Teachers of English to Speakers of Other Languages, Inc.

(TESOL)

Effects of Prereading Activities on EFL Reading by Brazilian College Students

Author(s): Loni K. Taglieber, Linda L. Johnson and Donald B. Yarbrough
Source: TESOL Quarterly, Vol. 22, No. 3 (Sep., 1988), pp. 455-472
Published by: Teachers of English to Speakers of Other Languages, Inc. (TESOL)
Stable URL: http://www.jstor.org/stable/3587289 .
Accessed: 20/06/2014 21:54

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

Teachers of English to Speakers of Other Languages, Inc. (TESOL) is collaborating with JSTOR to digitize,
preserve and extend access to TESOL Quarterly.

http://www.jstor.org

This content downloaded from 62.122.73.86 on Fri, 20 Jun 2014 21:54:17 PM

All use subject to JSTOR Terms and Conditions
 TESOL QUARTERLY,Vol.22, No.3, September1988

ofPrereading
Effects on
Activities
EFLReadingbyBrazilianCollegeStudents
LONI K. TAGLIEBER
Federalde SantaCatarina,Brazil
Universidade
LINDA L. JOHNSON and DONALD B. YARBROUGH
ofIowa
University

This studyinvestigated theeffectsof threeprereading activities

(pictorialcontext,vocabularypreteaching, and prequestioning)
and a controlconditionon the readingcomprehension of 40
undergraduate BrazilianEFL students. In a Latinsquaredesign,
all subjectsread fourdifferent readingpassages,each passage
underone of the fourconditions.Immediately afterreadinga
passage,subjectsansweredan 8-itemopen-endedtestand a 10-
itemmultiple-choice analysisofvariancetestson
test.Multivariate
thetwo measuresrevealedsignificant effectsforprereading and
passage. Further through
investigation Tukey's HSD revealedthat
all three prereadingactivitiesproduced significantly higher
multiple-choicescores than the controlcondition.Vocabulary
preteaching resultedin increasedcomprehension comparedwith
thecontrolbutwas significantly lesseffective thantheothertwo
Resultsof thestudyare interpreted
strategies. througha schema-
theoreticviewofthereadingprocess.

Studentsof Englishas a foreignlanguage experienceconsiderable

in comprehending
difficulty Englishtextswhenreadingthemfor
the firsttime. Not only do these textsusually contain unfamiliar
vocabulary,but theyalso may containunfamiliar
conceptsand
culturalallusionsthatmake comprehensiondifficult.In addition to
elements that make comprehendinga text in the native language
difficult,thereare sourcesof difficulty
thatare unique to readingin
a foreignlanguage.
Traditionally,attemptsto improvethecomprehensionof textsfor
EFL studentshave focusedon familiarizing
the studentwiththe
vocabulary needed to comprehend the passage (Bernhardt,1984).
However, such instruction
is unlikelyto raise thestudents'interestin

455

This content downloaded from 62.122.73.86 on Fri, 20 Jun 2014 21:54:17 PM

All use subject to JSTOR Terms and Conditions
reading the text or to prepare them for the culturallyand
conceptually novelelementsofthetext.
Studies of prereading activities for native speakers have
demonstrated the facilitative effectsof activatingreaders'prior
knowledgerelevantto understanding thenew text(Mayer,1984).
Not onlydo prereadingactivitiespreparenativespeakersforthe
conceptsthatfollow,but by makingthe readingtaskeasier and
connecting thenewcontentmoremeaningfully to priorknowledge,
prereading activitiesmakereadinga moreenjoyabletask(Hansen,
1981).
Schema theory(Rumelhart& Ortony,1977) may help explain
why prereading activities improve reading comprehension.
According to this theory, meaning is constructed through
interactionbetween the reader's schemata (i.e., knowledge
structuresin memory)and thetext.Schemataare activatedduring
readingand meaningfully relatedto theknowledgeto be learned
fromthe text(Rumelhart, 1981). If the readerlacks appropriate
schemata or fails to activate them, comprehensionmay be
impaired.Prereading areintendedto activateappropriate
activities
knowledge structures or provideknowledgethatthereaderlacks.
EFL readers'comprehension also depends on theirschemata
(Carrell& Eisterhold, 1983).If these readersare facedwithhighly
unfamiliarcontent,particularlymaterialswith many culturally
loaded concepts,comprehension ifnotimpossible,
willbe difficult,
because the readers lack appropriatebackgroundknowledge
(Hudson, 1982; P. Johnson,1981,1982; Steffensen, Joag-dev,&
Anderson, 1979).
Threeprereading techniquesthatseemedmostpracticalforEFL
learnerswereexaminedinthepresentstudy.Two ofthetechniques,
pictorialcontextand vocabularypreteaching, have been shownto
improvecomprehension of native speakers Englishundersome
of
circumstances (e.g., Hudson,1982).The third,prequestioning, has
earned some researchsupportfor both native and nonnative
speakers(e.g.,Royer,Bates,& Konold,1984).
A largebodyofresearchsupportstheeffectiveness ofpresenting
picturesto aid L1 students' readingcomprehension (Bransford&
M. D. Johnson, 1972; Dean & Enemoh, 1983;Dean & Kulhavy, 1981;
Rasco, Tennyson,& Boutwell,1975; Schwartz& Kulhavy,1981;
Sherman,1976).Two studieshavedealtwiththeeffects ofpictures
on L2 readers'comprehension of texts.Hudson (1982) foundthat
the techniqueof displaying,discussing,and writingpredictions
about pictures was significantly more effectivefor reading
comprehension than a
presenting vocabulary/written prediction
activitybefore studentsread. However, the activitywas not more

456 TESOL QUARTERLY

This content downloaded from 62.122.73.86 on Fri, 20 Jun 2014 21:54:17 PM

All use subject to JSTOR Terms and Conditions
effective overallthana taskthatinvolvedreadingthetext,takinga
test,and thenrereadingthetextand retakingthetest(thissecond
testscorewas includedin theanalysis).Omaggio(1979)foundthat
of thepictureshe presentedbeforereading,onlytheonerelatedto
thegeneralthemeofthepassageimprovedcomprehension.
The value of preteachingvocabularyto nativespeakersas a
methodto improvecomprehension is a good deal morecompli-
catedandhasonlyrecently beguntoreceivesupportintheresearch
literature (Kameenui,Carnine,& Freschi,1982;Stahl,1983).So far,
preteaching vocabularyhas been effective whenresearchers have
employedcontrived paragraphs witha highincidenceofrarewords
and littlecontextualinformation or when theyhave pretaught
relevantvocabularyoverextendedperiodsof time.The consensus
of theorists ofreadingcomprehension, however,is thatvocabulary
playsan essentialrole in Ll readingcomprehension (Anderson&
Freebody,1979;Davis, 1968).
In L2 instruction, what is taughtis not usuallythe vocabulary
importantto the story,but rathervocabularyimportant to the
lesson. These words may have limited value in facilitating
comprehension ofthestory.Atpresent, twostudieshavedealtwith
preteaching relevantvocabularyto nonnativespeakersof English
(Hudson, 1982; P. Johnson,1982), and neitherdemonstrated an
effecton comprehension ofsubsequentreadingpassages.
The thirdprereadingactivityinvestigated in the presentstudy
was prequestioning. Similarto thepreposedquestionsadvocated
by Singer (1978) for teaching active reading to children,
prequestioning is taughtby havingthe instructor ask questions
about a passage and the studentsanswer with new questions.
Throughprequestioning, studentssetpurposesforreadingand ask
questionswhose answersrequire understanding the text. The
methodis supportedbyresearchinbothL1 reading(Hansen,1981;
Singer& Donlan, 1982) and in foreignlanguagereading(Royer
et al., 1984).
These threeprereadingmethodsare intendedto help EFL
studentsovercomethreemajorproblemsthatinterfere withtheir
comprehension: (a) lack of vocabularyknowledge,(b) difficultyin
using language cues to meaning,and (c) lack of conceptual
knowledge (Steffensenet al., 1979; Yorio, 1972). Limited
vocabularyknowledgeoccasionallyresultsin readersstoppingto
identifyunfamiliarwords and disruptscomprehension because
readersmay forgettheearlierpartof thesentence(Seliger,1972;
Yorio,1972). This is also a problemwithL1 readers,particularly
thosewithdecodingproblems(LaBerge& Samuels,1974;Samuels,
1977).Limitedvocabularyknowledgealso disruptscomprehension
EFFECTS OF PREREADING ACTIVITIES 457

This content downloaded from 62.122.73.86 on Fri, 20 Jun 2014 21:54:17 PM

All use subject to JSTOR Terms and Conditions
whenthemeaningof a sentenceor paragraphturnson knowledge
of certainwords.Preteaching vocabularymayhelp addressthese
problems.
The secondprobleminvolveswhatYorio(1972) calls the"triple
process" of recallingsyntacticcues, making associations,and
predicting futurecues, a process that Li readers perform
automatically (Goodman,1970).EFL learners, due to interference
fromunfamiliar wordsand grammatical findit difficult
patterns,
simultaneously to rememberearliertextualinformation, predict
whatis coming,and connectphraseand sentencemeanings.The
prereading activitiesofprovidingpicturesand prequestioning help
studentsmake predictionsabout contentthatmay assistthemin
resolving meaningproblemswhenthesyntaxinterferes.
The thirdmajorproblem,lack of conceptualknowledge,is also
common to L1 readers. According to Anderson,Reynolds,
Schallert, and Goetz (1977),themessageof a textcan be distorted
if there is insufficient correspondencebetween the schemata
containedin the text and the schemataby which the reader
assimilatesthetext.Such distortion can happenparticularly when
readersdo notsharethewriter'sculture,beliefs, and assumptions.
Thisproblemcan seriously interferewithEFL readers'comprehen-
sion.It is well exemplifiedin a studyby Steffensenet al. (1979),in
whichcollegestudents fromIndiaandtheUnitedStatesreadletters
describingIndian and Americanweddings.Subjectsread more
rapidlyand recalledmoreinformation whentheyread thepassage
abouttheirown culturethanwhentheyread thepassageaboutthe
foreignculture.Furthermore, theyproducedmoreculturally based
distortionsof the foreignpassage. The prereadingactivityof
pictures, inparticular,mayhelpreducesomeof thesedistortions.

METHOD
The presentstudywas undertaken becauseoftheevidenceofthe
effectiveness ofprereading inL1 readingand thescarcity
activities
of researchin thisarea in EFL reading.The purposeof thestudy
was to determinewhetherthe readingcomprehension of EFL
studentswould be improvedwhentheydid different prereading
activities(pictorialcontext,vocabularypreteaching, and preques-
tioning).Four readingpassagesof between500 and 1,650words
were read by BrazilianEFL students, who thentookopen-ended
and multiple-choice tests.Based on theresultsof previousstudies
with native speakersof English,it was expectedthat subjects'
comprehensionscores would be higherwhen the readingwas
458 TESOL QUARTERLY

This content downloaded from 62.122.73.86 on Fri, 20 Jun 2014 21:54:17 PM

All use subject to JSTOR Terms and Conditions
preceded by any of the threeprereadingactivitiesthan when
readingwas notprecededby a prereading
activity.

Subjects
The subjectswere40 sixth-semester EFL students at a collegein
thesouthern partof the stateof SantaCatarina, Brazil.Theywere
selectedrandomlyfroma groupofstudents whohad scoredin the
midrange(30-73)ofcorrectitemsona 105-item Englishproficiency
testthatcoveredlistening comprehension, vocabulary,grammar,
andreadingcomprehension. The testwas a shortenedversionofthe
Michigan Test of English Language Proficiency(Division of
and
Testing Certification, 1963),usedroutinely by thefaculty at the
The
university. resultsfor studentswho scored in themidrangeon
theshortened version,althoughnotcomparablewiththenormsof
thefull-length
test,suggestedaverageabilityforthesestudents.

Materials
Materialsconsistedof four modern English passages, each
representing a different genre,whichwere takenfromillustrated
collectionsof readingsused by Brazilianstudentsof English.A
groupof fiveEnglish-language professorsat Universidade Federal
de Santa Catarinaexamineda pool of fiveexamplesfromeach
genre(fairytale,legend,nonfiction article,and fable)and madeits
selectionon thebasisofthefollowing criteria:(a) length(500-1,600
words),(b) number (at leastthree)and appropriateness (nonambi-
guity)of illustrations,
(c) appropriateconceptualand vocabulary
and (d) use of contemporary
difficulty, English.
The lengthcriterion reflectedthe need forpassagesthatwere
bothrepresentative of thoseusuallyread by studentsand nottoo
long for thetime constraints of theexperiment. The fairytalewas
the longest(1,646words),whereasthe otherthreetypesranged
from500-750words.It was necessaryto use two fablesto have a
combinedlengthofslightly morethan500 words.
For each passage three pictures were selected from the
illustrations
accompanying theoriginalpassageand weremadeinto
slides.One ofthethreepictureswas generalinnature, and theother
two were more specific.Earlierstudieswithnativespeakersof
English(Brody& Legenza,1981;Dean & Enemoh,1983) suggest
thatglobal picturesare moreeffectivethanpicturesthatpresent
specificscenesfroma text.
The finalconsiderationswere thatthe passages be appropriately
difficultfor the subjects and be in contemporaryEnglish. These

EFFECTS OF PREREADING ACTIVITIES 459

This content downloaded from 62.122.73.86 on Fri, 20 Jun 2014 21:54:17 PM

All use subject to JSTOR Terms and Conditions
judgmentswere made by the five English-language professors,
based on theirexperiencewithsimilartextsreadby similarstudents
and on theirknowledgeofcontemporary English.
The final selectionswere a fairytale, "The Laziest Man in
America"(Stuchl,1979, pp. 169-174);a legend,"Paul Bunyan"
(Reader'sDigestAssociation, 1978,p. 356); an article,"Thanksgiv-
ing Day" (Willson, 1982,pp. 153-154);and two shortfables,"The
Ducks and theFox" and "The Cat and His Visions"(Lobel, 1983,
pp. 5, 19). They differedon a numberof dimensions, including
length,priorcontentfamiliarity,
vocabulary, and writing style.One
of the selections(the fables) included considerabledialogue,
whereastheothersweremoreexpository.
The purposeof thisheterogeneity was to ensurethatanyeffects
found for prereadingtechniqueswould not be the resultof a
However,the intentwas not to
particularpassage characteristic.
investigate differences and in any case,
in passage characteristics,
thedesign,withitsintentional confoundings of order,length,type,
wouldnotbe able to illuminate
and othercharacteristics, anysuch
differences.

Instrumentation
For each passage,8 open-endedand 10 multiple-choice English-
language questions were written. Both tests included items
and
addressingtext-explicit text-implicit information.
(inferential)
Care was taken to avoid includingwords or conceptsdirectly
discussedin the prereadingactivitiesto preventsubjectsfrom
answeringquestionsabove chance level withouthavingread the
passage.The 18 questionsforeach passagewerepilottestedwitha
group of 8 subjects from another Brazilian university.The
directionsand testquestionswere testedand revisedfollowinga
pilotstudywith21 students at thesameinstitution.
The eightopen-endedquestionswere scoredby two teachers,
whose scores were combined to produce the finalscore. The
interraterreliability(intraclasscorrelationadjusted for pooled
ratings)was estimatedto be .85. Internalconsistency reliability
estimates(coefficient alpha) rangedfrom.59 to .69 forthe items
specificto each passage.
The 10 multiple-choice items specificto each passage were
summedseparatelyto yield a totalscore foreach passage. The
internalconsistencyestimates fortwoofthepassageswereverylow
(.19 for"Thanksgiving Day" and .36 forthefables),indicating that
items were not consistentlymeasuring the same thing. Internal

460 TESOL QUARTERLY

This content downloaded from 62.122.73.86 on Fri, 20 Jun 2014 21:54:17 PM

All use subject to JSTOR Terms and Conditions
consistency estimates
fortheothertwosubtestswerean acceptable
.50 and .60.
Finally,in additionto the tests,a 25-itemparticipant
question-
nairewas developedto obtainotherinformation to helpinterpret
thedata.

Procedures
The designwas a 4 x 4 Latin square,such thatsubjectswere
randomly assignedto groupsof10,witheachgroupreadingall four
passages and receiving all four treatmentsbut in different
combinations. Passageand orderof readingwereconfounded, but
treatment was not.Thus,all subjectsservedas theirown controls;
that is, treatmentand passage were within-subjects variables.
Treatmentconsistedof threedifferent prereading activities:
(a)
pictorialcontext,(b) vocabularypreteaching, and (c) prequestion-
ing. There was also a controlcondition,in whichtherewas no
prereading activity.
The fourgroupsof subjectswere randomlyassignedto four
different rooms.Next,each groupwas randomlyassignedto a
different conditionforeach readingpassage.For example,on the
first
day,when"The LaziestManinAmerica"was read,10 subjects
in one room read the passage precededby the pictorialcontext
prereadingactivity;in thenextroomvocabularypreteaching was
done; and so on. For each subsequentpassage,thefourtreatment
conditions wererotatedamongthefourgroupsso thateach group
receivedeach treatment. For each ofthefourexperimental groups,
a helpingteacherwas randomlyassignedto supervisesubjects'
whilethefirst
activities, authormovedfromone grouptoanotherto
carry out the prereadingactivities.On the firstday of the 2-day
experiment, subjectsread and tooktheshort-answer and multiple-
choicetestson "The LaziestMan in America"and "Paul Bunyan."
On the followingday, they read and completedthe tests on
"Thanksgiving Day" and thefablespassage.
In the pictorialcontextcondition,the firstauthorshowed the
subjectsthreeslidesofpicturesrelatedto thecontentofthereading
passage. Whilesubjectslooked at the pictures,the experimenter
encourageddiscussionby havingthemdescribethecontentof the
pictures,trytoconnectthethreepictures, and,finally,makea guess
aboutthecontentofthereadingpassage.
In the vocabularypreteachingcondition,subjectswere taught
eightwordsfromthereadingpassage.The wordswereselectedby
five Englishinstructors at the UniversidadeFederal de Santa

EFFECTS OF PREREADING ACTIVITIES 461

This content downloaded from 62.122.73.86 on Fri, 20 Jun 2014 21:54:17 PM

All use subject to JSTOR Terms and Conditions
Catarinaon thebasisof (a) theirimportanceto understanding the
passage and (b) the likelihood that they would not alreadybe
knownto thestudents. The experimenter presented these wordson
the board in meaningful, but unrelatedsentences.Studentstook
turnsreadingthe sentencesand predictingthe meaningsof the
words.Whena wordwas notadequatelydefined,it was defined
through classdiscussion.
Prequestioning consistedof givingsubjectsa one-sentence oral
summary of thereadingpassage and asking them to formulatesome
questionsthattheythought thepassagemightanswer.For the"Paul
Bunyan"passage, for example,the firstauthorsaid, "This text
containsa storyabout a lumberjackcalled Paul Bunyan.What
questionsdo youhavethatyouthinkthestorymightanswer?Make
a listofyourquestionson paperinEnglishorinPortuguese." These
questions were subsequently written on the board, where they
remainedwhilethestudents read thepassage.
This technique intentionallycombines prequestioningwith
receivinga veryshort(i.e., one-sentence-long) summary,which
thendirectsstudents'self-generated prequestioning. The rationale
forgivingthe one-sentence summarywiththe prequestioning is
twofold.First,the summaryguides the prequestioning activity,
therebyfocusingthe students'prequestionson the passage to be
read. Withoutthe introduction of the topic of the passage, the
prequestioning activitycould remain vague and abstract.Second,
thereis no theoretical orpracticalreasonfornotpresenting theone-
sentence summary.It takes very little class time or prior
preparation.Moreover,it was not the purposeof thisstudyto
investigate theteachingofa generalprequestioning technique.
Each of the prereadingactivitieslasted 10 minutesand was
carriedoutimmediately beforethereadingtask.Whenparticipants
in thecontrolconditionreceiveda passage,theybeganthereading
taskimmediately afterreceivingtheirinstructions. All prereading
activitiesweredirectedbythefirst author.Becauseparticipants had
to wait occasionallyfortheresearcher to conducttheprereading
activities,
participants weregivenpassagesnotrelatedto thestudy
to read while they waited. Time limitsforreadingthe passages
rangedfrom8 minutesforthe two fables(takentogether)to 20
minutesfor the fairytale. Studentswere given 20 minutesto
completeeach open-endedtestand 5 minutesto completeeach
multiple-choice test.
A multivariate analysisof variance(MANOVA) withrepeated
measureswas used to examinethetwo variablesof passage and

462 TESOL QUARTERLY

This content downloaded from 62.122.73.86 on Fri, 20 Jun 2014 21:54:17 PM

All use subject to JSTOR Terms and Conditions
 Scoresof themultiple-choice
prereadingactivity. and open-ended
testswerethedependentmeasure.

RESULTS
Thehypotheses testedwerethatsubjectswouldattaindifferent
comprehension scores with passages precededby different
prereading andthatall prereading
activities activities
wouldresult
in highercomprehension scoresthanthecontrolcondition.
The
comprehension measures werethecombined and
multiple-choice
the combinedopen-endeditem scoresof the subjects.The
multivariate Wilks's
criterion, lambdawiththealphafortheF setat
.05, was used to test ResultsfromtheMANOVA
significance.
indicateda significant
effect forbothpassagetypeandprereading
(seeTable1).
activity

TABLE1
SummaryofMANOVA ofTestsof
andRecall
Comprehension
Reading

Source df F

Betweensubjects
Groups 6, 70 0.92
Within
subjects
Passage 6, 214 11.68*
Prereading 6, 214 10.60*
Passagex prereading 12,214 1.16

"p < .05.

Univariateanalysesof varianceand Tukey'sHSD test(citedin

Kirk,1968) were used to determinethattherewas a significant
passageeffectonboththeopen-endedquestions, F (3, 108)= 20.48,
p< .05 (see Table 2), and multiple-choice
questions,F (3, 108) =
7.98,p < .05 (see Table 3). The averageopen-endeditemscoreon
the"PaulBunyan"passage(M = 15.50)was significantly lowerthan
for "Thanksgiving Day" (M = 31.75), for "The Laziest Man in
America"(M = 35.12),and forthetwo fables(M = 38.25). The
averagemultiple-choice itemscoresforthefourpassageswerein
the same rankorder,but Tukey'sHSD testsuggestedsignificant
differences onlybetweentheaveragescoresfor"PaulBunyan"(M
= 4.68) and the fables (M = 5.98).

EFFECTS OF PREREADING ACTIVITIES 463

This content downloaded from 62.122.73.86 on Fri, 20 Jun 2014 21:54:17 PM

All use subject to JSTOR Terms and Conditions
 There was also a significant prereadingtreatment effecton
subjects'comprehension scores,F (6, 214) = 10.60,p < .05 (see
Table 1). Subsequentunivariateanalysesindicateda significant
effectonlyon themultiple-choice itemscores,F (3, 108) = 23.62,
p < .05 (see Table 3), but not on the open-endeditemscores,
F (3, 108) = 0.78,n.s.Furtherinvestigationof thetreatment effects
on the multiple-choice scores,usingTukey'sHSD test,revealed
significant differences betweenthe controlcondition(M = 4.08)
and each of thethreeprereading activities:
vocabularypreteaching
=
(M 4.90),pictorial context =
(M 6.05),and prequestioning (M =
6.35). There were also significant between
differences vocabulary
preteaching and prequestioning and betweenvocabularypreteach-
ingand pictorialcontext(see Table 4).

DISCUSSION
Theoristsin L1 readinghave suggestedthatprovidingstudents
withsome assistancebeforereadingcan help themunderstand a
text.Some researchwithL1 readerssupports the effectivenessof
prereading activitieson readingcomprehension.
The findingsof the presentstudy indicate that prereading
EFL students'
activitiesalso facilitate comprehension. Of thethree
prereading activitiesexamined in the study, vocabulary preteach-
ing,althoughsuperior to the control condition, was less effective
thanprequestioning and pictorialcontext.It couldbe thatalthough
knowledgeof thesewords' meaningswas essentialforadequate
comprehension to occur,heightened backgroundknowledgefrom
theothertwoprereading made students
activities moreable to use
contextto arriveat a satisfactory meaning for the passageseven
when all the words were not known. Stanovich (1981), in his
discussion of an interactive-compensatory model of reading
fluency,cites evidence thatreaderscompensateforpoor word
recognition abilityby relyingon contextual information.
The presentstudysupportsHudson's (1982) contentionthat
studentsmay use theirbackgroundknowledgeabout a reading
selectionto overrideproblemstheyare havingwiththelanguage.
Such a strategywas thoughtto be preventedby weak second
languageability(Clarke,1979;Cziko,1978).
A secondreasonthatvocabularypreteaching was less effective
thanthe two otherprereadingactivitiesmay be due to how the
wordswerepresented. Although thewordswereputina contextby
illustrating their meanings in sentences, the sentenceswere not
meaningfully connected.Moreover, there was no attemptto relate

464 TESOL QUARTERLY

This content downloaded from 62.122.73.86 on Fri, 20 Jun 2014 21:54:17 PM

All use subject to JSTOR Terms and Conditions
 TABLE2
SummaryofUnivariate ofVariance
Analyses fortheEffect
of
PassageandPrereading
onOpen-EndedTestScores

Source SS df MS F

Betweensubjects
Groups 1,785.46 3 595.15 0.59
Errorbetween 36,404.37 36 1,011.20
Withinsubjects
Passage 12,301.71 3 4,100.50 20.48*
Prereading 466.72 3 155.58 0.78
Passagex prereading 884.68 6 147.44 0.77
Errorwithin 21,628.12 108 200.26
Op< .05.

TABLE3
SummaryofUnivariate ofVariance
Analyses forthe
Effect
ofPassageandPrereading
on Multiple-Choice
Scores

Source SS df MS F

Betweensubjects
Groups 27.12 3 9.52 1.39
Errorbetween 245.08 36 6.85
Within
subjects
Passage 32.72 3 14.93 7.98*
Prereading 130.07 3 44.18 23.62*
Passagex prereading 18.42 6 3.04 1.62
Errorwithin 202.23 108 1.62

*p < .05.

TABLE4
Tukey'sHSDTestforDifferences
Among
Treatments
Prereading on theMultiple-Choice
Items

Different
from
Treatments M1 M2 M3 M4

M, = 6.35 0.30 1.45" 2.28*

M2 = 6.05 1.15" 1.98"
= 4.90 0.83*
M3
M4 = 4.08

Note: M1 = prequestioning; M2 = pictorial context; M3 = vocabulary preteaching;

M4 = control.
*p < .05.

EFFECTS OF PREREADING ACTIVITIES 465

This content downloaded from 62.122.73.86 on Fri, 20 Jun 2014 21:54:17 PM

All use subject to JSTOR Terms and Conditions
thewordsto thecontentof thereadingselections.In theHudson
(1982) study,vocabularypreteaching was conductedsimilarly but
includeda prediction component.Studentswereactivelyinvolved
in writingpredictionsabout how the words mightrelateto the
selection. Nevertheless,Hudson's vocabulary activitydid not
surpassperformanceof those who did a read-test/reread-retest
activityor a pictorial/prediction activity.
Third, the observationsof the experimentersuggest that
participantsin this study reacted differentlyto vocabulary
preteaching thanto theothertwo prereadingactivities.The latter
two activitiesappeared to produce a deeper and more active
involvement ofthesubjectspriortoreading.Discussionofwordsin
sentencesunrelatedto each otherand unrelatedto theupcoming
readingmaynothavebeen novelorinteresting to thestudents.
Pictorialcontextand prequestioning, the two schema-building
tasksof the presentexperiment, bothdealt witha few concepts
relatedto thereadingpassageand werepresentedina larger,more
unifiedcontext.It mightbe thatprereading activitiesthatintroduce
wordsand conceptsfromthepassagein a moreglobalcontextare
more effectivein evoking or building readers' appropriate
schematafora passage thanprereadingactivitiesthatintroduce
wordsinisolation.Even whenwordsareevocativeofthecontentof
a passage, as in Hudson's (1982) study, and studentsmake
predictions withthem,thereis no guaranteethatthestrategy will
surpass direct knowledge of thetask at hand.
The success of the pictorialcontextconditionof the present
experiment in surpassingperformance of vocabularypreteaching
adds weighttoHudson's(1982)conclusionthata picture-discussion-
prediction conditionproduceshighertestscoresthana vocabulary-
preteaching-with-prediction condition.
That prereadingaffectedmultiple-choice scoresbut not open-
ended scoresis open to severalinterpretations. First,in a previous
study(Taglieber,1981), the first
author found that whenthereading
comprehension of EFL students with low- and average-level
English proficiency is measured only through open-ended
questions, the resultscan be misleading. These students oftenhave
in
difficulty expressing their answers clearlyin the foreign language
even iftheycomprehended whattheyread. Lee (1986)foundthat
subjectsrecalledmoreidea unitsofan L2 passagewhenwriting in
theirnativelanguagethanin thelanguageof thepassage.
Second, this findingmay reflectdifferencesin the type of
knowledgemeasuredin each typeof testand thedegreeto which
prereadingactivitiesaffecteach type of knowledge. It is possible
that prereadingactivitiesfocused more on details than on central

466 TESOL QUARTERLY

This content downloaded from 62.122.73.86 on Fri, 20 Jun 2014 21:54:17 PM

All use subject to JSTOR Terms and Conditions
ideas ina passageand thatforthisreason,subjectsremembered the
detailed informationneeded to answer the multiple-choice
questions.
Finally,a shortageof time for readingthe passages and for
answeringthe open-endedquestionsmighthave produced this
result.Of thesubjects,61%declaredon thequestionnaire thatthe
timeallowed forreadingeach passage was insufficient, and 46%
statedthatthe timeforansweringtheopen-endedquestionswas
insufficient. Futurestudiesshouldprovidemoretimeforreading
and answering open-endedquestions.
It should be noted that althoughthe internalconsistency
estimates
reliability forthemultiple-choice itemsfortwoofthepas-
sagesweredisappointingly low (.19for"Thanksgiving Day" and .36
forthe fables),theydid not seriouslyunderminethe experiment
because of theuse of multiplemeasuresand thepoolingof scores
overpassages.However,theyintroduced additionalerrorintothe
statistical
modeland reducedthepowerof theexperiment to find
treatment differences. The low reliabilities no doubtreflectvery
easyand verydifficult as wellas pooritems,especiallyinthearticle
"Thanksgiving Day." The itemsalsorequiredifferent skillsandmay
wellbe reflecting performance subfactors.
Futurestudiesshouldinclude20 or more itemsso thatother
psychometric approaches(e.g.,factoranalysisand subfactor scores
or itemdeletion)can be broughttobearon theproblemofinternal
consistencyin these measuresof reading comprehension with
nonnative speakers.
The passageeffectwas mainlydue tothe"PaulBunyan"passage,
whichhad thelowestmean score.Lookingat theanswerson the
participant questionnaire, one is temptedto attribute thisresultto
the difficulty of the passage (even thoughpassage is completely
confoundedwithotherfactors).The "Paul Bunyan"passage was
rated as the mostdifficult of the fourselectionsby 76%of the
It is
subjects. possible that thispassagewas inherently moredifficult
thantheotherpassages.A legendoftenmakesimplicit references to
knowledge about of
ways thinking, of and
ways acting, ways oflife
of the people of the particularculturalsettingabout whom it is
written.Thus,certainculturally specificconceptsin theAmerican
legend may have made it difficultfor the subjects,especially
becausethereseemsto be no counterpart to thislegendin Brazilian
folklore.Analternative is that
interpretation thetestquestionswere
simply more difficultfor thispassage.
Because the passages differfrom one anotheron several
confoundeddimensions(e.g., length,vocabulary,content,style,
and readerpriorknowledge,to mentionseveral)and because the

EFFECTS OF PREREADING ACTIVITIES 467

This content downloaded from 62.122.73.86 on Fri, 20 Jun 2014 21:54:17 PM

All use subject to JSTOR Terms and Conditions
independenttestsof comprehension foreach passage cannotbe
equated fordifficulty, littleproductivespeculationcan be offered
aboutthemaineffectforpassage.It is bestviewedas a productof
the methodology.The value of includingsuch heterogeneous
passages is to improvethe externalvalidityof inferences to the
different kinds of passages that studentsmust comprehendin
typicalforeignlanguageinstruction. It is also worthnotingthatthe
prereadingtreatment did notinteract withpassagecomprehension
as measuredon eithertheopen-endedor themultiple-choice test.
Resultsof this studysuggestthatprereadingactivitiesmight
become a usefultool forteachersof EFL to facilitatestudents'
readingcomprehension. Teachersmaywantto use one of thethree
prereadingactivities,depending on students'needs and the
ofthetext,or theymaywantto combineall threein
characteristics
thesameclass.
These prereading activitiesalso mayhave value forindependent
study. Prequestioningin particularcan be taughtto students
wishingto improvetheircomprehension. Moderntextbookson
readingforEFL studentsalreadypresentpicturesand othertypes
ofaids beforetextselections. The findings ofthisstudysupportthe
usefulnessof thisidea.
Prereadingactivitiesare also motivational devices.Accordingto
the questionnaire,the studentsin this experimentsaid they
appreciatedand foundithelpfulto do prereading before
activities
a
reading passage. They suggested that prereading activitiesbe
used more frequentlyin theiractual foreignlanguage classes.
Activitiesof the typeused in thisstudymightnot onlyincrease
students'comprehension of thepassagestheyread,butmightalso
makereadingmoreenjoyableand thusencouragemoreextensive
reading.Additional readingwill,in turn,no doubtresultin a better
grasp of thelanguage.
Futureinvestigations of prereadingactivities mightbe designed
to addressthreeissuesnotwell addressedby thisstudy.First,the
prereading inthisstudywereverymuchconfounded
activities with
the experimenterproviding the activities. The alternative
explanation-that it was participants' responding to their
perceptions of theexperimenter's expectations or to thenoveltyof
the experiment ratherthanto the prereadingtechniques-isnot
plausiblefortheseresultsin lightof thelargepassage effects, the
differences betweenprereadingtechniques,and the participants'
responseson the questionnaire. However,futurestudiesneed to
investigatewhether can
participants implement thesetechniquesin
independentstudy and whether other instructors can use themas
beneficially.
468 TESOL QUARTERLY

This content downloaded from 62.122.73.86 on Fri, 20 Jun 2014 21:54:17 PM

All use subject to JSTOR Terms and Conditions
 Second, futurestudiesneed to address the extentto which
differentkinds of passages benefitfrom differentprereading
techniques.For example,highlyconcretepassages mightbenefit
equallywell frompictorialcontextand prequestioning techniques,
whereasabstractpassagesmightbe betteraddressedby preteach-
ingvocabularyand prequestioning techniques.
Third,each oftheseprereading treatmentscan be viewedas one
exampleof a generalclass of treatments.
For example,fora more
global contextand possiblymoreinteresting activity,vocabulary
preteachingmightinvolvesemanticmapping(D. D. Johnson&
Pearson,1984) or the predict-o-gram (Blackowicz,1986). Future
studies could investigatethe relativeeffectiveness of general
prequestioningtrainingversus passage-specific prequestion
generation.
In theinterim,EFL teachersand students can be well servedby
of
adaptiveuse prereadingtechniques.Teachersand studentscan
try the various techniques and develop their own personal
knowledgeabout what worksand does not workforthemwith
differenttextsand readingpurposes.

THE AUTHORS
LoniK. Taglieberteaches
graduate EFL coursesattheUniversidade Federalde
SantaCatarina,
Brazil.
Sheispresently
organizinga cross-disciplinary lab,
reading
thefirst
of itskindin Brazil,whichwillbe opento bothPortuguese andEFL
andlatertostudents
students, from other
departments.Thisarticle
isbasedonher
doctoral
dissertation of
(University Iowa,1985).
LindaL. Johnson, Assistant
Professor
of Rhetoric
and Education, directsthe
ReadingLab andteachesrhetoric,
reading,andreading educationcourses.
She
spentthreesemestersteaching
readingandstudyskillsinShahAlam,Malaysia,
duringthe1985-1986academicyear.Herresearch
interestisinadvancedreading
andlearninginthedisciplines.
Donald B. Yarbrough,Assistant
Professor
of Educational
Psychology,
taught
as a secondlanguage
English for3 yearsinWestGermany andcurrently
teaches
coursesinthecognitive
graduate oflearning
psychology andthepsychology
of
Hismajorresearch
writing. interests
arethepsychologyofwriting
andlearning
fromtext.

EFFECTS OF PREREADING ACTIVITIES 469

This content downloaded from 62.122.73.86 on Fri, 20 Jun 2014 21:54:17 PM

All use subject to JSTOR Terms and Conditions
REFERENCES
Anderson,R. C., & Freebody,P. (1979). Vocabularyknowledgeand
reading (Reading Education Rep. No. 11). Urbana: Universityof
Illinois,CenterfortheStudyof Reading.(ERIC DocumentReproduc-
tionServiceNo. ED 177470)
Anderson, R. C., Reynolds,R. E., Schallert,D. L., & Goetz,E. T. (1977).
Frameworksfor comprehendingdiscourse.AmericanEducational
ResearchJournal, 14,367-381.
Bernhardt, E. B. (1984).Towardan information processing perspectivein
foreign languagereading.ModernLanguageJournal, 68,322-331.
Blackowicz,C. L. Z. (1986). Makingconnections:Alternatives to the
vocabulary notebook. Journal ofReading,29, 643-649.
Bransford, J. D., & Johnson, M. D. (1972). Contextualprerequisites for
understanding: Some investigations of comprehensionand recall.
Journal of VerbalLearningand VerbalBehavior,11,717-726.
Brody, P. J., & Legenza, A. (1981). Can pictorialattributesserve
mathemagenic EducationalCommunication
functions? and Technology
Journal, 28,25-29.,
Carrell,P. L., & Eisterhold, J.C. (1983).Schematheoryand ESL reading
pedagogy.TESOL Quarterly, 17,553-573.
Clarke,M. A. (1979).ReadinginSpanishand English:Evidencefromadult
ESL students. LanguageLearning, 29,121-150.
Cziko,G. A. (1978).Differences infirst-
and second-language reading:The
use of syntactic, semanticand discourseconstraints. CanadianModern
LanguageReview,34,473-489.
Davis, F. B. (1968). Researchin comprehension in reading.Reading
ResearchQuarterly, 3, 499-545.
Dean, R. S., & Enemoh,P. A. C. (1983). Pictorialorganization in prose
learning.Contemporary EducationalPsychology, 8, 20-27.
Dean, R. S., & Kulhavy,R. W. (1981).Influenceof spatialorganization in
proselearning. Journalof Educational Psychology, 76,57-61.
Divisionof Testingand Certification. (1963). Michigantestof English
language proficiency.Ann Arbor:Universityof Michigan,English
LanguageInstitute.
Goodman,K. S. (1970).Psycholinguistic universalsinthereadingprocess.
In P. Pimsleur& T. Quinn(Eds.), The psychology ofsecondlanguage
learning(pp. 135-142).Cambridge:CambridgeUniversity Press.
Hansen,J. (1981). An inferential comprehension strategyforuse with
primary gradechildren.The ReadingTeacher,34,665-669.
Hudson, (1982).The effectsofinducedschemataon the"shortcircuit"
T.
in L2 reading: Non-decodingfactorsin L2 reading performance.
LanguageLearning, 32,1-31.
Johnson, D. D., & Pearson,P. D. (1984). Teachingreadingvocabulary
(2nded.). New York:Holt,Rinehart and Winston.
Johnson,P. (1981). Effectson reading comprehensionof language
complexity and culturalbackgroundof a text.TESOL Quarterly, 15,
169-181.

470 TESOL QUARTERLY

This content downloaded from 62.122.73.86 on Fri, 20 Jun 2014 21:54:17 PM

All use subject to JSTOR Terms and Conditions
Johnson,P. (1982). Effectson reading comprehensionof building
background knowledge.TESOL Quarterly, 16,503-516.
Kameenui,E. J.,Carnine,D. W., & Freschi,R. (1982). Effectsof text
construction and instructionalproceduresforteachingwordmeanings
on comprehension and recall.ReadingResearchQuarterly, 17,367-388.
Kirk,R. A. (1968). Experimental design:Proceduresforthe behavioral
sciences.Belmont, CA: Wadsworth.
LaBerge,D., & Samuels,S. J. (1974). Toward a theoryof automatic
information processinginreading.CognitivePsychology, 6, 293-323.
Lee, J. F. (1986). On the use of therecalltaskto measureL2 reading
comprehension. StudiesinSecondLanguageAcquisition, 8, 201-211.
Lobel,A. (1983).Fables.New York:Harper& Row.
Mayer,R. E. (1984).Aidsto textcomprehension. EducationalPsycholo-
gist,19,30-42.
Omaggio,A. C. (1979).Picturesand secondlanguagecomprehension: Do
theyhelp?ForeignLanguageAnnals,12,107-116.
Rasco, R. W., Tennyson,R. D., & Boutwell,R. C. (1975). Imagery
instructions and drawingin learningprose. Journalof Educational
Psychology, 67,188-192.
Reader's Digest Association.(1978). Americanfolkloreand legend.
Pleasantville, NY: Author.
Royer,J. A., Bates,J. A., & Konold,C. E. (1984). Learningfromtext:
Methodsof affecting readerintent.In J.C. Alderson& A. H. Urquhart
(Eds.), Readingina foreign language(pp. 65-81).London:Longman.
Rumelhart, D. E. (1981).Schemata:The buildingblocksof cognition. In
J. T. Guthrie and
(Ed.), Comprehension teaching: Research reviews
(pp. 3-26).Newark,DE: International ReadingAssociation.
Rumelhart, D. E., & Ortony, A. (1977).The representation of knowledge
in memory.In R. C. Anderson, R. J.Spiro,& W. E. Montague(Eds.),
Schoolingand theacquisitionof knowledge(pp. 99-135).Hillsdale,NJ:
LawrenceErlbaum.
Samuels,S. J. (1977). Introduction to theoretical
modelsof reading.In
W. Otto, C. W. Peters,& N. Peters(Eds.), Reading problems:A
multidisciplinary perspective(pp. 7-41).Reading,MA:Addison-Wesley.
Schwartz,N. H., & Kulhavy,R. W. (1981).Map features and therecallof
discourse.Contemporary EducationalPsychology, 6, 151-158.
Seliger,H. W. (1972).Improving readinginEnglishas a secondlanguage.
EnglishTeachingJournal, 27,48-55.
Sherman, J.L. (1976).Contextual informationand prosecomprehension.
Journal ofReadingBehavior,8, 369-379.
Singer,H. (1978). Active comprehension: From answeringto asking
The
questions. ReadingTeacher,31, 901-908.
Singer,H., & Donlan,D. (1982).Activecomprehension: Problem-solving
schemawithquestiongeneration forcomprehension of complexshort
stories.ReadingResearchQuarterly, 17,166-186.
Stahl,S. A. (1983).Differential wordknowledgeand readingcomprehen-
sion.Journal ofReadingBehavior,15,33-50.

EFFECTS OF PREREADING ACTIVITIES 471

This content downloaded from 62.122.73.86 on Fri, 20 Jun 2014 21:54:17 PM

All use subject to JSTOR Terms and Conditions
Stanovich,K. E. (1981). Attentional and automaticcontexteffectsin
reading.In A. M. Lesgold& C. A. Perfetti(Eds.), Interactive
processes
inreading(pp. 241-267).Hillsdale,NJ:LawrenceErlbaum.
Steffensen,M. S., Joag-dev,
C., & Anderson, R. C. (1979).A cross-cultural
perspective onreadingcomprehension. ReadingResearchQuarterly, 15,
10-29.
Stuchl,V. (1979). Americanfairytales (V. Gissing,Trans.). London:
OctopusBooks.
Taglieber,L. K. (1981). How to facilitatereadingcomprehension of
Englishtextsto nativespeakersof Portuguese.Unpublishedmaster's
thesis,UniversityofIowa, Iowa City.
Willson,G. F. (1982).Thanksgivingday.In New bookofknowledge(Vol.
18,pp. 153-154).New York:Grolier.
Yorio, C. A. (1972). Some sourcesof readingproblemsfor foreign
languagelearners. LanguageLearning, 21, 107-115.

472 TESOL QUARTERLY

This content downloaded from 62.122.73.86 on Fri, 20 Jun 2014 21:54:17 PM

All use subject to JSTOR Terms and Conditions