four exams for both groups. In addition, after completing each of the four tests, subjects also engaged in global predictions by predicting the overall grade they thought they would receive. - .#egn Teit[5;coris.(and Standard OeVicitit?ns)fur PME - 89:88: (7.5) -iS.sii, .(9.16) (9,16) 72.(X (1't..8:2) 15.25' 714' P.071-, 168:89: 0. RESULTS The results or this study indicated several sipificanandings. [Fins41Meated-', measums analysis revealed a- Maio .eftect for. glob (6i15:79, p<i.0197), atid=fiO. interaction, between ['test and'groUp (F306=1.11, p< 3481) As thOWnliii-thc,:aimin (and itandsid deviations) in Table 1,^PLAE inbjects ,sooied;lkigher.ihen TM subjects on-ali exams. Second, a ,chl-square analysis indicated thatstatistically .signiiiCanichingess,m favor of the PLAE condition occurred between Tl and13 (=1.57.,P9),171:#4:,- [ SCOICS T T4 (x1=16.149, p<-.001), T2 and T3 (2*12.26, p<.001), and 13 (x2=25.57, p<.001) in mbjects' .ability to globally predict test scores, No:siariaticiilchanges occurred between the groups on T1 versus Tior T2 versus T3 ((2=-L63-and, 2.12 respectively). Table 2 indicates the frequency of under, exact, and over global predictiMis tor each test for the two groups. The Cochran's Q TCS2, used to detemiine irtkere*Cre overall changes in subjects' abilities tn globally predict their grades,:Wakapitittically significant, x2.7.81, p<.05. Stewart's extension ofMcNeznar's ttst-(iewart4955),.used to determine the nature of these changes, indicated that subjects in ibe, -PLO group changed from making oveigractions to making exact predktions they took T4. Such change did not occur in the TM group. As shown in tible4,,the on'sy Agnificant change that occurred in TM was between T1 and T3 v;fhen there wasan increase in ovmpredictions. Table 2 --- Frequency of Under. Over, and Exact Global Predictions for PLAE and TM Under Test 1 3 Test 2 Test 3 4 Tat 4 3 2 PLAE Over Exact Under 12 11 11 11 19 4 2 4 9 1 23 8 TM Over 15 13 12 12 12 12 8 9 iddeof Contrasts for Global.PrediCtions for PIAE and-TM ... thwT2. 4:*: T4 T4 -1'34 T4 . Rise; Under tfrIar .039 :000 .077 .423 269 '.038 .269* .39: 8* .077 .039 300* 369* 423* Undei X)verz-k . Exact , /14 423* .730' :138:- .071 .d8s 7.036 T="Test. *Statistical significance. p<.05. Means (and standard deviaSons) for mean on-line predietions are inchr4e0ia t Table 4. There was a statistically significant interaction ,betiveen:-giorrii 4.10114iti, prediction, F3,153=746; p<.0002. Simple effects for diffiretieeiheti4CkgrOjikait gros each level of on-line prediction indicated no statistical diffetences for TI and T2 (p<.634 and .149, respectivelY). However, itatisilealik,ii " -00 differences, favoring the PLAE subjects, were found for t3 tinri T4" (F=9:56;, p<.00), and F= 26.80, p<.0001, respectively). These results indicated that abilitiei - - to predict on-line accurately were dependent on the test subjects took. DISCUSSION AND CONCLUSIONS The results of this study indicated that subjects trained in PLAE performed statistically better over the four content area exams. It could be argued that the significance found in this study was a result of test order rather than an improvement in strategy control and regulation. However, the fact that there were no statistically significant differences in the vean item difficulty levels of the tests weakens this argument. In addition, scores for both groups on TI, the most difficult of the four tests (.59), were e Table 4 Mean Proportions of Correct On Line Predictions (and Standard Deviations) for PLAE and TM - -4. Tests 1 PLAE TM 76.19 (8.00) 77.10 (6.94) 2 3 4 74.85 (8.43) 70.86 (11.31) 74.00 (5.24) 69.29 (6.11) 85.42 (4.64) 74.26 (10.01) ---- :-.i' --....: -,.....-: ir::el,Tr' V." 326 , Literacy Theory and Reseich - higher than were the scores for T2 and T3. Furthermore, T3 (.66), the easiest of the four tests, had the lowest scores for both groups. The statistically significant difference between PLAE and TM over the four exams gains practical significance when examining the letter grade differential across the, four tests. Specifically, for two of the exams (T1 and T4), the PLAE group received; Bs, and the TM group received Cs. On T3 PLAE subjects received C grades, and,the, TM subjects received Ds. On T2 there was a half-grade difference groups. At first glance one letter grade or one half-grade difference between the Ortitit might seem inconsequential. However, when it is noted that these high-risk:000s, were predicted to perform at a 1.78 or D + level in university course ism*, a hilt or whole grade beyond a D + would make a difference between probation ar4 staying in school. Given the ñ that most college freshmen have not developed executive control over their independent learning, more opportunities for vertical transformations with PLAE would probably make these initial effects even more pronounced. The results of this study also indicated that subjects train*A in PLAE became statistically more aware metacognitively as indicated by both global and on-line predic- tions. It should Ix noted that these differences between the groups became more pronounced over time. As noted in Table 4, the PLAE Group increased their on-line predictive ability from 76% on the first exam to 85% on the fourth exam. However, the TM group declined in their ability from 77% on the first exam to 75% on the last exam. In addition, the interaction between group and prediction indicated that the two groups predicted on-line with equal degrees of accuracy for the first two tests. But by the last two exams, those in the PLAE condition statistically improved metacognitively, whereas those in TM declined or remained stagnate. Hence, had this study ended after the first data collection point, the statistically significant differences favoring PLAE would not have been noted since both groups were initially equivalent in their abilities to predict their on-line performance. Not only did PLAE subjects improve in their abilities to pre:act oz.-line, but they also statistically improved in their global predictions. It is interesting to note, however, that no dramatic change was evident for tF PLAE group until T4, again indicating the importance of giving students sufficient time to learn to control and regulate new strategies. PLAE may have facilitaten metacognitive performances because subjects had to specifically define each of the four tasks, select appropriate strategies, construct a task-specific plan of action, monitor and evaluate that plan of action, thus encouraging strategy control and regulation. In contrast, TM subjects, with a knowledge of the same Phase I strategies, appeared not to be able to control, regulate, and monitor those strategies in an appropriate fashion to the four different tasks and texts. The TM subjects did not perform as well on the tests, nor did they grow in their abilities to metacognitively assess their global and on-line performances. These findings have implications for research as well as for professionals helping students to become more autonomous learners. Long term training seems particularly important when conducting research on at-risk students who nee. -,-,.werful interventions coupled with lengthy training across a variety of tasks and texts. 329 '43 -; Anderson. T. H., & Armbruster, B. B. (1984). Studying. In P. D. Pearson (Ed.), Handbook of reading research (pp. 657-680). New York: Longman. Bransford, J. D. (1979). Human cognition: Learning, understanding and remembering. Belmont. CA: Wadsworth. [I: r: r- Jenkins, J. J. (1979). Four points to remember: A tetrahedral model end memory experiments. In L. S. Cennack & F. I. M. Craik (Eds.), Levels and processing in luanan memory (pp. 429-446). HiEsdale, NJ: Eribaum. Kluwe, R. H. (1987). Executive decisions and regulation of problem solving behavior. In F. E. Weinert & R. H. Kluwe (Eds.), Metacognition. motivation, and understanding (pp. 31-64). Hillsdale, NJ: Eribsum. Maki, R. H.. Berry, S. L. (1984). Metacomprehension of text material. Journal of Experimental Psychol..)gy: Learning. Memory. and Cognition, 10, 663-679. Nist, S. L., & Simpson, M. L. (1989). PLAE, a validated study strategy. Journal of Reading, 33 182-186. Nist, S. L., Simpson, M. L., Olejnik, S., & Mealey, D. L (1989). The relation between setkelected text learning variables and tesi performance. Manuscript submitted for publication. Pressley, M., Snyder. B. L.4evin, J. R., Murray, H. G., & Ghatala, E. S. (1987). Perceived Readiness for Examination Pelformance (PREP) produced by initial =ding of text and text containing adjunct questions. Reading Research Quarterly. 22, 219-236. Simpson, M. L.. & Nist, S L. (1984). PLAE: A model for planning sucoessful independent learning. Journal of Reading, 28, 218-223. Stewart, A. (1955). The comparison of frequencies in matched samples. The British Journal of Statistical Psychology, 10, 29-32. Weinstein, C. E. (1988). Executive control processes in learning. Why knowing about how to learn is not enough. Journal of College Reading and Learning, 21, 48-56. 33 U CORRECTING MISCONCEPTIONS: EFFECT OF TYPE OF TEXT Katherine Maria Colkge of New Rochelle Joanne Mons Johnson Scarsdale School District 4, A number of studies (e.g., Alvermann & Hynd, 1989; Dole,_1989;.Maria, have found that considerate texts and refutation texts (texts that refer directiy,,to,an existing misconception and then cortect it) promote learning ofscientific.infonnation that contradicts children's and adults' misconceptions more than the inconsiderate texts conunonly found in science textbooks. For example, Maria,(1988) fond that fifth-grade students who read an experimenter-constructed considerate refutation text learned the scientific explanation for seasonal change better than those who read an :nconeirterate text from a fifth-grade science book. The considerate text discussed and directly refuted the common misconception that summer is warmer because the earth is closer to the sun; the inconsiderate text made no reference to the misconception. The main purpose of the present study was to extend this work on the role of text it: the correction of misconceptions by investigating whether presenting scientific information about seasonal change in a narrative, which is also a refutation text, would promote learning more than presenting it in an expository text, which is either a considerate refutation text or an inconsiderate nonrefutation text. Narratives whose primary purpose is to present content area information are common in tradebooks on science topics (e.g., Cole, 1986) and in articles in children's science magazines (e.g., Ranger Rick published by the Natione Wildlife Federation). This type of text is really a hybrid of narrative and expository text, therefore, we are referring to it as soft expository text, a term suggested by S. Valencia (personal eommunicarion, December 2, 1988). There are a number of factors which suggest that soft expository text may be easier for children to understand than regular expository text. In general, even considerate c pasitory texts are more difficult for childr..,r. to understand than narrative texts. Very few children are aware of expository text sttuctures (Englert & Hiebert, 1984; Richgels, McGee, Lomax, & Sheard, 1987). Since chiMren are mt male of these structures, they cannot use them as an aid to comprehension am; learning. On the other hand, even young children seem to be aware of narrative structures and are able to use this awareness as an aid to comprehension of stories (Mandler & Johnson, 1977; Stein & Glenn, 1979). Freedle and Hale (1979) imnd that when young children used narrative structure in organizing their recalls of expository texts, they Ind bener recalls. They siggestee that one might create a schema for expository prose by transfer329 331 ';$' Literacy Theory and Research ring students' already existing narrative schema to expository text. Pincus, Geller and Stover (1986) reported one way of promoting this type of transfer with middle school children. Using magazine articles that were not organized chronologically, they chronologically numbered events in the article and provided frames to help the children organize information into a summary with a narrative structure. When given this support, children were better able to identify important events and summarize the article rather than simply copying information from it. Soft expository textstructures. may be another w..y of facilitating the transfer from narrative to expository text schemata since this text type includes elements of both types of structure. Interest is another factor that may make stories easier for children to comprehend than expository text. Stories seem to be inherently more interesting than expository material. Many authors of content area textbooks have attempted to increase children's interest in these texts by inserting narrative anecdotes at critical points (Hidi & Baird, 1988). However, a number of studies (Garner, Gillingham, & White, 1988; Hidi & Baird, 1988) have found that when narrative anecdotes are inserted into exposition, children remember the interesting narrative information which is unimportant but not the important content area information. Since the effect of soft expository text on learning of scierice information, specifically, had not been previously investigated, it was not clear in such cases whether children might focus on unimportant story details and fail to learn the science information embedded in the story. However, the focus of this study was not on structure alone. As Horowitz & Samuels (1987) point out, since structure and content cannot be separated, one should seek an effective match of content and structure. Science content is often difficult for children to learn because it is not part of their everyday experiences. Their expe.iences may support a misconception rather than a scientific explanation, for example, a child who experiences getting warmer by getting closer to a source of heat (a fireplace or a radiator) may use that experience to reason that it is warmer in summer because the earth is closer to the sun. A refutation text that directly refutes this misconception may create cognitive dissonance, but children often cope with this dissonance by considering real life and scientific knowledge as separate and unrelated (Dole & Smith, 1987) The decontextualized and impersonal way in which science information is presented even in a considerate refutation text may foster this separation of scientific knowledge and experience. Presenting scieme information that contradicts children's misconceptions in a story about children who confront the contradiction in attempting to answer a real life question may help to overcome this separation. The soft expository text in this study focusI on the problem experienced by Katie, a little girl who had movei from A. -trona to New York. Katie realizes that explaining seasonal change by saying that the earth is closer to the sun in summer and farther away in winter can't be ,rue if it is summer in July in New York and winter in July in Australia. She convinces her friends to seek another explanation which will resolve this contradiction. Thus the correction of the misconception in this text was the focus of the story. In addition to considering the effect of different types of texts, a second purpose of this study was to determine whether, before reading the texts, the percent of seventh graders who had the misconceptions about the topic of seasonal change would be smaller than the percent of firth graders with the misconception. Another question was whether those seventh graders who did have misconceptions would correct them 0,6*4#Y- to ti*ve rese.4e4:*sisie of Sb.s*Set-9PeTstiOes.*.e.',- betteç able to ke,t1re,et' ir,*isCOee*011s about ab. str.41,4 scic 004 P0144444' at A:i?gpcfo*ppotOott,t-tago.,Dobo.otti-er,hind . reVeiled:thit Contrary tci piagethin 'theory,- Minya-410 ai *ell aiChildrentold, ---dfie',adaCenntephont. (Linn, '198.6): A. final question was wh.ithói fifth an&sev aders :w.041.4 be #ffei:ctC4': diff0ePtlY1* the Mick* Yti# of' texts, soft eiPositOry text be Mere, facilitative for :,,bOth ,flith.!aria Say seventh graders -who ilaye Morn experiende.with regular ei,tOi type as helpful' in correcting their misconeeptions'ai the,SofteXPOsitalyieXt: .-- METHOD Subjects All seventh-grade students in the gifted and talented programsiniwo school systems (N=123) and all fifth-grade students in the, gifted aati.taleatiickpror grams of one of these school systems (N=129) were tested to deteanine Whether tlitY had a common Misconception about seasonal change, that is, that it is' hotter because the earth is closer to the sun and colder-in winter becatire tkiiaitkiifaittles away from the sun. Since standardized reading scores overall were low in theth systems, children were chosen from gifted programs to reduce the incidence of decoding problems. In both school systems chilu..,in were chosen for the gifted and tatentedprograrn on the basis of teacher recommendation and performance on an IQ test ind a teit of achievement. Hnwever, in both systems cut off points varied. Total reading pereentile scores on the Degrees of Reading Power (DRP) Reading Test for New Yark State Elementary Schools, Grade 6 (Touchstone Applied Science Associates, 1988) for the seventh graders who had this misconception ranged from 49 to 99. Total reading percentile scores on the Metropolitan Achievement Test 6, Elernentary Form M (Balow, Hogan, Farr, & Prescott, 1985) of the fifth grade s who had the misconeeption ranged from 28 to 99. The children were from diverse ethnic backgrounds and from lower and middle class socioeconomic levels. Unlike subjects in the Maria (1988) vtudy, these children had already received science instruction related to the scientific reasons for seasonal change. This topic was part of the fifth-grade curriculum so that the fifth graders had received this instruction 4 months before being pretested, whereas the seventh fraders in the same school had studied the topic 2 years before with the same teacher. This teacher's instruction provided concrete experiences for the children; for example, they used balls and flashlights to demonstrate how the tilt of the ball affected the slant of light rays. However, he did not make any reference to the targeted misconCeption in his inftruction, and after instruction 64% (N=82) of the fifth graders had the misconception. In the same school 97 seventh graders were tested for the misconception. Since, only 42% of them (N=41) had the misconception, it was necessary to test seventh gmders from the gifted and talented program of the other school district to ensure sufficie-n 332 Literacy Theory and Research subjects in each of the groups. Eighteen seventh graders were tested in this second program and 50% had the misconception (N=9). They too had received science instruction on the topic of seasonal change in fifth grade but with a different teacher. Materials The inconsiderate expository text (1E) was a 760-word sectiona with fourth-grade Raygor readability score (Raygor, 1977) taken from the fifth-grade book of the Holt Science Series (Abruscato, Fossaceca, Hassard, & Peck, 1986). It considered the topic of seasonal change but made no reference to the misconception. The considerate expository text (CE) was 1,075 words long and had a sixth-grade Raygor readability rating. It was constructed by us according to Armbruster's (1985) guidelines; for example, information was built step by step and key ideas were repeated and highlighted by use of bold print and spacing. This text also discussed the common misconception and directly refuted it. Both the IE and the CE texts, which also had been usPA1 in the Maria (1988) study, highlighted and defined scientific vocabulary such as revolution and summer solstice. The considerate soft expository text (CSE) was constructed for this study. It was 1533 words long and had a fourth-grade Raygor readability score. It focused on the same key scientific concepts as the other two texts but in the context of a story in which a group of children sought the answer to Katie's puzzling question. "Why was it summer in July in New York and winter in July in Australia?" Although it did contain key vocabulary, there was less emphasis on definitions since this emphasis did not fit the narrative nature of the text. All three texts contained diagrams noting the earth's position in relation to the sun. The diagrams in the CE and CSE texts also noted distance of the earth from the sun. Instruments Three tests were asz.; A 12-item vocabulary multiple-choice test was used as one pretest of prior knowledge about the topic of seasonal change. This test included all 7 technical vocabulary words relating directly to seasoc..: change defined in the 1E and CE texts Definitions of more general words such as tilt and slant were specifically related to the topic of seasonal change. A misconception test, which was given as a pretest, immediate posttest and delayed posttest, was a 10-item multiple-choice test developed by Marshall (1987) for use with prese-vice teachers Every item on this test contained a distractor related to the misconception that the earth was close to the sun in summer and farther away in winter Several slight revisions were made to make the test more suitable for children. These two tests were also used in the Maria (1988) study. An application test, which was given as an immediate and delayed posttest, was developed for this study It contained two questions. ln the first, children were presented with four diagrams showing the position of the earth in summer and winter and the distance of the earth from the sun at those seasons. On Iv one of the diagrams was correct; the others were incorrect either because the tilt of the earth was pictured incorrectly or the distance of the earth from the sun was incorrect. The children were directed to choose the correct diagram and write an explanation for their choh.e. The 334 333 Correcting Misconceptions second question asked children for the scientific explanation of why it is warm in summer and etAd in winter. Design and Procedure Pretesting with the vocabulary and misconception multiple-choice tests took place in class groups and was carried out by the classroom teachers. Children who, scored more than 5 on the misconception test were not considered to have the misconception and were not used as subjects. Since a 2 x 3 (Grade x Type of Text) design was used in the study, the 50 seventh graders and the 82 fifth graders who had the misconception were randomly assigned to three groups: Inconsiderate Expository (IE), Considerate Soft Expository (CSE), and Considerate Expository (CE). One month later the children read the texts and were again tested in class groups. Several children who had the misconception were absent, resulting in a final total of 47 seventh-grade subjects and 75 fifth-grade subjects. One month later, classroom teachers administered the misconception test and application test to the children again. Forty five seventh-grade and 66 fifth-grade subjects were present for the delayed testing. RESULTS In determiLing whether the groups in each grade differed in reading, separate ANOVAs were computed since each grade had taken a different test. Total Reading DRP NCE scores were available for 42 seventh graders and total reading NCE scores from the MAT 6, Elementary Form M were available for 70 of the fifth graders. Reading scores did not differ for either the seve. graders (F(2, 41) = .31, p<.74) or the fifth graders (F(2, 69) = 2.77, p<.07). Table 1 below gives the means and standard deviations for the three conditions in the two grades. Prior knowledge was determined by the scores on the vocabulary and misconception pretests. Two-way ANOVAs were computed for each of these dependent measures. Vocabulary pretest scores (Grade 7: IE M=7.27, SD = 2,22, CSE M=7.12, SD =2.06; CEM= 8.31, SD= 1.32; Grade 5: IEM= 6.74, SD= 2.16; CSEM=7.89, St) = 1.67, CE M= 6.32, SD = 1.87) did not differ for grade (F(1, 118) = 1.97, p<.17) or condition (F(2, 118) = 1.50, p<.23). Misconception pretest scores also did not differ for grade (F(1, 119) = .23, p<.64) or condition (F(2, 119) = .33, p<.72). M;sconception pretest means and standard deviati ...s are found in Table 2 along with immediate and delayed post test means and standard deviations. Separate two-way ANOVAs were also computed for each of the posttest measures (the immed,ate and delayed misconception and application tes.$). There was a diffet ence between grades (F(1, 118) = 17.42, p<.001) and conditions (F(2, 118)=23.29, p<.001) on the immediate misconception posttest. Seventh graders (M =7.83, SD =1.80) scored higher than fifth graders (M= 6.72, SD = 1.91) and a Scheffe test (p<.001) indicated that children who received the CSE text and those who received the CE text scored higher than those who received the IE text. On the delayed misconception posttest, there was no difference between the two grades (F(1, 104) = 3.55, ' bat Table t Veans (and Standard Deviations)'af Total Reading NCE Scóres far,Drent4eit, Condfflôiñ GradesS and 7 c?,ti coOftioly Coasiciffite:S5ikEiosit031 (CSE) 6nskleritellthiY (CE) `69.i9 'OVA - 67.46 (15:03) 58.0,2-(1455)- Table 2 Means (and Standard Deviatians) of Misconception Test Scores in DifferentIc41 Conditions and at Different Times of Testing for Grades Sland 7 Tittle-of rTeSt.hig Text Condition Grade 5 Inconsiderate Expository (1E) Considerate Soft Expository (CSE) Considerate Expository (CE) Grade 7 Inconsiderate Expository G Considerate Soft Expository (CSE) Considerate Expository (CE) hniazikel Pretest PosteSt 3.84 (1.17) 3.48 (1.16) 3.25 (1.48) 5.11 (.94) , 7.70 (1.61) 6.86 (1.98) 5.21-(2,4), ,7.41412,14: 5.58 (1.73): 3.27 (1.03) 3.29 (1.36) 3.64 (.93) 6.47 (1.73) 8.65 (1.00) 8.29 (1.90) 4.56 (2.00)' ; 5.4612:23): 5.62 (2,09): pr 06) but there was a main effect for condition (F(2, 104)=5.52, p<.005). Scheffe test (p<,01) indicated that once again children who received the CSEJteict: scored higher than those who received the IE text. However, there was no diffeienCe between the CE and IE texts. There were no interactions on either, the,iniffiediate misconception posttest (F(2, 118) = .16, p<. 86) or the delayed misconeePtion pCstteat (F(2, 104) = 1.24, p<.30). A MANOVA indicated a main effect for tithe-of (F(2, 198) = 150.45, p<.001). Paired tests indicated that child:en Scored higher-on, the immediate micPonception test than they did On the delayed misconception test 0(104) = 7.03, p<.001) and higher on the delayul test than they did on the pretest' (to Go .9.32, p<.001). Scores on the application test ranged from 0 to 9. Otildren received 1 point for choosing the cecrect diagram. Their explanadons were scored for presence of critic& ideas which were part of the scientific explanation of seasonal change. Cnildren alio received 1 point if they had no statement of the misconception in any of their explanations In independently scoring the application tests aecording to this scoring system, there were only 5 disagreements which were resolved by discussion. ne immediate application test differed by grade (F(1, 118)=5.87, p<.02) ancL condition (F(2, 118) = 18.32, p<.001), but there was no interaction (F(2, 118) = .23, 'OWE , isoofic ns eans (and Standard Deviations) of Appfication-Test Scores in Different Text onditions'and:cit Dif*ent Times of Testing for pt.ades 5 ,ant1:7- Text Condition 5 11:514er*MiPositotY (IE) considerate Soft Expository (CSE) Coasiderate Expository (CE) Pra,:de 7 Inconsiderate Expository (LE) Coesiderate Soft Expository (CSE) Considerate Expository (CE) 1.76 (1:31) 4.15 (1.86) 3.29 (1.98) .V.811(1:57)' .30 (1.44) 2.80 4.00, (2.63) 128 (2.29) 4.18 (2.37) 3.4009) 1:98'(12.63) 3454;99) p<.80). Seventh graders (M=3.94, SD= 2.39) scored higher than fifth gradets (M =3.21, SD = 1.99). Scheffe tests at p<.001 indicated that children who fetid the CSE text did better than those who read the LE text. However, again there,wai no difference between the CSE and the CE texts nor between the CE and'IE The delayed application test also differed by grade (F(1, 104) =.60, p<.02) and condition (F(2, 104)=5.80, p<.004). Seventh graders (M=3.34, SD=2.22)`scoredhigher than fifth graders (M = 2.51, SD =1.94) and a Scheffe test (p<.05), indieated that children who read the CSE text scored higher than those who read the CE text or the TE text. Once again there was no interaction (F(2, 104)= .36, p<30). A MANOVA indicated that children in both grades scored higher on the immediate test than on the delayed test (F = 5.83, p<.02). Table 3 contains the means and standard deviations for the immediate and delayed Application Tests. DISCUSSION All the measures indicated that the subjects had learned the scientific explanation of seasonal change befter with the considerate soft expository text (CSE) than with the inconsiderate text (1E). However, only the delayed application test indicated better learning with the considerate soft expository text (CSE) than with the considerate expository text (CE). That the difference showed up on the delayed test, however, sugpsts that the considerate soft expository text may make the information more memorable. The consistently better scores achieved by children who read the consider- ate soft expository text (CSE) suggests that further studies should be done using this text type to present science information, particularly information that corrects misconceptions. In contrast with the results of the Maria (1988) study, the considerate expository (CE) text was only better than the inconsiderate expository (1E) text on one measure, the immediate misconception test. Perhaps the children's previous instruction on the 336 Literacy Theory and Research, topic was responsible for this finding. Future studies s:ryild directly compare children who have received instruction with those who ha ve not. Seventh graders did differ from fifth graders at the outset of the study and after reading the texts. Fewer seventh graders (42%) than fifth graders .(64%) had the misconception despite the fact that fifth graders had studied the topic only a few months lefore and seventh graders had studied the topic 2 years before. Of course, many things may have happened to the seventh graders in those 2 years which may have consolidated the learning for them. Seventh graders learned the new information more easily than the fifth graders as measured by three of the four dependent measures. Why the delayed misconception test should have a different pattern of results is hard to explain. The fact that the considerate soft expository text was more helpful for both seventh and fifth graders raises the question of whether presenuot, ci...ract science information that contradicts misconceptions in a story about a real-life problem-solving situation may be helpful for older children and adults. This is another question that should be tested in future studies. REFERENCES Abniscato. I . Fossace:a W Hassard. 1 & Peck, D. (1986). dolt Science Level 5. New York. Holt, Rinehart & Wintscn. Alvermann, D , & amd, C R (1989, November) The influence of discussion, demonstranon and text on the learning cr; counterintuitive science concepts. Paper presented at the meeting of the National Reading Conference, Austin. TX. Armbruster, B (085). Content area textbooks. A research perspective. In I. Osborn, P. T. Wilson, & R C Ander:on (Eds ), Reading education. Foundations for !aerate America (pp. 47-60). Leungion, MA: Lexington Books. Balow, 1 H , Hogan, T P , Farr, R C., & Prescott. G. A. (1985). Metropolitan Acluevemem Tests (6th ed.). San Antonio, TX: Psychological Corporation. Cole, J. (1986 The magic schoolbus at the waterworks. New York. Scholastic. Degrees of Reading Power Test (1988). New York: The ColIrcA Board. Dole, 1 A (1989. November) The effects of refiaation an.4 constderate texts coi learning concelaually eiuy and difficult science concepts Paper presented at the meeting of the National Reading Conference, Alstin, TX. Dole. 1 A , & Smith. E L (1987, December). When prior knowledge is wrong. Reading and learnins from science text Paper presented at the meeting of the Nauonal Reading Conference, St. Petersburg, FL. Englert, C S , & Hiebert, E (1984) Children's developing awareness of text structures in expository material',. Journal of Educational Psychology, 76, 65-74. Freedle, R , & Hale G (1979) Acquisition of new comprehension schemata for expository prose by transfer of a naintive schema In R. Freedle (Ed.), New directions in discourse processing (Vol. 2, pp. 121-135). Norwood. NJ: Ablex. Gamer, R . Gillingham, M G , & White. 1. (1988). Effects of "seductive details- on macroprocessing and microprocessing in adults and children. Unpublished manuscript. Hidi S , & Baird, W (1988) Strategies fa mcreasine tcxt-hased interest and students' recall of expository texts. Reading Research Quarterly, 23, 465-483. Horowitz, R , & Samuels, S (1987) Rhetorical structure in discourse processing. In R. Horowitz & S. 1 Samuels (Eds ). Canorehending oral and written language (pp. 1-52). New York. Academic Press. Linn, M C f1986) Science In R F Dillon & R 3. Strtnberg (Eds.), Cognition and instruction (pp. 155-204). Orlando: Academic Press. 338 :§ Correcthtg Misconceptions It,fandler, J., 4ft Johnson, N. (1977). Remembrance of things parsed: Stacy structure and recall. Cognitive 111-151. (1981pCcember). He4sing fdth graden learn with science text. paperlitsented-at therneeting of the Natioitil Reading Conference. Thcson, AZ. Marshall,"N: (1987, Dicember) . IAA text faili so meet reader cOectatiOns ;Ps* Ins**1 at ti.le meeting of the National Reading Conference, Si.,Petenbnrg, FL. Atethn1cfou1ngatchernaasatebon :Pincus, A. R. It GcBer,- EA3`. ,-& Stover,' E M ugni..491: Y to understanding arid summarizing event based magarIne articles. :fair/kat of Resâltg i-49, :Raygor, A. L (1977): the Raygor readabUity estin*:-A qUicir al;d eaiy waitor:rfere.. ,P. D. Pearson (Ed.);/1e#Ing:, TheitY, reseairh, and practice (Pp. 259-263);;PvientYSlith Yrasbc of the National Reiding Conference, Clans" SC. Richgels, D. J., McGee, L. M., LOmax, R. G., &Shard, C. (1987). Awaishess orfoar teit strucfniei: Effects on recall of expository text. Reading Releatch Quarterly, 22, 177-196. Stein, N., & Glenn, C. (1979). An analysis of story comprehension in elementary school Children. In R. Freedle (Ed.), New &rations in discourse processing (pp. 53-120). Norwood, Ni: Ablex. Touchstone Applied Science Associates (1988). Degrees of reading power (DRP) let t for ihe ?kw York State elementaty schools. Brewster, NY: Author. 339 ,, - . , .-. , ' ' ,*.- I T , t , . , 'PM O'ER:En:KO -wgRitatICNOWVEDavyiig_SEOF OAPRiOORGANIZERkOSLCaLEOE:DEVRIMWEOAL 1,;.E*DER'4' SUMWRIZATIPN'JAIW.C. 910.1**NRPN'Pf- ilPEPO$0070ct Ernest BalsOthy State University of New York at Geneseo , Renee Weisberg Beaver College . Though often recommended to students as a cognitive !Canting strategyoiso,Of graphic organizers has received little empirieid examination .(Holler kpinSeretiik 1984). Research has recently confrrinedihat trating in use of,graphic orgarriZers beneficial effects-on Sturients' comprehension (literkowiZ 1986;.:cuii-Rozel*iti 1989) 4,141,ities (Weis, az kBalaithy, PBP: In 44.09- *Wei*, . sucii itudies often haVe used artificially copstraqed texti with Consistent Oigani.Zatiotyl patterns. Real-life material encountered by stUdents ismrely so well-niganiii4Cilittlert & Tierney 1081), and queitions have:been raised 4s to the praetteal, trans." fer benefits of such training (Hare, Rabinowitz, & gchiebie, '1089). Thepurpnse of this research was to investigate the transfereffects.of trainingin the use of graphic organizers and summuy writing on college divei6pMenteireadera" recognition of the compare/contrast text structure. Content area textbook:Miter:Isis, were used in whielrthe organizational structure is not as clearly, ippaient artificially constructed passages used for training The researchesi,songht :to rfeternrine` whether these less able readers could use the strategies they ha beentaught tOrecbgnize the specific text structures with which they, had been successful during training. The students' ability to tr.:seer training was examined in the light of their prior knowledge of the passage's content. An earlier study of similar design (Weisberg & Balajthy, 1989) used a younger and less able population of remedial high School students. Results indicaieet- that transfer of training did occur on measures designed to determine recognition of text structure and to assess summary writing ability. This held true for hoth moderate andlow prior knowledge passages. For the comprehension 'assessment, hoWever, the trained group outscored the controls only on a mOdenue, not on alow, prior knoWledge passage. Tbe researchers suggested that prior knowledge on the latter passage Was so low that subjeets could not bring their strategies to bear in effeetively improving compretension. In the present study, as with the earlier study, these effects were examined using low and moderate prior knowledge passages. 339 34 0 340 Subjects The study was conducted at a northeastern state college of liberal arts am sci7 ences. Subjects were a college freshman population Oaz.645) required to take .ntlevek opmental reading/study skills course. Three groups of studerits.were **Oa in the developmental course: (a) Educational Opportunity Program (BOP) 'tticIents; !!!': mitted to the college based on low high school performance.and orkotionale nee4; (b)-Talented Opportunity Program (TOP) studenu, admitted through a special-,Prograin to encourze admission of minority students; and (c) Special Talent Athletes (STA) students, almitted because of exception athletic ability, despite poor high school performance. Subjects were ra4dotrly assigned n experimental or control group. Due to absences, only 52 students were inclucLu in the final analyses. The mean score on the comprehension subtest of the Stanford Diagnostic Reading Test (Karlsen, Gardner, & Madden, 1984) for the raxperimental group was 49.97 and for the control group 51 53 The overall mean score for both groups combined was 50.78, corresponding to the 41st percentile and to a grade level of 11.3. Procedures Pretest Prior to training, all subjects were administered a true-false test of prior knowledge on several topics, including the two to be included in the posttest passages. Ten questions were included for each topic. The pretests were used to verify which passages presented topics associated with low and with moderate prior knowledge i'or these subjoets. Training Instruction was ;entered on a collection of eight readings consisting of scientific expository text, each of veh:ch had a comparison-contrast internal organiza- tion The compare/contrast text structure presents special challenges for less able readers (Englert & Hiebert, 1984; Rafael & Kirschner, 1985; Richgels, McGee, Lomax, & Sheard, 1987), who have difficulty summarizing even easier text structures (Head & Buss, 10S7). Henk and Stahl (1989) have verified that even college developmental readers hxle difficulty comprehending this structure. Five of the readings were taken from science textbooks and were adapted to reflect tightly constructed organizational patterns The three other passages were taken directly from the textbooks and, though they did have a central comparison-contrast pattern, the organizational structure was not as clearly presented. Students in the experimental group were trained by one of the researchers to follow this basic procedure: (a) Read the passage to identify topics and categories of comparisons, (b) use underlining and annotation to identify and organize comparisons and contrasts, (c) using telegraphic writing, complete a graphic organizer (see Figure 1 for an example of a student's graphic organizer), and (d) incorporate thc comparisons and contrasts into a summary statement. Instruction included explicit rules and modeling for constructang graphic organizers and writing summaries. Experimental subjects received four training sessions of 341 Prior Knowledge and Graphic Organi: r Transfer Organic Compound* In Living Things I Protein Carboholrafes Vits..4ins - Hydrogen Oxygen Nitrogen Carbon Meat fish From good Woe end beans For hemoglobingives blood red color ener.gy also Rings dist Chemical fre In body Car, n - Carben- 21emants Hydroggn OxYZell Ofnfeller Examples Fried foods Boger Vouch Sias* Source Uses energy HYdrogea of energy Figure 1. Sample student graphic organiztr. 40 minutes each over a 2-week period, as well as shot. homework assignments for three of the sessions. Initial training was carried out through whole-group instruction and modeling in the first session. The last three sessions followed this schedule: I. The instructor collected homework and briefly modeled the process of constructing a graphic organizer and a summary for the homework passage. 2. Students were divided into cooperative learning groups of three to increase motivation and group problem solving. 3. Each group was given the same reading assignment find told to study as a group for a possible posttest. Each group was required to annotate passages, then construct a graphic organizer and a summary. Work was monitored by the instrictor. 4. Graphic organizers and summaries were collected. A short answer fill-in comprehension test was administered on one occasion. 5. The passage was discussed, and the instructor modeled construction of thc graphic organizer and the summary. The control group was not entirely untrained in the procedure During the four training sessions for the experimental group, the control group carried out various comprehension-oriented activities that did not deal with either graphic organizers or companson-contrast relationships. They did, however, receiv a one-half hour presentation that introduced them to comparison-contrast graphic organizers and summaries so they could complete the posttest. Posuests. Both groups were administered a posttest consisting of two scientific comparison-contrast passages. One passage was directly taken from Jantzen and Michel (1986) and the other from Heimler and Price (1981). The passages were administered in counterbalanced order to eliminate effects of order. One of ex transfer pas- 342 17, 342 Literacy Theory and Rest* It-, sages, titled "Organic Compounds," was designated as a moderate prior knowledge passage based on pretesting described above. The other passage, titled "Types of Fish," was designated as lo% prior knowledge. Subjects were instructed to read and annotate the passage. They then constructed a graphic organizer and a sammary All materials were collected and the subjects completed an immediate retention multiple-choice test of 10 items. The qUestionS,, which addressed all major comparisons in each passage, were created by the ri%eami- ers and verified for passage dependency by them and by the students' classmom instructors. Scaring and Data Analysis Prior to administering the posttest, the researchers and their assistants had created a master template of the comparison-contrast idea structure for each posttest passage. This w is carried out by parsing the text into idea units and constructing a grid o corn: Arisons and contrasts within the passage. Subjects' posttest graphic org . and summaries were separately scored against the template. The graphic organizers and summaries were evaluated separately by two persons (the researchers and/or their assistants). Each subject's score was the percentage of items on the master template which had been included. Minor differences were resolved in discussion. RESULTS Results were analyzed using MANO`rA with three dependent task variables: (a) graphic organizers, (b) summarizing, and (c) comprehension. The between-subjects factor was Group (trained and untrained). The within subjects factor was Prior Knowledge (low and moderate). (See Table I .) The main effect for Group was statistically significant, multivariate F(1, 51)=10.88, p< .01. The effect for Prior Knowledge was also statistically significant, multivariate F(1, 52)=42.87, p<.0001. In addition, the effect for Task was significant. multivariate F(2, 102)=196.84, p<.0001. Two interactions were significant, for Group by Task, multivariate F(2, 102)=3.77, p<.05, and for Task by Prior Knowledge, multivariate F(2, 102)=9.98, p<.0001. The experimental group mean across all tasks was 62.30%; the contiol group mean was 55.49%. Univariate analyses indicated that the experimental group outscored the controls in both grRphic organizer, F(1, 51)=13.40, p<.001, and summa- rizing F(I, 51) = 6 25, p< .02, scores. Effect sizes, calculated according to procedures described in Cohen (1977), were .84 and .48, in the large and moderate ranges respectively There was no significant difference on the comprehension test. Combined experimental and control group mean scores across all tasks were 64.80% on moderate prior knowledge passages and 51.93% on low prior knowledge passages Individual univariate analyses indicated significant differences between prior knowledge conditions for both summarizing and comprehension, F(1, 43) = 15.54, p<.001 and F(1 , 43) = 30 31, p<.001. Moderate prior knowledge scores for summarizing and comprehension were 16.51 and 17.63 percentage points higher than for 343 3 343 Prior Knowledge and Graphic Organizer Transfer Table 1 Mean Percentages (and Standard Deviations) by Group Graphic Organizer Summary Comprehension Yest Experimental Group (n=26) Low Prior Knowledge Moderate Prior Knowledge 68.96 (17.64) 74.23 (14.36) 25.32 (24.83) 44.63 (25.24) 71.28 (13.99) 87.81 (12.11) Control Group (n=26) Low Prior Knowledge Moderate Prior Knowledge 56.91 (19.25) 60.61 (16.94) 19.75 (21.30) 33.46 (15.35) 69.33 (19.28) 88.06 (13.27) low prior knowledge, with effect sizes of .95 and 1.45, both in the large range. The difference for graph.c organizers was only 4.48 points, and the univariate analysis indicated no different. DISCUSSION The results of the present study validate previous research with elementary and secondary grade poor readers suggesting the benefits of instruction in strategies that direct readers' attention to text structure (Weisberg 4 Balajthy, 1989, 1990). As in the previous study with secondary readers (Weisberg & Balajthy, 1989), these developmental college students were able to transfer their training in the use of graphic organizers and summaries to real-world textbook materials. This ability is critical for effective comprehension and retention of content area material. Subjects in the experimental group did not obtain higher comprehension scores than those in the control group. In the study at the secondary level (Weisberg & Balajthy, 1989) with similar results, the authors had suggested that the prior knowledge of the topics (mean of 31.88% comprehension) in the passages had been so low that subjects could not bring their strategies to bear in effectively improving memory of the passage. In the present study, a different explanation might be offered. The overall mean comprehension score was 79.12%, a high score considering that prior knowledge of the topics was low to moderate. This suggests that comprehension ability of the subjects was fairly high, a suggestion that is verified by examination of the mean comprehension ability of the subjects on the Stanford Diagnostic Reading Test. A previous study by the authors (Balajthy & Weisberg, 1989) had found that training in generative learning strategies has greater effect on the comprehension of poorer college readers than better. It may be that the highei _omprehension ability of the subjects in the present study was not readily amenable to improvement. The main effect for prior knowledge was expected, since readers comprehend (Afflerbach, 1986; Balajthy & Weisberg, 1989; Johnston, 1984; Weisberg & Balajthy, 1989) and summarize (Pratt, Luszcz, McKenzie-Keating, & Manning, 1982; Weisberg & Balajthy, 1989) higher topic familiarity passages better than passages with low topic familiarity. An examination of the task by prior knowledge interaction indicated, 344 344 Literacy Theory and Research however, no significant differences due to prior knowledge on performance in the graphic organizer task. It may be that these college readers, who am more skilled than most subjects in the stud'ts cited above, are betel able to recognize and diagram text structure, even when the content is less familiar. Although this finding merits further investigation, the possibility that use of graphic organizers may be a generative strategy relatively unaffected by level of prior knowledge for college developmental students is an issue of potential importance for cognitive learning strategy research and teaching. Previous research by the authors (Weisberg & Balajthy, 1989) has suggested that lower prior knowledge passages are more amenable to instructional effects than higher prior knowledge passages. The present study did not substantiate these earlier findings. Differences in results between the present and earlier studies may be due to the population, to differences in levels of prior knowledge on passages, or to a combination of both Additional research might investigate whether the effects of training for college students are less affected by prior knowledge Ian for the younger tercters in earlier studies. Although the results of this study offer strong idence for the usefulness of generative learning strategies, and for the transfer of training in these strategies to real-world textual materials, the limitations of this transfer must be noted. The actual transfer passages were similar to the adapted training passages in a variety of ways and the readability levels were similar. Lengths of passages varied from 400 to 900 words, but many real-world tasks involve reading selections which are much longer. The training was carried out using a specific text structure, the comparison-contrast structure, and both the adapted and transfer passages employed that stricture. Whether training in one text structure will transfer to another structure is an issue not addressed, nor is the issue of transfer between content areas. As Tobias (1987) has noted, there is little existing evidence that learning strategies tansfer across content areas. Informal conversation with the subjects in the study showed that many appreciated the learning strategy They suggested that the strategy made them aware of authors' use of the cumparison-contrast structure, so they would be more likely, to spot its use in textbooks and employ their already well-developed ability to analyze the structure of concepts presented They also suggested that the training convinced them of the importance of using text structure as a tooi in retaining information, and they appreciated the usefulness of the graphic organizer as an effective method of spatially reorganizing information from text. REFERENCES Affieqbach, P (1986) The influence of prior knowledge on expert readers importance assignment processes In 1 A Niles & R V Lalik (Eds ), Solving problems of literacy. Learners, teachers, and researchers (pp. 30-39). Rochester, NY: National Reading Conference. Balaithy, E & We'sberg, R (1989, October) Effect of transfer to real-world subject area materials from training in graphic organizers and summarizing on developmental readers comprehension of the compare Icoi.trast text structure in science expository text. Paper presented at the meeting of the College Reading Association, Philadelphia (ERIC Document Reproduction Service No. ED 300 771) 345 Prior Knowledge and Graphic Organizer Transfer 345 Berkowitz, S. J. (1986). Effects of instructich: ia ma organization on sixth-grade studcnts' memory for expository reading. Reading Research Quarterly, 21, 161-178. Cohen, J. (1977). Statistical power analyses for the behavioral sciences (2nd ed.). New York: Academic Press. Enght, C. S & Hiebert, E. H. (1984). Children's developing awareness of text structure in expository material. Journal of Educatioxal Psychology, 7tr, 65-75. Gnri-Rozenblit, S. (1989). Effects of a tree diagram on students' comprehension of main ideas in an expository text with multiple themes. Reading Research Quarterly, 24, 236-247. Hare, V. C., Rabinowitz, M., & Schieble, K. M. (1989). Text effects on main idea comprehensiim. Reading Research Quarterly, 24, 72-8Z. Head, M. H., & Buss, R. R (1987). Factors affecting summary writing and their impact on reading comprehension assessment. In J. E. Readence & R. S. Baldwin (Eds.), Research in literacy: Merging perspectives (pp. 25-33) Rochester, NY: National Reading Conference. Heimler, C. H. (1981). Focus on physical science. Columbus, OH: Merrill Henk, W. A., & Stahl, N. A. (1989). Companson-contrast text structures and the college developmental reader. Journal of Reading, 32, 494-499. Holley, C. D., & Dansereau, D. F. (1984). Spatial learning strategies. Orlando, FL: Academic Press. Jantzen, P. G., & Michel, J. L. (1986). Life science. New York: Macmillan. Johnston, P. (1984). Prior knowledge and reading comprehension test bias. Reading Research Quarterly, 19, 219-239. Karlsen, B., Gardner, E. F., & Madden, R. (1984). Stanford Diagnostic Reading Test, Blue Level (3rd ed.). Cleveland, OH: Psychokigical Corporation. Pratt, M. W., Luszcz, M. A., MacKenzie-Keating, S., & Manning, A. (1982). Thinking about storiei: The story schema in metacognitiou. Journal of Verbal Learning and Verbal Behavior, 21, 493-505. Rafael, T. E., & Kirschner, B. M. (1985, April). The effects of instruction in comparelcontrast text structures on sixth-grade students' reading con; vhension and writing predictions. Paper presented at the meeting of the America.i Educational Research Association, Chicago. Richgels, D. J., McGee, L. M., Lomax, R. G., & Sheard, C. (1987). Awareness of four text structures: Effects on recall of expository text. Read sg Research Quurterly, 22, 177-196. Schallert, D. L., & Tierney, R. J. (1981, December). The nature of high school textbooks and learners. Overview and update. Paper presented at the meeting of the National Reading Conference, Dallas, TX. Tobias, S. (1987). Learner characteristics. In R. M. Gagne (Ed.), Instructional technology. Foundations (pp. 207-232). Hillside, NJ: Erlbaum. Weisberg, R., & Balaithy, E. (1989). Transfer effects of instructing poor readers to recognize expository text structure. In S. McCormick & J. Zutell (Eds.), Cognitive and social perspectives for literacy research and instruction (pp. 279-286). Chicago, IL: National Reading Conference. Weisberg, R., & Balaithy, E. (1990). Development of disabled readers' metacomprehension ability through summarization training using expository text. Results of three studies. Journal of Readina, Writing, and Lean. ig Disabilities, International, 6, 18? -202. Weisberg, R., & Balaithy, E. (In press). Improving disabled readers' summarization and recognition of expository text. In T. Rasinskr (Ed.), 1990 Yearbook of the College Reading Association. Kent, OH College Reading Association. 346 WHAT DETERMINES COURSE ACHIEVEMENT? AN INVESTIGATION OF SEVERAL POSSIBLE INFLUENCES ON ACADEMIC OUTCOMES ALn, J. Pace, Karol Walters, and John K. Sherk, Jr. University of MissouriKamas City The literature on studying presents something of a para..1cm. On the one hand, both researchers and practitioners frequently offer the seemingly reasonable proposal tat by using appropriate study stratekies and practices, students should be able to improve scholastic performance. (Rohwcr, 1984). The efficacy of various procedures appears supportable by extant theory, laboratory research, and the informed observations of school and college instructors. On the other hand, it has been diffieult to demonstrate a clear and unequivocal relation between particular studying stralegies and real-world course achievement (Schumacher, 1987). This picture is supported by studies such as those by Nolen (1988), Pace, Peck, and Sherk (1986), and Pace, Sherk, Peck, and Baldwin (1985), which found no statistical relationship between self-reports of students' studying practices and measures of academic achievement. Despite the limitations of such procedures, r consensus seems to be developing that the complexity of actual learning situations, especially the variety of contextual factors that influence individual course outcomes, limits the probability that particular study strategies will produce demonstrable effects. Schumacher (1987) recently has underscored the importance of context for understanding the role of studying in learning. Other analyses of academic studying, such as those by Biggs (1984), Nolen (1988), and Thomas and Rohwer (1986), emphasize that a number of personalogical and situational factors, such as motivation and course expectations, influence strategy t se and academic outcomes in complex ways. Addi- tionally, previous academic acifievement, as reflected, for example, in cumulative GPA, and academic aptitude, such as SAT scores (Schuman, Walsh, Olson, & Etheridge, 1985), are known to be among the best predictors of future academic performance. Prior subject matter knowledge is also a crucial determiner of later success in courses in that field. Further, Schuman, Walsh, Olson, and Etheridge (1985), in a series of large-scale investigations with college students, consistently found that class attendance could better explain variations in subsequent course grades than hours studied. Thus, as Thomas (1987) has argued, a single linear model of "good" studying practices may be inappropriate, and researchers need to look much more closely at the contextual or situational factors that affect students' performance and perceptions in particular courses, as well as instructors' expectations, to derive an accurate picture *A the factors influencing acack.nic achievement and individual studying decisions rr,:br 347 3247 348 Literacy Theory and Research The study described here is an initial effort in this direction and was designed to obtain a picture of the variety of factors that affect course achievement within specific college courses. METHOD Subjects and Setting To accomplish this aim, the cooperation of two instructors (A and B) of the same introductory college economics course was obtained. This course was held during the fall semester of the academic year. Initial enrollment was 93 (52 female and 41 male) in A's class and 72 (35 female and 37 male) in B's; 80% of the students in both classes were freshmen and sophomores. Fewer than 10% of the students in either class repiesented ethnic minorities. Students in each class were given a general explanation of the purpose and scope of the project, and their participation was solicited. Those students who agreed to participate were asked to give the investigators permission to obtain their cumulative GPAs and, if possible, their scores on a measure of mathematics aptitude, such as the SAT-M or ACTMath. Students were also asketd to allow their course grades and examination scores to be made available to the investigatcrs. In A's class, the mean GPA, on a 4-point scale, was 3.09 (N = 38), with a range from 2.19 to 4.0; in B's class the mean GPA was 2.99 (N = 35), with a range from 1.89 to 4.0 The mean ACT-Math score in A's class was 21.3 (N = 26); the range was from 7 to 36. For B's class, the Act-Math mean was 20.8 (N = 29), with a range from 8 to 33. The mean age of the students in each class was 21 (N = 38, for both classes), with ranges from 18 to 40 in the first class and 18 to 47 in the other. Procedure Instructors provided couiae outlines and descriptions of their Intentions and purposes for the course Each class was also visited periodically by one of the investigators, three times to address the whole dass (to introduce the study,, distribute material, etc ) and at other times, only briefly, to collect information fr-m individual students. Although both instructors were teachilg the same aurse and agreed on its general content, they approached the course somewhat differently. One instructor used a textbook thai he had written, and his lectures corresponded closely to the sequence of topics in tha text. The other instructor used a different textbook, and his lectures tended to diverge more from the assigned reading. Participating students were requested to provide various kinds of information at different times &ring the semester. At the beginning of the course, and again toward the end, they were asked to state their level of aspiration in the course in terms of an expected grade. In addition, they were asked to provide information about theit majors, their reasons for taking the course, the hours per week they worked, the hours per week they had available for studying, both in general and for this specific course, their age, and the total number of hours in which they were enrolled during that 348 -f 349 Determining Course Achievement semestet. A randomly selected group of students in each class was also asked to keep narrative logs over the course of the semester, in which they recorded information about how and why they studied outside of class. ANALYSES AND RESULTS Approximately two-thirds of the participating students in each class were econom- ics ms. Most of the nonmajors reported they were taking the course to satisfy a college core requirement. Instructor A gave a total of three exams including the final, whereas Instructor B gave five exams including the final. All exams used multiple-choice and short-answer supply questions predominantly. Each instructor developed his own exams; therefore, different tests were given in the two classes. The first examination in each class was held toward the end of the first month of the semester. The mean score on the first exam in A's class was 42, Niith a range from 32 to 54. In B's class, the mean score on the first exam (a different one) was 29.2, and the range was from 17 to 39. The mean finai grade, on a 4-point scale, was 2.6 (N = 89) in A's class and 2.42 (N = 62) in B's class. Results were comparable (2.67 and 2.47, respectively) for the 38 students in each class who participated in the study. To assess the effect of the many factors examined on final course achievement, several of these variables were entered into multiple regression analyses (one for each class), with final exam soon. as the criterion variable in each case. The final exam in each class was cumulative, in that it covered material from the entire semester Separate analyses were performed for each ,.lass, as the exams and conditions within each class were not comparable. Final course grade was not employed in the analysis, since it was determined, in part, from other variables used. Complete data from 34 students were available for one class (A) and from 31 for the other (B). Other students did not choose to participate, and, in some cases, data for participating students were not available or not provided by the instructors. Since very few of the students took the SAT exams, SAT scores are not included. Insufficient data were provided for some of the variables, such as expected grade, to use them in the overall analyses. The mean number of hours per week students in A's class reported having available for studymg (overall) was 23.3, the comparable figure for B's class was 19.5. Students in A's class reported working an average of 17.3 hours per week, while the average for Ws l-lass was 18.4. Students in both classes were enrolled for an average of 13 5 hours. Predictor variables entered into each analysis included cumulative GPA, ACTMath score, hours enrolled, age, hours of work per week, estimated total hours per week available for studying, and first exam score. Each analysi: (one for each class) h explained a significant portion of the obtained showed that the only variable variance was first exam score (44% in one case, 469 in the other). In each case, cumulative GPA an,: ACT-Math score were each either moderately or highly correlated with final exam score, brt they also showed a strong relationship to first exarr score and thus did not independently predict final course performance Other variables, 349 350 Table 1 Intercorrelations Between Variables Incbded in Regression Analyses Variables GPA Hrs. Enr. Age Hrs. Work Hrs. Stud Exam I Final Exam GPA Hrs. Enr. Age Hrs. Work Hrs. Stud. Exam 1 Final Exam GPA - .151 .325 - .228 - .115 .541 .364 - .150 .171 - .300 .335 .375 .339 -- --.132 ---.300 Class "A" (n = 34) -.133 -.446 .123 .045 - .007 - .058 .217 - .043 .041 - .033 Class "B" (n = 31) -- - .245 - .527 .291 --- -- .105 .023 -.111 -.101 .055 .020 - .479 .004 -.165 - .323 - .091 .691 --- .171 .13" .665 Work = hours per week Note. Hrs. Enr. = Number of semester hours for which student is enrolled; Hrs. student works; Hrs. Stud. = estimated hours per week available for studying. minimal and inconsuch as age, hours worked, and hours available for studying, had regression sistent effects. The intercorrelations between the variables included in the analyses are shown in Table 1. 6 in The studying logs kept by a subset of students in each class (8 in one case, when studying. In the other) indicated that they primarily read their texts or notes discussions, were mendoned. some cases, however, other activities, such as group the time they stopped In their logs, students noted the time they started studying and also commented on for each occasion that they studied for this course. Students Although problems they were having with the class, the instructor, and/or the text. differed in their class observations and interviews revealed that the two instructors the way tht approach to teaching, this difference seemed to have little effect on about students were quite systematic students studied. In addition, although some economics at studying, others were much less so. Ten of these 14 students studied the period of 9 least once a week, and 9 of these 10 studied at least 1! times over weeks that they kept the logs. CONCLUSIONS They indicate These results are not surprising, but they are nonetheless interesting. prior aptitude college course, itself influenced by that initial perfonnance ir. auch a regardless and achievement, is the best predictor of final achievement in that course, reflect, to of which instructor taught it. Not only does the initial exam in a course 359 t 351 Determining Course Achievement some thgree, students aptitude for and interest in that course, but it also may be encouraging to those who Jo well and discouraging to ti.ose who don't. Other variables thought to influence course achimiement may inherently have less statistical variability, and also the direction and intensity of their influence may differ across students. That is, factors such as hours of work per week, age, and initial expectations may affect different students differently. It is not the case that they are not important, but probably that their importance varies, suggesting again the complexity of motivational and situational variables in academic achievement. Further research is needed to study these factors more carefifily and closely. In particular, this investigation underscored the difficulty of obtaining a real pic- ture of what is going on in large, introductory college courses, which tend to be highly impersonal. Typically, there is little class discussion and little ongoing student involvement. Feedback to students tends to be infrequent. Given these circumstances, students probably develop "survival strategies" for negotiating success in such cuirses. Survey procedures which use paper-and-pencil measures, however, offer few clues as to how students may be doing this, or into how much actual learning may be going on. To obtain such information more qualitative or narrative methods may be needed to gain insight into the memges students give themselves concerning course expectatims, events and outcomes, and how they deal with them. REFERENCES Biggs. J B. k 1984). Leamaig strategies. student motivation patterns. and subjectively perceived succeu In J. R. Kirby (Ed.). Cognitive strategies and educatwnal performance (pp 111-134). Orlando. FL Academi, Press. Nolen. S. B k1988). Reasons fur stidying Motivationa. ..ientations and study strategies. Cognition and Instruction. 5. 269-287. Nice. A. J.. Peck. A. R.. & Sherk. J K.. Jr. (1986. December). The relanonship between students self reports of text studying practices and course achievement. Paper presented at the meeting of the National Reading Conference. Austin. TX. Pace. A. J.. Sherk. J. K.. Jr.. Peck. A. R . & Baldwin. D. (1985). The relation between students' selfreports of studying procedures and measures of reading comprehension In I A Niles and R V Lalik (Eds.). Issues in literacy. A research perspective (pp 233-237). Rochester. NY, National Reading Conference. Ruhwer. W. D.. Jr k1984). An invdatiun to an educational psychology of studying Educational Psycholo gist, 19. 1-14 Schumacher. G. M. (1987). Executive 4.ontru1 in studying In B K Britton & S M Glynn (Eds ). Executive control processes .n reathng (pp. 107-144). Hillsdale. NJ. Erlbaum Schuman. H.. Walsh. E.. Olson. C., & Etheridge. B (1985). Effort and reward The assumption that college grades are affected by quantity of study. Social Forces. 63. 945-966 Thomas. J W (1987. Apn1). Proficienci at academic sndying. Paper presented at the meeting of the American Educational Research Association. Washington. DC. Thomas. J W.. & Rohwer. W. D.. Jr (1986). Academic studying. The role of learning strategies Educational Psychologist. 21. 19-41 35 CHILDREN'S ABIUTY TO UTILIZE THE JTIEMONIC KEYWORD METHOD: AN EDUCATIONAL APPLICATION WITHIN FOURTHGRADE CLASSROOMS Nancy L. Williams University of North Carolina at Charlotte Within the last decade, research in the area of mnemonics has focused Lizon the keyword method, a technique using imagery to link new and known information (Pressley, Borkowski, & Johnson, 1987). Here, the learner pairs a new fact with a perceptually similar keyword, incorporates both in an interactive visual image; and then, when encountering the new fact, uses the image/keyword to generate an appropriate response. Although the results of this method are imprssive (Atkinson, 1975; Levin, Johnson, Pittelman, Levin, Shriberg, Toms-Bronowski, & Hayes, 1984; Pressley & Levin, 1978), educators (Graves, 1986; Sternberg, 1987) have expressed concern over the experimental conditions used to test the technique. That is, the vast majority of studies ha. _ been conducted under laboratozy-type situations (for reviews, see Pressley, Borkowski, & Johnson, 1987; Pressley, Levin, & Delaney, 1982). To investigate actual classroom application of the keyword method, some studics have used content material (Konopak & Williams, 1988; Levin, Morrison, McGivern, Mastropieri, & Scruggs, 1986; Williams & Konopak, 1988), with only one study using ecologically valid materials (Williams, Konopak, & Readence, 1989). The Williams et al. study (1989) successfully attempted to teach fourth graders how to utilize the keyword method within the context of a science lesson over days. The promising results, however, are limited by a minimal amount of instruction. The present study, therefore, was an effort to extend previous research on the efficacy of the keyword method by teaching children a study strategy over a period of 2 weeks within an ecologically valid context. Specifically, fourth graders were instructed in either the mnemonic keyword method or in notetaking/outlining, a common study strategy suggested in fourth-grade curricula. Teaching included direct instruction, guided practice, and independent practice within a classroom setting. METHOD Subjects Subjects wtre 106 heterogeneously grouped fourth graders enrolled in a rural elementary school. The four intact classes were randomly assigned to either a mne353 354 Literacy Theory and Rem:eh monic keyword group or notetaking/outlining group. To insure equivalency of groups by ability, standardized reading test percentile scores (Comprehensive Assessment Program, 1983) were collected and compared. The mnemonic Ltyword group with 53 subjects had a mean of 47.40 (SD m,24.45), whereas the notetaking/outlining group, also with 53 subi-cts, had a mean of 52.81 (SD 21.51). Two preliminary t-tests indicated no statistically significant differences between the two treatment conditions in regard to reading ability or prior knowledge of tit:: content material. Materials Instructional materials included passages 3elected from a fourth-grade basal reader (Early, Canfield, Karlin, Schottman, & Wenzel, 1983), not used by the classes, for initial study strategy instruction, researcher-developed charts, and a chapter on Midwest climate and resources from the social studies textbook (Cangemi, 1986) utilized by the fourth-grade classes and selected by the fourth-grade teachers for strategy application. Assessment materials iLcluded a test of prior knowledge, adapted from Zakaluka, Samuels, and Taylor's (1986) procedure, immediate and delayed multiple-choice and probed recall tests following the social studies instructional unit, and study strategy questionnaires. The test of prior knowledge (a phrase on Midwest climate and resources) was presented to the subjects who were then given 5 minutes to list words related to that topic The multiple-choice assessment consisted of 14 questions based upon the information contained within the chapter in the social studies textbook. Independent judges examined the items for content validity and to designate literal or inferential level questions The probed recall consisted of four phrases targeting significant information contained within the chapter in the social studies textbook. dere, the subjects were required to write what they remembered about thesc target phrases. The delayed multiple-choice and probed recall tests were reordered versions of these assessments. The final assessment measure included two forms of s study strategy questionnaire The initial form addressed present study skills whereas the second form, administered after instruction in social studies, asked the subjects to report the employed study skill, its helpfulness, and use. Procedure The researcher, following the daily schedule of the subjeis, conducted all instruction and testing within the four intact classes. During the first week, the researcher met with suhjects in reading classes for strategy Instruction. After completion of the initial study strategy questiolk .ire, the subjects were instructed in either mnemonic keyword or notetakingloutlining study strategies. Durins the second week, the researcher instructed the four classes in the social studies period, using either mnemonic keyword or notetaking'outlining methods to study Midwest climate and resources. Mnemonic keyword group During the first week, on Day 1, subjects completed the study questionnaire and were given initial instsuction in the mnemonic keyword method On Day 2, after a review, subjects practiced this study strategy using a 353 Fourth Grarkrs Utilizing Mnemonic Keyboard Method 355 passage from a fourth-grade basal reader (Early et a)., 1983). On the third day, subjects were presented with another passage and worked within small grnups for application of the strategy. In the second week, on Day 1, subjects were administered the test of prior knowledge, reviewed the mnemonic keyword method, and introduced to the chapter. After establishing a purpose for mading, the subjects read a passage about the climate of the Midwest, compared it to the clicate of the Northeast, and discussed the amount of rainfall in the Central and Great Plains. They were then provided with the keyword Clem Ant Jr climate and discussed the passage using researcher-developed charts illustrating Clem Ant visiting the area. An independent written activity and discussion followed. On Day 2, after a review, the next section of the chapter, that of resources and industries of the Midwest, was introduced to the subjects. Vocabulary was taught using the keyword method. Here, the subjects worked in small groups to generate keywords, which were shared with the entire class. The subjects were provided with a purpose for reading and instructed to silently read the passage. After reading and discussion of the main topics, the subjects were presented with keywords for the resources and industries of the Midwest. These included. (a) resot rces represented by a sorcerer who transformed such resources as corn, wheat, milk, and cattle iuto respective products manufactured in the Midwest, and s'b) industries representel, by trees that contained flour and cereal, cheese and other dairy products, and meat. Discussion and a written activity concluded the lesson. On Day 3, after a review, the subjects, in small groups, were asked to read tire textbook passage about Mark Twain. Then, after reading the passage, they were to generate their own key words tu help there remembe. :he important facts. These keywords and images were illustrated on the chalkboard and shared with the other subjects. On Day 4, the subjects completed an immediate probed recall, an immediee multiple -choiee assessment, and the questionnaire. Then 5 days later, prior to more ins action in social studies, the subjects were administered the delayed measures. Nutetak.ng,enaloung group. In contrast, the notetaking;outlining group followed thc same proeedures in lesson format and materials but with instruction in using notetaking and outlining strategies instead of mnemonic keywords. More specifi call), during the first week, subjects were taubht notetaking and outlining strategies using researeher prepared charts developed from a basal series not utilized by the school system (Early et al., 1983). During the second weeic, subjects were presented resew cher de eloped outlines of chapter sections dunng guided instruction and en couraged to generate new outlines within small gruups as independent practice. Scoring Soring for the immediate and delayed multiple-choice tasks included one point for a correct response, with a possible total of 14 points for each testing measure. A coefficient alpha, calculated mint, the Kuder Richardson formula, resulted u. a reli ability coefficient of .54 for this task. On the probed recalls, one point was given for each correct answer in each of the four probes. These tasks were evaluated by indepen 35 4 356 Litm...7 Theory and Research t Table 1 Means (and Standard Deviations) for Immediate and Delayed Probed Recall and Multiple-Choice Tasks on the Social Studies Instructional Unit Group n Mnemonic Keyword 53 Notetaking/Outlining 52 Inunediate Recall MC 2.66 (2.17) 3.25 (2.23) 6.50 (2.58) 6.90 (2.39) PelaYccr Reeall 4 MC 50 2.66 6.14 (2.34): 50 (1.94) 3.48 (2.24) Not e . Maximum score...14. dent judges trained in the scoring procedure. Based upon perceeaige of agreement, the interrater reliability was .92 on this measure. Or the study strategy questionnaires, all strategies were first categorized by type and then furtha subcategorized by helpfulness and use. RESUL2S To assess differences between groups aft-r strategy instruction in social studies, a repeated measures MANOVA was conducted on the immediate and delayed probed recall and multiple-choice taskb. No statistically significant results were found on the immediate probed recall task, F(I, 97) = 2.12, pc.. 1482, on the immediate multiple- choice task, F(1, 97)= 1.02, p<.3147, on the delayed probed recall task, F(I, 97)=3.83, p<.0532, or on the delayed multiple-choict task, F(1, 07)= .76, p<.3846. (See Table I for mcans and standard deviations.) On the study strategy questionnarie thc self-reported strategies were first catego- rized according to type and then percentages were calculated for each method by experimental group The initial study strategy questionnaire indicated that 83% of the mnemonic keywor! qibjects did not use a study method during social_ studies, whereas 53% of the notetakingioutlining group did not use a study strategy. The remaining subjects in both groups reported using study guides, reviewing, and rehearsal to gain knowledge in social studies. After instruction in social studies, the majority of subjects in both treatment conditions r-ported adopting the taught strategy and finding it to be a useful technique to help them remember the information. A greater percentage of the subjects in the mnemonic keyword group, however, reported using that method, whereas over onethird of the notetaking'outlining group continued to use previously acquued study strategies. (See Table 2 for percentages.) DISCUSSION In taking into account the generalizability of the results of this study, several limitations must be considered. First, as the study extended over a 2-week period, 3.5. -4.1.1111111. 357 Fourth Graders Utilizing Mnemonic Keyboard Method Table 2 Percentages of Study Strategies Reported for the Social Studi-s Instructional Phase Group n Mnemonic Mnemonic Keyword 53 85% (98%) 0% Notetaking/Oudining ,g, 52 Notetaking 0% 65% (100%) Reading Rehearsal Other 6% (100%) 29% (100%) 6% (100%) 2% 3% (100%) 4% (100%) (1130%) Note. Percentages of usefulness are indicated in parenthesis. several subjects were absent on one or more days, and as a consequence not included in the data analysis for both testing periods. Secondly, instruction was limited to one social studies series, whereas results may differ when other content area textbooks are employed. Finally, the reliability of the multiple-choice test was low, indicating that the test itself may have affected the results. Given these limitations, the following conclusions can be drawn. First, although no statistically significant mults were found on the testing measures, die mnemonic keyword method was as equally successful as the notetaking/outlining strategy as a study technique for fourth-grade students. That is, subjects in the mnemonic keyword group scored as well on all testing measures as those subjects in the notetalcing/ outlining group. This indicates that when given the opportunity to study content material using the mnemonic keyword method, fourth graders would most likely remember as much information using this strategy as they would using the more familiar study strategy of notetaking/outlining. Further, an examination of the ind:vidual reading ability levels indicates a wide range, with many subjects in both conditions reading below grade level. As previous studies (Konopak & Williams, 1988, Peters & Levin, 1986) have found similar nonsignificant results with poorer readers, the lack of statistical significance in this study is not suprising, gi:.en this wide range of ability. Also, the analysis of the initial study strategy questionnaires provided additio.:11 insight. Although many subjects reported no study strategies prior to instruction, more subjects in the mnemonic group tridicated a lack of study skills. This lack of what Pressley, Borkowski, and Johnson (1987) label "specific strategy knowledge," coupled with poor reading skills, may also have accounted for the lack of ..atistical signifit.ance on the posttest measures. Pressley et al. (19E7) describe this condition of specific study strategy within the context of the development of imagery and mnemonic skills. They contend that good readers are capable of relating metacognitively with the text and consequently are able to determine what is important. Further, mature thinkers are more capable of integrating prior knowledge w id. new information, a basic premise of the keyword method. Future research in this area of mnemonic keyword strategies could focus on the ability of subjects of various grade levels to use and consequently generate keywords and images based upon reading ability and specific study strategy knowledge. This research could address the amount of time and direction needed for young students to adopt and use mnemonic strategies as part of regular class instruction. 358 REFERENCES Atkinson. R C ""5) Mnemotechnics in second-language learning. American Psychologist, 30, 821-828. Cangemi, J. (1986). Regions. New York: Holt, Rhinehart & Winston. Early, M , Canfield. G R. Karlin, R., Schottinan, T A., Syrygley, S. K., & Wenzel, E. L. (1985). New frontiers. New York: Harcourt Brace, Jovanovich. Graves. M (1986) Vocabulary learning and insmiction. Review of Research in Education. 13, 49-89. Konopak, B C , & Williasrs. N L.. (1988). Eighth graders' use of mnemonic imagery in recalling science content information. Reading Psychology, 9, 233-250. Levin, I R ('983) Pictorial strategies for school learning. Practical Illustrations. In M. Pressley & J. R. Levin (Eds 1, Cognitive strategy research. Educational applications (pp. 213-237). New York: Springer-Veilag. Levin, J R., Johnson, D., Pittleman, S., Levm, K. S., Shriberg, L. K.. Toms-Bronowski, S., & Hayes, B (1984) A comparison of semantic and mnemonic-based vocabulary-learning strategies. Reading Psychology, 5. 1-15. Levin. J R . Morrison, C. R., McGivern. J E., Mastropien. M. A.. & Scruggs. T. E. (1986). Mnemonic facilitation of text-embedded science facts. American Educational Research Journal, 23, 489-506. Pressley. M Bork J G & Johnson, C J. (1987). The development of good strategy use. Imagery and related mnemonic strategies. In M A. McDaniel & M. Pressley (Eds.). Imagery and related mnemonic processes Theories, individual differences, and applications (pp.. 274-297). New York. Springer-Verlag. Pressley, M . & Levin. J R (1978) Developmental constraints associated with children's use of the keyword method of foreign language vocabulary learning Journal of Experimental Child Psychology, 24, 53-59. Pressley, M Levin.] R , Delaney. . H D (1982) The mnemonic keyword method. Review of Educational Research. 52, 61-91 Sternberg, R J (1987) Most vocabulary is learned from context. In M. McKeown & M. Curtis (Eds.), The nature of vocabulary acquisition (pp. 89-105). Hillsdale. NJ. Erlbaum. Williams. N L & Konopak, B C (1985) Sixth graders' use of mnemonic imagery in recalling content material In J R Readence & R S Baldwin (Eds.). Dialogues in literary research (pp. 125-131). Chicago: National Reading Conference. Williams. N L , Konopak, B C . & Readence, J E (1989). Effects of mnemonic imagery training on fou.rth graders' recall of coni.nt matenal In S McCormick & J. Zutell (Eds.), Cognitive and social perspect es for literacy research and instruction (pp 287-292). Chicago. National Reading Conference Zakaluka. B L Samuels. S J . & Taylor. B M (1986) A simple technique for estimating prior knowledge: Word assoiation Journal of Reading. 30. 56-60 1; A DESCRIPTIVE ANALYSIS OF GOOD READERS' AND WRITERS' CONCEPTS OF AUTHORSHIP AT GRADES ONE, THREE AND ME RGbert J. Nistler University of North Texas Rearchers such as Burke-LeFevre (1987), Ede (1985), Eisenstein (1979), and Foucault (1975) document the evolution of authorship as a concept in Western civiliza- tion. Beginning in the early Middle Ages, interactions between authors and society have served to expand, elaborate, and further define what it means to be an author. An expressive-romantic image of authorship which developed in the early nineteenth century depicts an author as isolated from the social world, a contemilative individual bowed over books, striving to get in touch with experience, truth, reality, history, cr tradition. The continued presence of this expressive-romantic image of authorship in society today exerts a great influence on instructional practice in schools (Cooper, 1986). Emphasis on writing as an indiNidual activit:, has led to the structuring of assignments, courses, and methDds of evaluation reflecting a tacit assumption that invention is th,-. private, asecial act ,.: a writer for the purpose of producing a textan assumption which fails to acknowledge that invention is often a collaborative process (BurkeLc Fevre, 1987). In response to more traditional writing programs that promote writing as a solitary endeavor, recent ,-,-esear...,h on children's w ...ten composition documents how children within a "community of authors" develop as writers (Atwell, 1987, Calkins, 1983; Graves & Hansen, 1983, Hall, 1989). The importance of such development and its relevance to current instructional practice is noted by Lamm (1989) in her claim that "One of the key facets of Whole Language instruction is authorship. As children learn to view themselves authors, they become more aware of what authorship means" (p. 704). The Hawaii Department of Education (1985) concurs in its policy statement endorsing child authorship, saying. "Children whose writing is published see themselves as creators of wie.as, as producers of language, as functioning members of the language community" (p. D5). Adult concepts of authorship are thoroughly, descnbed in interviews with professional authors, suai as those published in the seven-book series, Writers at Work. The Pans Review Intentews (PlimptcL, 1984), as well as in autobiographical accounts of professional v. nters, for example, Donald Murray (1984). However, a similarly extensive body of literature describing children concepts of authorship does riot exist. The disparitm between traditiona elementary school writing programs and actual authorial practices, the t.all by researchers to implemont elementary writing programs 359 358 ", 360 Literacy Theory and Research based on real authoring experiences, the important role childrec's concepts of author- ship play in their development as competent readers and writers, and the lack of previous research focusing directly on children's concepts of uuthorship led to this investigation. The research questions guiding this study were: (a) What Concepts of authorship were revealed in the oral and written language of.children engaged in 9 bookmaking task and pre- and post-bookmaking interviews?, and (b) How do these concepts differ for good readers and writers in first, third and fifth grades? METHOD Participanu The study was conducted during spring semester at two elementary schools in a small, middle-class, central Texas city. The building principal at each school identified two first-, third- and fifth-grade classrooms in which it was thought children were expected to do a fair amount of writing. Thirty-six students from these classrooms were selected as potential subjects based on teacher judgment and standardized test scores indicating strengths in reading and writing abilities. Because standardized test scores were not made available to the researcher, classroom teachers ranked their students according to test scores ard shared those rankings with the researchers. Test score and teacher judgment rankings were weighted and combined resulting in a list of 12 good readers and writers at each grade level. A general description of participants is provided in Table 1. Final selection participants was dependent upon a screening interview conducted b the researcher. All 36 students were interviewed to determine their willingness to participate in the project and capability to discuss their experiences in reading and writing. Procedures In a manner similar to that of Emig's (1971) study of the composing process of 12th graders, data were collected throughout three phases of student participation. In Phase One a pre-bookmaking interview probed each child's perceptions of self as a..thor and decision-maker regarding writing. Questions includ (a) Does anyone at home read books to you? About how often? (b) Does anyone at school read books to ou? About how often? (c) Are you a reader? (d) Do you have some favonte books? (e) Can you name some of them? (f) Do you know what an author is? What does an author do? (g) Arc you an author? [If yes] What makes you an author? [Lf no] Could you be an author? [If yes] What would you have to do? [If no] Why not? (h) Can you write? [If yes] What do you write? Where do you write? When do you wnte? What is the writing like that you do at home?..What is the writing like that you do at school? (i) Who reads what you write? (j) Who makes decisions about your writing? (k) Can you give me some examples of those decisions? In th second phase of data collet:lion, a bookmaking activity placed each child in the pp 'ion of being an author charged with the task of producing a book using writing matefals supplied by the researcher. One weekend for first graders and one s. lakIren's,ConciprofAuthorship Table 'I besCription of Participants Student Ethnicity Firit Grade* Classroom I-A Jose Bob Mary Hispanic Anglo Anglo Classroom 1-B Lorie Cindy Lisa Anglo Anglo Anglo 70=41. Teacher Rating Test Score Rink Combined Test TeaCheriR4. 1 2 3 1 4 6 Third Grade Classroom 3-A Ben Miguel Kay Anglo Hispanic Anglo 1 5 2 3 3 3 4 NA 4 Classroom 3-B Lynne Molly Alicia Anglo Anglo Hispanic 1 1 6 2 3 3 NA 4 1 Fifth rrade Classrooms 5-A/B Juan Joy Sue Joan Tim Mae Hispanic Anglo Anglo Anglo Anglo Anglo 1 3 2 3 2 4 3.6 2.4 3.0 2.4 NA 1.8 1 2 6 7 8 10 Standardized tests had not been administered prior to this investigation. week fur Jurd aad fifth graders separated the time when children were informed of involvement in this composing phase and the physical act of writing. During the bookmaking process each child worked individually on development of a written text. Over the course of one week appmximately 30 minutes for writing were allotted each day. The actual use of this time and whether more was needed was determined by each writer. Data collection concluded with a post-bookmaking interview in which each participant was asked to describe personal authoring processes and compare them to those of other authors. Questions included: (a) Go through your book page by page so you can tell me wh t you did on each page; (b) If you had more time, what else would 362 Literacy Theory and Research you add 1 tc) If you had more time, would you change anything? (d) What did you do that's just like what an author does? (e) Do authors do anything that you didn't do? (f) Do authors do anything that you cannot do? For all interviews, this structosed outline of questionn was a means of gathering consistent information across participants and was developed from literature onstudents' perceptions of reading and writing processes (Calkins, 1986;;-Graves, 1983; Graves & Hancen, 1983), professional authors' beiiefs about authorihip.,(Murrny, 1984; Plimpton, 1984) and the resealcher's earlier tillot studies explonnislAth*.4 concepts of authorship. The combination of semiformal guided interviews 24,44..ter.F fonneace task addressed methodological concerns for using interviews to gather data. on cognitive functions as noted by Bogdan and Taylor (1975), Ericsson and Simon (1980), and Garner (1987). Data Analysis Data consisted of transcripts representing approximately 30 hours of interviews and dialogue with children during their writing, the students' books, and field notes describing participant composing behaviors. All data were analyzed at each grade level for what they revealed about children's concepts of authorship. Early analysis was guided by coding data according to general aspects of authorship derived from the literature (Calkins, 1986; Moffett, 1981; Plimpton, 1984). In tesponse to research pestion one, narrative descriptions outlined individua: and within-gioup concepts of authorship held by participants in this study. These descriptions, organized categorically, formed the data base for addressing the Aid research question regarding differences in concepts of uthorship among go,...te-kvel groups. As data analyses pmgressed, categories for describing concepts of authorship could be subsumed under three major Limas perceptions of self as autb^ sense of audience, revision and editing, and authors as decision makers. RESULTS Perceptions of Self as Author During pre-bookmaking interviews, 17 of the 18 children stated d-nt authors write books When asked if they were authors, children answered affirmatively less often for each succeeding grade level. By fifth grade, no children ..onsidered themselves authors. In first grade 5 children c nsidered tliembives authors and Justified this status in varying ways Factors contributing to this -,erception were generally related to the ability to Lomplete written schoolwork that tivc-y error free, and to nonschoolbased experiences writing stories, or in two caseb, to w citing books. An overwhelming concern for t...rectness accompanied children's reports of their writing when they addressed aspects of their own composing in terms of what they believed other authors might do. Of highest priority were good handwriting, correct spelling, and other mechanics of wrirng. Ent graders did have a perception that they were authors, Children's Concept of Authorship 363 which developed out of writing experiences, but this sense was tied to the physical or external characteristics of the production of text. Unlike the near unanimity in first grade, only 3 of the third graders considered themselves authors, 2 because they wrote stories, and one because he occasionally assembled his stories into a book. The remaining 3 children were indecisive and could not claim full 1-.!..tus as authors. Unlike the first graders, the third graders demonstrated a shifting emphasis from external concerns with the production of test to an internal ct,ncern with meaning. They separately noted concentraem on the subject, decision making, the sense of a story, use of stylistic devices for effect, and correct punctuat:on. Third gralers acknowledged dif:erences between their authoring procenes and those of other authors. They recognized that authorship extended beyond their capabilities to produce a book of appreciable length. However, as writers of stories or short books, they felt that they might qualify a, authors. No fifth graders believed that they qualified for authorial status, although 5 felt they could become authors by writing a book. Three children felt special training such as college w as necessary for learning how to write books. Tat trend toward the internal aspects of writing noted between first and third grades continued into fifth grade. Fifth graders thought they wee.. most like ether authors because they thought about or planned for their writing. Five children discussti this prewriting aspect of their composing processes. Publishing and length of text were attributes of authoiship that fifth graders felt separated them from other authors. Children in this study identified authors as writers of books. Across all gradeievel groups, one common theme related to children's beliefs that authors write books emerged during data analysis. The hooks that authors write are publishI,J, and for the most part, nea;y all first, third, and fifth graders exhibit a limited understanding of the publishing prouess. In general, children understood publishing to involve sending books somewhere to get '*checked out" and copied. In effect, publication as a means of sharing their wnting was removed from ...hildren's composing processes. Sense of Audience, Revision and Editing Childic., displayed great differen,:.., in their sense ot audience as it related to editing and revision proes.,es. First giaders viewed authorship as a rig:-.t or wrong proposition. They displayed an awareness of an audience for their writing only to the extent that texts were prepared for evaluation, usually by teachers. This limited understanding of a relationship between author and auilence restricted the range of options avadabk dunng composing. In preparing their texts for reading by ()thus, ti.c first graders were mainly concerned with editing. They generally made changes to text immediately after a word a:, written and rarel; reread completed text. In contrast to editing, revision remair IA a relatnely unknown and annecessary consideration for first-grade authorship. Consequently,, it was not surprising that similanties ...hildren noted between their own composing behaviors and those of other authors relates to the correctness of texts. First graders did not demonstrate the sense of process that G.:ayes and Hansen (1983) descnbe as capable of 'lifting children into thinking more about information 3,62 364 Literacy Theory and Research and the content and organization of what authors actually do in writing" (fa. 182). First graders involved in full authoring experiences during the Graves and Hansen study generally acquired this "sense of process" by the end of October. However, in April, during the time of this study, the first-grade participants' "sense of process" remained limited to editing concerns, at least to the degree that students regarded correctness as their teachers' criterion for evaluation. Third graders' concepts of authorship were characterized by an understanding of the relationship between author and audience as evidenced by their use of a range of revision options_ They did most revising at the vt ord, sentence, and paragraph levels, Lod demonstrated a concern not only for correctness but also for the "sense" of what they composed by making changes that they felt would make the teAt more enjoyable or easier to reac Overall, third graders wrote to a wider audience than teacher, self, and family by including classmates and generalized pul,... Awa-eness of revision, however, did not preclude a pn,portionately gre..-; concern for editing While involved in writing stories, third graders consister ly reread their work, checking for errors. This editing focus was guided by a sth of innerreader, and reportedly reflected teachers' emphases on spdling and mechanics of writing. Fifth-grade participants iisplayed the broadest se se of audience. They were able to make specific references to points in their texts where they had written in a particular fashion with ti-eir .caders in mir.d. Fif*!- graders tended to go beyond third graders' "sense" of their text to include setting ai.d character development. To a greater extent than first- and third grade participants, fifth graders exhibited a balanced emphasis on editing and revising In that respect, fifth graders reflected a diminished concern, relative to other participants, for the physical appearance of their texts. All reread their writing throughout all stages of composing, revising at the word, sentence, paragraph, and composition levels. Efth graders also exhibited a heightened awareness of multiple stages in the , imposing process, with half of them formally reading over their frYts after they were written. Authors as Decision Makers Crilkin5 (1986) describes an author as one empowered to make decisioiis a,out writing decisions involving such things as style, audience, content, length, and fullness of levelopment Children in this study identified people they felt were most responsible for making decisions in their writing. Students vaned in their responses by identifying one, two or three different decision makers for their writing. These responses included "parents," "teachers" and "self." At first grade, "teachers" and/or "parents" appeared in all children's responses. Although three children named themselves ("self") as decision makers, It was in 'iition to responses of "teacher" and "parents." First graders believed that deci- sions :lade in their writing were contrued by others, especially their teachers. Deci- sions were described as evaluativejudging the overall neatness of the text ann handwriting When children identified themsehes as decision makers, they identified the same types of decisions as ascribed to their teachers. neatness and handwriting. While composing, first graders generally exercised few options available to them '163 , Children's Concept of Aze.;wrship 365 as authors, possibly in part because of a lin..ed awareness of the range of authorial options involved in writing a book, but more likely because of pekreptions that authorship implied writing to do school work. Third graders displayed a greater amount of control in the decisions made regarding their writing. The perceived role of the teacher diminished a great deal from first grade, as did the role of parents. Parents and/or teachers represented f.,3% (3/6) of third graders' responses, the same percentage as "self." In contrast to first-grade responses of 'self " that were alwr.js accompanied by "teacher," only one child in third grade added "teacher" to his resporre of "self." UnliLe first graders, the third grders professing to bc in control of their decision making did not focus on a.. physical aspects of their texts. Tbeir reported decisions related to the wise of text, its appeal to their audience, its form and its topic. Those children reporting the teacher as decision maker showed a similar shift in emphasis from evaluative concerns regarding correctness to control f topic and form. Third graders were becoming auva.-e of their conhol over options but were not yet fully able to exercise them in practice. Three of the chi:dren expressed uneasiness and an unwillingness to handle options in topic or development of text. 1;,- 2 of these children, classroom writing exp:triences, as they described them, had prepared them to write only within well-defined guidelines. Waen that structure was removed, they struggled. Both children reported doing little (.1 no writing outside of school, whereas the other 4 third graders descnbed writing as a favoriu. leisure-time activity. Perhaps the ..reased opportunii.es outside of school for making .lecisions in their own writing had led the other thira graders to be less dependent on extenni controls in thei. rising. Fifth graders demonstrated the greatest awareness of themselves as decision mak ers in their wnting, an aware..ess supported in options the.y elected while composing. In ..iorts ..ecision making in their vv...ing, parents were no longer mentioned as a fawn, and the listing of "teacher" or "self" became an absolute choice. Children identified themselves as exclumve decision makers in their writing in 67% (4/6) of the greatest the responses. Discussion of their responses indkated that they percei control over wink, (..hok.e and limited control over length and form. Both children who named the teadier as deLision maker oted deusions regarding the correctness of their text. While composing, fifth graders displayed an awareness of options available for the genre in vvhiLlt they Lhose to wnte, distinguishing their stories by style. Awareness of a responsibility to audieme led fifth graders to explore a wide variety of options for inv&.v mg readers in the text. Underlying these explorations was an awareness of self as final decision maker not present in the other writers in this study. In keeping with Calkins statement regarding authors and deciskin making, this ision making in their writing raise questions perceptions of analysis of child:children's perceptions of themse:.s as regarding information shared ear'ki ?uthors. In "perception of self as author." there was an inverse relationship between author status and grade level, whereby, as grade level Increased, perceived authorial status decreased. In ,..untrast, with regard to control of decision making in writing, as grade level inaeased, the amount of (..hild zontrol of decisions increased, implying a -orresponding increase in authonal status- writers empowered to make decisions. 366 Literacy Theory and Research DISCUSSION The early school writing ....xperiences reported by the children in this study were controlled by teachers who emphasized the physical aspects of writing. The results or this control were most evident in first graders' concepts of authorship. The classroom experiences of authoring, as described by students, had been limited to writing in which the length, topic, form, and organization were not under their control. The typical format guiding children's writing was suitably named "controlled compo- sition." According to Graves and Hansen (198:)), "Children realize autho-- have options because they do the following in both the reading and xriting proc.:sses: exercise topic choice, revise by chcice, observe different types of* composing and become exposed to variant interpretations" (p. 182). The findings of this study support this position aad suggest that early writing experiences need to addres, aseects of composing not represented in first graders' concepts of authorship. Frequent and varied experiences with composing processes from prewriting to publishing are an avenue to developing concepts of authorship not displayed by first graders in tin._ study. Third- and fifth-grade students also reported limited opportunities for involvement in full authoring prucesscs, a consequence in part of the influence ot testing. The state of Texas has mandated testing of children ft:4 ri.iiiimum pmficiency in math, reading, and writing The tests (TEAMS f-- Texas Educational Assessment of Minimum Skills) are administered .mnually to odd numbered grade levels. As reported by children in this study and their tea'ters, much of the writing that occurred in school was modelled upon the format of the test, and according to student comments, in some classrooms "writing instruction" ceased after testing was completed. Despite teachers' recognition of children in this study as good readers and writers, writing instruction iocused on "minimum skills. Ms. Readel, a t. i-grade teacher, commented, "It is only after four years of teachhg 'iat I have learned what skills are called for on the "TEAMS' " She noted how this knowledge has altered the focus of her writing instruction Similarly, Ms Piooks, a fifth-grade wader, stated. "We are a minimum competency district. We don't have time in the day foi the 'fnll of 'free writing'." Increased knowledge about children as writers would empower teachers to rely less on 'he content of standardized tests for instructional direction. For example, the third grade tetchers involved in this study expr;ssed concerns about .onflicting infonnatior provided at writing workshops they attend. Ms. Sand noted that some presenters at workshops call for attention to mechaniLs and a finished product, while others advocate a process approach. Ms Bing spoke of presenters in this manner. "They all say something different. Some encourage the marking of errors, while others do not " Tea;hers' oncerns are reflected in children's reports uf instructional emphases in classroom writing activities. The findings suggest a need for consistent and ongoing pre and inserv ice teacher education in process writing as well as means hr dealing with the issue of standardized testing. Participants at all grade levels in this study identified two major differences between their composing processes and those of other authorsother authors write pieces of considerable length and have their books published. For most third and fifth graders, length was a critical attribute of books by Otht.1 authors, a length considered 365 19t. Children's Concept of Atahorship 367 unattainable by the children. However, the writing resulting from this project proved to be children's longest, continuous pieces of writing, a finding that demonstrates children need regular opportunities to write for extended periods of time, to carry over a single project, and to develop an area of interest over time. A number of children ..ited factors of solitude and unintemipted w .g time when explaining their requirements for fully involving themselves in their composing. For others, writing was more of a social act. When providing authoring experiences, sensitivity to differences in children's appreiches to composing is indicated. Few of the older participants' concepts of authorship included a perception of themselves as authors, since they felt incapable of writing "books." The differences between children's perceptions of themselves and of professional writers as authors indicates a need to provide children with numerous experiences related to authoring. Insights into authors as people and as fellow writers coukl be gained by extending invitations to published authors for classroom visits, writing to them, and showing examples of their work in progress. The notion of showing work in progress could also help children come to a greater understanding of revising and editing processes. Children could be reassured that w:yri. La progress is a recursive, messy and complex process, even for professionals. Finally, the inverse relationship between perceived author status and control of decision making suggests that movement toward empoweied authoring may not acces sarily coincide with one's per-eption of self as author, but rather depend more on the number and quality of opportunities to engage in full authoring processes. The impor tance of involving children in true authoring experiences both at hnme and school is once again supported. As an initial exploration into children's concepts of authorship, this set, was necessarily limited, in scope, to those experiences related to authoring a book, in setting, to individua.! writing seions, and, in selection of participants, to good readers and writers in first, t:Iird, and fifth grades. It is for future research to extend the findings of this study using altemative populations, settings and methods. From a greater understanding of children's concepts of authorsh., will ,..ome clearer applications for elementary reading and writing programs. REFERENCES Atwell. N (1987) In the middle. Upper Montclair, NJ: Boynton/Cook. Bogdan, R. & TayIo, S (1975) Introducucm to qualaanve research methods New York. John Wiley Burke-LeFevre, Kay (1987). Invention a a social act Studies in writing and rhetoric Carbondale South ern Illinois University Press. Calkins, L. M. (1983). Lessons from a child. Portsmouth, NH: Heinemann. Calkins, L. M. (1986) The art of teaching writing. Portsmouth, NH. Heinemann Cooper, M. (1986). The ecology of writing. College English, 48, 364-375. Edc, L. (1985, November). The concept of authorship. An lustarical perspective Papa presented at the meeting of the National Council of Teachers of English, Philadelphia, PA. Eisenstein, £ L. (1979). The pnruing press as an agent of change. Cambndge Cambridge University Press. Emig, J. (197!). The I. orrposing processes of twelfth graders. Urbana, IL. National Council of Teachers of English. 366 368 Literacy Theory and Risesich- Ericsson, K. t Simon, (1980). Verbal reports as data. Psychologkal Review, 87, 215-251. Foucsult, M. (1975). What is an author? Partisan Review, 42, 603-614: Garner, R. (1987). Maacognition and reading comprehension. Norwood, NJ: Ablex. Gralfes, D. (1983). Writing: teache-s and children at work. Portsmouth, NH: Heinemann. Graves, D. & Hansen, J. (1983). The author's chair. Language Arts. 60, 176-183. Hall, N. (1989). Writing with reason. Portsmouth, NH: Heineinainn-. Hawaii Department of Education. (1985). Children as authors handbook. Honolulu: Office of Instmcdonal Services. Lamme, L. (1989). Authorship: A key (amt of Whole Language. The Reading Teacher, 42, 704-710. Moffett, J. (1981). Coming on center. Upper Montclair, NJ: Boyiton/Cook. Murray, 1). M. (1984). Write to learn. New Yorlr. Holt, Rinehart & Winston. Plimpton, G (Ed.) (1984) Writers at work. The Paris Review interviews (sixth series). New York. Viking. 367 s CONSTRUCTING CONVERSATION: PEER RESPONSES TO STUDENT WRITING Sarah J. McCarthey Michigan State University The literature on peer groups in writing suggests a variety of purposes for peer groupt depending on teachers' and students' goals. (a) responding to writing, (b) thinking collaboratively, (c) writing collaboratively, and (d) editing writing (DiPardo & Freedman, 1988). Further, groups provide a forum for discussing the writing process, generating ideas, understanding the functions of an audience, and providing support for engaging in writing (Gebhadt, 1980). Much of the research has focused on the goals of responding to writing and editing writing. This research supports the idea that students learn about writilig through talking about texts with other students For instance, Nystrand (1986) found that students produced better revisions and recon- Leptualized their writiug through participation in groups, while Gere and Stevens k1 )85) found that students attended to the actual text more than the teacher did Groups can facilitate writing in a variety of ways especially given more structured tasks 1984), whereas students arranged collaboratively can solve problems in writing (Freedman, 1987). Students as young as fourth grade can expect and receive substantive help through peer conferencing w her. _ anferences focus c- improving the hor's draft (Dahl, 1988). Another line of research suggests that COE ersations among teachers and students about texts are not only valuable for increasing writing performar,.. hut are valuable in increasing other forms of liLracy. Tannen (1987) regards orality and literacy, speaking and writing, not as dichotomous, but rather as overlapping and intertwined The benefits claimed for teachers and students discussing texts together include wnter's knowledge becoming available i- talk, the beginner's work being supported through questions, comments, anti suggestions of others, and giving the beginning writer the opportunity to practice orally ways of using written language (Calkins, 1987, Flono-Ruane, 1988, Graves, 1983). Students can then transform the onversation-based knowledge and strategies into their independent writing Additionally, students who engage in talking about their texts reveal their beliefs about literacy as well as their thought processes (Daiute, 1989). For learning from peers to occur, students need opportunities to interact wit" one another. The realities of classroom life with its :equitable distnoution of knowledge and authority, however, ...an undermine opportunities for students to understand their wntilie through responses from the teacher and peers (Cazden, 1986; Florio-Ruane, 19%). Traditional classroom norms limit opportunities for students to interact, with the result that peer interactions are rare. Large group instruction with the teacher in 369 368 370 Literacy Them and Research control and children working alone on individual tasks persists me American schools (Cazden, 1988; Good lad, 1984). Ordinarily, teachers dominate instructional talk and control access to the floor. The teacher selects topics fot discussion, asks questions to which he or she kn-iws the answer to find out what students know about a topic, and allocates turns after students bid for the floor (Coulthard, 1977). The typical pattern of interaction is for teachers to initiate instructional talk, for students to respond, and for teachers to evaluate their responses (Mehan, 1982). The teacher, then, has the au' iity and control over the conversational interaction within the classroom. Child ,:arn at a very young age that the teacher makes virtually all initiating moves and mat students are expected to respond to the teacher's initiation (Willes, 1983). These norms limit opportunities for students to try out their own ideas, to confront alternative theories to the teacher's, and to respond to their peers. Several current writing programs encourage transforming traditional patterns of teacher saident interaction into more dynamic student-centered and student-controlled interactions by creating opportunities for students to learn from one another (cf. Calkins, 1987; Graves, 1983). In these ptograms, teachers organize the classroom to support daily writing, publish students' work in a variety of forms, and interact with students through conferences in which they encourage children in their writing. Students choose their own topks to write about, discuss their work with peers, and share their writing in more formai settings called "share sessions" or whole-group response sessions. These share sessions provide opportunities for students to participate in discourse that enables students to confront alternative ideas, to enact complementary roles, to have a relationship with an audience, and to Ly out new ideas (Cazden, 1988). Share sessions differ from traditional classroom inte un in several ways. First, the goal is for students to share their written texts with iers, not for the teacher to find out what children know already about a topic. Secord, the share sessIons are focused on the student/author who sits in a special chair designated as the author's chair" and calls upon students to respond to the text (G-aves & Hansen, 1983). The student/ author may control topic selection by asking students for speufic help on a prublem the author has 1. -d, the teacher's role is as an additional respondent or one who takes on the role or clarification of discussion. Research on share sess;ons is important because the dialogue in which students engage can provide a means of finding out what students know about text through the ways they talk about text Research is needed to examine how students' mteracuons change in settings that actively seek to alter traditional norms. PURPOSES OF THE STUDY This study follows from previous research that highlights slodents learning from another about texts and examines student learning in particular setting where the teacher has provic+.1 opportunities for students to mteract with one another. dent learning was investigated through two questions that guided t'ais study of one first-grade classroom (a) How do t'ae conversational strategies children use change nne' 369 371 Peer Responses to Snuient Writing over the come of a school year? (b) How does the conten1 of what children say in the share sessions change during the course of the school ye---? METHOD Context Students. The focus of this suidy was the first-grade classroom of Emily Johnson.' Twenty students of various ethnic backgrounds including Black, Hispanic, Asian, and Caucasian were in Ms. Johnson's classroom, located in anlementary school in the New York City Public Schools. The students were not grouped by ability for instruction, but were provided who'. -group instruction, small heterogeneously grouped instruction, or individual instmction during writing titre. Teacher. Emily Johnson is an expuienced elementary school teacher who has taught in the New York City Public Schools for 4 years. During the summer preceding the 1987 school year she became involved in the Teachers College Writing Project whett she received extensive instruction in helping students learn to write. Program. The Teachers College Writing Project consists of two major aspects: (a) on-going workshops including a 2-week intensive Summer Institute and 10-12 half-day workshops throughout the school year; and (1-) en-site training in which a teacher-trainer demonstrates the writing process in cla.srooms with teachers and students. The purpose of the Writing Project is to involve students in the process of what real authors dorecording ideas, planning, organizin texts to make sense of their lives (Calkins & Harwayne, 1987). The role of the teacher is to establish a predictable structure as a vehicle for the teacher and students to interact daily about writing. The focus J this study was on one part of that predictable structurethe share sessions in which several student/authors read their pieces aloud to the whole group and the other students respond to the texts. Classroom. Ms. Johnson's classroom is organized arcrind a rug that occupies a central place in the racm. Tables and chairs that students are free to use art situated throughout the room. Books are on display and accessible to children during the day During writing time children may choose where to sit to write; they sit at tables, in chairs, use pillows, or sit on the floor. Ms. Johnson calls the share sessions by announcing that it is time for students to share. The students sit on the floor in a circle around the rug. The teacher also sits on the floor. The student/author who has been designawa by the teacher earlier during the writing time is called upon by the teacher to share his or her work. The student goes to the "author's chair" and reads his or her text. The student/author then calls on students who have their hands raised to resixed to the text. During the share sessions, two to three students share their pieces that are either considered "finished" by the author or are still in progtess. All names of teachers and students are pseudonyms. .37o 372 Literacy Theory and Research Data Collection Procedures Ms. Johnson's classroom was observed three times during writing time over the course of the 1987-88 school year (October, December, and May). Each qf the writing periods consisted of about 1 hour divided among a "minilesson" in which the teacher explained a concept to the students, writing/conferencing time in which students wrote and the teacher spent time with individuals discussing *heir writing, and the share sessions. The observations were audiotaped using a wireless microphone that the teacher wore.The thserver transcribed the observations into narratives containing the actaal dialogue of the members of the class. Although this study is limited by having only three data points, it has the advantage of providing data over time. ANALYSIS The analysis is rooted in classroom discourse theory r.nd methodology outlined by Cazden (1986) as well as conversational analysis detailed by West and Zhnmerman (1982) New categories have been generated to talk more specifically about Ms. Johnson's classroom. Initially, several sources were used to define the unit of analysis. The idea of "speech events" defined as recurring, bounded events with a clear beginning and end with mnsistent rules for pan Icipation (Cazden, 1988; Hymes, 1972) was combined with the notion of "literacy events" including "occasions in which written language is integral to the nature of participants' interactions and their interpretive processes and strategies" where "participants follow socially established rules for verbalizing what they know from and about the written material" (Heath, 1982, p. 50) to form the "text/speech event" as the unit of analysis. The unit began when a student/author read his or her piece, included the conversation during his or her allotted time to read and respond, and ended with another student being called upon to "share." Several different types of analysis of each of the "text-speech eveuts" were performed. Establishing Categories The conversational sq.ategies students used and the r,ontent of students talk were analyzed using both qualitative and quantitative analyses. Conversational strategies were divided into three categories. (a) praise, (b) asking or answering questions, and (c) challenging or defending statements. Examples of praise included statements such as "I like your story," Asking questions included examples such as "Were there things9" whereas answer;ng questions included responses that answered specific questions The category of defending or challenging statements included students challenging the response of a student by giving a different opinion, whereas defending ctatements consisted of supplying more infoimation or a rationale for including something in text. The centent of student responses was categorized into either (a', focus on idea or (b) focus on mec!,,mics and logistics. The category of ideas included features of the stories such as ever 3, characters, setting ui more general concepts such as scientific 371. .c. 373 Peer Responses to Student Writing concepts. The category of mechanics or logisucs included attention to surface features of the text such as number of pages or logistics such as "Can I see the pictures!" Procedures The unit of talk that was analyzed consisted of each complete thought expressed by a student other than the actual reading of the text. Each unit was coded for both the type of conversational strategy and the type of content. Scoring was done by adding up the number Thf tall:es in ...ach category and dividing the number of tallies by the total number of complete thoughts to produce percentages. Interrater reliability was established through having a secund rater read through each set of the transcripts from each of the three share sessions (October, December, and May) and rank each of the share sessions using the initial categories that were described_ Fa instance, the rater ranked each session according to which se4,sion had the most focus on ideas, which had some focus on ideas, and which had the least focus on idea., :here were no discrepancies between the two raters. The qualitative analyses focused on emergent patterns tc e:aborate upon, corroborate, and pri, . ;de a mean.:4fcl context for the numerical data. txamtites were selected from the emergent patterns to pr. Je e,,idence for the results presented in the next section. RESULTS Table I shows the changes in the types of contersational strategies in which the students engaged over the cocrse of the year. From the beginning of the year to the next two data points, a s.gnificant ...,ange occurred in the type of strategies students used. Whereas 5:dents engoted in providing generic praise, while doing very little challenging of each other in October, in May students did not engage in generic praise at all. Instead, students engaged in asking and answering questions or challenging other.. A noticeable t' ..d is the continual ;ncrease in students challenging one another or defending their own texts. !II Table 2, the trend towards students' en6aging in increasingly more taik about ideas lather than mechanics or logistics is apparent. Whereas in Octobet therc was less ta'k auvut ideas than on mechanics or logisti,s, December's data show a greater -uiphasis upon ideas than on mechanics. In Ma; , a nigh percentage of the talk is Table 1 Students' Conversational Strategies Category October December May Praise Question-Asking/Answering Challenging/Defending 51-1% 0% 68% 32% 55% 45% 44% 6% 372 IIIMIN=rFTM. 3 4 Literacy Theory and Research Table 2 Content of Student Talk Category October December May Ideas Mechanics/Logistics 50% 50% 68% 32% 85% 15% OmMEINIft fecused on ideas in the children's texts. These trends are supported by more qualitative analyses and examples of the talk in which the teacher and students engaged. Octobes In the October share session, students did not take a very active role in the discussions. The teacher did much of the talking and encouraging of mdent interaction, while offering her own evaluations of the students' responses. Tbe following excerpt displays how students wert relatively inactive, responding only to teacher prompting: Shannon: I like that story. Teacher What did you like about it? Shannon. I like that because [it was] ftin and I liked the part about being seasick. Teacher Good When examining the content of students' responses, it seems that students had not yet developed a repertose "A issues about texts to which they might respond. The students' conversational strategies consisted mainly of offering generic praise of an- other student's story. The sequence f 'flowing Laurie's reading of her story shows another example of how students' resronses ccnsisted of praising of the story as a w hole Linda replied that I was a "nt...e" story and, when prumpted, gave another general kind of answer "I like that story." Teacher. Anybody want to say anything about that? Linda: [It was; nice. Teacher Why did you think it was nice? Linda: I like that story. Teacher Why? Linda: I like that part about when they went to school in the smw. Teacher (to Laurie) That makes you feel happy .aiten they tell you that about your writing, doesn't it? Students did focus on sot= of the ideas that the itudent/author had expressed after the teacher prompted and it is clear that there was some attention to the ideas within the story However, students tended to focus their responses on logistical types of issues as illustrnted in the retrk.ases to Jack's story: Cathy: Did you do all the pages? kli the pages in the book how corm you only reed rew]? Jack: (1' s not respond) Teacher is it finished yet, Jack? Jack: I think I can make another page tomorrow. 3 73 Peer Responses to Student Writing In this response both the teacher and the students seemed concerned with such issues as the number of pages in the story and whether the piece was finished. December In December, there seems to be a shift in the student's role during the share sessions. It was the student/authoi who was in charge of turn-taking a . responded to what she felt was important. The following discussion took place after Emily read 'aer story about summer. Emily: (reads her text) The sun. tr. I live in this house. I live in this house in the summer. This is the house where my little. . . . I have a little shelter in the woods. Hi ho hi ho hi ho. This is my rabbit and he goes tweet tweet tweet. This is the biggest rabbit . . . rainbo w. This is Jennifer and Douglas walking in the park. This is Jennifer and Dougas going to the park. One day Jennifer the boy and the girl went to the park. Students: (make noises) Emily. Not Jennifer and Douglas. It is another one (continues reading). One day the boy and the girl went to the park. T...acher. Comments? Qu:stions? Only peopk. with hands up are ,:alled on, should be talking. (Emily calls on Ron.) Ron: (Why) were there little things? Emily: It was summer. Ron: I know. Were there little thing3? Teacher I know what you mean. Everything she wrote about w as little. Snails and what else? Ron: Rabbits. Teacher: Le her comment on that, please. Eveiything is little because if I made them big, I wouldn't fit everything in, right? This is a folded paper. Rick: I think you have too many staples. Teacher There are a lot of things to say, but let Emily call on you. Alison: Why didn't you say kissing? It was Tina and Matt. Teacher Other questions? Juanita: How come tweet, tweet, tweet? Emily. The rabbit cliO'n't go tweet, tweet, he went ha ha. It was a bird that weht tweet, tweet and his name was tweet, tweet. And there was a rabbit who we.... Hi-o. Raise your hand if you w-...(a u.. ask me his name. Teacher. That is a good question. I want to see what it has to do wIth the rest of the story. Emily: I wanted to. Jason: Is Tina and Matt . . . why did you pick them? Emily. Because it was a boy and a girl and I didn't want to get anybody excited. In the December excerpt, the students' voices were much more apparent. Ron had ts . s; opportunities to ask his questior. of "Were there little things?" and participated in re were "rabbits" whe,- askeu by the teacher. Tne the dialogue by responding that student, author's voice came across morc clearly as well. She h.di several opportunities to explain why she included certain aspects in her text such as "If I made them big, I wouldn't fit everything in, right?" and she had the opportunity t.z explain that "the rabbit didn't go tweet, tweet, tweet." She defended losing the characters she did by saying, "Because it was a boy and a girl and I cll.. want to get anybody excited." 374 376 Literacy Theory and Research These examples provide evidence that students were taking a greater role in share sessions. A change occurred in December in the nature of the content of what students said and to what they Attended in the texts they heard. In the December share session, students asked more questions that were related to the text such as, "Were there little things?" and "Why didn't you say kissing?" or "How come tweet, tweet, tweet?" Of the seven smdent responses, only one was about a logistical or noncontent related issue, "I think you have too many staples." The other responses had to do with characters in the story such as "It was Tina and Matt" or questions asking about the inclusion of certain elements. The student/author was able to provide a rationale for what she included in her text. She provided details and se-med to take the questions seriously Unlike the October session, students did not '..st provide generic praise, but rather tried to understand what the author was trying to say. Since there seemed to be many places in the text that were unclear to the audience, the respondents did sessra to show a genuine interest in understanding the text. The December se.sions shows that students were beginning to ask for explanations about other,' texts and were beginning to challrnge why student/authors may have included certain aspects in their texts. In general, the talk among students appears to have more complexity than in the October session. Students seemed to be trying to respond to the content of the text and were less focused on logistics such as numbers of pages in the stories. May In tJ May session, students took an even more activr rile as participants by challenging one another The following sequence after Shelley reads her story demonstrates an e .mple of students' focusing on ideas and being v:illmg to challenge and defend ideas expressed in the texts and in subsequent tal, Shelley reads her story. "This is a song. Rain rain, go away. Rn. rain, go away, come back another day. I don't like the rain. Do you know why I don't like the rain? Because the rain gets in your face I like the sun. Do you know why I hke it? It's because it does not, does not get in your face." [when another student interrupts] I'm not finished . . . . Teacher OK Now thinking about Shelley's question tbat she had of you, let's see if your questions can help Shelley. Billy How come you say rain can get in ycur face and sun can't get in your face: Shelley: It can't. but it doesn't come in your eyeball. Billy Because the sun can go down, because it still can come down. It gets on you like Just . . . it can only like vanish . . Anytime I look . . [when a student tries to interrupt] Wait I'm not finished. Teacher Well, what do you think Billy is saying? Joshua: I think he said in the night the sun don't come up. Billy:1 sah, the sun can come, the sun can come down, but it can come up. Shel ey: But it can't come down like the rain comes. Billy: I kaow that but rain hits water. Rain is . . . . Teacher- I think what shes doing, Billy. I think she has two different things, and . think that she is comparing the two. Do you set how she is comparing the two? She's saying how she U.Aesc't like the rain - id the reason she doesn't like it is because it can get in her face. She like./ the lin and the reason she likes the sun is because it doesn't come down and touch, get in he. `,ce. It can. You can feel the heat. Is that what you are talking about? 375. ti Pee* Respmses to Student Writing 377 Billy: Ycah, and it feels like . . . . Teacher. And it hurts your cyes. Yeah, but she mems really come down, Billy, and touch you. Billy: Like it rain . . . . Teacher. OK, Billy, you calk to Shelley about that later about this. Maybe m writing tomorrow, but right now there are other people that need to talk to Shelley. In this excerpt, the studentiauthor seemed to be able to explain what she meant in her stoty. The student/author's authority was challenged, however, by the student, Billy, who persisted in suggesting +hat the sun "can come down" and "get on you like klie rain]." The teacher encouraged the students to try and .nplain their conflicting positions. Billy persisted in his reasoning, whereas Shelle) continued with her explanation about the differences between thf: sun and the rain. -Ate content of what students discussed in May reflected students aking a trlre active role in the discussions. The focus of ihe interaction was on a mora general idea in the textwhether you could really feel the sun on your face in the same way that you can feel rain. There was no specific praising of ideas or parts of the story, and little emphasis on surface features such as number of pages. Instead, the stidents were struggling with a concept as well as trying to understand the meaning of the text. Whereas early in the year students generally made comments about the whole story, , and in December, students asked more questions and probed the specific content of :he text, in May there was a real dialogue among members of the group with argumentation and an attempt at explanation. At least one studeat was challenging and refutmg the idea implicit in the text. There was a lack of clarity at several points in the conversation as the students grappled with the idea of whether the sun and the rain are felt in the same way. The conversation seemed to move from comparing the sun and the rain to a discussion about the rising and setting of the sun, perhaps because of the use of the words "come up" and "come down." As the students engaged in discussion precipitated by the question, "How come you say rain can get in your face and sun can't get in your face?" the interaction becalm. complicated. For instance, when one student was asked to rephrase the boy's question. he says, "I think he said in the night thc sun don't come up" indicating a possible misunderstanding of Billy's question. The conversational data sugrst that the ideas and discourse itself are more complex in the May interactions. This discussion also highlights the overlap ...aong understanding concepts, talking about texts, and the texts themselves. DISCUSSION By comparing the October, December, alid May interactions of students within ,ne share sessions, shifts in several areas seem to emerge during the course of the school year. Whereas students were not very active in the beginning of the year, different members of the class v ere in control at different moments in time in the May session. Increased student participation was reflected by the changes in conversational strategies. There seemed to be an increase iii challe:.ing and defending ideas and a sharp decrease in using genenc praise. In terms of content, students seemed to focus 376 378 Literacy Theory and Research increasingly on t"-:. ideas of the texts, rather than on surface features. Distssions began to focus on concepts and ideas expressed in texts near the end of the year. Lhwever, the classroom share sessions did not unde_go a complete transformation in which students learned to become completely self-sufficient during the share sessions. Although the dialogue became richer and more complex while focusing on ideas later on in the year, it was not clear if students really understood and developed the ideas of others. Students did respond and challenge ideas, yet students did not necessarily build upon one another's ideas in a tracedble pattern. These limitations may reflect the difficulty of engaging young students in rich, complex, and sustained dialogue. For students to learn to engage in complex dialogue to learn from one another, it seems apparent they will need many more opportunities in a variety of contexts to become accustomed to responding to their peers. It is hopeful to note that at least in , 1 1 ";;;:l one classroom, during writing time, students had opportunities to engage in rich discourse about texts and seemed to have benefited from these discussions. Because this study was limited to three share sessions in one school year. it seems that future research could extend and elaborate upon the changes discussed in this paper Mar: more examples of share sessions are needed in other classrooms to provide additiorai data of student learning when students have th.. opportunity to engage in discussion of text In addition, it seems that an important contribution of future research vv ould be to examine oAat individual students learn from these share ses s ions . REFERENCES Calkins, L. M (1987) The art of teaching writing. Portsmouth, NH. Heinemann. Calkins, L M , & Harwayne, S (1987). Writing workshop. A woad of dee, ence. Portsmouth, NH. Hememann. Cazden, C B (1986) Classroom discourse In M C Wittrock (Ed.), Handbook of research o teaching (pp. 432-463). New York: Macmillan. Cazden, C B (1988) Classroom discourse The latiguage of teaching and :earning. Portsmouth, NH. Heinemann Coulthard, M (1977) An introduction to discourse analysis. Essex, Englaid. Longman Group. Dah!, K L (1988) Peer conferences as social contexts for learning about r..vicion In J. C. Readence & R S Baldwin (Eds ), Dialogues in literacy research (pp 307-315) Clucago. National Reading Conference. Daiute, C (1989, Play as thought Thinking ategies of young writers. Harvard Educational Reciew, 59, 1-23 DiPardo rreedman, S W (1988) Peer response groups in the writing classroom. Theoretic foundations and new directions. Review of Educational Research, 58, 119-149. Florio- Roane, S (1988) How ethnographers of commumcalivo study wnting in school. In J. C. Readence & R S Baldwin (Eds.), Dialogues in literacy research (pp 269-283). 'Thicago. National Reading Conference. Florio-Roane, S (in press) Instructional t,onversations m learmng to write and learning to teach. In B. Jonv ), Dimensions of thinking A framework for curnculum and instruction (Vol. 2). Hillsdale, NJ. Llbaum. Freedman, S W (1987) Peer response in two ninth grade classrooms (Tech Rep. No. 12). Berkeley, CA: University of California, Center for the Study of Writing. II 379 Peer Responses to Student Writing Gebhardt, R. (1980). Teamwork and feedback. Broadening the base of collaborative writing. College English. 42, 69- '4. Gere, A. R., & Stevcns, R. (1985). The language of writing groups: How oral response shapes revision, In S. W. Freedman (Ed.), The acquisition of written language: Response and revision (pp. 85-105). Norwood, NJ: Ablex. Goodlad, J. I. (1984). A place called school. New York: MacGraw-Hill. Graves, D. H. (1983). Writing: Teachers and children at work. Ponsmouth, NH: Heinemann. Graves, D. H., & Hansen, J. (1983). The author's chair. Language Arts. 60, 176-183. Heath, S. B. (1982). What no bcdtirne story means: Narrative Is at home and at school. Language in Sociay, 11, 49-75. Hillocks, G., Jr. (1984). What works in teaching composition. A mem-analysis of experimental treatment studies. American Journal of Education, 93, 107-132. Hymes, D. (1972). Introduction. In C. Cazden, V. John, & D. Hymes (Eds.), Functions of language in the classroom (pp. vi-xii). New York: Teachers College Pius. Mehan, H. (1982). The structure of classroom events and their consequences for student performance. 'In P. Gilmore & A. A. Glatthorn (Eds.), Children in and out of school (pp. 59-87). Washington DC: Center for Applied Linguistics. Nystrand, M. (1986). Learning to write by talking about writing. A summary of research on intensive peer review in expository wnting at the University of Wisconsin-Madison. In M. Nystrand (Ed.), The structure of written communication (pp. 179-211). Orlando, FL: Academic Press. Tannen, D. (1987. 1 The orality of literature and the literacy of conversation. In J. Langer (Ed.), Language, literacy and culture: Issues of sociey and schooling (pp. 67-88). Norwood, NJ: Ablex. West, C., & Zimmermans, D. H. (1982). Conversation analysis. In K. R Scheret & P. Ekman (Eds.), , New York. Cambridge UniHandbook oi methods in nonverbal behavioral research (pp. 506 versity Press Wines, M. (1983). Children ,r..0 pupils. A study of language in early schooling. London. Routledge & Kegan Paul. 378 EARLY LITERACY STRA1EGIES: ACTIVITIES REPRESENTED IN CURRENT BASAL READERS Lesley Mandel Morrow and Rachel Parse Rutgers University t For more than 150 years basal readers have been the most commonly used instructional materials in American elementary education, estimated to be in use in at least 95% of the schools (Farr & Roser, 1979; Yarington, 1979). Increasingly, basal programs are being adopted for kindergartens as well. In their Report card on Basal Readers (1987), GooCanan, Shannon, Freeman, and Murphy place the basal within its twentieth century context. Relying on Thorndyke's Principles of Learning plus scientific mandgement techniques of the 1920s, they report, publishers created "a sequential, all-inclusive set of iLstructional materials [capable of teaching] all children to read regardless of teacher competence and regardless of learner difference" (p. 133). Teachers in the 1920s had little professimal preparation and even le.ss access to scientific knowledge of the processes of eading and writing. By contrast, Lachet education progrxrs today are quite sophisticated and provide this information. Coupled with current notions of empowering teachers to make primary decisions about instruction, it is logical a. ask to what eaten( basals continue to rely on prescripfive .tructures, and to what extent they reflect research on literacy development that has been reported in the past 20 years. A number of arguments support the use of basals: Their reading selections rend to be of high qualit eachers' manuals suggest systematic instruction; materials are tions of the read'ing process (Shannon, I983a). On the other based on scientific hand, teachers find it difficult to choose from the many ...:tivities suggested without deviating from the basal programs' s, pe and sequence charts, and several studies found adminisn-ative constraints on teachers so demanding that some teachers followed manuals to the letter, simply to protect their jobs (Duffy, Roehler, & Putnam, '987; Shannon, 1983b, 1987). Halliday (1975) proposes that it is the functional u of laiguage that motivates language development. Some critics charge that, within the conflict of basal readers, the function of text i often lost to reading readiness and skill oe.ntation which tends to separate words from meaning (Goodman et at , 1987; Morrow, 1989k. It has even been suggested that reading readiness pregrams in particular ere sheoretically and practically inconsistent with the way young children learn to read (Teale & Sulzby, 1987). 381. 37 382 Literacy Thecry and Research Early Literacy and Basal Readers Research in cognitive -elopment and language acquisition has changed attitudes and ideas concerning literacy development. The concept of emergent literacy (Clay, 1966) suggests that chiP n acquire some knowledge about reading and writing long before formal education begins. Literacy development begins early in life, in dynamic, interactive relationships among communication skills (e.g., reading, writing, speaking, and listening) and within social contexts (e.g., family, community, sibling and peer relationships), whether or not those skills are fully dm:loped or entirely conventional (Tea le, 1982). A child's scribble writing, narration of a story from illustration, attention to page sequence, awareness of left-to-right progression, and differentiation of voice tones to distinguish between conversation and "reading" all are evidence of emergent literacy (Sulzby, 1985; Tea le & Sulfoy. 1987). By contrast, basal reading programs have tended to rely on three primary assumn- tions about reading readiness: (a) that a child has no prior knowledge of languav each letter and story is introduced as if the child is completely unfamiliar with it, (b) that literacy develops through rote repetition and drill, and (c) that reading acquir,ition is essentially a structured, hierarchical process which moves from letters to words to sentences and paragraphs (Hiebert & Pap'terz, in press). A report for NCTE's Commission on Reading stated that "the sequencing of skills in a basal reading series exists not because this is how children learn to read, but simply because of the logistics of developing a series of lessons that car. be taught sequentially day after day, week after week, year after year" (Weaver & Watson, 1988, p. 1). Given the fact that basals play such an important role in reading instruction and more of a role than es er before in early childhood education, it is essential that findings from recent research be incorporated 2tit0 basal materials. Researchers in early literacy suggest that to becomc literate, young children must learn the functional uses of literacy, plus concepts about books and print (clay, 1985, Lomax & McGee, 1987, Sulzby, 1985; Hiebert, 1981), that recognizing letters and words includes phonemic awareness, grapheme phoneme correspondence, and decoding and encoding strategies (Lomax & McGee, 1987, Mason & McCormick, 1979), that both listening and reading comprehension follow from an understanding of language (Pyson, 1984, Morrow, 1985; Sulzby, 1985), that emergent literacy includes compo:ation and writing (Clay,, 1985; Ehri, 1989; Harste, Woodward & Burke, 1984); add that the acvmplishment of these goals takes place in environments rich in literacy materials that foster interest in reading and the opportunity to engage in and learn recoi...aended strategies for literacy development "arrow, 1982; Morrow & Weinstein, 1982; 1986). A study by I-het, and Papierz (in press) of 1988 editions of kindergarten basal materials noted that althoagh many studies have dealt with content issues concerning basal materials for first grade on, none had dealt with kindergarten materials. The study reported here was an attempt to determine to what extent current basal readers include accepted traditional as well a.s newer st.4tegies that have been determined over the past 20 years to promote literacy development in early childhood. More specifically, the study asked (a) With what frequency do current and traditional strategies for early literacy development appear in lesson plans for kind:rgarten and firstgrade bac& materialo and (b) Are there differences in the fretiuency and suggested 3S0 Basals and Early Literacy 383 use of such elements between the main lesson:- and supplementary sections of the teachers' manuals? METHOD Materials The kindergarten and first-grade books from six sets of 19C.; basal readers, chosen for their widespread use, were selected for analysis. Kindergarten kits were aot included. Because the publishers' separate descriptions of their programs were quite similar, the gist for all six is paraphrased here: The reading program comt)ines aspects of a literature-based, whole language, meaning-oriented approach that encourages critical thinking. The series uses a temes with activities that utilize reading, writing, speaking, and listening to motivatt interest. Selected lessons in phonics are provided to help the studem become an independent reader. In their program descriptions, two of th .. publishers emphasized a skills approach more than did the others. Procedure Research assistants analy.. d lesson plans for all 506 stories in the 59 books included in the study. They identified the number of times the lesson plans suggested activities that have been found to promote early literacy development, more specificall: . (a) comprehension development, (b) concepts about books and print, (c) Ianaage development, (d) rich literacy environment, (e) reading attitudes and independent reading, (f) word recognn on skills and phonemic awareness, and (g) writing development. Within each of these major cateyories were numerous subcategories which researchers have found vital. The list was composed from an extensive review of research in emergent literacy over the past 20 years (see Table I). Activities found in the main portions of le*son plans were oded separately from those found in supplementary sections of manuals. Supplemen ary sections had differ- ent labels from publisher to publisher, for example, enrichmert, whole language activities, vtional lessons, and so forth. For purposes of the study, supplementary was defined as lessons not required within the program at hand. Elements occasionally fit into more than one category and were therefore counted in both. During practice sessions, the 12 research assistants analyzed manuals, familiarizing themsel .es both with the manuals and the elements to be identified. Reliability among the 12 was determined by asking each to analyze the same lesson pka for each of the six series at each grade level, kindergarten and first grade (a totql of 12 selections for each assistant). Calculations indicated the following reliat..aty quotients. comprehension cif' Aopment, 9J%; concepts about books and print, 88%; language development, reading attitudes and independent reading, 93%; rich literacy nvironment, Id recognition and phoncmic awareness, 88%; writing development, 90%. 3 8,1 Literacy Theory and Research 384 Table 1 List of Strategies Select for the Basal Reader An, ,lysis With Their Associated References Comprehension Development Literal, inferential, and critical activities; pre- and post-story discussions; repeated readsmall group book ings; retelling stories with or without pictums or props; shared book readings; resolution (Anderson, Mason, & Shirley, 1984; discussion; use of setting, theme. plot episodes, i,s34, 1985, 1987; Teale, 1982; Teale & Sulzby, Bower, 1976; Crowell & Au, 1979; Morrow, 1987). Concepts About Books and Print Attempted readings focusing on illuvrations and print; asking child how one begins to read; connecting oral and written language, differentiating print from pictures; differentiating words and letters; figuring front, back, top, bottom of book; relating print and pictures; talking wondering of title, author, illustrator; tracking print, turning pages; using environmental print; Sulzby, 198r, Mason, 1980; Morrow, 1987, 1989; what books are for (Clay, 1979; Combs, 1985; Tovey.& Kerber, 1986). Language Development Following directions, speaking in sentences; seeing syntax; tying language to meaning and 1973; function; vocabulary dzvelopment (Dyson, 1984; Halliday, 1975; Morrow, 1989; Smith, Taylor, 1983). Reading Attitudes and Independent Reading Checking out books from ciassroom library, children reading to each other; children sharing reading; content activities related to literature; cooking with books; involving parents in reading program; literacy in play, literature selections in basal; literature used for skill development; self in recreational reading time; storytelling; teacher reading to children; teacher reading to repetition, familiar sequence, cumulative patterns); school; using predioable elements (rhyme, 1989; using school library kAnderson, Wilson, & Fielding, 1985; Bridge, 1982; Cullinan, & SebGreaney, 1980, Holdaia ay, 1979; Morrow, 1982; Morrow & Weinstein, 1986; Stewig este, 1978). Rich Literacy Environment Book checkout system, book-making nu.erials; feltboard and felt stories; literacy center (library corner and writing area), multiple copies of books; open-faced bookshelves; pillows; READ posters; rocker; roll movies; taped stories and headsets; varied types of literature (Big Books, biography, concept books, fables, fairy tales, informational books, magazines, newspaper, novels, picture story books, poetry, realistic books, song books, and wordless books); writing materials (Bissex, 1980; Goodman, 1984; Hiebert, 1981; Holdaway, 1979; Ingham, 1981; Morrow, 1984; 1985; Morrow & Weinstein, 1982; Teale, 1978). Word Recognition Skills and Phonemic Awareness Blends; digraphs; environmental words; final consonant sounds; idetification of letters; initial consonant sounds, thyme; sight words; syllables; syntax; visual and auditory discriminaLomax tion; vowels; word familits; word taught through conter liiiebert, 1981; Juel, 1990; 1989; 1979; Mason & Steward, 1990; Schickedanz, & McGee, 1987; Mason & McCormick, Sulrby & Teale, in press). Writing Development Cluldren sharing writing, conferences; copying, editing; functional writing; invented spellof experience irg; pre-writing discussion, recreational writing time; revision; story dictation; use (Clay, 1975; Dyson, 1984; Ferreiro, 1978; Hamm charts; using writing folders; writing stories & Burke, 1980; Hiebert, 1981; Morrow, 1989; Sulzby, 1986). 382 Bards and Early Literacy 385 Table 2 Frequency of Occurrence-Elements of Comprehension Main Retell Story: Pictures Or Pr.tps Retell Story: No Pictures Pre-Story Discussion Post-Story Discwion Setting Theme Plot Episodes Resolution Repeated Readings Shared Book Readings Nanation Literal Activities Infetential Activities Critical Activities Small Group r. ussion Totals 12.18 9.27 20.34 25.21 12.83 6.83 Kindergarte Suppi. 4.27 4.83 6.50 3.00 6.42 1.25 8.50 3.17 21.68 2.80 8.33 81.83 68.50 38.56 8.92 2.27 7.83 2.83 11.67 25.55 23.17 6.83 328.95 111.09 3.42 1.25 ist Grade total Main 16.45 10.71 11.50 16.00 11.17 10.95 43.67 23.31 11.83 17.67 11.17 12.50 8.50 7.33 6.17 14.1.1 26.84 28.21 19.25 8.08 11.92 4.42 23.94 10.63 11.17 20.50 Suppl. l' 13 6.67 10.53 11.55 11.08 Total 21.88 72.45 28.83 66.98 20.33 25.00 17.33 19.17 31.03 28.21 26.00 16.67 14.92 81.89 63.63 34.67 34.17 15.75 40.89 27.39 12.67 10.33 97.00 53.55 37.72 440.04 400.92 211.93 612.85 93.50 94.05 61.73 1116.55 RESULTS Comprehension Development Table 2 presents the frequency of elements in comprehension development and its subcategories. Generally, these elements appeared more frequently in first-grade books than in kinderwten and in main lesson plans than in supplements. Literal and inferential questions were the most frequently used categories said outnumbered Fitical or app.led questions throughout. While pre- and post-story reading ssions were suggested a great deal, the following techniques were suggestd less frapiently:,story retelling, attention to story stucture (setting, theme, plut episodes, resolution), shared book readings, narration, repeated readings, and small group discussions. Concepts about Books and Print Table e presents the frequency of elements that conwn concept show. ooks and print. Elements in this category appeared for the most part more often in firstgrade books thsn in 'undergarten and in main lesson plans than in supplements. Elements appearing lost freytently were relating print to pictutts and discussion of stoty titles, and thor4 appearini, less frequently were attention to book elements such of as front, back, top, bottom, differentiation between print and pictures, discuk illustrators, page turning, direc ionality, connections between oral and written language, differentiation of words from letters, environmental print, attempted reading, and asking children how they learn to read. ' 7 Tel* 3 Frequency of Occurrence-Conrepts,41)003oolcs:4Priitt Main 1p4scioB04.%*eizai 6.83 4.50, 7.50 .trik Back, Top tc._Bottokii .0.75: 5.25 16.75. '19,75 17:27 41:55; '3.10 3.17 8.00 5.50 12:40 8.75- 6.91 4.00 3:42 0.00 0.00 0.50 0.00 5.00 4.00 3.11 5.25 5.33 5.27 8.56 6.60 oo 4.11 5.83 5.00 2.08 7.50 omo ..26.61 42.30 Diffeientiatet_Print & Picture, 'nd-Pictine Related 12.75 13.85 titk- 3.10 3.17 7.50 5.50 Atithoi111u-stator Page Turning, Where to Begin Reading Read Left_ tnRiglrt Tracking Print Connections Between Oral & Written Language Differentiates Words & Letters Uses EnvironMente Print AttemptectReading-Illust Attempted Reading-Print Asks Child How One Reads Totals _SupOL 11.91 7.11 4:42:., 6.f' 9.36 7.00 11.17 10.27 10.64 10.67 6.58 10.00 omo 7.12 130_ 148.91 159.82 14-.10 0.00: o.u7 10.67 10.29 1.67! 4.50. _ 1233; '; 1 3.83,- -410.833.83 .14.50= 3.625:96 7.467 omo 1404 69.12' 228.94' Language Development 5~. Table 4 presents the frequency of elements concerninglanguage.4eVelopMe4 Frequtncies were greater in first-gkade books _than in kindergarten:for,all'elernents'in : this category and in main lesson plans ghan insupplements.MostireqUeni;Were.genL, - concern* vocabulary and following direction& Second most frequent** Spealang,, in sentences and relating language to meaning. Least frequent were items conceinint, , syntax and relating language to function. Table 4 Frequency of Occurrence-Language Development Main Vocabulary Follows Directions Syntax Speak in Sentences Tied to Meaning (Units) Tied to Function Totals Kindergarten Suppl. Total Main 64.25 52.84 9.89 42.73 35.77 26.03 44.33 19.50 6.00 21.83 20.00 8.67 108.58 64.56 55.77 34.70 75.47 62.25 27.78 43.67' 29.25 28.08 231.52 -120.33 351.85 26630' 72.34 15.89 1st Grade Suppl. Total 67.13 14.59;-. 26:20 26.59. 36.42 19.66 '53:98` 70.25! 65.67 47.75, 208.41 474:99: , -. "- - 'fai:614141d4i rable5 _regifenCy cf -e) c c ur ren c e -R ea ding Attiuidcs anctItidepent,R-egdt:ng. **e*iiiii: Rtióa ado tei*eraicadici rot*, :C111"*4***04, 27.17 ron*if*XiVirieS,Rslated to Lit. 10.70 tic Beoki Reid. liuig CthBs iotach Other -46.0 8.0. -0.00' 33.55 1.00 3,67 8.50 2.58 3.17 2:67: 44:25 . 8:90, 2.67: 2.67 0.00 2.00 5.36 1.83 0.00 LOO 4 6:61 113-.S6 4.42 3.17 8.0 18.0a. 5.50 16.17 28.27 26.99 8.93 1.50 7.50 126.44_ 222.84 160.18 1.50 PiediCtithletieinents:Usedi Rhyme, 'dim:dative Patterns Repetition, 11.51 'Traeher,Readi in School Parents & Reading 7.83 17.83 4.83 0.00 2.00 Tétals 96.40 Nuinber of Liternture Selections LiteraturefOr Skill Delielopment 4.67 20.44 9.16 4.10 1.50 11.88 25:50 6.50 5.25 6.00 193.90' .- 35*.6:8- Reading Attitudes and Independent Reading Table ,.piesents the frequency of elements that concern indePendenp *ding; and; aOtudes about-reading. Frequencies were,- greater in first-,graden'ibbohs,lhan, in '1.0.44firartP! al01.9110 FPO of the elq111511ts. in thi cgcgory, appead mq f#.4115.41.),r; in siipPlementary seCtions- than in mainlesson plant. Wit: frequent:ietiyities*Ok teachers' reading to students and actiVities that related toliteratnie, With 04404* punt next in frequency. Less 'frequently -suggested were recreational-reading, book sharing, and storytelling, among others. Rich Literacy Environment Table 6 presents the frequency of elements that concern a rich-literacy environment, trequencits were greater in first-gi.4e books than inldr4ergarterrand inanPple., mentaly sections than- in main lesson planS.:MoSt frequently,a4,0441' were the uie of taped :'Stories .;.and-, headsets, bookolng and; in first gra4, . ris,adirit postrs'. Different types of literature appeared fat More-freci*Mlif.M.firat-graijehOOkk. , than, in- iihdergarten. -Genres moat represented; were OciMe stOrk:.boOks2,and,PdetiY, fOilowectly,,fairy tan, inforthationaniteraturi, Big Books, .fahlet, and-,realistie Lo*cyTable 6 1re4tien4 of Omni mce-=Rich'Literacy Eiwironinent Main et Total "Literriery CenterIL_Rirazy Corner 0.00 3.011; ,3..00 Book cheC,k;Otit SYStern _2:30' 0.00- 0.00. 0.011 01*#-F.10.41Y*.WI',a 0.00 0.67 0.00 0.00 4.33. 0.00 0.86 '7,4:-W..**4-4*- , MuNpleC.otties Of Books Fe*irfct & _Felt-Stories RolEKtivies-- TipertStories & Head Sets Pillcivis Rocker Read-Pagers Writing Materials Book Making Materials Types' of Literature Big'Books Biography Concept Books Fables Fairy Tales Informational Magazines Newspapers Novels Pict= Story Books Poetry Realistic Song Books Wordless Books 1.33 0.00 5.67 0.00 0.00 1.33 0.33 2.43 4.39 0.33 0.00 1.67 5.33 0.50 0.00 0.00 0.00 S.39 7.07 0.3:. 0.67- 5:67 483- -0318:, _ 00, 11.37 1.50 2.17 2.10 1.83 6.17 5.36 4.67 4.50 4.05 0.83 2.67 5.00 6.33 3.58 4.50 5.33 7.07 1.50 7.23 0.33 3.58 6.17 7-.4-5i 10.67 10.63, 16.29: _ _1344' . 0.50 0.50 1.33 1.33 21.89 5.67 4.30 0.00 0.00 0.00 4.78 4.17 1.33 10.01 1.33 15.67 0.00 0.00 7.17 7.83 8.93 0.00 0.00 8.50 8.17 2.83 0.00 13.50 12.33 3.67 7.57 1.50 1.67 1.171.83 1.67 3.50 41.95 101.76 143.71 33.37 19.41 4.00 :1.33, -7.42. 6.00:=__ ,-_. : 0:00,,. .-. 'ili.ii 6:06- - :i 436: 4.566.33" '16.83, ,- 12.54 12.39 5.72 1.58 1.50 0.50 ...133.:-:72'7.- 0.83" 9.24 8.67 6.33 1.67 10.50 40:30: --' 1.61-. 9.17 -8.61' . 6.33 -1:67: 23* , 12.66- 1732.50 .i.P 125.41 '13:44,: 4.:08: .39 "./--: ...,,-_ ..: 268.78 Phonemic Awareness and Word Recognition Elcills Table 7 presents the frequency of elements that concern phonethic aWareness and word recognition skills. Frequencies were -gfeater in firstliader:books ,.than 40. kindergarten. Visual and auditory cliserbnination, :environmental sight wOrdseidentifi; cation of letters, and thyme, however,- appeared: tho're°' ofte'a 111-k404140a?0,1*-Most elements appeared more often iii _main lesraa plant thaii in suPpOrneatarSinateri, Ws.- Most freqUently suggested'elements:, Within -the.- Icingergartei*aterjOttpficeOed: letter_identifieition,- initial, consonant scrup4s, and ,visuaI the first-grade texis, initial consonant' sounds, vowels, identifiCation ccf letterbien4S, and final consonant sounds. . 9111101 Tablel FrOUency of Occurence-Word Recognitiou:ond PhOnentic:AfForetlesf -K.o.44arienMen '§1001r '4-0-3.1` 11442# PiSaithiri/Oon '4411.r:WOr* Words ?dentifinadon of,-Letten ,lisitia1COnsonarit Sounds final Consonant Sounds yo*els 13104 Pitr.iPhs 'Wird Families ithythe Syntax Syllables Words Taught with Context Clues Totals 4567- -24;59: 34.98' :151:00: 8.50. 611 71.77 64.34 23.00 15.00 14.17 4.17 9.17 14.88 22.33 0.00 23.73 25.71 25.51 3.33 377.30 1.33 0.50 5398 , 47482, 41020, -1467' 97.48 60g, -14: 89.85, 26.33: 16.33 14.67' '12.83:.. 48.11: 5117:- 5.50 NA.; . 51994, "4:-,891-,7., 1.33 1.58 19.00 3388 15.71 15.21. 4.08 0.00 33:33 26:42 0.0057.06 39.83 3.1.00' ,f:1083 186.18 563.48 10.75- 22:61' -13*" 18331 32.87_ ,19.171:,.,62,43; 566,44- '*:* **: Writing Developmetu Table 8 presents the frequency of elements that concern voting. GenernilY; fre., quencies were greater in first-grade hooks than in kindergarten, and for the Opat-i*tin supplementary sections than in main lesson plans. 'Most, freqUeritli,SUggeit#in kindergarten materials were wriOng st6ries, gm) dictation,,a0 Oe'FcP: of 40-11!efiPP charts, and, in first grade, writing stories, story dictation, and shared Writing. Summary Table 9 summarizes the resultc of all aneyses represented, above.- Genet:01y, in both first-grade and kindergarten books,..greatest frequencies were: giYeii 1ordrecognition and phonemic analysis, comprehensiim Oeyelopinent,And languige,deyelif, opment, in that order. In all three caregories,-niore erdphisiS wins pit4441f11:.4-iiPlge books than in kindergarten, and orrelements in main lassOnplani.than,in suppleinentary, sections. Next most frequent were, elements:in writing Aind in, readingatfitUdes and independent reading, more often in-, siipileMentart Seetions,thanin,niairi-ieSson, plans, and ,in first-grade books than in ,kiridergartsii.;LenSt freqUent*aitggesteitlkere cpncepts, about ,poks and-print, and rich fiteitiO,cniiionninent,:the, former mOre often appearingirt main lesson- plans, the latter in-sopplemetit4tik sections Praginatically,;it must be notetthet,tear&ei-do not use, Sitiiipkrnentary sections often, either because of lack of tithe or becausetwy feel they are not as-important as main, lesson plans. *103 , Frequency of Occurence=-Writing'ppylopitieffr, :1st:Grade:. Main iitieiiinionid*dng.Tirn-e -Kr.i1.#400.,0 Stoili'DiCtatiOn :ce9iP8: F4titiii#5e. Charts Om: ---Tiii., 10:67' 13.99' 2.83- -18.K 13.83- -9.48 11.25 5.0 , 8:5-6.".' _ 33-.38 23:St ..1312; 3.83 Wiiiii*FOlders &notional Nriting O*1 Stialling 1.33 583 5.76 2.83 8.84 6.50 3.50 3.00 7.17 8;59 13.87 10.94 10:50 8.67 3;50 494 3.00' 17.89 1.67 1.67 Totals 1.67 2.17 0.00 0.00 0.00 56.1 165.4 !r.f3'41=;. Tg.5:1 1940: SJ*Ivritigi PreWritink Discussion Refision Conference Editink Sul 15.68 8.09 161.4 " DISCUSSION Results of the study indicate that the basals analyzed do incorporate bOdatra:cll=-., tional and more recently espoused strategies for develOping literacY ineIyhildhpo Such cturent strategies, however, is prOviding rich litetn4 eny,:iOnOents,J0gOstino., writing activities, iiroinoting positiVe attitudqa toward iiingingi,anct'e46904sing#1,4-.,:. pendent reading were not is well representedilS the MOreiraditionat **a/0i_ appeared more often as supplententary idea§ rattier dvin aS ,insitxtessi*plan **OS. Although ctirrent concepts of,emergeritliteracy,:Suggest,conettirent:devetapinent,ot reading and- writing, with equal- emPhasis on itiategiesifOr.thedeVelOpOnt.Ofbotiii, word recognition remained the preetninent skill to,be deiTioped-Withini:ti**00., Informally, the- research analyst& fonnd mtich ,to praise in:.the-basat **As:. illustrations and formatrwere- aimed*. and; Would engage ,nioSt, Children, ..actiVities, were varied and interesting;-and inuclvef:theliterature inchided *ascii high qoaritY::,,, On the other hand,-the numbeyand-cainplexity of lessen 14ans-Were **helm= mg, prompting-fears thatterchers Would'Iose sight of priorities lan&feet inadequateto follow- every plan 'wall its dctail. Althotiih selectioni represented some ot "di best of children's, literature, they were sarrounded-brmunerous Skill tessonsspthàtone wonders if the joy of literature coald'inryive. The results of the-study miseseerat questions about the natyre and role of beast readers: (a) Should tbe- major roie of abasal remain it& traditional oneitoproikkr the -direct-instruction. of skills?(b).Shettld liters:Ore* :usertior,SIditinstructl*.-CT... should it:- be Suggested : within the basat- primarily: for- ntetivation, for,r,044ng trade books? :(C) 'Shotild :basal in'ograms- seynainf iliglisy-protiiiiiiive or shotdd-diey..Ociflie, into process and resource guide&by,:oriiicktenc:hers'Oan design their ciwriiiiitroetioitik c. , , ter alik 0-s Sa*mary4Early Literacy Categories - . Welfiii)011 ... :#,e-4!(.1.inIgAtiiNike zRich'Illiterlicy-; Environment Wri:ci "ii911 , _ Itsaii ..8n4 Writing DCVelopment tili*lie* : 1:4"-.4rge -8iippl _Total, 328.9 106.6 11,14 f440.0 148.9' 10019' 231'.5 964 129.3 126.4 351:9, . . ,266:5,.... , ,:1602, 41.9 1013 143:6 377.3 56.1 186.2 105:4 563.5 161.4 Main -c.:0,0or,pw 0. Book* Print Langiiaage_:Oelopment Icfick_rgFten 423 222.8 '04 83.31. 5004 198.4 luf,, -694, :- 208* ......, fi!2.,13:,.. ;,228.9_';,-. J`ti14,.,-; .. 193,..9: ;-..-.3.54t '12.5.4: '400, 3-30- .010., ,1904.:- -*8`.9 strategies? (d) Can basal materials do justice to a whole lae;'roadi,orthight it mom effectively be nurtured by the teacher through.EteMay:deyelnPinent throughout the entire sehopl day? (e) Eecause theY ha* been Succellfulin:in4pneat,, ing many;of -the- newer strategies into their miterials,,',Can ensuing-editions continue that trend and,bioaden program one that includes materials for skill- developinent al 'we iM6W,"diern, along with numerous separate paperback selections of children'i,literatuie,relatiVely free Ot' -prescriptive instruction. The children's literature wouldbc considered as ir.iiiP§rtant as, the rest of fig skill materials by giving equal emphasis through.the,thne aliOtedfor their use. Whatever the answers to these questions, it seems_obvious that anY reading,pipt, gram in early childhood must respond with at least equal eittPhasis'to,eyiden4e cited in behalf of emergent literacy as it does to traditional aiguinents for skilfleyelOpment: Basal publishers have the resources to make a difference. Educators andeducatinind , researchers need to make sure they send publishers the best message about howtteracii; learning takes place. REFERENCES Anderson, R. C., Mason, J., & Shirley, I.. (1954). The reading group: An experimental investigation of a labyrinth. Reading Research Quarterly, 20, 6-38. Anderson, R. C., Wilson, P. T., & Fielding, L. G. (1985, December). A new focus on frce-re4Xling. Symposium presentation at the meeting of the National Reading Conference, San Diego, CA. Bissex, G. (1980). GNYS at work: A child learns to write and read. Cambridge, MA; 1farvard University Press. Bower, G. (1976). Expc, ments or story understanding and recall. Quanerly Journal of Experimental Psychology, 28, 511-538. Bridge, C. (1982). Predictable materials for beginning readersLanguage Arts, 58, 103-107. Clay, M. (1966). Emergent reading behavior. Unpublished docoral disscrtstion, University of Auckland, Aticklaid., New Zealand. Clay, M. (1975). What did I write? Auckland, New Zealand: Heinemann. '14.2 Clay, M. (1979). Reading: The patterning of complex behavior (2nd ed.). Auckland, New 7ealand: Heinemann. Clay, M. (1985). Early detection of reading difficulty (3rd cd.). Portsmouth, NH.: Heinemann. Combs, M. (1987). Modeling the reading process with enlarged texts. The Redding Teacher, 46;423-426. Crowell, D., et Au, K. (1979). Using a scale of questions to improve listening comprehension. Language Arts, 59, 555-570. CuWman, B. (1989). 'Litenuore and the child (2nd ed.). New York: Harcourt, Brace, lovinnVich. Duffy, G., Roehler, L., & Puna.% J. (1987).,Putting the teacher in control: Basal reading it...litho-Oki aid' instructional decision making. Elementary School Journal, 87, '351,366. Dyson, A. H. (1984) . Emerging alphabntic literacy in school contexts Written CommunicatIon ,,`I, 5=55. Ehri, L. (1989). Movement into wordsreading and spelling. In 1.'14: Mrison connections (pp. 65-79). Needham Heights, MA.: Allyn &Bacon. Farr, R., & Roser, N. (1979). Teaching a child to read. New York: Hartonn, Brace, Ihvionvich. Ferreho, E. (1978). What is written in a written sentence? Journal of Education, 160,14-49. Goodman, K., Shannon, D., Freeman, Y., & Murphy, S. (1987). Report card on basal readers. New York: Richan3 C. Owens. Goodman, Y. (1984). The development of initial literacy. In H. Goelman, A. Oberg, & F. Smith (Eds.), Awakening to literacy (pp. 102-109). Heinemann. Greaney, V (1980). Factors related to amount and type of leisure reading. Reading Research Quarterly, ., 15, 33 -357. Halliday, M. A. K. (1975). Learning how to mean. Exploration in the development of language. London: Edward Arnold. Harste, J , & Burke, C (1980) Examinir instructional assumptions. The chilo as informant. Theory into Practice, 19, 170-178. Harste, I, Woodward, V , & Burke, C. (1984). Language stories and literacy lessons. Exeter, NH: Heinemann. Hiebert, E W (1981) Developmental patterns and interrelationships of preschool children's print awareness. Reading Research Quarterly, 16, 236-260. Hiebert, E , & Papierz, J (In press). The match between kindergarten reading materials and the emergent literacy construct. Early Childhood Research Quarterly. Holdaway, D. (1979). The foundations of literacy. Sydney: Ashton Scholastic. Ingham, J. (1981). Books and reading development. London: Heinemann. Juel, C (1990) The role of decoding in early literacy instruction and assessment, In L. Morrow & J. Smith (Eds ), Assessment for instruction in early literacy (pp. 135-154). Englewood Cliffs, NJ: Prentice Hall. Lomax, R G McGee, L. M (1987). Young children's concepts about print and readmw. Towards a model of word reading acquisition. Reading Research Quarterly, 22, 237-256. Mason. J (1980) When do children begin to read. An exploration of four year old children's letter and word reading competencies. Reading Research Quarterly, 15, 203-227. Mason, J & McCormick, C. (1979). Testing the development of reading and linguistic awareness (Tech. Rep. No. 26). Champaign, IL: University of Illinois, Center for the Study of Reading. Mason, I & Stewart, I , (1990) Emergent literacy assessment for instructional use in kindergarten. In L. Morrow & J Smith (Eds.), Asse-sment for instruction in early literacy (pp. 155-175). Englewood Cliffs, NJ: Prentice Hall. Morrow, L M (1982) Relationships between literature programs, library corner designs and children's use of literature. Journal of Educational Research, 73, 339-344. Morrow, L M (1984) Reading stories to young children. Effects of story structure and traditional questioning strategies on comprehension. Journal of Reading Behavior, 16, 273-288, Morrow, L M (1985) Retelling stories: A strategy for improving children's comprehension, concept of story structure and oral language complrxity. Elementary School Journal, as, 647-661. Morrow, L M. (1987) Promoting inner-city children's recreational reading. Reading Teacher, 41, 266-274. Morrow. L M (1989) Literacy development in the early years Helping children read and write. Englewood Cliffs, NJ: Prentice Hall. 39 Ogifelq. tes,, V V VV V V 231, IMow LM., Weinstein, C. 5.;(19E.1.2) t;InSeafintntIii.44W,81.11.e.PtAteraPre tbOugh'PO87a'4,114 .'phisicil design Changes: ElementiseySehOOlojiiiat, 83,431,-131. .MortqW, L M *._Wekisistei,i1Jypyit!t reading The: ,29f Prorpain o'n children's, uSe Of library centers:Iteridint Research Quarterly ,- 2,1,,,33,0+.346,-, L. Morrow 090: The P;ace of,si:eeille tkilia,MOniieliOOl and kindergarten Ii D Strickland siliteracy: Youni ehi:Mrin learn to rend a;d write (iip: 96=106): *4; - 4 0E64 tire treatiaini of COMinercial reading Materials in reading methods textbooks. Reading Work103;-40.r.157:-(19.83titire use-of iOmasercialreacling materials in American elenienta..-ry:41O-*/2_ !ReSeafchdiso:iiii, 19 68=85: 'Sharrinni; 0.-(19,87):COMMereialitidinkmaterials, a technologieal ideologyiaticiffie#tt114* of teachers Erfeinen46:: so,* triuris'a4.:90, Smith; F. (1973)'.;:iisAcilingulstics andreading. New York: Hólt,,Rineiart & Win460. Stewis, J. V., SeSehesia,, S.- (Eds.): (1978). Using liteianire in the ilenieniary clasSropm. UrbanaIL: NationakcOnnell; of Teieliers of Engliah. Sulzby, E. (1983). Children's'emergenureading of favorite storybooks. Reading Research eitarterly, JO; 458-451. Sulzby, E. (1986). Kindergartners as writers mrd readers. In M. Farr (Ed.), Advances in Writing,resegeh: (Vol. I.) Children's early writing. (pp. 127.200). NorwOod, Ablex. Sulzby, E. d.. Teale, W. (In press). Emergent literacy. In P. D. Pearson, R. Barr, M. L. kind!, P. Mosenthal HandlwOok of reading research (2nd ed.). New York: Longman. Taylor, D. (1983). Family literacy. Exetes,t111: Heinernann. Teale, W. (1978). Positive environments for lemming to read: What studies of earl/ readers tell us. itinguali Arts, 55, 922-932. Teale, W. (1982). Toward a theory of how children learn to read and write naturally. Language Arts, 59, 555-570. Teale, W. (1987). Emergent literacy. Reading and writing development in early childhood. In J. Readence & R. S. Baldwin (Eds.), Research in literacy: Merging perspectaes (pp. 45-74). chicagO, IL: National Reading Conference. Teale, W., & Sulzby, E. (1987). Access, mediation and literacy acquisition in early childly.rod. In I). Wagner (Ed.), The future of literacy in a changing world. (pp. 111-136). New Yorl.:: perganiOn Press. Tovey, D. R., & Kerber, J. E. (Eds.). (1986). Roles in literacy leanting: A new perspective. Newark, DE: International Reading Association. Weaver, C. & Watson, D. (1988). Basal readers and the state of American reading instruction: A calor action. A statement of the Commission on Reading, Urbana, IL: National Council of Teachers of English. Yanngton, D. ,1979). The great American reading machine. Rochelle Park, NJ: Hayden Books. ,'T-7:7',!,'rk TOFViGSINCLDBASA1EADING PROGRAMS, KINDERGARTEN THROUGH; SE-COND -GRADE: AN INVESTIGATI9N OF CHANGgsylacai19s3it)49$9 James FloOctaiscl ianèLàp San Diego State-University The types of writing that have been includt4 in. hitta1 rr-1,141",s,,f110:040gir throughout our history. They have changed from an exchtSiVe deflection of rrelOis treatises- to a More eclectic Collection of Writing:typer- VeneZkY (1987) has, Aiggi#tpd that tho tyPea of writing hi ;ea4int prOgrit.Ps political, and economic constraints ,of the era in, Whick,Mey*ere :POhliihed; Chall (1967) and Smith (1965, 1986) have also argued that the:Content "ikhasala htta..4e- pended on-the instructional ideology of the author and/or .publisher:it,,the thue,Of publication. Discussions about the types of writings- to'be included in basal*programa aeeriito. be tied to one of three positions; these three position's often.initaince.404 OeleetiOns and reflect differences in instructional emphases within the hasal Pregtains.,,PrOPonentr of the flrat position hold that basal readeri are priMarilyinairtictional tOolt Or teaching: decoding,strategiestools to be-used and then discarded whinattidentsliave ifia**cr, beginning 'reading skills, and can read"real" books. In' this- View, tho-aeleCiiOniand . .developrnent of the content of basals are subjugated to the learning of doCedirig Basal readets designed from this perspective often contain brief constrained teita-and are quasi-literary. The second view of the purpose of the basal reader in inatruction is to introdUee children to "quality" literature from .their earliest, Years. Altbonglr:thOie this viewpoint may also believe tha. .asal readers,sheuld include proviaiOns jhrdeitei= opment of decoding strategiea, they emphasize that basair aro -essentially YitaSary anthologies and should include "proven," "miality" literature. There proPonents, argue that basal readers should be the textbooks for the "discipline" of iiteratitre;.and that euly and continuous exposure to qUality literature ensures a lifettitic habit of reading. Those who hold the third view of the role of the basal in instmetion 5elieVp, that basal readers should offer instruction to ensure all, 'tippets of reading conipe-tence, ranging froth the ability to decode to the ability tolearti frOm Ottiplei teitr..:1)4o, nents of thia view' argUe that reading,instiuCtiOn, is ratili Offeredotitalde, of the-timer designated for tile instrUction herefore,:ehildien need: to' read'..a ,4riestt (*keit types, and they need, to be taught the skiiirthat,they will tie«, tO toMprehend, content area goa teehnicat textbboka as WO as-literature.. This,: poaitiatt adVOCater that basal traders should contain a wide variety Of **tug tyjfas knd dactialeatafuictudiag " 'I,' :396- charts, schedules, diagrams, and graphs) so that children will be exposed to and will learn to read a wide, variety of text types. It seemed to us that the third view most closely mgached the claims that contertpo7 rary basal designers and publishers made in their promotional materials. We collected sample statements inclUded in teacher's manuals of widely used basal reading programs: . . [contains] a variety of literary forms. . . [offers] a variety cf selections including poetry, fantasy, realistic ..4on,and non-fiction. . . [provides] a balanced represerestion of traditional stories, fables, iolktales, fairy tales, myths, legends, fantasies, plays, songs, and poems, as well is conten3porary stories and articles of various kinds. . . [contains art] ideal balance between fiction and appropriate factual material. In addition to fiction, plays and poetry, the program contains varied and interesting materials designed to impart knowledge and help pupils appreciate good 'Writing in all forms. . . . [includes] literature that encompasses every form of writing, from plays, folktales and poems to fascinating articles in social studies, science and other content areas. STUDIES OF TYPES OF WRITING INCLUDED IN BASAL PROGRAMS PRIOR TO 1986 Data from investigations of basal content, however, do not support these claims for breadth. For example, two studies showed that the majority of selections-in basal readers were literary, that is, stories, poems and plays (Durkin, 1981; Olsen & Dinner, 1976). Both of these studies found few expository, informational selections in basals. In fact, Olsen and Dillner found that 90% of basal selections were stories. The =stilts of these studies, igowever, were disputed in a more recent study by Schmidt, Caul, Byers, and Buchmann (1984). They found that stories represented only one-third.of the total selections, and that information articles accounted for thc same number of selections as stories. However, these investigators examined only second-, fourth-, and fifth-grade texts, thereby rendering their data inconclusive. A STUDY OF EIGHT BASALS, 1986 To build on this work and to address the question of the types of writing that were included in the 1983 basal reading programs, Flood and Lapp (1986) examined each book from preprimer to second reader for eight basal series published in 1983. They chose to examine these eight programs because each was approved for adoption in California and each had a comprehensive program (i.e., materials ranging from readiness level to Grade 2). The programs included: Ginn Re4ding Program, Harcourt Brace Jovanovich Bookmark Reading Program (Eagle edition), Holt Basic Reading, Houghton Mifflin Reading Progrnm, Laidlaw Reading Program), Lippincott Basic Reading (Harper & Row), Macmillan Reading Series R, and Scott Foresman Reading 39a '-'1:ypes of Writing Included In BasatheadintProgrants Program..The results from that study indicated that,bisal progrania Were still overyThehningly literary Txtotal,pereentage of literary selections (narratives, pberas and plays):was 76%,(491A) Storiei; 23% poeins;4% PlayS),,stid*OnlY:13% of the selectiOns 'NY* OtiecificaW-nqa.tefarY (6xPo.sicorYriafothizgoFt 1400, A 5.00.40*.Y.1,11Y14 Of the.data; counting the nUMber Of pages, alio* to each tYpe of,Writing, indicaterkthit ea-programs weie eveninore literary- in nature thitii.origipally_lhOugh41396 ..ourt* pages contained literary texts, only 896- of the pages contained exposifOrWinfOrmition trats. AN ANALYSIS OF CONTEMPORARY BASALS, 1089 We decided to do a follow-up study to compare contemporary, basals i.Vith!the basals of 1983 to see whether basals had changed. Wordd thfty still be overwhelmingly literary? To do this, we examined eight conwrehensive prom** that were approved for adoption in California in 1989: (a) HaMourt Lai-mute Pro-00i, (h),Harceurt Imagination Program, (c) Heath Reading Program, (d) HoltReatling44Ograria, (e) Houghton-Mifflin Reading Program, (f) Macmillan Connections, Program, (g) McGraw-Hill Reading Program, and (h) Scott, Foresman Reading Program. Method System for classifying types of discourse. The same classification scheme as used in the 1983 study was used; it included six general categories: (a) Narrative (Fiction), (b) Poetry, (c) Play, (d) Exposition, (e) Biography, and (f) Hybrid (the form of narrative and the funNion of exposition). Procedure. We Jetermined the variety of types of writing in the 1989 basal programs by using ine same two analyses that were used in 1983: first, we deterniined toe number of selections representing the six writing types; and second, Ind pasilly a more accurate reflection of basal content, we examined the nuinber of pages devoted to each type of writing. It has been argued that a single measure might not aCcurately reflect the overall content of a particular book because it is possible that a text Might contain one lengthy expository piece and five brief stories. If !he first tally were used exclusively, the basal under inspection would appear to be essentially narrative; conversely, if page counts were used alone, the readers would seem to be heavily expository. Results The scores for the total number of selections for each type of writing included in the basal readers are presented in Table 1. The most frequent form of writing at each level was narrative, with an overall mean of 49%. Expository writings were the second most frequently included type of writing, accounting for 28% of the selections, and poems accounted for 22% of the selections. In Table 2, the results of the comparison of the number of selections included in basal programs between 1983 and, 1989 are presented. The data indicatet; that the programs have changed considerably since 1983. 39 :raamnarra ; :/ r).; -1- - . .s r- Tablo-1' ((ina erce kia84- Ô1 SE 11-7511* ofRithlg Thit'utk &wild-Gm* Rfia464, P89. 14****4 . -, 144fPri* Secoad4rode: 311:,(52) 2i1 '(.0);- 14 (2 .0, 43-(30, _ -Plays- 4 -(0) Yexic!.. 3 If* _ , , '9 (3).- "(0) 1.NOgroPhy _ 10): 4 10- "(0) Table'2. - A,ComPoris'ori af the P.ercentoge_of WritineType Sekctknistri98 31sn44009:4114Sai;Rcaderi (PreprinseriThrough Grude 2)- TofWdtpg 1983 51 25 4,a0sition 1141 . =22: 4 0 MosisPhy Hybrid- 0-- 49 ?tti 4 1 -f The data indicated that the0 had heed a Substantial shhtt in thenthnheeof 490Pr-aYlCa.ta. 424 -9/14t inclo4c4 In i)atal MO* folio 1983 . C.0 4970.`74.-110466:, doubled:(14* to:2816). HO/sib:T. the data &Om the scOr410.4s4*,-4TeP11110e.r, of PAO dP,v9tP4 to tack-air-414 -ty0e, did not a*tritiat-O'titia-faid418;_:Tha: of page* allotted to narrative, for exanople, Aciwid.that the *oks *ft 72% nrst1ve aod 'a_tat aYeatil.1.44110gly.4*ani: (*airy 0004, 141).."*ASec Tal#1-41 The cpiisArisatibetwied the -1923 =41989 Pectia4s'aba*ed_that littia-chfiage had actually oOciured;.41,Table 4; the dria-illuittate chit lack of chine. pie Nicks remain more thza 80% literary. Discussion Our-inspection .shows-that -basis are still predominantly, litcraiy, .with-n%.ot the selections ana 81% of the pages containing. either narratives, .-Peenis) cc plays. EimberMore, *e eoeld net, Onctr in the bial-thatiusia. one Osuaterit iiipip1,014or guiding teach** in desigahitinstruction for the different'kindi Of*riti,ng kkluiJed la the readerir The procedures rpcionitheirdij vierellotileIficiqty differ*** thoce rooOnithe-Odod tor rotor Pr!rriOvt. This is lifC4040.Cf badaraaa t$4tac-ikt-E004 otresearch lñdicazingthi rdiel4structiCsi'in 'readietexpOlito:tyteita shttld be 0.80'4 _ = rkiligThfts uiBasalReo Tahle4 A copporisort:of the Percentakeof Pages Allotted to Writingitii**44 ,. Reafltif,(ereprimers rhrOugligiode 2) . . 1983 Nitive P.,04:61 ,EXiio0fion ilifilys l!OkaPhy 'Hybrid 73 6 ii,. 11 5 1 4 ; approaChed differently from narrative:texts-if the.intcactita beiWeeu the text is to be'sticcessful. inspecting basalt fOr, genre' 00:i:e: not,,however,..addressihe isSue of what hisalkihM0 BirPferOng to Oto three diveT1011.01406400.:, of the, basalliafrUctlipP, we .can:1440 st4OP thAi-'0iir;=.-0A:.#0;41.4404dqiuMF:fr address the- tin.; paispectiyethat of baS.4aa aids u) deCo4447-. (111,2tOly"OdresiOd lvithout:4 conSider the fami4cationa of the Janet tWaiiCisitia#:0046,hasala.heptedAmtnffitlY anthologies Of-iitmiture of should they rePriseRtA 0:44i4i*ot is a balanced presentatiiin of text typeSTIn'hat-Wayi CaricP4riii be for (kalioli '1.001 booStor4zy tex.ts in their 'Sasi#V6.074ris-,' Basali as 'llteroiy tinthologiel..in:reeein Yearar.,kgril4d-cal of condncted on text aS a critical i,iariehlikaitecting crimiehen4c-ii-'#*baiA 4014P..re.,410 be ttiO'i:eXt that is,-usr4.49'i4OSSitdre.1114:**4*;t*iii*.#4,i5'44f presented ula fónnattht is ripilior to F440.::041,eti's 40,4111.1-*Orfc o,t 132*PIPAPR.,7qi.101,Pwat.s e4gc structures, which rçadcrs establish in memory through 00641.ro tO !Oji0i-C410 . readers to encode and retrieve information. His work was conducted with stories (narratives), and the abstract knowledge structures that he discussed were labeled schemata. In the 1970s, several researchers developed story models for representing the underlying organization of a story that could be u to determine whether children actually haa schema for story texts (Mandler & Johnson, 1977; Stein & Gleig4,1979;1 Thorndyke, 1977). Their findings suggested that children as yoing as 4. Yearsof,age, have acquired knowledge of the structure of text, and they use this schematic fgpvytedge to organize and remember information. From this data, it can be argue4 that the knowledge of structure enables a child to understand and recall story events. In addition to stories, researchers have analyzed other types of writing to determine the underlying structure that serves as a schematic framework for understanding text (Berknwitz, 1986; Floot:, Lapp, & Farnan, 1986; Kintsch & Van Dijk, 1978; Meyer, 1984). Meyer (1984) contends that competent readers approach texts with knowledge about the ways in which texts are conventionally organized; readers select the most appiopriate schema for processing a given text. She also points out that recall is impaired if the appropriate schema is ill-formed or unavailable. Although there is no single answer to the question of vihether basals should I= literary anthologies, we believe that an effective basal program should contain a balanced presentation of text types if one of the reasons for using the basal is to get students ready for other types of texts that they will be encountering throughout their school years. A balanced presentation of text. To determine the nature of a balanced presentation of text types, it is important to consider again what the role of the basal reading program should be. Since there is no conclusive answer, we suggest that a basal should introduce students to the reading processes they will need and to the types of textual material they will be required to read throughout the curriculum. Basals should include a variety of writing types to ensure that children will be exposed to many different types of discourse. Children should learn strategies for dealing with many types of writing before they are asked to read complex content textbooks. Several researchers have argued that this early exposure is critical to later reading success (Calfee & Curley, 1984; Lapp & Flood, in press). In fact, 33 years ago, Squire (1957) argued for teaching children how to read many different forms of writing, contending that the form of a text represents the author's thinking, and that an understanding of form gives the reader the ability to relate ideas in memory. We believe that a well-balanced basal is a composite of all of the types of writing the reader will encounter in daily school experiences. The reader should be provided with the opportunity to rehearse learning-to-read strategies with texts that are representative of those in all of the books the reader will encounter. REFERENCES Bartlett, F. C. (1932). Remembering. Cambridge: Cambridge University Press. Berkowitz, S (1986) Effects of instruction in text organization on sixth grade students memory for expository reading. Reading Research Quarterly, 2/, 161-178. 397 . tjIesofgIncisäled in Basal 12- ending Programs &lift, R.. &,..Curlex, R. (1984): Structures Of piose in content areis. IriJ.-Flcod (Ed.), Understanding 'readiniconigrekensiOn (pp. 161-180: Newark:DE: internatiotiat_icAai4A4o0iiition. 'Chall 1 ,(1967)::Lecirning to read: The great 4einiii.gesy icOlilig 'Comprehension instruetion in:fiVe basal reading series. -:,ge_ading .Reiiarch VuOrterly, 16,- 515.-544. Lapp D (1986).,TYpes of textr The, match beiween *hat `itiadents read in imisali _, . 'Flood 3 they eficOnntei in tesir Acading`Research Qynrieriy,.2j, 2844291z._ Finian, N. (1986). A reading-Writini iiioeedike 'diet hittch4expesitoiy, paragn MO; sirixtere..Rin4ing-Teeeher, 39, 556=562: Van DiAc,T.A. (1978). Toivaid a model of text compreliensiou and produCtian'firjdoietical Review,-24,. 196-214. Lapp, D., & Rood, J. (1986). Teaching students to read. New York: 1491iiilan. Lapp, D., & Flood, J. (in press). Teaching rrnding to every child (3M ed.). New Yoric Mierniltair Mandler, J. M. , & Johnson, N. S. (19,77). Remembrance of things parsed:' Story stnictire 'and *Ali. Cognitive Psychology, 9, 111-151. Meyer, B. J. F. (1979). Organizational patterns in prose-and their use in reading. In M. L. Kainit & A. J. Moe (Eds.), Beading research: Studies and applications (pp. 109-117). Clemson, sq Nati:nee1' Reading Conference. Meyer, B. J. F. (1984). Organizational aspects of text Effects on reading comprehension and amilications for the classroom. In J. Flood (Ed.), Promoting reading comprehension (pp. 113-138). Newaric;iDE: International Reading Association. Olsen, J., & Dillner, M. (1976). Learning to teach reading in the elementary school. New York;Macmillan: Schmidt, W., Caul, J., Byers, J., & Buchrnann. M. (1984). Content of basal reading selections: Implications for comprehension instruction. In G. Duffy. L. Roehler, & J. Mason (Eds.), Comprehension instruc- tion (pp. 118-131). New fork: LotInan. Smith, N.B. (1965). American reading instruction. Newark, Da International Reading Association. Smith, N. B. (1986). American reading instruction (2nd ed.). Newark, DE: International Reading Association Squire, J. (1957). The necessity of form in literature. Urbana, IL. National Council of Teachers of English. Stein, J. L., & Glenn, C. G. (1979). An analysis of story comprehension in elementary school children. In R. 0. Freedle (Ed.), Advances in discourse processes (Vol. 2, pp. 203-220): New directions in discourse processing (pp. 203-220). Norwood. NJ: Ablex. Thorndyke. P. W. (1977). Cognitive structures an comprehension and memory of narrative discourse, Cognitive Psychology, 9, 77-110. Venezky, R. (1987). A history of the American reading textbook. Elementary School Journal, 87, 247-265 398 1; . . RgAOING STRATGIES QF MARdiNALOi';iiiiTE40.8YSIC.#0; /Vc*.wmpy,' .-- verna-._*!420,0*,e14.#=' Q=vera College, Ctly'UOiitii* OfNeri YOik , The lack of *,quate literacy skills on the part of an,inamaiitigly140,401*.ttthe American adult population, currently.eatimated to be 27 (Daniels, 1988), is new considered to be one oftko most imPoit0t,jAss.ies-*** societk: Business and industry., are -bearing the lirrunt- ok,.nrie *44 IrOnediUgi,tinit practical effeets: low-level litetacy skills. among morkera. Yet, there has been relatiyely little research eopeerriklg,,t4e,,ii**41y4.13tefite worker population. This can be pattially-,aniihuted:'0.th .1*t that has been done on adult literacy in general -(barkeMiiald, '1986; :Pinkie, 17)In additicn, the specific phenomenon of low-level worker literw- tad its inn)* retry.* little understood: ; Adult litelacy researchers have recendy begun to invi agate the actUallieVela of literacy demand imposed by various work places and whedief or less difficult tasks than the tests they are given for jobs or tbe.:schOOt 4404 tbat, they hold would indicate they can handle. The results of.itstarch inggeStS'':thitevin though some workers hold jobs that demand a higher level appear to possess, these wa..kers are often able to read an4 tOdaritanci,i400?-0:'::./ materials their work requires (Mikulecky, 1981; Moe, Rush; bk.Storlie,/1979)'. The fact that these workers are able to perform job-related reeding tasksat better than their tested reading levels has not yet been adequately explained,by existing reScareb.Vitle' research has been done on the already existing strengths these adults 'bring to die reading process. Because strategy use is considered an important variable. in how well people read and do not read, and how they understand the reading process, a full explanation of the phenomenon must include an exploration of the readers' use of strategies. This paper will present a case of one such worker who participated in a study examining strategy use on job and nonjob related materials. METHOD Informant Cliff Parker (a pseudonym) is one of 4 participants in the larger study who work as car cleaners for the Tran.sit Authority of a major East Cont city. This report focuses on Cliff to provide indepth information about his reading behaviors. -"' On bisjob,, ctdreleans .and does minet:inainteqnceof subway:cals.ustiiiiyjn.,,,..-t-,14e'silbAyY*44,11.0*4,160.tecttcf itceP-cUrrent,by.readinialt materiali Ot!Oti*-, in his payroll e4vel4e, *It *1)04-:-.10,10trit _0*(44? completely *.refci't0.4.-WhO nerr-C414.Y.'W.ift3eci*14_51nArttl*.-f,14u0:-...;-kiiow and .1fii4e.tst01.4:P6r4iSV s104114Os-aa4 at. 61714 *h..0 *1344, k pail 4r*iter 4._Tc14 A:4*. C:hf.f*I0 attending sn *044 cia4 given tr hik-114ic$ at t.t*, qme of thç stuHe is a natiye at,'; eiith giacie'level of the Siosson Oralkeiding -Test (Slo,.7.sent.,490),andhad,beetker*)yett, by the transit Authority for l years at the time of the'study. Procedure To determine Cliff's use of strategies when reading, three measures we.,..,adraitristie tered: a retrospective task, an introspective taslk,, and an interviewferthe:fut Ora tasks Cliff was given-eight passages .:to rcad. aim job-relatect PasorigeS,;;tilte#2:intriet. frow his manual .and four nonjob rela ted passages taken, froth, s*.irclizeci',Fet,t4*,t, tan: at reading levels that approximated the job passages, 14.1th-.-42th 1o...0e:10_00d. above. Two passages from each type-were USed for theintrospectiVetask.Where Cliff stopped at the end of each sentence while be was reading to talk abonttiS reeding-, process, and two of each type were used for the retrospective task, Where Cliff:talked. at the end of each passage about what he was thinking or feeling while' readingilic passage. Cliff's responses to these eight passages were taped and transcribed. Tkie txnscriptions were divided into response urri (a unit contains one coMpleter thought er idea) by two raters test for reliability. There was 98.2% agreement. Backresponto was then analyzed for the display of strategies. Twenty-five perc P. of the-lotal response units were additionally coded by a second reading specialist. Reliability-was 90.6%. Discrepancies were discussed and resolved. Tentative categories of strategies were developed based on the literatureon reading and metacognitive strategies (Block, 1985; Kavale & Schreiner, 1979; Olshavsky 1976-77) and subdivided into general and specific strategies following the model.of Block's (1985) work. However, to get as coMplete a picture of Cliff's reading as possible, the coding system was set up to allow for the display of any other strategies that might develop from Cliff's responses. The third measure administered to Cliff was an interview questionnaire adapted from the Paris and Meyers (1981) and Gambrell and Heathington (1981) metacogtritive questionnaires and the Diehl-Mickulecky Job Survey (Diehl & Mickulecky, 1980). This measure gives Cliff's own assessment of himself as.a reader. All three measures were administered to-Cliff on a one-on-one basis in four separate sessions held directly after his upgrading classes at-the comiriunity college site where the classes were held. In additionorruchof the information thatforms, the following profile of Cliff came from ,informal,ciinversatiorts. either- beicire Or:after sessions or during.the thank-yen lunch 'thatiberesepiciier:inidlvith:dlitf:.:#roothis` data an indePth picture of Cliff SS an ipiliVittuji.,44 cs ;1-reacler was develOped: T 405' Reeding Strategies of Marginally Literate Workers RESULTS Cliff' as a Reader Cliff, an outgoing man, was receptive to participating in the. study in spite of some initial nervousness when he talked about his reading, as well lis.:OCCiiiionaL insecurities about "whether he was doing it right." His regionses Usually 'in91144 an example or an anecdote from his experiences and throughout the stlidklile . to enjoy talldng about himself in this ,way. Cliff conSiders himself only a fair reader, but he likes to read, mostly4Ooka,abOut chess. Cliff enjoys reading both for the pleasure it gives him in anclof itSeltalidal0 because it gives him the information he needs to pursue and hnprove at his hobbiei. The environment he reads in is very important to Cliff. Cliff goes Ate a rcióm by himself to read. "If I'm relaxed, maybe got, you know, a little somethingIo nibble on, and something to drink right there, I'm comfortable . . . I like to be by Myself, look you know." He cannot read on the subway, but will "look at a book . . . at it, I don't ready be, I don't read it, you know. It's just something where I won't have to be looking at people or something like that." Part of the reason why Cliff likes to read by himself is that, quite often, he reads out loud. Cliff says, "I may even read it to myself and then I like it ..-24:1 then I want to hear how it sounds. I don't know what that does." In an informal disc..ssion Cliff gave an example of how reading aloud helps him in reading. He talked about how much he enjoyed reading Shogun (Clavell, 1977), a lengthy adult novel about Japan. If it grabs my, and I don't stop, it 'list goes on . . . Shogun was like that . . . I didn't read it in a night , . . . but I finished it in a sitting . . . say two, or three nights. I just sat up, I stayed up all night reading it. I'd go to work and I'd come home and I'd get right into it. My wife . . . thought I was crazy. The revelation that Cliff had read Shogun was surprising. Cliff's reading assessznent indicated that this book would be much too difficult for him even with his high interest in the subject matter. However, in follow-up discussions Cliff talked about his strategy for handling hard material, he locks himself in his room and reads aloud. That, he feels, is the only way he can get through many of the reading materials he is confronted with. Cliff on the Job Not surprisingly, this enthusiasm for reading does not continue in the work environment, a place where it is impossible for him to read under the ideal conditions he describes. When Cliff is at work, he reads the newspaper and regularly checks the bulletin boards to see if any bulletins have been issued for his department. He says that this is because he works different jobs and they each have a different schedule. He does not read his rule book or other job-related materials unless it is absolutely essential to do so. "If a bulletin comes out I would reread it, but I'm not going to go get my rule book and just read it . . . . If they want you to hrow alything they'll put it in the booth." When asked how he reads the things he has to read on tiv. job, Cliff states, 4,94, Ihe'if sOii)ething,lreallyneed4O,10-011#0 dO*ni Yo4 kn'0*, and discuss it, maye with my *ife ;OF -*#4,0*, illustrates; Cliff relieS'on other People to help huwo '*e°4t5,APY ,dePliu.1-4F913,30b4 'When Cliff dOe's need Tto refer: to, a I1iiinUart,' hei; likely to ip is oi*P4,10401 manuslwi the elc00-s lstatiO, in: the POP.4ra "Ore.c.Pion0fr'fq* 0,*ple,'fP=-P.IMPIK?Ota:f40#- Hesyi don'OciiiiiN ho*to fili it gut; inaybe sbecculdhelpme kilt* is,iiinallY-theene,thare that:Lein deal so it'S' basicalli: the e the' inforiiiiiion in thebooth. ':MO-StlY the '1)604 you'ii it. Cliff's Use of Strategies Cliff talked a lin during the sessions. However, only a relatively,sthattimpOtlion,,.. ef stratOkkr4 weTe'OSPlaye4 in his responses. This wac heeenSe, so many of Clit6 comments 'were tangential to the teit. given .this style, it TS not stirprishig then that of thegeneftd.StrategieS;that,Cliff used on both job ,and nonjOb relatert.passages,,the most'fresnentiy;04,9***. "uses geneial knowledge and association" and "reacts to kit" (See,,Tahle. example, of L"reacts to the text," front, the Job 2 passage, is _Ciiff'S:coOnentliiho4 accidents: "Basically I waS just thinking, -I was ivoi,..:tering, it neyer,liaS, 4ipoe,40 me, and I 'know some peopb that it has happened ton." An eXstnOieggiff:Ose of general knowledge with regard to the Nonjob l:passage was hiScothmeliti:"IN*Seen articles on TV about- foxes, you know, they're hunting_ Wein in: calif** and:Out west and stuff, and I know they're cageylittle creatures." theernphisis.onth*twO strategies "uses general knowledge and' assoCiation" and "reaCtsjklext",seents tó be in keeping with CM'S focus on himself, his knowledge and his exiSerrences,,,,A1-, though these two strategies were the overwhelming method of choiCe4a14#,S4get, they occurred substantially more often on nonjob passages. This-pattern vat different . from that of the other 3_ study participants in the larger stuclY Who used,,motegeneral, strategies on their job-related protocols,- presumably because,theit,greaterlathifiirity with the content of job-related-materials enabled them to apply,a brOadei knowledge base to these passages. Cliff, of the 4 participants, seemed tichave,(or at least ,exptessed) the greatest nonjob related overall knowledge base, Which may be why his pattern of response to the nonjob related passago differed so greatly. Cliff used a variety of specific strategies on,the passages he read (10 in all) Such as "just reading," "paraphrase," "statingithe topic or gist," "summarize" and' "reread." "Just reading" was a strategy usedby Cliff alone. By"Just reading" Off meant "just [going) through the text." "Basically I know it, it's important, but sittinghere reading it I just really just went_throughlt." It is clear that Cliff has a lot of different:strategies at his disposal, yet when asked directly, in an interview question if he wat aware of any reading stptiegieS, Usedhe' stated, "I ,donl know how to answer that." it is also'intereiting: to:note that:40' of' the 12 strategies exhibited on thsintrospective jOb-telated passages were Useclon'One. passage which'6Iiff found diffienit. Thisluggests that Cliff's range of grated- nie incraties in response to' the particular challenges OUapasSage. -4.fdrik411), 4.iteteit.:ErWork,cri :fsf. . ," ^ , , A sis o f St.tategyUODkrinkthi-Retr.espeetivi iiiutIstitpspectfyi Xasia,f;r1 IY,'.,., ;7; ,4, - '''. ,a,..,'4,.',3 -',-.":,, r' ,;,;?-6, ,, _ yledgcvf, 6-, eacts,toeiV :dynka' P§trate.814. , 4 4 2 Reieadmg Wordolvrng 1 kaaiisci,e010ce, solving 7 'Se4iiLtv4i4 -hit 'reading: _ 3 3 anTs' mate content 1 7 eMINO, T'he,sttategy that Cliff employed also ,seerned ticAeteriainedl*:. rt. 104 confl*ted Aviih.t140414Ad two strategies on introspective tasks that, Mf. -did nt* appear lireVionsly: antiCipites contentr 'and loOka like SoMe,'FtiOte As in exaMplo of anticipates content Ja Cliff's ,COmMent TA MO tnid.61off. Cliff's'ase Of ti)e.`-`iiitegratea-infeethatiOni'' etrinteik by, q, eOpithent' on the inine.pnsOige, Where be Conneetett infornia::tion.froiri:41#04ier paOgraP4 4y, 44Ying, ik4rigi.40;:i40 verified thati"' Clearly, -t4 the introspective' task 'helped _cliffter.foCuS hIs reiPtinses 'in-' way (hat-scsieMtai, wider .range of suth strategies Having to.StoP at-the, end' of organize Cliff and Oroviiied helped fran*WOric for his idens: CONCLItION .4 , Cliff was Able to read at a higher' thap.piiected,.leyel And 4crOse effectiVely wide range of reading strategies on liotki"44. material that',Wis ch'ffienit tor hint 4.s. Was* case with ti*Olier-3:0fficiOant'S'in*: 4:094y1;-Pravio4.s*Licliq i44410,04' & Y44.001044, 40,0i command stri*ITPT.104.4'' oico4ib= cfiff -*wed hirk(40lf a4 a fak re*r At °F'-,FE-LPP!9RFf(''f_?r,44.14gisOPd*4,0.4 4fid was 4.41.d'of. WhOic140,`,(e.t net able to 'insViorsitiestiiiiii,..abeittL-W ,s01414:he'ilindytliek, 2.0.2,44; , 408 Literacy Theory and Resear Ch Although Cliff displayed a wide variety of strategies he tended to rely on two, favorites during .7.1e administered tasks ("uses general knowledge and association"' and "reacts to the text"). Johnston (1985) found a siniilar pattern in his study of adult low level readers and suggests that it may be because reading is most comfortable when it can be related to what is known. Reading is intertwined with-Cliff's life because of its usefulness in,the develotiment of hiS hobbies and interests. Cliff manages his readhigin context making it a very personal endeavor and controlling, his environmetitso -that he caryeMplOY, what he sees as his most effective strateth:- reading out lea, Partly lieetiUSe:',Of his reliance on this strategy he finds it difficult to funttion as a reader in a public setting,. such as the job site, and compensates for this by pretending that he can.re4 better than he can or avoiding print and relying on social networks to get the inforination he needs to function. Cliff, then, uses a combination of effective strategies, prior knowledge and background information and nontext-based strategies to navigate his literacy world that he has entered both by choice and need. Clearly there are many strengths that Cliff and others like him bring to the reading program and to the literacy environment at the workplace. REFERENCES Block E L (1985) The comprehension strategies of non-profioent native and non-native readers of English A descriptive study of process in progress. Dissertation Abstracts International, 46. 3663A-3664A. (University Microfilms No. DA86-03,872) Clavell. J. (1977). Shogun. New York: Athenum. Daniels, L A (1988, September 7) Illiteracy seen as threat to U.S. economic edge. The New York Times. p. 88. Darkenwald. C (1986) Adult literacy education A review of the research and pnonties fur future inquiry. New York: Literacy Assistance Center. Diehl, W A . & Mikulecky. . L (1980) The Diehl-Mikulecky Job Literacy Survey, In Job literacy (pp. 65-78). Bloomington, IN: Indiana University, Reading Research Center. Gambrel!, L B , & Heathington. B S (1981). Adult disabled readers' metacognmve awareness about reading tasks and strategies. Journal of Reading Behavior, 13, 215-222. Johnston, P H (1985) Understanding reading disabilities. A case study approach. Harvard Educational Review, 55, 153-177. Kavale, K , & Schreinder, R 0973) The reading process of above average and average readers. Reading Research Quarterly, 15, 102-128. Mikulecky, L (1981) Job literacy The relationship between school p. eparanon and workplace actuality. Bloomington IN: Indiana Universi7, Reading Research Center. Moe, A Rush. R T , & Storlie, R. (I 19) The literacy requirements of the licensed practical nurse on the job and in a vocational traim e prvgram. Unpublished manuscript, Purdue University, West Lafayette. IN. Olshavsky, , I . (1976- 1977) Reading as problem solving. An investigation of strategies. Reading Research Quarterly, 12, 654-674. Paris. S G & Meyers, M (1981) Comprehttsion monitoring, memory and study Arategtes of good and poor readen. Journal of Reading Behavior. 13, 5-22. Parker, 1 M (1987) Adult literacy What do we know and how do we know it? Unpublished manuscript. New York University, New York. Slosson, R L (1963) Slosson Oral Reading Test East Aurora New York. Slosson Edutation Pubhcations. 4C4 - i '. ,:''':-1, :^ ,, 1V.''' ,....., , _ ..- ;,.. 'THEW 'w/Wir .:2-41EA0iisitiitiFtmB:iittia,ititu?4'i AW4001EsFQF TOPti$1*V ,, ... -4-,... _,.., ,k...._-, ..,, __ -__44,-. ,c,..,,--:., :§400444tigio: ,.,..:. Pi!-v.4.64.#10-4.66-14s, -..-::. x,... ,-i, ,,,,,,. -SO.trOfejpi)e: .,...,i .t.. I.Inivirsity of corkla -,:::).. -_,: ,,..: ..., An. iMptirtiint 'tp disfinguisiiin suggeted in, a text from the less .iiiiportini.ontia. ,B..esettichin fist 1 ,-, that Suclya skill niay'bedeVeloOrtientaf in ;nature. POrexamP,I*oiti 4 that:younger readers do not iiiscerif 4einnfinate icians'fixini xtgio:fci*e4e#,:.(e.*.?, tiemey, Bridge, 4 'dein, 1078-79),.* to lesS able readers'i0fOrth ai..*O'LlirOfilhia:, tisk- aS their More Ole-counterparts (e.g.,. *Gee; -1911) , '0th:evil AeisearChfhita,..._,, .propoaed a 'poSsible relationship- betWeen the reitsier'S atitiriness-Of the- struCtiirj:fir a:, ,..., ---,:. text and the :.moognition of iniportant ideas (e.g.;.ticGeo, 140; *yet, '104; Me*? , _Brandt, '&11tith, 1980). iteading-in nfOreign language is cognitlY03, more deMeorafg,Man .1r,..,4108 in-.-ilr native langUage. Previous resenicn has ShoWn that already existing reading Ociiiiin, Lt can help the reader with a second langUage (L2) reading tasic.:fOr instanoeig.0, (1989) deMonstrated that prOficient 1.2 readers,Who AvereaWare-Of teXt.StMeitire.!4, E.:nglish, were able to transfee:their awareriest to a foreign hingnage.ieadiritikanul: , that thiS awareness aided in both the recollection and,retenthiri OfAens.,,iiieSented: in the text, HoWever, few investigations of L2 reading hive explored the ty0 OthifOrittation that the 12 reader_reMembers. The second,langnago researchers who have investigated what LZreaders.-raine* ber have obtained few, generalizable-results. Tor, example;- Connori(ic,44).:6;t:14P9 difference in the dumber of superordinate and suborOinate ideas:that the-14licade0, in her study remembered. However, becanse this and oilier Studies Oa similar, nahue' have used students , studying English, as a secondlanguage"(SL) aslublccti, tie, researchers have not been able to explaiutheir findings in termsOf the ri;ader's-alleady,existing LI Ithowiedge due to the varied largnage baCkgrOundi of the SUhjects..MOreover, these studies have required subjects* demOnstrate cciMprehensionbylsYrithig a free recallintnglish-(L2),, a procedure whiCh likely inasks What the reader actnally understne4 (Bernhar44;:1987.; Lee, -190), Thia.paper;,;then4resetits the secondiportion of ,a,tWopart study (Hagui,,19,89) thafinveitigatediherelationship'betweeri aWareness of textstrUCturein'a na0VC.-10gnage(English):And ;reading -hi it, SecOTICklititg4age,:(Spaniih); As-in the*St. Parti thia pinion Cominties'th 16olcat- the transfer Of aWarenesS-Stext Stinenue.frbni-Ltio-_,, . --::: , . ...-.4 .:,. .4.1 ft i . ... 014 its effects on L2 trndint,0,041PrOtOinti. ***14s ,ft.1=4,PPOotf 'of7.016 st44 ciatnittO this: iSs.ne in teOlg of 06:tqtal niOher: of wiwow,,.. ret#i#4 this Onninnnt,the ship'hetween'awarenesk of text *Om; !tag the t!'10 (InestiO#Aili*Ohe OS** (41#-Ai.n4*.nnis.- of fOtittliCtInn,ittV ated, With Ahe recalLof 1,46###t ideas n a :c#4:n441;4.)iPnct #ruCtPFP*10Y', readers recall a grOet FpOrtiO Of 4400aq 41.00. thin Ininnt le4def0 1*-§64y differs fp*, prOlo4s Fc$4011.: in *0 ilin***40044,..giltsttstibjects,Were able:te denaonitiate-an eiiistiniOarenes*Of tektfitruettire;WIA secon4 the free:recall task 4e.d to-itis6ss. coinprehefition; in 12 Wg.i'denedkpithe-native language of the subjects. Subjects The subjects in this study were high school students enrolled-in their,feurth.yei of Spanish as a foreign language. From an original pool of 92 SubjeCtS:Attendingj area high schools, 62 met the criteria necessary to remain in the itudY; f(a);ademotistrated awateness of text sttuctdre in English, _and (b) being a native speaker.:bt English. The 62 subjects comprised 34 seniors, at juniors, and 8 sophOinare4 23, were male and 39 were female. (See Hague, 1989 for a mote detailed descriptien,of the prior awareness test used in this study.) In addition to the measure used to determine prior awareness of text struetura-in, English, data were collected on each subject's reading achieyernent in both languages. In English, the reading comprehension percentile score of the StanfonlIteldeythiefit: Test (SAT) was obtained for each student; the ave,age score was 86.8%. In spanisk, the Spanish version of the Cooperative Foreign Language Tests (MLA), (Modem Language Association, 1963) was administered; the average score was 31.88: out of a possible 50 points. Procedures Four passages on dehydration (Meyer & Freedle, 1984) were translated into Span.! ish for use in the study. The four passages contained essentially the same information, differing only in text structure (i.e., comparison, problem/solution, causation, collection). Subjects were randomly assigned to one of the four passages. During thedata, collection, subjects read the aasigned passage and wrote a free =all. One weeklater; subjects wrote a delayed recall without rereading the passage. Because L2 productive skills lag behind receptive skills, the free recalls welt written in English tci maximirt the subjects' ability to convey everything that was understood during4eading.: The recalt protocols were scored by-the researcher:and'hO eolleigues,trained in the scoring procedure. Each immediate and delayed protecol*as-given four scorea; The first score was a text structure awareness scett, a score. that. Was ifitended'to reflect the degree to which the subject used the same text structure in .writing tIte free( =all as the author used in writi4 the passage. Meyer (1975, 1977) has hypothesize& 411 12 Text Stmcture that this procedure opens a window to the thinking processes used during reading and is a valid indicator of the reader's awareness of a text's structure. An a-point scale was developtzi for each passage, with a score of 8 reflecting total use of the text stnicture in question and a score of 1 reflecting no use of the appropriate text structure (Meyer, Brandt, & Bluth, 1980; Riehgels, McGee, Lomax, & Sbeard,,1987). The second score was the total number of i'lea units contained in the- tecal1 protocol. Four templates w-ze developed using Meyer's ofiginil content -struCtures (B. Meyer, personal comme_acadon, October 31, 1987), and they were used to quantify the number of idea units that each subject wrote in the protocol. EackPassage contained 38 i&.at units. (See Hague, 1989 for the analyses of total idea units.)' The third and fourth scores were the pucentage of superordinate and subordinate idea units contained in the free recall protocol. The four templates used to determine the number of major and minor idea units were also developed from the original content structures provided by Meyer. The superordinate idea units were those from levels one and two of the content structure, whereas minor ideas were those from levels three and below. The comparison, problem/solution, and causation passages each had the same number of superordinate and subordinate idea units. However, since the collection passage differed, raw scores were converted to percentages and used as the unit of analysis for each passage. RESULTS AND DISCUSSION The data were analyzed using the General Linear Model. English reading ability (SAT) and Spanish reading ability (MLA) were used as covariates. The text structure awareness score was treated both as a continuous and categorical independent variable in different analyses. Time and the structure of the passage were categorical indepen- dent variables. Dependent variables included the percentage of superordinate idea units, the percentage of subordinate idea units, and the difference between the percentage of superordinate and subordinate idea units. Is Awareness of Text Structure in 12 Associated with Recall of :mportant Ideas? The free recall protocols were used to determine the percentage of superordinece idea units that each subject recalled at the time of immediate recall. The protocols were alb.) used to measure each subject's awareness of text structure in U. The first analyses used English reading ability, Spanish reading ability, text structure awareness, and passage type as design factors with the pere ntage of superordinate idea units recalled as the outcome of interest. Table 1 presents summary statistics ibr the average percentages of superordinate idea units that were recalled from each passage at the time of immediate and delayed testing. At the time of immediate recall, there were no significant interactions. There were, however, statistically significant main effects for awareness of text structure, F(1, 55)=33.61, p = <.001 and for the type of passage, F(3, 55)=9.03, p= <.001. Additional analyses revealed that each increase in degree of text structure awareness was associated with an increase of 4.77% in superordinate idea units. Examining , .-1 Sienngriy-$tatistkS far the Average Pereentage a f SuperordiOietdeal'tnitiftehi, ltritnedlate and' Delayed Meng Paksage C-60-00 -PrObkni/Soluti6n. £114ktion Colleaioh- bri*diat' -4349 *0.76 32.15 _15:413 , further the passagemain effect, the Fisher ProCedure indiCiited,stitiaticallySig*atif differences between,the col ...arison and prOblein/solUtion.pastsaies comparison and causation passages (p='<.091), the CollectibninttproblA/sobition pa.ssages (p=<.001), and between the colleetion arid clutatiOn pisOge*(pF-ccipi)., Therefore, thcre was a COnsistentipositilie-relationsbir-bet;?ae44**K-11;*k structUre and recall for all four OSSiges, .1311t..the, average peateritage of, stiparOrdinite idea units recalled was greater for sonic types of passages thin fointhers. Regardleks of the degree of awareness, student:recalled more superordinate ideaS..frointhe cornparison'and collection passages. _ At the time of delayed recall, there was a statistically significant interaetion between awareness of text structure and the passage that was read, F(3, 52) = p...7 <01. This relationship was further explOred by building seParate regtestion Models for each passage, revealing a potitive relationship between text sipcittie *Wariness and the proportion of superordinate idea units nailed for erici passage, tad! *rime= in degree of text structure awareness was atsociated with, an inciease- Ot5.13%(p = <.02) of superordinate ideas on the comparison passage, .81% (0 = ...15,58)`an the problem/solution passage, 5.3% (p= <JO-, on the causation ;pass*, -and 11.13% (p = <.001) on the collection passage. Fisher's test of Least' Significant bitference (LSD) indicated a statistically significant diffeienee betweenihecnnipirison and problem/solution passages (p= .048) and between the collection andpmblem/ solution passages (p = <MI). The results of these analyses, using text structure awareness as a continbus variable, suggest that awareness of text structure in L2 is positively related to therecall of important ideas in a foreign language text. Regardless of the passage that was reads, an increase in awareness of the structure of that passage was paralleled by an increase in the percentage of superordinate idea units contained in the recall protocol. Atilie time of immediate recalk students remembered_motemperordinateidels.fromibe_ comparison and collection passages. On the delayed recall;,_- the relationshipbOieen text structure awareness and recall -was stronger for the-comparison and collection passages than for the problem/solution and causaticin pastages. Meyer and Freedle (1984) have argued that texts -vary in their degree of orgaiiiiation from loosely structured text such as a collection of descriptions to mnre highly structured texts such as comparison or problem/solution. Accordingly, thestnictufpaware readerusiz the organizational pattern as a memory-aid and hat bettarretall 'of Strikhke _ 413: the important ideas in a more structured tut because they are highly interrelated. The present studY only partially supports this notiort with respect to readera of-a foreign language: iotigh A v a r e n C s s of t e 4 s t r k a c . t u r e anci rt?zalk' 9fin*id* wposifivey related, students Who read the comparison and collection PassigeS rernernbeied Mere superordinate ideav than those:Who read fix`prebleM/solution iz.dall of Main ideaa-on the Collection-, passage SeettiktO de Meyer (1)84); however,- experienced Similar results and eonj ,ec(uredT high number of-Sue'enzirdinate idea Units recalled.Waa4ue4o pore unita being in this passage,than in the other passages: An alternitike exilifiatiOn may be that good reader's have had'nfore exposure to this typerif Passage thin tO the, tither types, and they have developed some other strategy to CoMpepsate, for- thelick -of organization. This might be particularly true for foreign language readersaS uuch of their initial reading experience is with this type of text. Do Structurepware 12 Readers Recall a Greater Proportion of Importatu rdeas2pan Minor Ideas?' The second set of analyses examined the proportion of superordinate idea units in relation to the proportion of subordinate idea units that were recalled by the foreign language readers. In these analyses, the construct of text structure awareness MO treated as a categorical independent variable rather than.as a continuous measure: Because we were interested in differences among the consistent users, the inconsistent users, and the nonusers of text structure, we created three levels of text structure awareness. Subjects scoring a 4 or better on both the immediate and delayed text stricture awareness scales were categorized as consistent Lsers of text structure. Subjects scoring 4 or better on the immediate measure and below 4 on the delayed were considered inconsistent users of text structure. Those subjects who scored below 4 on both the immediate and delayed were labeled nonusers. Table 2 presents the average percentage of superordinate and subordinate idea units recalled by eaeh level of text structure awareness on each passage. A mixed model using Spanish re, 'Ing ability, English reading ability, level of text structure awareues, type of passage, and time as design factors and the diffetence in superordinate and submlinate idea units a .e dependent variable -evtaled that the three-wa:, interaction between time of recall, level of text structure awateness, and passage was not significant, F(6, 45)=1.23, p= <.308. Therefore, it was concluded that any relatior.ship between the level of text st um awareness and the passage that was read in terms of the difference in superordinate and subordinate idea units was the same at the immediate and delayed times of recall. The data were then reexamined collapsing across time. After eirninating time from the design, there was a statistically significant twoway interaction between the level of text stucture awareness and the passa read, F(6, 45) = 2.63, p= <.029. In other words, the difference between the propoidon of superordinate idea units recalled and the proportion of subordinate idea units recalled varied according to the combination of passage and level of text structure awareness in Spanish. The following discussion SUMIIICAMS these relatiaaships (see Figure 1). 0 14,* TrketOighktititr- -,,tiii: 26..,45 '. -'.I.042'.. -4,-.66,- ...,U±4,3., *17 q4.001' 41,:eirel'4 _ 34:1-1 34.50 400 :4Ye.t3 Col*tion -'*-0:ir-t .59.87 48.09 21.06 LeYeP WO- 400 43;25- 49.20' -1:$:.$,;_ twut 6640 24.5 474 5415 31,4937.06, . Note., Level 1=Consittent users,;leyelZaIncoasistait iisas; Lcrel i=;Iprcas:as., On, the comparison passage, the consistent ,users reca11048:59A, mbre.:sti nate idea's Om- Subordinate ideas; thd inconaiStent nada recakled-',*irap*: note thr -ooh9r4inate; *4- ifie '4060§ers', .r6i4i164.001): 41.**H. _40*** than 'suoOrcunate: ideas. Fi.ter''S,1...sp ppoot*:ipino4toza:mtOco/F.0" differePee`hetWOO: 020'0)90-4:Lot' 9Ocit aild the loc*****19Oet*(ikri: otherpsirwise'cOntraSts4ere.signifieant.. On the tOilection passage, Il*.cOnsistent'usess reealle4 2.5.*Inore#Perorri*laa..:;.,- Wets than s0001144i4eisi'-the ***gni tioerO *04 ii.11:k..0.9r0--,..,' , labeidiilate,iiteiiiiii:tts,i-_ _.' o1., recollect 1 -9.6%'06re .0409141atO dor& itldigOte4 aitatiOtjc011ir 00440( difference .-L,***. the 4.0iigat*:11,!*.= the nontisers (p = < '.IIII3). No oth0,pitOrile . contrasts to*ogsgoot: On-tho 'COlisoll, _on POOOokO,' the- FO,9.,, -000'. *it** sti*Oisii6 4';; idea unitsiban suonimnate Sekpnitit awl Ime.uOuukeiji, recalled 84.* ma*, the: inconiistent Users .reetilledliAlik-,,niore,:anbOrdiUile,:, , . . .. ,,ficliv-YSti 1!",to' OOper-T9,314e*: ideOs..-NPoc-of the Ott.it**c9Otr.Oot* IYOO'10009.44 --Pet wcreany of "t* ol cllffer,O990 9O, the: tir9hlerOloOlotictkpOoliage'stittiO4e0r Oigoille,Ot. 311e r.0044'0, llIerie A. 4OlyOcO Ottgitlf ti,:inerOt 040.04To5qr,44cis US: her a iteAtOr Pt9Potti9.9 of Ole, hrOcrtoot:i4COO t 48,4 n),*;t:44#44 ill 'ttoel.eCtiOn- FOitheti490.,-the-*91tO Pteoe9t 0919 e*itkoCC that th*fen40.04.4* t9 the ii*r's***048: .0,f* structure ,141 Olq--40$4,100.4'dr. .0/IA0# Os* t.Px!stnia.tiq.-zigc414'4'.#1041011Y---04#:00000611.°0141.10-14,04,14P` readert Villa *ere eitherinOttonalitebt lu, -, - -.. -_...,i1*,:cfj,te#:Strnetuie:o*:* - .. -.Mama ' , ,, 79, ,211` -22. ;32: 20, 14. 14: 1i10, 6 6 4 ANIVAINEINFAIN - 2 0 -2 -4 Coasterilma Problem/Sol C202441414 C011ooti4a, ;;. Figure I. Average Difference in ProPortion of gUperortlinate and Units for Each Level of Text-8tructureAWareness. Idea --e..--- use text structure at all. These findings would seem tesupport Mer s (1979) fion that readers who are aware ot a text's OureProcessjextAifferentlY4lian ! readers_ who do riot have this awareness and,- conSeqUently, reinemberunMe,:ef:: main ,ideas contained in a passage than therelativelTmiaor,ideas. The erratic performance of the inconsistent tisera of text StOctiute.otythie_e1±*tierty passage,-however, pmvides the th9st FetIvincing 1040ce of ki).?Ss..Thl,c.g.alOW4,51. awareness of text structure and the ability- torticall'the.iMPott*itieas:.,e0ntaineci;1# ..t. . .. r:--- :_ , -;,-' -,Y 1,4 an 12 text.. Their frilktre to- activate, consYtenSgy 4.4. 4**,0* ,pf-11,:i -400*, structure wag paralloled by an increasedtOndeocy tO focus on ininor dCiailg:nrit*, onthetofthpsageOnthccomprlsonpasgóilthis;gpappeaed :`Of not to distinguish betweCnimportant ind Oh* details, iCcallingeri both brother words,, the ineonsistent users of text struetureprif (1977) Would predict a reader unawareof structure to perform, approaching* text in a random and unsystematic- manner. CONCLUSIONS Success in school depends on a student's ability:to make-the transition between learning:to read -and'reading to learn: Ag-10portant part;of reading tt0eartl..ia, the' al4Ity io. disiialaish the 100-P ' ia100#0,i4.40.1,40...a. t,0,#,f0iii:R.60*1.*..416 ios; -hrip9e4t1t. §40-!arly, as stuckrita Of a f'..004t1,4340:20:prigregg.irl tbeit itiiily'Oiihat ',I,-,k,. i. , --,,-_,;--: 416 Literacy Theory and Resetic language, their success is increasingly dependent on their ability to read to learn also. And, to succeed, they, too, must be able to differentiate the important ideas and minor details in a rending selection. The results of this study provide us with some clues as to the inforination.that second language readers remember. Awareness of text structure hill, is tcconstruct that certain readers are able to engage during an 12readiUg4ask, and, a this 014Y demonstrates, it is positively associated .with the recolieetiOn of the impOrtantlie4S., in a text Not only was there a positive relationship between awareness Of.text istOctUte, and the recall of superbrdinate ideas, but structure-aware readers also ternembered a greater proportion of main ideas than minor ideas. Several implications may be drawn from the results of this study. First, more research is needed to determine the LI reading skills that are potentially uiefil to the L2 reader. Specifically, we need to know what skills the 12 reader already =Sesses, and under what conditions such skills transfer to the L2 reading task. Second, b:ause this study has shown that awareness of text structure is related to the recall of important information, there is a need to explore the instructional ramifications of text structure awareness. Finally, there is an implied need to examine the types of structures that are inherent in texts written in different languages. Although students may be aware of the structures embedded in their nativ.. language, this awareness would be of no practical use if similar structures do not exist in the L2 text. However, identification of and instruction in the structures typical of the target language may assist L2 readers in their attempts to read to learn. REFERENCES Bernhardt, E (1987, April) Teaching and testing reading and listening. The immediate recall protocol. Paper presented at the Meeting of the Central States Conference on the Teaching of Foreign Languages, Columbus, OH. Connor, U (1984) Recall of text Differences between first and second language readers. TESOL Quarterly, 18, 239-257. Hague, S (1989) Awareness of text structure. The question of transfer from LI to U. In S. McCormick & J Zutell (Eds ), Cognitive and social perspectives for literacy research and instruction (pp. 55-64). Chicago: National Reading Conference. Lee, J (1986) On the use of the recall task to measure L2 reading comprehension. Studies in Second Language Acquisition, 8, 201-212. Madden, it., Gardner, E , Rudman, H., Karlson, B., & Merwin, J. (1983). Stanford Achievement Test (6th ed.). San Antonio, TX: The Psychological Corporation. McGee, L (1981) Good and poor readers' ability to distinguish among and recall Ideas on different levels of importance. In M L Kamil (Ed.), Direction., in reading. Research and instruction (pp. 162-168). Washington, DC: National Reading Conference. McGee, L (1982) Awareness of text structure. Effects on children's recall of expository text. Reading Research Quarter! 17, 581-590. Meyer, B (1975) Identification of the structure of prose and its implications for the study of reading and memory. Journal of Reading Behavior, 7, 7-47. Meyer, B (1977), The structure of prose. Effects on learning and memory and implications for educational practice In R. C Spiro & W. E. Montague (Eds.), Schooling and the acquisition of knowledge (pp. 179-200). Hillsdale, NJ: Erlbaum. 412 , , ,04t0910,'AFi$OstiSqgkO., tompREHEIsism, The, close, relationship between': &it language ,(14) and,: :SOL_ reOing suggests that bOth share' a nUmber of *00<inippoe* prrKessesci, and SOtne`,, ha K Skilla are transferred to',L2 reading (e:g Ctilinnini 1979 ;Keller; 1988;' Sittig,,1987;::Skiittiahh-Kingas-&-Tonkonnia,:197 . :StUdieS,inVestigating,differenCes:botWeenLt and,L2, that; inefflcient',Iower leVel:prociasing O0140191.1*.4*4._infl=nv *nth* *WO* leaflets. Eye.ñioyèmént research, kOi ,OjOinPle;' Gliet more ftequent fliations than Li readers :(e.g. readeititf:42W0iikik, sap, 1989). SiMilarly, Ober 'studiei::h0e less acenrate in word:recognition (e:g., Albert igc- Gbler; 497, SegsloWitz, 1916; 'Vases,190). ,beficient itienioir, capacity and' inhibits text-frinn being integrated*a:MeanhigfAaequenOe Daneman 4:` OnMenter, 1980;,1Kintoh, ,1974';''Perfo,ttk 1987; Moreover, if' readers are overly involved 'in lower-104 ProceSsittg 5,40100,141, attenti9naloapacity is available for higher-leVel cagnitiye sietiyity-SUCh textnal :information and drawing upen prior knowledge-4-:, and 09:40'0404-iinffeiii; The low-quality verbal procesiing skills of Li readera:apparentlfS, attribiitabla to limited linguistic knowledge :(e.g,clarke, 19N_;,16ik04,1978f 45kotA0 Eskey, 1988; Hudson,1982). little is known, hoWeve: aspects of linguistic kriewledge influence verbal procisSint andreading,.4:1)*-0#0sion skills. The purposes' of the present study:were (a) to identify:fag* contribütihg to L2 reading- ce_ipiehension, (b)to explere the, effeCts of differenikkapoiaAL2' linguistic knowledge on reading cornprebension,' and (c), to 641-00, factorS,differendating good from poor L2 readers. Tl*:,,study was , conducted: ivith,:heiOnnint readers of Japanese. In the section that fellows; the Japan* oithographic,syitems as , well as grammatical features of Japanese are briefly described. JAPANESE LANGUAGE Orthography Two distinctive orthographic systems are used in Japanese: Nana and Kanji...1.4iyi is a sound-based scriPt called a syllabary, ir which the syllable is the basiC '419' 'refieientation: Two kinds of syilabaries are used; fliragatia and.Kitalcana. Tr,16seddildmeans "Chinese tharicterk." Kanji,iS *Meaning 7based scriPt; call6criiThigOgraPhy, wherein one character rePreselits tl*,itieaning,, Of :.,410ole.,Wart1:Or.,niOrphet* The twe f.s'ysteOlirs 4,tbari6s )0i9g#Ptv,-;(140.9709.145,r0:47.* rcfitlY .11.604.1 jaPa?le.se,§, ,enteNes-1300* et fadfi Was 10.4i4,`,1,h1,101,0e$,..(40,9* kiragana uied':;prititarily:fot .*Ords: Such:: cate:Maticing4articlei,4 adjective,,andjaaarb'infleCtiOns,.,whileiKatakana uSed.eXClusiyely4or,::, rro words 'froiti Weitern:lingua*. Kanji is used, ptelusiveir fOr,conteat,*Ordi,i,Ji (1981) estimates that inian aVerage sentence-about 65%, is, ebinpriied;of 25-30% of. Kanji, and 4% of Katakana. , klorphosyntactic System Japanese is an agglutinative language, highly marked for caSe. Tgicaiyi is an SOV (subject-object-verb) language, exhibiting,rtiany:Ot*:Mo features associated with its classification (Clancy, 1985). For eXaMPle, ixelt case-marking particles (i.e., a grammatical marker indicating a eaSif role; ofiitioUnal are all.postpOgitional, and major recursive devim such as suboinate.clanses;atid, relative clauses are generated in the left to head nouns. For syntactic marking, Japanese makes use of both case-marking, pglieleK au, word order. Presumably, this dual syntactic marking System influences pMeedures used in both sentence production and comprehension. In fact, recent studies of Japa,, nese sentence processing provide evidence that children acquiring Japanese 4. isheir L I formulate an interpretive strategy that takes both word order and caselnarkink, particles into account (Hakuta, 1982; Hayashibe, 1975). It is of particular inrereStt, therefore, to determine how morphosyntactic features specific to Japanese iniluenee the reading comprehension strategies of L2 learners of Japanese. METHOD Subjects Subjects were college students enrolled in a first-year Japanese language program during the 1987-88 academic year. Out of 59 first-year students, native-speakers of Korean and Chinese (N =20) were eliminated because of their experience with Chinese characters in their LI. The remainder of 39 students were the subjects of the. present study (35 English, 2 Spanish, 1 Arabic, and 1 Portuguese speakers). The data were collected at the end of the first quarter after the students had studied Japanese for approximately 50 hours. The major objective of the Japanese program was to develop .communicative skills in both spoken and written Japanese. Classroom instruction focused on oral communicative activities. Although a brief explanation was given on ...itch grammatical point in the class, little time was spent on mechanical drills. Insteat, students were required to do grammar drills individually at a Language Laboratory. Since a main 415 - *opprOtrofiii '4 .text.wai Written itt,the authentic Japanese script, Hirangana syllabary wasAntroduced; at the, Very ; beginning Of the- quarter. As, soon Jts -students ,hird, niasteted'1,11#00;41; .KanyChsracters Wert intioduced`a few at a time. A total Of 49.,-cliaticjet:S:fititl.000.74=. -mately" 80 kinji words composed with the 40 basic'charneteraWare taught dniing.thej first eitrarter. Test Batteries Language proficiency measures. The language proficieney ,test sections: grammar and vocabulary. Tho grammar, compOnentindirded:tWii-0,14§,:en: tions, designed to measure knoWledge Of word-forMation rtr*rand des. Vocabulary knowledge-waTineafured throtigh throi-Subtesia: (a) tranSiatiOn;:(b) word grouping, and (c) sentence compietion. Multiple-choice items were icsioi the word grouping and sentence completion tests. Reading comprehension measures. Two reading measures were utilizato obtain information regarding comprehension at different levels: a cloie, tet fer:locid-leyel (or intra-sentential) understanding and a paragraph comprehension test for mei:6'1;1,0AI text-level comprehension. For the -doze test, a short -paraglisyntietieal1y;and lexically controlledwas constructed. A sixth-word-deletion fermat' was _used. The test contained apnroximately 120 words totaling 20 deletions. In scoring,:an-"actept able" scoring criteria was utilized: Any word considered semantically and syntaotidaily appropriate was counted as correct. This form of scoring is generally correlated more highly with other 12 reading comprehension measures than an "exact" method (e.g., Alderson, 1953; Shohamy, 1983). The paragraph comprehension test consisted of four paragraphs. These were syntactically and lexically controlled, and of variable length (80-150 words): k comprehension test of five short-answer questions was devised for each paragraph. These questions assessed the accuracy of the reader's conceptual synthesis of the content, as well as command of factual knowledge presented in the test materials. The subjeets responded in English. To control for background knowledge, test content in both doze and paragraph comprehension items was based on familiar topics so that the subjects could handle them with general knowledge. Word recognition speed test. Word recognition speed was tested in three discrete conditions: Kanji (Chinese characters), unfamiliar syllabary, and familiar syllabary. In the Kanji condition, 30 frequent Kanji words were given and the subjects were required to write the meaning of each word in English (i.e., word translation) as quickly as possible. In the unfamiliar syllabary condition, 30 words taken from the text, normally written in Kanji, were presented in a syllabary. Because the subjects were not visually familiar with the stimuli words (although ley were phonetically and semantically), they were forced to go through a symbol-by-symbol processing to obtain the meaning. In the familiar syllabary condition, another 30 words from the textbook were presented in customary syllabary (Hiragana) form. A 3-minute response time was used in each of the three tests, given 2 days apart. Symbol identification speed task. Thirty nonsense syllabary symbol strings, each consisting of two to four mbols, were constructed for this task. The symbol strings CYate4j0.104 scr,etii, 06.44, '15i:e00:14.0:d!Fat4)04.k seC.00); **ctkiPlin*Flr,4 t*:.0 PiPt.Ok 0.6 ei 'ktci.b.4a4i*.10414jancs,e'(R9thin-All*OPtic't4iiiit) 444610**8.0:'-i:ES4AdMinistration cf the tests were given during 'clitsi,periods of.- the:40**a: activities. RESULTS AND DiSdUSSiON Factors Influencing the DeveloPment of 1,4;;ReddingComi#teisce Table'l-presents,the,condatiotia. :arnanglinguiatio1MOwleflge,(i,ew9147.40.4t tion rules, -Cate-marking ,partielesi wdrdlecagnition*It#.#01:ide009404=a4i44ing'-c0741****** anti P4iiiiiiph c011x..eliOsiOk. !!.0411S4480 partielei 'and Vocabulary) were-highly cerrelated:,..with tke ;WO radink COMPrehensiOk measures. The correlation was plitiedru2ly high betweett'particW1,00Wledge'and;d0Zet.. and -betweeo vocabulary ,knowle4ge arld'imMigraPh cOMPiehensiOn.,.; knowledge was correlated with doze slightly Mote highlY.than With paragraph,e0iprcy hension. Overall, however, the cOrielation, betweeii Word-forniatimi *IW reading comprehension was not strong relative to the other- two isPOts:ot knowledge (vocabulary,and case-particles). 'High correlations were also found-between the two reading compreliensio&m* sures and word recognition. Symbol identification was highly eorrelate0 with the dózetest, but not with-paragraph cbinprehension. tinally, a relativelYhigh.correiatiOn found between the two reading test scores. To isolate the factors contributing to L2 reading comprehension, stepWiSe regression analyses were conducted for each of the two reading comprehension,theai Table I Correlations among Linguistic Knowledge, Verbal Processing Skills, and Reading Comprehension 1. Word-Formation Knowledge 2. Particle Knowledge 3. V:zzlulirry Knowledge 4. Word Recopition 5. Symbol Ltentification 6. Cloze Test 7. Paragraph Comprehension .47 .59 .77 .56 .48 .67 .73 .70 .67 .65 .55 .81 .54: .39' .56 12 CoMpreltension - -- sums (see Table 2). The following five factors were entered into the regression model as predicting variables: word recognition speed, symbol identi&cation speed, particle knowledge, knowledge of word-formation rules, and voestiplark knoWledge..Iwthe doze test, 'particle knowledge is of the greatest sipificaece,_seconnting,for4a1MOSt -60% of the-variance. This result can be exilained, ii *4 by the!, fact,',th4t. case- particles comprise 30% of the deletion items. It Li iiiiPelta4 t0:0,0e;:hOW0ett that case-particles provide essential information regarding the seinantk ie4tionahips . among content words. Presumably, this information is,vital for-,Mearting Oon'Oriretion particularly when the context does not offer adequate chiesai iS"the.cale"0 the el.* test. Vocabulary knowledge was also found statistieally, significant in this inalYsit, the remaining variance after removing the portion aecOunted for explaining 15% by particle knowledge. In contrast, in paragraph comprehension vocabulary knowledge was found to be the most statistically significant factor, accounting for 64% of the variance, followed by particle knowledge and symbol identification, being responsible for 12% and 3% of the reinaht.,ag variance, respectively. The regression analyses thus revealed that vocabulary knowledge was a significant factor in both of the two reading measures. This result is consistent with findings from previous Ll and L2 reading research, and suggests that knowledge of content-Word meanings significantly enhances reading comprehension. Particle knowledge was also found to be statistically significant in both reading tests, indicating that in addition to individual word meanings, information regarding the semantic relations among the words is essential to sentence comprehension. Interestingly, vocabulary knowledge was foutid to be the most significant factor in the paugraph comprehension test, whereas particle knowledge was of the greatest significance in die clore test. Different factors thus accounted for the variance in each test to varywk, degrees, suggesting that knowle4,e sources activated during the task performancc are constrained by the nature of the task requirement in different reading measures. Despite the fact that high correlations were found between word recognition and two reading comprehension measures (r= .67 and r 54), word recognition was not Table 2 Summary of Stepwise Regression Analysis Steps and Variables Paragraph Comprehension 1. Vocabulary Knowledge 2. Particle Knowledge 3. Symbol Identification .81 .64 42.29 .0001 .55 .39 .12 11.79 3.45 .0022 .0463 .73 .70 .59 .15 35.52 .0001 .0012 .03 Core Test I. Particle Knowledge 2. Vocabulary Knowledge 13.44 "*. riiè , Tat Pezjbrmnmce 4t G661 4?"-117,7-_,_ . ,_ , _ 10-000Y, ,, 134.05;-',/,044:ifti7-,(1:0,641tiont.40.9). VouMiy -=' (W.inl 15 ,, 25,7', :0:6 0 4 133. ,160 , -,,.7.ki.fij- ,87, P9-''C'.. .. . , ..-r-, , ..-----,-,,:: ..,...-.-).,, -''-.,?' ,-. 4631:-.-. -,v-..-.0*-- , ' ' -'14,41'.'-..- !t,16 :_8 -,,:-.:-.:::=,,::-.4 47.0 ,i.ip!' 1,14,:?.--s,>;.',, -,.%.v.-: 4:;* t 3.-,f ;Orr kiti -2:521 439 , -,0i4f *0-:._ ,. . .._, '3.'18; .irs *-4`"*C:i'at4-,',.--41' (Faniiba 4.6.0 YI.00:- Recognition ,.-:**4)' 444i :24 . Y.94.b,4.41i--1.', Woof ' . ..,...., -1.1,t-%-,;,---449 te,-c, Jig' -,17,8.. t4;_ ::',',".-;,--'.47,1`,- 'kite =' Maiinuith with the vsilsbtcnames , , . saxes ... ., Sze' _ ificticatiii ......--- -.In the Parentheses -.... ,,... .. , ,, -',,, ',-,,' '7' :: f9:04A4t4g.04lir 'iikoiOtt4ot for 004. '410110Y, 04704154#400,0,9-44i9i9.#7,. tw sccw - ,significant -for Me ,. eiriie test,: in ,SPite- of the: iiii0r, cool/4o ': between .(i', -,0,1). Ail, 0, 440' OPottOo-o; O__ _r i*:#4;00.0474iitit 441i .41# ,bitI4'11..ifro !*e_._, ,aotiPii*tcd ,....; ,-,...,,, , froi4 140--Otiela00* ThA. ..,si..,P4iiiPC*.ar..*-0-.06-#:, Of* Ittortig t4atiOnski0 .4**et:t *OA fe:POgn4i9n and k9i..;Y914.40#..-(r ar PartitIeknpwledge.(r-= .9)'as well as that between syMbOl- identification kid ,,,knowiedie (r= 36). : r , .7,- .1,,, , , . _ti .. ., . - . . ...- , Factors Differentiating Good4rom Poor42:::Readers Bek4 Cri the two reOink Omprektensip.olot Korea, a Sr9913 9f "g9O0.7=Fsti4r#), 7: and- a-- gronli -Of pooe' le-aderac-we01:8490,94, TWo highest z,lo and theIOWcat:lirtest-bcorei, Were-choir:0:ga** first 39 cibJects The means -and-the standard- deviations of thctwogroupsfoc.sàvàn ieitaeorei are ilist#1- ' in.Table3. t. To identify factort diffeTeatiating thp two groups, a stepwise:diseriMinantiniiiixiii?, was perionned. The' fetiewi4 6e-factor:aver:A entered**, iiii4itpiy.,904911-07-. tion knowledge, pirticle knoWlefige;;Vocabblari:IMOWledger wriri.re.cajoboo speed and .aYinbOl iidcntifioatiOn-liPetd-'It-Was5:fcrind most significant factor.distingoishingtood;troin,pOpeizrearkrst9kOWed bY pAilt,i0):14607: tificatiOn speed (See Tablo4). the data thus demonatrates that,,knoWledge-Of eitie,marking partzoles,1411e highly significant faa* not 9niy facilitating reading cOo4)040#10`!*t tliffe*OhAa4t, 064 frOm &OF Li roadOisof-104n#9.:T4o-a9Arnitaing:4* 42* .fea:toles shnaling the syMactie stilmtkireb4 the ti!rget lang*ige'(4Se-mar4413.40,00;, . in this caie) 04, a critical role in .i,c'ad49Oini*j*9kto.n, a494.414: L2 readers utili#,thit;knowledgO ifteetfVelY: in thelif.edintirebeo4,0,000.4, turn,. anggeaf that the granunatical f6.444*,9941:t9:a900* c9n1094,F,'44 'fa9gOaii4o9i&; Aa:0494.a9d-441er.k-14,009--044)1.10;,49t#149, ts S. , Table 4 Summary pf Stepwise Discriminant Analysis Step iariables 1. Particle Knowledge 2. Symbol Identification Increase in R2 .58 .19 system, utilizing both word order and case-ir Icing particles. Althotigh:thOnOst preferredor canonicalword order in Japanese is soy, in continon Oral dikourse word-order is quite flexible preceding the Main'verb (Clancy, 1985). Unlike the.Wordorder dominant languages such as English, the syntactic structure of.Japanese,timiiot alwayv be perceived in a linear fashion. Moreover, Japagescis mheavilY'Conteat, dependent language and permits eitaniiVa ellipsis=that is, lin-guiStiC;eledientS, that can be understood from the context are often omitted (Kune, '1978; Tsiitsui,, 19,84). In fact, in a pragmatically appropriate context, elliptical sentences _(e.g., sentences without a subject or a direct/indirect object) are not only- grammalically Correct, but also stylistically more preferable in many cases. Thus, the extensive ellipsis eitereised in Japanese discourse makes word-order even less reliable as a syntactie-Maildng device. Consequently, as is demonstrated in the 7resent study, knowledge of.casemarking particles and ability to utilize this knowledge are crucial for sentenee comprehension in Japanese. Hence, the present findings corroborate those from earlier studies, and suggest that ability to use grammatical knowleckt significantly influences comprehension, and therefore accounts, at least in part, for irmividual differences both in LI and 12 reading (Cowan, 1976; Cziko, 1980; Tyler & Nagy, 1985). Symbol identification speed was another significant factor separating good from poor 1.2 readers. As shown in Table 3, the difference in word recognition between the two groups was also greater than other factors. These results, coupled with high correlations between verbal processing speed and reading comprehension, seem to suggest that efficient verbal processing skills are also important for successful reading perforrnalA.e. Interestingly, the difference in vocabulary knowledge was relatively small despite the fact that this factor was found statistically significant in both of the two reading comprehension measures. This finding would seem to indicate that vocabulary knowledge alone might not be sufficient for fluent reading, and that L2 readers need to develop skills to retrieve this knowledge with an adequate speed. To summarize, the data from the present szyly demonstrate that (a) vocabulary knowledge significa lily contributes to 1,2 reading comprehension; (b) two types of grammatical knowledge (word-formation rules and case-marking particles) have differ- ential effects on text comprehension in Japanese; and (c) particle knowledge and symbol identification speed are two major factors differentiating good from poor 1.2 readers. These results, in turn, suggest that (a) knowledge of c6ntent word meanings play an important role in reading comprehension, (b) some syntactic features inherent in the target language constrain essential knowledge sources utilized in the L2 reading comprehension process, and (c) efficient verbal processing skills are related to successful reading performance. Because of the limited sampling, the study findings may not 426 UteracY '&0tY and Restart% be generalizable to other L2 learners at different proficiency levels. Nonetheless, The studY, represents an examination of relatively tiexplored-yet significant7;--dititensions of L2 reading research. Subsequent studies may verify the fi4ings of the *sent results. REFERENCES Albert, M. & Obler, L. (1978). The bilingual brain: Nescropsychological and neurolinguistialispects bilingualism. New York: Academic Press. Alderson, J. C. (1983). The clue prOcediw and preficiency in English as a foreign' language,:in W.'. 011er, Jr (Ed.), Awes ir language testing 'research (pp. 205-217). RoWleY,'IVIA:,NeVArriHouse: Clancy, P. M. (1985). The acquisition of Japanese-In D. I. Slobin (Ed.), language acquisition .(pp. 373-524). Ilillsdale, NJ: Eribaum. Clarke, M. (1980). The shott circuit hypothesis of ESL reading-or when language competence Interferes with reading performance. Modern Language Journal, 64, 203-209. Cowan, J. R. (1976). Reading, perceptual strategies and contrastive analysa. Language Learning, 16, 95-109. The crosilstudyef Cummins, J. (1979). Linguistic interdependence and the educational development of bilingtialchr Review of Educational Research, 49, 222-251. Cummins, J. (1984). Bilingualism and special education. Issues In assessment and pedagogy.tlevotIon: England: Multilingual Matters. Cziko, G (1978). Differences in first and second language reading: The use of syntaetic, semantic, and discourse constraints. Canadian Modern Language Review, 34.473-489. Cziko, 0. (1980) Language commence and reading strategies. A comparison of fira and second language oral reading enors. Language Learning. 30. 101-116. Daneinan, M., & Carpenter, P. A. (1980). Individual differences in the perception of color. Quarterly Journal of Experimental Psychology, 28, 387-394. Devine, J. (1987). General language competence and adult second language reading. In J. Devine, P Carrell, & D. E. Eskey (Eds.). Research in reading in English as a second language (pp. 73-85). Washington D.C.: TESOL. Eskey, D (1988). Holding in the bottom: An interactive approach to the language problems of second language readers. In P. Carrell, J. Devine. & D. Eskey (Els.), Interactive approach to second language reading (pp. 93-100). Cambridge. MA: Cambridge University Press. Haketa, K (1982) Interaction between particles and word order in the comprehension and production of simple sentences in Japanese children. Developmental Psychology, 18, 62-76. Hayashibe, H (1975) Word order and particles. A developmental study in Japanese. Descriptive and Applied Linguistics, 8, 1-11. Hudson, T (1982) The effects of induced schemata on the "short circuit" in L2 reading: Non-decoding factors in L2 reading performance. Language Learning, 32, 3-31. Kintch, W. (1974). The representation of meaning in memory. Hillsdale, NJ: Edbaum. Koda, K (1988) Cognitive process in second language reading. transfer of LI reading skills and strategies. Second Language Research, 4, 133-156. Koda. K (1989) Effects of LI orthographic representation on U phonological coding strategies. Journal of Psycholinguistic Research. 18, 201-222. Kuno. S. (1978). Danwa no bounpoo [Discourse Grammar]. Toky "aithukan Publishing. Macnamara, (1970). Comparative studies of reading and pro..., solving in two languages. TESOL Quarterly, 4, 107-116. 011er, 3 W , Jr (1972). Assessing comi=tence in ESL: Reading. TESOL Quarterly, 6, 313-325. Oiler, J W , Jr & Tullius, J. (1973). Reading skills of non-native speakers of English. 1RAL, XI(1), 69-80. Perfetti, C A (1987). Cognitive and linguistic components of readin ability. In 13. R. Foonnan & A. W. Siegel (Eds ), Acquisition of reading skills. Cultural constraints and cognitive universals (pp. 11-40). Hillsdale, NJ: Erlbaum. 4.1.1 12 Coniprihension. Perfetti, C. A. & Iona Id, A. M. (1979). Coding arsi comprehension in skilled reading and implications frit- reading instruction. In L. B. Resnick & P. Weaver (Eds.1, Theory and practice c early reading (1;13,00-84),;11iflidiik, NJ: &Mani. Rolla, C: (15188) Tianifir-of cognitive -academic competence and L2 reading in s rui*I'Zinibqtat*, primary schoOl. TESOL.dranherly, 22, 3037318. Saito, Y. (1989; Deceinber), The effecu qf ailicle deletion in Engli.sh on the cognkre jirocisies rejleded` cf English: In tite eye moYeine;Itsiind itiesacoritivedWarenfP cf nadve retiairs iik4alxlar4e 4n eitiracking ;1,4 y:,PaPir PielOted at ti* &rig, G.- (1987)-.'High.leiel reading in the first Isitgaiget,Some c;:intiaritive pin*: data: 110;,,Detiiiie,_ P. Can:e11; kIX.E;Eskey (Eds.), Research in readingin Nell:hal a sieo lartguaie(ip.'165-121:1). Washington TESOL. Skutnabb-Kingas, T., & Toukornaa, P. (1976). Teaching migrant child:erg mother toneue ód kerning, the language if the host courury in the context of mdck-ultural situation of thi Helsinki: The Finnish National Commission for UNESCO. Sakamoto, T., & Malan, K. (1973). Japan. In J. Downing (F4.), Comparative Reading (pp. 449".165). New Yodg Macmillan. Shohamy, E.,(1983). Interrater and intraramr reliability of the cral interview nod coccurrentvalkfity with doze procedure in Hebrew. In J. W. 011er, Jr. (Ed.), Issues In language testing, researth, (pp. _ 229-236). RowleY, MA:-Newbury House. Segalowitz, N. (1986). Skilled reading in the seccod language. In J. Vaid (Ed.), Language processing in bilinguals: Psychiolinguistk and muropsyrhological perspectives (pp. 3-20). Hillsdale, NJ: Edbaum. Stanovich, K. E. (1982). Word recognition skill and reading ability. In M. G. Singer (Ed.), eaMpetent reader, rksabled reader: Research and application (pp. 81-102). Hillsdale, NJ: Edbautn. Taylor, 1. (1981). Writing systems and reading. In G. E. Mackinnon & T. G. Waller (Eds.), Reading research: Advances In theory and practice (Vol. 2, pp. 1-51). New York: Academie Press. Tsutsui, M (1984). Particle ellipsis in Japanese. (Doctoral dissertation, University of Illinois at MinnaChampaign). Dissertation Abstraas International, 45, 11A. Tyler, A., & Nagy, W. E. (1985). The role of derhational sigNses in serusnce cornprehension (Tech. Rep. No. 357). Champaign: University of Illinois. Center for the Study of Reading. Vasos, H. (1983). Semantic processing of bilinguals. Unpublished doctoral dissertation, Cott.ordia University, Montreal. _ 4? 2 PROGRAM FOR 1989 NATIONAL READING CONFERENCE Austlit, Tens--No*embek 2843eizeinber2: ._ pug Minas's mad* and - 001.0 Literacy la ploy: The Impact et ilteney-entiched play amities hohavier 001.1 The bitireace of Ottracpenriched ility settings cm Preatisailetl',Fonesptiors *fp* .- Ssitan'R: -Newnan, University of 1.oevell, and Kerby Roakie, Tohn carat UrSiVersitY 001.2 Increiehog children's literacy behavior dm* playtime* liseitilan unheraey nasiirbia and adult asidtikeg. Carol Ilukelich, University of Delaware 001.3 Ler** Weeny behavior &dog play *mach physleti &sip amen. Lesley Mandel 002.0 002.1 002.2 Moe w, Rutgers University Observing literacy: What prownke and hmervice Seachaeleara from pm pad letters Pm pal Warm A km foe *Maw writing etrakgies. Jill Burke, Team A&M University, and Fetich Garcia Longnecker, Bryan 4depnodent School District From drat grade to langoage arts methods Amen Sethi role negotlatialia he Pes pal letters. Barbara S. Tyler, Northwestern Cficiehoora State linivallitY. and TranY Rulings, Medford Elementary School 002.3 Language growth la pen pa/ letters. (Artie:dm Impfications for ekmentary kapage arts and teacher edacatien. Los Ann Baron. Texas ARM University, and Dianna Micron, Sam Houston Elementary School 003.0 003] Readers, writer; and the totartau Intertextuallty and moves to authoeity in welling trom sources. Sams Greene, Carnegie Mellon University 003.2 003.3 004.0 004. i 004.2 014.3 005,0 005.1 What's new tn reading and writing: Prior knowledge as an interteztual construct. John M. Ackermsn, University of Utah Composing from sour= Task, text, nod Interim*. Nancy Nelson Spivey, Carnegie Mellon University Issues in text revision: A changed perspective Investigating the mearch literature base in text revision. Judy Nichols Mitchell, Univeraity of Arizona The effects of text change on readers. John M. Bradley, Univereity of Arizona Text factors he modified texts: The case tithe excerpted text. Jonme M. Golden, University of Delsware Classroom dialoeue and dircusdoe Reeder responae: A way to promote higher ceder thinking skills In the classroom. rtsibicia R. Kelly end Nancy I. Farms, United States International University 005.2 005.3 006.0 006.1 006.2 What do stadents In teactertaided diandons and peer dkenselems kern &heat the processes of small gimp discomke? John F. O'Flahavon, University of MarylandCollege Park Classrooms sad literacy: Critical thinking throe:1h reflective inquiry. loan A Mime. University of Pennsylvania Tbe transmission of literacy c,z ?me-Income homes Idmtifiable roles low-Income s....)th:rs play daring book sharing time. Patricia A. Edwards, Michigan State University, and Bonnie M. Kerr, University of Illinois Low4neome matters using cooperative small groups ts a model for training other low- 429 423 7:1 income mothers/fathers to share books with their young children. Janie Everett and Nancy Weems, Louisiana State University 006.3 Low-income mothers supporting their children's development in writing. Kathleen A. Copeland, University of Illinois, and Patricia A. Edwards, Michigan State University 006.4 A case study of one low4ncome mother learning to share books with her four-yearold daughter. Patricia A. Edwards, Michigan State University, and Georgia Earnest Garcia, University of Illinois 007.0 Students' comprehension processing: Farnminheg the impact of evolving lesson implementation during a teacher growth process 007.1 007.2 007.3 A history teacher's verbal behaviors and decision making: Impact on eightir grade students' processes of comprehension attainment. Nancy I). Padak, Kent State Univeisity. Student inferences and comprehension processing: An analysia of the development of comprehension processing in history lessons formed on discussion and problem solving. Bonnie C. Wilkerson, St. Charles School District, Illinois An investigation of a teacher's and student's verbal interactions in reading and discussing history texts during the teacher's change ht teaching style. Jane L. Davidson, Northern Illinois University 008.1 Changing language arts instructional practices: A case study of one school district. Kathleen A. Hinchman. Syracuse University, Alice Boijonis, SUNY College at Oswego, and Bren T Price, Cynthia Hawkin, Ellen Molinar, Richard Tabo and Jane Woodward, Marcellus Central School 008.2 008 3 008 4 009 1 009 2 009 3 Condderateness of postsecondary reading texts: A content analysis. Jeanne Shay Schuir.m and Georgeann Ross, University of Miami Discourse types in Canadian basal reading programs. Sharon Murphy, York University Captioned video technology and television-based reading instructhm. Patricia S. KosEnen and Linda B. Gambrell, University of Maryland The long and the short of first word recognition. Luci imncomb and Philip B.Gough, University of Texas at Austin The differential influence of knowledge of signals to importance. Danielle Michaud, McGill University A developmental study of the processing of orthographic information in children varying in reading ability. Evelyne Comm and Dale M. Willows, Ontario Institute for Studies in Education 009 4 010.0 010 1 Project InfoNet: Description of an integrated CD-ROM data base computer network and an analysis of reading disabled student search processes and comprehension of text. Dump. F. Shell and Christy A. Horn, University of Nebraska-Lincoln Becoming a teacher of literacy A study of teacher change during writing process instruction. Barbara Moss, The University of Akron, and Richard T. Vacca, Kent State University 010 2 010 3 011.0 011 1 011 2 011 i 012 0 Becoming a teacher of reading: An ecological inquiry. Susan B. Argyle, Sbppery Rock State University, and JoAnne L. Vacca, Kent State Uoiversity From the college classroom to the elemratary classroom: Becoming a teacher of titeracy. Nancy D. Padak and Olga G. Nelson, Kent State University Main idea Background knowledge and main idea comprehension: Knowledge activation and nonstrategic comprehension. Richard D. McCallum, University of California-Berkeley The main idea in reading: 1979-1989. Mary Frances Graham, University of North Carohna Effects of reading science text for main ideas. Linda L. Johnson, University of Iowa Patterns of success at literacy learning among low-SES urban children in the early gasdes 4.24 °NRC Prograin, Austin, Terns, 1989 431 012.1 Patterns of successfnl and less successful behaviors el low-SES children during School fitting tasks. Ellen McIntyre, University of Cincinnati 012.2 Extent and patterns of learning and growth in eonceptual understanding of written 012.3 hnguage by low-SFS children. Victoria Putne 1I-Gates, UniversitY of Cineirsti Learning to read and writein inner-city schools: An nnalysis of learners engaged in real 013.0 rendingind real Inking:Karin L Dahl, Universiti of Cincinnati Prior knowledge and learning from scialce text: The effects of text type and instruction on changjng naive conceptionn 013.1 Correcting misconception= Effect of type of text. Katherine Maria, College of New 013.2 Rochelle, and Joanne Mons Johnson, Scarsdale New York School District. The effects of refutation and considerate tmEts on learning conceptually easy and difficalt science concepts; Janice-Dole; University-of Utah- 013.3 014.0 014.1 Tim influmce of discussion, demonstration, and text on the learning of counterintuitive wknce concepts. Donna Alvermann and Cynthia R. Hynd, University of Georgia. Veinal recall and comprehension of stories Verbal participation and story comprehension during story book reading. Lesley Mandel Morrow and Jeffrey Smith, Rutgers University 014.2 Verbal recall: A procedure for messing inferendng strategies. Susan B.Neuman, University of Lowell 014.3 Verbal rehearsal and reading comprehension performance. Linda B.Gambrell, Deborah Mille, Susan Kin and Joan Thompson, University of Maryland 015.0 Word and language development 015.1 A lonenidinal study of the role of cognitive development in the evolution of young 016.0 children's Concept of Word. Beth R. Spencer, Agnes Scott College The language development of early readers. Maty E. Huba and Ramisetty-Mikier Suhasini, Iowa State University Toward an episodic model of word recognition. Jim Wagner and Alison Fisher, Brock University and Alisa Cantwell, McMaster University The effects of varietal comprehension interventions 016.1 The effects of integrated learning strategies on young childrens' comprehension, 015.2 015.3 vocabulary and summarization ability. Eileen Margaret Carr, University of Toledo, Mary 016.2 Bigler, Eastern Michigan University, and Cyndi Morningstar, Ply,-..out:z Canton Schools Effects of active comprehension on attitudes and motivation in reading. Ruth Helen Yopp, California State UniversityFullerton 016.3 Effect of story mapping instruction on first-grade children's understanding of trade 017 1 017.2 books. James F. Baumann, Purdue University Coming to terms with the terminology of knowledge. Patncia A. Alexander, Texas A&M University, Diane L. Schallert, University of Texas at Austin, and Victoria Chou Hare, University of Illinois at Chicago The incidental learning of word meanings by kindergarten and first grade children through repeated read aloud events. John J. Pikulski and Cy.ithia B. Leung, University of Delaware 017.3 017.4 018.1 018.2 Development of strategic rmtling in S basal reading programs. Jimmie L. Russell, Oklahoma Baptist University, and Maureen S. Siera, Northeastern State University Self-Correction strategies of disabled readers. Barbara M. Fleisher, Beaver College What determines course achievement? An investigation of several possNe tatLences on academic outcomes. Ann J. Pace, Karol Walter, and John K. Sherk, Jr., University of MissouriKansas City Examining dictionary definitions: Research in progress. Margaret u. McKeown, University of Pittsburgh 018.3 018.4 019.0 020.0 Reading difficulties, metacognition and affeet. Christina E. van Kraayenoord, John Elkin, and Adrian F. Ashman, University of Queensland A syllabic-unit approach to teaching word identification to 4th and 6th grade disabled readers. John Shefelbine, University of Texas The teaching of literature across the grades Teaching and Waring processes in Ilterature,Judith A. Langer, SUNYAlbany National study of the teaching of literature in the secondary school. Arthur N. Applebee, SUNYAlbany The teaching of literatzre in the elementary school. Sean Walmsley SUNYAlbany Perspectives on teachers Teacher characteristics in selected Australian and American cinema= Fifteen years, 021.0 later. Roberta L. Berglund and James P. Raffini, University of WisconsinWhitewater, and Lorraine McDonald, Catholic College of Education Teachers and the use of basal readers: An examination of needs fulfiihnent. Mary Alice Barksdale, West Virginia University Portfolios of elementary literacy instruction: Findings of Stanford's Teacher Assessment Project. Linda G. Vavms, Stanford University Sharing the responsibility for emergent literacy development 021.1 Shared book readbg in an early start program for at-risk children. Jana M. Mason, 021 2 Bonnie M. Kerr, and Shobba S. Sinha, University of Illinois Intergenerational literacy intervention. Linda M. Phillips, Institute for Educational Research & Development, and Jana M. Mason and Bonnie M. Kerr, University of Illinois 021 3 Implementing an early literacy instructhmal model for anrisk kindergarten children. 022.0 Janice P. Stewart, Rutgers University, and Jana M. Mason, University of Illinois Spesling development 022.1 Elementary students' invented spellings: What do they mean? Mary E. Hitchcock, 022 2 Southeastern Oklahoma State University, and Gan E. Tompkins, University of Oklahoma Concept of word and spelling development in beginning readers. Beth R. Spencer, Agnes Scott College 022 3 023.0 023 1 023 2 The relationship between phonemic awareness and spelling ability ammg children in first and third grade. Priscilla L. Griffith and Patricia E. Hanley, University of South Florida Comprehension instruction for atrisk populations: Emerging trends Story grammar and scaffolded bstruction: Teaching literature to low-performing high school students. Joseph Dimino, Long Beach Unified School Disztrict Graphic organiser instruction: Teaching learning disabled students to comprehend and recall science texts. 'ynthia C. Griffin, University of Florida, Deborah C. Simmons, Vanderbilt University, and Edward J. Kamennui, University of Oregon 023 3 Success in reading and writing: Evaluating the curricular experiences of lowashieving children. Melinda Lindsey, Boise State University 0214 An evaluation of two major bilingual education programs: The fourth year in a 024.0 longitudinal study. Susan Schneider, El Paso Independent School District Adult literacy 024 1 lAteracy to% analyses of key entry level jobs in the banking industry. Lany Joseph Mikuieckv, Indiana University 024 2 024 3 025.0 025 1 Reading strategies of marginally literate workers. Verna Haskins Denny, Literacy Assistana Center Assessment sf adults in a workplace environment: The development and validation of a reading and wining measure. Rita M. Bean, University of Pittsburgh Improving ameatment A scale for assessing motivation for reading in subject-matter areas. Mark W. Conley, 46 Michigan State University . NRC Proirrini, Akiiin,--Tixds, 1989 025.2 Annan children's perceptions of the writhag process. Rebecca P. Harlin, State University Collegeluffalo 025.3 026.1 anon* the distracters: Insights on anessment. Deana Seja, Penntylvala State University Trends in reading research: A content armlysis of the NRC .Yearhoolcs. Jeanne 'Shay Schurnm, Uninrsity of Miami, -John Konopak, Lead:hum Stete'Univeisity, R. Scat:Baldwin, UniVersity of Misuri; and John 13: 'Readeniclijuisiani S*'-Unirt*ler. 026.2 The effect ottani/won lise_qatudons pose vicetachereask ier9r .f *id dinte litegSture 026.3 026.4 Assessing tarberSt Conceptlaboat writlei'SallyE) Lipa,'SUNYGeneseo A continuum if comprehension in the understanding of stories by children. Cheryl M. 027.1 027.2 027.3 027.4 028.0 028.1 v28 2 028 3 029.0 029 1 029.2 029 3 030.0 030 1 030 2 030 3 '44.. M Wed*n UnivelOty of MjArtkr_k# Columbia Hruper, SUNYGeneseo Old yoga and new dimensions. Rajendra Prasad Rajgum, Government Holksz Science Cohege How mathematics teachers think about reading. Marjorie Siegel and Raffaella Botasi, University of Rochester Differences between professional and pre-professional teachers' reading habits and attitudes. M. Cecil Smith and Norman A. Stahl, Northern Illinois University Teaching-as-inquiry: A context for qualitative data analysis and collection. Sharon Vincz Andrews, Indiana State University Potentials and realities in university/school projects: Teacher education, writing across the curriculum and college companion courses Writing across the curriculum in a HA school setlingr Assumptions and realities of collaboration, teacher change, and curriculum assessment alignment. Barbara M. Hunter, Sangamon State University Developing expertise in preservice teaching. Lucy Ann Dahlberg, Governms State University Investigating a university/college writing project in two high school settings: The effects of a freshman composition course. Avon Crismore, Indiana UniversityFurdue University at Ft. Waym. Hetping students read subject matter texts. Factors inguendng student-generated questions. Mary C. Shake, University of Kentucky Effects of prior knowledge and transfer of Maluku in graphic orpnizers and summarizing for older below average readers' comprehension of the compare/contrast text structure. Renee Weisberg, Beaver College, and Ernest Balajthy, Slate University of New YorkGeneseo Children's ability to utilize the mnemonic keyword metimd: An educational applicalion with% fourth-grade classrooms. Nancy L. Williams. Louisiana State University When children rend expository text to answer questions Modeling the process of reading expository text to answer questione. Thomas H Anderson, Martha A. Wis, and Linde A. Meyer, University 4 Illinois An experimental investlgaeon of whet students learn when reading expository text. Joseph W. Guenther, Parklend College Fourth and fifth graders' probkate and strategies in =firming inferential questions, Jennifer Lynn Meyer, University of Illinois 031 0 031 I Using retelling as on instructional straitry to improve rending comprehension Retellings: Comprehension strategy as wen as measur:. Don Williams, East Texas State University 031 2 031 3 032.0 Report on use of retelling as an instructknud strategy with adok..ant students of English as a second language. Pi A. twin, Univenity of Mums Story retelling: An active msd strategic mans for awn:using reading comprehension of narrative text. Kyle David Shanton, University, of Arizona A final . eport: The Stratbam evaluation project fAr4 y. 434 032.1 032.2 032.3 033.0 033.1 Literaiy,Theery and, TteiiiiiCk'l The language patterns in evaluation claatrooms. Jane Hansen, University of New Hampshire The path of the chlidren's values. Mary L. Comstock, UniversitY of. New Hampshire The path of the teachers' values. Ann B. Vilna,- University of New Hampshire Basal texts Are basal reading programs draughts? A comparison of types of writing:included in bead reading programs: 1983 and 1989..James Flood and Diane LappirSan,Diego iate University 033.2 What's new about the new reading materials? -Richard D. McCallum; University ot California-Berkeley, and Elizabeth Bondy, University.of Florida 033.3 Emergent literacy strategies: Activities represented in basal readers. Rachel Parse and Lesley Mandel Morrow, Rutgers University 034.1 Kindergarten teachers' knowledge of whole4angnage literacy learnin. Richard P. Ambrose and 13 rferly J. Bruneau, Kent State University 034.2 Skipping through the text: Which words do readers fixate? Paul W. Kerr and George W. McConkie, University 034.3 034.4 035.1 Illinois Librarians: The key to a comprehensive reading program. Mary L. Piersma and Diane D. Allen, University of Alabama in Huntsville. Effects of text considerateness and question generation on recall. Barbara G. Lyman, University of Delaware Preschool teachers' mental models of ideal classrooms: Conceptions of reading and writing. Lea M. McGee, Boston College, and Donald J. Richgels, Northern Illinois University 035 2 035 3 035 4 036.1 037.0 037.1 037 2 Reading In Alzheimer's Disease: Implications for normal reading. Rhoda Au, Boston VA Medical Center Humane literacy: Literacy competence and the ways of knowing. Sheridan Blau, University of California-Santa Barbara Headings and prior knowledge in the search and recall of text. Stephen Clark Wilhite, Widener University Tbe myth of teaching. James Hoffman, University of Texas-Austin Teachers, children and tasks: Complex Interactions in literacy instruction Students' metacognitive response to ambiguous literary tasks. Martha Rapp Haggard, Sonoma State University Novel perceptions of reading: An examination of students' task definitions in a literaturebased reading program. Susan B Murphy, Virginia Tech, and Luther Kirk, Montgomery County Schools 037 3 First-grade children's constructs of reading taska In the cher/room. Pauline Harris, 037 4 Teacher and student perceptions of literacy tasks: Looking for congruence. Candace S. University of Wollongong, Australia Bos, University of Arizona, and Carol V. Lloyd, University of Nebraska at Omaha 038.0 038 I Research narrative In literacy studies: Transformations of self and genre Instractknal choice in language arts: Reality or illusion. Robin L. Hensley, Texas Woman's Uni versity 1138 2 Heroes in reading teachers' tales. James R. King, Texas Woman's University 0..,0 3 Reactions to change: The results of implementing the Right to Read program In one 039.0 039 1 039 2 039 3 Texas school. Venetia M. Braune, Texas Woman's University Word identification and spelling Polysyllabk decoding and spelling for middle school students. Patricia M. Cunningham, Wake Forest University The relationship of spelling and word recognition. David A. Koppenhaver, University of North Carolina at Chapel Hill Investigating the relationship between word identification and spelling. Dorothy P. Hall, University of North Carolina at Greensboro At' A...Y.16mM 040.0 040.1 Readin instruction and learning to read in the inner city The reading preparedness and achkvement of inner city first graders. Laurie Nelson, National College of Education 040.2 040.3 041.0 041.1 041.2 041.3 042.0 042.1 042.2 042.3 043.0 043.1 043.2 043.3 044.1 044.2 Developing low-performing fourth grade inner city, students' ability to comprehend narrative. James Mosenthal, Michigan State University The effectiveness of an after school tutorial reading program with low-performing inner clty students. Darrell Morris, Appalachian State Unive.sity Literature-based and whole-language reading programs The effects of a literature-based approach to reading instruction in grades 1-6. Barbara M. Taylor and Barbara J.. Frye, University of Minnesota Cast studies of six low-prokieney readers: Literacy learning in a whole-language classroom. Joan R. Bock, University of Cincinnati Growth in comprehension, metacognition, and writing in Uteretcre-based classrooms: A follow-up study. Patricia Hagerty and Mary Katherine Owens, Adams County, CO #12 School District, Elfrieda H. Hiebett and Charles W. Fisher, University of ColoradoBoulder Assessing written retellings: A comparison of methods Using a macrostructure to analyze retellings. David A. Hayes and John Ponder, University of Georgia Using narrative analysis to analyze retellinp. Kam Williams, University of Central Florida Using a holistic approach to analyze retellings. Connie Ulmer and Jane White, East Texas State University Phonemic awareness Correlates of phonemic awareness in pre-school children. Ann Hail, University of Texas at Austin The effects of phoneme awareness training and repeated readings on Intermediate-grade level disabled readers. Peter A. Dewitz and Martha Skilliter, The University of Toledo Phonological awm ss and phonological procesning: An analytic review. Lois G. Dreyer, Teachers College, Columbia University Videotape feedback: Effect on preservice teachers' knowledge of and attitudes towards content area reading. Trine Zych and Beverly E. Cox, Purdue University Understending primary teachers' referral practices. Kathleen A.Broikou, SUNV Geneseo 044.3 A multidimensional analysis of teacher-led reading group discussion at the second-, fourth-, and sixth-grade levels. John F. O'Flahavan, University of tvlaryland-College Park, and Douglas K. Hartman and P. David Pearson, 'aniversity of Illinois 044.4 Instruction and standardized testing in the kindergarten: Breaking the cycle. Carol A. Hodges, State University College at Buffalo 045.1 045.2 045.3 046.1 047.1 Reflecting about the reletionships between reading to learn and sclem instruction. A study of inservice teachers. Rosary V. Lalik and George E. Gleason, Virginia Tech Standardized measurea of reading achievement for placement of children in special and remedial programs: A nationwide snrvey of state practices. Jeannie L. Steele, UniversitY of Nonbem Iowa, and Kunis Meredith, Univrsity of Iowa The effects of a storybook reading program on the literacy development of at-risk kindergarten children. Ellen O'Connor, Bayonne Public Schools, and Leslie Mandel Morrow, Rutgers University Literacy in school and out. Lauren Resnick, University of Pittsburgh What's happening in the strum? A report on current literacy issues and programs. Nancy Eberhart, Ohio State Department of Education, Wilmer S. Cody, Louisana State Department of Education, W. N. Kirby, Texas Education Agency, and Donald L. Bemis, Michigan Department of Education 048.0 Instructional decision making in exemplary middle school reading classrooms '.7.-i , _.. . _ 436 :, ,,,' , ,... ,.. 5 '''''',-^ : :i: -a' g; '", . 048,1 048.2 'Litiiiii4 :the* ilicliteiii4::-T, - Instrnetional decision making in exemplary middle achool reading classrooms. LeslieAnn. , Patterson and Joan Pronty,' Sam Houston State Uniiersity . ,.,-i., Inatructional declaims: Integrating literatarti vicabulary; critical thiainiqad writing instruction for, inefficient 'Lliiiddie . school readers: Annette 'Spikes,- 'Conrni.InciePeadeut District, Travis Junior High , 048.3 Instrisctionsit docisiontiv Seikekcted materials for 'math-grade reader& Linda Bilis, Lovelsdi Independent *Sc:hoell Distal 049.0 049.1 Towanta theory of practletf-.Whole liagaige and tbeat-Zhlt kaiser Whole linguae 'Vt., traditiMig indruction: The reseircla ,. milieu. John E. ,Bertrand, Tennessee State University 049.2 Four models derived fermi two types of instructkinal programs: R'halteicber* datielsat' children do la whok-langnage and traditional dassrooms.Camle F. Stice,-tenneisee -Stare . University .. _ ' 41 Salkut differences in the artifacts produced in whole-language and traditional classrooms. Nancy P.Bertrand, Middle Tennessee State University 050.0 Tbe Unending study of reading acquisition Grades kindergarten-2. Linda A. Moyer, University of Illinois 050.3' 051.0 051.1 051 2 051.3 051.4 052.0 052.1 052.2 Authentic literacy, authentic tasks: Learning hi bilingual, cooperative, and whole- language contexts Dialogue journals and literature logs: A comparison of native and L2 discourse tasks. Maria de la Luz Reyes, University of ColoradoBoulder Constructing meaning through sharing: A priority in authentic tasks. Charles W. Fisher and Elfrieda H. Hiebett, University of ColoradoBoulder Capitalizing on Hispanic students' strengths: The search for "ability aren". Ofelia Miramontes, University of ColoradoBoulder Authentic literacy tasks and cooperative learning. Michael S. Meloth and Paul D. Deering, University of ColoradoBoulder CompariN amessment practices A descriptive stmly of the reflecdve statements of preservice teachera in a reading clinic setting. Ball= I. Walker, Eastern Montana College The comparative value of shilis and holislic assessment in the elementary reading Comparing two dinnostic Nrocedures. Ronald P. Carver, University of MirouriKansas City 053.0 053,1 Students and families at risk The San Antonio Literacy Project: A study of students at risk. Rosalind Horowitz and Lucy Frontera, University of TexasSan Antonio 053.2 053.3 054.1 Families at risk: Parental pesceptions of young children's literacy development. Jill Fitzgerald, Dixic Spiegel and Jim Cunningham, University of North Carolina at Chapel Hill Narrative skills and literacy learning: An examination of story-ing and reporting among low-Income and middie-hicome first graders. Deborah A. Hicks, University of Delaware Elementary teachers' rationales for instructional decisions. Leslie Ann Panerson, Sam Houston State University. and Fran Mote and Susan Thompson. Huntsville Independent School District 054.2 ', Kindergar'=. Darrell Moths, Appalachian State University Grades 1-4. Connie Joel, University of Texas at Austin program. Peter A. Dewitz, University of Toledo, and I_ annum Scheibal and Vicki Johnson, Kirkwood School District 052.3 -- , _ 049.3 050.1 050.2 = Construct vaiidity of the degrees of reading power test. Thomas H. Estes and Herbert C. Richards. University of Virginia, and Elizabeth WetmoreRogers, Albermarie County Public Schools 4-30- --; -,,' :-.;' ...,..-) ,,,,, -,.'.. 4 -7,' ,.-,4 A --,=--'i - '',1.1 --,' 1 NRC Program, Austin, Texas, 1989 437 054.3 Th2 effects of an Integrated curriculum on etudents' summarization of scifsece selections and on teachers' and students' perceptions ot teaching and learning. Frances E. Halliday, McGill University 055.1 The we of response journals in a reading methods course: A barometer of students' 055.2 karning. Diane D. Allen, University ofAtabama in Huntsville An exploratory and confirmatory factor_ analysis of the Learning and Study Strategies Inventory (LASSI). Stephen C ninik and Sherrie Nist, University of Georgia 055.? An exploratory study of practicing teachers' use of study guides in content arta classrooms. Carol L. Peterman, David B.Dunnin, and M. Carrot Tanta; Penland State University 056.0 056.1 056.2 056.3 057.0 057.1 Multiple perspectives on early literacy developments A conlunction and syntheds of Piagetian, semiotic, awl' transitional knowledge interpretations of conceptual change A *sedan interpretation of the nature and development etif children's spontaneous questions during storybook reading. David B. Yaden, University of Houston Searching for competencies: A semiotic perspective on early literacy learning. Deborah Wells Rowe, Peabody College, Vanderbilt Univexsity Empirical and theoretical perspectives oa transitional knowledge in emergent literacy. George Kamberelis, University of Michigan Preservice teachers' knowledge structures Exploring the relationship between presentee teachers' knowledge structures about reading and their instructional effectiveness. Sara D. Weidier. State University College at Buffalo 057.2 Exploring preservice and expert content area reartag teachers' evolving knowledge 057.3 structures. Sherrie L Shugarman, University of Dayton A longitudinal study of the evolution of preservice teachers' knowledge structures. Beth Ann Hemnann. University of South Carolina 058.0 Orthographic knowledge and oral reading of beginning and transitional readers: 058.1 Interactions of pattern and prosody Reading fluency In beginning readers and expression In practiced oral readinv Links with word knowledge. Donald R. Bea, Jul Sharon Cathey, University of Nevada-Reno 058.2 Normal and disabled spellers on achievement levels one through four: Ore*. graphic awareness and oral reading. Marcia A. liremini and Jo Worthy, University of Virginia 058.3 Reading and spelling connections In third-grade students, II. Jerry Zucca, Ohio State University 059.0 059.1 059.2 059.3 060.0 064.1 Responding to literature Insight into literature: Learning to Interpret inside view and character plans in fiction. Cheryl Rappaport Liebling, BBN Systems and Technologies Corporation A child's developing sense of theme as a response to Literature. Susan S. Lehr. Skidmore College Reader awareness of emotions in liter ature: The reader/character connection. Jim Banton, Stanford University Factors affeding teachers' practices In reading comprehension instruction School and district influsona on reading insaruciion: The testing arid instruction connection. Virginia Richardson and Patricia Anders. University of Arizona 060.2 Relationships between teachers' beliefs and practices In reading comprehension Instruction. Carol V. Lloyd, University of Nebraska at Omaha, and Deborah Tidwell, University of Arizona 060.3 Research Into practice: Effects of a practical arguments staff development program. Patricia Anders, Mary Lynn Hamilton, and Virginia Richardson, University of Arizona 061.0 MIA Dev(opments in second Immune reading research: A focus on Hebrew, Japanese, and EDO& Factors affecting foreign Marine text comprehension. Kaiko Koda, Ohio University 061.2 The effects of article deletion in English on the cognitive processes of native and Japanese readers'of Yoshiko Salto, Ohio State University 061.3 Deve:spenent of beide read* skirls in L2: The case of Hebrew. Esther Geva, Ontario Institute for Smdies in Education 062.0 062.1 Talk and literacy Comartsding conversatkm: Peer responses to stndent writing. Sarah J. McCarthey, Michigan State University 062.2 062.3 Center talk: Gateway to literacy. Mary Williams Aylor and Donna J. Camp, East Central Univessity, Florida Oral language lire. to literacy: Grade 1. Karen F. Thomas and Steven Rinehart, West Virginia University 063.1 063.2 063.3 064 I A case study of reBective coaching in collaboration with teachms implementing emergent literacy concepts in their kindergarten protrams. Beverly J. Bnmeau, Kent State University Validating Webb's Hypothesis: Students' judgments of good and peor analogim in high school biology. Thomas W. Bean, University of Hawaii at Hilo, and Sian Cowen, Garden Grove High School Developing bockground for expository text: PReP revisited. Linda A. Moiner, Univasity of Colorado/Boulder 065.0 A collaborative project Two perspectives plus. Lenore H. Ringlet, New York University, Hindy M. Litt, Community School District #2, and Michael J. I-nrker, New York University Understanding expository texts: Collaborative reading in the elementary classroom. Anne Goudvis, University of Illinois The ten best eas for reading teachers: A study of values 065.1 What are some alternate sources of best Ideas? Sara B. Nix, University of Califor- 064.2 065 2 065 3 066.0 066 I nia-Riverside What are some unique best ideas? Hallie Yopp, California State University-Fullerton What are the 10 best ideas? Edward Fry, University of California-Riverside Preservice and inservice teachers' reasoning. Stake that claim: The content of pedagogical reasoning. Victor M. Rentel and Gay Su Pinnell, Ohio State University 066.2 Teachers' rationales for their adaptive Instructional actions: The reasons teachers give 066 3 for doing what they do. Janet Johnson and Laura Rockier, Michigan State University Learning to be reflective: The tiltkicg processes of student teachers involved In a coarse on reading diagnosis. Jerome A. Niles, Rosary V. Lan, Susan B. Murphy, and Charies Lucado, Virginia Tech, and Beverly J. Bnweau, Kent State University 067.0 067 1 Reducing the risks: An investigation of literacy learning students at risk Social Issues for literacy: Fifth grade is another world. Marsha West, Clarke County School District 067 2 Engagement and community in a second-grade classroom. Barbara Michalove, Clarke County School District 067 3 061 4 067 5 A cniversity-palle school research team investigates reducing the risks for young literacy learners. JoBeth Allen, University of Georgia Curricular innovation and teachers' values: Teachers make decisims that reduce risks. Sherrie Gibney, Clarke County School District Sing a song of Jonas: Multiple risks, multiple strengths. Betty Shockley, Clarke County Szhool District 068.0 Writing development 432 Iti.i111111111.1,- NRC Program, Attila, Texas, 1989 068.1 439 Adolescent writers' revising stratrgies. Camlyn Sue Andrews-Beck, Kent State University 068.2 068.3 069.0 069.1 069.2 Learning to write persuarively: A study of writing instruction in the social studies class. Martin A. Rybczynski, Bowling Green State University Changes In the struttureof third-grade students' written composition. Ruth M. Caswell and A. Rota= Wilson, Texas Woman's University Reading and writing: Processes and 'Malin The relationship of domain akin, and general reasoning ability to reading comprehension and writing expression: Au Interactive conceptualization. Duane F. Shell, Christy A. Hor, and Roger H. Bruning, University of Nebraska-Lincoln A description of tbe meaning making strategies used by seven subject., &ring reading and writing. Sarah H.Martin, Eastern Michigan University 069.3 The nature of the reading/writing relationship: A study of experienced readers and writers. Mary Alice Barksdale, Painne Gerlac, and W. Michael Rixi, West Vir8inia 070.1 Initiation into the research community: Learning front process, collaboratoon, analysis. Panel: Jane White, East Tens State University, Michael Sampson, Zast Texas State University University, Pat Human, East Texas State University, Carol DeRita, Potter Elementary School, Linda Lewis-White, Gabe Allen Elementary School, Lory Fetzer, Gabe Allen Elementary, Scott Beesley, Roundnee Elementary School, and Sandra Bierman, East Texas State University 071.0 071.1 071.2 071.3 012.1 072.2 Helping the learning dissbled reader negotiate meanings Responding to narrative texts: Responses of students Identified as lamming disabled Cynthia Brabson, Indiana University, and Patricia Tefft Cousin, California State University-San Bernardino Teaching decision-making via narrative text. Joanna P. Williams, Columbia University Interactive learning to facilitate learning disabled students' transition from novice to expert. Margaret A. Gallego, Michigan Suitt University, and Grace Z. Ours and David J. Scanlon, University of Arizona Researching teacher theories: How do we study THEORY? Sharon C. Lee, University of South Dakota, Jill Burke, Texas A&M Umversity, and Leslie Ann Parenton, Sant Houston State University A logical foundation for feature-based theories of letter and word recognition. John E. McEneaney, University of Georgia 072 3 072.4 Literacythe final destination? Sambre Mina Rarnakant, Devi Ahliya University The relationship between scientific literacy and high school biology textbooks. Carol V. 073.1 073.2 Lloyd, University of Nebraska at Omaha School contexts of literacy Instruction. Mrty E. Robbins, Texas Woman's University The components of tiull decoding. Cynthia L.Peterson, David S.Kreiner, and Philip B. Gough, University of Texas at Austin, and Wesley A. Hoover, Southwest Educational Development Lab 073.3 A struggle for control: Four conceptions of the literacy curriculum. Patnck Shannon, 074.0 074.1 Emergent and conventional literacy: Issues in development and transition The transition to conventional literacy: Theoretical consIderations. Elizabeth Sulzby, 1714.2 University of Michigan Markers of cognitive change in the transitional period between emergent and conventional ifieracy. George Kamberelis. University of Michigan 074.3 Differential patterns of spoken and written language development in second-grade University of Minnesota/Duluth children. June E. Barnhart, Northern Illinois University 075.0 Student research strategies 433 Literiicytkoiy ind.keseakh 075.1 Intentionality and awareness of cognitive peveemes in text search...John T. Guthrie, 075.2 The%roin olthborat1on and inference frs arguing from: rooms. Victoria Ellen Stein, University of Maryland University of Azizona - , 075.3 Whet does it tske for college students to "buy In" to study strategy nse? Maribeth Caealdy Schmitt, DePahw University 076.: InstrectiOna effKts in the cannections between earlireading and !paling . The kiiineste or phonies instenition on spelling progress: Laurie Nell* Natleatel c2Polio 076.1 of Education._ 076.2 076.3 - The infinence dreading !Ekstrom:don on the spelling ot bawd ind non-band wank:Rebecca Barr, National College of Education - The interactive rehaiondilp between spelling and reading. Edmund H. Hendeison, University of Virginia (Delivered by 'Thomas Gill, University o4 077.0 077.1 Graphic aids to comprehension: Trends and Imes Graphics and coo:ceche:Wow Tres& In research and practice. Lawrence B. Friedman and Mama Banker Tmzinsnin North Central Regional Educational Laboratory 077.2 Graphic ads in printed and computer-mediated texts. David Ranking, University of 077.3 Effective and ineffective graphic representations in besal readers and social studies 077.4 textbooks. Barbara M. Hunter, Sangamon State University Visuals ar.a rtetoric: Argument, audience, discourse and metatliscomac. Avon Crismore, Indiana Unlversity-Pindue University at Ft.Wayne 078.0 078.1 078.2 Contedetiliting literacy Instruction within the middle school classroom: An Integrative reset, ch programme Tbe yelationehlp among literacy tasks, tar:her-dna conversations and students' learning in rodal studies damn Mak W. Anlis, McGill University The differential Induence of knowiedge of elgoale to hoPortenoe on eleith graders' accaracy In representing content and organize.= of essays. Danielle Michaud, McGill 2 University 078.3 078.4 079.0 079.1 079.2 The nature of task systems. Donna Goloff, McGill University The effects of an integrated curriculum on students' eumniarhadon of science selections and on teachers' and students' perceptions of teaching and learning. Frances E. Halliday, McGill University Assessing reading comprehension whollstically An overview of the project. Jerome C. Harste, Indiana University Exploring think-elands as assessment. Insights and patterns. Cheryl Ann Kelleher, Margaret E. Chas, Yueh-Hung Tam, Al Meng, Chaedar Alwasilah and Cibel Ceareak, Indiana University 079.3 77.xploring retellings as assessment Intights and patterns. Paul Michael Chandler, Naomi Ono, ?alba Sad Mustaphe and Cite Tomiella, Indiana University 079 4 Exploring Lee:writes as amassment: Insights and patterns. William P. Matz, Caroline Bevnstock, lob), Gansauge Copenhave, Terry W. Hughes, and Tunothy L. Farley, Indiana University 080.0 080.1 Reading and writing in content arm Research and report wad. I throne+ the grades: A critique of the theoretical, research, 080.2 and pedagogical litizatnre. Sandra Streaky, Harvard University Elementary students' acqubition asocial studies knowledge throe:0 discount. 1. Michael Gee, Ohio Stine University 081.1 Multicultural children's literature hi the 1980s. Kathryn Meyer Reimer, University of illinois ,...4111.111 441 NRC Program, Austin, Texas, 1989 081.2 Thinkbsg-alood: An examination of Its transfer to other learning situations. 3 081.3 Eh linger, Wbona State University Toward an understanding of literacy development across various writing contexts. Marcia Guddemi and Heidi Mills, University of South Carolina 082.1 A critical review- of investigations of sexism in basal readers published in 1929-1985. Lynda R. Medd= Alma College 082.2 A comparison of instroction in three compnheasion activitiet: Question/answer discussion, discussion of written answers to compreisensicat quids% and oral corrections. Vienna K. Moore, Oglethorpe University, and David Wendier, Dr. Martin Luther College 082.3 083.1 084.0 084.1 084.2 The relationship between teacher conceptions and student mencognitive mint end performance In writing. Kathleen L Fear, Albion College Motivated literacy. Mazy McCaslin Robrkemper, Bryn Mawr Paradigms for teacherlresearcher collaboration in secondary school literacy instruction The effects of reciprocal teaching in a secondary school ESL life sdence classroom. Diane Lapp and James Flood, San Diego State University, and Doris Alvarez, Hoover High School Collaborating to understand teaching and learning in the 9e:canary school. Joseph Ruhl, Lafayette Jefferson High School, and Deborah Dillon and David O'Brien, Purdue University 084.3 Meaning of literacy in the lives of at-risk students in a rural mcondary school. Doan Alvennann, University of Georgia, and Richard Umpleby, Burke County Comprehensive High School 085.0 085 1 085.2 085.3 086.0 086.1 Learning as authoring: A collaborative study in a transition-flria grade CialWOOM Learning as authoring. Timothy T. O'Keefe, R. Earle Davis Elementary School Learning across sign systems. David J.Whitin, University of South Carolina Halliday's perspeclive on learning in language end mathematics. Heidi Mills. Univ, .ity of South Carolina Preschoolers' reading The development of story schema and book language knowledge in '--ser cfty kindergarteners exposed to two different programs. Lynne R. Putnam, George aington University 086.2 A comparison of Intellectually superior preschool accelerated readers and nonreaders: Three years later. Jeanne M. Bums, Southeastern Louisiana University. and Martha D Collins, Louisiana State University 086.3 Emergent writing and rereading among "academically able" preschoolers. Beverly Otto, 087.0 087.1 Northeastern Illinois University, and Elizabeth Sulzby, University of Michigan Affective dimensions of literacy Discourse involvement: An investigation of a cognitive/motivational construct in academic tasks. JoyLynn H. Reed and Diane L. Schallert, University of Texas at Austin 087.2 087 3 The effects of proximal goab on students' attitudes towards learning, Eileen Margaret Can, University of Toledo The structure of emotional response in reading; Quantitative and qualitative analyses. Ernest T. Goetz, Mark Sadosb, Arturo Olivarez,Jr., Ayxa Calero, and Pamela Gamer, Texas A&M University IT 088.0 088.1 Instruction, strategies and metacognition of high and low achievers In first- and thirdgrade whole-language classrooms Thinking about reading and writing: Metacognition of students in whole-language classrooms. Laurie Jundt sad Joyce Downing, University of Colorado-Boulder 088.2 Reading and writing instruction in wh.ole-language classrooms. Tonda Potts, St Vram Valley (CO) Schools, and Jacqueline Papiem, University of Cobrado 43 5 Litency Theory and Rasarch 088.3 089.0 089.1 089.2 089 3 090.0 090.1 090.2 Knowing about reading and writing: Strategies of students ha whok-langsage classroom. Anne Goudvis, University of Illinois, and Nancy Burton, University of, Colorado-Boulder Authorship Authomlalp and critical reeding. Timothy Shanahan, University of Illinois at Chicago k descriptive analysis of good readere sad writers' concepts of aitimeshinat grades I, 3, and S. Robert James Nistier, Univeraity of North Texas Teal mati text, reader meets writer. Trevor Henry Cairney, Rivezina-Murray Institute for Higher Education Early litesacy-development Composing knees: A case study of learning to write. Donald J. Richgels, NorthernIllinois University, end Lea M. McGee, Boston College First graders' selkelected wridar A developmental study*. Elizabeth G. Pryor, Revere (Ohio) Schools, and Nancy D. Padak, Kent State University 090.3 OM 091 2 091 3 Au investigation of children's concepts of the purpose and nature of reading in different instroctional settings. Penny A. Freppon, Thomas Mote College Teacher as researcher: Using student journals to evaluate college instruction.Jane White. East Texas State University Exploring the reading-aloud curriculum genre: A soda-semiotic perspecdve. Christine C. Pappas, University of Illinois at Chicago Developing reading fluency and story comprehension using a redtation lesson framework. James V. Hoffman, Mary Ellen Isaac, Nancy L. Roser, and Cindy Farts*, University of Texas at Austin 091 4 Reading and spelling development aflame kindergarteners. Susan S. Robinson, Iowa State University, Janice Parkinsor Moines Public Schools, and 092 1 Teacher as researcher: Using multiple measures to evaluate Instruction. Kann Feathers, East Texas State Univasity The literacy symbol in the dossroom coutest. Joanne M. Golden, University of Delaware, Consultants 092 2 and Annyce Gerber, Indian Oasis Elementary School 092 3 092 4 Changing teacher behaviors to improve chlidren's oral reading. Sharon Arthur MOOfe and David W.Moore, Arizona State University-Wen S pelting error patterns of blind and sighted childreo. Dale M. Willows, Fivi Chitiri, and Deborah Mayne Ontario Institute for Studies in Education 093 0 Evalration of the teaching and learning of literacy by the stakeholders in a schol system 093.1 The composition of the report card: Making the grade hi the impute axis. Peter 093 2 Afilerbach and Janet Lynn Norton, Emory University, and Peter Johnston SUNY-Albany Teachers' evassIkan of children's literacy development. Peter Jot= usn and Paula B. Weiss, SUNY-Albany 093 3 094.0 094 1 The relationship between stakeholders' evaluations of literacy instruction. Paula B. Weiss and Peter Johnston, SUNY-Albany Teacher education English as a second language teachers and literacy: Learning from a graduate level course on bilingualism and literacy. Eleni Kokkino. University of Illinois at Chicago. and NgocDiep ml Nguyen. Illinois Resourx Center 094 2 Collaborating in coursework and in classrooms: The impact of pre-aervkauservice 094 3 teacher partnerships on risk-taking with literacy Instruction. Jean Anne Clyde and Mtrk W. F.Condon, University of Louisville From reading methods' coursewark to actual classroom diagnosis and InstructionWhat transfer? Christy Foley, University nf Guam Whole-language Instruction and special populations 095.0 43 NRC Program, Austin, &cat, 1989 095.1 Whole lasianage and prtnutry ESL students: The effects of models on writing development. Lee Gunderson and Jon Shapiro, University of British Columbia 095.2 095.3 096.0 696.1 096.2 096.3 Effects of whole language on tannage delayed children. Sally E. Ups, SUNY-Genesco Effects of whole language on Sow SES children. Rebecca P. Harlin; State University College-Buffaln Affective dImensions of literacy II Prior knowledge and content-related attitudes: Testing the Mathewson hpaothests. D. Ray Reutzel and Paul M. Hollingsworth, Brighina Young University Affective involvement and its effect on comprehemion. Robert Gas/tins, WNY-Albany Children': attitudes tom:4 reading: Secondary analysis of data farxin a Dadaist study. Mich"41 C. McKenna, Deonis L Kea, and Randolph Ellsworth, Wichita State Univesity 097.0 097,1 097 2 Ten processing Carat network theory and text comp ension. S. lay Samuels, University of Minnesota Dynamic working memory as a corn tett of reading: Separating memory skill and domain-specific -NW! David S. Kreir el Patrick lames Carroll, University of Texas at Austin 097 3 098.0 098 1 098.2 098 3 099.0 099 1 099.2 099 3 100 1 100 2 101 1 Explaining the effects of a comprehension taxonomy: Levels of processing and transferappropriate processing. Mark W.Conley, Michigan State University College reading: Issues of assessment for a diverse student population Issues and theoretical models for college reading assessment. Michele L. Simpson and Sherrie Nig. University of Georgia Testing and teaching as eimultaneous events: Interactive assessment of college developmental renders. William G. Brom, Eastern Michigan University Learning and study strategies of college students in four countries: A comparative study. Bonnie C. Higgmson. Murray State University. Norman A. Stahl, Northern Illinois University. Lce Sung-ho. Hanyand University, and Ming-y, Yang. Yunnan Normal University Classroom studies of literacy inetruction The diffesential effects of three procedures for teaching strategic reading: A research report. Annemarie Sullivan Palincsar. Yvonne Mane Dam Judith Win. Barbara S. Snyria. and Darmelle Stevens. University of Michigan The nature of task system>. Donna Goloff. McGill University Teacher and student Inter:drib patterns during interactive and direct instructional practices. Margaret A. Gallego, Michigan State University Comprebemion and recall of narrative vs. expository text: A developmental study. Lauren Leslie. Marquette University Opportunity and the development of competence: An investigation of culture and control in student/Gs.. 'ter interaction over the course of a Mar-month reading tutorial. Paula B. Weiss. SUNY-Albany The construction of narratives by normal and poor readers. Rose-Msne Weber. 102.0 SUNY-Albany Effects of a cross-age tutoring program on first g:aders' listening comprehension. Joanne L. Ratliff. University of Central Florida. Bonnie W. Maddock. 'voltam County Schools. and Ray R. Buss. University of Northern Iowa Parents' perceptions of children's meting and writing development in a whole-Language kindergarten program. Beverly J. Bruneau. Timothy V. Rasinski. and Richard P. ianbtose. Kent State University Assessment and decision making ln ti.r schools 102 1 Decision-making from a variety of perspectives. Michelle Commcyras. University of 101 2 101 3 Illinois, Bill Johnston. University of NC-Wilmington. Therasa Rogers and Patncia Scharer. Ohio State University. Mary Roe, Ahem Rodriquez. Judith K. Sheik). and Anne C. Stalling:1. University of Illinois 444 Literacy Theory and Research 102.2 A qualitative approach to studying decision making in the schools. Diane Stephens and 103.0 The teacher's role in reading and writing instruct:en Teacher: Orchestrator or authoritarian during meaning construction? Judith G. Gasser, P. Davis! Pearson, University of Illinois 103.1 Texas Woman's University 103.2 The effects of at-uctural factors of expository texts on teachers' judgments of writing quality. Beverly E. Cox, Pardue University 103.3 Reading as conversation: Collaboration in the teacher-student s /fling conf Melanie Sperling, University of California-Berkt: 104.0 104.1 Whole language in different ethnofinguistic situation Indian students' literacy development in a wbele-leaguage daearoom. Karen Guilfoyle, 104.2 Documenting a whok-language clamant: A pilot study anyaine Marie Beefier, University of Arizona 104.3 105.0 105.1 University of New England, and William P.Bintz, Indiana University Ethnolinguistic differences in reading-writing relationsh4so. Victor Froese, University ut British Columbia Reader response s.ad engagement Thinking in play: A young chlid's response to literature. Shelby Anne Wolf, ::-..mford University 105 2 Fifth graders' responses to reading a play: An anminallon of knowledge revealed tirough reader response and speech set interruption points. Laura B. Smolkin, University of Houston 105 3 The effect of reader stance on students' personal understandbm of literature. Joyce E. Many, Louisiara State University 106.0 106 1 Reading and learning from tort in the classroom: The teacher variable An interactive Instructional model of reading: The teacher's role in learning from text Mariam Jean Dreher, University of Maryland 106 2 106 3 Teaching students to learn rim text: Preservice content teachers' changing view of their role through the window of student-professor dialogue Journals. Thomas W. Bean and Jan Zulich, University of Hawaii at Hilo A comparative study of the goals, plans, and monitoring strategies used by Infinendial teachers in contrast to non-Influential teachers. Robert B. Ruddell, University of California-Berkeley 106 4 Factors affecting teschers' instructional decision making: Text materials and teachers' knowledge about the reading process. Nancy Rogers-Zegarra, Sonoma County Office of Education 107.0 107 1 Processing and attention difficulties Auditory perceptual processing in shit and disabled readers. Jon Shapiro, Umversity of British Columbia, Gary W. Nix, Richmond Education Clinic, and Stephen F. Foster, University of Limburg 107 2 Visual and auditory perceptual and short-term memory processing of dyslexks. Chene 107 3 Attention deficit disorder: A review of research literature. Cassandra K. Meents, De Jong, California State University-Los Angeles SUNY-Albany 108.0 108 I Discourse structure in learning in LI and L2 Discourse structure research in the 1990s. Rosalind Horowitz, University of Texas-San Antonio 108 2 Discourse structure in visual narratives of the deaf. Madelyn Maxwell, University of 108 3 Texas-Austin Discourse structure in science for ESL learners. Carolyn Kmsler, University of Texas-San Antonio 436 NRC Program, Austin, Texas, 1989 108.4 445 Deaf strdents' use of structure In ESL text processing. Tane A. Akamatsu, Michigan State University 108.5 Discourse structure in foreign language reading. James Davis, Penn State University 109.1 Teachees' developing a&-ghts about the use of children's ilteratve for langnage and literacy growth. Nancy L. Roser, James V. Hoffman, Cindy Fairest, Mary Ellen Isaac, and Jennifer Battle, University of Teaas at Austin 109.2 Expanding the role of technology In teacher preparaticn: A pilot study. Elaine Yates- 109.3 How bawds teach strategies to derive word meaning from context. Peter J. L. Fisher, National College of Education, and Alicia Maier and Judith C. Smith, Naperville Selma: Hendrix and Ritchie S. Kelley, Pennsylvania State University District 110.1 A description of teacher-talk during kindergarten sharing time. Debra S. Elliott, University of Northern Iowa, and Sharon Arthur Moore, Arizona State UniversityWest 110.2 Teachers "talk" about student writing processes and writing !attraction: An analysis of a computer conference "text". Lawrence B. Friedman, Ninth Central Regional Educational Laboratory, and James McCullough, Petoskey High School Ilr 3 Lexical cohesion in comprehension and composition: A synthesis at research " 111.0 Richard B. Speaker, Jr., John G. Barnitz, and Joan P. Gipe, University of New Ork Studies of teachers and teaching in early literacy Intimation for ct-risk children 111.1 111.2 Characteristics of teachers who are particularly snccessful in accelerating at-risk first graders' progress In reading. Carol A. Lyons and Nora L. White, Ohio State University Early literacy intervention study. Diane E. DeFord and Eleanor Hanuettan, Ohio State University 111.3 112.0 112.1 112 2 112.3 Teachers' applications of theoretical concepts to new Instructional settings. Gay Su Pinnell and Andrea McCarrier, Ohio State University Comprehension strategies Sometimes people miss main ideas and do not realize it. Michael Pressley, University of Maryland, and Elizabeth S. Ghatala, ' Tniversity of Houston Knowledge, use, and control of an Interactive cognitive strategy for learning from content area texts. Candace S. Bos and Elba I. Reyes, University of Arizona Stratem generalization instruction for dbabled readers. barna Kim-Sang Chan, University of Newcastle 114 0 Vocabulary studies 114 I . The effects of context on the spontaneous instantiation c general terra by third, fifth, and seventh graders. James J.Martin-Rehnuann, Syracuse University 114 2 114 3 115 0 The effects of preteaching vocabulary on word knowledge and content area text comprehension. Donna Lynn Mealey, University of Georgia Fourth graders' knowledge of definitions and how they work. Judith A. Scott ant_ Villiam E. Nagy, University of Illinois Developing reading-writing connections 115 1 The effect of an intwated writing strategy on fourth graders'comprehension of content are. marterial. Michael A. Martin and Sarah H. Munn, Eastern Michigan University, and 115 2 The reading-writing connection: An instructional intervention in fourth grade. Gerry Bonnie Konopak, Louisiana State University Shiel and Charles H. Clark, Western Illinois University 115 3 Using the whole picture. Nelly Hecker, Shirley A. Rine, and Garman B. Smith, Furman University 116 0 116 1 116.2 The Lextle framework: In theory and practice Testing the power of the LeAle theory. A. Jackson Stenner, MetaMetrics Construct definition and the Lexile theory of reading comprehensi.na. Malbert Smith, Computerland Literacy Theory and Research 116.3 Applications of the Lexi le theory. Dean R. Smith, Meta Metrics 117.0 Teaching ESL and limited English programs Student roles En peer review responses. Ann S. Schlumberger and Kate W. Mangelsdorf, 117.1 University of Arizona 117.2 117.3 The role of metacognition in facilitative transfer of expository comprehension abilities: The unilingual and cross-lingual cases. Robin Ave lar La Salle, Stanford University Inters...tive journal writing strategies for English language and literacy development of young LEP shidesits. Shareen Abramson, California State UniversityFresno, Ileana Seda, Pennsylvania State University, and Caw ly Johnson, Hidalgo Elementary School 118.1 An exploration of Imagery reports during and after reading: Imagery modalities and 118.2 elements of prior knowledge. Shirley Long, Transylvania University Children's choice of and placement in reading materials. Rhoda Q. Spiro, SUNYAlbany 118.3 Immersion into two academic disciplines k longitudinal study. Maureen A. Mathison, Carnegie Mellon University 119.1 Effects of mental imagery training on gifted students' creative writing. Ellen Jampole, 119 2 Bonnie, Konopak and John E.Readence, Louisiana Stan University An mullysis of formal measures of early literacy. P. David Pearson and Anne C. Staliman, Univei-ity of Illinois 119.3 College students in remedial reading classes interacting with stories written in 120.1 121.0 121 1 chronological time order and time-shift. Judith Entes. Baruch College Reading-writing connections: The relationship among three research traditions. Timothy Shanahan, University of IllinoisChicago Impacting thinking through reading instruction Teaching and learning for America's future. Beau Jones, North Central Regional Education Lab 121 2 Thinking-reading-writing: Literacy for problem solving and communication. Robta Calfee, Stanford University 121 3 121 4 Effects of challenging at-risk students to think during reading: The 35-minute principle. Stanley Pogrow, University of Arizona A new approach to middle school reading programs: Expanded thinking. Cathy Collins, Texas Christian University 121 5 California's new direction in reading and thinking development. Francie Alexander, California State Dept. of Education 121 6 122.0 122 1 122 2 The Texas Education Agency Project: Identifying the most effective reading and thinking strategies in the state. Victoria Bergin, Texas Education Agency Literacy learning and instruction in the content areas Reading instruction in science at the transitional grades: Perceptions vs. practice. Bonnie Konopak, Nancy Cothern, Ellen Jampole, Mary Margaret Mitchell, Janie Everett, Lennie Holomon, Nar.cy Weems, Rita R. Dean, and Leslie S. Arceneaux, Louisiana State Unitiersity The mismatch between what learners know about a topic In history and what texts assume. Isabel L. Beck and Margaret G. McKeown, University of Pittsburgh 122 3 What color is my chalk: Literacy Instruction in a secondary earth science classroom. 123 1 The politlin of whole language. Panel: Diane Stet, Roger Alan Stewart, Purdue University ns, University of Illinois, Judith Newman, Mount Saint Vincent University, and Susan Church, Halifax County Bedft.rd District School Board 124.0 124 1 Comprehension instructional effects Effects of text-induced mood upon story recall. Grover C Mathewson, Florida International University 124 2 4. The effects of PLAE upon students' test performance and metacognitive awareness. Sherrie Nist and Michele L. Simpson, University of Georgia NRC Program, Austin, Texas, 1989 124.3 447 The effects of prior knowledge and differential levels of processing in notetaking on ronceptual and factual recall of lecture material. Christy A. Horn, Roger H. Bruniz., :Ind Duane F. Shell, University of Nebraska-Lineoln 125.0 125 1 Macrocontexts to fa 'that learning: Creating meaningful contexts for instruction Macrocontexts to facilitate learning: Social interaction and the construction of integrated knowledge. Deborah Wells Rowe, Peabody College, Vanderbilt University 125 2 Macrocontexts to facilitate learning: Guiding principles for cnnizalum implementation. 125 3 Jennifer R.Goodrnan and Kim McLarty, Peabody College, Vanderbilt University Macrocontexta to facilitate learning: Theoretical perspectives. Victoria J. Risko, Peatody College: Vanderbilt University 126 0 Recent developments in second language reading research: Tbe ....a=nonly taught languages 126 I 126 2 126 3 127.0 127 I What L2 readers remember: Is it related to their awareness of text structure? Sally A. Hap:, Duval County (FL) Schools, and Stephen Olejnik, University of Georgia LI and L2 models of reading: Quantitative and quslitative evidence. James Davis, Penn State University Assessing second language reading proficiency. E zabeth Bernhardt, Ohio Star University Story book reading and its effets Development in written language: Oral and wr'c a monologues in emergent storybook reading by bilingual children. Liliana Barro Zecker and Elizabeth Sulzby, University of Michigan 127 2 Storybook reading events and literacy acquisition: An analysis of the questions children asked and the answers parents gave them. David B.Yaden, University of Houston 128 I The effects of concept mapping and sentence combining instructIcn on the writing process. Ely Kozminsky and Daniel Lugasi, Ben-Gurion University 128 2 Teachers in transition: Moving from basal readers to Uterature-bised reading in basal dorni :sited schrwt. ...!'-ta W Combs, University of Nevada-Reno and Maureen S. Siera, 128 3 Cblidren's perceptions e: reading, reading materials and reading instruction. Trevor Northees:ea State Universey F'enry Cairnoy and Margie Lays, Riverina-Murray institute for Higher Education. 128 4 The effects e a self-generated learning strategy on the prose-processing abilities of tiementary stunonts: A qualitative and quantitative analysis. Rosemary Barone Lonberger, Arkansas State Unia -ratty 129 I A a exploratory stud r of third-and fifth-grade readers' comprehension of four types of Tournure idioms. Sasan M.Koloski and Odarka S.Trosky, University of Manitoba I 2r, 2 How do basal readers communicate their theories to tbe thoughtftd, reflective teacher? Peter Afflerbach, Emory University 129 1 Precocious reading achievement: A critical review of the literature and the initiation of a longitudinal study. Aileen Webb Tobin, U S Army Ordnance Center & School, and John J. Pikuiski, University of Delaware 129 4 130 0 130 1 130 2 130 3 Effects of grouping and difficulty of materials on reading achievement. Michael Karml, Olno State University, and W.Chnstine Rauscher, Naperville Illinois Schools. Critical Issues in literary education Critical issues in schooling of children at risk. Trika Smith Burke. New York University. Richard Valencia, University of Texas-Austin, John Baugh, University of Texas-Austin, Richard Allington, SUNY- Albany, and Robert Calfee, Stanford University Critical issues on assessment. Peter Johnston, SUNY- Albany, Karen Wixson, University of Michigan, and Sharon O'Neal, Texas Education Agency Critical issues in *hole language. Judith Newman, Mt St. Vincent University, Francie Alexander. California State Department : Education, and Ann McCallum, Fairfax County Schools 4 41 :. -." 13E0 131:1-- 1311 131.3 132.0 132.1 132.2 "Resekrik"perspectives on tmcker devekiinsent Oics:vieW. Jeny Niles, Virginia Tech,. Sandra "Hollingsworth, Mil/ashy. of California Beikeiy Wii;re we've-bees in remora oss trischer e's,hSe#i*:000:14L Al vain,*,..itniveranY. ., . A..; i., 01,641 fir 1e0c*.7 eal!olitY.a0:11-'*.iiiiiiii.:11164-4051,#171igs1iy,'Olai.Afiz,9P, wifilil*Y411,4*Erfl.i*,! 0**,44 big :11014 ikkfeNoes:- 'OvaileiV.'/C.Ssen Featierti; , , East. Triai - Siairbilyeriity! . Fivi iiiiiis moilo ii"ii70,, i:Istio &ipso:1444d panelvIeny Niles, VisgiltiaTech, portal!: Ogle, National College of Education, and P. David Pearson, University of IllinOis 133.0 -. 4' Teacher asiesansent workshop 442 ,..,,