ED 463 521

TITLEINSTITUTIONPUB DATENOTE

AVAILABLE FROM

PUB TYPEEDRS PRICEDESCRIPTORS

IDENTIFIERS

ABSTRACT

DOCUMENT RESUME

CS 510 846

Reading Recovery[R] in Maine, 1991-94: State Report.Maine Univ., Orono. Coll. of Education.1994-00-0029p.; Published by University of Maine Center for ReadingRecovery.University of Maine, College of Education & HumanDevelopment, 5766 Shibles Hall, Orono, ME 04469-5766. Tel:207-581-2438; Fax: 207-581-2423; Web site:http://www.ume.maine.edu/-cel.Reports Descriptive (141)MF01/PCO2 Plus Postage.*Early Intervention; Grade 1; Primary Education;Professional Development; Program Evaluation; ProgramImplementation; Public Schools; *Reading Difficulties;*Reading Instruction; *Remedial Instruction; State ProgramsClay (Marie); *Maine; *Reading Recovery Projects; State Role

Reading Recovery is an early intervention program that helpslow-achieving first-grade children to become independent readers who need nofurther remediation. Reading Recovery was introduced by Marie M. Clay who, inthe mid 1970s, developed procedures with teachers and tested the program inNew Zealand. In 1984, the success of the program in New Zealand ledresearchers at Ohio State University to introduce Reading Recovery to theUnited States. In February of 1990, Kathryn Manning, consultant to the MaineDepartment of Education, organized a group of 26 Maine educators to go toOhio to see Reading Recovery teacher training in action. In the school year1990-91, Maine teachers began to be trained in Reading Recovery. This statereport delineates the implementation of Reading Recovery in Maine schoolsfrom 1991-94. The report is divided into the following sections: What ProgramParticipants Say; Program Overview; Program Evaluation Results; ProgramImplementation; and The Reading Recovery Lesson. (Contains 11 references.Also lists Reading Recovery training sites in Maine.) (NKA)

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

1

PERMISSION TO REPRODUCE ANDDISSEMINATE THIS MATERIAL HAS

BEEN GRANTED BY

TO THE EDUCATIONAL RESOURCESINFORMATION CENTER (ERIC)

U S DEPARTMENT OF EDUCATIONOffice of Educational Research and Improvement

EDUCATIONAL RESOURCES INFORMATIONCENTER (ERIC)

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

0 0 Minor changes have been made to'cr improve reproduction qualityC4C>1-1 Points of view or opinions stated in thisif) document do not necessarily represent

official OERI position or policyV)

R 1E ilt N (7;

i 'kr AStINF

.11 If , i'r 041.61

B TCOPYAVAILABLE

READINGRECOVERY®

INMAINE

1991-94

cffrzie Weport

3

Ti bile f Cont nts' ;z. i'Lly,46PEL 77i kw,-1._ 23'

What Maine Program Participants Say

Program Overview

Program Evaluation Results

Program Implementation

The Reading Recovery Lesson

References

Training Sites in Maine

5

7

12

21

24

29

31

2

Being able to see children's faces when they realize theycan read is the only explanation needed to understand

our dedication to this task.

From left to right: Kathryn Manning, Division of Special Services - Maine Department of Education; PaulaMoore, Director University of Maine Center for Reading Recovery; Marie Clay, founder of Reading Recovery,Governor John McKernan, Jr; Leo G Martin, Maine Commissioner of Education; Robert Cobb, Dean ofthe University of Maine College of Education

The State Report is published by the University of Maine Center for Reading Recovery, Orono, Maine.

Dean of the College of EducationRobert Cobb

University Site CoordinatorRosemary A. Salesi

Trainer of Teacher Leaders and Center DirectorPaula Moore

Teacher Leaders and Adjunct FacultyLaura Cook, Melanie Ellsworth, Margaret Hawkins, Claire Hurd, Patricia Jackman,

Judith Karam, Sandra Lowry, Gael Romei, Nancy Todd

Graduate AssistantJudith Bradshaw Brown

Administrative AssistantSusan Russell

3

5

What Maine Program Participants Say

The effects of Reading Recovery extend farbeyond the children served. In questionnairesadministered to parents, administrators, and

classroom teachers, as well as ReadingRecovery teachers and students, individualreactions to the program were collected.

Parents:"Everyone should have the chance to have a child come home and say, 'I can read!' I'll never forgetthe feeling."

"He was a mad and angry child because he couldn't read. Now he's happy and wants to read a lot."

"My child's school experience was headed into the negative zone and this really helped to turn itaround so that she feels better about herself"

"When I was going to school, they didn't have these programs and I had reading problems which endedup in me dropping out. I believe that without this program, my child too, in years to come would havegot frustrated and dropped out of school. Thank you for this program."

Classroom Teachers:

"We've never had this kind of progress in reading at the first grade level before."

"Children's self confidence increases and they begin to participate more in whole group literacy activi-ties."

Administrators:

"We have been able to take 'hard core' learners with real difficulties in reading (and other multipleareas) and turn them into readers."

"Reading Recovery has provided the students with a program that shows results in a short period oftime. The self-esteem of the students is greatly improved by their success in reading."

Reading Recovery Teachers in Training:

"My view (of reading) has changed a great deal. Perhaps most importantly my understanding of read-ing independence has increased. In my years of teaching, never have I been as successful in gettingkids so independent so rapidly"

"I have much more knowledge of the 'practice' of teaching reading. I realize that it is a lot more thanjust providing a rich repertoire of experiences."

Reading Recovery Children:

"It makes me glad because I get to read to my mother"

5

Program Overview:1.11-trisata

IntroductionReading Recovery is an early intervention programthat helps low-achieving first-grade children tobecome independent readers who need no furtherremediation. Reading Recovery is also a powerfulteacher training model characterized by a robusttheory, a tradition of continued learning for teach-ers, and a strong system of colleague support. It isa unique system intervention that can reduce read-ing failure, increase teacher involvement, and lowerremediation rates.

Reading Recovery operates within entire educationsystems through three key programs:1. intensive daily one-on-one instruction for chil-dren who are at risk of reading failure;2. an in-service program through which educatorsare instructed in proven Reading Recovery tech-niques; and3. a network of Reading Recovery educators andadministrators who monitor program results andprovide support to participating teachers andinstitutions.

Program HistoryReading Recovery was developed by Marie M.Clay who conducted observational research in theMid 1960s that enabled her to design techniquesfor detecting children's early reading and writingdifficulties. In the mid 1970s, she developedReading Recovery procedures with teachers andtested the program in New Zealand. The successof this pilot program led to the nationwide adop-tion of Reading Recovery in New Zealand in theearly 1980s.

In 1984, the success of the program in NewZealand led researchers at the Ohio StateUniversity to introduce Reading Recovery to the

United States. In 1993-94, approximately 9,000Reading Recovery educators in 42 states, theDistrict of Columbia, and 4 Canadian provinceswill provide services for an estimated 60,000 chil-dren in more than 2,000 school districts. In Mainealone, 128 teachers served 858 children in 71towns and cities. It is projected that, in 1994-95,218 teachers will provide services to 1,505 childrenin 117 Maine towns/cities.

Reading Recovery in MaineIn February of 1990, Kathryn Manning,Consultant for the Division of Special Services atthe Maine State Department of Education, orga-nized a group of 26 Maine educators to go to Ohioto see Reading Recovery teacher training in actionand to attend the conference. In that group wereteachers and administrators from Westbrook,Bangor, Wiscasset, and Bethel, as well as otherdistricts. In school year 1990-91, Bangor andWestbrook sent Teacher Leaders Sandra Lowryand Patricia Jackman to be trained in Texas andNew York. Also, in 1991 Paula Moore was sent,by the University of Maine, to New Zealand totrain as a university Trainer of Teacher Leaders.Bangor and Westbrook trained the first 23 teachersin Maine in school year 1991-92. Bethel sent aTeacher Leader to be trained in Ohio in that year.

In 1992-93, Moore trained Teacher Leaders at theUniversity of Maine from Caribou, Howland,Harrington/Machias, Wiscasset, South Portland,Benton, and one for the University of Maine. Fortymore Reading Recovery teachers were trainedthrough Bangor, Westbrook, Bethel, and at a subcenter of the University in Ellsworth. A state steer-ing committee formed and met for the first time in1992. Members of that committee include theReading Recovery Teacher-Leader Trainer/Center

7

Director, Site Coordinators, Teacher Leaders,Reading Recovery teachers, administrators, a parent,and representatives of the Maine State Departmentof Education Division of Special Services, theMaine Facilitator Project of the National DiffusionNetwork, and the University of Maine.

Sixty-five more Reading Recovery teachers weretrained in 1992-93. In December of 1993,Governor John McKernan, Jr., allocated discre-tionary funds to support Reading Recovery train-ing and continuing contact with trained teachers.The state legislature allocated additional funds tohelp support Reading Recovery training and imple-mentation through 1995-96. Kathryn Manning wasselected to receive the National Teacher LeaderAward for her advocacy of Reading Recovery

In the spring of 1994, the University offered pilotcourses for classroom teachers at three sites. Thiscourse came at the request of districts now servedby Reading Recovery who wanted primary teach-ers to learn instructional practices which supportchildren during and after the Reading Recoveryintervention and to improve overall literacy instruc-tion in the primary grades. The Center sponsoredresearch which explored the results of thisapproach and which was used to inform the

Rible 1. Reading Recovery in Maine

Children TrainingServed Sites

Teachers-inTraining

I

1991-92 161 2 23

1992-93 493 4 40

1993-94 858 10 65

I 1994-95 1,505* 10 87

*Projected

instruction in courses offered in fall 1994. TheCenter for Reading Recovery sponsored an insti-tute for classroom teachers and school changeagents with a New Zealand teacher trainer inJanuary 1994.

In 1994-95, Moore is training a Teacher Leader forBelfast. Eighty-seven Reading Recovery teachersare training at 10 sites. Each of the regional train-ing sites is offering a year-long course for K-2classroom teachers to facilitate work with smallgroups. A pilot program to train special educatorsis underway in southern Maine. Dr. Marie Clay,founder of Reading Recovery, visited Maine inOctober 1994 to meet with state stakeholders.During the spring semester, a visiting TeacherLeader from New Zealand will take over theUniversity Teacher Leader duties. During schoolyear 1994-95, Reading Recovery in Maine expectsto serve 57 districts, 117 towns and approximately1,505 children.

Northeast Regional Reading RecoveryConferenceAmong its many professional development activi-ties, Reading Recovery is the primary sponsor ofthe highly successful Northeast Reading RecoveryConference and Reading Recovery Institute. Thisannual conference draws teachers, administrators,and Reading Recovery personnel from throughoutNew England, New York, and Canada. InNovember 1993, Maine hosted the conference inPortland attended by 1,250. Featured speakersincluded Barbara Watson, national director ofReading Recovery in New Zealand, and CarolLyons, US. National Director of Reading Recovery.

Reading Recovery Council of NorthAmericaMaine participates in the Reading RecoveryCouncil of North America. The initiation of thisprofessional organization is considered a milestone

8

in the development of Reading Recovery. In orderto disseminate research and program results, theCouncil founded a new international journal focus-ing on early literacy, Literacy, Teaching andLearning, that will begin publication in 1994.

Program for ChildrenReading Recovery gives children a second chanceto succeed before they enter a cycle of failure.Children are selected for the program based onauthentic measures of assessment and teacher judg-ment. Their regular classroom instruction is thensupplemented with daily, one-on-one lessons.

Reading Recovery lessons are designed to helpchildren to develop strategies for hearing sounds inwords, representing messages, and monitoring andchecking their own reading and writing. Theteacher systematically records the child acting on avariety of texts in each lesson and uses theseobservations to form the basis of the next lesson.Individual instruction continues until the child hasdemonstrated the ability to succeed as an indepen-dent reader, can read at or above the class average,and can continue to learn without later remedialhelp.

Chikiren . . . continue to learn without later

remedial help.

Almost 80% of the children in Maine who havecompleted a Reading Recovery program havebecome independent readers. Initial data fromMaine is consistent with numerous other studieswhich have shown that Reading Recovery helps alarge majority of low-progress readers achieve con-tinued reading success.

Program for EducatorsThe remarkable progress that children make inReading Recovery demonstrates that reading fail-ure is not a foregone conclusion for at-risk stu-dents. The key to success for such children isspecialized teaching that will enable them toimprove quicklybefore they are labeled as fail-ureswithout disrupting their regular classroomcurriculum.

In Reading Recovery the teacher training beginswith a year-long curriculum that integrates theoryand practice and is characterized by intensive inter-action with colleagues. Following the training year,teachers continue to develop professionally throughongoing contact with their colleagues and instruc-tors. Teachers-in-training teach children whilebeing observed by their colleagues and get feed-back on their practice. They reflect on their teach-ing in the light of literacy theory and peer critiqueover an extended period of time. Reading Recoveryteachers-in-training become literacy experts withhighly developed observational skills and a reper-toire of intervention strategies that can be tailoredto meet the individual needs of students.

Reading Recovery as a SystemInterventionAs the scope of the instructional program suggests,Reading Recovery is not a teaching methodologythat can be packaged and delivered through a set ofmaterials, a workshop, or a series of courses.Reading Recovery is even more than a program forchildren and educators. It is a program for schoolsystems that want to impact the educational oppor-tunities for at-risk students. The collaboration ofthe school, the State Department of Education, andthe university promotes change within the systemto impact instruction for all children.

The program is adopted by a Maine school districtor consortia of school districts that have made a

9

9

long-term commitment to early literacy in a region.These Reading Recovery sites send an experiencedteacher to the University of Maine. Following thetraining year, these specially prepared TeacherLeaders return to their home districts and workfull-time teaching children, training ReadingRecovery teachers in a region, and performingother duties related to the operation of a site. InMaine, the regional Reading Recovery teachertraining sites are located in Bangor, Belfast,Benton, Bethel, Caribou, Ellsworth,Harrington/Machias, Howland, Orono, SouthPortland, Westbrook, and Wiscasset.The benefits of incorporating Reading Recoveryextend well beyond the success of individual stu-dents who complete the program. The resultsachieved by the teachers and children involved in

Reading Recovery demonstrate for the entire dis-trict the impact powerful teaching can have on low-progress children. Through interaction withReading Recovery teachers, classroom teachersoften begin to construct new theories about howchildren learntheories that tend to carry overinto classroom instruction.

Districts that have adopted Reading Recovery havethe additional benefit of lower cost for special ser-vices. Reading Recovery has been shown to reducethe rate of retention, special education placements,and remediation beyond first grade. And no time islost delivering the services that will effect thesechanges. At most sites, teachers undergo trainingoutside of regular school hours, and they actuallybegin working with students as the training begins.

The Reading Recovery Network

is a cooperative effort among institutions and educators . . .

UNIVERSITYTrainers of Teacher Leaders

Ma ne Department ofEducation

D'vlsion of SpecialServices

TRAIN

SCHOOL DISTRICTSTeacher Leaders TEACH CHILDREN

SCHOOLSTeachers

. . . that extends throughout Maine.

1 0

1 0

Reading Recovery as a Network ofEducators and InstitutionsInstitutions and educators that have adoptedReading Recovery become part of an extensivenetwork to support early literacy. In 1993-94, TheReading Recovery in Maine network included 57school districts, 215 Reading Recovery teachers, 11Teacher Leaders, a university Teacher Leader'Trainer, and a State Department of Education con-sultant. These individuals and institutions worktogether to preserve the integrity of ReadingRecovery and improve its effectiveness as an earlyintervention program in Maine.

Future Directions in MaineThe implementation of Reading Recovery inMaine has presented some unique problems andopportunities. Reading Recovery personnel fromthroughout the state are actively involved in find-ing solutions to the early literacy and learningchallenges that affect the future success of chil-dren.

With the help of allocations from the state leg-islature, it is hoped that Reading Recovery willbe available to all Maine children who need itby school year 1996-97.

Courses for the classroom teacher, now beingoffered at all sites, will continue to expand tomeet the expressed needs of Reading Recoveryschools.

Courses will be offered to special educatorswho desire training in Reading Recovery tech-niques for hard to teach children.

The Reading Recovery network will focus onfostering strong literacy teams in each ReadingRecovery school to ensure successful imple-mentation of the program.

The Reading Recovery network will worktoward improving the cost-effectiveness ofReading Recovery with regard to the number oflessons per week, when Reading Recoveryinstruction begins, and decision making byschool teams.

11

Program Evaluation Results

IntroductionThe success of Reading Recovery has been care-fully documented since its inception. Pilot studiesin New Zealand and the United States demon-strated that the program empowers children in thelowest 20% of their class with the strategies neces-sary to read at or above grade level in an average offifteen weeks. Follow-up studies in both countriesfurther show that Reading Recovery children con-tinue to read at an average or better level afterreceiving the intervention, reducing the need forlong-term remediation. These results have beenreplicated regionally throughout North America,and they continue to be supported by the work ofthe National Data Evaluation Center at Ohio StateUniversity, which tracks the progress of everyReading Recovery child in the U.S. and Canada.Each state compiles data for documenting state, aswell as national, progress.

Program Evaluation Design andDefinitionsChildren are selected to participate in ReadingRecovery based on teacher judgment and theresults of Clay's Observation Survey. The six tasksthat make up the Observation Survey are also usedto indicate progress over time. Three of the sixtasks are used as dependent measures to documentthe progress of the Reading Recovery children forthe national and state studies. These tasks include:

Writing Vocabulary: Children are asked towrite all the words they know how to write. Thescore on this test is the number of words spelledaccurately. There is no ceiling for the scores onthis measure.

Dictation: Children write a dictated sentence. Inscoring, children are given credit for every sound

they represent correctly, thus indicating the child'sability to analyze the word for sounds. This mea-sure has a ceiling of 37.

Text Reading: Children are told the title of aselection, given a brief, standard introduction, andasked to read text materials in graded levels of dif-ficulty. The child's text reading level indicates thehighest level of text read at 90% or above accuracy.Levels are drawn from a basal reading system thatis not part of Reading Recovery instruction or usedin any first grade classroom. This measure has aceiling of Level 30 (i.e. sixth grade reading level).

Using these measures, Reading Recovery childrenare compared to a random sample of first gradechildren. This year, Maine also charted theprogress of waiting list children to compare withReading Recovery and random sample children.The following terms describe the children in thestudy:

Reading Recovery Program Children: Allchildren who receive 60 or more lessons in theprogram. Sixty lessons is considered the minimumamount of learning time for children to makeprogress. Children having less than sixty lessonsare excluded from the program evaluations as theyhave not had adequate learning opportunity in theprogram.

Discontinued Reading Recovery Children:Those children who successfully met the programcriteria of reading at or above the average of theclass.

Random Sample Children: Those childrenwho were randomly selected from the populationof first grade children. Children who received any

1212

Reading Recovery lessons, those on the waitinglist, and special education students were deletedfrom the sample.

State Random Sample: 526 children from thestate (in 1993-94) were randomly selected, with 5-8 from each Reading Recovery school. This groupprovides a basis for determining an average rangefor comparison as a state Average Band.

Waiting List Children: Children who, at thebeginning of the school year, were judged to be at-risk of reading failure in their schools, but who didnot have the lowest scores on the ObservationSurvey and, therefore, were not selected forReading Recovery As Maine is under implementedat this time (i.e. there are not enough ReadingRecovery teachers to cover the needs of the lowestachieving students in all schools), only the lowestachieving students received Reading RecoveryTherefore, the next lowest achieving students were"waiting" for Reading Recovery services.

Pretest and post-test scores on the ObservationSurvey are used to determine progress for all chil-dren. Data is sent to The Ohio State University foranalysis. Descriptive statistics (e.g., numbers ofchildren discontinued, mean performances forReading Recovery, random sample, and waiting listchildren) are calculated and the results returned tothe University of Maine.

Summary of Maine Results for1991-1993

1991-92. During the first year of Reading Recoveryin Maine, 161 children were served through twosites. Of the 121 children who received 60 or morelessons, 101 (84 percent) successfully completedthe program (discontinued) as independent readers.

1992-93. In the second year of Reading Recovery473 students received services through four sites.

Of the 372 who received full programs, 296 (80percent) were discontinued as successful readers.The proportion of discontinued children whoachieved end-of-year scores equal to or exceedingthe state average band ranged from a low of 87percent for the Writing Vocabulary to a high of 97percent for Writing Dictation. Ninety-three percentof the discontinued children achieved end-of-yearscores on the Text Reading measure equal to orexceeding the average band. The proportion of pro-gram children who achieved end-of-year scoresequal to or exceeding the state average bandranged from a low of 76 percent for Text ReadingLevel to a high of 90 percent for Writing Dictation.Thirty-four percent of the children discontinuedprior to April 1. Students discontinued beforeApril 1 continued to make progress on all mea-sures after the intervention lessons were with-drawn, particularly on Text Reading on which theirmean performance rose from Text Level 12 (1stgrade) to 18 (2nd grade).

Maine Program Evaluation Results for1993-94

The following questions guided the program evalu-ation for 1993-94:

1. What proportion of Reading Recovery programchildren were discontinued?

2. How did Reading Recovery and waiting list stu-dents compare to a group of randomly selectedpeers at the end of first grade?

3. What was the progress from entry through end-of-year testing for children discontinued from theprogram prior to April 1?

13

lable 2 Summary of Observation Survey Scores for Discontinued and Program Reading Recovery Childrenand Waiting List Children

Measure Term Discontinued Program Waiting List

Writing Vocabulary Fall 5.11 4.53 7.97Spring 52.04 48.34 40.15

Dictation Fan 8.14 7 13.2Spring 35.39 34.39 33.0

Text Reading Level Fall 0.78 0.68 1.03Spring 18.4 15.96 12.89

1. What proportion of Reading Recoverychildren were discontinued?In year three of the program in Maine, of the 858

children served, 471 (74%) were discontinued asindependent readers. Table 2 shows beginning andend scores for those children.

2. How did Reading Recovery studentsand waiting list students compare to agroup of randomly selected peers at theend of first grade?Figures 1, 2, and 3, display the growth rate forReading Recovery discontinued and program stu-dents and waiting list students compared to that ofa random sample of their peers on measures ofwriting vocabulary, dictation and text reading.Seventy-eight percent were at or above average lev-els in writing vocabulary (Figure 1), ninety-fivepercent on dictation (Figure 2), and sixty percentin reading (Figure 3), indicating that many in thisgroup made accelerated progress and caught upwith their peers.

At the beginning of the year, the waiting list chil-dren all scored higher than either discontinued orprogram children on all measures.

The random sample comparison illustrated in Figx1, 4 and 3 provides a very rigorous test for ReadingRecovery children because the aven2ge band inMaine is drawn from the middle and upper levelachievement groups. Reading Recovery students,chiklren waiting for Reading Recovery servkes, andspecial education students are excluded fn9m therandom sample.

Though the waiting list children were not the low-est achievers in their first grades, they were never-theless considered at-risk of failing to learn how toread. In many schools where there were not enoughReading Recovery teachers to serve the at-risk firstgraders, waiting list children received alternativesmall group help. However, it is noteworthy that, atthe end of the year, Reading Recovery studentsscored higher on all measures than waiting listchildren. The most meaningful measure of thethree is the text reading level. Waiting list childrenscored at level 12, which is below first grade read-ing level, while program children scored at level15, grade one reading level, and discontinued chil-dren scored at level 18, which is considered to begrade 2 reading level.

14

1 4

Fig. 1: Progress of Total Reading Recovery andWaiting List Chiklren Compared to a Random

; Sample on Vocabukry

60 -

50 -

40 -

30 -

20 -

10 -

0

Sept.

Average Band

May

Discontinued Program Waiting List

Fig. 2: Progress of Tbtal Reading Recovery andWaiting List Children Compared to a RandvmSample on Dictation

40 -

35 -

30 -

25 -

20 -

15

10- ..

5 -

0

Sept.

Average Band

Discontinued Program

May

Waiting List

Fig. 3: Progress of Ibtal Reading Recovery andWaiting List Chiklren Compared to a RandomSample on Text Reading

24 -

22 -

20 -

18 -

16 -

14 -

Average Band

. w12-

10 -

8-6-4 -

2 -

0

Sept.

,

May

Discontinued Program Waiting List

3. What was the progress from entrythrough end-of-year testing for childrendiscontinued from the program prior toApril tA goal of Reading Recovery is to help childrenbuild self-extending systems that allow them tocontinue to learn without extra help. Children whoenter the program early in the first grade year arelikely to be released mid-year and then areexpected to make progress through participation inregular classroom instruction alone. The extent towhich this goal is reached is indicated by assessingthe progress made from mid-year to end-of-year by

15

15

the group of children who are discontinued duringthe year. Average rate of progress for this group ispresented in Table 3.

Bilk i Progress Rate of Students DiscontinuedPrior to April I.

I Measure Entry Exit End-of-Year

Writing Vocabulary 478 44.07 52.88

I Dictation (Max = 37) 8.09 34.83 35.42

Text Reading Level(Max = 30)

0.83 14.03 20.02

Discontinued children entered the program with anaverage text reading level score of .83, discontin-ued with a score of 14.03, and ended the year withan average reading level of 20.02. To put this inperspective, they entered as nonreaders, discontin-ued at a level considered to be the end of firstgrade, and at the end of first grade year reached alevel equivalent to second grade reading level asusually defined in traditional reading systems.

Data from the 1993-94 project indicated that ReadingRecovery was successfully implemented during itsthird year and that the program has potential for help-ing Maine's at-risk students become successful read-ers in their first year of school. As before, surveys ofclassroom and Reading Recovery teachers, parents,and administrators were very positive. Some of thosecomments are presented at the beginning of thisreport. The expansion of the program also providedan opportunity to identify some of the problems to besolved if Reading Recovery is to achieve its potential.

National Program Evaluation ResultsAs Reading Recovery has grown, the academiccommunity has shown interest in various effects of

the program. Researchers have compared ReadingRecovery with other intervention prOgrams, evalu-ated its cost-effectiveness, and studied its long-termeffects on children. Others have explored suchareas as the success of the teacher training compo-nent and the impact of the program on learningdisabled students. This research, combined withthe data collected each year on children whoreceive the program, provides answers to some ofthe most commonly asked questions about ReadingRecovery.

How do discontinued Reading Recovery studentscompare to their peers at the end of first grade?Reading Recovery students, all of whom begin firstgrade at the bottom of their class, make considerableprogress as a result of the program, especially whencombined with effective classroom instruction.

The first end-of-year study on Reading Recoveryin the United States (Pinnell, Deford, & Lyons,1988) indicated that 73.5 percent of the 136 ran-domly assigned Reading Recovery students werediscontinued from the program. Over 90 percentof the discontinued students were performing at orabove average on four measures of reading abilityat the end of first grade, and more than 70 percentwere performing at or above average on three othermeasures of assessment. At the end of the year, thegain score of the Reading Recovery students on anationally normed standardized test, CaliforniaTest of Basic Skills (CTBS), was 8.6 compared toa score of 2.4 earned by a similar group of ran-domly assigned first graders who had receivedanother form of compensatory education.

Researchers at Texas Woman's University foundthat the 1,789 Reading Recovery students who suc-cessfully completed the program performed at anaverage or better level on three measures of readingand writing ability at the end of their first gradeyear (Askew, Frasier, & Griffin, 1993).

16

Individual Reading Recovery sites documentedsimilar results in their annual reports. The Halifax,Canada (Talwar & Hill, 1993) site reported that inthe spring of 1990, their discontinued ReadingRecovery students read an average text level of 15,compared with an average first grade band of 11-19.At the end of the school year 1991, the discontin-ued Reading Recovery first graders were readingan average text level of 16, compared to an averageband of 11-21, and in 1992, discontinued ReadingRecovery students read at an average level of 16,compared to an average band of 15-22.

In 1992-93 (National Diffusion Network, 1993),83 percent (22,493) of all children in NorthAmerica who had received a complete ReadingRecovery program were discontinued. When com-pared to a random sample of classmates at the endof the year, 85 percent of these students scored ator above the average band range on writing vocab-ulary, 94 percent on dictation, and 83 percent ontext reading.

Are the gains made in Reading Recoverysustained over time?

Research indicates that Reading Recovery studentsnot only become average or better readers in firstgrade, they develop a self-extending learning sys-tem, which enables them to continue learning atleast as quickly as their peers in later grades.

A follow-up study to the Pinnell et al. (1988) studyshowed that students served in Reading Recoverymaintained progress in second, third, and fourthgrades (Pine 11, 1989). Fourth grade ReadingRecovery students demonstrated that they couldaccurately read text at the sixth grade level orabove. Additionally, these children proved to beexcellent spellers, producing spellings on a fifthgrade level spelling test closer to conventional thantheir randomly selected peers.

Smith-Burke, Jaggar, & Ashdown (1993) tested174 second grade children who had successfullycompleted Reading Recovery as first graders in1990-91. Their performance on several measureswas compared to that of a grade level, randomsample of 177 children. The following results high-light the strong residual effects of the program:

Eighty-nine percent of the Reading Recoverychildren scored within or above the averageband on text reading compared to 80 percent ofthe random sample, and 23 percent of theReading Recovery children scored above theaverage band.Ninety-six percent of the Reading Recoverychildren scored at grade two or above, com-pared to 89 percent of the random sample.At the end of second grade, the average ReadingRecovery child was able to read passagesroughly equivalent to fifth grade basal readingmaterial with at least 90 percent accuracy.

How does Reading Recovery compare toother early intervention programs?

Large scale and local investigations demonstratethat Reading Recovery is a particularly effectivemethod to improve the reading acquisition of at-risk children.

A recent study (Pinnell, Lyons, Deford, Bryk, &Seltzer, 1994) compared Reading Recovery withfour other types of early intervention: (1) an indi-vidual tutorial program similar to ReadingRecover, but taught by a teacher with an abbrevi-ated training program; (2) Direct InstructionalSkills Plan (Cooter & Reutzel, 1987), an individ-ual tutorial taught without Reading Recovery byexperienced reading teachers; (3) a small groupintervention taught by trained Reading Recoveryteachers; and (4) a control group, which received astandard federally funded remediation program.

17

1 7

The final report concluded that Reading Recoverychildren performed significantly better than chil-dren from an equivalent control group and thethree other intervention programs. ReadingRecovery was the only group that scored better onall tests, showing long-term improvements in read-ing. At the end of 70 days of instruction, ReadingRecovery children were reading five levels ahead ofchildren who received regular remedial readinglessons. Even though the control group continuedto receive lessons for the rest of the year, ReadingRecovery children were still three reading levelsabove the remedial group average when all chil-dren were tested the following fall.

Another investigation supported the findings ofthis study. Reading Recovery students were com-pared with a group of students similarly at-risk anda reference group comprised of average-perform-ing first graders (Gregory Earl, & O'Donoghue,1993). The Reading Recovery students receiveddaily Reading Recovery lessons plus regular class-room instruction. The comparison group receivedregular classroom instruction, plus necessary inter-vention services (ESL, special education, parentvolunteers, private tutors). The reference groupreceived regular classroom instruction only.

Researchers reported that Reading Recovery stu-dents scored higher than comparison students onend-of-year measures, that the performance ofReading Recovery students improved at a fasterrate than their at-risk peers who did not receiveReading Recovery, and that Reading Recovery stu-dents made significantly greater gains than boththeir average-achieving classmates and the compar-ison group based on results of the WoodcockReading Mastery Test, the MetropolitanAchievement Test, a spelling assessment, and amiscue analysis.

Is Reading Recovery cost-effective?Evidence indicates that Reading Recovery canreduce costs associated with at-risk students by low-ering retention rates and thereby reducing the needfor remediation and special education referrals.

Dyer (1992) found that while Reading Recoveryrequires an initial and ongoing investment, itsimplementation is educationally sound and reducesthe necessity of more commonly used means ofintervention. The study concluded that school dis-tricts implementing the program will realize sig-nificant long-term cost savings through reductionsin grade retentions, remedial Chapter 1 services,and special education placementssavings thatcan more than offset the short-term costs of imple-menting and operating the program.

Using figures from an average school district inMaine to compare Reading Recovery per studentexpenditures to those of other interventions, indica-tions are that investment in Reading Recoverycould save a district up to $39,740 per student.This estimate is based on a student who isassigned to special education over six years com-pared to a student who does not require specialeducation services after receiving a successfulReading Recovery intervention in grade one. SeeTable 4.

ilible 4 Cost Comparison of Interventions in aTypical Reading Recovery District in Maine

Reading Recovery $2,332 per student

I Retentions

I Chapter 1

$4,254 per student I

$1,243 per student$14,000 (six years)

I Special Education $7,012 per student$42,072 (six years)

18

1 8

Researchers have also examined ReadingRecovery's ability to reduce first grade retentions,the need for further remediation, and the numberof students classified as learning disabled, withpositive results:

One study found that the first grade retentionrate in a school district that had implementedReading Recovery dropped from 4.3 percent inthe three years before implementation to 2.9percent four years after system-wide implemen-tation (Lyons & Beaver, in press).

The same study showed that the district reducedits enrollment in learning disabilities classroomsat the end of first grade frbm 1.8 percent of thefirst grade in the three years before full imple-mentation to .64 percent in three years afterimplementation.

Another study documented the experience of adistrict that reduced its first grade retentionssignificantly in the five years following theimplementation of Reading Recovery whichresulted in considerable savings (Lyons, Pinnell,& Deford, 1993).

In Maine, Westbrook Assistant Superintendentof Schools Robert Hall reports that his districthas eliminated two Step-up Program teachers(transitional program) at a savings of approxi-mately $60,000 per year as a result of ReadingRecovery

19

19

Program Implementation

1. Teacher Leader Districts: How TrainingSites DevelopedIt generally takes a school district two years todevelop a Reading Recovery Training site: one yearto have a qualified member of its staff trained as aTeacher Leader at the University of Maine and asecond year to establish a training site and begintraining teachers.

The Application ProcessTo become an approved training site, a school dis-trict (or consortium of districts) begins by applyingto the University of Maine Center for ReadingRecovery to have a'qualified member of its teach-ing staff trained as a Teacher Leader. As part ofthe application process, prospective sites mustsecure financial support within the district andobtain the approval of the district superintendent.

The applying district also selects an administratorin the district to assume administrative responsibil-ities for Reading Recovery This site coordinatoroversees the preparation of the facility, managesthe budget, negotiates contracts, and acts as admin-istrative liaison with the Reading Recovery net-work.

The Training YearApplicants are selected for the program in thespring, and the year-long residency program beginsthe following fall. The Teacher Leader training is agraduate program taken for credit at the Universityof Maine. The program for Teacher Leadersincludes five components:

1. Graduate-level courses consisting of a clinicalpracticum, a seminar in theory and currentresearch, and supervised fieldwork;

2. The daily teaching of four Reading Recoverystudents;

3. Field requirements, including assisting withthe training of Reading Recovery teachers, con-ducting colleague visits to observe other classmembers teaching a Reading Recovery lesson,and visiting other Reading Recovery sites;

4. Preparation for implementing ReadingRecovery in their district; and,

5. Attendance at a number of professional devel-opment activities including the NortheastRegional Literacy Conference and ReadingRecovery Institute.

During the training year, Teacher Leaders workwith their site coordinators to prepare the site forits first year of operation. They inform appropriategroups about the program, prepare the space wherethe teacher training classes will be held, ordermaterials for teacher training, and assist in theselection of appropriate teachers for the trainingclass.

Implementation YearFollowing their training year, Teacher Leaders andsite coordinators work together to maintain the site.Teacher Leaders train new teachers, collect data onchildren served, and prepare an annual site report.They also participate in a variety of continuingcontact events and activities, including nationalconferences and training seminars, in order to fur-ther their own professional development. In subse-quent years, Teacher Leaders visit previouslytrained teachers and conduct continuing contactsessions.

3ehind the GlassExtensive use is made of a one-way glass for traininglessons. Once a week during the training year, twoteachers each work with one of their children individu-ally behind a one-way glass while the rest of the teach-ers-in-training observe from the other side of the glass.Guided by the Teacher Leader, the teachers engage inan intensive discussion of what they are watching.After the lessons, teachers participate in a critique ses-sion. Use of the one-way glass has been proven to beone of the most powerful components of staff develop-ment in Reading Recovery

1

S.*

Teachers-in-training continue to work full-time in theirschool districts as they receive instruction in Reading Recovery procedures. The most commonarrangement during the training year and subsequent years is for the teacher to spend half a day teach-ing Reading Recovery students and the second half in other teaching duties.

2. Teacher Districts: implementingReading Recovery in a SchoolTo implement Reading Recovery at the classroomlevel in districts where the program has beenadopted, qualified teachers enroll in a year-longacademic course taught by a certified TeacherLeader. The courses are offered for graduate creditthrough the University of Maine. Through interac-tive clinical experiences and theoretical studyguided by a Teacher Leader, teachers learn how toimplement all components of a Reading Recoverylesson and to select teaching procedures appropri-ate for individual students.

Implementation ModelsReading Recovery has been implemented in Maineusing a wide variety of models. Each day, ReadingRecovery teachers are required to spend half theday (two and one-half hours minimum) workingone-to-one with children (usually four). Theremainder of the day is assigned to various other

teaching and support functions. Districts havereported using the following configurations forassignments of teachers:

Chapter 1 remedial reading or SpecialEducation teachers spend half of their day inReading Recovery and the other half workingwith individuals or small groups using otherinstructional strategies.

Two teachers share a first grade classroomwhere one teacher teaches the class and theother uses Reading Recovery with individualchildren and then they switch roles for the sec-ond half of the day.

Kindergarten teachers teach one session andthen spend half a day in Reading Recovery.

Migrant education teachers use ReadingRecovery in extended-day sessions.

Half-time teachers are employed as ReadingRecovery teachers.

22

21

The Costs of ImplementationThe costs of adopting Reading Recovery includethose associated with the establishment of a site, aswell as ongoing site maintenance. Start-upexpenses include training fee, materials, andexpenses for the Teacher Leader-in-training; theinstallation of a one-way glass at the new site forteacher training; a portion of the site coordinator'ssalary during the training year; and tuition forcourses. Following the training year; new sites pro-vide funding for Teacher Leader salaries, continu-ing contact for Teacher Leaders, site staff support,and training materials. For specific informationregarding costs, contact the Center for ReadingRecovery at the University of Maine.

The Benefits of ImplementationImplementing Reading Recovery requires a sub-stantial commitment on the part of the district. Theintegrated nature of the instructional programs forchildren and educators, the use of quantitative datato measure the results of the intervention on allchildren served, the strong professional develop-ment modelthese and the other features of theprogram simultaneously ensure its effectivenessand demand a high level of support from partici-pating individuals and districts. This level of sup-port is justified by the accelerated growth achievedby Reading Recovery program children and thetransformation of teachers who become truechange agents in their districts.

Key Elements of Reading Recovery

Reading Recovery is an early intervention program that supports accelerated learning.

Reading Recovery serves the lowest achieving children.

Reading Recovery is effective with diverse populations.

Children develop a self-extending system of learning to read and write.

Student outcomes are sustained over time.

Reading Recovery teachers serve children as part of their training.

Reading Recovery provides continuous professional support for teachers.

All Reading Recovery teachers, Teacher Leaders, and trainers work with children daily.

Program success is directly tied to student performance.

Reading Recovery is cost-effective.

Reading Recovery is a nonprofit program.

23

The Reading Recovery Lesson

Reading Recovery teachers use a battery of sixmeasures called the Observation Survey to selectthe lowest-achieving children in their classrooms(see box). In addition to regular classroom readinginstruction, these children receive daily ReadingRecovery lessons.

The first two weeks of each child's program aredesigned to develop the student's strengths. Thisperiod, referred to as roaming around the known,is comprised of a variety of literature-based activi-ties that build the child's confidence and establisha rapport between teacher and child. The teacheruses this time to learn about the child's ability andbuild a foundation for the individualized lessonsthat will follow.

Each lesson includes six components:

Reading many known stories

Reading a story that was read once the day

before

Writing a storyWorking with magnetic letters

Working with a cut-up sentence

Reading a new book that will be read indepen-

dently the next day

During these reading and writing activities, theteacher provides just enough support to help thechild develop the effective strategies that indepen-dent readers use. This teacher assistance supportsthe process through which children learn to pre-dict, confirm, and understand what they read.Writing opportunities are essential for developingstrategies for hearing sounds in words, representingmessages, and for monitoring and checking theirown reading and writing.

Selecting and Evaluating ReadingRecovery Children

At the beginning of each academic year, children at-risk of reading failure are selected for ReadingRecovery using classroom teacher judgment andresults from the Observation Survey. Looking acrossmeasures, teachers select children who are the lowestachievers. The Observation Survey is also used toevaluate children who receive the program. The fol-lowing six measures comprise the diagnostic tool:

1) Letter Identification: Children are asked toidentify 54 different characters, including upper- andlower-case letters and conventional print for "a"and "g".

2) Word Test: Children are asked to read a list of20 words drawn from the words used most fre-quently in early reading material.

3) Concepts about Print: Children are asked toperform a variety of tasks during a book reading.These tasks, presented in a standard situation, checkon significant concepts about printed language, suchas directionality and concept of word.

4) Writing Vocabulary: Within a 10-minuteperiod, children are asked to write all the words theyknow. The score on this test is the number of wordsspelled accurately.

5) Dictation Test: Testers read a sentence to thechildren, who write the words, indicating their abil-ity to analyze the word for sounds.

6) Text Reading Level: Measures of text readinglevel are obtained by constructing a gradient of textdifficulty, then testing for the highest level read withaccuracy of 90 percent or better. Levels are drawnfrom a basal that is not part of Reading Recoveryinstruction.

24

The framework of a Reading Recovery lessonremains fairly consistent from child to child.However, each lesson is unique. The child andteacher have their own interactions which deter-mine the direction each lesson may take. Theteacher constantly observes the child's reactionsand questions. All will vary based upon thechild's responses. Books to be used in the lessonare chosen specifically with each child in mind.Books are selected from a variety of little booksfrom numerous publishers for their appropriatenessof natural language, meaning, and level ofdifficulty.

At the beginning of each lesson the child readsfamiliar books. These books were introduced inearlier lessons and have been placed in a groupspecifically for the child to read with ease, confi-dence, and fluency. Some problem-solving mayalso occur in this part of the lesson, although the

primary focus is to ensure student success with aminimum of teacher assistance. After the familiarbook, the child reads a book that was read oncethe day before. The teacher keeps a detailedrecord of the child's behavior for use in selectingthe appropriate teaching strategy.

Next the child writes a story. This allows the childthe opportunity to observe the connectedness ofreading and writing. The child writes indepen-dently and is assisted by the teacher in areas whereassistance is needed. The teacher's involvementwill decline as the child becomes more indepen-dent over time. A sentence written by the child iscut up and the child reassembles it using visualinformation and language structure.

Each day the teacher selects a new book to intro-duce to the child at the end of each lesson.

25

24

Working with rooks and StoriesReading Recovery students typically work with anentire book or a complete story, rather than withunconnected sentences or word lists. By readingand writing continuous texts, children learn to usemany different aspects of printincluding letters,words, sentences, and picturesto understandcomplete stories just as successful readers do.

Accelerated LearningThe goal of Reading Recovery is accelerated learn-ing. Children are expected to make faster thanaverage progress so that they can catch up withother children in the class. The majority ofReading Recovery children typically reach an aver-age reading level after 12-16 weeks of dailyinstruction. During this period, they continue towork in the regular classroom for all but 30 min-utes each day.

Work from StrengthsAccelerated learning is possible because ReadingRecovery teachers base their instruction on carefulobservation of what each child already knowsabout reading and writing. This approach createsefficiency, as the individualized instruction thatfollows "will work on these strengths and notwaste time teaching anything already known"(Clay, 1993, p.3).

Independent LearningThe goal of Reading Recovery is not just toimprove the reading and writing ability of children,but to help them learn how to continue improvingon their own so that later remediation is unneces-sary. With the assistance of their ReadingRecovery teacher, children learn the strategies thatgood readers use. Reading Recovery instructioncontinues until the child has a self-extending sys-tem for literacy learning.

Thirty Minutes of ReadingRecovery ...

Readirvg Familiar BooksThe child is able to read an entire book,exhibiting behaviors indicative of good readers.The teacher supports those behaviors throughappropriate and well-chosen questions orprompts.

2. Reading a Book Thai. Wee ReadOnce Me Day Beare

The child reads the new book from the previ-ous lesson independently while the teachernotes reading behaviors. The teacher recordsimportant information to be used in makinginstructional decisions, selecting teachingpoints to be used after the reading.

3. Magna& Legter WorkThe child learns how to discriminate and dis-tinguish between letters and how words andword parts work.

Wrifing a MaryThe child composes a story about a book reador a personal experience. Through joint prob-lem-solving, the child and teacher worktogether to write the story. The child writes asindependently as possible.

26

25

5. Working with a Cut-up SentenceAfter writing the story, one of the sentences iswritten on a sentence strip and cut up. Thechild uses knowledge of the sentence to searchand monitor for cues while reassembling thestory

6. Reading a New BookThe teacher introduces a new book, providing aframework for the meaning and language struc-tures the child will meet. This book shouldoffer a little more challenge than previousbooks read in the lesson, but be well within thechild's reach.

27

2 6

References

Askew, B., Frasier, D., & Griffin, M. (1993).Reading Recovery report 1992-93 (Tech. Rep.No. 4). Denton Texas: Texas Woman'sUniversity.

Clay, Marie M. (1993). An observation survey.Auckland, New Zealand: HeinemannEducational Books.

Dyer, PC. (1992). Reading Recovery: A cost-effectiveness and educational outcomes analysis.Spectrum: Journal of Research in Education,10(1), 110-119.

Gregory D., Earl, L., & O'Donoghue, M. (1993).A study of Reading Recovery in Scarborough:1990-1992. Annual Site Report of theScarborough School District. Ontario:Scarborough School District.

Lyons, C.A., & Beaver, J. (in press). Reducingretention and learning disability placementthrough Reading Recovery: An educationallysound, cost-effective choice. In R. Allington &S. Walmsley (eds.), No Quick Fix. New York:Teachers College Press.

National Diffusion Network. (1993). 1992-93Discontinuation data (Research Rep.).Columbus: Reading Recovery National DataEvaluation Center.

Pinnell, G.S. (1989). Reading Recovery: Helping at-risk children learn to read. The ElementarySchool Journal, 90(2), 159-181.

Pinnell, G.S., Deford, D.E., and Lyons, C.A.(1988). Reading Recovery: Early intervention forat-risk first graders. Arlington, VA: EducationalResearch Service.

Pinnell, G.S., Lyons, C.A., Deford, D.E., Bryk,A.S., & Seltzer, M. (1994). Comparing instruc-tional models for the literacy education of high-risk first graders. Reading Research Quarterly,29(1), 9-38.

Smith-Burke, M.T, Jaggar, A., & Ashdown, J.(1993). New York University Reading Recoveryproject: 1992 follow-up study of second graders(Research Rep.). New York: New YorkUniversity.

Talwar, J., & Hill, S. (1993). Interim report.Reading Recovery: Halifax, 1989-1992 (Tech.Rep.). Halifax, Nova Scotia: Halifax DistrictSchool Board.

292 7

Reading Recovery Training Sites in Maine

,BANGOR

Robert Ervin, Site Coordinator945-4400Sandra Lowry, Teacher Leader945-4844

BELFAST

Sally Lec lair, Site Coordinator338-4020Rebecca Mailloux, Teacher Leader338-1858

BENTON

Susan Giorgetti, Site Coordinator453-4941Judith Karam, Teacher Leader453-4248

BETHEL

Ann Holt, Site Coordinator824-2582Melanie Ellsworth, Teacher Leader665-2227

CARIBOU

Arthur Benner, Site Coordinator496-6311Nancy Todd, Teacher Leader493-4250

ELLSWORTH

University Center for Reading Recovery581-2438

HARRINGTON/MACHIAS

Ronald Ramsay, Site Coordinator483-6681Gael Romei, Teacher Leader483-2920

HOWLAND

Ed Paul, Site Coordinator732-4141Laura Cook, Teacher Leader732-4141

ORONO

University Center for Reading Recovery581-2438Rosemary A. Salesi, Site CoordinatorPaula Moore, Teacher Leader Trainer

SOUTH PORTLAND

Cheryl Jensen, Site Coordinator799-4845Margaret Hawkins, Teacher Leader879-7122

WESTBROOK

Robert Hall, Site Coordinator854-0800Patricia Jackman, Teacher Leader854-0847

WISCASSET

Roy Bishop, Site Coordinator882-6298Claire Hurd, Teacher Leader882-7585

31

University of MaineCenter for Reading Recovery

Orono, ME 04469

32

29

1:1

U.S. Department of EducationOffice of Educational Research and Improvement (OERI)

National Library of Education (NLE)

Educational Resources Information Center (ERIC)

NOTICE

Reproduction Basis

IC

This document is covered by a signed "Reproduction Release(Blanket)" form (on file within the ERIC system), encompassing allor classes of documents from its source organization and, therefore,does not require a "Specific Document" Release form.

This document is Federally-funded, or carries its own permission toreproduce, or is otherwise in the public domain and, therefore, maybe reproduced by ERIC without a signed Reproduction Release form(either "Specific Document" or "Blanket").

EFF-089 (3/2000)