Three Standardized Tests: Can a Storytelling Curriculum Measure Up?

advertisement
loumal of Early Chilrlhood Teacher Education, 28:251-275, Z0O7
Copyright O National Association of Early Childhood Teacher Educators
ISSN: 1090-i027 pdnt/ 1745-5642online
DOI: 10.1080/10901020701555564
!) Routledge
Taylor&FrancisGroup
fl \
@
N
-c
s
(.)
o
One Authentic Barly Literacy Practice and
Three StandardizedTests:Can a Storytelling
Curriculum MeasureUp?
c{
j-i
'o
'c
o
tL
o
o
>
d)
c)
=
=
o
o
PATRICIA M. COOPERI,KAREN CAPO2,BERNIE MATI{ES2,
AND LINCOLN GRAY3
lNew York
University, New York, New York, USA
University Center for Education, Houston, Texas, USA
3JamesMadison University, Harrisonburg, Virginia, USA
2Rice
The current study was designed to assessthe vocabulary and literacy skills of young
children who participated in an authentic literacy practice, i.e,, Vivian Paley's
"storytelling curriculum," over the course of their respectiveprekindergarten or kindergarten years, We asked: How do prekindergarten and kindergarten age children,
who participate in the storytelling curriculum over lhe course of the school year, perform on pre- and postmeasuresofAGS/Pearson Assessments'Expressive Vocabulary
Test (EVT| the Peabody Picture Vocabulary Test (PPW) (3'" ed.) Form IIIA, and
Whitehurst's Get Ready to Read!, as compared to those young children in the same
grade with similar backgrounds and in the same or similar school settings who did not
participate in the storytelling curriculum? Resuhs show that in comparison to sameage children in like settings,participants in the storytelling curriculum showed significant gains in both vocabulary knowledgeand literacy skills. Thesefindings underscore
the possibility of supporting both beginning and experiencedteachers in using authen'
tic literacy activilies to prepere childrenfor literacy learning, while maintaining their
service to a wide range of other developmental issues.They also call into question the
prevailing trend to abandon such classroom practices in favor of a skills-centered
approach to curriculum.
Introduction
ln The Child and the Curriculum,Dewey (190211990)writes of the tensionbetweenpsychology and subjectmatter in the early grades:
"Discipline" is the watchword of those who magnify the course of study;
"interest" that of thosewho blazonthe "Child" upon their banner.(p. 188)
In reading education, the tension between a child-centered vision of learning and a subject-matterbasedone has long been representedas authenticvs. skills-basedinstruction.
In actual practice, a skills approachdominated early reading curricula in the first and second grades throughout most of the last century, though the well-known "great debate"
flourished as to exactly what skills should be emphasized(Chall, 1968; Sadoski,2O04;
Stahl, 1998). A "reading readiness"curriculum, including work on shapes,fine motor
I 1 November
2006.
30 October2006;accepted
Received
to PatriciaM. Cooper,I l0 E. 14Street,#1507,New York,NY 10003.
Addresscorrespondence
E-mail:pmc7@nyu.edu
251
252
c!
iir
A
c\|
,g
(g
(L
=
3
o
P. M. Cooper et al.
skills, and conceptslike sequence,dominatedthe kindergartenyear well into the 1970s,
when direct instruction in skills becamemore common. The late 1980s and 1990switnessedthe rise of the "reading wars" betweenproponentsof phonics and whole language
approaches(Stahl). In recent years, the reading wars have given way to a "balanced"
approach(Sadoski),where authenticpracticesare comingledwith skills-basedones."Balanced literacy" is a popular idea with many teachersand teacher educators, who tout its
mix of affective and cognitive learning opportunities. Growing support for it, however,
has not countered several other trends in early literacy to be addressedin this articletrends that are threatening the future of authentic literacy activities in prefirst classrooms.
First is the pressfor accountability.In a shift from decadesof practice,the questionof
what, how, and when children learn to readis no longer the exclusivedomain of universities, teachereducationprograms,or local school districts.Increasingly,it is the privy of
federal agencies,political appointees,and for-profit entities.Under their combinedinfluence, accountabilityin early literacy learning meanssomethingnew in early childhood
of literacy subskills, often in conjunction
education:standardizedtests or assessments
goals. A prime example of an accountthat
target
specific
with state and federal standards
First Initiative (Title I, Part B) of the
is
the
Reading
ability-driven early literacy endeavor
(NCLB),
legislation passedin 2001
national
education
the
No Child Left Behind Acr
Statesand school disL2o2a.html).
(www.ed.gov/legislation/FedRegister/fi
nrulel2002-4/
five
literacy subskills in
then
test
must
teach
and
tricts that receive Reading First funding
fluency,
phonics,
vocabulary,
and comprehenphonemic
awareness,
the primary grades:
"high
associoutcomes
stakes"
funding,
other
(Yell
In
addition
to
& Drasgow,2005).
sion
rankings.
pupil
school
include
retention
and
ated with mandatedtesting
Despitethe widespreadcontroversysurroundingthe suddenrise in testingyoung children, it should be noted that the trend is not limited to children in kindergarten and the primary grades.Meisels (2006) reportsthat the "most extensiveuseof high-stakestestinghas
taken place in Head Start" (p. 2),the well-known interventionprogram for low-income 3and 4-year-olds.
The second concern of early childhood educatorsaround early literacy instruction is
the "push down" of formal reading instruction-in its historical sense of decoding and
encoding-from first grade to kindergarten, from prekindergarten to the nursery classroom. Content usually incorporateshighly formalized instruction in phonemic awareness
and activitiesrelatedto letter recognition,reproduction,and word study.Critics claim that
in order to cover test content, teachersmust make inappropriate demands on young and
very young children, including too much inactivity requiredby direct instructionand premature expectationsaround symbolic and metacognitivethinking. Just as significant is
that time spent on these academicsubskills meansa concomitantdecreasein traditional
curricula around play, story, and movement that spur oral languagedevelopmentand
allow for teacherguidancein language(Dickinson,2002; Neuman & Roskos'2005).
Early childhood educators' third and perhapsgreatestconcern is the move away from
free-rangingoral languageactivities.More than any other, this shift is likely to have prolonged impact on early literacy development,and, by extension,on all later academic
achievement(Bielmiller, 2006; Dickinson & Tabors, 2001; Scarborough,2002; Snow,
Without a strongoral languagebase,
2002; Snow, Burns, & Griffin, 1998;Watson,2OOZ).
young children are lesslikely to meet successin early reading.They are also likely to have
difficulty in making a timely leap from speechto writing, and thus fail to reap the reciprocal benefits of writing on reading proficiency. Clearly, the implications of an anemic oral
languageprogram in prefirst classroomsaffect all young children's experiences.Yet,
research suggests that children who enter first grade with the language habits and
Can a StorytellingCurriculum Measure Up?
@
N
v
A
N
;j
'x
(L
253
dispositionsreflective of middle classusageand child rearing practicescan still perform
adequatelyor better (Snow, Tabors, & Dickinson, 2001). The stakesare far greaterfor
children who dependon very early schoolingto preparethem to meet the languageexpectationsofthe first and secondgradecurriculum (Strickland,2000).
One persistentresponseto the oral languagedisparity betweenmiddle-classchildren
and their peersfrom lower socioeconomicgroupshas been a call for changesin parenting
norms (Barbarin,20O2).This proposalhas obviouspractical,and, to some degree,ethical
limitations.The good news is that early childhoodprogramshave long played a key role in
bridging the gap between home and school. The bad news is that given the very real limitations of time, resources,and district demands,many early childhood educatorsmust
decide,if they are given the chanceat all, betweenthe long-term advantagesof a holistic
oral language approach to early literacy and a curriculum that emphasizestest content. A
Hobbesianchoice?Or a false dichotomy?
The Problem
>
o
c
3
Influential curriculum guides for early childhood curriculum, including Learning to Read
and Write: Developmentally Appropriate Practicesfor Young Children (Neuman, Copple,
& Bredekamp,L999),describelanguage-rich,authenticactivitiesin the serviceof literacy
development,such as restaurantor firehouse"centers,"and writing activities like "postman" or snacksign-up.While theseactivitiesmeetthe usualcriteria for authenticlearning
experiences,the early childhood community has resistedassessingtheir educationalefficalled for in today's
cacy through the type of standardizedtesting or formal assessments
generally
assumedthat,
(Meisels,
is
it
2006).
For
one
thing,
climate of accountability
is
not
easily tested.
activities
impact
of
these
counterparts,
the
skills-based
unlike their
in
today's eduauthentic
curricula
measure
of
efficacy,
without
some
standard
However,
prevalent
that
they fail to
charge
vulnerable
to
the
increasingly
left
cational climate are
press
for
accountability
the
Does
this
mean
that
young
literacy.
children's
early
advance
spells the veritable end of authenticliteracy practicesin the early childhood classroom?
As noted, this presentsa theoretical and professional conflict around best practice that is
relevant to all young children (and to those who would prepare their teachers). But, a
skills-dominantcurriculum, lacking in rich and free-rangingoral languageopportunities
and play with text, poses more serious harm to young children already on the margins of
school preparednessby virtue of their inexperience with middle-class language experiences.These include children from low-income families, those whose home languageis
not StandardEnglish, and thosewhosefamilies are disengagedfrom school expectations.
This dilemma leads us to the following questionsand the focus of our study: What
happensif we take a second look at standardizedtesting and authentic literacy experiences?What might we learn from standardizedmeasurementsabout the role play-based,
language-richcurricula play in early literacy development?In other words, can the case
be made that various authentic literacy activities do not "measure up" before they are
investigated?
The Study
This article reportson a study of the impact of an acclaimedauthenticearly literacy activity, Vivian Paley's (1981) "storytellingcurriculum," on prekindergartenand kindergarten
children's early languageand literacy developmentover the course of the school year. Participantsin the study included,but were not limited to, low-incomeand English-language
254
@
N
rir
A
N
:0
n
!j
;
E
3
o
P. M. Cooper et al.
learners,who are typically marginalized by oral languageand text-related expectationsof
the standardearly literacy curriculum. After conducting a review of the relevant research,
we accepted Meisel's (2006) claim that "accountability calls for information about
whetheror not somethinghappened"(p, 17).Given the high ratio of teacher-childinteractions around language and text inherent in Paley's storytelling curriculum (described
below) analogousto that found in middle-classhomes,we hypothesizedthatregular participation might have a positive impact on participants' vocabulary and knowledge of
beginning skills-two essentialcomponentsof an early literacy skill set. We looked to
well-known and commonly used assessments,
designedto measurethe degreeto which
individual children possesstheseparticular attributes(Gullo, 2006). We asked:How do
prekindergartenand kindergartenagechildren, who participatein the storytellingcurriculum over the course of the school year, perform on pre- and postmeasuresof the Expressive Vocabulary Test (EVT), the Peabody Picture Vocabulary Zesr (PPVT), and
Whitehurst's Get Ready to Read! (GRTR!) as comparedto those young children in the
samegrade with similar backgroundsand in the sameor similar school settingswho did
not participatein the storytellingcurriculum?Mindful of the many negativeaspectsassociated with testing of young children, we sought to eliminate or reduce these by, first,
allowing the children time to be comfortablewith the researchersthrough repeatedvisits
to the classroomprior to assessment,
and, second,allowing the children to decideagainst
pafiicipation in both pre- and posttesting.
Significance
The significanceof this study lies in its mergerof severalconcernsfacing the early childhood community today: the push for accountabilityregardingyoung children's early literacy learning;the push down of academicexpectationsinto the prefirst grades;the push in
of culturally biased expectationsaround oral language and literacy experiences;and,
finally, the push out of curriculum that focuseson broader developmentalissues.The
"usesof storytellingin the classroom"(Paley, 1990)embraceinclusive,play-based,holistic relianceon story in the serviceof psychosocialdevelopment,which has beenlinked to
literacy and narrative development (Cooper, 2OO5; Mclane & McNamee, 1990;
Nicolopoulou, 1996, 1997).The storytellingcurriculum's impact on young children'soral
languageand literacy subskill developmenthas not been assessedpreviously, however.
An underlying question for this article is whether it "measuresup" as a prototype for
authenticliteracy activities that serveboth developmentand academicendsfor all children.
Oral Language Development
The body ofliterature on young children's oral languagedevelopmentis large and beyond
the scopeof this article. We have, therefore,limited our review to the characteristicsof
oral languagedevelopmentat school entry that is consideredrelevantto short- and longterm literacy learning. A separatesection is provided on the storytelling curriculum.
A salient thread throughout the researchis that young children's early literacy success,and thus their long-term profiles, is most predicatedon their oral languagecharacteristics at school entry when they are between5 and 6 yearsold. More notably, successful
characteristicsare highly correlatedwith middle-class(especiallyWhite and professional)
usage, as well as competenciesderived from middle-class child rearing practices around
languagein the first 5 years of life (Snow et al., 2001). Vocabulary developmentis one
example. Hart and Ristley (1995, 2003) find that by their third birthday middle-class
Can a Storytelling Curriculum Measure Up?
N
-c
I
s
-i
N
;J
:p
(L
c)
o
O
;
0)
(o
E
3
o
255
children "have heard 30 million more words than underprivilegedchildren." This fact
alone advancesmiddle-classchildren's receptiveand expressivevocabularies,linguistic
ability, reflectivereasoning,abstractthinking, and overall backgroundknowledgein ways
privilegedby schoolexpectations.
Similarly,Snow and Dickinson (1991) write that middleclasschildren are also habituatedfrom infancy into the understandingand use of decontextualizedlanguage.The Early Childhood Learning s/rd), (ECLS) (U.S.Departmenrof
Education, 1998) reports that young White children-significantly overrepresented
among the middle class-regularly outperform their African American peers--disproportionately underrepresented-onmeasuresof early reading subskills,including beginning
sounds,ending sounds,and letter recognition.
Middle-class parents also advantagetheir prefirst children by directing their attention
to text and print in ways that cultivate comprehensionskills. For example, story-related
talk often includes such observationsas, "The third little pig was the smart one, wasn't
he?" Understandingthe reciprocity between print and talk is assistedthrough regular
informal experiences(such as respondingto environmentalprint, learning letters, and
drawing) that lead to writing. Finally, Heath's(1983) seminalstudy of parent-child talk in
different racial and socioeconomiccommunitiesfinds that middle-classparentsfurther
advantagetheir preschoolchildren educationallythrougha style of verbal communication
that prefiguresnot only the teacher'sforthcoming question-and-answer
expectations,but
orientstheir children towardsthe overall demandsof texts as relatedto school tasks.
Early childhood advocacyorganizations,in particularthe National Associationfor the
Educationof Young Children (NAEYC), have long emphasizedauthentic,or "naturalistic" (Gronlund,2006), oral languagepracticesbetweenteachersand children. NAEYC's
highly influential program of Developmentally Appropriate Practice (Bredekamp &
Copple, 1998)recommendsthe following:
Teachersencouragechildren's developinglanguageand communicationskills
by talking with them throughout the day, speakingclearly and listening to
their responses,and providing opportunities for them to talk to each other.
(p.127)
Further recommendationsinclude engagingindividual children and groups in conversation about real experiences.Watson (2002) writes that certain forms of talk, such as metalanguage and abstraction, are more relevant in literate cultures as a result of literate
influences,and that children's proficiency in them leadsto a bidirectional learningcurve
wherein the more competent they are in and about the language of texts, the more texts
they can read, and competencyincreases.when applied to early schooling, CochranSmith (1984) and othersreport that "read-alouds"are instrumentalin this process.Parallel
to and woven throughout languageexperience,Bredekamp and Copple also advocate
embeddingprint and writing opportunitiesin authenticactivities,including:
learning particular names and letter-soundscombinations and recognizing
words that are meaningful to them (such as their names,names of friends,
phraseslike "I love you," and commonly seen functional words like exlt).
(p.l3l)
The historical bias towards autheltic literacy activities in prefirst early childhood
education reverberatedin the whole languagemovement of the 1980s and 1990s (see
Goodman,1986; Goodman& Goodman,1979).The differenceis that. whereasthe former
256
N
rir
:c!
jj
'^
'tr
n
L.
>
m
E
3
o
P. M. Cooper et al.
was long associatedwith the prereadingphase,the latter is seen as reading and writing
instruction in the primary gradesand throughoutelementaryschool. Stahl and Miller's
(1989) review of the literatureon whole language,however, found its greatervalue in a
"readiness"perspective,including kindergarten,ratherthan a beginningreadingand writing one.
Teachersare seento play different roles in fostering languagedevelopment.Dickinson and Sprague(2N2) suggestthat, althoughteachersmay becomesurrogatementorsof
languageand continue to foster young children's oral languagedevelopmentaway from
home, they may possibly underutilize their potential impact on languagedevelopment
when engagedin direct instruction.They report that teachersengagemore in intentional,
efficacious conversationsand more talk about past and presentevents during free play
than in large group times focusedon specificskills. Others(Stahl& Miller, 1989;Watson,
2002) caution that an emphasis on oral language development without a concomitant
emphasison featuresof print will not advanceemergingliteracy in the classroom.
The Storytelling Curriculum
Paley introduced the storytelling curriculum in Wally's Stories: Conversations in the Kindergarten(1981) without referenceto its usefulnessto oral languagedevelopmentor literacy subskills. For Paley, its main purposethen, and in all recent iterations, is to offer
young children an opportunity to tell their classmates"what they are thinking about"
(p. 66). Its tools are what Paley refers to as storytelling and story acting, also referred to by
Cooper (1993,2005) as dictation and dramattzation.In brief, though the storytelling curriculum varies somewhatfrom classroomto classroomand teacherto teacher,it has two
basic components.The first is the dictatedstory, in which the teacher-or scribe-is an
active participant,freely asking questionsthat help clarify the child's intention ("Is this
part a dream or did the little boy wake up? Should I write that?"). Sometimes,she offers
assistance("Are you thinking of the 4th of July?"), and, in a variation on Paley's original
plan (Cooper,2005), sometimesinstruction("Look here, a thword, like we talked about
in the mini-lessontoday"). Story contentis almost always,if not exclusively,determined
by the child. The secondcomponentis the dramatizationof the story by the authorand his
or her chosen classmates.In this part, the teacheracts as director, while the remaining
children constitutethe audience.Again, the teacherreservesthe right to interrupt,as she
talks with the author and actors about staging, improvising dialogue, and so on. As a
result,the ensuingdramaresemblesrehearsalmore than openingnight.
Guidelinesfor implementationcan be derived from Paley (1981; 1992,1997,2004).
Cooper (L993,2005) offers a methodologicaloverview basedon extensiveprofessional
development with teachersin the Houston area and colleaguesin the School Literacy and
Culture Project in the Rice University Center for Education. One feature of the storytelling
curriculum that is especially relevant to the study is its inclusion of second-language
learnersin classroomswhere all instructionis conductedin English. Teacherscooperate
on a variety of methods, from key vocabulary and peer assistanceto small objects and
"show me," to help the child seehis or her story written down and actedout, howeverlimited his or her English proficiency. (SeeFigure I for a methodologicaloverview.)
Though dictation and dramatizationactivitieswere not new to early childhood education when Paley (1981) wrote Wally's Stories: Conversationsin the Kindergarten, she
gets credit for bundling them into a unified and regular focus of classroomlife (Cooper,
2005). The storytellingcurriculum fits Donovan,Bransford,and Pellegrino's (2003) definition of an "authenticlearning"opportunityin that it allows participantsto "meaningfully
Can a Storytelling Curriculum Measure [Jp?
zJ/
Methodologicar overview of parey's storytelring curricurum
c{
Storytelling/Dictation
o
$
A
N
.T
'-
'tr
!i
o
o
>'
ct
C)
E
=
o
Where children sil The child/author should sit to the left of right-handed teacher/
scribe and to the right of left-handed ones so that the teacher's arm does not block the
c h i l d ' sv i e w o f t h e w r i t i n g .
carbon paper Carbon paper is used to make an instant copy of the story so that
the child can take one home and the teacherhas one to read to the classand later put in
the child's file or portfolio or ClassBook of Stories.
Name and date Before the child begins to dictate, the teacherwrites his or her
name in the left-hand corner and the date in the right. The teachershould say out loud
what she is writing.
How to begin First-time storytellers might need some help to get started. Others
are too shy or inexperiencedto dictate stories at first. It's okay to offer suggestions
until the child gets used to the process.("I really like those new sneakers.would you
like to tell a story about the day you went shopping?")Sometimesit's merely a matter
of offering a beginning or a way in. ("Some storiesbegin 'One day' or 'Once upon a
time' or 'Once there was a little boy.' Would you like to startthat way?")
Length Stories are limited to one page due to time restrictions.Children who
pressto tell more shouldbe taughtaboutinstallments,chapters,and "to be continued."
Subiect matter The fewer the restrictions on subject matter the better (except the
obvious-bathroom stories,explicit sexuality,unkind descriptionsof other children,
and so on,)
Echoing As the teacherwrites, she echoesback to the child what he or she has
just said. ("One-day-a-bear-came-to-dinner."
This keeps both teacher and
child on track as it calls attentionto the words being written.
A hesitunt storyteller If the child hesitatesbetween thoughts, the teachercasually
encouragesher or him to proceed.("Yes? And then what happened?"or "Okay. Go on.
I'm ready.")
Wrifing and narative developmenl The teacherhelps the child expand on his or
her thought by engaging in a conversationabout the story. ("Wow! You must have
beenreally scaredwhen the monstercame.Did you scream?Would you like to put that
in the story?What did the baby do that madeeveryonelaugh?") Sometimesit helpsfor
the child to think aheadto the dramatization.("Tell me what the kids are going to do
when they are tigersin the play. Maybe you could put that in your story aheadof time.")
Skill development The teacher indirectly points out or asks questions about
decoding,such as beginning sounds,doubleconsonants,and rhymes.Occasionallyshe
asks the child to spell a word that is a challengefor him or her. ("Do you remember
how to spell floor?") Grammar and grammarand punctuationminilessonscan also be
easily inserted.("Where shouldI put the quotationmarks?")
Figure 1. Methodologicaloverview of Paley's storytellingcurriculum.
258
N
-c
rif
(f)
A
N
.-T
(o
'tr
I
ri
o
(J
;
d]
o
=
3
o
P. M. Cooper et al.
Editing and revision Editorial questionsregarding sequencing,narrative development, and so on can be askedof the storytellerat any time. (..so, your mama took you
to the storeand then to school.Is that right?) Dictatedstoriesare rarely revised,though
they can be if a child desiresto.
Read the story after it is done When the child is finished dictating, rhe reacher
rereadsthe story to make sure she "got it right." She automaticallymakesany changes
the child requests.
Choosing the cast After finishing reading the story, the teacherreminds rhe child
that he or she needsto choosea cast.First, she calls attentionto the possiblecast by
underlining the charactersin the written story. It is assumedthat the author will play
the lead,though this is not a requirement.The child chooseshis castfrom the classlist,
noting who has not had a turn yet in this cycle. He or shemay chooseas many characters as there is room on the classroom's"stage" (dramatizationarea).Usually this will
be four to six actors and covers the main charactersin most stories.When necessary,
the audienceis askedto "imagine the rest" of the characters.The teacherwrites the cast
namesand their partson her copy of the story becauseit is difficult to rememberwho's
who when it comes time to dramatize later in the day.
Extra tip Teachers should be upbeat, involved scribes. ("No kidding? Oooh,
that's scary." "Hey, I like this part where the fire engine talks." "A deep blue, gooeygobbly day?-I love itt")
Story Acting/Dramatization
Keep it simple Do not think in termsof rehearsalsor props.
Gather the class in u semicircle The teacher begins by announcing who wrote a
story that day. In turn, she asks eachauthor to come standbesideher while she reads
the classhis or her story.
Calling the cast The teacherannounceswho in the classwill play which roles and
ask them to come stand"offstage." (A small rug helps to mark the spot. Actors move
on to the stageas required.)
Reread while cast acts out the story As the teacherrereadsthe story once more,
the chosenactorsact as the story line dictates.
Dialogue The teacher pausesbefore any dialogue to see if the child remembers
his or her lines. If not, the teachersimply repeatsthem, and the child repeatsafter her.
Improvisationis welcome,exceptwhereit changesthe author'sintent or distractsfrom
the overall play.
Directing the action The teacher should feel free to interrupt the dramatization
with suggestions.("Zoe, the little bear is very upset to find his porridge eatenall up.
Can you look upsetlike the little bear would?")
Curtain call After "The End" the actors ioin hands and take a bow while the
audienceclaps.
Figure l, (Conrinued)
Can a Storytelling Curriculum Measure Up?
Dramatization of Adult.authored Stories
@
c!
v
A
N
jj-
259
children love to act out adult-authoredbooks,too. This is a key opportunity for them
to learn how good storiesare constructed(beginnings,middles, ends, problem, solution, and so on). It also extendstheir vocabularyand knowledgeof sophisticatedsentencestructure.The methodis the sameexceptusually the teacher,not a child, chooses
the cast.Dramatizationof a favorite book can occur as many as five to a dozen times
beforethe children want to move on.
Figure1. (Continued).
(!
'tr
t'
0)
o
;
c0
E
3
o
o
constructconceptsand relationshipsin contextsthat involve real-world problemsthat are
relevantand interesting."To be clear, "real-world problems" in the courseof very young
children's storytellingare mostly psychosocialin nature(e.g.,sibling rivalry) and, though
often borrowedfrom real life (e.g.,good guys vs. bad guys) are usuallyfiltered throughthe
imagination.As such, it is ideally suitedto children's explorationsin language,from the
use of new vocabulary, to moving in and out of tenseand time frames, to the articulation of
decontextualized and abstract thought, exemplary of what Snow et al. (2001) refer to as
"extended discourse" in which all new readersand writers must become proficient. (See
Figure 2 for samplestories and a sampleexcerptfrom transcript.)From this perspective,the
ability to tell coherent stories is a by-product of both scaffolding (vygotsky, 1978) and
function. For teachers,the storytelling curriculum provides a model through which they can
model narrative discourseand createopportunitiesfor children to try their hand at it.
Researchers
have investigatedthe storytellingcurriculum from a variety ofpsychosocial angles(Cooper, 1993;Dyson, 1997;HurwiCI,200l; Katch, 2001,2003; Nicolopoulou,
Scales,& Weintraub,1994;Sapon-Shevin,1998;Sapon-Shevinet al., 1998;Wiltz & Fein,
1996).Researchon its direct relationshipto literacy developmentincludesa focus on narrative development(McNamee,Mclane, Cooper,& Kerwin, 1985; Nicolopoulou, 1996,
1997).Wolff (1993) theorizesthat, as describedby Paley, telling and acting out stories
fostersthe child's ability to move beyond "here-and-now"languageto "there-and-then,"
an essentialtask of namativedevelopmentand comprehension.In addition, Cooper(1993,
2005), Dyson (2002), Dyson and Genishi (1993), and Mason and Sinha (1993) arguethat
the storytellingcurriculum offers informal-but-deepsupportfor young children's acquisition of secondaryliteracy subskills,including symbol making, and conceptsabout print.
Again, to date, no study has investigatedthe impact of the storytelling curriculum on
young children's pre- and postperforrnanceon standardizedand formal assessments
of
oral languagedevelopmentand literacy subskills.
Method
The study was designedto comparecontrol (3) and treatment(3) public school classroomsof
same-gradechildren in the same school district or school building, representingtwo school
districts.All teacherswere expectedto follow the school-mandatedliteracy curriculum.
Research Team
All membersof the researchteam were at the time of the study closely associatedwith the
School Literacy and Culture Project in the Rice University Centerfor Education.All have
260
P. M. Cooper et al.
Pretest story from Ramon, ELL prekindergarten male
@
N
October 14,2004
(g
$
(f)
A
N
I played in the block center.Build car. Right there gamecenter.Bus bumpy. play cars.
I like play playdough.Library. Sandtable.Play outside.
The End
ji
Posttest story from Ramon, ELL prekindergarten male
'.E
April27,20OS
n
f
(D
The Running Big Bad Wolf
x
m
=
3
o
A big bad wolf killed all the people.He ate them. He went to the park and ate the fish.
The shark was in the water. The shark ate the bad wolf. The shark was playing with
him friends. They drink water.The alligator was in the water, too, They were fighting.
The alligator won. He went to drink from the water fountain.He went to the park doing
exercise.The daddy at the water fountain said,"Stop fighting."
The End
Sample transcript of teacher (J/child (G) interactions during story telling/dictation
Giovanni's Story
October 16,2005
J:
"Stand over here 'causeI want to be close by you. A1l right Mr. Giovanni. Let's
put your name G - I - O - Vanni. This is Giovanni's sixth story! you ready?
Do you have a good idea for your story?What's it going to be about?"
G: "A silly wolf."
J:
"A silly wolf-I can't wait to hearit!"
G: "And he was flying!"
J:
"He is silly! He had wings?"
G:
"Yeah, he made it out of paper."
J:
"Oh, I cannotwait to hearthis story...so'A silly wolf was flying.' Is that how you
want to start it?"
G: (nods)
J:
(Echoes& writes) "Tell me againhow he did that."
G: "He made his wings out of paper."
pre-andpoststories
Figure2. Sample
andsample
transcipt.
Can a Storytelling Curriculum Measure Up?
J:
N
-c
o
o
v
5
261
(E&W) (Pause)
G: "He didn't know how to blow houses
downbecause
all of themweremadeout of
bricks."
J:
"Ah, OK." (E&W)
G: "He pickedon. . .thechicken.. .he,he,he.. .pickedon thewolf andhe wentlike.
'ow !" '
N
J:
.g
o
n
G:
J:
j
g
>
m
"He picked on him or he peckedon him? Which word?',
'?ecked
on him."
"He pecked." (Echoes as she writes) "The chicken, he pecked-that's a good
word"- goesback to writing "he peckedon the wolf and he said-What?"
G: "He said,'Yowl"'
J:
q)
"Yowch!" (Chuckles) "oK... So thar poor wolf, rhe poor silly, flying wolf got
peckedby a chicken,huh?" (Chuckles)"And he said, 'yowch!," (pause)
(5
G: "A dog came and bit him."
E
;o
J:
"He's having a very bad day." (E&W) "A dog came and bit him."
G:
"He said nothing."
J:
"He said nothing?He wasjust quiet aboutthat, huh? Do you want me to write 'He
saidnothing'?"
o
G: (Nods)
J:
(E&W)
G: "And then a bearcame."
J:
"Oh no, I notice your animals are getting bigger and bigger-first one was little,
then a middle-sizedanimal, now he's gonna get this bear." (Preparesto write.)
"Did you say a big bear orjust a bear?"
G: "A big bear."
J:
"Then a big bear-"
G: "A big, big hugeBEARI"
J:
(Chucklesat his enthusiasm)"A big, huge bearcame." (E&W)
G: "Then he knocked him on the head."
J:
(Chucklesagain) "He knockedhim on the head." (E&W) "Then what?"
G: "He stilled."
J:
"He stilled?You mean he was frozen like that?"
G:
"He is frozen and he can still move."
Figure 2. \Continued).
262
J:
N
-c
s
A
N
.q
J:
|t
::
o
;
o
"He was knocked on the headand that madehim like frozen, like that?" (Dramatizes)
G: "Uh-huh. And then someoneput water on him and the ice came off him."
J:
"oh! so he was still-he was frozen," (writing) "And then what happened?Someone did what?"
G:
(Unintelligible)
J:
"wait a minute, let's go back to the frozen part." (Rereads)"He was still, he was
frozen. You said somethingabout someonepouredwater..."
G:
"They poured water on the ice and then it, he, he, the water got off him."
J:
(Writing) "So, they pouredwateron the ice..."
G:
"Then it got off the wolf."
J:
(E&W) "Can I ask you a quesrion real quick? Who is they'/ Who poured rhe
water?"
G
4
o
"He can still move?"
G: "Uh-huh.He was frozenand..."
ji
F'r
P. M. Cooperet al.
G: "Jaylen."
J:
"Jaylen." (Laughs,writes "Jaylenpouredwater on the ice.")
G: "I want to make a long story!"
J:
"It's getting pretty long! OK, so now your wolf is not frozen any more, so then
what did he do-the flying silly wolf?"
G: (Pause)
J:
"He's not frozen any more so he can do whateverhe wants."
G:
"The bearput him on a jelly bean."
J:
"On a jelly bean? So now I've got a wolf sitting on a jelly bean? Is this true?"
(Laughs)"OK." (Rereadsthe last two sentences
quickly) "The wolf got on..."
G: "The title is 'The Wolf."'
J: "The title is 'The Wolf?' That's right, becauseit's all about a wolf. OK, so the
wolf got on. . .what color were thejelly beans?"
G: "Yellow."
J:
"OK, can I put that in your story?The wolf got on the yellow jelly bean."
G:
(Nods)
J:
(E&W) "Then what happened?"
G: "He ate it."
J:
(E&W) "Did he like jelly beans?"
Figure2. (Continued).
Can a Storytelling Curriculum Measure [Jp?
@
G:
(Nods)
J:
"Did you want me to write that-'He liked jelly beans'-orjust skip it?,'
N
o
6
G: (With enthusiasm)"Write it!"
(E&w) "oh, I wish I had an artisr who could draw a picrure of a flying wolf who
likesjelly beans...Allright." (Rereadslastsentence)
.i-
J:
A
C.l
ii
G: "Then there was a girl who ate the whole wolf."
-T
J:
"A girl-like
G:
(Nods)
J:
(E&W) "Then there was a girl. . .whardid shelook like?"
"She looked like an Indian 'causeshe was brown."
'^
f
o
;
c]
E
G:
a peoplegirl?"
J:
"Should I put that in?" (E&w) "Then therewas a girl. She looked like an Indian."
(Interruptionfrom child sayingthey found a cocoon.J says,"Good for you!" then
returns to the story.)
J:
"She looked like an Indian becauseshe was brown. (Pause)An Indian from India
or a native American?"
=
o
o
263
G: "From Asia."
J:
"Should we put 'She was from Asia?"'
G:
(Nods)
J:
(E&W) "All right." (Rereadslast two sentences."She looked like an Indian
becauseshe was brown. She was from Asia.") "Now tell me again what this girl
did."
G:
"She ate the wolf."
J:
"OK. She ate the wolf."
G:
"And then a boy came and a bear.They looked the same!"
J:
(Distractedby anotherchild) "And then a boy came."
G: "And then a bear."
J:
(E&W)
G: "And they looked the same!"
J:
"The boy and the bear lookedjust the same?"(E&W)
G: "It's getting closer!"
J:
"Yes, you've about got your whole page here." (Rereads'And then a boy came
and then a bear.They looked the same.')What did they do?"
G: "They went to a person."
Figure 2. (Continued)
264
P. M. Cooper et al.
"Who?"
"They went to an artist."
N
o
o
J:
s
"Can I write that?That's a good word." (E&w) "oK, and what did they do when
they found the artist?"
"They said, 'Hello,' and then they went back".
A
N
(Writes with no echo) "What did rhe artist do?"
JJ
.-?
"He was painting...he was paintinga flying wolf that likesjelly beans."
:s
"They said hello and then they went back. should I write what the artist was
doing?"
j
(Agrees)
c)
o
o
"So how should I say it?"
>
"The artist was writing a flying wolf that likesjelly beans."
Q)
(Chuckles)"OK, was he writing or painting?"
=
3
"Painting."
r\
"OK, The artistwas paintinga flying wolf...we're at the end."
"We're almost thereso we can get a long story."
"It's long, but you've got to leave me some spacebecausewe've got a lot of
friends in this story." (Finishesechoingand writing) "That liked jelly beans."
"It's almost getting closer."
"It's very close 'causewe have to have room to write here. Shall I read it to you
first, then we'll pick whose gonnabe in it? 'The Wolf. A silly wolf was flying. He
made his wings out of paper.He didn't know how to blow housesdown because
all of them were made of brick. The chicken, he peckedon the wolf and he said,
"Yeowch!" The dog came and bit him. He said nothing. Then a big, huge, huge
bear came and he knocked him--oh, I skipped a word-he knocked HIM on the
head.He was still. He was frozen.Jaylenpouredwater on the ice and it got off of
him. The wolf got on a yellow jelly bean.He ate it. He liked jelly beans!Then
there was a girl. She looked like an Indian becauseshe was brown. She was from
Asia. She ate the wolf. Then a boy came and then a bear. They looked the same.
They went to an artist and they said hello and then they went back. The artist was
painting a flying wolf that liked jelly beans.'That's awesome.So who's gonnabe
the silly (laughing)flying wolf?"
"Julius."
J:
"Of coursehe is." (Juliusis G's good friend and he had also walked up to the table
during the last few sentencesof the story.)
G : "Jerald's sonnabe the chicken."
Figure 2. (ContinueQ
Can a StorytellingCurciculumMeasure [Jp?
"I thoughtToby was gonnabe it, but that'sa goodpick. Jeraldneedsto be in this story."
(Another child. "Chicken? Chicken is a funny name,")
co
N
o
J:
"This class loves that word!" (Brief aside as J explains to researcher
that Toby
started that. "when he first came to pre-K he liked to call people .chicken
Head.'we askedhim not to call peopleanimal names,so then, in order to get the
namechicken in he would say somethinglike 'chickenpot pie' and it's the way he
saysit that everybodyjust roars,so now chicken'sa funny word.")
J:
"All right, we've got a chickenhere.who's gonnabe the huge,huge bear?"
rq
N
i)
..7
'6
n
'j
265
J.
"Oh we skippedsomebody.We needa dog."
G: (Answerspreviousquestion)"Allejandro."
J:
"We needa dog."
G : "That's gonnabe Ms. Mandelli." [The studentteacher.]
;
J:
o)
G : "Uh-hum."
E
J:
=
"Ms. Mandelli. She'll do a goodjob. And you said Ailejandro will be the bear?"
"And Jaylen-who's gonnabe Jaylen?Jaylencould be his own self."
G : "Toby. And Julius is going to be the chicken."
"No, you said Jerald,Julius is going to be the flying wolf."
(Julius:"I can be the chicken and the flying wolf too.")
J:
"No, you have to just be one. OK, we needa girl-an Indian girl from Asia."
G: "Shairah."
J:
"And we needa boy."
G: "That's gonnabe Alexia."
J:
"Alexia" (a girl) "is going to be a boy? And I need an artist-is thar going to be
you? Who's gonnabe the artist?"(Pause)"You're not in the story, do you want to
be in it?"
G: "I want to be the artist."
J:
"OK, I didn't want you to forget yourself.OK, we're done."
G: "Now it's time to clean up."
Giovanni's story
The Wolf
"A silly wolf was flying. He made his wings out of paper.He didn't know how to blow
housesdown becauseall of them were made of brick. The chicken, he pecked on the
wolf and he said,"Yeowch!" The dog cameand bit him. He said nothing.Then a big,
huge,huge bearcame and he knockedhim. He knockedHIM on the head.He was still.
Figure 2. (Continued).
266
@
N
o
s
P. M. Cooper et al
He was frozen.Jaylenpoured water on the ice and it got off of him. The wolf got on a
yellow jelly bean. He ate it. He liked jelly beansl rhen there was a girl. She looked
like an Indian becauseshe was brown. She was from Asia. she ate the wolf. Then a
boy came and then a bear.They looked the same.They went to an artist and they said
hello and then they went back. The artist was painring a flying wolf that liked jelly
beans."
A
The End
N
ji
Figure2. (ContinueA.
(s
'c
!j
>
c0
c)
(!
E
3
o
considerableexperiencein both implementingPaley's storytelling curriculum and conducting related professionaldevelopment.Cooper (1993, 2005) has written previously
about it. Capo and Mathes acted as mentors to the three treatmentclassroomteachers
aroundthe storytellingcurriculum.After severalvisits to eachclassroomto get acquainted
with the children so as to make participationin the study easier,they also conductedthe
pretesting and later the posttesting, Capo and two other researcherscollected data on
teacher-child interactions as stories were dictated and acted out. This latter data was not
analyzedfor this study.
Classroom Selection
The study took place in two public prekindergarten,kindergarten,and mixed-aged classrooms in lower-and mixed-income communities in southeastTexas. Treatment and control
classroomswere groupedas pairs with matching grades,community and family demographics, and school-mandatedcuniculum. Approximately 7570of the study children qualified for
free or reduced lunch. Roughly half of the participants,including all of the prekindergarten
classroomsand the treatmentkindergartenclassroom,were designatedas English Language
Learners(ELL). Although qualificationsfor ELL designationvariesfrom district to district
and even school to school, all of the children had been evaluatedby school personnelas
competentenough in English to participatein the study and perform reliably on the selected
instruments.All classroominstructiontook place in English. Both the kindergartenand
mixed-agetreatmentand control classroomswere in the sameelementaryschool building.
Becauseone of the prekindergartentreatmentclasseswas the only one in the building with
ELL students,district administratorsidentified a class in another elementary school that
matchedthe treatmentschool in ethnic makeupand family backgrounds.
Treatment Classroom Teachers
Each treatmentclassroomteacherhad participatedpreviously in a professionaldevelopment program conductedby the researchteam on Paley's storytelling curriculum. Each
had implementedit for at least I year prior to the study,and had servedas a mentor to new
residentsin the program.Dictation and dramatizationof children's stories,the core of the
curriculum, as well as ongoing dramatizationof quality children's literature,took place
routinely in each of theseteacher'sclassroomat least four times per week. In the study
year, each treatmentteachertaped the storytelling process.The storytelling curriculum
was consideredan addition to eachschool'srecommendedliteracv curriculum.
Can a StorytellingCurriculum Measure Up?
267
Control Classroom Teachers
N
!
\r
;i
c!
u
:g
f
o)
::
o
E
3
o
The control classroomteacherswere randomly assignedto the study by virtue of
the factors mentioned above. None of the control teachershad received professional
development in the storytellingcurriculum, though all had attendedan information session
on the
study. (SeeFigure 3 for an overview ofdescriptive data.)
Subject Participation
Approximately 92Vo of 124 children in both control and treatment classrooms
returnedpermissionslips to participatein pre- and posttesting.Regardlessof permission to participate,all children receiveda picture book for returning the permission
form' All teachersreceived a box of books to supplementtheir classroom library.
Every effort was made to obtain pre- and posttestmeasureson each eligible child,
though children were allowed to refuse to participateat any point, and severaldid.
Some scores could not be obtained due to continuing absences,families moving
away, and schedulingconflicts. In the end, pre- and postscoreswere obtainedfrom 95
children.
District permissions,identification of control classrooms,and parental permission
were obtainedby early November,when pretestingbegan.Posttestingbeganin mid-April,
beginning with the classroomsthat were testedfirst during the pretestingsessions,allowing for approximately4.4 monthsfrom pretestto posttestfor all subjects.
Treatment
Although all children in the classroom participated in the dictation and dramatization
activities, only those children with parental permission were followed for the study.
Participating teacherseach took dictation and then dramatizedchildren's stories4 days
a week. Stories from quality children's picture books or other forms of literature were
acted out by small groups of children on the 5th day. The dictation processgenerally
required l0 to 15 minutes of one-on-oneteacher/childinteraction per story. Two classrooms set the goal of recording and acting out two children's stories each day, while
the third teacher chose to do only one. Neither is an unusual schedule in storytelling
classrooms(Cooper, 2005), though Paley strongly statesa preference for greater frequency (personalcorrespondence).In each ofthe classrooms,teacherstook dictation at
a classroom table in full view of the other children in the class. Teachers sat directly
besidethe participatingchild, positioningthe paperso that the child could clearly see
what was being written down. Teachersechoedeach storyteller'sdictated words as
they scribed to ensure that the children's words had been captured exactly as intended
or modified, if necessary.Teachersalso pausedto ask both clarifying and extension
questions.Stories were limited to one page in length due to time constraints.At the
story's close, the author chose which classmateswould representgiven charactersin
the forthcomingdrama.All dramatizations
took placethe sameday in which the stories
were dictated. Children were asked to imagine props and nonessentialcharacters.
Teachershelped extend the children's dramasearly in the year by asking prompting
questionsor having the audiencesuggestpossibleways to act out a given part. As the
year continued,children took over greatercontrol of the director's role in their own
dramatizations.
'Q
I
ca
(\
E
o!:
+.?
{)v
x
v
:
A
c{
I
ra)
';.i
'tr
13
o
n
E"E= ; $
!.)
ea
=
cgy
c.)
tr\J
7
f+,qir
-..
9E
gR!
=
g'
s
:
+. csc6.
I r E9; !n
i,;q
A e ==
rE: 5
*
..
a9!t
o
0)
X>
9,
I
F
J
E;
F
ra:; :I 5 Er+s
ieFa
I
c)
3
E
S $:!72v
s
.=
>
Ra
g
o
I
5R
F.:r
e
;
s:Ec =i! EE?
fiF5'"
'.1
-!.1
RL
!
i:
b-
"
6-a
tlE
E
irr
e
9ie
EiEPei
E:a;,.==
\.,r
i5
E'
tr0..66
S,P
Bft t !>
E
Hgi E: E
Eou
E
c.t
0.
t"
(-)
i:
h tr
[ -E e
i; sE
F+HiEsFs
JgEE"
i e H ; e g t:
_ > oh E
:
tr
r
v;
a
^-=
+A
Fi;gEEgr
l!BE
g-
;,
!- F.:
ifla#Ef! e5,FE,t-ts:5;!-
*
;
E ,!,
ts!
7
q
a
rn
g.F's#
$gifu5,
F
!
?n
u!
&
Lxl
vx&
g:;ii'r
v -^
.92
Y^
s h
'B
\
a^
= F.=
tr.F
-di E
<
?. * >
i
EE!
d!
a=
ts
gr
'sa.2
gEaa;EF,EF*g
2 -l
# !EH
s s::- tsd
EiH
-;Eeefr
ds"!iir
t H:F
L;: !h E E;
oi
o
H
.o
-
O.
'd
g!e AE
b
-,8e i'a
E
IE;EEE ={ E
-a=sir$
as!;E gFgFe-s* !
E -q=;5* 5"o*=
; !
268
Can a Storytelling Curriculum Measure (Jp?
269
Assessments
N
\t
The researchteam chosecommonly acceptedand widely usedmeasures
of children's oral
language and early literacy knowledge. The prekindergartenchildren
were given the
Expressive vocabulary Test (EYT) and the peabody picture vocabulary zesr (ppvr),
as
well as Whitehurst's Get Ready to Read! (GRTRI) screeningtool for kindergarten
readiness.The kindergarten children were given the only the EVT and the ppvr.
N
Findings
'F
o
g
L:
o)
;
@
(!
=
=
o
A priori it was determinedthat subjectswould needto have valid pre- and posttestscores
on both the PPVT and the EVT to be includedin the analysis,as the basicmeasureof analysis is change from pre- to posttestscore.Through this criterion, a basic data set was
established,sincethesetwo measureswere administeredto all subjects.The basicdata set
included58 (32 pre-w26 K) treatmentsubjectsand37 (16 pre-w2l K) controls.
As indicatedin Figure 4, the importantresult is that the changesin both the PPVT and
GRTRI testswere significantly greaterfor the children in treatmentclassrooms(p . .05,
one-tailed).Changein each test was defined as the postscoreminus the prescorefor each
child. An independentsamplest-test(SPSS,chicago,IL, version r2.o,2oo3) was usedro
evaluatethe null hypothesesthat there was no differencebetweenthe treatmentand control classroomsin any of the changescores(PPVT, GRTR!, and EVT) versusthe hypothesis that the treatmentgroups improved more. Homogeneityof variancewas satisfiedfor
all comparisons(p > .5).
children in the treatmentclassroomsimproved an averageof 4.5 points (+ 1.0 sem)
on the PPVT, while children in the control classroomsimproved only an averageof 0.7
points(+ 1.3 sem).This differenceis significanr(t90 =2.23, p = 0.013, l-tailed) and rhe
x
2a
?
-
!q3
';
?z
C
270
@
c{
-c
o
v
;{
c\|
!
'i\
'tr
n
f
o
;
o)
=
3
o
P. M. Cooper et al.
effect size of this differenceis 0.49, which is consideredto be
a moderateetl'ectand to be
educationallysignificant.
children in the treatmentclassroomsimproved an averageof 4.0 pornts (+
0.7 sem)
on the GRTR!, while children in the control classroomsimproved only
an averageof 1.7
points (+ l.l sem).This differenceis significanr(r4t = l.g, p = 0.03b,
t_taileajandthe
effect size of this difference is 0.60, which is consideredto be between a moderate
and
large effect and to be educationallysignificant.
There was no significant difference between children in the treatment and control
classroomsin their changeon the EVT (190= 0.95, p = 0.175, l-tailed), though the differ_
ence was in the predicted direction with treatmentclassroomsshowing slightly greater
improvement (4.5 + 1.1) than control classrooms(2.9 + 1.3). standard power analysis
(Cohen, 1988) suggeststhat significantly greaterimprovement after treatmentcould be
demonstrated(a = .05, F = .2) with278 childrenin eachgroup or about six times the effort
hereinaccomplished.
Figure 4 shows the changes in each of the three tests in both treatment and control
classrooms.Positive changesindicatethat the scoreimproved. To summarizethe significant results,the PPVT and GRTR test scoresimproved more in the treatmentclassrooms.
Figure 5 shows the averagepre- and posttestscoresfor the PPVT and for the GRTRI
testsin the treatmentand control classrooms.None of thesemeansare different between
the treatmentand control classrooms(p t .13 for the pretestsand p > .65 for the posttests,
respectively).The trend in the treatmentclassroomsbeing more steeplyupward was evaluated with the change scores describedabove. Significance in the individual change
scores(Figure 4) and not in the group means(in Figure 5) indicatesthat each individual
needsto be their own control (by comparingpre- versusposttestscores)in order to detect
the effect of the treatmentin this sample.Such subject-to-subjectvariance within each
group is expected.
Discussion
In her most recent book, A Child's Work: The Importance of Fantasy Play (2005),Paley
writes that, if academicsuccessis the goal, the early childhood community cannotafford
to dispensewith a play-basedcurriculum.Like many advocates(Pelligrini & Galda, 1993)
PPVT
MPE
Figure5. Pre-andposttest
meanson thePPVTandGRTR!
Can a Storytelling Curriculum Measure (Jp?
N
(s
s
N
'tr
j
c]
0)
o
=
3
271
of a traditional, play-basedearly childhood curriculum in the kindergartenand
below,
Paley arguesthat the best preparationfor later academicsuccess,from symbolic thinking
to comprehensionto problem solving, is the developmentof the imaginationthroughplay.
This study does not reject or attemptto measurethe storytellingcurriculum's impact
on the young child's imaginationor other developmentalgoals.It was designedto ask if it
meets more standardizedgoals, specifically vocabulary (ppVT and EVT) and reading
readiness(GRTR!). Findings reveal that both English LanguageLearners and Englishonly speakerswho participatedin the storytelling curriculum made significant gains in
vocabularyknowledge,as measuredby the PPVT. They further suggestparticipationalso
significantly impactedprekindergartenchildren's performanceon the GRTRI Though in
the predicted direction, flat scoreson the EVT for both groups are hypothesizedto be
relatedto the children's English limitations,despitethe districrs' appraisalof rheir abilities. Researchersfelt that in contrastto the PPVT and GRTR!, the demandfor expressive
languageon the EVT exceededmany of the children's expertiseor comfort level at that
time.
The importanceof the findings relatedto the PPVT and GRTRI is immediateand significant for teachersand teachereducatorsinterestedin Paley's storytelling curriculum.
The findings extend the storytelling curriculum's educationalefficacy beyond its established psychosocialand narrativevalue to specificacademicgains increasinglynecessary
to justify implementation.The significanceis amplified by the fact that its participants
were consideredgreatly at risk for entering first grade without sufficient languageand lir
eracy skills, including lower income, nonstandardEnglish speakers,English Language
Learners, and other children whose home life cannot guaranteethe free-ranging oral languageexperiencesaroundthe creationand explorationoflanguage and text that are imitative of middle-classhomes and associatedwith early literacy successin school. The
findings should also be of great interestto teachers,teachereducators,and advocatesof
authenticearly literacy curricula in general.
In recentyears,the limited empirical evidencedemonstratingthe academicrelevance
of such practicesin the classroomhas left teachereducators,teachers,and other school
district personnelunpreparedto defend their selectionof them. First, looking beyondthe
storytellingcurriculum, the findings underscorethe gainsto be had in preservingthe early
childhood tradition of authenticearly literacy curricula,and call into questionthe prevailing trend to abandonsuchfree-rangingoral languagecurriculain favor of a skills-centered
approachto early literacy.Working backwardsfrom the outcomes,the findings ultimately
lead to the question that most concernsteachereducatorsand teachers:What notable
aspectsof the storytelling curriculum account for the children's progress in language
developmentand skill knowledge?That is, what happensinside the curriculum to support
early literacy successin theseareas?Which aspectsof the story telling curriculum might
be most appropriateto integrateinto the literacy curriculum of teachereducation?
Transcript analysisand story contentsuggestfour aspectsof storytelling curriculum
are plausibly responsiblefor its success,which may haveapplicationto other authenticliteracy curricula.The first is the way in which the storytellingcurriculum marrieslanguage
experimentationand explorationto young children's living and learning agendas.Dictation and dramatization,which dependon languageand little else, make it possiblefor
them to sharewhat happenedlast night, to satisfytheir curiosity about the squiggleson the
paper, or to act like superheroeson stagewith their friends. Hence, their motivation to
becomestorytellers-language users-increasesautomatically.In addition,it makessense
that as children get more experiencewith the processand more exposureto the possibilities of languageto convey their ideas more accurately,their vocabulary increasesand
272
6
N
o
sf
o
N
'o
(U
!j
o
;
o
-9
=
o
P. M. Cooper et al
languageuse matures,as doestheir knowledgeof print. A glanceat the sample
setof stories from ELL prekindergartner Ramon (Figure 2) reveals an over 200Vo increase
in the
sheernumber of words in his story from the beginningof the study to the end. Even
more
telling is his growth in sentenceconstructlonand semantics.
october 14,2004. "I played in the block center.Build car. Right there game
center.Bus bumpy. Play cars.I like play playdough.Library. Sand table.play
outside.The End."
April 27, 2005. "The Running Big Bad wolf A big bad wolf killed all the
people. He ate them. He went to the park and ate the fish. The shark was in the
water. The shark ate the bad wolf. The shark was playing with him friends. They
drink water. The alligator was in the water, too. They were fighting. The alligator won. He went to drink from the water fountain. He went to the park doing
exercise.The daddyat the waterfountainsaid, 'Stopfighting.' The End."
The secondfactor in the storytelling curiculum's successappearsto be the way in
which languageand knowledge of print is teacherfostered and scaffolded (Vygotsky,
1978).As the sampletranscript(Figure2) reveals,the processrequiresteachersto engage
in informal, but directed interactionswith individual children, highly imitative of discourseinteractionsin middle-classhomes.For example,the teacher(J) in the sampletranscript engagesthe child (G) throughlanguagethat is inferential,inviting, and personal,yet
it never fails to lead G on. Her openingstressesboth use and skills, startingwith attention
to the lettersthat begin his name,and movesquickly to clear narrativeexpectations,
J: "Stand over here 'causeI want to be close by you. All right Mr. Giovanni.
Let's put your name - G - I - O - Vanni. This is Giovanni's sixth story! You
ready?Do you have a good idea for your story?What's it going to be about?"
The child, in turn, acceptshelp from the teacher to both express and
expandhis ideasas sheputs them on paper.
J: "Oh no, I notice your animalsare gettingbigger and bigger-first one
was little, then a middle sized animal, now he's gonna get this bear.
Did you say a big bear orjust a bear?"
G: "A big bear."
J: "Then a big bear-"
G: "A big, big hugeBEARI"
J: (Chucklesat his enthusiasm)"A big, hugebear came."
He allows her to help him attend to the sounds of words and other
nuancesof language.
G : " H e p i c k e do n . . . t h e c h i c k e n . ,. h e ,h e ,h e . . . p i c k e do n t h e w o l f a n d
he went like, 'OW!'
J: "He picked on him or he peckedon him? Which word?
G: "Peckedon him."
J: "He pecked.(Echoesas she writes) The chicken, he pecked-that's a
good word-he peckedon the wolf and he said . . . . What?"
Can a Storytelling Curriculum Measure Up?
N
$
\r
(f)
N
ij
'.?
.G
6
;
c0
0)
-9
B
o
273
Thesetype of interactionsalso occurredwhen young childrendictatedportlons
or versions
of their favorite children's books, thus strengtheningthe children's relationship
to and
commandover common texts.
Teachersand teachereducatorsmay also want to considertwo other findings in the
datathat go well beyond the goals of this study,but that we hypothesizeimpact the storytelling curriculum's potential for languageand literacy development.They are worth nor
ing becausethey are so often overlooked in preparing young children for academic
achievement.The first is the inclusive natureof the dictation and dramatizationactivities.
As Ramon's stories suggest,no proficiency in any area of development,including language and subskills, is required to tell storiesor act them out. There is no possibility of
failure or needfor remediation.The only criterion is, as noted at the beginningof the article, "interest" (Dewey, 1902/1990).In this case,it is the desireto join the community of
storytellersand actors.Second,interestin participationis possibly increasedby the fact
that a common practice in the three storytellingclassroomsis the free choice of content.
Inspirationmay have come from a book. But it may also havecome from home,the television, the movies, or a peer.Borrowed and repeatedthemesare frequent.In this sense,the
storytelling curriculum validates who the children are, what they know, and what they
careabout.When combinedwith the other aspectsof the processdescribedabove,the storytelling curriculum offers young childrenfair and unrestrictedaccessto and growth in the
"discipline" (Dewey) of languageand literacy.
Conclusion
The positive impact of Paley's storytelling curriculum on young children's vocabulary
and skills knowledge suggeststhat it is a viable alternativeto the skills-dominant and
teacher-neutralearly literacy curricula increasingly prevalent in prefirst classrooms
aroundthe country. As such, it offers a model of what curriculum can look like and what
teacherscan do to supportearly literacy successthat is as child friendly and inclusiveas it
is effective.Findings direct researchers,
teachereducators,and teachersto continue their
advocacyof both the storytelling curriculum and other free-ranging,teacher-scaffolded
oral languageopportunitiesin the early childhood classroom.As discussed,the current
climate of accountabilityhas increasedthe needfor early childhood classroomsto provide
curricula that representfair and equitableimitationsof home life, relevantto the academic
demandsof the upper grades.To this end, the needfor early childhood teachersto retain
their historical focus on oral language-based
curricula that are directedat fair and equitable goalsfor young children-like the storytellingcurriculum-has never been greater.
References
Bielmiller, A. (2006).Vocabularydevelopment
andinstruction:A prerequisitefor schoolleaming.
In S. B. Neuman& D. K. Dickinson(Eds.),Handbook
of earlyliteracyresearch,Vol.2 (pp.4l-51).
New York: Guilford.
Chall,J. (1970).karning to read:Thegreatdebate.NewYork: McGraw-Hitl.
Cochran-Smith,
M. (1984).Themakingof a reader.Norwood,NJ: Ablex.
Cohen,J. (1988)Sralislicalpoweranalysisfor the behavioralsciences.
Hillsdale,NJ: Lawrence
Erlbaum.
Cooper,P. (1993).Whenstoriescometo school:Telling,writing,andperformingstoriesin theearly
childhoodclassroom.New York: Teachers& WritersCollaborative.
Cooper,P. M. (2005).Literacylearningandpedagogical
purposein VivianPaley's'storytelling
curriculum'. Journalof Early ChildhoodLiteracy,5(3), 229-251.
aa/
@
N
$
:{
c!
jj
'i
'tr
G
f
>
c0
o
=
=
o
o
P. M. Cooper et al.
Dewey, J. (1990). The child and the curriculum. Chicago: University of Chicago press. (Original
work published1902)
Dickinson, D. K. (2002). Shifting images of developmentally appropriate practice as seen
through
different lenses.Educat i onal Researc her, 3 I (1), 26-32.
Dickinson, D. K. & Sprague, K. E. (2002). The nature and impact of early childhood care environments on the language and early literacy development of children from low-income families. In
S. B. Neuman & D. K. Dickinson (Eds.), Handbook of early riteracy research, vol. 2 (pp.263280). New York: Guilford.
Donovan, S. M., Bransford,J. D., & Pellegrino,J. w. (Eds.).(zoo3). How people learn: Bridging
research and practice. Washington, DC: National Academy press.
Dyson, A. H. (1997). Writing superheroes:Contemporary childhood, popular culture, and classroom
literacy.New York: TeachersCollege Press.
Dyson, A. H. (2002). Writing and children's symbolic repertoires:Developmentunhinged.In S. B.
Neuman & D. K. Dickinson (Eds.), Handbook of early literacy research, vol. 1 (pp. 126-14l).
New York: Guilford.
Dyson, A. H., & Genishi, C. (1993). Visions of children as languageusers:Languageand language
education in education in early childhood. In B. Spodek (Ed.), Handbook of research in the educat ion of y o ung ch il d r en (pp.122*136). New York : Macmillan.
Goodman,K. (1986). What's whole in whole language?Portsmouth,NH: Heinemann.
Goodman, Y., & Goodman, K. (1979). Leaming to read and write is natural. In L. Resnick &
A. Weaver (Eds.),Theoryand pracrice in early reading (pp. 137-154).Hillsdale, NJ: Erlbaum.
Gullo, D. F. (2006). K today: Teachingand learning in the kindergafien year. Washington,DC:
National Association for the Education of Young Children.
Hart, B., & Risley, T. (2003). The early catastrophe.
Americon Educator, 27(l),4-9.
Hurwitz, S. C. (2001). The teacherwho would be Vivian. YoungChildren, 56(5),89-91.
Katch, J. (2002). Under deadman's skin: Discovering the meaning of children's violent play. New
York: Beacon.
Katch, J. (2003). They don't like me: kssons on bullying and teasingfrom a pre-school classroom.
New York: Beacon.
Mason, J. M., Sinha, S. (1993). Emerging literacy in the early childhood years: Applying a
Vygotskian model of Ieaming and development.In B. Spodek (Ed,), Handbook of research in the
educationof young children (pp.137-150).New York: Macmillan.
Mclane, J. 8., & McNamee, G. D. (1990). Early literacy. Cambridge, MA: Harvard University
Press.
McNamee,G. D., Mclane, J., Cooper,P. M., & Kerwin, S. M. (1985),Cognition and affect in early
literacy development. Early Child Developmentand Care, 20,229-224.
Meisels, S. (2006). Accountability in early childhood: No easy answers (Occasional paper, 6).
Chicago: Erikson Institute.
Neumann,S. 8., & Roskos,K. (2005).Whateverhappenedto developmentallyappropriatepractice
in early literacy? YoungChiLdren,60(4),22-27 .
Nicolopoulou, A. (1996). Children and narratives: Toward an interpretive and sociocultural
approach.In M. Bamburg,Namative Development:SixApproaches(pp. 179-215).Mahwah, NJ:
Erlbaum.
Nicolopoulou, A. (1997). Narative developmentin social context. In D. I. Slobin, J. Gerhardt,
A. Kryatza,& J. Guo (Eds.),Social interaction,social context,and language: Essaysin honor of
SusanErvin-Tripp (pp.369-390). Mahwah, NJ: Erlbaum.
Nicolopoulou,A., Scales,B., & Weintraub,J. (1994).Genderdifferencesand symbolic imagination
in the storiesof four-year-olds.In A. H. Dyson, & C. Genishi,C. (Eds.),The needfor story: Cultural diversity in classroom and community (pp. lO2-123). Urbana, IL: NCTE.
Paley, V. (1981). Wally's stories: Conversations in the kindergarten. Cambndge, MA: Harvard
University Press.
Paley, V. (1990). The boy who would be a helicopter: The uses of srorytelling in the classroom.
Cambridge,MA: Harvard University Press.
Can a Storytelling Curriculum Measure Up?
N
o
(!
s
A
N
ji
:g
(L
o
;
c0
o
E
3
275
Pellegrini, D. & Galda, L. (1993). Ten years after: A reexaminationof symbolic
play and literacy
research.Reading Research euarte rly, 2B(2), 162_175,
sadoski, M. (2004). conceptual foundationsof teachingreading.New york: Guilford.
Scarborough,H. (2002). Connecting early language and literacy to later reading (dis)abilities:
Evidence,theory, and practice.In S. B. Neuman& D. K. Dickinson (Eds.),Handbook of early
literacy research,VoL I (pp.97-l l0). New york: Guilford.
Snow, C. E.' & Dickinson, D. (1991). Skills that aren't basic in a new conceptionof literacy. In
E. M. Jennings & A. C. Purves (Eds.), Literate systemsand individual lives: Perspectiveson literacy and schooling (pp. 179-191).Albany, Ny: stare University of New york press.
snow, c. 8., Tabors, P. o., & Dickinson, D, K. (2001). Languagedevelopmentin the pre-school
years. In D. Dickinson & P. Tabors (Eds.), Young childrenlearning at home and school: Beginning with literacy and language (pp. l-26). Baltimore, MD: Brookes.
Stahl, S. (1998). Understanding shifts in reading and its instruction. Peabody Journal of Education,
73(3& 4),31-67.
Stahl,S., & Miller, P. D. (1989).Whole languageand languageexperienceapproachesto beginning
reading:A quantitativeresearchsynthesis.Reviewof EducationalResearch,39(l),97-116.
U.S. Department of Education. National Center for Education Statistics. (1998). Earty Childhood
Learning Study. Kindergarten class of 1998-1999. Washington, DC: Office of Educational
Researchand Improvement.
Watson, R. (2002). Literacy and oral language:Implicationsfor early literacy acquisition.In S. B.
Neuman & D. K. Dickinson (Eds.), Handbook of early literacy research, Vol. I (pp. 43-53).
New York: Guilford.
Wiltz, N. W., & Fein, G. G. (1996).Evolution of a narrativecurriculum: The contributionsof Vivian
Paley. YoungChildren, 51(3), 6l-68.
Wolfl D. P. (1993). There and then, intangible and internal: Namativesin early childhood. In
B. Spodek (Ed.), Handbook of research in the education of young children (pp. 42-56). New
York: Macmillan.
Download