DOCUMENT RESUME

ED 449 473 CS 014 230

AUTHOR Williams, E. Jane, Ed.TITLE The Early Literacy Learning Initiative (ELLI) at The Ohio

State University Research Report, January 1998.INSTITUTION Ohio State Univ., Columbus.PUB DATE 1998-01-00NOTE 26p.; Holes punched in the document may obscure some text.

Published annually. The Early Literacy Learning Initiativeis now called the Literacy Collaborative.

AVAILABLE FROM Literacy Collaborative at The Ohio State University, 807Kinnear Rd., Columbus, OH 43212.

PUB TYPE Numerical/Quantitative Data (110) Reports Research(143)

EDRS PRICE MF01/PCO2 Plus Postage.DESCRIPTORS *Beginning Reading; *High Risk Students; *Instructional

Effectiveness; *Literacy; Primary Education; ProfessionalDevelopment Schools; *Reading Achievement; *ReadingInstruction

ABSTRACTA study described and evaluated the Early Literacy Learning

Initiative (ELLI), a collaborative effort between individual schools andschool districts and the Ohio State University. The overall goal of ELLI isto significantly raise the level of literacy achievement of kindergarten,first, and second grade students by implementing instructional approaches andsafety nets for students needing additional support. Four measures were usedto evaluate the program. Five schools met selection criteria: has been anELLI school at least 4 years; the ELLI training model has been followed andimplemented; and the literacy coordinator has been at the school since thebeginning of his/her training. Preliminary results reveal that the majorityof these schools is demonstrating a consistent pattern of improvedstandardized test results in reading, through the Fall of 1997.Recommendations include: (1) full implementation of the ELLI training modelappears to be necessary to achieve good results; (2) teachers need to be surethat children are engaged in reading and writing instruction for sufficientamounts of time each day; (3) efforts with schools to improve home-schoolcommunication should be continued; and (4) literacy coordinators needadministrative support to produce the desired change. (Contains 12references, and 10 figures and 5 tables of data.) (EF)

Reproductions supplied by EDRS are the best that can be madefrom the original document.

The Early LiteracyLearning Initiative

,cn at The Ohio State University

1

PERMISSION TO REPRODUCE ANDDISSEMINATE THIS MATERIAL HAS

BEEN GRANTED BY

P. Sc_61,r,e

TO THE EDUCATIONAL RESOURCESINFORMATION CENTER (ERIC)

U.S. DEPARTMENT OF EDUCATIONOffice of Educational Research and Improvement

EDUCATIONAL RESOURCES INFORMATIONCENTER (ERIC)

This document has been reproduced asreceived from the person or organizationoriginating it.

Minor changes have been made toimprove reproduction quality.

Points of view or opinions stated in thisdocument do not necessarily representofficial OERI position or policy.

OCID

U

the early literacy

ELLIlearning initiativeat The Ohio State University

Research ReportJanuary 1998

BEST COPY AVAILABLE

ELLI Research Report

O rcipacip §t60:819wis

The Early Literacy Learning Initiative Research Report will be published annuallyby The Early Literacy Learning Initiative at The Ohio State University.

Principal InvestigatorGay Su Pinnell

Program DirectorAndrea McCarrier

Program CoordinatorsTina Henry, Joan Wiley, Peg Gwyther, Mary Fried

Director of ResearchELLI Evaluation CenterE. Jane Williams

EditorE. Jane Williams

Layout and ProductionAllen S. Coleman

The Early Literacy Learning Initiativeat The Ohio State University200 Ramseyer Hall29 West Woodruff AvenueColumbus, Ohio 43210

!,*

Contents

The Ear lly Literacy Learning ffnitiative (ELLDat The Ohi® State UniversityResearch ReportJanuary 1998

Introduction 1

Components of the Early Literacy Learning Initiative 1

Literacy Learning and Teaching Component 1

School-Based Leadership Component 2

Professional Development Component 2

Assessment and Research Component 2

The Growth of ELLI ... 2

University ELLI Training Centers 3

Bilingual Students within ELLI Training Centers/Sites 4

Research/Evaluation Design 4

Instrumentation ... 6

Results 8

Responses of Educators to the ELLI School Development Program 15

Summary and Recommendations 17

Summary and Discussion 17

Recommendations 19

References .20

The Early Literacy Learning ffnitiativeat The Ohio State University

Research ReportJanuary 11998

Introduction

The Early Literacy Learning Initiative (ELLI)is a collaborative effort between individual schoolsand/or school districts and The Ohio State University(OSU). ELLI provides long-term professional devel-opment and systemic support for the educators whoare responsible for providing children's first schoolexperiences. The overall goal of ELLI is to signifi-cantly raise the level of literacy achievement of kin-dergarten, first, and second grade students by fullyimplementing a combination of tested instructionalapproaches and safety nets for those students who re-quire additional support to achieve independence asreaders and writers. The purpose of this report is two-fold. First, a thorough description of the Early Lit-eracy Learning Initiative will be provided. Second,iiiireliminary research results will be presented and dis-

ssed.

Components of the Early LiteracyLearning Initiative

The Early Literacy Learning Initiative incor-porates a design for creating systems to guarantee earlyliteracy success for all children. The ELLI projectuses principles based on effective district, school, andteacher change. ELLI includes components relatedto literacy learning and teaching, school-based lead-ership, professional development, and assessment andresearch.

Literacy Learning and Teaching Component

Theoretical Base. Theories underlying theEarly Literacy Learning Initiative staff development

oouodel come from the research of Vygotsky (1978),ner (1983), and Clay (1991). All three theorists

cus on the role of the "more capable other" in as-sisting learners to achieve independence. By

placing the teacher into the "more capable other" role,the theory can transfer into innovative classroom prac-tice. Teachers systematically observe student learningand then use the knowledge they obtain through ob-servation to inform and/or guide instruction.

Theory also guides curriculum and instruction.It is important that instruction is not too simple or toodifficult, yet offers challenges for the learner. Vygotskyrefers to this as working in the zone of proximal devel-opment; Bruner, as scaffolding; and Clay, as sensitiveteaching.

Instructional Framework. Students learn lit-eracy skills during authentic reading and writing ex-periences. The ELLI instructional framework includessuch methodology as reading aloud to children, sharedreading, guided reading, independent reading, sharedwriting, interactive writing, writing workshop, and in-dependent writing. During many daily reading andwriting experiences, children are taught about letters,sounds, words and how they work.

Flexible Grouping. Teachers work with bothheterogeneous and homogeneous groups of studentsdepending on the teacher's instructional purpose.When it is appropriate, for example during readingaloud or writing workshop, teachers work with theentire class. At other times they meet with small groupsor individual students.

Safety Net. A required safety net for all ELLIschools is one-to-one Reading RecoveryTM tutoring ingrade one for students needing more help.

Home Outreach Program. A parent outreachprogram may include inexpensive little books that chil-dren first read in school and then take home. ManyELLI schools use the KEEP BOOKSTM program aspart of their parent outreach.

5

Materials. The school invests in two kinds ofbook collections. Teachers need access to carefullyselected children's books that can serve as springboardsfor literacy activities and sources of content knowledge.Classroom collections include a wide variety of litera-ture and informational books. There is also an adequatesupply of leveled books housed in a central locationfor use in guided reading lessons.

School-Based Leadership Component

Commitment. School leaders pledge a five-year commitment to the training and participation ofthe staff.

Local Leadership. A trained literacy coordi-nator works with a literacy team. This team is com-posed of primary classroom teachers, Reading Recov-ery teachers, Title 1 teachers, reading specialists, spe-cial education teachers, and the school principal to de-velop and implement a local plan to support profes-sional development.

Professional Development Component

Professional Development of Literacy Co-ordinators. Literacy coordinators in-training partici-pate in a year-long course that includes six weeks oftraining at OSU or other ELLI certified university train-ing sites: Lesley College, Texas Tech University, andGeorgia State University. After their initial year of train-ing, literacy coordinators attend yearly professionaldevelopment institutes.

The program model provides opportunities fordifferent levels of assistance for participants. The pro-cess involves a cycle of university assisted learning,peer interacting, and individual practice and reflection(Button, 1992). Assistance is provided through instruc-tion, coaching, and demonstration incorporating sitevisitations, videotape analysis and feedback by univer-sity trainers.

Staff Development and Support. Teachersparticipate in a long-term professional development pro-gram that integrates theory and practice and is con-ducted by a specially trained literacy coordinator. Lit-

2

eracy coordinators offer long-term support to the staffthrough study groups, in-class demonstration lessonilland coaching.

Assessment and Research Component

Reflective Practice. Reflection is a key assess-ment tool utilized by ELLI to strengthen the instructionalprocess. Teachers continually reflect on the effectivenessof their teaching through discussions, videotaping analy-sis, and systematic observation of students' progress.

Systematic Assessment. Both formal and in-formal measures are used to monitor student progress.These measures include tasks found in An ObservationSurvey of Early Literacy Achievement (Clay, 1993),running records of text reading, and standardized tests.Teachers use the information gathered to inform their in-struction.

Research and Development. The ELLI re-search design incorporates a five-year data collection pro-cess which provides evidence to analyze changes in stedents' literacy learning. The research design also evalates school change over time.

The Growth of ELIA

ELLI began in response to a need for a staffdevelopment model that offers classroom teachers on-going support in learning new ways of teaching read-ing and writing. Work to develop a framework for lit-eracy lessons and a model for staff development beganin 1986 as a collaborative effort between staff mem-bers at The Ohio State University (OSU) and ReadingRecovery, and classroom teachers from the ColumbusPublic Schools.

In 1989-1990 the OSU staff conducted a year-long pilot study in which kindergarten teachers fromColumbus Public Schools were taught how to teachfor strategies and skills in their literacy lessons. Thekindergarten teachers attended formal classes and werecoached by OSU staff. Their instruction was video-taped. With OSU support, the kindergarten teache.reflected upon their teaching. Examination of data froclassrooms provided evidence of gains in student

iiiichievement as measured by scores on Clay's Obser-tion Survey of Literacy Achievement (1993). OSU

staff began to work on a model for broader dissemina-tion of the training. The goal was to develop a modelfor expanding capacity at the building level by traininga local leader for each school. After one year of inten-sive training, these local leaders, called literacy coordi-nators, would develop the knowledge and skills neededto support the teachers in their buildings as these teach-ers took on or refined their literacy instruction. Schooldevelopment was seen as a long-term process requir-ing several years of effort.

The first group of literacy coordinators wastrained at The Ohio State University in 1993 1994.Since then the staff of The Early Literacy LearningInitiative at The Ohio State University has trained anew group of literacy coordinators each year. Thereare now 177 schools representing 77 districts in 20 states(Figures 1, 2, and 3).

University ELLI Training Centers

In order to make literacy coordinator trainingmore accessible, The Ohio State University ELLI staffentered into partnerships with four other universities:Lesley College, Texas Tech University, Georgia StateUniversity, and St. Mary's College. Lesley College isin their second year of training literacy coordinators.Texas Tech University began training literacy coordi-nators this past fall (Fall, 1997) and Georgia StateUniversity will begin training literacy coordinators inFall, 1998. In a few years a training center will openat St. Mary's College in Moraga, California. The fouruniversity-based ELLI training sites were selected fortwo reasons. First, the university trainers at all foursites have a strong commitment to Reading Recoveryas a safety net for at-risk first graders. Second, all fourinstitutions are located in different regions of the coun-try, making literacy coordinator training accessible tomore schools.

200

-a 175co

uiO

750

50Ez

150

125

100

25

Figure 1. The growth of ELLI from 1993 to 1998

1--1=p77.2,

93-94 94-95 95-96

Years

96-97

School

El District

Z State

97-98

Figure 2. Number of literacy coordinators trained eachyear at each training site

100

90

80

N 70

A'J co;5 60

..g 50

i6o

40

E c) 30Z 20

10

0 1 I

93-94 94-95 95-96

Training Year

E[13 LC-Texas

El LC-Warren

LC-Jackson

LC-Maine

LC-Lesley

LC-OSU

96-97 97-98

Bilingual Students within ELL][ Training Centers/Sites

Presently, bilingual teachers at 21 sites in sixstates are supported by bilingual literacy coordinators.These teachers deliver the ELLI instructional frame-work in Spanish. To assess student progress kinder-garten, first, and second grade Spanish-speaking stu-dents are administered Spanish versions of the ELLIFall assessments that are given to English-speaking stu-dents. Bilingual Benchmark books are being devel-oped and field-tested for use with other bilingual mate-rials by The Early Literacy Learning Initiative at TheOhio State University. Aprenda 2 (Harcourt BraceEducational Measurement, 1997), a standardized read-ing test for Spanish students, is administered in the Fallto second grade Spanish-speaking students.

Research/Evaluation Design

The overall goal for the Early Literacy Learn-ing Initiative is to raise the level of literacy achieve-ment of kindergarten, first, and second grade students.The ELLI research design institutes Fall-Fall data col-

4

lection using a variety of reading and writing assess-ments, which incorporates both individual and groupadministrations.

In order to monitor student achievement, boilindividual and group results are analyzed. The purposesfor collecting data on individual children are (1) to helpinform instruction by building on each child's strengthsand (2) to analyze the growth of individual studentsover time. The collection of group data provides op-portunity to look at the class/group as a whole. Schoolofficials then have the opportunity to evaluate curriculaand teaching methodology by generalizing group re-sults and/or looking at trends over time (school/gradelevel).

The goal of data collection in the first year ofthe project is to establish a baseline for the purpose ofhistorical comparisons. In the Fall of each year, mea-sures are administered to kindergarten, first, and sec-ond grade children. The literacy coordinators are theonly teachers implementing the framework during theirtraining year. During the next year, classroom teach-ers are just beginning to use the new approach. Thus,fall testing in the first two years of the project actual] ,form a baseline for subsequent years.

.0 0 0

00 Oo

o.°

00

0 ,.0

0

.r

10

Figu

re 3

.M

ap o

f E

LL

I Si

tes

acro

ss th

e U

nite

d St

ates

L

s r 1

t,,

o

l..

... .

...e

tt

,.....

...%

(--

.7..

0...

0it;

,^0

0it.

00 0

00 0

Ok

IN0

/a

0-: 0

° : 0

0.°*

0' 0

00

I I%

I1

00°0

%0:

0:0:

#--

...I

1o

oI

00

0../

1I..1

'..N

o

*N0

, Ii.

l'iti

..

0o

0 o

0

1

co

4...

..c".

...7

--

....

I'1.

0.1

1110

......

. ...

..1

101P

ow*

cO°0

° ..4`

e...1

......

.. I

1,

..-

s1

,6.

.$

fI

4o

°i

I. ,

if

/".

k

1 1

0*-0

.0

° 0

00

0

LE

GE

ND

:0

Bui

ldin

g T

rain

ing

Site

s

9

to--

N...

-16,

.....-

-4.

.-

i0

I)

I,jft

f":

0:00

po

0 0

o 0

4

'''..-

.-.-

S.1 l

O

00

00 0

*Uni

vers

ity T

rain

ing

Site

sD

istr

ict T

rain

ing

Site

sT

he E

arly

Lite

racy

Lea

rnin

g In

itiat

ive

at T

he O

hio

Stat

e U

nive

rsity

©19

9810

Evaluation is an essential feature of ELLI. Eachyear data are collected on the literacy achieve-ment of kindergarten, first, and second grade chil-dren. The evaluation process serves several pur-poses:

0 To establish baseline and measure progresstoward goals. When they enter the project,school staffs can use data to establish a base-line against which to assess their school improve-ment efforts. Developing a school program toits full potential is not a quick or easy process.It is one that may take several years. System-atic examination of children's achievement canserve as a measure and guide for the peopleinvolved in the process.

0 To design/revise the school literacy cur-riculum. Data are used for problem solving re-garding the school program. For example, theliteracy coordinator, building administrator, andprimary team look at composite results from theassessment. Particularly revealing are the re-sults of second grade assessment, which pro-vides a measure of the impact of the K 1 pro-gram. Team members can detect areas thatrequire more attention and teaching time.

0 To inform daily instruction of individualstudents. Most of the measures used in theevaluation design are those that provide mean-ingful information for teachers because they in-volve observations of children that inform man-agement of instruction and teaching decisions.Observational procedures are individually ad-ministered; the entire process sharpens teach-ers' awareness of what children know and cando.

0 To inform administrators and the public.Each year the literacy coordinator and schoolplanning team prepare a report that describesthe school program, goals accomplished duringthe year, and student outcomes. ELLI requiresthat the reading and writing data be collectedfrom every student in grade kindergarten, one,and two. Many schools collect additional datato inform their instruction and program design.

Instrumentation

Four measures are used to evaluate the EarlyLiteracy Learning Initiative. These measures werechosen since they best matched the goals of the project.All are described below.

1) Hearing and Recording Sounds in WordsHRSIW) (HRSW) (Dictation Task). This task

is a measure of the child's knowledge of rela-tionships between letters and sounds in words.The assessor reads a sentence to the child andthen reads it again slowly, asking the child totry to write the words. Products are scored asto the number of phonemes accurately repre-sented through sound analysis. Two dictationassessments exist: (1) five sentences to be usedin kindergarten and grade 1 (HRSIW) (Maxi-mum score = 37) (Clay, 1993); and (2) onesentence for grade 2 children (HRSW)(Maxi-mum score = 64) (DeFord, Pinnell, Lyons,& Place, 1990).

2) Benchmark Text Reading Assessment. Bend.mark texts were constructed to determine stu-dents' ability to read, with 90 percent accur-acy or better, text(s) at their appropriate gradelevels.

3) Fluency. A 4-point likert-type scale to recordratings is used by the teacher to assess eachchild's ability to read with fluency and phras-ing. Fluency and phrasing are characteristicsrelated to comprehension (Fountas & Pinnell,1996).

4) Gates-MacGinitie Reading Test (3rd Ed.,1989). The Vocabulary and Comprehensionsubtests of the Gates-MacGinitie Reading Test(Level K, Form 1) are administered to all grade2 students in each building. This test series,which is published by Riverside Publishing,has empirical norms for fall and spring, estab-lished in the fall of 1987 and spring of 1988.(MacGinitie & MacGinitie, 1989).

Validity. To ensure that the instruments beingeked measure what is intended, the validity of the second

grade HRSW, Spelling, and Fluency measures was as-sessed. To assess the validity of these measures, indi-vidual scores for the second grade population were cor-related with the students' scores on the Gates-MacGinitieReading Test Subscales: Vocabulary, Reading Compre-hension, and Total Reading. All correlations were highlysignificant at the 0.01 level (2-tailed) (See Table 1). Sincethese measures demonstrated a strong correlation withthe Gates-MacGinitie Reading Subscales, one can statethat the HRSW, Spelling, and Fluency assessments val-

Table 1

idly measure a child's ability to recognize and understand/comprehend what he/she has read, ranging from singlewords to passages.

Validity: Does the test measurewhat it is intended to measure?

Pearson Correlations between Fluency, HRSW, Spelling and Gates-MacGinitie Reading Test Subscales for1996-97 Second Grade Population

End-of-Year HRSW HRSW Gates Fall Gates Fall Gates FallBenchmark Phonemic Spelling 1996 NCE on 1996 NCE on 1996 NCE on

Fluency Rating Awareness Reading Comp Total Reading Vocabulary

End-of-Year.603**

(n=5318).626**

(n=5111).691**

(n=4399).712**

(n=4393).694**

(n=4414)Benchmark

Fluency Rating

HRSW.783**

(n=6087).672**

(n= 5159)'.718**

(n=5151).720**

(n=5175)PhonemicAwareness

HRSW.707**

(n=4938).772**

(n=4930).788**

(n=4948)Spelling

Gates Fall.953**

(n=5392).855**

(n=5392)1996 NCE on

Reading Comp

Gates Fall.956**

(n=5392)1996 NCE onTotal Reading

Gates Fall1996 NCE onVocabulary

**. Correlation is significant at the 0.01 level (2- tailed).

712

Results

Presented in this section are the criteria forthe selection of schools for analysis in this report;descriptive information on the selected sites; findingsand interpretation of the Gates-MacGinitie ReadingTest results; and Fall-Spring results of first grade stu-dents at two different schools.

Selection of Schools. Schools were selectedfor inclusion in this report based on the following cri-teria.

1) The school has been an ELLI school at leastfour years, thus making it possible to exam-ine results over time.

2) The ELLI training model has been followedand is being implemented.

3) The literacy coordinator has been at the schoolsince the beginning of his/her training; he/shehas not taken a leave of absence, transferred,or resigned during this time. It should benoted, however, that the student and/orteacher populations might have changed inthe building during this time. For example,

Table 2

at School E during the 1996-97 school yearmost of the school's student population walpnew because of a change in the district's bus-ing policy. At School A, as displayed in Table2, there has been a substantial increase in theFree and Reduced Price Lunch (FRPL) popu-lation since 1992-93 (FRPL: 1992-93,32.2%).

An example of changes in the teaching popu-lations occurred with the first grade teachersat School A; two of the four first grade teach-ers were on Maternity Leave for half of theschool year during the 1996-97 school year.

Five schools met the above criteria; one fromthe first training class (1993-94), School A, and theother four from the second training class (1994-95),Schools B E. Schools A, B, C, and E are urbanschools/districts, while School D is a suburban school/district. Table 2 provides Free and Reduced PriceLunch status for each school. The number and per-centage of trained teachers in each school by gradelevel for the 1996-97 school year are presented J.Table 3.

Free and Reduced Price Lunch Status (FRPL) for Selected ELLISchools for 1995-96, 1996-97, and 1997-98 School Years

School 1995-96 1996-97 1997-98

School A 58.0% 41.3% 51.4%

School B 98.7% 95.2%% 95.7%

School C 83.9% 84.7% 92.0%

School D 45.0% 47.0% 43.0%

School E 99.3% 98.7% 99.6%

13

e' A. 8

Table 3

Number and Percentage of Teachers Who Have Completed Initial ELLI Training at Each Site for 1996-

School Kindergarten First Grade Second Grade Total

n % n % n % n

School A 2 100% 4 a 100% 3 100% 9 100%

School B 2 100% 2 100% 2 50.0% 6 83.3%

School C 1 100% 3 66.7% 2 50.0% 6 66.7%

School D 4 100% 5 60.0% 5 0.0% 14 50.0%

School E 3 66.7% 4 100% 2 50.0% 9 77.8%

Two of the four teachers were on maternity leave during the 1996-97 school year.

Standardized Test Results. Standardized testresults on the Gates-MacGinitie Reading Test, NCEs' ,

for three different cohorts of second grade students dur-ing the 1995-96 through 1997-98 school years are pre-

flitted for the five schools, Table 4, p.10. These stan-dized test results for the second grade students in the

Fall of 1995, Fall 1996, and the Fall of 1997 on ReadingComprehension and Total Reading are also displayed byschool in Figures 4 and 5.

The second grade cohorts in four of five of theschools (80%) demonstrated NCE gains across the threeyears on the Gates-MacGinitie Reading Test in TotalReading and Reading Comprehension from the Fall

'An NCE, Normal Curve Equivalent, is a statistical transfor-mation of percentile ranks in which reading achievement isdivided into 99 equal units with a mean of 50 and a standarddeviation of 21.06. NCEs are generally considered to pro-vide the truest indication of student growth in achievementsince they provide comparative information in equal units ofmeasurement. It should be kept in mind that NCEs are basedon percentiles, which compare the student's performance inrelation to the general population. An NCE of 50 representswhere a student should be for his/her grade level. For astudent's NCE score to remain the same at posttest as at pre-test does not denote a lack of absolute progress; on the con-trary, it means that the student has maintained the same rela-tive position in terns of the general population. Even asmall gain in NCEs indicates advancement from the student'soriginal level of achievement. T, E ''-

of 1995, 1996, and 1997. The schools in which theELLI model has been fully implemented demonstratedan average NCE gain of 5.60 NCEs in Reading Com-prehension and 5.31 NCEs in Total Reading.

Standardized test results at School E were ex-amined further due to the change in the student popula-tion at the beginning of the 1996-97 school year. Thegroup of students who were new to the school in 1996-97 was examined in relation to two groups of students.These two groups of students were in classrooms of teach-ers who were ELLI trained: one group of students at-tended school at least 141 days out of 178 school days(79%)2, while the other group attended less than 141days. Results for the three groups of students are pre-sented in Table 5. Results should be interpreted with cau-tion due to the small number of students in each group.

The second grade students who were in first gradeclassrooms taught by ELLI- trained teachers and attendedschool at least 140 days scored higher than those chil-dren who were in ELLI classrooms but attended schoolless than 141 days. In turn the students who attendedschool less than 141 days outperformed the group of stu-dents who were new to the school.

2The attendance criterion was based upon a natural break inthe attendance distribution for all students in ELLI schoolsduring the 1996-97 school year.

9 1

Table 4

Standardized Test Results on Gates-MacGinitieSchool by Year a

eading Subtests for Second Grade Cohorts by

School

Fall 1995

Total

Fall 1996

Total

Fall 1997

TaalReading

ReadingComp

Readingfaun

ReadingReading Reading Comp

Li NCE NCE n NCE NCE n NCE NCE

School A 74 38.7 38.3 66 47.2 46.6 73 42.8 40.6

School B 41 29.6 29.2 37 35.9 35.1 45 44.7 41.5

School C 40 24.7 24.6 31 31.3 27.9 34 35.8 33.1

School D 105 44.6 42.5 87 52.1 50.1 88 58.9 61.8

School E 50 18.6 17.2 36 24.9 23.2 43 30.0 27.9

a Fall 1995 was the first year that a standardized test was administered for evaluation purposes within the ELLI project.Consequently, only one year of baseline data is available for Schools B E in this report; two years of baseline data will beavailable for future reports. No baseline data are available for School A.

Figure 4. Average NCE for second grade cohorts inReading Comprehension Fall 1995, Fall1996, and Fall 1997

70.0

60.0

50.0 n = 66

0Z 40.0

-n

E 30.00ct

20.0

10.0

n = 31

73

n -n

n = 74n 10

= 41= 4n

0.0

A

n

El 1995

1996

ED 1997

B C D E

School

10

Figure 5. Average NCE for second grade cohorts inTotal Reading Fall 1995, Fall 1996, and Fall1997

70.0

60.0

n= 0 1995

1996

03 1997

= 6650.0

40.0 n

30.0

20.0

n=n = 31

n = 7 n=

4- I n

10.0

n = 74n =

n =10n = 41n= n=

0.0A

SchoolD E

Table 5

Average NCE Scores on Reading Comprehension and Total Reading on the Gates-MacGinitie Reading Test for Second Grade Students by Attendance Groups for Fall1996 at School E

Sample Reading

Fall 1996

Total

Comprehension Reading

Students in ELLI classroomswho attended school more than140 days

Students in ELLI classroomswho attended school 140 daysor less

Students new to school

n NCE n NCE

9

8

11

32.8

29.5

18.1

9

8

11

32.0

27.6

17.1

11

60

50

ui 40

< 20

10

Figure 6. Average NCE in Reading Comprehension andTotal Reading for second grade students atSchool E Fall 1996

Absent < 141 Days

MI Absent > 141 Days

D New to School

Reading Comp Total Reading

Individual Schools with Fall-Spring Data.ELLI requires the collection of Fall-Fall data. Of thefive schools in this report, two collected additional Fall-Spring data to measure student achievement on all firstgrade children during the 1996-97 school year. Thetwo schools were Schools B and C. Assessments ad-ministered by both of these schools to examine theirschool program included Text Reading Level (TRL)3and HRSIW. Both of these assessments are part ofthe Observation Survey (Clay, 1993). Results are pre-sented below.

School B is an urban school serving approxi-mately 310 students. This school is a Title 1Schoolwide building. Seven special education classes,22% of the school population, are housed at School B.In addition to being an ELLI school with ReadingRecovery support, over 100 community memberscome to the school each week to tutor students one-on-one in reading and writing.

Measures of Text Reading Level were obtained by constructing a gradi-ent of difficulty for text drawn originally from a basal reading system.A child's text reading level indicates the highest level of text that he/shereads at 90% accuracy or above.

it;, 12

Figures 7 and 8 display the stanine frequencydistributions of Text Reading Level and HRSIW score.for first grade students at School B in the Fall and Springof the 1996-97 school year. For stanines, as withNCEs, when a child remains at the same stanine levelfrom pretest to posttest means that the child has main-tained the same relative position in terms of the generalpopulation. Even a small gain denotes growth beyondwhat would be expected for that time period and gradelevel.

4 Stanines are standardized scores in which the range of reading achievment is divided into 9 equal units with a mean of 5 and a standardeviation of 2. Stanines of 1, 2, and 3 are below average; 4, 5, and 6 areaverage; and 7, 8, and 9 are above average.

1 "

Figure 7. Number of first grade children achieving TextReading Level stanines at School BFall - Spring 1996 -1997

16

14

12

10

8

6

4

2

0

ii 1

/

/

0--- FaII -TRL

"""' D ""' S pr-TRL

\I \

A

1

, t, . o'0 cf

2 3 4 5

Stanines6 7 8 9

12

10

8

6

4

Figure 8. Number of first grade children achievingHRSIW stanines at School BFall - Spring 1996 - 1997

. . 00 If

#

z2..,

0.

c - ---.0

0

---- Fall-HRSIW

- D - Spr-HRSIW

0 n 0 , o

1 2 3 4 5 6 7 8

Stanines

9

0 Observant Teachers. One literacy coordina-tor commented that teachers have a better un-derstanding of the learning process. Anotheragreed and said that teachers in her building "lookat children more on a continuum from emergentto independent." Another literacy coordinatoradded to the conversation by saying that at herschool "they look at what children can do, notwhat they can't."

Principals. All five principals from the build-ings highlighted in this report were also interviewed.When they were asked about the impact ELLI has hadon their buildings, their responses touched a variety oftopics.

0 Collegial Support. Several principals talkedabout the increased collegiality among staffmembers. One principal (Principal, School E)said that he has noticed a general increase incollaboration and teamwork since the schoolhas had a literacy coordinator. He attributedthis change to the leadership role of the literacycoordinator. Another principal stated that shehas seen stronger collegial relationships withingrade levels as a result of teachers participat-ing in ELLI training. Yet another spoke of theliteracy-specific training for classroom teach-ers on-site when asked about strengths of theprogram:

"Personally I like the availability of literacy-specific training for classroom teachers on-site. Having a literacy coordinator work-ing with classroom teachers has improvedliteracy teaching which, in turn, has led toincreased learning among students (Princi-pal, School D)."

ole of Literacy Coordinator. The buildingprincipals valued the work being done by theliteracy coordinator. One principal said thatthe staff in his building has become more re-flective practitioners. He stated that the literacycoordinator has helped teachers to think moreabout how children learn. This, in turn, has

caused them to change teaching practices. Hesaid that they have a much better idea of whleach child needs to learn in order to become abetter reader (Principal, School E). Anotherprincipal noted that "ELLI training enrichesthe staff and therefore enriches the literacylearning of children (Principal, School D)."

The principal at School A stated that the suc-cess of the program is linked directly to the supportand supervision of the literacy coordinator.

"The ELLI program was implemented fouryears ago at (School A) with the training ofour ELLI coordinator and the establishmentof our partnership with the Ohio State Uni-versity. Subsequent years brought ELLItraining for all staff, kindergarten throughsecond grades, to use as the delivery sys-tem for teaching reading and written lan-guage. The overall success of this programcan be directly attributed to the consistentsupport and supervision from our ELLIcoordinator, (LC, School A). She maintaicontact with classroom teachers and pro-vides technical expertise, encouragement,and program supervision. I cannot stressenough the importance of the ELLI coor-dinator position in the success of this pro-gram. The simple training of teachers andcasting them off to teach with newlylearned skills is not enough to realize themaximum benefits of this well designed andproven program (Principal, School A)."

Several principals mentioned the importance ofsite-based leadership. One said that having the sup-port at the building level makes learning easily acces-sible to everyone (Principal, School B).

0 Taking on New Instructional Practices. Notall teachers are eager to try out new instruc-tional approaches. Some literacy coordinatorsdeliberately trained the most enthusiastic teach-ers first, hoping that their enthusiasm willspread to more reluctant teachers.

A principal from one building said,

"Initially some teachers in our buildingwere reluctant to try out new ways of teach-ing. But they did notice the children's workdisplayed by the teachers participating inthe initial ELLI training. The followingyear they had students in their classroomsthat loved to read and write. These morereluctant teachers are now beginning tochange some of their instructional prac-ticeswithout my insistence (Principal,School B)."

The principal at School D supported this sameview:

"Some principals mandate that teachers tryout new approaches for teaching. I thinkthat when they come to it themselves, theyown it. We can force it on people, but itmakes for an unhappy staff and may notlead to the results you expect. This will notcreate long-term change. Change takestime.

In my building, some teachers were reluc-tant in the beginning, but as they saw theresults and value in it, they have begun toincorporate these new instructional prac-tices (Principal, School D)."

0 Teaching and Learning. Principals also ob-served that the climate for teaching and learn-ing has changed since the school became anELLI school.

"Of course the impact on student achieve-ment is impressive. But what really im-pressed me was the first time a child cameto read to me. I am from the old school. Ifa child stops, you tell him the word. Whenthis child got stuck, he knew just what todo. He cross-checked and self-correctedhis own reading. These young readers arebeing taught how to help themselveshowto use strategies. It is wonderful to see suchindependence at such an early age (Princi-pal, School B)."

17

In addition to the principals of the five schoolsstudied here, principals from two buildings that trainedliteracy coordinators last year already noticed changesin their buildings. Kim Marshall, principal at MatherSchool in Boston, stated that he has never seen a pro-gram that teachers have taken to so readily. He saidthat ELLI is off to a strong start in kindergarten, first,and second grades.

Parkway Elementary School is in the first yearof ELLI implementation. A literacy coordinator wastrained in 1996 - 1997 and is now working with teach-ers in their initial training. Even before this first fullyear of implementation Ellen Desoriers, principal, no-ticed that teachers were informally sharing ideas andthat literacy teaching and learning was increasing inthe school. She continues to be impressed by thechanges she has noticed in her building saying:

"I've never before been in a situation in whichteachers know so much about teaching the skillsof reading and writing. Without a doubt there ismore reading and writing going on in our school."

She remarked that even kindergarten children under-stand critical concepts related to literacythat readingand writing are useful tools to convey messages. Thedaily schedule at this school includes three to four hoursof English language arts instruction. Social studies andscience are incorporated into this block. An additionalhour is spent on mathematics instruction.

Summary and Recommendations

A summary and discussion of the results, andrecommendations for implementation and future re-search are presented in this section.

Summary and Discussion

The Early Literacy Learning Initiative is a com-prehensive approach designed to provide long-term sup-port to schools working toward successful literacyachievement for every child by the end of second grade.

A strength as well as a weakness in the researchconducted in ELLI schools is that this is not an experi-mental or controlled study; the data are from kinder-garten, first, and second grade classrooms where many"things" are going on each day. These events do notalways relate positively to teaching and learning, butthey are part of the process. For example, many chil-dren move from one school to another during a schoolyear; consequently, many schools have high mobilityrates. In addition, there are daily disruptions to teach-ing schedules, substitute teacher shortages, changes inthe teaching staff, etc. Nonetheless, the preliminaryresults from schools where the ELLI training modelhas been fully implemented look very promising.

Preliminary results from schools in which theELLI training model has been fully implemented forfour or more years reveals that the majority (80%) ofthese schools is demonstrating a consistent pattern ofimproved standardized test results (NCEs) in readingfrom the Fall of 1995 to the Fall of 1997. More spe-cifically, the average NCE gain on the Gates-MacGinitie Reading Test in Reading Comprehensionacross schools over the three years was 5.60 NCEsand 5.31 NCEs in Total Reading.

Standardized test results from one of theschools in which there was a change in the school popu-lation from the 1995-96 to the 1996-97 school yearwere analyzed further to find out if the NCE gain wasdue to ELLI or to new students moving in to the school.Analyses revealed that the students who had ELLItrained teachers in first grade and had attended schoolat least 141 days (79%) scored higher on both ReadingComprehension and Total Reading than students whohad ELLI trained teachers in first grade but had at-tended school less than 141 days. Both of these groupsof students, in turn, outperformed the students whowere new to the school. Consequently, it can be saidthat the NCE scores from the second grade studentswho were new to this school in 1996-97 did not out-perform the students who had been at this school infirst grade classrooms of ELLI trained teachers in Read-ing Comprehension and Total Reading. Nonetheless,these results should be interpreted with caution due tothe small sample size of each group.

18

Unfortunately, it is not clear why the NCscores for one of the schools decreased from Fall 199to Fall 1997. Preliminary review of the data revealedseveral possible explanations: 1) the Fall 1996 cohortof second grade students was an exceptional group,i.e., there is a larger number of high- ability students inthis class compared to other cohorts of students at thisschool; 2) two of the four ELLI trained first gradeteachers were on maternity leave during 1995-96; and3) ELLI training was revised from 1993-94 training tothe 1994-95 training, the effects of which are unclearsince the other sites trained during 1993-94 did notfollow the ELLI training model. Furthermore, since1995 was the first year that the standardized test wasadministered, no baseline data are available for SchoolA. Until additional analyses are performed, no expla-nations can be given.

Two schools collected additional Fall-Springdata on first grade students during the 1996-97 schoolyear. Both schools had Text Reading Level (TRL)and Hearing and Recording Sounds In Words(HRSIW) data on all first grade students. Fall anSpring stanine results revealed similar trends for thtwo schools. Each group of students showed a posi-tive shift in the stanine distribution from Fall to Spring,with the greatest shift demonstrated by HRSIW results.

Interview data from literacy coordinators, prin-cipals, and teachers at the five schools revealed nu-merous changes as a result of the implementation ofELLI. Several of these changes include increased col-legial support, change in the teacher talk, improved lit-eracy teaching, and increased learning and enthusiasmfor reading and writing among students. Overall theimplementation of ELLI has reformed literacy educa-tion in these schools.

23

ecommendations

For Implementation. Based on the findings, thefollowing recommendations for implementation are in or-

der.

1. Full implementation of the ELLI training model

appears to be necessary to achieve good re-sults. Analyses of data from schools not fol-lowing the ELLI model (and not included inthis report) are showing inconsistencies inGates-MacGinitie Reading Test results acrossyears. It is recommended that each literacycoordinator be assigned to only one school.

2. Research has shown that the amount of timestudents spend engaged in reading, i.e., en-gaged time, effects student achievement(Burstein, 1980; Fisher, Filby, Marliave,Cahen, Dishaw, Moore, & Berliner, 1978;Slavin, Karweit, & Madden, 1989). Eventhough teachers are trained how to help chil-dren become more strategic readers and writ-ers, teachers need to be sure that childrenare engaged in reading and writing instructionfor significant amounts of time each day inorder to achieve optimum results. Childrenwill not become proficient readers and writ-ers as quickly unless they read and write andreceive instruction in reading and writing ev-ery day.

3. Efforts with schools to improve home schoolcommunication should be continued so par-ents will realize the importance of childrenattending school daily. In addition, efforts tohelp reduce mobility rates, such as workingwith social service agencies or providing trans-portation so children can stay at the sameschool all year, need to be explored to helpindividual children.

4. Based on comments from principals and lit-eracy coordinators, it appears that literacycoordinators need administrative support inorder to provide the desired change. In

19

order for ELLI to be effective, a literacycoordinator needs to be able to fulfill theresponsibilities of a literacy coordinator andnot be used for other functions in the school,i.e., be assistant principal, substitute teacher,or pulled away for other administrativetasks. We would not expect children tomake the same gains in reading and writ-ing if the classroom teacher were absentmost of the time; likewise, we cannot ex-pect teachers to make changes in theirteaching practices if the literacy coordina-tor who supports their learning is assignedto do other duties.

For Further Research. Based on the findingspresented, the following research studies are being pro-posed:

1. The Early Literacy Learning Initiative at TheOhio State University is pilot-testing a dis-trict-level training centers model and has en-tered into collaborative agreements for train-ing literacy coordinators with several schooldistricts. Each district has a district-leveltrainer of literacy coordinators trained atOSU. Training a district level person requiresa minimum of two years of training. After thedistrict trainer has completed training, she/heis responsible for training building level literacycoordinators. Data from the pilot sites willbe evaluated to determine the effectivenessof the district-level training model.

2. It makes intuitive sense that children in full-day kindergarten programs may leave kin-dergarten knowing more about reading andwriting than children in half-day kindergartenprograms; thus, first grade teachers are ableto begin instruction at a higher level. SomeELLI schools provide full-day kindergarten,while others have half-day kindergarten.Consequently, a study needs to be con-ducted to look at the effects of children infull day versus half-day kindergarten pro-grams.

24

3. Presently, all literacy coordinators are ableto refine their training by attending yearlyinstitutes and receiving on-site visits fromuniversity trainers. More intensive modelsof support are being explored. For ex-ample, some previously trained literacycoordinators have been invited to attend amonthly study group at which time they dis-cuss how to be more effective in their roleas literacy coordinator. Further investiga-tions are being conducted to determine ifthis approach really does help improve theirprofessional development/training.

References

Burstein, L. (1980). Issues in the aggregation of data.Review of Research in Education, 8, 158-233.

Button, K.A. (1992). A longitudinal case study ex-amination of the theoretical and practicalchanges made by three urban kindergartenteachers during participation in early literacytraining. Unpublished doctoral dissertation.The Ohio State University, Columbus.

Bruner, J.S. (1983). Beyond the information given:Studies in the psychology of knowing. NewYork: Norton.

Clay, M.M. (1991). Becoming literate: The construc-tion of inner control. Auckland, New Zealand:Heinemann Education.

Clay, M.M. (1993). An observation survey of earlyliteracy achievement. Auckland, New Zealand:Heinemann Education.

DeFord, D.E., Pinnell, G.S., Lyons, C.A., & Place,A.W. (1990). The Reading Recovery follow-up study (Technical Report, Vol. III). Colum-bus, OH: The Ohio State University.

Fisher, C.; Filby, N.; Marliave, R.; Cahen, L.; Dishaw,M.; Moore, J.; & Berliner, D. (1978). Teach-ing behaviors, academic learning time, and stu-

20

dent achievement: Final report of phase IIIBeginning Teacher Evaluation Study. S

Francisco: Far West Laboratory for EducationalResearch and Development.

Fountas, I.C., & Pinnell, G.S. (1996). Guided Read-ing: Good first teaching for all children. Ports-mouth, NH: Heinemann, p. 81.

Harcourt Brace Educational Measurement. (1997).Aprenda 2 (Segunda edicion). San Antonio,TX: Harcourt Brace Educational Measurement.

MacGinitie, W.H., & MacGinitie, R.K. (1989). Gates-MacGinitie Reading Test (Third Edition). Chi-cago, IL: The Riverside Publishing Company.

Slavin, R.E., Karweit, N.L., & Madden, N.A. (1989).Effective programs for students at risk.Needham Heights, MA: Allyn and Bacon.

Vygotsky, L.S. (1978). Mind in society. Cambridge,MA: Harvard University Press.

Coordinator Training 011

U14 1 ViGliS ittv Tirialmim Suter

25

the early literacy

ELLIlearning initiative

at The Ohio State University

Reproduction Release

U.S. DEPARTMENT OF EDUCATIONOffice of Educational Research and Improvement (OERI)

National Library of Education (NLE)Educational Resources Information Center (ERIC)

Reproduction Release(Specific Document)

Page 1 of 3

CS 014 230

I. Document Identification:

Title: The, &rly LiferQdy Learning 1-41-614/11/e... Me, chid

Authos%Si(27 7i& 11/1 iier5 1.744)/

dage WWI arn 6 i4 ay aa P //mallCorporate Source: 74 6 Cisie ,..574 74e, amiiterzsvyPublication Date: CM k qr y / 9 9 7

II. Reproduction Release:

in order to disseminate as widely as possible timely and significant materials of interest to the educational community, documents announced in themonthly abstract journal of the ERIC system, Resources in Education (RIE), are usually made available to users in microfiche, reproduced papercopy, and electronic media, and sold through the ERIC Document Reproduction Service (EDRS). Credit is given to the source of each document,and, if reproduction release is granted, one of the following notices is affixed to the document.

If permission is granted to reproduce the identified document, please CHECK ONE of the following options and sign the release below.

/ermission is granted to the Educational

Resources information Center (ERIC) toreproduce and disseminate this material in

microfiche or other ERIC archival media (e.g.

electronic) and paper copy

The sample sacksa shown below WI be affotad to athwel Idocuments

PtntassioN ro REPRODUCEDISSEMINATE THIS MAMMAL HAS

BEEN GRANTED BY

TO THE EDUCATIONAL RESOURCESiNFORMATION CENTER (ERIC)

Level 1

Permission is granted to the Educational

Resources information Center (ERIC) toreproduce and disseminate this material in

microfiche and in electronic media for ERICarchival collection subscribers only

The sample sacker shown below *3 be affixed to all Level 2Adoasnents

PERMISSION TO REPRODUCE AMDISSOMATE itiATIIRAL IN

FaCRWMIKZ. AND IN ELIECTROUIC NEVArca ERIC CCUSCTION 41,1134GRONRS =X.

NAS BEEN GRANTED Wir

TO flr= solzemoNAL nesouncesINFORNARCW Man (SAC)

2A

Level 2A

Permission is granted to theEducational Resources information Center(ERIC) to reproduce and disseminate this

material in microfiche only

The sante sticker shown below vat be effaced to al Level 2Bdocument

P5RENSSON to MPROINJC6 AnDCOSSOJINAYE TI OS MATERIAL di

ECCROFSONE ONLY HAS BEEN GRANTED ri

28

-TO TIE EDUCATTONAL RESOURCES

nocardariczn c;orrEs IETHC)

Level 28

Documents will be processed as indica1451 provided reproduction quality permits. If permission to reproduce is granted, but neither box is checked,

documents will be processed at Level 1..;

http://ericeece.org/reprorel.html 2/4/01

Reproduction ReleasePage 2 of 3

I hereby grant to the Educational Resources Information Center (ERIC) nonexclusive permission to reproduce and disseminate this document as

indicated above. Reproduction from the ERIC microfiche or electronic media by persons other than ERIC employees and its system contractors

requires permission from the copyright holder. Exception is made for non-profit reproduction by libraries and other service agencies to satisfy

information needs of educators in response to discrete inquiries.

Signature:

Printed Name: /2irAt k, L., 5 e,44r et"'

Position/Title: AS,,506/Se.-- ,ri/ eSS or-

Organization: 7n oh/o ...5A,71-e, an/144:1"/y

Address: ao 0 if0.7,6 eye," 'Ad/ , 02 ? it/ vie,eirc,#, 6 /arfry 6 a5, y#Vc3 a/a

Telephone Number 6 /1/. a FR, a veoFAX: 6/4.,g9,1. 7 q60

E-mail address: 6.chater, i @ osq,edt,

Date: a.... 6_0/

III. Document Availability Information (from Non-ERIC Source):

If permission to reproduce is not granted to ERIC, or, if you wish ERIC to cite theavailability of this document from another source, please provide

the following information regarding the availability of the document (ERIC will not announce adocument unless it is publicly available, and a

dependable source can be specified. Contributors should also be aware that ERIC selection criteria are significantly more stringent for documents

which cannot be made available through EDRS).

Publisher/Distributor.

Address:

Price:

IV. Referral of ERIC to Copyright/Reproduction Rights Holder:

If the right to grant a reproduction release is held by someone other than the addressee, please provide the appropriate name and address:

Name:

Address:

V. Where to send this form:

Send this form to the following ERIC Clearinghouse:

Karen E. Smith, AcquisitionsERIC Clearinghouse on Elementary and Early Childhood Education

University of Illinois at Urbana-ChampaignChildren's Research Center

51 Gerry DriveChampaign, IL 61820-7469

phone: (800) 583-4135

http://ericeece.org/reprorethtml2/4/01