Education and Training in Developmental Disabilities and Training in Developmental Disabilities The...

Education andTraininginDevelopmentalDisabilities

Focusing on individuals withcognitive disabilities/mental retardation, autism, and related disabilities

Volume 43 Number 1 March 2008

March

2008E

ducation

andTraining

inD

evelopm

entalDisab

ilitiesV

ol.43,

No.

1,p

p.

1–132

Education and Training in Developmental DisabilitiesThe Journal of the Division on Developmental Disabilities,The Council for Exceptional Children

Editor: Stanley H. ZuckerArizona State University

Editorial Assistant: Hannah H. HainlineArizona State University

Consulting EditorsMartin AgranReuben AltmanPhillip J. BelfioreSharon Borthwick-DuffyMichael P. BradyFredda BrownMary Lynne CalhounSharon F. CramerCaroline DunnLise Fox

David L. GastHerbert GoldsteinRobert HendersonCarolyn HughesLarry K. IrvinJames V. KahnH. Earle KnowltonBarry W. LavayRena LewisKathleen J. Marshall

John McDonnellGale M. MorrisonGabriel A. NardiJohn NietupskiJames R. PattonEdward A. PollowayThomas G. RobertsRobert S. RuedaDiane L. RyndakEdward J. Sabornie

Laurence R. SargentGary M. SassoTom E. C. SmithScott SparksFred SpoonerRobert StoddenKeith StoreyDavid L. WestlingJohn J. WheelerMark Wolery

Education and Training in Developmental Disabilities is sent to all members of the Division on Developmental Disabilities of The Councilfor Exceptional Children. All Division members must first be members of The Council for Exceptional Children. Division membership duesare $25.00 for regular members and $13.00 for full time students. Membership is on a yearly basis. All inquiries concerning membership,subscription, advertising, etc. should be sent to the Division on Developmental Disabilities, 1110 North Glebe Road, Arlington, VA 22201.Advertising rates are available upon request.

Manuscripts should be typed, double spaced, and sent (five copies) to the Editor: Stanley H. Zucker, Special Education Program, Box872011, Arizona State University, Tempe, AZ 85287-2011. Each manuscript should have a cover sheet that gives the names, affiliations, andcomplete addresses of all authors.

Editing policies are based on the Publication Manual, the American Psychological Association, 2001 revision. Additional information isprovided on the inside back cover. Any signed article is the personal expression of the author; likewise, any advertisement is the responsibilityof the advertiser. Neither necessarily carries Division endorsement unless specifically set forth by adopted resolution.

Education and Training in Developmental Disabilities is abstracted and indexed in Psychological Abstracts, PsycINFO, e-psyche, Abstractsfor Social Workers, International Journal of Rehabilitation Research, Current Contents/Social and Behavioral Sciences, Excerpta Medica,Social Sciences Citation Index, Adolescent Mental Health Abstracts, Educational Administration Abstracts, Educational Research Abstracts,and Language and Language Behavior Abstracts. Additionally, it is annotated and indexed by the ERIC Clearinghouse on Handicapped andGifted Children for publication in the monthly print index Current Index to Journals in Education and the quarterly index, Exceptional ChildEducation Resources.

Education and Training in Developmental Disabilities Vol. 43, No. 1, March 2008, Copyright 2008 by the Division on DevelopmentalDisabilities, The Council for Exceptional Children.

Division on Developmental DisabilitiesThe Council for Exceptional Children

Board of DirectorsOfficers

Past President Phil ParettePresident Polly ParrishPresident-Elect J. David SmithVice President Emily BouckSecretary Toni MerfeldTreasurer Amanda Boutot

Members

Leslie BrounLinda LazNikki MurdickAngie Stone-MacDonald (Student Governor)Dianne ZagerDeborah Wichmanowski

Executive Director

Tom E. C. Smith

Publications Chair

Jack Hourcade

Communications Chair

Darlene Perner

The purposes of this organization shall be to advance the education and welfare of persons with developmental disabilities, research in theeducation of persons with developmental disabilities, competency of educators in this field, public understanding of developmental disabilities,and legislation needed to help accomplish these goals. The Division shall encourage and promote professional growth, research, and thedissemination and utilization of research findings.

EDUCATION AND TRAINING IN DEVELOPMENTAL DISABILITIES (ISSN 1547-0350) (USPS 0168-5000) is published quarterly inMarch, June, September, and December, by The Council for Exceptional Children, Division on Developmental Disabilities, 1110 NorthGlebe Road, Arlington, Virginia 22201-5704. Members’ dues to The Council for Exceptional Children Division on DevelopmentalDisabilities include $8.00 for subscription to EDUCATION AND TRAINING IN DEVELOPMENTAL DISABILITIES. Subscription toEDUCATION AND TRAINING IN DEVELOPMENTAL DISABILITIES is available without membership; Individual—U.S. $40.00 peryear; Canada, PUAS, and all other countries $44.00; Institutions—U.S. $175.00 per year; Canada, PUAS, and all other countries$179.50; single copy price is $25.00. U.S. Periodicals postage is paid at Arlington, Virginia 22204 and additional mailing offices.

POSTMASTERS: Send address changes to EDUCATION AND TRAINING IN DEVELOPMENTAL DISABILITIES, 1110 North GlebeRoad, Arlington, Virginia 22201-5704.

Education and Training in DevelopmentalDisabilities

Editorial Policy

Education and Training in Developmental Disabilities focuses on the education andwelfare of persons with developmental disabilities. ETDD invites research andexpository manuscripts and critical review of the literature. Major emphasis is onidentification and assessment, educational programming, characteristics, trainingof instructional personnel, habilitation, prevention, community understanding andprovisions, and legislation.

Each manuscript is evaluated anonymously by three reviewers. Criteria for ac-ceptance include the following: relevance, reader interest, quality, applicability,contribution to the field, and economy and smoothness of expression. The reviewprocess requires two to four months.

Viewpoints expressed are those of the authors and do not necessarily conform topositions of the editors or of the officers of the Division.

Submission of Manuscripts

1. Manuscript submission is a representation that the manuscript is the author’sown work, has not been published, and is not currently under consideration forpublication elsewhere.

2. Manuscripts must be prepared according to the recommendations in the Pub-lication Manual of the American Psychological Association (Fifth Edition, 2001).Standard typewriter type, laser, or high density dot printing are acceptable.

3. Each manuscript must have a cover sheet giving the names and affiliations of allauthors and the address of the principal author.

4. Graphs and figures should be originals or sharp, high quality photographicprints suitable, if necessary, for a 50% reduction in size.

5. Five copies of the manuscript along with a transmittal letter should be sent to theEditor: Stanley H. Zucker, Special Education Program, Box 872011, ArizonaState University, Tempe, AZ 85287-2011.

6. Upon receipt, each manuscript will be screened by the editor. Appropriatemanuscripts will then be sent to consulting editors. Principal authors will receivenotification of receipt of manuscript.

7. The Editor reserves the right to make minor editorial changes which do notmaterially affect the meaning of the text.

8. Manuscripts are the property of ETDD for a minimum period of six months. Allarticles accepted for publication are copyrighted in the name of the Division onDevelopmental Disabilities.

Guardianship: Its Role in the Transition Process for Students withDevelopmental Disabilities 3ERIN M. PAYNE-CHRISTIANSEN and PATRICIA L. SITLINGTON

Inclusive High School Service Learning Programs: Methods for and Barriersto Including Students with Disabilities 20STACY K. DYMOND, ADELLE RENZAGLIA, and EUL JUNG CHUN

Using Pivotal Response Training with Peers in Special Education to FacilitatePlay in Two Children with Autism 37LAURA R. KUHN, AMY E. BODKIN, SANDRA D. DEVLIN and R. ANTHONY DOGGETT

Effects of Perspective Sentences in Social StoriesTM on Improving theAdaptive Behaviors of Students with Autism Spectrum Disorders and RelatedDisabilities 46SHINGO OKADA, YOSHIHISA OHTAKE, and MASAFUMI YANAGIHARA

Effectiveness of the Picture Exchange Communication System as a FunctionalCommunication Intervention for Individuals with Autism Spectrum Disorders:A Practice-Based Research Synthesis 61KAI-CHIEN TIEN

Preschool Teacher Perceptions of Assistive Technology and ProfessionalDevelopment Responses 77JULIA B. STONER, HOWARD P. PARETTE, EMILY H. WATTS, BRIAN W. WOJCIK, andTINA FOGAL

Effects of the STAR Intervention Program on Interactions between Camperswith and without Disabilities during Inclusive Summer Day Camp Activities 92CHRISTINA M. BOYD, JEFFREY L. FRAIMAN, KELLY A. HAWKINS, JENNIFER M. LABIN,MARY BETH SUTTER, and MEGHAN R. WAHL

Use of a Handheld Prompting System to Transition Independently ThroughVocational Tasks for Students with Moderate and Severe IntellectualDisabilities 102DAVID F. CIHAK, KELBY KESSLER, and PAUL A. ALBERTO

Peer-Implemented Time Delay Procedures on the Acquisition of ChainedTasks by Students with Moderate and Severe Disabilities 111JANET READ GODSEY, JOHN W. SCHUSTER, AMY SHEARER LINGO,BELVA C. COLLINS, and HAROLD L. KLEINERT

States’ Eligibility Guidelines for Mental Retardation: An Update andConsideration of Part Scores and Unreliability of IQs 123RENEE BERGERON, RANDY G. FLOYD, and ELIZABETH I. SHANDS

Manuscripts Accepted for Future Publication in Education and Training inDevelopmental Disabilities 2

Education and Training in Developmental Disabilities

VOLUME 43 NUMBER 1 MARCH 2008

The Division on Developmental Disabilities retains literary property rights on copyrighted articles. Up to 100copies of the articles in this journal may be reproduced for nonprofit distribution without permission from thepublisher. All other forms of reproduction require permission from the publisher.

Manuscripts Accepted for Future Publication in Educationand Training in Developmental Disabilities

June 2008

Teacher’s perceived efficacy and the inclusion of a pupil with dyslexia or mild mental retardation:Findings from Sweden. Lise Roll-Pettersson, Stockholm Institute of Education, Department ofSpecial Education, PO Box 34103, SE-10026, Stockholm, SWEDEN.

Descriptive analysis of classroom setting events on the social behaviors of children with autismspectrum disorder. Brian A. Boyd, Maureen A. Conroy, Jennifer M. Asmus, Elizabeth L.W.McKenney, and G. Richmond Mancil, University of North Carolina at Chapel Hill, UNC School ofMedicine, Campus Box 7122, Bondurant Hall, Suite 2050, Chapel Hill, NC 27599-7122.

Further conceptualization of treatment acceptability. Stacy L. Carter, 4518 20th St., Lubbock, TX79407.

Child-therapist interaction patterns in ordinary and adaptive toys. Hsieh-Chen Hsieh, Departmentof Occupational Therapy, Fu Jen Catholic University, NO. 510 Jhongjheng Rd., Sinjhuang City,Taipei County, 24205, TAIWAN.

Comparison of assessment results of children with low incidence disabilities. Dennis J. Campbell,AmySue Reilly, and Joan Henley, Department of Leadership and Teacher Education, University ofSouth Alabama, UCOM 3130, Mobile, AL 36688-0002.

Teaching an algebraic equation to high school students with moderate developmental disabilities.Bree A. Jimenez, Diane M. Browder, and Ginevra R. Courtade, Department of Special Education,UNC Charlotte, 6928 Rollingridge Drive, Charlotte, NC 28211.

Do parents prefer special schools for their children with Autism? Javier Moreno, Antonio Aguilera,and David Saldana, Developmental and Ed. Psychology, Universidad D Sevilla, Camilo Jose Celas/n., 41018, Sevilla, SPAIN.

Utah’s alternative assessment: Evidence regarding six aspects of validity. Karen D. Hager andTimothy A. Slocum, Department of Special Education and Rehabilitation Counseling, University ofKentucky, 229 Taylor Education Building, Lexington, KY 40506-0001.

Comparison of PECS and the use of a VOCA: A replication. Ann R. Beck, Julia B. Stoner, andStacey J. Bock, 4100 College of Arts and Sciences, Illinois State University, Normal, IL 61790-4100.

Parent-delivered community-based instruction with simultaneous prompting for teaching commu-nity skills to children with developmental disabilities. Elif Tekin-Iftar, Anadolu University, Engel-liler Arastirma Enstitusu, Eskisehir, 26470, TURKEY.

Address is supplied for author in boldface type.

Guardianship: Its Role in the Transition Process for Studentswith Developmental Disabilities

Erin M. Payne-Christiansen and Patricia L. SitlingtonUniversity of Northern Iowa

Abstract: The purpose of this qualitative study was to explore: (a) the underlying beliefs of those involved indetermining the need for guardianship for young adults with developmental disabilities, (b) the overarchingframeworks or theories that might explain some of the more predominate beliefs, and (c) the relationship oftransition assessment, transition planning, self-determination, and age of majority to the guardianship process.The authors found that planning for guardianship was separated from the transition planning process andthat full guardianship had become the set path for every student in the educational program. The authors madethe following recommendations: (a) schools must begin with the assumption that each individual has thepotential to lead his/her own life--from there supports in areas of need can be developed; (b) both the transitionplanning and guardianship process should be based upon an ongoing assessment of the student’s strengths,needs, preferences, and interests; (c) schools must recognize students as emerging young adults, and preparethem to assume a variety of adult roles by helping them develop and practice self-determination skills; (d) thetransfer of rights at age of majority should be seen as a key point in the transition process; and (e) in workingto prepare students for adult life, instructional and support staff need to be aware of the wide variety ofalternatives to and options within the guardianship process.

Even though guardianship is a profound de-cision with serious implications both for andabout the person labeled as having a develop-mental disability, the concept of guardianshiphas received little emphasis in the literature ofthe special education field. It seems logicalthat the determination of the need for guard-ianship should be made based upon an ongo-ing assessment of the student’s strengths,needs, preferences, and interests, as part ofthe transition planning process. Identifyingthe supports needed by the student as he/shemakes the transition to adulthood should beincorporated into the transition planning pro-cess, so that less intrusive alternatives to guard-ianship may be possible. Training in self-

determination should also provide skills thatwill assist the individual in taking control ofhis/her adult life, and advocate for these sup-ports. The transfer of rights at age of majorityis an ideal time for decisions regarding guard-ianship to be made.

Studies on Guardianship

The majority of research on guardianship hasfocused on the concerns of elderly people(Bulcroft, Kielkopf, & Tripp, 1991; Iris, 1988;O’Sullivan & Hoffman, 1995; Peters, Schmidt,& Miller, 1985). In particular, these studieshave focused mainly on the abuses of anddifficulties in monitoring guardianship (e.g.,Bulcroft et al.; Kritzer, Dicks, & Abrahmson,1993; O’Sullivan & Hoffman; Peters et al.). Anumber of authors have undertaken an anal-ysis of court records of guardianship hearingsfor elderly persons. Analyses of court recordsin Florida (Peters et al.), Ohio and Washing-ton (Bulcroft et al.), Wisconsin (Kritzer et al.),and Maryland (O’Sullivan & Hoffman) iden-tified several similar concerns. These con-cerns included: (a) the questionable validity

The authors express their appreciation to Dr.Deborah J. Gallagher for her active role in thedesign and implementation of this study and toCrystal Stokes for her editorial assistance during thefinal stages of manuscript preparation. Correspon-dence concerning this manuscript should be ad-dressed to: Patricia L. Sitlington, Department ofSpecial Education, Schindler Education Center 154,University of Northern Iowa, Cedar Falls, IA 50614-0601. Email: [email protected]

Education and Training in Developmental Disabilities, 2008, 43(1), 3–19© Division on Developmental Disabilities

Guardianship / 3

of the assessment or rigor used in determiningincompetency; (b) late notifications to allegedwards; (c) lack of participation of (or even thepresence of) the alleged ward; (d) inadequateindependent counsel to serve the allegedward; (e) the frequent assignment of a fullguardian when a partial or limited guardian-ship may have sufficed; and (f) a lack of mon-itoring of annual reports filed by guardians(both low numbers of reports filed and thefailure to notify or sanction guardians whofailed to file an annual report).

Guardianship and individuals with disabilities.Concerns have also been raised regarding theuse of guardianship for individuals with dis-abilities (e.g., Endicott, 1988; Hoyle & Harris,2001; Pepper, 1989), and possible alternativesto guardianship have been proposed(Pierangelo & Crane, 1997; Racino, 1993; Wil-ber, 1991). As Stancliffe, Abery, Springborg,and Elkin (2000) pointed out, “One of thedangers of guardianship is that it can easily gobeyond protecting rights and seriously inter-fere with self-determination if guardians exer-cise control in areas where persons couldmake their own decisions or engage in collab-orative decision-making with support from sig-nificant others” (p. 409).

Stancliff et al. (2000) examined levels ofpersonal control exercised by 76 adults withmental disabilities, as related to their guard-ianship status. They found that individualswith no guardian exercised more personalcontrol over their lives than did those with alimited guardian. Similarly, those with a lim-ited guardian exercised more personal con-trol than participants with a full guardian.These significant differences remained, evenwhen controlling for competency in self-deter-mination.

Millar and Renzaglia (2002) conducted anin-depth analysis of 221 court records ofguardianship hearings for young adults be-tween the ages of 17 and 29 with a disabilitywho were living in one of nine counties inMichigan. The found the following: (a) 120full guardians and 101 partial/limited guard-ians were appointed, but distinctions betweenthe powers of these two guardianship typeswere often found to be minimal; (b) thewards’ primary disability was most often re-ported as “mental impairment”; (c) over 50%of the wards in the sample were 18 years of

age; and (d) over 90% of the young adultswere still in public schools at the time thepetition was filed. In addition, petitions weremost often filed by family members of thewards (74.7%), usually the mother. Motherswere the ones typically appointed as legalguardians.

Millar (2003) extended the findings of Mil-lar and Renzaglia (2002) using the same courtfiles to ask additional questions. She foundthe following reasons for petitions for guard-ianship: to make all decisions (37.1%), ward isnot capable of making informed decisions(33%), specific tasks which the ward is unableto perform (13.5%), to assist the ward withdecision making and specific daily living tasks(16.3%), and none given (1.8%). Millar alsofound the alleged ward was present at thehearing 86.8% of the time. If the individualwere not present, the reason offered was thatattendance would subject the individual toserious physical and/or emotional harm.

In addition, Millar (2003) found that eval-uations tended to use standardized and norm-referenced intelligence tests. In all 221 courtdecisions, the judges stated that “clear andconvincing evidence” was provided and thatthe ward was an individual with a developmen-tal disability and required a guardian. Of the105 full guardians, 88.2 % had appointmentsfor an indefinite term. Of the 101 limitedguardians, 97% had durations of five years,the legal limit in the state. Thirty-three per-cent of the wards indicated no preference asto whom they believed should be appointedguardian; thirty-two percent indicated thesame preference as their petitioner did.

Based upon these findings, Millar (2003)suggested: (a) changes in the way evaluationsare performed, to include evaluation of adultdaily living skills (including decision making)on an ongoing basis throughout the guardian-ship term; (b) education for attorneys andjudges in the area of disability, with an empha-sis on the fact that persons with disabilities canand do lead quality adult lives when given theappropriate support; (c) education for fami-lies and educators related to the guardianshipprocess and ramifications of guardianship im-position, well before the student reaches theage of majority, (d) increased participation ofthe alleged ward in the guardianship hearing;and (e) the selection and monitoring of

4 / Education and Training in Developmental Disabilities-March 2008

guardians who are knowledgeable about com-munity resources, housing options, account-ing, and public benefits for their wards.

Within almost all of the previously discussedstudies there are similar difficulties: (a) prob-lems in the assessment of incompetency, (b)inadequate due process procedures, (c) as-signment of too high a level of guardianship(full when limited would have sufficed), and(d) poor monitoring of guardianships onceassigned. Researchers are unsure, however, asto why these difficulties seem so widespread.

While these studies are interesting and use-ful, they fail to provide insight into the beliefsand attitudes of those involved in the processand how those beliefs and attitudes shapedthe guardianship process. Also missing fromthese studies is the voice and beliefs of indi-viduals with disabilities, those most affected bythe guardianship process. Additionally, nostudies have looked at the impact of theschool system, or specifically, special educa-tion, on guardianship for young adults withdisabilities. Within the schools, transitionplanning, mandated by law, is designed toprepare students for all of the adult roles theywill assume. Discussion about guardianship,with the potential for the removal of rightsand decision-making powers from the individ-ual, should occur within the framework offour closely related concepts—(a) transitionassessment, (b) transition planning, (c) self-determination, and (d) transfer of rights atthe age of majority (Flower, 1994; Hoyle &Harris, 2001).

The Guardianship Process

To fully understand the results of this study, itis important to understand the guardianshipand transition processes.

Our State’s Guardianship Process

Guardianship, including the process, termsand definitions used, varies by state. In thestate in which this study was conducted, aguardian is defined as a “person appointed bythe court to have the custody of the person ofthe ward” (Iowa Administrative Code;IAC§633.3(20)). A conservator is defined as a“person appointed by the court to have cus-tody and care of the property of a ward”

(IAC§633.3(23)). “Ward” refers to the individ-ual who has been assigned a guardian. Addi-tionally, there are two types of guardianships –full guardianship and partial guardianship.Full guardians have all rights allowed by lawover their ward, while partial guardians havespecific rights over their ward as assigned bythe judge during the court hearing(IAC§633). We will first review our state’s legalprocess regarding guardianship, followed bythe process regarding conservatorship.

Legal process for guardianship. The legalprocess for obtaining guardianship in thestate takes the following steps. First, an indi-vidual files a petition for guardianship(IAC§633.552). The petitioner (who is notnecessarily the proposed guardian) can beanyone. The petitioner files a document thatlists the alleged ward and his/her contact in-formation, as well as the proposed guardianand his/her contact information. (“Allegedward” is the legal term used in our state torefer to a person who has not yet been as-signed a guardian but for whom a guardian-ship petition has been filed.)

After a petition is filed, notice is sent to thealleged ward (IAC§633.554)). The noticemust inform the alleged ward that a petitionhas been filed on his/her behalf requesting aguardian, that he/she is entitled to represen-tation, and of the rights that could possibly betaken away from him/her if assigned a guard-ian. Second, the court determines if the al-leged ward needs a lawyer. The court canassign counsel if the alleged ward is incapableof voicing the need or if the alleged ward isindigent (the attorney fees, unless the allegedward is deemed indigent, are paid for bythe alleged ward) (IAC§633.561 and IAC§-633.673). The lawyer’s obligations are de-scribed in the state’s code (IAC§633.561(4)).He/she must advise the alleged ward of theproceedings, must advise the alleged ward ofhis/her rights, must “personally interview” thealleged ward, and must file a report sayinghe/she has done these three things.

A hearing with a judge is the third step. Theburden of proof is on the petitioner for aninitial assignment (IAC§633.551). The lawyerfor the petitioner must first “prove by clearand convincing evidence” that the allegedward is indeed incompetent (IAC§-633.556(1)). Incompetency is determined by

Guardianship / 5

looking at whether the person has: (a) “a de-cision-making capacity which is so impairedthat the person is unable to care for the per-son’s personal safety or to attend to or providefor necessities for the person such as food,shelter, clothing, or medical care” or (b) “adecision-making capacity which is so impairedthat the person is unable to make, communi-cate, or carry out important decisions con-cerning the person’s financial affairs”(IAC§633.3(23)).

Next, there is a determination as to who willbe the guardian. The only requirements arethat the guardian be a “qualified and suitableperson who is willing to serve in that capacity”(IAC§633.559). The rights that will be sub-sumed by the guardian are then identified.The court is to first consider a limited guard-ianship. There are, however, no specific defi-nitions of “limited guardianship” containedwithin the state’s code. Limited/Partial guard-ians have only certain rights granted. Full/Plenary guardians have all rights allowed aguardian. These rights are divided into twocategories: those that the guardian can doonly with court approval and those that theguardian can do at his/her own discretion(IAC§633.635).

With court approval, the guardian maychange the ward’s residency to a more restric-tive one, arrange for elective surgery or non-emergency medical treatment, and consent tothe withholding/withdrawal of life-sustainingprocedures. Without court approval, theguardian may decide where the ward will live;select his/her educational program; choosethe individual’s clothes, furniture, vehicle andother personal effects; determine what profes-sional care and counseling might be needed;and make any other decisions specified by thecourt in the initial hearing.

Legal process for conservatorship. As men-tioned previously, while guardians have rightsover the person, conservators have rights overthe property. Conservators “have a right to,and shall take, possession of all the real andpersonal property of the ward” (IAC§-633.640). They have the responsibilities ofprotecting, preserving, and investing theward’s assets (IAC§633.641). Powers of theconservator are broken down similarly tothose of a guardian--with and without courtapproval. With court approval, the conserva-

tor may invest funds, execute leases, makepayments (to the ward, to agencies providingservices, to the guardian, to anyone who hascustody of the ward), and carry out any otherduties specified by the court (IAC§633.647).Without court approval, the conservator maycollect income, defend or sue for the ward,sell/transfer personal property, vote in proxyat corporate meetings, and receive propertyfor the ward (IAC§633.646).

Fulfilling the guardianship/conservatorship role.After the hearing and assignment of a guard-ian, the guardian is responsible for filing threetypes of reports (IAC§633.669). The initialreport is to be filed within 60 days. From thenon, reports are to be filed annually. A finalreport must be filed when guardianship isterminated (for any reason). Conservatorshave slightly different requirements, but thesame type and frequency of reports(IAC§633.670 and IAC§633.671).

Hearings may be held to modify or termi-nate guardianship/conservatorship. The bur-den of proof rests on the guardian if he/sherequests termination or modification(IAC§633.551). If the ward requests termina-tion/modification, the ward must first “makea prima facie showing of some decision-mak-ing capacity”; then the burden rests on theguardian to prove that the ward is incompe-tent (IAC§633.551). Changes may includeadding additional restrictions, removing re-strictions, or terminating the guardianship.

Alternatives to Full Guardianship

Because guardianship is a powerful legal con-trol, many alternatives to guardianship areavailable. Often, these alternatives to guard-ianship offer monitoring and assistance to in-dividuals with disabilities, but do not requirethat they define themselves as “incompetent,”nor that they give up their role in the decision-making process.

Typically, these alternatives can be groupedby type of service. These services include assis-tance in financial, educational, vocational,and/or daily living matters. Financial servicesmay consist of: (a) a representative payee, adesignated person who can receive Supple-mental Security Income (SSI) or Social Secu-rity Disability Insurance (SSDI) payments for aperson with a disability to assist the individual


in budgeting and spending; (b) a specialneeds trust, a fund created to insure that aninheritance goes to the person designated (ifthis person has a disability) and the assign-ment of a trustee to help with money manage-ment; (c) joint bank accounts (Pierangelo &Crane, 1997), an account that would allow asecond individual to assist the individual witha disability with banking and budgeting; and(d) daily money management services (Wil-ber, 1991), where a non-profit organizationassists with financial affairs.

Educational services consist of educationalpower of attorney (P. Ehrenman, personalcommunication, November 14, 2001), a doc-ument that allows parents to retain the rightsto make educational decisions beyond the ageof majority. Vocational services typically con-sist of supported employment, a program toprovide on-the-job support to an individualwith a disability through the use of naturalsupports (supports occurring in the environ-ment) or external supports, often using thesupport services of adult providers (Butter-worth, Hagner, Kiernan, & Schalock, 1996).

Supported living services are also available,ranging from living in a group home to one’sown apartment or home, with supports pro-vided by adult service providers, as needed.Some services have multiple functions, includ-ing power of attorney, a document that allowsan individual appointed by the person with adisability to make decisions in one or moreareas of adult life. This role can also be termi-nated by the individual with a disability.

Additionally, within the state’s code, it isspecified that a limited or partial guardianshipshould first be considered (IAC§633.560), be-fore consideration of a full or plenary guard-ianship. Because limited/partial guardianshave fewer rights over the ward and becausethese rights are based on specific, focusedneeds of the ward, as determined in the courthearing, limited/partial guardianships areconsidered to be somewhat less restrictive andare therefore to be considered before fullguardianships.

Transition to Adult Life

The Individuals with Disabilities EducationImprovement Act (IDEA 2004) mandates in-dividualized transition planning for move-

ment to all areas of adult life. The definitionfor transition services in IDEA 2004 is a coor-dinated set of activities for a child with a dis-ability that:

● is designed to be within a results-orientedprocess, that is focused on improving theacademic and functional achievement ofthe child with a disability to facilitate thechild’s movement from school to post-school activities, including postsecondaryeducation; vocational education; integratedemployment (including supported employ-ment); continuing and adult education;adult services; independent living or com-munity participation; and [602(34)(A)]

● is based on the individual child’s needs,taking into account the child’s strengths,preferences and interests. [602(34)(B)]

IDEA 2004 requires that a student’s IEP beupdated annually at age 16 to address thefollowing:

● appropriate measurable postsecondarygoals based upon age appropriate transitionassessments related to training, education,employment, and where appropriate, inde-pendent living skills;

● transition services needed to assist the stu-dent in reaching these goals, including thestudent’s course of study (e.g., career andtechnology education, college preparationcourses)

● a statement that the student has been in-formed of the rights (if any) that will trans-fer to him or her on reaching the age ofmajority--no later than one year beforereaching the age of majority under Statelaw.

The entire thrust behind transition services isthe preparation of young adults with disabili-ties for the variety of emerging adult roles theywill assume, with all the rights and responsi-bilities accorded to an adult. There are fourmajor concepts that are a key part of thetransition planning process: (a) transition as-sessment, (b) transition planning, (c) self-de-termination, and (d) transfer of rights at ageof majority.

Guardianship / 7

Transition Assessment

IDEA 2004 clearly provides a mandate for in-cluding transition assessment in the IEP pro-cess. Transition assessment is an integral partof the educational process for students withdisabilities during the secondary school yearsand serves as the foundation for planning foradult roles. Transition assessment encom-passes age-appropriate methods to assist stu-dents in identifying individual needs,strengths, preferences, and interests and inobtaining information on future living, work,and education environments.

Sitlington, Neubert, Begun, Lombard, andLeconte (2007) defined transition assessmentas follows:

Transition assessment is an ongoing processof collecting information on the student’sstrengths, needs, preferences, and interestsas they relate to the demands of current andfuture living, learning, and working envi-ronments. This process should begin inmiddle school and continue until the stu-dent graduates or exits high school. Infor-mation from this process should be used todrive the IEP and transition planning pro-cess and to develop the Summary of Perfor-mance document detailing the student’s ac-ademic and functional performance andpostsecondary goals. (pp. 2–3)

Transition Planning

As stated in IDEA 2004, transition planning isindividualized and is based upon the “stu-dent’s needs, taking into account the stu-dent’s strengths, preferences, and interests”(20 U.S.C § 1401 (30)(B)). A focus on prepa-ration for and transition to a successful adultlife can be seen throughout transition plan-ning and education, which should begin assoon as the student enters the school system(Sitlington & Clark, 2006).

The skill and knowledge domains under theconcept of transition are broad. Sitlingtonand Clark (2006) identified the following nineskill and knowledge domains: communicationand academic performance skills; self-deter-mination; interpersonal relationship skills; in-tegrated community participation skills;health and fitness skills; independent/ inter-dependent daily living skills; leisure and rec-

reation skills; employment skills; and furthereducation and training skills.

Transition strives to prepare students forthe wide variety of adult roles they may as-sume. Therefore, the goal or outcome of theprocess of transition should be a high qualityof life (Halpern, 1993). This long-term plan-ning process brings together the student, fam-ily, school, outside agencies, and others toplan for and to prepare students for the vari-ety of roles they may assume as adults, includ-ing such roles as worker, student, parent,friend, and citizen.

Within this broad focus on transition, thereare many issues to consider. In terms of assum-ing adult roles, it is important for the studentto be prepared to accept the responsibilitiesthat come with adulthood. These responsibil-ities include the demands upon individuals toact autonomously and to self-advocate, as wellas to assume the legal rights that are given atthe age of majority.

Self-Determination

In recent years there has been a major focuson the concept of self-determination for allyoung adults with disabilities, but particularlyindividuals with developmental disabilities.The term “self-determination” attempts to en-compass the themes of choice, control, andpersonally meaningful success for individualswith disabilities (Field, Martin, Miller, Ward,& Wehmeyer, 1998b). For students to transi-tion successfully, it is critical for them to de-velop the skills considered under the conceptof self-determination. However, studies focus-ing on the self-determination of individualswith disabilities indicate that adults with dis-abilities, particularly those with developmen-tal disabilities, have low levels of self-determi-nation (Wehmeyer & Metzler, 1995).

Many authors have defined self-determina-tion. Field, Martin, Miller, Ward, and Wehm-eyer (1998a) summarized these definitions asfollows:

Self-determination is a combination ofskills, knowledge, and beliefs that enable aperson to engage in goal directed, self-reg-ulated, autonomous behavior. An under-standing of one’s strength and limitationstogether with a belief in oneself as capable


and effective are essential to self-determina-tion. When acting based on these skills andattitudes, individuals have greater ability totake control of their lives and assume therole of successful adults. (p. 2)

Research is steadily mounting suggesting thatenhanced self-determination may play a rolein improving adult outcomes for students withdisabilities, including employment status (We-hmeyer & Palmer, 2003; Wehmeyer &Schwartz, 1997) and participation in postsec-ondary education (Field, Sarver, & Shaw,2003). As a result, promoting students’ self-determination is now an important compo-nent of recommended practices in the transi-tion of youth with disabilities to adult life(Council for Exceptional Children, 2003;Field & Hoffman, 2002; Field et al., 1998b).

Models act to guide the development ofcurricula and instructional strategies, thus act-ing as an overarching framework for moreconcrete instruction and facilitation of self-determination (Field, 1996). Several profes-sionals have developed models outlining in-struction in self-determination (Abery, 1994;Field & Hoffman, 1994; 1996; Halpern, Herr,Doren, & Wolf, 2000; Wehmeyer, 1992; 1997;Wehmeyer, Palmer, Agran, Mithaug, & Mar-tin, 2000). Most recently, Test, Fowler, Wood,Brewer, and Eddy (2005) developed a concep-tual framework of self-advocacy, a conceptclosely associated with self-determination.This framework, based upon a comprehensivereview of the literature, included the compo-nents of knowledge of self, knowledge ofrights, communication, and leadership.

Wehmeyer and Schalock (2001) also of-fered what they termed “essential characteris-tics” of self-determined behavior: (a) choice-making skills; (b) decision-making skills; (c)problem-solving skills; (d) goal-setting and at-tainment skills; (e) independence, risk-taking,and safety skills; (f) self-observation, evalua-tion, and reinforcement skills; (g) self-instruc-tion skills; (h) self-advocacy and leadershipskills; (i) internal locus of control; (j) positiveattributions of efficacy and outcome expect-ancy; (k) self-awareness; and (l) self-knowl-edge.

We will refer to self-determination as anindividual’s autonomy and voice. While thismay indeed simplify the concept when com-

pared to other authors’ definitions, it clarifiesthe meaning and allows for discussion of theneed for self-determination for all people, in-cluding those with developmental disabilities.

As transition planning occurs, it is necessaryto attempt to develop and nurture an individ-ual’s level of self-determination so that he/sheis able to fulfill adult roles, exercising one’sautonomy and voicing one’s opinions. Stu-dents must be taught and given the opportu-nity to exercise skills related to self-determina-tion (Wehmeyer & Schalock, 2001). Thispreparation for the assumption of adult rolesmust be planned for through the transitionprocess. Within this process, the transfer ofrights at the age of majority, as well as therelated need to act in a self-determining way,need to be addressed.

Age of Majority and Transfer of Rights

The term age of majority refers to “the age atwhich an individual is no longer considered aminor and, as such, becomes legally able toexercise rights accorded to adults in that stateor province” (Lindsey, Wehmeyer, Guy, &Martin, 2001, p. 3). The legal rights attainedat the age of majority would allow individuals,including those with developmental disabili-ties, to express their voice, to have power withthat voice, and to have autonomy. In essence,this transfer of rights is a necessary part oftransition planning as youth assume adultroles and act in a self-determining way.

Thirty-three of the forty states that re-sponded to a survey conducted by the Na-tional Association of State Directors of SpecialEducation (NASDE, 1999) indicated that theage of majority in their state, which is whenthe transfer of rights occurs, was 18. IDEA2004 requires that students with disabilitiesand their parents are made aware of the trans-fer of rights at least a year before the studentreaches the age of majority. Individuals whohave reached the age of majority, regardless oftheir disability label, are considered to havethe rights accorded to an adult in that state,unless the individual has been declared legallyincompetent (Lindsey et al., 2001).

In their position statement on the age ofmajority and individuals with mental disabili-ties, the Council of Exceptional Children’sDivision on Mental Retardation and Develop-

Guardianship / 9

ment Disabilities cited their concern that thediscussion regarding age of majority requiredby IDEA may “lead to a circumstance whereparents and family members will feel com-pelled to obtain guardianship or other legaldecision-making status over their son ordaughter when they might not otherwise doso” (Lindsey et al., 2001, p. 13). They alsoclarified that the way in which the school han-dles this notification of the transfer of rightswill affect this possibility.

The legal basis for the personal choice andcontrol emphasized within the concept of self-determination comes through the transfer ofrights that occurs at the age of majority.“Transfer of rights” refers to the shift of re-sponsibility from parent to child upon attain-ing the age of majority. This transfer of rightsoccurs for young adults with disabilities just asit occurs for those without disabilities. Uponattaining the age of majority, the individual isviewed in the eyes of the law as an adult,capable of making and responsible for his/her own decisions (Lindsey et al., 2001).

Transition and Guardianship

To assure that students are prepared to as-sume the diverse duties of adulthood, carefuland thorough transition planning, with abroad-based focus, is essential. For studentswith disabilities, this transition planning ismandated. Guardianship, which may beabused, can work against the goals of transi-tion planning. In attempting to understandhow guardianship and transition are related,it is important to understand that, for studentswith disabilities, the transition planning pro-cess undergirds the movement from school toadult life. A seminal time or moment in thistransition is at the age of majority when thetransfer of rights occurs. It is essential thatongoing transition planning has led up to thismoment. Part of this transition process is thedevelopment of such life-long skills as self-determination, which provides students withvoice and autonomy. As mandated by IDEA2004, the transition planning process must bebased upon transition assessment.

A major part of this transition is the devel-opment of needed supports to allow studentsto act as self-determining adults. These sup-ports will be individual to each student, and,

for some, might include a form of guardian-ship or an alternative to guardianship. It ishere that the connections between guardian-ship and transition are established. Each indi-vidual student will have a variety of needsupon entering the adult world.

The supports in place to meet these needsmay include natural supports, those availableto all in the adult world, or may be moreformal and planned by those involved in thetransition process. Regardless of the type ofactual support, those that are planned shouldbe as non-intrusive as possible, allowing theindividual with a disability to retain as muchindependence and autonomy as possible (Weh-man, Revell, & Brooke, 2003). For some indi-viduals with developmental disabilities, guard-ianship may be seen as a necessary support.However, before opting to declare a youngadult legally incompetent and removing cer-tain rights, all other options and alternativesshould first be considered.

Purpose of Study

Very little research could be found on theissue of guardianship for young adults withdisabilities, particularly as it relates to the pro-cess of planning for the transition to adult life.In particular, no studies could be located thatfocused on the values, perspectives, and socio-logical frameworks of those directly involvedin the guardianship process - individuals withdisabilities, their families, teachers, supportstaff from intermediate units, and lawyers.The purpose of this study was to explore: (a)the underlying beliefs of those involved indetermining the need for guardianship forone young adult with mental disabilities, (b)the overarching frameworks or theories thatmight explain some of the more predominatebeliefs, and (c) the relationship of transitionassessment, transition planning, self-determi-nation, and age of majority to the guardian-ship process.

Method

In order to understand guardianship, the be-liefs of those involved in the process and therelationship of this process to planning for thetransition to adult life, we undertook a quali-tative study, focusing on the beliefs of those


involved in the guardianship process for oneyoung man.

Selection of Qualitative Methodology

We chose to explore guardianship using aqualitative methodology, because there are somany complex issues imbedded within thetopic. Often, related to these complex issuesare responses that would be considered opti-mal in their social acceptability. In exploringthe topic, we wanted to reach beyond thesemore socially acceptable responses in an at-tempt to understand what ideas and under-standings inform and motivate people whoare involved in the guardianship process. Wewanted to understand the way a small group ofinter-related people view the complex issue ofguardianship.

The letter and the spirit of a law often differsomewhat from the actual practice of the law;we attempted to delve into this issue as well.The spirit of laws pertaining to guardianshipattempt to limit guardianship orders as muchas possible and offer the alleged ward a voice;the letter of the law indicates that the allegedward is to have his/her own council and re-ceive notification of all legal action beingtaken. It was important to understand howpeople negotiate the difference between howa law is put into practice and what the spirit ofa law intends.

Participants

Purposeful sampling (Glesne, 1999) was em-ployed to obtain information from individualswho had been involved in the guardianshipprocess in some way or another. The guard-ianship process is likely to include some or allof the following individuals: those with disabil-ities, their families, teachers, intermediate ed-ucation unit staff (if applicable), and lawyerswho have worked with individuals with devel-opmental disabilities. The intermediate edu-cation unit staff member who is head of aparent education project located a family, theSmiths (a pseudonym), who had recentlygone through the guardianship process withtheir son, Evan, who has a developmental dis-ability. From the Smiths we branched out,speaking with a current teacher, the interme-diate education unit staff who were involved in

the student’s education, and the lawyer whowas employed by the family.

Evan and his family. The Smiths, a familyof four, live in a rural area of a midwesternstate. Larry (Evan’s father) works at a tractorimplement factory in a city about 40 minutesfrom their home; Connie (Evan’s mother)works in the same city’s county courthouse.Katie, 22, and Evan, 20, both live at home withtheir parents.

Important to this study on guardianship,and something that will certainly be asked, isthe question, “Can Evan make it on his own?”This question, which is asked about many in-dividuals, with and without disabilities, is dif-ficult to answer. Evan does indeed have signif-icant problems in all academic areas. He readsand writes poorly. He has well-developed so-cial skills and excellent manners. Though hisopportunities to practice his decision-makingskills have been limited, both by his young ageand by the protective blanket that surroundshim, he is aware of his likes and dislikes andmakes choices regarding his day-to-day lifewith much success. It is our opinion that Evanwill need various supports throughout his life.What these supports should consist of, how-ever, should be determined by Evan, his fam-ily, and those who support him, based uponhis future education, opportunities forgrowth, and personal freedom.

The school and intermediate education unit staff.Evan’s current teacher and two intermediateeducation unit staff, both of whom work pri-marily at Evan’s school, Lakewood, acted asparticipants. Evan’s teacher, Laura Jenson, isin her third year of teaching; she has taughtEvan for the past two years. Before teaching,she worked in a group home setting and cur-rently works part time for another adult ser-vice provider supervising employees with dis-abilities. Bonnie Potter has been a socialworker for the intermediate education unitfor 27 years; she has spent the last 15-16 yearsat Lakewood. John Pitts, the intermediate ed-ucation unit school psychologist, has been aschool psychologist at Lakewood his entirecareer, for 24 years. Before working at Lake-wood, he worked with individuals with severeand profound disabilities in an institution set-ting.

Lakewood is a school run by the intermedi-ate education unit for students with disabili-

Guardianship / 11

ties. Typically, individuals served at Lakewoodtend to have more significant developmentaldisabilities. The school and its grounds houseall needed facilities, including a cafeteria, agym, a swimming pool, accessible bathrooms,a “life skills” area, and classrooms. Many op-portunities are provided for experiences inthe community, including exploration andtraining experiences in community busi-nesses.

The school population is low, allowing forclassrooms of 5-8 students. Students range inage from 2-21 years and are typically groupedin classrooms by age and the perceived sever-ity of their disability. It is not atypical for astudent to spend his/her entire educationalcareer at Lakewood. Almost all school staffknow each student and their family membersby name; the resulting environment is warmand protective. Families of the students placedat Lakewood are very loyal and have lobbiedfor the school’s continued existence. Fromcasual discussion with family members, it ap-pears that families of students feel the schooloffers their students a safe, protected environ-ment where the students are among otherindividuals with developmental disabilities.

The lawyer. Geoff Ryder acted as theSmiths’ lawyer for the guardianship proceed-ings. He has been in practice for the past 25years, and while he has received no specifictraining on working with individuals with dis-abilities, he estimated that he has been in-volved in two to three guardianship proceed-ings a year since he began his career.

Data Collection

Data collection focused on three sources ofinformation: interviews, archival records, andfield observations. All data were collected bythe lead author. Interviews took place with allof the participants identified in the previoussection. A primary interview was scheduled atthe convenience of the participant. Follow-upinterviews to clarify statements or request ad-ditional data were used as necessary. Archivaldata collection focused mainly on court doc-uments and material used by intermediate ed-ucation unit staff; field observations were alsoused to gain a clearer picture.

Interviews. The semi-structured interviewsincluded both close-ended and open-ended

questions. The close-ended questions pro-vided a general overview of participants’ opin-ions, while the open-ended questions alloweda better understanding of the participants’attitudes, beliefs, and sociological frameworks.Interview questions focused on several areas,including general background, familiaritywith the guardianship process, and beliefs.Questions on the participants’ general back-ground focused on demographic informationand interactions with young adults with dis-abilities and young adults with disabilities whohave guardians. Questions on the partici-pants’ involvement with guardianship focusedon their role in the guardianship process andawareness of and attitudes toward the scope ofguardianship powers available. Questions onthe participants’ beliefs focused on beliefsabout the type of young adult with a disabilitywho might need a guardian; beliefs on therole of the age of majority in precipitating theguardianship process; and attitudes towardthe concept of self-determination.

All interviews took place at the convenienceof the participant and each was tape-recordedwith the participant’s permission. Each inter-view was transcribed by the lead author. Allinterviews were coded to identify emergentthemes throughout. Memos and notes werewritten while coding, to help clarify themesand issues identified by the participants.

Archival data and field observations.Through contact with participants, we wereable to identify other individuals who had ob-tained guardianship of a young adult with adisability and who were willing to allow for thereview of court documents of the guardian-ship proceedings. In addition to court docu-ments, we were given other archival data inthe form of informal handouts received by theparticipants from school and intermediate ed-ucation unit staff. The main observations fo-cused on Evan as he interacted with his familyin their home.

Self as Researcher

Because of our interest in this topic, experi-ences, and past reading in this area, webrought to this study some pre-theoretical dis-positions. We were concerned that notenough questions were being asked as peoplelabeled as having a disability were having per-


sonal rights removed through guardianship.Removal of these rights appeared to occurwith relative ease. As a result, we entered thestudy with some skepticism and concern re-garding guardianship. We attempted, deliber-ately, to challenge our beliefs and our cogni-zance. Identification of these concernsallowed us to constantly evaluate and re-eval-uate our interpretation of data, as well as alertreaders to our dispositions.

The Smiths’ Story

We first met the Smith family at their home;Connie opened the door, explaining thatLarry was still out working in the garage butwould drop in shortly. Katie and Evan weresitting in the living room. Katie worked dili-gently, preparing materials for the day care bywhich she is employed. Evan and Connie hadbeen watching the news in the living room.Evan joined us as Connie detailed his history.

Evan was born on November 2nd, five weeksearly, and with serious complications. He wasmoved from the small, local hospital to a largeuniversity-affiliated hospital an hour away.There he was diagnosed with Pierre RobinSyndrome, which is characterized by such fea-tures as a large tongue, cleft pallet, and a smalljaw. Evan underwent many surgeries, had atracheotomy, and had a feeding tube inserted.When he finally was sent home almost a yearlater, he began early intervention with thelocal intermediate education unit, which pro-vides support services. The staff members pro-vided physical therapy for Evan and instruc-tion for his parents. At the age of 2 1/2 hestarted preschool at Lakewood, a specializedschool for children with disabilities adminis-tered through the intermediate educationunit.

Shortly before Evan’s 18th birthday, theSmiths contacted a lawyer in their county ofresidence and obtained guardianship andconservatorship over Evan. Lakewood recom-mends, even stresses, that all parents obtainguardianship and conservatorship over theirsons/daughters with disabilities, so the Smithsopted to do so. Lakewood staff has on handthe names of two lawyers in the county whoroutinely file guardianship papers, but as theSmiths live in a different county, they locateda lawyer on their own. The lawyer they con-

tacted recommended that the family apply forfull guardianship and conservatorship of Evanso that there would be no additional costs tothem if they decided they needed more powerover aspects of his life. The $150 fee was paidand the papers were filed. Then, because thislawyer and county choose to waive the courthearing in cases where the parents of a “spe-cial needs child” are the petitioners and pro-posed guardians/conservators for their son/daughter, the papers allowing Evan’s parentsto become his legal guardian were signed bythe judge with no further discussion. TheSmiths now file annually a brief form onEvan’s whereabouts and well being to main-tain guardianship, as well as a more detailedform accounting all his funds to maintain con-servatorship.

Beliefs, Perspectives, and SociologicalFrameworks of the Participants

After an initial interview with the Smiths, thelead author interviewed Evan’s teacher, inter-mediate education unit staff at Lakewood, andthe Smiths’ lawyer; she then returned for sev-eral interviews with the family. After speakingwith these individuals, transcribing the inter-views, making notes, and reading and re-read-ing the transcriptions and notes, we were ableto identify several important themes recurrentin the data. First and foremost, as a whole,participants demonstrated few reservations orsecond thoughts when concluding that Evanand other students at Lakewood neededguardianship. Second, participants felt theneed to protect Evan and other students andbelieved the way to do this was by pursuingguardianship. Third, participants lackedknowledge regarding alternatives to guardian-ship. Finally, the process in place for guard-ianship planning was separate from the stu-dents’ transition planning. In the followingsections, we elaborate on and substantiate thisassessment.

Preemptive Conclusion of Need for Guardianship

A declaration of incompetence is the first ma-jor step in determining the need for a guard-ian. An implicit statement is made regarding astudent’s competency by advocating for orstating that there is a need for guardianship. It

Guardianship / 13

is very important to look at how the determi-nation of the need for guardianship, or a stu-dent’s competency, are made.

Evan’s placement at a segregated school forstudents with disabilities and his status as anindividual with a disability undoubtedly con-tributed to his being viewed as “incompetent”and therefore in need of guardianship. JohnPitts and Bonnie Potter felt that Evan’s atten-dance at Lakewood, a segregated school forindividuals with disabilities, was indicative ofhis lack of or level of competence.

If school staff, educated in working withstudents with disabilities, feel that placementin a certain school is enough to determinecompetence or need for guardianship, it isnot surprising that other adults, with less train-ing in working with students with disabilities,might draw similar conclusions. Geoff Ryder,the Smiths’ lawyer, also concluded Evan was inneed of guardianship. When asked how hehad determined this, he responded, “Now thisis when we have a special needs child – I relymore upon the parents. That’s because nor-mally they aren’t going to go to the cost andthe expense of this unless necessary.” So, if aparent went to the time and expense to re-quest guardianship for a “special needs” child,Ryder indicated that he would assume theyoung adult was incompetent and proceedwith the guardianship.

Troublingly, Ryder also noted that in otherguardianship cases, he may be called upon toact as a guardian ad litem for an alleged ward.In this case, he would determine competencyand identify what actions he believed to be inthe best interest of the alleged ward. In doingso, he stated that he would contact the allegedward, meet them, and have a conversation. “Alot of times,” he said, “they aren’t able tocommunicate.” In this statement, he was refer-ring to an alleged ward’s ability to communi-cate in a traditional sense, specifically, orally.He tied great importance to the ability tocommunicate orally. In referring to a particu-lar case, he noted that, even though staff doc-tors at a state mental health institution hadlabeled a woman incompetent, he, from talk-ing with her, knew that she “had a certain levelof understanding” and judged her competent.

These assumptions about the need forguardianship or incompetence appeared tobe based on perceptions of disability, the

meaning of a school program placement, andthe label of “special needs.” No individualmentioned specific capabilities of Evan or ofother students. Related to these assumptionsof incompetence is the need to protect indi-viduals viewed in this way.

Paternalism and the Related Push forGuardianship

The Smiths have concerns for Evan, as allparents might for their child. They wonderwhat he will do when he grows up, when hewill leave the house, and if he will be able tomaintain and prosper on his own. Addition-ally, there are other concerns related to Evan’sperceived ability to care for himself. Connie,his mom, admits, though, “it’s going to be, Ithink, a bigger leap for me” to adjust to hisgrowing up and leaving the house. Theschool, though, seems to have used these nat-ural concerns of parents to strongly encour-age the Smiths to obtain guardianship.

When asked why she had pursued guardian-ship, Connie cited Lakewood’s emphasis onacquiring it and the importance they hadplaced on it. She added that she did havesome fears for Evan. When asked to elaborate,she said she was concerned he might be“taken advantage of.” These vague suggestionsof what might occur to a young adult with adisability, if left without a guardian, werenoted in several other interviews. Laura Jen-son, Evan’s teacher, stated the following inregards to what she might tell a parent duringan IEP meeting:

A lot of parents ask, “Why would I want tobe my student’s or my child’s guardian?” Wejust make suggestions, “If you do want to beyour son’s guardian a good reason would beso someone couldn’t come along and takeadvantage of their money,” which is a certainpossibility. . .

Along similar lines, Potter, the school socialworker, provided some concerns regarding astudent’s ability to protect or care for them-selves.

Really, our kids are very gullible and can betaken advantage of easily and that’s wherewe come from with the parents. Our kidscan be talked into something very easily.Unfortunately there are people out there


who are not trustworthy, and when our kidsleave here (Lakewood) they go to otherprograms, you know, or they go to grouphomes, or into the workforce and they canbe talked into something that is illegal orthey can be taken advantage of and if theyare their own guardians, they can sign pa-pers or get involved with something theydon’t understand if somebody talks theminto it and then they are responsible for it ifit’s some illegal activity.

Each participant stressed the need to protectand look after Evan and other young adultswith developmental disabilities. Evan andother students at Lakewood were consideredunable to make decisions and/or choices or, ifthey were allowed to do so, these decisions orchoices would result in negative consequencesdue to poor decision-making capacity. To pro-tect Evan from this, guardianship was needed.While the paternalistic desire to protect andkeep safe may stem from love and concern forEvan and other young adults with disabilities,it acts to hinder his development as an adult.He is not seen as an adult, but as an eternalchild, forever to be looked after.

Lack of Knowledge Regarding Alternatives toGuardianship

No other alternatives to guardianship werediscussed with parents. Staff had no knowl-edge of or was reticent to discuss the legalprocess behind guardianship or of any alter-natives to guardianship. Initially in the inter-view with Jenson, Evan’s teacher, she was veryanxious and kept repeating that she was un-sure she would be able to help us, as she knewvery little about guardianship. Her role in theprocess at Lakewood is to refer students on tothe intermediate education unit staff, theschool psychologist, Pitts, or the school socialworker, Potter. Jenson said she had no knowl-edge of the legal process that guardianshiptakes and that she relied on Pitts and Potter toprovide information and resources to the par-ents of students at Lakewood.

When speaking with Potter, though, she ad-mitted that she also had no knowledge of thelegal process. Pitts also knew little; when askedif he had any idea of the process, he repliedwith a shrug and a shake of the head, “Not

really.” So while Lakewood staff was unfamil-iar with the legal process surrounding guard-ianship, they recommended it without discuss-ing or exploring with parents possiblealternatives to full guardianship. Though nu-merous and substantive, no alternatives werediscussed by staff at Lakewood. It seems thatany alternatives to full guardianship, even lim-ited guardianship, were not discussed. Theseoptions could have been provided as part ofthe transition planning process.

Separation of Planning for Guardianship fromthe Transition Planning Process

Transition planning should be the drivingforce behind the movement from secondaryschool to adult roles. Without adequate tran-sition planning, students leave school unpre-pared, and parents are left in even more un-certainty and worry about their child’s future.At Lakewood, planning for guardianship hasbeen separated from transition planning. Theresult is a push for guardianship for all stu-dents without consideration of other aspectsof their adult life. The resulting situation isproblematic and includes the use of a blanketpolicy towards one particular option (in thiscase, guardianship), the use of the notifica-tion of the transfer of rights as a warning toparents, and a lack of emphasis on or belief inself-determination.

In the first interview with staff working atLakewood, the lead author spoke with LauraJenson, Evan’s classroom teacher. When askedif it is assumed that guardianship is appropri-ate for all the students at Lakewood, she re-sponded, “Well, we leave that up to the par-ents to chose. We just, we just simply ask ifthey’ve started the process and recommendthat they do if they want to.” As you can see,Jenson states that this decision should bemade by the family, but does note that Lake-wood takes a more aggressive path, by recom-mending guardianship. The intermediate ed-ucation unit staff were much more certain ofthe role that Lakewood should play, advocat-ing clearly for any family with a student at theschool to obtain guardianship. Bonnie Potter,the school social worker, made numerousstatements to this effect, including “We feelguardianship is a real necessity for our kidshere in the building.” The school psycholo-

Guardianship / 15

gist, John Pitts, felt similarly. “I would encour-age,” he said, “all parents that have studentshere (at Lakewood) to pursue this, to obtainguardianship so that, so that they don’t runinto any difficulties or problems on down theroad if they don’t.”

The first dilemma is that while Lakewoodstaff, in their interviews, used words like “rec-ommend” and “encourage,” it appears thatthe role they play in this decision is muchstronger than recommending and encourag-ing. Connie Smith, Evan’s mom, went throughthe process. She said that different membersof the school staff mentioned guardianship atevery meeting she had with school staff foralmost two years prior to Evan’s 18th birthday,including parent-teacher conferences, annualreviews, three-year re-evaluations, and meet-ings about the work experience program.Connie noted that even after the family hadobtained guardianship, school staff still veri-fied that she had completed the process.

Lakewood staff also discussed, with Evan’sfamily and with the lead author, their empha-sis on the transfer of rights at the age ofmajority. This tenet was used by staff as amajor push for guardianship. The age of 18,which is the age of majority in our state, be-came the deadline for the completion of theguardianship process. As Laura Jenson, Evan’steacher, stated, “18 years old is when we rec-ommend they have it done, because whenthey turn 18, if it’s not done by 18, that stu-dent is automatically in charge or is guardianof their finances and situations. . .” BonniePotter, the school social worker, discussed theconcept in much the same way, saying thingslike, “Parents don’t realize what it meanswhen kids turn 18.” Pitts, the school psychol-ogist, also shared stories with parents aboutthings that could happen if a student did nothave a guardian at the age of 18, such as howthey could buy a car or could move into theirown apartment. The transfer of rights becamea major reason why parents were encouragedto pursue guardianship. The transfer of rightsand the powers that came with that were dis-cussed in a way that highlighted only the neg-atives of this transfer, leaving the Smiths witha sense of alarm and dread. In essence, thenotification of the transfer of rights was usedas a warning and a further reminder of theneed for guardianship.

The blanket policy towards guardianshipand the questionable use of the transfer ofrights tenet, when coupled with the aforemen-tioned theme of lack of knowledge of alterna-tives, leads to the potential abuse of trust byLakewood staff. Families often have long-standing relationships with the school andtrust the input they receive from school staff.Evan’s family, who have interacted with Lake-wood staff since he was 2 1/2 years, arepleased with the school. Connie felt, in devel-oping Evan’s IEPs, that the school staff knewbest. She stated, “Because they’ve worked withso many kids in the same category as him thatthey can say, ‘well, you know, maybe thiswould be best for Evan.’” Connie and thefamily have been happy with the decisionsmade by Lakewood staff and feel that it is thebest place for Evan. She trusts what staff atLakewood think and relies on the staff forsupport and input in working with her son. IfLakewood staff is not representing the widearray of options available and are advocatingfor one decision, guardianship, there is thepossibility for an abuse of trust to occur.

An underlying problem is that staff at Lake-wood is unable to see their students as beingor becoming self-determined individuals. As aresult, with the guardianship discussion sepa-rated from transition planning and self-deter-mination, we see these decisions being madefor students and families, with their care andprotection in mind, as opposed to with stu-dents and families.

Many of these problems stem from the factthat the issue of guardianship has been re-moved from the concept of self-determinationand the transition process, which brings to-gether the student, family, school, adult ser-vice providers, and others involved with thestudent. Instead, it is a blanket policy, cover-ing all students at Lakewood. Alternatives arenot discussed and are therefore not madeavailable to parents. The wide variety of adultroles that the student may assume is not fullyconsidered alongside the issue of guardian-ship. A consideration of self-determinationand the assumption of the rights that aretransferred at the age of majority are notpresent in the decision regarding guardian-ship.


Discussion and Recommendations

Lakewood’s preemptive conclusion regardingthe need for guardianship for all students isproblematic. The root of the problems sur-rounding the issue of guardianship at Lake-wood is that the planning for guardianship isseparated from the transition process. Theresult is that the student, family, school, adultservice providers, and others are not making afully-informed, well-planned group decision.Discussion of the individual student’sstrengths, needs, preferences, and interests inrelation to the guardianship process has notoccurred. Alternatives and options that maywork for the student have not been discussed.Guardianship has not been considered in con-junction with the assumption of other adultroles. Full guardianship becomes a set pathfor every student, as opposed to a process,such as the transition process that is individu-alized for each student and family accordingto their needs and desires

Based upon our observations, we make thefollowing recommendations related to therole of the guardianship process within thetransition planning process. First, both thetransition planning and guardianship pro-cesses should be based upon an ongoing as-sessment of the student’s strengths, needs,preferences, and interests. This is a requiredcomponent of the transition planning process(IDEA 2004). The transition assessment pro-cess is an ideal vehicle to identify thestrengths, needs, preferences, and interests ofthe individual. The IEP team can then identifythe supports and accommodations needed bythe individual to transition to adult life. TheAssociation for Persons with Severe Handicaps(TASH; 2003) in the TASH Resolution onAlternatives to Guardianship urged “the devel-opment and promotion of the use of accom-modations and supports people need to makechoices and decisions, to have their prefer-ences recognized and honored, and to havetheir rights to self-determination protected.”In this same resolution TASH committed tothe promotion and use of alternatives toguardianship.

Second, schools must recognize students asemerging young adults, and prepare them toassume a variety of adult roles. Third, schoolsshould prepare students for the assumption of

these adult roles by helping students developand practice self-determination skills. Devel-oping students’ autonomy and voice will allowthem to participate in their transition plan-ning to a greater degree and will allow them toadvocate for themselves in the future.

Fourth, the transfer of rights at age of ma-jority should be seen as a key point in thetransition process, rather than a warning, orperhaps even a threat. Lindsey et al. (2001)addressed this concern, offering a reminderthat schools cannot make judgments on thecompetency of students and voicing a concernthat notification of this transfer could leadparents towards guardianship. As they stated,if schools “adopt a philosophy of supportingstudents to become more self-determined andto become meaningful participants in theplanning process” (p. 13), then notificationwill not act as a threat. They also noted thathow schools address this will greatly affect par-ents’ concerns. Millar and Renzaglia (2002)recommended that the IEP team could eventailor the IEP goals and objectives with theaim of preventing the imposition of guardian-ship.

Fifth, in working to prepare students foradult life, instructional and support staff needto be aware of the wide variety of alternativesto and options within guardianship. Knowl-edge of guardianship is necessary for allschool officials, but especially if schools have astrong relationship with parents, and parentstrust the input of school officials, as they do atLakewood. School staff should be familiarwith the legal proceedings and should edu-cate themselves and families regarding possi-ble alternatives; otherwise, full guardianshipbecomes the only option presented to fami-lies. Additionally, those outside the educa-tional arena who interact with individuals withdisabilities (i.e., lawyers) need education inthese same areas.

Finally, the discussion regarding guardian-ship must be intertwined with the transitionprocess, as this will help ensure that students,families, school staff, adult service providers,and others are considering all adult roles andall options to help the individual succeedwithin these varied adult roles. It seems logicalthat the determination of the need for guard-ianship should be made based upon an ongo-ing assessment of the student’s strengths,

Guardianship / 17

needs, preferences, and interests, as part ofthe transition planning process. Identifyingthe supports needed by the student as he/shemakes the transition to adulthood should beincorporated into the transition planning pro-cess, so that less intrusive alternatives to guard-ianship may be possible. Training in self-de-termination should also provide skills that willassist the individual in taking control of his/her adult life, and advocate for these supports.The transfer of rights at age of majority is anideal time for decisions regarding guardian-ship to be made.

References

Abery, B. (1994). A conceptual framework for en-hancing self-determination. In M. F. Hayden &B. H. Abery (Eds.), Challenges for a service system intransition (pp. 345–380). Baltimore: Brookes.

The Association for Persons with Severe Handicaps(TASH). (March, 2003). TASH resolution on al-ternatives to guardianship. Retrieved March 28,2006, from http://www.tash.org/resolutions/res02altguardianship.htm

Bulcroft, K., Kielkopf, M. R., & Tripp, K. (1991).Elderly wards and their legal guardians: Analysisof county probate records in Ohio and Washing-ton. The Gerontologist, 31, 156–164.

Butterworth, J., Hagner, D., Kiernan, W., & Scha-lock, R. (1996). Natural supports in the work-place: Defining an agenda for research and prac-tice. Journal of the Association for Persons with SevereHandicaps, 21, 103–113.

Council for Exceptional Children. (2003). What ev-ery special educator must know: Ethics, standards, andguidelines for special educators (5th ed.). Arlington,VA: Author.

Endicott, O. (1988). Decision-making time onguardianship. Entourage, 3(4), 15–18, 47.

Field, S. (1996). Self-determination instructionalstrategies for youth with learning disabilities. Jour-nal of Learning Disabilities, 29, 40–52.

Field, S., & Hoffman, A. (1994). Development of amodel for self-determination. Career Developmentfor Exceptional Individuals, 17, 159–169.

Field, S., & Hoffman, A. (1996). Steps to self-determi-nation: A curriculum to help adolescents achieve theirgoals. Austin, TX: Pro-Ed.

Field, S., & Hoffman, A., (2002). Preparing youth toexercise self-determination: Quality indicators ofschool environments that promote the acquisi-tion of knowledge, skills, and beliefs related toself-determination. Journal of Disability Policy Stud-ies, 13, 113–118.

Field, S., Martin, J., Miller, R., Ward, M., & Weh-

meyer, M. (1998a). A practical guide for teachingself-determination. Reston, VA: Council for Excep-tional Children.

Field, S., Martin, J., Miller, R., Ward, M., & Wehm-eyer, M. (1998b). Self-determination for personswith disabilities: A position statement of the Divi-sion on Career Development and Transition. Ca-reer Development for Exceptional Individuals, 21, 113–128.

Field, S., Sarver, M. D., & Shaw, S. F. (2003). Self-determination: A key to success in postsecondaryeducation for students with learning disabilities.Remedial and Special Education, 24, 339–349.

Flower, D. (1994). Guardianship and self-determina-tion. Minneapolis, MN: Minneapolis Institute onCommunity Integration. (ERIC Document Re-production No. ED368109)

Glesne, C. (1999). Becoming qualitative researchers:An introduction (2nd ed.). New York: Longman.

Halpern, A. S. (1993). Quality of life as a conceptualframework for evaluating transition outcomes. Ex-ceptional Children, 59, 486–498.

Halpern, A. S., Herr, C. M., Doren, B., & Wolf, N. H.(2000). Next S.T.E.P.: Student transition and educa-tional planning (2nd ed.). Austin, TX: PRO-ED.

Hoyle, D., & Harris, K. (2001). Rethinking guard-ianship. Retrieved November 29, 2001, from TheCenter for Self-Determination web site: http://www.self-determination.com/publications/rethink.guard.htm

Individuals with Disabilities Education Improve-ment Act of 2004, PL 108–446, 118 Stat. 2647.

Iris, M. A. (1988). Guardianship and the elderly: Amulti-perspective view of the decision-makingprocess. The Gerontologist, 28, 39–45.

Iowa Administrative Code (IAC). Iowa Code Section633. Retrieved May 14, 2003, from the Iowa StateLegislature website: http://www.legis.state.ia.us/IACODE/2001/633/1.html

Kritzer, H. M., Dicks, H. M., & Abrahmson, B. J.(1993). Adult guardianships in Wisconsin: How isthe system working? Marquette Law Review, 76,549–575.

Lindsey, P., Wehmeyer, M., Guy, B., & Martin, J.(2001). Age of majority and mental retardation: Aposition statement of the Division on Mental Re-tardation and Developmental Disabilities. Educa-tion and Training in Mental Retardation and Devel-opmental Disabilities, 36, 3–15.

Millar, D. S. (2003). Age of majority, transfer ofrights, and guardianship: Considerations for fam-ilies and educators. Education and Training in De-velopmental Disabilities, 38, 378–397.

Millar, D. S., & Renzaglia, A. (2002). Factors affect-ing guardianship practices for young adults withdisabilities. Exceptional Children, 68, 465–484.

National Association of State Directors of SpecialEducation. (1999). Age of majority. Washington,


DC: OSERS. (ERIC Document Reproduction Ser-vice No. ED 432872)

O’Sullivan, J. L., & Hoffman, D. E. (1995). Theguardianship puzzle: Whatever happened to dueprocess? Maryland Journal of Contemporary Issues, 7,11–51.

Pepper, C. (1989). Guardianship: Friend or foe ofAmerica’s frail and elderly? Journal of Elder Abuseand Neglect, 1, 65–74.

Peters, R., Schmidt, Jr., W. C., & Miller, K. S. (1985).Guardianship of the elderly in Tallahassee, Flor-ida. The Gerontologist, 25, 532–538.

Pierangelo, R., & Crane, R. (1997). Complete guideto special education transition services: Ready-to-usehelp and materials for successful transitions from schoolto adulthood. West Nyack, NY: The Center for Ap-plied Research in Education.

Racino, J. A. (1993). A qualitative study of self advocacyand guardianship: Views from New Hampshire. Syra-cuse, NY: Community and Policy Studies. (ERICDocument Reproduction Service No. ED 374610)

Sitlington, P. L., & Clark, G. M. (2006). Transitioneducation and services for students with disabilities (4th

ed.). Boston: Allyn & Bacon.Sitlington, P. L., Neubert, D. A., Begun, W., Lom-

bard, R. & Leconte, P. (2007). Assess for success: Apractitioner’s handbook on transition assessment (2nd

ed.) Thousand Oaks, CA: Corwin.Stancliffe, R. J., Abery, B. H., Springborg, H., &

Elkin, S. (2000). Substitute decision-making andpersonal control: Implications for self-determina-tion. Mental Retardation, 38, 407–421.

Test, D. W., Fowler, C. H., Wood. W. M., Brewer,D. M., & Eddy. S. (2005). A conceptual frame-work of self-advocacy for students with disabilities.Remedial and Special Education, 26, 43–54.

Wehman, P., Revell, W. G., & Brooke, V. (2003).Competitive employment: Has it become the“first choice” yet? Journal of Disability Policy Studies,14, 163–173.

Wehmeyer, M. L. (1992). Self-determination: Criti-

cal skills for outcome-oriented transition services.The Journal of Vocational Special Needs Education, 15,3–7.

Wehmeyer, M. L. (1997). Self-directed learning andself-determination. In M. Agran (Ed.), Student-directed learning: Teaching self-determination skills(pp. 28–59). Pacific Grove: Brookes/Cole.

Wehmeyer, M. L., & Metzler, C. A. (1995). Howself-determined are people with mental retarda-tion? The National Consumer Survey. Mental Re-tardation, 33, 111–119.

Wehmeyer, M. L., & Palmer, S. B. (2003). Adultoutcomes for students with cognitive disabilitiesthree years after high school: The impact of self-determination. Education and Training in Develop-mental Disabilities, 38, 131–144.

Wehmeyer, M., Palmer, S. Agran, M., Mithaug, D., &Martin, J. (2000). Promoting causal agency: TheSelf-Determined Learning Model of Instruction.Exceptional Children, 66, 439–453.

Wehmeyer, M. L., & Schalock, R. L. (2001). Self-determination and quality of life: Implications forspecial education services and supports. Focus onExceptional Children, 33(8), 1–16.

Wehmeyer, M. L., & Schwartz, M. (1997). Self-de-termination and positive adult outcomes: A fol-low-up study of youth with mental retardation orlearning disabilities. Exceptional Children, 63, 245–255.

Wilber, K. H. (1991). Alternatives to conservator-ship: The role of daily money management ser-vices. The Gerontologist, 31, 150–155.

Zimny, G. H., Gilchrist, B. J., Grossberg, G. T., &Chung, S. (1991). Annual reports by guardiansand conservators to probate courts. Journal of ElderAbuse and Neglect, 3, 61–74.

Received: 23 August 2006Initial Acceptance: 13 October 2006Final Acceptance: 15 February 2007

Guardianship / 19

Inclusive High School Service Learning Programs: Methodsfor and Barriers to Including Students with Disabilities

Stacy K. Dymond, Adelle Renzaglia, and Eul Jung ChunUniversity of Illinois at Urbana Champaign

Abstract: The purpose of this study was to determine methods for and barriers to including students withdisabilities in high school service learning programs (HSSLPs) with non-disabled peers. Focus groups wereconducted with adult stakeholders at five schools nominated as having exemplary inclusive HSSLPs and atleast 3 years experience implementing such programs. Methods for including students with disabilities addressedthe categories of activity selection and structure, collaboration, expectations, encouragement, grouping, andmodifications. Barriers clustered around the areas of teacher attributes and experience, organizational structure,planning, resources, and student characteristics. Implications for inclusive education, universal design forlearning, access to the general curriculum, Individualized Education Programs (IEPs), and instruction arediscussed.

Service learning is a form of pedagogy thatenables students to meet their educationalgoals while providing service to the commu-nity (Fertman, 1994). It offers hands-on, prob-lem-based learning that is focused on meetingcommunity needs and enhancing school-com-munity collaboration (Fager, 1996; Perkins &Miller, 1994). Service learning offers a depar-ture from traditional pedagogy by linking ac-ademic content, standards, and vocationalgoals addressed in the classroom setting withhands-on service activities that meet authenticneeds in the student’s community (Gent &Gurecka, 1998; Hamilton & Hamilton, 1997).Almost half of all high schools nationally arereported to have service learning programs(Skinner & Chapman, 1999).

A service learning project typically includesfour components: 1) learning (i.e., prepara-tion for the activity), 2) service (i.e., perform-ing the actual service), 3) reflection (i.e., pro-cessing what was learned), and 4) celebration(i.e., of accomplishments) (Fertman, 1994;Gent & Gurecka, 1998). For example, stu-dents might initially learn about homelessness

in class and then engage in service activities ata local homeless shelter or soup kitchen. Afterthe activities, students reflect on what theyhave learned. Educational goals (e.g., math,social studies, literacy, social skills) are embed-ded across classroom and community activi-ties. Some of the benefits that have been at-tributed to service learning include anincreased appreciation for diversity, height-ened self-esteem and motivation, access to rel-evant learning contexts, the development ofcitizenship skills, an understanding of how towork collaboratively with others, and the es-tablishment of connections to the community(Allen, 2003; Billig, 2000; Briscoe, Pitofshy,Willie, & Regelbrugge, 1996; Eisler, Budin, &Mei, 1994; Fager, 1996; Nelson & McFadden,1995).

Service learning is increasingly being em-ployed with students with disabilities. As aform of pedagogy, it supports many of thewidely accepted tenants of effective curricu-lum and instruction in secondary special edu-cation. For example, it addresses academic,social, vocational, and life skills curriculumcontent in settings where the skills can beapplied (Brill, 1994; Burns, Storey, & Certo,1999; Everington & Stevenson, 1994; Yoder &Retish, 1994). Because service learning fre-quently occurs in the community, studentsreceive instruction in inclusive settings with

Correspondence concerning this article shouldbe addressed to Stacy Dymond, University of Illi-nois, Department of Special Education, 288 Educa-tion Building, MC-708, 1310 S. Sixth Street, Cham-paign, IL 61820. Email: [email protected]



people who do not have disabilities (Burns etal.). This in turn increases their visibility ascontributing members of society (Everington& Stevenson; Kleinert et al., 2004) and ex-pands the notion of “community-based in-struction” to include volunteer work (Burns etal; Dymond, 2004). It also provides a vehiclefor connecting students to socially significantprojects where they can make a difference intheir school and community (Gent &Gurecka, 1998; Muscott, 2001). All of thesepractices have the potential to assist studentsto generalize skills from school to real-life ap-plications, build competence across skill areas(e.g., social, academic, behavioral, life skills),develop career awareness, and prepare for thetransition to adulthood (Brill; Burns et al.;Dymond; Everington & Stevenson; Kleinert etal.; Carty & Hazelcorn, 2001; Muscott; Yoder& Retish).

Descriptions of service learning programsthat include high school students with disabil-ities have profiled segregated and inclusiveprograms. In segregated programs, serviceprojects are completed solely by students withdisabilities. Typically these projects involvestudents from one particular disability group(e.g., emotional and behavioral disorders, se-vere disabilities, learning disabilities) and theservice project is completed by an entire spe-cial education class. Examples of segregatedservice learning programs are widely availablein the literature (see Abernathy & Obenchain,2001; Everington & Stevenson, 1994; Frey,2003; Jackson, 1996; Krajewski & Callahan,1998; Carty & Hazelkorn, 2001; Muscott,2001).

In inclusive service learning programs, stu-dents with disabilities work alongside theirpeers without disabilities to complete serviceprojects. They may participate in service learn-ing as part of an inclusive class in which theyare enrolled (see Gent & Gurecka, 1998; Yo-der & Retish, 1994) or as part of an extracur-ricular school club (see Kleinert et al., 2004).While the instructional goals and type of par-ticipation may vary among students, inclusiveprograms focus on group effort and includeall students in planning, implementing, andevaluating the project (Brill, 1994; Gent &Gurecka; Kleinert et al.).

Conceptualizing service learning within aninclusive paradigm appears to be an emerging

trend within the field of secondary specialeducation. This is not surprising given thecontinued movement within the field towardinclusive education and the increasing em-phasis placed on access to the general curric-ulum (Dymond & Orelove, 2001; IDEIA,2004). Gent and Gurecka (1998) provide aparticularly compelling justification for inclu-sive service learning. They argue that suchprograms meet the needs of all students be-cause they easily blend academic and func-tional skills, and promote critical thinking. Asa result, it is possible for students with verydiverse abilities to actively participate in mean-ingful ways. For students with disabilities whoneed community-based instruction, servicelearning allows them to receive such instruc-tion with their same-age peers within a mean-ingful context. Furthermore, performingservice activities alongside peers without dis-abilities in the community enables studentswith disabilities to be seen as competent,equally contributing members of society.

Methods for including students with disabil-ities in inclusive high school service learningprograms (HSSLPs) are not clearly defined orunderstood. In our review of the ERIC,PsychInfo, and National Service LearningClearinghouse databases from 1990-2005, wefound eight peer reviewed articles that ad-dressed inclusive HSSLPs, only two of whichreported empirical data. Although none ofthese articles specifically investigated methodsfor including students with disabilities, theydo provide some direction for practitionersregarding promising practices. Methods forincluding students in inclusive HSSLPs consistof matching students’ skills with service activ-ities that they are capable of successfully com-pleting (Yoder & Retish, 1994), pairing stu-dents with and without disabilities to carry outprojects (Gent & Gurecka, 1998; Kleinert etal., 2004; Yoder & Retish), actively engagingstudents with disabilities in planning serviceprojects (Kleinert et al.), modifying materials(Gent & Gurecka), and having students withdisabilities teach students without disabilitieshow to perform tasks (Brill, 1994). In inclusiveHSSLPs, the Individualized Education Pro-gram (IEP) goals and objectives for studentswith disabilities should be linked to the servicelearning experiences (Brill; Gent & Gurecka;Kleinert et al.). This enables students to pur-

Inclusive High School Service Learning Programs / 21

sue the general curriculum as well as individ-ualized goals.

Barriers to including students with disabili-ties in inclusive HSSLPS were visibly absentfrom the literature. Yoder and Retish (1994)identified lack of time for engaging in servicelearning as one barrier. Students engaged inservice learning expressed interest in volun-teering for longer periods of time or on amore frequent basis than was possible. Giventhe relatively few articles addressing inclusiveHSSLPs, it is surprising that little informationis known about the barriers schools face inincluding students with disabilities.

Inclusive service learning appears to be apromising practice for assisting students withdisabilities to access the general curriculumand address other important curriculumgoals. It also incorporates a number of tenantsof effective practices for educating studentswith disabilities. In order to understand howinclusive service learning is an effective peda-gogy at the secondary level, more informationis needed about the methods for and barriersto including students with disabilities. Hence,the purpose of this study was to explore theperceptions of stakeholders from inclusiveHSSLPS about effective methods for includ-ing students with disabilities and barriers thatlimit or prevent their participation.

Method

One focus group was conducted with adultstakeholders in each of five inclusive HSSLPsin the state of Illinois. An “inclusive servicelearning program” was defined as one inwhich students with and without disabilitiesparticipated alongside each other to completea service learning project. Students with dis-abilities were the providers of service, not therecipients or beneficiaries of the service.

Participants

A combination of criterion and snowball sam-pling procedures (Patton, 2002) were em-ployed to select five Illinois high schools forparticipation. Criterion sampling allowed theselection of schools that met a pre-determinedcriterion of excellence while snowball sam-pling narrowed the number of schools to themost information rich cases. Exemplary

schools were initially identified as those receiv-ing distinction within the last five years as aNational Service Learning Leader Schooland/or a Prairie State Service LearningLeader School. The former designation in-volves national recognition by the Corpora-tion for National and Community Service(http://www.leaderschools.org/) and the lat-ter involves recognition by the Illinois StateBoard of Education (http://www.isbe.net/learnserve/). Sixteen schools met this crite-rion. Officials from the Illinois Learn andServe program, service learning coordinatorsfrom high school Leader Schools, and admin-istrators from regional superintendents’ of-fices were also contacted by telephone to ob-tain recommendations of exemplary inclusiveHSSLPs. This resulted in the identification ofeight schools, four of which were duplicativeof the list of leader schools.

To purposefully select a wide range ofschools with different experiences, informa-tion was gathered from the Interactive IllinoisReport Card (http://iirc.niu.edu/) about thegeographic location, size, socio-economic sta-tus, and ethnicity of the schools. Each of thefour schools that were both nominated as ex-emplary and identified as a leader schoolproved to be different along these dimensionsand thus were contacted first. Service learningcoordinators from the nominated programswere interviewed individually by telephone toobtain more information about the program(participants, inclusion of students with dis-abilities, courses using service learning, exam-ples of service learning projects, program’shistory).

Decisions about the inclusion of schools inthe study were made collaboratively by theauthors based on findings from the telephoneinterview. In addition to being identified asexemplary, criteria for inclusion in the studyincluded having at least three years of experi-ence implementing an inclusive service learn-ing program. All four schools nominated asexemplary and recognized as leader schoolsmet the criteria and agreed to participate.Two additional schools found to complementthe demographics (i.e., geographic location,size, socio-economic status, and ethnicity) ofthe selected schools were then contacted forinterviews. Although both schools met the cri-teria for participation, one school lacked suf-


ficient people to participate due to recent staffturnover, thus the other school was selected.We would have included both schools in thestudy had each agreed to participate.

The five schools selected for participationserved grades nine to 12 and had five to 12(M � 8.6) years of experience including stu-dents with disabilities in service learning.These schools were purposefully diverse withregards to geographic location, school size,socio-economic status, and ethnicity (see Ta-ble 1). All of the schools included studentswith mild disabilities. Three schools includedstudents with moderate disabilities and twoincluded students with severe disabilities. Oneto two individuals coordinated the servicelearning program at each school. These coor-dinators simultaneously assumed other re-sponsibilities at the school including teaching(social studies, foreign language, physical ed-ucation, special education) or guidance coun-seling.

One focus group was conducted at eachschool. Each focus group was composed ofadult stakeholders who were knowledgeableabout the service learning program and itsinclusion of students with disabilities. Viewsfrom at least three stakeholder groups (e.g.,

administrators, service learning coordinators,general education teachers, special educationteachers, related services personnel, parapro-fessionals, parents, and community members)were included in all of the focus groups. Par-ticipants were chosen by the service learningcoordinator at each school in consultationwith the focus group moderator. Two of thefocus groups included parents and/or com-munity members while three were composedexclusively of school personnel. Focus groupsranged in size from three to six participants(M � 5) and all participants received a smallmonetary stipend for participation in the two-hour interview.

Data Collection

The procedures selected for moderating thefocus groups adhered to the guidelines pro-vided by Krueger and Casey (2000). One re-searcher served as the moderator. She askedthe pre-identified questions, probed for moreinformation, summarized responses, and en-sured all participants had equal opportunityto participate. A second researcher partici-pated as an assistant moderator. She recordednotes on a flip chart and led the post-session

TABLE 1

School Demographics

Demographics School A School B School C School D School E

Number of students 144 362 1,984 595 2,193Percent with disabilitiesa 18 14.5 16 13 11.8Average class size 17.5 20 25.7 7.4 15Teacher to student ratio

across the schooldistrict

1:13 1:16 1:21 1:19 1:19.3

Percent low income 19 15 3.4 40.3 31.5Geographic location rural rural suburban urban suburbanPercent by ethnicity

White 100 99 89 23 41.6Black 0 .3 1 62 11.2Hispanic 0 0 5 14.5 40.3Asian 0 .6 4.5 .7 5.8Native American 0 0 .5 .2 0

Disability level of SLparticipants

Mild Mild,Moderate,

Severe

Mild Mild,Moderate,

Severe

Mild,Moderate

Note. SL � service learninga Includes students with an IEP, 504 plan, and those receiving speech and language services


debriefing following data collection. Althougha digital voice recorder was utilized to recordeach session, the flipchart notes allowed par-ticipants to see the ideas that had been gen-erated previously and enabled the moderatorto summarize the group’s responses prior tomoving on to the next question.

Each focus group was held at the partici-pants’ school in a room that was free fromdistractions. The moderator began each inter-view by facilitating participant introductions,clarifying the roles of the moderators and par-ticipants, explaining the purpose of the focusgroup, discussing confidentiality and how theresults would be used, and reviewing the focusgroup questions. The moderator and assistantmoderator convened following each focusgroup to discuss, clarify, and summarize themain points of the session.

Five structured questions were asked withineach focus group interview. The first threequestions gathered data on the participants’service learning program and their beliefsabout the key elements of inclusive HSSLPs(see Dymond, Renzaglia, & Chun, 2007). Thefourth question requested participants to de-scribe the methods they found most effectivefor including students with disabilities in ser-vice learning. The final question requiredthem to identify what, if any, barriers theyexperienced in including students with dis-abilities in service learning. Findings fromthese last two questions are the basis for thisarticle.

Data Analysis

Data were analyzed in two stages (Patton,2002). The first stage involved a qualitativeanalysis. This method was chosen because itallowed a list of specific methods and barriersto emerge from the data, and the use of in-ductive analysis to determine categories thatcut across schools. The second stage involveddetermining the number of schools that iden-tified each method and barrier. This analysiswas performed to clarify the extent to whichthe methods and barriers were prevalentacross schools.

Coding. The digital recording of each fo-cus group session was transcribed verbatim.Utilizing a content analysis procedure (Pat-ton, 2002), each transcript was read several

times in order to develop a list of codes todescribe each method for and barrier to in-cluding students with disabilities in servicelearning. Once the codes were finalized, tran-scripts were re-read and a code was assigned toeach method or barrier. A second researcherindependently reviewed all coded data to ver-ify the appropriateness of the assigned codesand the consistency used in applying thecodes. Where differences of opinion existed,the two researchers discussed the coding untilthey arrived at agreement. The final codes arelisted as methods in Table 2 and barriers inTable 3.

Data reduction. Using an inductive ap-proach, methods across all five schools wereexamined and grouped into categories. Meth-ods that were similar were included in thesame category if they were complementary inmeaning and helped to define the category.Each method was assigned to only one cate-gory. This process was repeated for the list ofbarriers. As with the initial coding, a secondresearcher independently reviewed the codesassigned to each category to confirm theirappropriateness. Differences of opinion werediscussed between the two researchers untilagreement was obtained about the codes in-cluded in each category. The final categoriesare defined in Tables 2 and 3.

Following the qualitative analysis, a fre-quency count was performed to determine thenumber of schools that identified eachmethod and barrier. This allowed for compar-isons in the magnitude of responses and thelevel of agreement among schools. All meth-ods and barriers, regardless of the number ofparticipants or schools that identified them,were included in the analysis. We anticipatedthat the methods and barriers identified byeach school might vary due to the nature ofthe school (i.e., geographic location, size, eth-nicity, percent low income, type of studentswith disabilities served) so we purposefully se-lected schools that were diverse across thesedimensions. The decision to retain all data wasmade because the primary purpose of thestudy was to capture the range of practitionerexperiences and beliefs about methods andbarriers. Deleting ideas mentioned by onlyone person or school would have diminishedour understanding of practices advocated bystakeholders in diverse, inclusive HSSLPs.


Trustworthiness. Guba and Lincoln (1989)indicated that rigorous qualitative analysesuse procedures that support “trustworthi-ness” of the data. In this study, trustworthi-ness was ensured by a) analyzing data tran-scribed verbatim from a high quality digitalvoice recorder, b) using a two-step processto understand the data, and c) verifying

interpretations of the data between tworesearchers.

Results

Data from this study are organized around: a)methods for including students with disabili-ties in inclusive HSSLPs and b) barriers to

TABLE 2

Methods for Including Students with Disabilities in Inclusive HSSLPs (N � 5)

Category Methods # of Schools

Activity selection and Give students choices 4structure (n � 5) Place students in situations where they will be successful 4

Choose activities depending on students’ ability level 3Include more hands-on activities 2Offer a variety of activities 2Assess students to determine their capabilities 2Start with smaller, teacher initiated projects 1

Collaboration (n � 5) Collaborate with general and/or special educationteachers

3

Obtain support for service learning from all staffinvolved with students

2

Provide enough staff 2Obtain information about students’ abilities 2Obtain information about methods for ensuring

students’ success1

Provide teachers with information about students’disabilities

1

Discuss service learning opportunities during IEPmeetings, 504 meetings, or transition meetings

1

Expectations (n � 5) Have the same expectations for students with andwithout disabilities

3

Treat students with disabilities like everyone else 3Expect students to participate 2Allow students to fail 1

Encouragement (n � 4) Allow student ownership of the project 3Get to know the students with disabilities 3Ask students how they would like to be involved in

service learning2

Allow students to have an active role in decision-making 1Grouping (n � 4) Pair students with and without disabilities to complete

projects4

Include normal proportions of students with andwithout disabilities

2

Modifications (n � 3) Provide additional instruction 2Provide supplementary instruction in the special

education classroom1

Modify the rules when needed 1Modify grading practices 1

Note. HSSLPs � high school service learning programs; n � number of schools contributing methods to eachcategory


including students with disabilities in inclusiveHSSLPs. Tables 2 and 3 provide a list of themethods and barriers that emerged and thenumber of schools that identified methodsand barriers within each category.

Methods for Including Students with Disabilities

Methods clustered within six categories: a)activity selection and structure, b) collabora-tion, c) expectations, d) encouragement, e)grouping, and f) modifications (see Table 2).Participants provided varying levels of detailregarding the methods they employed. In fact,many initially struggled to identify specificstrategies for including students with disabili-ties. A common remark among participantswas that the methods they employed were

ones they found to be effective with all stu-dents, regardless of whether they had an iden-tified disability.

Activity selection and structure. Participantsacross all five schools emphasized the impor-tance of selecting appropriate service learningactivities for students with disabilities. Activi-ties should be selected that are “at their level”and will enable them to be “successful” and“feel safe” with engaging in the project. Twomethods were suggested for ensuring studentsuccess. One is to informally assess students todetermine activities that match students’strengths and promote active participation.Understanding students’ skills prevents teach-ers from assigning students to activities thatextend beyond their abilities. A secondmethod is to invite students with disabilities to

TABLE 3

Barriers to Including Students with Disabilities in Inclusive HSSLPs (N � 5)

Category Barriers # of Schools

Resources (n � 5) Not enough transportation 3Lack of money 2Too much effort required to coordinate the inclusion 2Lack of administrative support 1Not enough people/staff 1Staff turnover 1Lack of proper equipment (e.g., washroom, chairs) in the

community1

Not enough time to work with students with disabilities 1Teacher attributes and Lack of knowledge about students’ disabilities 3

experience (n � 4) Negative attitudes toward students with disabilities 2Low expectations of students with disabilities 1Lack of experience with students with disabilities 1Need for staff development and training 1Teachers don’t have the patience to work with students

with disabilities1

Organizational structure Schedules of students with disabilities are not flexible 3(n � 4) Students with disabilities have a shortened school day 2

Too many students with disabilities in service learningclasses

1

Students with disabilities need to be more visible in theschool

1

Planning (n � 3) Lack of communication among staff about students withdisabilities

3

Not enough time to co-plan 2Student characteristics

(n � 2)Students with disabilities are afraid of participating in the

service activities2

Note. HSSLPs � high school service learning programs; n � number of schools contributing methods to eachcategory


participate in service learning projects initi-ated and planned by the teacher. As studentswith disabilities become more comfortable,they can be encouraged to design their ownservice learning projects.

Participants also spoke about methods forstructuring activities to promote the participa-tion of students with diverse abilities. Provid-ing “choices” and “options” was viewed as akey method for encouraging reluctant stu-dents to participate and creating student own-ership for the service project. Other methodsinclude offering a variety of tasks and incor-porating hands-on activities. The morehands-on the activity and the more choicesand types of tasks available, the more likely allstudents will be able to participate in somecomponent of the activity. As one participantcommented, when activities are structuredthis way, “their disabilities aren’t so notice-able.”

Collaboration. All five schools deemed col-laboration essential. Discussions across focusgroups focused primarily on collaborationamong school personnel. To effectively in-clude students with disabilities, general andspecial education teachers need to meet reg-ularly, collaborate to develop curriculum,share information about each student’sstrengths and needs, and provide support toeach other. In one school, the special educa-tion teacher provided general educationteachers with a list of the strengths and weak-nesses of each student included in his/herclass. This information helps the general edu-cator understand the student’s disability, ad-aptation requirements, and instructional sup-port needs. Participants at another schooldiscussed student participation in servicelearning more formally during IEP, transition,and/or 504 plan meetings. During thesemeetings, the student and his/her team re-view service learning requirements, determinehow the student might benefit from this formof instruction, and identify the supports andaccommodations necessary for participation.

Teaching assistants and other support staffare considered essential collaborators in pro-grams that include students with severe dis-abilities or a large number of students withmild disabilities. These individuals provide in-struction to students and help those withwheelchairs move to various locations during

activities. One participant noted that thereneeds to be “support staff specifically gearedtowards those students” because they needmore “one-on-one and hands-on” than otherstudents. Another commented that having suf-ficient staff “facilitates the inclusion.” In addi-tion to working directly with students withdisabilities, participants recommended in-cluding support staff in meetings pertainingto student participation in service learning.

Collaboration was also defined as havingsupport from the school faculty for the pro-gram. As one participant noted, “I think thatyou gotta have all the teachers involved andbuy into it for it to work.” At another school,faculty buy-in was viewed as having teachers atthe school who were flexible about lettingstudents out of classes to participate in servicelearning. Even though those teachers werenot leading the service learning activity as partof their class, they allowed students to partic-ipate in service learning as long as they madeup the work they missed and maintained goodgrades.

Expectations. Participants’ expectations forstudents with disabilities were uniformly highacross all schools. Those who spoke about ex-pectations were adamant that teachers havethe same expectations for students with andwithout disabilities, and treat students withdisabilities the same way they treat studentswithout disabilities. For one participant, pro-viding equal treatment included allowing stu-dents to fail. “Everybody fails at somethingand a person with a disability has the right tofail just as much as the next person.” Althoughother participants did not echo this position,it demonstrates the extent to which this indi-vidual viewed the importance of providing eq-uitable experiences for all children.

A few participants felt they treated studentswith disabilities differently from their non-dis-abled peers. For example, one participant ad-mitted:

I think subconsciously, I’m gonna make itconsciously, even a lot of time, we tend totreat those kids a little differently. I don’tknow that it has anything to do with servicelearning, but I think sometimes we maytend to accept some of their behaviors alittle more readily for them than we mightfrom some other kids.


Another participant commented that shewasn’t always sure about students’ capabilitiesand would feel bad if her expectations weretoo low. Although these types of commentswere voiced somewhat reluctantly, they under-score the issue inherent in determining ap-propriate expectations for students with dis-abilities.

Two rationales for having high expectationswere provided. One rationale emphasized theimpact high expectations have on studentswithout disabilities. When teachers have highexpectations for students with disabilities andtreat them like all other students, studentswithout disabilities are more likely to interactwith and accept students with disabilities asequal members of the class. The other, morecommon, rationale voiced suggests that ser-vice learning activities do not limit the partic-ipation of any student and, thus, should notimpact the expectations one has for studentswith disabilities. One participant summed thispoint up succinctly. “There’s no distinction ifa kid has a disability or not. It (service learn-ing) crosses every barrier there is. I don’t carewhat your limitation to learning is.” Anotherparticipant was emphatic in denying the needfor students with disabilities to have specialtreatment. She stated, “The LD kids, it’s like, Iain’t given you a break cuz you have a learningdisability. Just go and do it. What’s the prob-lem?”

Across schools, participants’ focused on anexpectation for active participation in activi-ties. Because choices are available within ac-tivities, students are able to self-select out ofcompleting tasks that are difficult for them oraccentuate their weaknesses. For example, if astudent experiences difficulty with writing, hewould likely choose a different part of theservice learning activity to complete that al-lows him to use his strengths. High expecta-tions for participation were not necessarilytied to high expectations to perform at gradelevel in content area subjects.

Encouragement. Participants across four ofthe schools agreed that it is important forteachers to encourage students with disabili-ties to participate in service learning activities.They do this by “asking” students to partici-pate, inquiring how they would like to be in-volved, and negotiating the role they want toassume. One participant explained it this way:

We just need to encourage them more. Ithink some of the kids with needs or disabil-ities sometimes sit back and don’t want tobe involved because they don’t think theycan or they don’t think they should, andthey don’t want anybody to know.

In three of the schools, participants advocatedspending time getting to know students inorder to make them feel comfortable with theteacher, the other students, and the activities.Teacher encouragement was viewed as amethod to help build students’ confidence.

In addition to encouraging students to par-ticipate, students with disabilities also need tobe encouraged to assume an active role indecision-making. One service learning coordi-nator was adamant about demanding studentinvolvement as evidenced by this statement:

Involve the kids from the get go. Ask theiropinions. Seek them out. Ask them. If onething is repeated more and more than any-thing else over the last ten years to me isthey always say how do you get all these kidsto do all these things? And the simple an-swer is you ask them. Seriously, if you stopand think about it, almost anybody woulddo anything for you.

When students are encouraged to provide in-put into the design of the activities, they gainincreased ownership for the project and itssuccess. As one participant noted, without thistype of ownership, “the service is not going tofly.”

Grouping. Participants at four of theschools described strong rationales and strat-egies for pairing students with and withoutdisabilities to perform service learning. Whenthinking about the benefits of diverse group-ing, one participant commented, “We had astrength here, and a strength here, and astrength here. As a group they’ll worktogether, but if individually, they had to doall three (tasks), they’d struggle.” Teachersgroup students so that each person can con-tribute a different, yet complementary,strength to the group’s project.

Some participants carefully pair students towork together and some allow partners toemerge on their own. At one school, studentswith disabilities are paired with individuals


who are “more experienced” in service learn-ing than the student with a disability. Experi-enced students were described as individualswho are “more mature” or “advanced” (e.g., astudent taking Advanced Placement courses).In another school, students with disabilitiesare paired with students who can serve as “rolemodels.” For example, students with behaviorproblems are often paired with students whodo not have behavior problems.

An alternative strategy to grouping is tohave students select their own partners. Par-ticipants at one school indicated they do notpurposefully group students. They believe thathaving students work with partners is impor-tant and that students need to choose theirown partners. When partners do not work welltogether, the adults ask another student to“help out.” They do not force students to worktogether. Students always have choice abouttheir partners. In order for this strategy towork, teachers need to be observant and sen-sitive to student interactions, and be preparedto “step in” when problems arise.

Although participants elaborated on strate-gies for grouping students with and withoutdisabilities to work together, they also statedthat the class as a whole (i.e., the large group)needs to include a normal proportion of stu-dents with disabilities. Over-representation ofstudents with disabilities inhibits the effective-ness of the teacher and the ability of all stu-dents to adequately learn and participate. Ifcontrols are not put on the number of stu-dents with disabilities enrolled, it is easy forthese classes to become a “dumping ground.”

Modifications. Modifications that facilitateinclusion of students with disabilities in ser-vice learning include providing additional in-struction, monitoring students with disabili-ties more frequently than students withoutdisabilities, providing supplementary instruc-tion in the special education classroom (e.g.,pre-teaching behavior expectations, providingconsequences for misbehavior in the commu-nity), and modifying the rules and gradingpractices. Instructional modifications wereonly discussed within three schools and noparticipants provided specific examples (be-yond those listed above) of the types of mod-ifications they employ.

Barriers to Including Students with Disabilities

Five categories of barriers emerged. These cat-egories include a) resources, b) teacher at-tributes and experience, c) organizationalstructure, d) planning, and e) student charac-teristics (see Table 3). Although all fiveschools contributed to the list of barriers, par-ticipants at two of the schools indicated thatno barriers currently exist to including stu-dents with disabilities in service learning. Theyadmitted that barriers might exist if they wereasked to serve more students with disabilities,or students with more severe disabilities orbehavior problems.

Resources. Participants across all five schoolsidentified resources that were needed to im-prove the inclusion of students with disabili-ties. The two primary resource barriers citedwere money and transportation. Limitedfunding for materials needed to completeprojects can curtail the number and type ofservice learning projects available. Funding isalso needed for specialized buses to accommo-date students who use wheelchairs. Sincethese students cannot ride the regular schoolbus, there is an increased cost for an extra buseach time a student with a wheelchair per-forms service in the community. Lack of trans-portation for service learning activities thatoccur after school or on weekends was alsocited as a barrier.

One school experienced difficulty findingappropriate equipment in the community.Students with disabilities need accessiblewashrooms and specialized chairs that maynot be available in all community settings.This impacts students’ access to some servicelearning activities.

Staffing was viewed as another barrier toincluding students with disabilities. At oneschool participants commented that studentswith disabilities would be more successful ifthe staff had more time to work with them.There needs to be a peer mentor, anotherservice learning student, or a staff personavailable to provide support when needed. Atanother school, staff turnover was viewed as abarrier to maintaining the program in gen-eral. Position cuts at the school eliminatedmany teachers who had been implementingservice learning within their curriculum. Al-though pockets of people exist to champion


the program, time is needed to attract andtrain new teachers to incorporate servicelearning in their courses.

A concern was articulated about theamount of effort required to coordinate ser-vice learning programs and the impact thatadding students with disabilities might haveon a program. It could be very overwhelmingto a service learning coordinator if he or shewas expected to include all students with dis-abilities at once, particularly if that programhad not included those students previously.Additional staff support would be necessary toinclude large numbers of students with dis-abilities, particularly if the students have se-vere disabilities.

Finally, lack of administrative support wasviewed as problematic (either currently or inthe past). Some felt that the administrationwas not supportive of service learning in gen-eral and had difficulty seeing how it “fit” withthe general curriculum and overall academicprogram. This barrier was not specific to in-cluding students with disabilities. Othersthought their administration was not knowl-edgeable about the needs of students withdisabilities and offered minimal support toadvance their participation in the community.

Teacher attributes and experience. Partici-pants from four schools viewed their own lackof knowledge and experience with studentswith disabilities as a potential barrier to in-cluding students with disabilities. Teachersand community members need more informa-tion about students’ disabilities, the “prob-lems” associated with the disability, and howto adapt the curriculum to meet studentneeds. All school faculty need to becomemore competent in working with studentswith disabilities.

In addition to lack of training, another bar-rier is the negative attitudes toward studentswith disabilities possessed by some adults. Atone school, where students with severe disabil-ities are just beginning to participate in servicelearning projects, a participant noted, “I don’tthink people know what to expect of studentswho have more moderate to severe disabili-ties. I think there’s fear out there.” At otherschools, participants admitted that someteachers have low expectations for studentswith disabilities and lack the patience neededto work with them. One general educator/

service learning coordinator eloquently com-mented on the attitude he aspired all teachersto embrace.

It’s a person with a disability. It’s not adisability on a person. We need to thinkbeyond what we think they may or may notbe capable of. We don’t allow those barriersto be put in front of the child without adisability.

This participant stressed the need for moreadults to “think outside the box” when deter-mining how to include students with disabili-ties.

Organizational structure. Three barriersemerged related to the organizational struc-ture of the school. First, the schedules of stu-dents with disabilities prevent or limit theirparticipation in service learning. Some stu-dents are on a shortened school day or leaveearly to go to work. Others participate in com-munity-based instruction at times that overlapwith service learning activities.

Second, classes that offer service learningneed to be monitored so they do not becomea “dumping ground” for students with disabil-ities. Over-representation of students with dis-abilities makes the classroom dynamics diffi-cult to manage. Participants suggested puttinga “cap” on the number of students with dis-abilities in each class so that all students areable to sufficiently benefit from service learn-ing.

Third, students with disabilities need to bemore visible throughout the school buildingand “more integrated into the population.”When students with disabilities are served pri-marily through self-contained special educa-tion classrooms, it is important for their class-rooms to be integrated into the generalclassroom areas rather than placed in a sepa-rate part of the building. These students alsoneed to be served in more general educationclasses in order to facilitate their inclusion inthe service learning program and the school.

Planning. Four schools identified plan-ning as a barrier to including students withdisabilities. Insufficient time exists for staff todiscuss the needs of students with disabilities,collaboratively plan activities, or coordinateactivities and staff efforts across the school.For example, several participants voiced con-


cern that they were not always aware of whichstudents in their classes had disabilities. Oneparticipant noted, “When I get some of thesekids, I don’t know what is wrong. I don’t knowthat there’s a problem. I don’t know anythingabout these kids.” Communication breaksdown and teachers are not informed of stu-dents’ disabilities, IEP objectives, and accom-modation needs. In addition, some partici-pants were not sure if all students withdisabilities at their school participated in ser-vice learning. This was particularly true whenthe school housed full-time self-containedclassrooms.

Student characteristics. Some students withdisabilities have low self-esteem and do notwant to be placed in a situation where theyface potential embarrassment. They lack con-fidence in their abilities and are fearful ofparticipating in service activities. Participantsviewed student characteristics as a barrier attwo of the schools. They also acknowledgedthat the self-esteem and confidence issuesfaced by students with disabilities are alsoprevalent among some students without dis-abilities. These student characteristics, morethan the presence of disability, negatively im-pact students’ willingness and ability to partic-ipate in service learning.

Discussion

Findings from this study provide a preliminaryexamination of the methods for and barriersto including students with disabilities in HSS-LPs alongside their peers without disabilities.Methods emerged in the categories of activityselection and structure, collaboration, expec-tations, encouragement, grouping, and mod-ifications. Barriers clustered around the cate-gories of resources, teacher attributes andexperience, organizational structure, plan-ning, and student characteristics. These meth-ods and barriers support and expand on thoseidentified previously in the literature on inclu-sive HSSLPs.

There are several limitations that should beacknowledged prior to discussing the find-ings. First, we did not observe the servicelearning programs at each school, therefore itis unclear the extent to which the methodsand barriers identified by each school wereactually present. Second, since stakeholders

from five schools generated the data for thisstudy, it is possible that the findings are notrepresentative of the range of methods andbarriers experienced by all inclusive HSSLPs.Third, school stakeholders (i.e., general edu-cators, special educators, paraprofessionals,principals) were the dominant group presentwithin the focus groups. Additional methodsand barriers might have been identified ifrepresentation from parents and communitymembers was higher within each focus group.Fourth, the data do not take into consider-ation the perspectives of students with andwithout disabilities. Their views may be verydifferent than those of adults. Finally, a limi-tation of the focus group methodology is thatit did not allow participants across schools tointeract and comment on the methods andbarriers identified by each school. As a result,the number of schools indicating eachmethod and barrier (as reflected in Tables 2and 3) may underestimate the actual numberof schools where the methods and barrierswere present.

A Philosophy of Inclusion: The Unspoken Method

Within each school there appeared to be anoverall shared commitment for inclusive ser-vice learning. Each stakeholder, regardless ofhis or her role, conveyed the importance ofincluding students with disabilities in servicelearning and spoke positively about his or herexperiences. At each school, there was neveran instance of service learning being champi-oned by only a single person. Nor was inclu-sion an idea that was “owned” or “advocated”exclusively by the special education staff. Par-ticipants were united in their belief that allstudents, regardless of ability, could andshould participate in service learning.

While no one specifically discussed the im-portance of having a philosophy of inclusion,one must wonder whether an unspoken andunacknowledged method for including stu-dents with disabilities in service learning is thepresence of a group that philosophically sup-ports inclusive education. Participants acrossschools were able to identify barriers that limitthe inclusion of students with disabilities inservice learning, but they also spoke openlyabout changes they would like to see to im-prove the overall inclusion of students with


disabilities at their school. For example, someparticipants felt the classrooms for studentswith disabilities should be located alongsideother general education classrooms ratherthan in a non-classroom wing of the building.Others indicated that students with disabilitiesshould be included in more general educa-tion classes. They stated that improving thevisibility and inclusion of students with disabil-ities across the school day would enhancetheir inclusion in service learning projects be-cause students and teachers would “alreadyknow them.”

Barriers to Inclusion or Inclusive ServiceLearning?

Barriers to inclusive service learning identi-fied by the participants are consistent, in manyways, with barriers to inclusive education citedin the literature. Issues related to resources,staff knowledge and expertise, planning time,and program organization and logistics arewidely acknowledged as constraints on the im-plementation of effective inclusive education(Janney, Snell, Beers, & Raynes, 1995; Mas-tropieri et al., 2005; Pearman, Huang, & Mell-blom, 1997; Pivik, McComas, & Laflamme,2002; Scruggs & Mastropieri, 1996; Werts,Wolery, Snyder, & Caldwell, 1996). Althoughtwo schools indicated that no barriers cur-rently existed to including students with dis-abilities in service learning, participants werequick to point out that adding more studentswith disabilities or students with more severedisabilities or behavior problems could beproblematic unless additional resources andtraining were available. They were able to en-vision events (e.g., reduced funding, changesin administration, teacher attrition) thatcould create barriers to including studentswith disabilities in the future.

Given the similarity between barriers iden-tified in this study and those cited in the in-clusive education literature, it seems plausibleto infer that the barriers may be more charac-teristic of an inclusive practice as opposed to aphenomena specific to inclusive service learn-ing. We suggest that it is not the educationalpedagogy of service learning, per se, thatcauses or creates the barriers. It is the diffi-culty with implementing the practice of inclu-sive education (that transcends curriculum

and pedagogy) that creates the barrier. Meth-ods for overcoming barriers to inclusive edu-cation may well prove effective in addressingsome of the barriers encountered by inclusiveHSSLPs. It remains unclear which barriers, ifany, are specific to inclusive service learning.

The Goal: Participation vs. Skill Acquisition?

For the schools in this study, inclusion wasdefined almost synonymously with participa-tion. Implicit in this definition, as articulatedthrough numerous examples, was an empha-sis on active engagement and ensuring that allstudents made a contribution to the serviceproject. Whether the student completed a taskin whole or in part, with help or not, did notdiminish the value of the participation. Infact, partial participation and collaborationwith peers was clearly valued, as evidenced bythe many descriptions provided of how teach-ers grouped students with and without disabil-ities to complete projects. Participation was akey term reiterated across schools andthroughout each interview.

We found the emphasis placed on students’active participation to be both refreshing andnoteworthy. At the same time, we questionedwhether the methods discussed for includingstudents with disabilities extended appropri-ately far enough to ensure their access to thegeneral curriculum and attention to individu-alized IEP objectives. In previous discussionswith the participants (see Dymond et al.,2007), we found strong support for connect-ing service learning to the curriculum. Partic-ipants were adamant that inclusive HSSLPsshould link to both the academic curriculumand to a functional life skills curriculum. Wewere, therefore, intrigued when these sameparticipants failed to mention the connectionbetween service learning and the curriculumas a method for including students with dis-abilities. Only one school’s participants sug-gested talking about service learning as part ofan IEP, 504 plan, or transition plan, but nomention was made of how decisions weremade about the curriculum to be addressedthrough service learning.

Across focus groups, stakeholders discussedthe importance of allowing students to chooseactivities and to select activities that matchedtheir skills. Yet, it seemed that the focus of the


service learning activity for students with dis-abilities was more highly focused on participa-tion than it was on learning new skills. Stu-dents were channeled into activities thatmatched their skills and preferences ratherthan working on new skills or applying emerg-ing skills. If students perceived an activity astoo difficult, they were allowed to self-selectout of the activity and choose a different one.The role of the teacher in directing studentlearning and teaching new skills was visiblyabsent from the list of methods for includingstudents with disabilities. One might arguethat this is appropriate since an importantcomponent of service learning is student par-ticipation in designing service projects. In ouropinion, if service learning is to be linked tothe curriculum, there needs to be a carefulbalance between allowing students to assumeownership and responsibility for projects andteacher oversight regarding the acquisitionand application of new skills. Connectionsneed to be made that illustrate not only thatstudents with disabilities can be successfullyincluded (i.e., participate), but that they canalso achieve and learn as a result. It was notclear from the focus groups whether the in-tent of service learning for students with dis-abilities was to assist them in learning andpracticing new skills, or whether it was to allowthem to apply skills they had already masteredto new situations. We believe there should bea balance.

Are High Expectations Enough?

A truly admirable trait of participants was theemphasis they placed on having high expecta-tions for students with disabilities. They con-sidered students with disabilities capable. Inaddition, they believed that the structure ofservice learning activities (e.g., hands-onlearning, activity-based projects, studentchoice, variety of tasks) was what allowed stu-dents with very diverse abilities to participate.Service learning, as a form of pedagogy, elim-inated potential barriers to including studentswith disabilities, thus increasing teachers’ ex-pectations for students to be successful.

In essence, participants viewed servicelearning as a form of universal design forlearning (UDL). The premise of UDL is thatthe curriculum should be designed with the

needs of all students in mind from the start. Ifthe curriculum is designed appropriately, theneed for modifications and adaptationsshould be greatly diminished or non-existent(CEC, 1998; Hitchcock, Meyer, Rose, & Jack-son, 2002). In the present study, many partic-ipants initially expressed difficulty with identi-fying methods for including students withdisabilities in service learning. They indicatedthat the methods they used were no differentthan the ones they used with students withoutdisabilities. Some even stated that servicelearning was “the method” for including stu-dents with disabilities alongside their peers.Perhaps the service learning experiences atthese schools embodied the principles of UDLsuch that specialized adaptations and modifi-cations were rarely necessary.

Although numerous methods for includingstudents with disabilities in service learningwere ultimately identified, we were surprisedto find limited mention of teaching methods.It is the absence of these methods that bringsto question whether the expectations partici-pants had for students with disabilities weresufficiently high. That is, did the participantshave adequate knowledge and expertise inworking with students with disabilities toclearly identify reasonable goals and effectiveteaching strategies for helping students learn?The nature of service learning may indeedmake it a UDL strategy, but UDL does notpreclude the need for teaching. UDL shouldmake the curriculum accessible. An accessiblecurriculum is not equivalent to an easier orless demanding curriculum (CEC, 1998; Ork-wis, 1999; Rose & Meyer, 2000).

Participants themselves seemed unclearabout how their expectations for students withdisabilities could or should be any differentthan they would be for students without dis-abilities. While they expressed support forholding students with disabilities to the samehigh standards as students without disabilities,they were not always sure what was reasonableor what the students were truly capable ofaccomplishing. This was evident by some ofthe barriers they identified, including lack ofknowledge about students’ disabilities and ac-commodation requirements, a need for moreexperience and training in working with stu-dents with disabilities, and a desire for bettercommunication and collaboration between


general and special educators. If informationabout the needs of students with disabilities isnot regularly shared and discussed, it becomesless surprising that strategies for teaching stu-dents with disabilities failed to emergestrongly in our list of methods.

Implications for Research and Practice

The methods and barriers identified by par-ticipants in this study offer practical directionfor high school personnel seeking to includestudents with disabilities in service learning.Their words and examples offer insights thatcan only be gained from their experienceswith implementing an inclusive program.Based on the collective findings acrossschools, we believe there are several pointsthat warrant additional consideration byschool and community stakeholders.

● An inclusive HSSLP is bound together by agroup of adult stakeholders from diversedisciplines that embody a philosophy of in-clusion. The philosophy of this group needsto extend beyond the day-to-day operationof the service learning program to includegoals toward increasing the overall inclu-siveness of the school itself. The presence ofan inclusive school philosophy will supportand enhance implementation of an inclu-sive HSSLP.

● All students, regardless of their ability, arecapable of participating in service learning.“Participation” is important and each stu-dent with a disability needs to be activelyengaged, at his or her own level, to theextent they are capable.

● There is a need to move beyond defininginclusion as participation. Teachers andother adults who support students in learn-ing need to create a balance between allow-ing students to apply skills they have alreadymastered and assisting them to learn newskills and practice emerging ones. Althoughservice learning can be conceptualized as aUDL strategy, it does not preclude the needfor specialized instructional strategies.

● Greater attention needs to be devoted toensuring that methods for including stu-dents with disabilities in service learningtake into consideration the curriculumneeds of the students. IEP objectives should

be infused into service learning projectswhere appropriate.

● In light of IDEIA (2004) and the No ChildLeft Behind Act (NCLB, 2002) measurabledata must be gathered on the performanceoutcomes of students with and without dis-abilities who participate in service learning.These outcomes need to address curricu-lum linked to the state standards as well asperformance on IEP objectives that addressother curriculum areas such as life skills.

● School and community members need ad-ditional training in how to teach studentswith disabilities. They need to be informedof the unique needs of the students withdisabilities for whom they are responsible.Time for teachers to collaborate and shareinformation is essential to consistently meetthe needs of students with disabilities acrosshigh school classes.

Additional research on inclusive HSSLPsshould focus on validating the methods andbarriers identified through this study. Thisinformation could provide useful informationabout the extent to which the findings fromthis study are representative of other inclusiveHSSLPs and may help to extend the list ofmethods and barriers identified. Researchshould also seek to identify effective methodsfor overcoming barriers that prevent or limitthe participation of students with disabilitiesin HSSLPs.

A more thorough examination of the meth-ods used to include students with disabilitiesin HSSLPs is also needed. Observations ofinclusive HSSLPs should be conducted to de-termine the extent to which the practices ad-vocated by schools are present and result indesired student outcomes. The impact of stu-dents’ disability level (i.e., mild, moderate,severe), teacher to student ratio, general edu-cation teacher experience in working with stu-dents with disabilities, and role of the specialeducation teacher should also be investigatedin relationship to methods employed. Finally,we believe there is a need to define effectivemethods for linking the general curriculumand IEP to service learning activities. Addi-tional work is needed to investigate how stu-dents with disabilities can address challengingcurriculum content within HSSLPs and the


supports school personnel need to help stu-dents reach their potential.

References

Abernathy, T.V., & Obenchain, K.M. (2001). Stu-dent ownership of service-learning projects: In-cluding ourselves in our community. Interventionin School and Clinic, 37(2), 86–95.

Allen, R. (2003). The democratic aims of servicelearning. Educational Leadership, 60(6), 51–54.

Billig, S. H. (2000). Research on K-12 school-basedservice-learning: The evidence builds. Phi DeltaKappan, 81, 658–664.

Brill, C. L. (1994). The effects of participation inservice-learning on adolescents with disabilities.Journal of Adolescence, 17, 369–380.

Briscoe, J., Pitofshy, J., Willie, J., & Regelbrugge, L.(1996). Service learning and school to work: A part-nership strategy for educational renewal. Alexandria,VA: National Association of Partners in Educa-tion.

Burns, M., Storey, K., & Certo, N. J. (1999). Effect ofservice-learning on attitudes towards studentswith severe disabilities. Education and Training inMental Retardation and Developmental Disabilities,34, 58–65.

Council for Exceptional Children. (1998). A curric-ulum every student can use: Design principles for stu-dent access. Retrieved February 22, 2006 fromhttp://www.cec.sped.org/osep/udesign.html

Dymond, S. K. (2004). Community participation. InP. Wehman, & J. Kregel (Eds.), Functional curric-ulum for elementary, middle, and secondary age stu-dents with special needs (2nd ed.). Austin: Pro-ed.

Dymond, S. K., & Orelove, F. P. (2001). What con-stitutes effective curricula for students with severedisabilities? Exceptionality, 9, 109–122.

Dymond, S. K., Renzaglia, A., & Chun, E. J. (2007).Elements of effective high school service learningprograms that include students with and withoutdisabilities. Remedial and Special Education, 28,227–243.

Eisler, J. A., Budin, H., & Mei, L. (1994). The stu-dent service and philanthropy project. ERS Spec-trum, 12(2), 20–27.

Everington, C., & Stevenson, T. (1994). A givingexperience: Using community service to promotecommunity living skills and integration for indi-viduals with severe disabilities. Teaching Excep-tional Children, 26, 56–59.

Fager, J. (1996, July). Service learning in the northwestregion. Northwest Regional Educational Labora-tory.

Fertman, C. I. (1994). Service learning for all students.Bloomington: Phi Delta Kappa.

Frey, L. M. (2003). Abundant beautification: An

effective service-learning project for students withemotional or behavioral disorders. Teaching Excep-tional Children, 35(5), 65–75.

Gent, P. J., & Gurecka, L. E. (1998). Service-learn-ing: A creative strategy for inclusive classrooms.Journal of the Association for Persons with Severe Hand-icaps, 23, 261–271.

Guba, E. G., & Lincoln, Y. S. (1989). Fourth genera-tion evaluation. Newbury Park: Sage.

Hamilton, S. F. & Hamilton, M. A. (1997). When islearning work-based? Phi Delta Kappan, 78, 677–681.

Hitchcock, C., Meyer, A., Rose, D., & Jackson, R.(2002). Providing new access to the general cur-riculum: Universal design for learning. TeachingExceptional Children, 35(2), 8–17.

Individuals with Disabilities Education Improve-ment Act (IDEIA) of 2004, P.L. 108–446, 108thCongress (2004).

Jackson, R. O. (1996). A foxfire gardening serviceproject for the elderly. Teaching Exceptional Chil-dren, 28(4), 64–69.

Janney, R. E., Snell, M. E., Beers, M. K., & Raynes,M. (1995). Integrating students with moderateand severe disabilities into general educationclasses. Exceptional Children, 61, 425–439.

Kleinert, H., McGregor, V., Durbin, M., Blandford,T., Jones, K., Owens, J., et al. (2004). Service-learning opportunities that include students withmoderate and severe disabilities. Teaching Excep-tional Children, 37(2), 28–34.

Krajewski, J., & Callahan, J. (1998). Service-learning:A strategy for vocational training of young adultswith special needs. The Journal for Vocational SpecialNeeds Education, 21(1), 34–38.

Krueger, R. A., & Casey, M. A. (2000). Focus groups:A practical guide for applied research (3rd ed.). Thou-sand Oaks, CA: Sage.

Mastropieri, M. A., Scruggs, T. E., Graetz, J., Nor-land, J., Gardizi, W., & McDuffie, K. (2005). Casestudies in co-teaching in the content areas: Suc-cesses, failures, and challenges. Intervention inSchool and Clinic, 40, 260–270.

McCarty, B., & Hazelkorn, M. (2001). Reflection:The key to social-emotional change using service-learning. Beyond Behavior, 10(3), 30–35.

Muscott, H. S. (2001). Service-learning and charac-ter education as “antidotes” for children withegos that cannot perform. Reclaiming Children andYouth: Journal of Emotional and Behavioral Problems,10(2), 91–99.

Nelson, B., & McFadden, D. (1995). A refuge forreal-world learning. Educational Leadership, 52(8),11–13.

No Child Left Behind Act of 2001, P.L. 107–110, 115Stat. 1425 (2002).

Orkwis, R. (1999). Curriculum access and universaldesign for learning. ERIC/OSEP Digest #E586. Re-


ston, VA: ERIC Clearinghouse on Disabilities andGifted Education. ED 437767

Patton, M. Q. (2002). Qualitative research and evalu-ation methods (3rd ed). Thousand Oaks, CA: Sage.

Pearman, E. L., Huang, A. M., & Mellblom, C. I.(1997). The inclusion of all students: Concernsand incentives of educators. Education and Train-ing in Mental Retardation and Developmental Disabil-ities, 32, 11–20.

Perkins, D. F., & Miller, J. (1994). Why communityservice and service-learning? Providing rationaleand research. Democracy and Education, 9, 11–16.

Pivik, J., McComas, J., & Laflamme, M. (2002). Bar-riers and facilitators to inclusive education. Excep-tional Children, 69, 97–107.

Rose, D. H., & Meyer, A. (2000). Universal designfor learning. Journal of Special Education Technology,15, 67–70.

Scruggs, T. E., & Mastropieri, M. A. (1996). Teacherperceptions of mainstreaming/inclusion, 1958–

1995: A research synthesis. Exceptional Children,63, 59–74.

Skinner, R., & Chapman, C. (1999, September).Service-learning and community service in K-12 publicschools. Washington, DC: U.S. Department of Ed-ucation, National Center for Educational Statis-tics.

Werts, M. G., Wolery, M., Snyder, E. D., & Caldwell,N. K. (1996). Teachers’ perceptions of the sup-ports critical to the success of inclusion programs.Journal of the Association for Persons with Severe Hand-icaps, 21, 9–21.

Yoder, D. I., & Retish, E. (1994). We did it: Servicelearning with special populations. Democracy andEducation, 9, 25–27.

Received: 14 June 2006Initial Acceptance: 28 July 2006Final Acceptance: 10 October 2006


Using Pivotal Response Training with Peers in SpecialEducation to Facilitate Play in Two Children with Autism

Laura R. Kuhn, Amy E. Bodkin, Sandra D. Devlin and R. Anthony DoggettMississippi State University

Abstract: This study evaluated the ability of peers in special education to implement pivotal response training(PRT) with two students with autism in order to increase social interactions. Peers were taught the strategiesusing modeling, role-playing, and feedback. After training, peers implemented PRT strategies with the childrenwith autism. Picture prompts were provided to assist peers in recalling the strategies, but were completely fadeduntil peers could implement the procedures with no instruction from observers. Increases in opportunities torespond were observed, as well as responses and initiations of social interaction by the children with autism.

In his original description of autism in 1943,Leo Kanner described problems related to thedevelopment of social relationships as inher-ent in the disorder (McConnell, 2002). Sincethis finding in 1943, researchers have thor-oughly explored this characteristic, suggestingmultiple forms of interventions to enhancesocial interactions among children with au-tism.

A significant body of research has focusedon peer-mediated interventions (Goldstein,Kacamarek, Pennington, & Shafer, 1992; Stah-mer, 1999; Goldstein & Ferrell, 1987; Garri-son-Harrell, Kamps, & Kravits, 1997). Peer-mediated interventions involve teaching peersspecific strategies to direct, respond, and re-inforce children with autism (Goldstein et al.,1992). These strategies enhance the commu-nicative interaction among children with au-tism, while minimizing the need for adult im-plementation and prompting. Additionally,the peers learn appropriate social behaviorwhile assisting other peers in developing acommunicative repertoire. However, manypeer-mediated interventions include specificscripts and limited toys or activities, whichmake generalization of these skills to new set-tings difficult.

A specific peer-mediated strategy developedby Koegel, Schreibman, Good, Cerniglia, Mur-phy, and Koegel (1989) has been effective inincreasing play behaviors in children with au-tism (Pierce & Schreibman, 1995; Thorp,Stahmer, & Schreibman, 1995). Pivotal re-sponse training provides a more naturalisticapproach to peer-mediated intervention (Mc-Connell, 2002). This strategy does not includespecific scripts with limited toys or activities.Rather, it promotes a variation of peerprompts to elicit a larger range of responsesfrom children with autism, thus promotinggeneralization and maintenance of interac-tions. Children with autism often engage inrepetitive behaviors with toys rather than thetoy’s common uses (Terpstra, Higgins, &Pierce, 2002), but with the implementation ofPRT children possess little opportunity to iso-late themselves and engage in repetitive be-haviors. Furthermore, initiating play is a be-havior not often observed among thesechildren and research has demonstrated thatPRT increases initiating behaviors (Pierce,1993).

Pierce and Schreibman (1995) suggest thatPRT is effective in the enhancement of socialinteractions among children with autismwhen implemented in a school setting by typ-ical peers. However, without an interventionin place with adult supervision, typical peersare most likely to select other typical peers asplaymates (Goldstein et al., 1992). Further-more, due to the need for a continuum of

Correspondence concerning this article shouldbe addressed to Sandra D. Devlin, Department ofCounselor Education, Educational Psychology, andSpecial Education, Mississippi State University, Box9727, Mississippi State, MS 39759.


Pivotal Response Training in Special Education / 37

placements for students with disabilities, somechildren with autism are included in a typicalclassroom for part of the day and spend theremainder of the day in a special educationsetting, while others may spend the bulk of theday in a special education classroom.

The purpose of this study was to evaluatethe ability of peers in special education toimplement PRT in the special education set-ting. Across-the-day interventions imple-mented by typical peers are promising (Strain& Hoyson, 2000). Thus, if PRT is imple-mented in both the regular and special edu-cation classrooms, the student(s) with autismwill have opportunities to engage in socialinteractions in both classes throughout theentire day.

Method

Participants

Participants were Colin, an 8-year-old Cau-casian male, diagnosed with autism, and Wil-son, a 7-year-old Caucasian male diagnosedwith autism, and the treatment agents: fivepeers in special education (two in groupone, three in group two). As pre-determinedby the Battelle Developmental Inventory(BDI), Colin obtained a total standard scoreof 65 on the communication domain, indi-cating deficits in both receptive and expres-sive language skills. These results were ob-tained when Colin was age 3 years, 4months. More current test results were un-available. Colin’s educational placement wasthe special education classroom for the ma-jority of the school day.

As determined by Project Memphis, a cri-terion-referenced instrument that assesses achild’s level of development in several areasincluding language, Wilson’s expressive lan-guage skills were the equivalent of a child of14 months when in fact he was at the age offour. On the BDI, Wilson obtained a stan-dard score of 65 on the communication do-main, equivalent to a child aged 13 months.More current test results were unavailable.Wilson also attended special educationclasses all day.

Of the peers chosen to participate in thestudy, one peer had an educational diagnosisof mild mental retardation, three had diag-

noses of specific learning disabilities, and onehad a diagnosis of developmentally delayed(see Table 1). Four of the peers chosen at-tended the same special education class as thetwo children with autism. Three spent thebulk of the day in the setting, while one peerattended only morning classes in the samespecial education classroom.

The research took place in an empty class-room in the children’s school in a rural south-eastern town. The special education teacherreported that the two students with autismengage in some self-stimulation and repetitivebehaviors if not under direct supervision, butengage in little to no social interactions unlessprompted by the teacher or other peers.

Materials

Training materials included a variety of toyswith which a small group of children couldplay. Toys included Legos®, cars and trucks,airplanes, blocks, and dinosaurs. These toyswere used during baseline and treatment ses-sions. Six picture prompts were used to teachpeers the strategies. Rewards (i.e., stickerchart and candy) were used when peers werecooperative in learning the strategies and suc-cessful at implementing the strategies duringtreatment. A video camera was used to tape allsessions.

Dependent Measures

Interaction opportunities. Opportunities tointeract, or peer prompts, were provided by

TABLE 1

Peer characteristics

Child Age Race Gender ED1 TISEC2

A1 7 African-American M MMR3 All dayA2 8 African-American M SLD4 2 hoursB1 6 African-American M DD5 All dayB2 8 African-American F SLD All dayB3 7 African-American M SLD All day

1 Educational Diagnosis2 Time in Special Education Classroom3 Mild Mental Retardation4 Specific Learning Disability5 Developmental Delay


the trained peers during the sessions. Numberof prompts provided were observed and re-corded.

Responses. The target children’s responsesto peer prompts were observed and recorded.A response was defined as a verbal, gestural, orphysical indication that the child understoodand/or answered the peer (e.g., answered apeer’s question, made eye contact whenprompted by peer, nodded to answer a peer’squestion, etc.).

Rate of responses to prompts. Rate of re-sponses to prompts was defined as the numberof responses divided by the number ofprompts presented.

Initiations. Initiations were defined as be-ginning a conversation with a peer without aprompt, or approaching a peer to play with apeer without a prompt (e.g., handing a peer atoy, helping a peer with an activity, etc.).

Data Collection and Experimental Design

Data collection was completed individually forWilson and Colin by reviewing video-tapedsessions. Protocols included opportunities torecord target children behaviors (verbal andphysical responses, initiations) and whetheror not prompts were provided by peers.

A multiple baseline design across peergroups was implemented. This design wasused to control for reactivity, such that targetchildren’s behavior did not change as a resultof an increased number of play sessions dur-ing baseline. Additionally, differences in be-havior among peer groups can be analyzedusing this design.

Data Collectors

Two investigators were present at all sessions.Both were graduate students supervised by afaculty member. The faculty member re-viewed all video-taped sessions for inter-rateragreement. All investigators were trained inpivotal response training, data collection pro-cedures, and procedures for collecting treat-ment integrity and inter-rater agreement.

Inter-rater Agreement and Treatment Integrity

Inter-rater agreement was calculated for 100%of the intervals. Reliability was calculated by

percent of agreement on each event (totalnumber of agreements divided by total num-ber of agreements plus disagreements). Vid-eo-tapes of sessions were reviewed by all inves-tigators for reliability. Percentage agreementfor all intervals of each behavior was 92%.Treatment integrity was assessed using achecklist of each step in the treatment (e.g.,prompts were provided by observers, promptswere faded, reinforcement was provided byobservers contingent upon each peerprompt). Treatment integrity was assessed on33% of the sessions by reviewing the video-tapes. Treatment was implemented with 98%integrity.

Procedure

The pivotal response techniques that were im-plemented by peers of children with autism inthe special education classroom were derivedand defined from the sets of procedures de-veloped by Pierce and Schreibman (1995) andKoegel et al. (1989). The strategies were mod-ified to facilitate comprehension by all stu-dents in the study. Prior to training, peers tobe included in the study were chosen based onfunctioning level and compliance.

During baseline, all toys were placed in themiddle of an empty classroom in the chil-dren’s school. Several different rooms wereused, dependent upon the school’s scheduleof activities. The rooms did not have desks;thus play space was blocked off in the centerof the rooms with cones or with a large rug.Students (two children with autism andgroups of two or three peers) were told to playwith the toys. Other than the boundaries ofthe play space, no other instructions orprompts were delivered during baseline. Base-line sessions lasted for ten minutes. At the endof the probe, students were instructed to helpput the toys away before they returned to class.

Peer Training

For the next several weeks, peers of the chil-dren with autism participated in 20-minuteintensive training sessions, two to three daysper week. Training ended when observersnoted mastery of strategies among the chil-dren, or demonstration of skills correctly 80%of time. The following behaviors and modified


definitions were selected by the authors fromPierce and Schreibman (1995) to teach thepeers to implement:

1. Paying attention. Ensure that the targetchild is attending before delivering aprompt (i.e., “Wilson, look at me.”).

2. Child’s choice. Offer an option of differentactivities to the child in order to maintainhis interest (i.e., “Would you like to playwith the airplane or the dinosaur?”).

3. Reinforce attempts. Verbally reinforce thechild after attempts at play or social inter-action (i.e., “I like the way you’re playingwith that car”).

4. Extend conversation. Ask questions or talkabout topics related to play (i.e., “Do youhave Legos® at home?”).

5. Turn taking. Model appropriate play andthen offer the child a turn (i.e., “This ishow you play with the car. Now, it’s yourturn.”).

6. Narrative play. Provide descriptions of playactions (i.e., “I’m flying this airplane to theairport.”).

The observer(s) held a picture prompt infront of the peers demonstrating each actionprior to modeling the behavior. After thepeers observed the prompt and the behavior,they practiced. Feedback was provided by theobservers until students demonstrated the be-havior correctly. They role-played with eachother and took turns playing the peer who wasto initiate an interaction with the child withautism. During each training session, strate-gies were reviewed until the peers could lookat the prompt and explain what he was sup-posed to do. Training lasted for eight 20-minute sessions. One peer moved to a differ-ent school and was replaced with another peerin Group A. Thus, Peer 1 of Group A receivedtwice the training that Peer 2 and the peers inGroup B received.

Implementation

Peers then began to generalize strategies to aplay setting for treatment implementation. Asin baseline, the group was told to play to-gether with the same toys. Observersprompted peers with the picture promptsfrom training when needed. Ten prompts

were delivered during the first session. Thenprompts were gradually faded until peerscould implement strategies independently bythe final treatment sessions. Contingent uponeach occurrence in which a peer engaged inan interaction and/or delivered a prompt to achild with autism, he received a sticker on achart. At the end of the session peers receiveda prize for earning ten stickers, which is anaverage of one interaction per minute. Treat-ment steps were identical for group two, ex-cept there were three peers in Group B in-stead of two. As in baseline, the sessions wereten minutes in length and were videotaped.

Results

Results of the study indicate improved socialinteraction for target students and peers. Pos-itive changes were noted for number of op-portunities for interactions, responses to peerprompts, and initiations of conversation andplay.

Results of the number of peer prompts, orinteraction opportunities, presented to Wil-son and Colin during baseline and treatmentwith Groups 1 and 2 are presented in Figure 1.For Wilson, opportunities to interact occurredan average of less than one time per sessionduring baseline with Group A. During treat-ment, peers offered prompts for social inter-actions an average of 16 times per session.Group B offered an average of less than oneopportunity for interaction per session duringbaseline, even though baseline was extendedfor Group B. During treatment, the groupincreased prompts to approximately four persession. Opportunities to interact, or peerprompts, were exhibited to Colin an averageof two to three times per session in baselineand 18 times during treatment with Group A.Group B offered approximately two promptsper session in baseline, and increased promptsto over three per session during treatment.

Results of Wilson’s and Colin’s responses topeer prompts are exhibited in Figure 2. WithGroup A, Wilson responded to peers onlyonce during baseline. During treatment, Wil-son responded an average of 13 times persession. Colin responded to prompts fromGroup A less than once per session duringbaseline and over 13 times per session duringtreatment. He responded to prompts from


Group B between one and two times per ses-sion during baseline, and two to three timesduring treatment.

Wilson’s and Colin’s rates of responses toprompts appear in Figure 3. Wilson re-sponded to peers an average of 20% of the

Figure 1. Total number of prompts.


time prompts were presented during baselinewith Group A. The rest of the prompts wereignored by Wilson. During treatment, Wilsonresponded to prompts an average of 84.16%

of the time prompts were presented. Colinresponded to prompts 18.7% of the time theywere presented during baseline with Group A,and 73.8% of the time during treatment. Dur-

Figure 2. Total number of responses to prompts.


ing play sessions with Group B, Colin re-sponded to 41.3% of the prompts during base-line and 70% during treatment.

Initiations of interactions exhibited by Wil-son and Colin are presented in Table 2. Wil-son exhibited only two total initiations across

Figure 3. Total number of responses.


five sessions during baseline of Group A. Dur-ing treatment, he exhibited a total of five ini-tiations across only three sessions. With GroupB, an increase of initiations was not exhibited.During baseline, Wilson initiated an interac-tion only once across seven sessions. He didnot initiate play at all during the three treat-ment sessions. Colin initiated only one inter-action during baseline of Group A, and in-creased to eight initiations during treatment.Colin displayed an increase in initiations afterPRT implementation of Group B by an aver-age of one initiation more per session. Duringbaseline, Colin initiated interactions an aver-age of two times per session. During treat-ment, he exhibited an average of 3.25 initia-tions per session.

Discussion

Results indicate that some peers with disabili-ties can successfully implement pivotal re-sponse training with children with autism. Par-ticularly with Group A, consisting of one peerdiagnosed with a specific learning disabilityand one with mild mental retardation, socialinteractions increased among target childrenand peers. The teacher described Group Bpeers as lower functioning and less coopera-tive than Group A peers. This factor may con-tribute to the difference in results betweenGroups 1 and 2. Wilson experienced moresignificant gains with Group B than did Colin.This finding may be explained by the differ-ence between functioning levels of Wilson andColin. Colin was described as more sociableand higher functioning than Wilson, possiblyhigher functioning than a couple of the peers

in Group B. Thus, one explanation for Colin’slack of significant gains with Group B is thathe possessed more social skills initially thanone or two of the peers in Group B. Anotherexplanation may be that Group B containedone more peer than did Group A. Perhaps thenumbers of peers in the groups influencedthe social gains of the target children. Futureresearch should examine the differences insocial interactions contingent upon peergroup size.

A reason for the increases in rates of re-sponding may be attributed to the fact thatpeers were taught to use different levels ofprompts: verbal, gestural, and physical. Thus,if a target child did not respond to an oppor-tunity during baseline, the peer ceased to tryto interact. However, during treatment if thetarget child did not respond to a verbalprompt the peer continued to prompt him byrepeating the prompt, using a gesture, orphysically helping the target child to respond.

These findings indicate that peers with dis-abilities can be successful at implementationof pivotal response training. Thus, childrenwith autism who attend both regular and spe-cial education classes can receive the trainingacross a school day. Perhaps lower functioningpeers selected to implement PRT need moreintensive training and more programming forgeneralization from training setting to playsetting. Perhaps some peers would benefitfrom more learning trials during the trainingof steps for implementation. Research relatedto the different methods of training for peerswith different disabilities would contribute topresent PRT research so that practitionersmay individualize peer training sessions.

After training implementation was termi-nated, the peers were observed in their class-room and continued to implement PRT withthe children with autism. A limitation of thisstudy is that more generalization data was notcollected. Thus, the long-term effects of PRTare not available. Future research should eval-uate the effects of PRT over time.

Teacher training of PRT strategies may ben-efit the peers and children with autism in thatbooster sessions could be conducted to en-hance the generalization and maintenance ofplay and social skills. Peers and children withautism both benefited from PRT in that they

TABLE 2

Initiations of Play

Child

Average initiations per session

BL1

Group ATx2

Group ABL

Group BTx

Group B

Wilson 0.4 1.7 .14 0Colin 0.2 2.0 2.14 3.25

1 Baseline2 Treatment


learned to interact with others in order toenhance play and social skills.

References

Garrison-Harrell, L., Kamps, D., & Kravits, T.(1997). The effects of peer networks on social-communicative behaviors for students with au-tism. Focus on Autism & Other Developmental Disabil-ities, 12, 241–254.

Goldstein, H., & Ferrell, D. (1987). Augmentingcommunicative interaction among children be-tween handicapped and non-handicapped pre-school children. Journal of Speech and Hearing Dis-orders, 52, 200–211.

Goldstein, H., Kaczmarek, L., Pennington, R., &Shafer, K. (1992). Peer-mediated intervention: At-tending to, commenting on, and acknowledgingthe behavior of preschoolers with autism. Journalof Applied Behavior Analysis, 25, 289–305.

Koegel, R. L., Schreibman, L., Good, A., Cerniglia,L., Murphy, C., & Koegel, L. (1989). How to teachpivotal behaviors to children with autism: A trainingmanual. University of California, Santa Barbara.

McConnell, S. R. (2002). Interventions to facilitatesocial interaction for young children with autism:Review of available research and recommenda-tions for educational intervention and future re-search. Journal of Autism and Developmental Disor-ders, 22, 351–371.

Pierce, K. (1993). Teaching an autistic child to re-spond to and instigate social initiations via pivotalresponse training. Unpublished data, Universityof California, San Diego. Pierce, K., & Schreib-man, L. (1995). Increasing complex social behav-iors in children with autism: Effects of peer-im-plemented pivotal response training. Journal ofApplied Behavior Analysis, 28, 285–295.

Stahmer, A. C. (1999). Using pivotal response train-ing to facilitate appropriate play in children withautistic spectrum disorders. Child Language Teach-ing & Therapy, 15(1), 29–40.

Strain, P. S., & Hoyson, M. (2000). The need forlongitudinal, intensive, social skill intervention:LEAP follow-up outcomes for children with au-tism. Topics in Early Childhood Special Education, 20,116–122.

Terpstra, J. E., Higgins, K., & Pierce, T. (2002). CanI play? Classroom-based interventions for teach-ing play skills to children with autism. Focus onAutism & Other Developmental Disabilities, 17, 119–126.

Thorp, D., Stahmer, A., & Schreibman, L. (1995).The effects of sociodramatic play training on chil-dren with autism. Journal of Autism and Developmen-tal Disorders, 25, 265–282.

Received: 14 June 2006Initial Acceptance: 25 August 2006Final Acceptance: 15 December 2006


Effects of Perspective Sentences in Social StoriesTM onImproving the Adaptive Behaviors of Students with Autism

Spectrum Disorders and Related Disabilities

Shingo Okada, Yoshihisa Ohtake, and Masafumi YanagiharaUniversity of Okayama

Abstract: This study examined the effects of adding perspective sentences to Social StoriesTM on improvingthe adaptive behaviors of students with autism spectrum disorders (ASD) and related disabilities. In Study1, two students with ASD read two different types of Social Stories: Social Story without perspectivesentences (SS without PS) and Social Story with perspective sentences (SS with PS). ABC or ABCA designswere used, with an SS without PS presented in the B phase and an SS with PS presented in the C phase.A visual inspection revealed that Social Stories were likely to be effective in reducing inappropriatebehaviors even without perspective sentences. In addition, adding perspective sentences appeared to haveno impact on further improving the target behaviors. In Study 2, a perspective sentence was added,characterized as specific, valuable, and contingent to a Social Story in the SS with PS condition. AnAA’BA’CA’ design was utilized, with a permanent visual step poster in the A’ phase, an SS without PSin the B phase, and an SS with PS in the C phase for a student diagnosed with attention deficithyperactivity disorder. A visual inspection revealed that adding a perspective sentence to a Social Storycontributed to further improvement of the target behavior. Based on these findings component andparametric analyses on Social Stories are recommended in future research.

Social StoriesTM is a highly appealing strategyfor improving the adaptive behaviors of stu-dents with autism spectrum disorders (ASD,Sansosti, Powell-Smith, & Kincaid, 2004). So-cial Stories use an individualized written shortstory with illustrations to help individuals un-derstand social situations where they oftenhave difficulty due to failure to understand ormisunderstanding important social cues(Gray, 2004).

The underlying belief is that many prob-lems exhibited by students with ASD in social

situations may be caused by their deficits inreading the social script or others’ thoughtsand feelings that are embedded in a givensocial situation. Thus, if the situations are de-scribed in ways that they can understand, theirproblem behaviors are believed to decreaseand their adaptive behaviors will increase(Myles, Trautman, & Schelvan, 2004; Sansostiet al., 2004). In other words, Social Stories arebelieved to serve as an interpreting bridge forstudents with ASD and related disabilities, de-lineating how people behave and peoplethink, and what social cues should be at-tended to social situations where they fre-quently have had or are likely to have (Gray,2004).

To serve this interpreting function, a SocialStory may include six types of sentences: de-scriptive, perspective, affirmative, directive,control, and cooperative (Gray, 2004). Briefly,descriptive sentences provide information aboutsocial rules governing the target situation andobjective facts or events occurring there. Per-spective sentences describe others’ thoughts orfeelings associated with the target situation.

This study is based on a thesis submitted by thefirst author, under the supervision of the secondand third authors, to the Faculty of Education atOkayama University for the master’s degree. Wethank Dr. Brenda Smith-Myles, at the University ofKansas, for giving us useful resources and comment-ing an earlier version of the manuscript. Correspon-dence concerning this article should be addressedto Yoshihisa Ohtake, the Faculty of Education at theUniversity of Okayama, 3-1-1 Tsushima-naka,Okayama-Shi, Okayama 700-8530, JAPAN. E-mail:[email protected]



Affirmative sentences emphasize a value under-lying a particular fact. Directive sentences pro-vide information about how to behave in thetarget situation. Control sentences consist of adescription developed by the student to helpretrieve important information in the targetsituation. And finally, cooperative sentences de-scribe who will help the individual and how tosucceed in the target situation.

In brief, Social Stories focus on describingsocial situations in which a target behavioroccurs, but not on directing how to behave.This is the essence that distinguishes SocialStories from other strategies such as directinstruction, visual cue card, or role playing.Therefore, descriptive sentences, perspectivesentences, or other sentences describing so-cial situations should be predominantly usedin a Social Story (Gray, 2004).

As the number of practitioners using SocialStories has increased, researchers have beenprompted to determine if this strategy is trulyeffective. For example, Kuttler, Myles, andCarlson (1999) successfully applied a SocialStory intervention to eliminate inappropriatevocalization and dropping to the floor by achild with ASD. In this study, an ABAB designwas utilized to demonstrate the functional re-lationship between the Social Story interven-tion and improvement in the target behavior.Similarly, Hagiwara and Myles (1999), em-ploying a multiple baseline across settings,demonstrated the effectiveness of a multime-dia Social Story intervention for three chil-dren with ASD in terms of improving hand-washing and on-task behaviors. Further,employing an ABAB design, Lorimer, Simp-son, Myles, and Ganz (2002) presented empir-ical evidence showing that a Social Story itselfwas responsible for reducing vocalization,which interrupted adult conversations andwas followed by tantrum, exhibited by a childwith ASD.

In yet another study, using a multiple base-line across subjects design, Scattone, Wilczyn-ski, Edwards, and Rabian (2002) demon-strated that Social Stories themselvescontributed to decreasing chair tipping, star-ing girls, and shouting by two children and anadolescent with ASD. Finally, using an ABABdesign, Bledsoe, Myles, and Simpson (2003)showed that a Social Story tailored for an ad-olescent who exhibited eating-related behav-

iors appeared to be responsible for reducingthese problem behaviors.

More recently, research has started address-ing which components of Social Stories areimportant for changing target behaviors. Forexample, Kuoch and Mirenda (2003) used anACABA design, with a children’s storybook inthe C phase and a Social Story in the B phase,to determine whether the content of SocialStories contributed improved social behaviorsor increased adult attention following readinga Social Story with adults. Results suggestedthat the content of the story, rather than in-creased adult attention, contributed to im-proving children’s adaptive behaviors.

Together, previous studies have revealedthat Social Stories themselves are effective inimproving various types of adaptive behaviorsof individuals with ASD. However, only onestudy has thoroughly examined which compo-nents of a Social Story intervention are re-sponsible for improving adaptive behaviors.That is, no studies have addressed how eachtype of sentences (e.g., descriptive, perspec-tive, affirmative, directive) contributes to theimprovement of adaptive behaviors, orwhether or not all components of Social Sto-ries are necessary to achieve the goal.

This study focused on the contribution ofperspective sentences. Typically, descriptivesentences are predominantly used in a SocialStory to describe what happens in the givensituation, when and how it happens, and whyit happens. In contrast, very a few sentencesare used to describe the perspectives of oth-ers. Therefore, it is unknown, for example, ifadding a few sentences to a Social Story todescribe the thoughts or feelings of otherscontribute to reduce problem behaviors orincrease desirable behaviors in individualswith ASD or related disabilities.

The following two research questions weredeveloped for this study:

(a) Is a Social Story effective in improvingadaptive behaviors even though the storydoes not include any perspective sen-tences?

(b) Does a perspective sentence contribute toimproving adaptive behaviors?

Effects of Perspective Sentences / 47

Study 1

Method

Participants

Two students, both of whom were enrolled ina same special school for students with cogni-tive disabilities, participated in Study 1.

Taro. Taro was a 12-year-old boy withmoderate mental retardation. Although hehad no referrals to a licensed pediatricianabout a diagnosis of autism, his score on theChildhood Autism Rating Scale (CARS) ad-ministered by the primary investigator, was30.5, indicating a mild to moderate level ofautism.

According to his main special educationteacher, Taro usually engaged in putting min-iature cars in a line and singing a favoritephrase of a commercial song during free time.Although Taro made eye contact and initiatedinteractions with teachers, his teacher de-scribed that he always used the same phrasesin a non-reciprocal manner. He used a verbalmode of communication to request and rejectobjects or social interactions.

According to his teacher, his reading andwriting levels were equivalent to first grade.However, he rarely utilized a daily schedulewith written words, specifically developed tohelp him move to the next activity smoothlyand independently. Social Stories had beenused to reduce his aggressive behaviors and tostay calm in an auditorium when this studycommenced. His teacher reported that hisproblem behaviors did not occur when thisstrategy was implemented.

Kenji. Kenji was a 13-year-old boy withautism and moderate mental retardation. Ac-cording to a Japanese version of the AdaptiveMaturity Scale (Asahide-gakuen-kyoiku-ken-kyu-sho & Nippon-shinri-tekisei-kenkyu-sho,1980), the level of his adaptive behavior wasequivalent to 7 years old. According to hisformer teacher’s records, he often had trou-ble with his peers because he did not knowhow to respond to their jokes. When he wasnot assigned a specific task, he tended to en-gage in repetitive behaviors such as hand-flap-ping and shoulder-patting. His main specialeducation teacher noted that he sometimestalked about his favorite topics such as local

cable TV or sign language. According to histeacher, his reading level was equivalent to thefourth grade. He did not use his personalschedule cards to regulate his behaviors. Hetook risperidone and fluvoxamine when thisstudy was conducted.

Settings

Observations of the two participants tookplace in each of the students’ classroom. Bothclassrooms were physically structured so as toprovide an individual work area, a group workarea, and a meeting area. Each classroom hadtwo special education teachers, who were incharge of five to six students.

Taro. The observational setting for Tarowas an area surrounded by shelves on twosides and a wall and a closet for dressing oneach of the remaining two sides. In the area,six desks with chairs for the students werearranged in a line. On the front wall, a visualreminder was posted to notify the “chair of theday.” During the morning meeting time, thechair of the day stood in front of the class-mates asked them to tell the date, the day ofthe week, the weather, the schedule, and thelunch menu. The remaining five students satin their chairs and responded to the direc-tions or questions made by the chair. One ofthe teachers was close to the chair to providenecessary support. The other teacher sat be-hind the remaining students and helped themparticipate in the meeting.

Kenji. Kenji was observed during breakswhen sitting in a sofa along a wall of his class-room. The sofa was 6 feet wide, so if onestudent lay down, there was no space left forother students to sit. During the break, theremaining five students typically engaged inindependent work, changing clothes (Theyhad two types of clothes; for commuting andfor studying) and free play. The two teachersin the classroom rarely interacted with stu-dents, but provided support necessary for stu-dents to complete their own tasks.

Target Behaviors

The primary investigator contacted formerand current teachers of the participatingstudents to identify the behaviors aboutwhich they were the most concerned. After


identifying several important behaviors, theinvestigator conducted direct observation todetermine which behavior occurred mostfrequently among the identified behaviors,and this, in turn, was selected as a targetbehavior for the present study.

Taro. For Taro, persistent and aggressiveverbal behaviors were selected as a target be-havior. A functional assessment revealed thatthese behaviors were related to his eagernessto be in the chair on Fridays. In his classroom,the chair of the day was rotated among the sixstudents the five school days, Monday throughFriday. Because the number of the studentswas not five but six, he was not allowed to bethe chair every Friday. Throughout the days ofthe week when he was not allowed to be in thechair on Friday, his persistent and aggressiveverbal behaviors dramatically increased, espe-cially before and during the morning meet-ing. For example, he removed the picture ofthe classmate that was put up to indicate thechair of Friday, instead putting his picture up.In addition, he repeatedly said to the class-mate who was Friday’s chair, “I’ll never makeyou the chair;” “You should be in the chair(the days other than Friday).” Sometimes, thisverbal behavior escalated, culminating in call-ing the classmate’s name out loud and repeat-edly saying, “You should not come to school,”or hitting the classmate.

Preventing aggressive verbal behaviors re-lated to the chair on Friday was deemed effec-tive in reducing hitting. Therefore, any of thefollowing behaviors were targeted to preventthe escalated behaviors:

● utterances related to his eagerness to be inthe chair on Friday (e.g., “I will be in theFriday’s chair,” “I will do it, I will do it, I willdo it . . .”).

● utterances related to negative attitudes to-ward the classmate who was the chair ofFriday (e.g., “[student’s name] is Boo [Boorepresents a sound of buzzer, meaning in-correct], ” “I will never make you the chair,”“Get out of here”).

Kenji. For Kenji, sitting neatly on a sofawas selected as a target behavior. He typicallyspent his free time sitting on a sofa. However,he tended to put his leg up on the sofa, to liedown there, or shake the sofa hard, blockinghis classmates from sitting there. Although his

teacher was not seriously concerned about thisbehavior, his mother had placed a high prior-ity on working on changing this behavior.

His target behavior was recorded as occur-ring when one or more of the following be-haviors were observed:

● placing his shoulder below the top line ofthe backrest on the sofa.

● placing either leg on the sofa.● exhibiting repetitive behaviors such as lock-

ing or hand-flapping.

Materials

Two types of Social Stories were developed foreach participant. One was a Social Story thatdid not include perspective sentences (here-after referred to as SS without PS); the otherwas a Social Story that included perspectivesentences (hereafter referred to as SS withPS). Gray’s Social Story guidelines do noteliminate the inclusion of the perspective ofthe individual from the concept of perspectivesentences (Gray, 2004). However, Gray as wellas the literature has emphasized the impor-tance of understanding of perspectives of oth-ers. This study, therefore, used the term “per-spective sentences” in ways that meant thethoughts and feelings of “others.”

Specifically, the perspective sentence in-cluded in the SS with PS for Taro was “Wheneverybody complies with the turn, everybody feelsgood because everybody can be fairly in the chair.”For Kenji the statement “So many people in thehotel thought I am cool. Many people around mewatched me do something and think I am cool” wasincluded in the SS with PS. Both Social Storieswere developed according to the guidelinesproposed by Gray (2004), which includes butare not limited to (a) use of positive expres-sions, (b) a ratio of 2:1 or more between sen-tences describing social situations and sen-tences directing how to behave, (c) use of theterms “usually” and “about” to describe thetarget social situation as accurately as possible,and (d) delineating of the target social situa-tion according to the functional assessmentresults. The text of the Social Story used foreach participant is included in the Table 1.

For Taro, the SS without PS consisted of sixpanels, and the SS with PS consisted of sevenpanels. Each page included two panels; there-


fore, one Social Story consisted of three orfour pages. One or two sentences and oneillustration were included in one panel. ForKenji, all texts in both types of stories wereprinted on one page, with three illustrationsexhibiting a point of “cool sitting” for each.

The story for Taro was written by the pri-mary investigator and modified based on feed-back from his teachers and the second author,whereas the story for Kenji was written by oneof his teachers and modified based on feed-back from the primary investigator and thesecond author. The teachers confirmed thatthe complexity and length of the stories wereappropriate for the two students.

The text and illustrations of the Social Storywere printed on A4 white paper, which waslater laminated and bound by a metal ring onthe upper left-hand side. The text, developedby PowerPoint 2002TM for Taro and by Ichi-taro 2005TM for Kenji, was placed on the up-per part of each panel. In the illustrations,icons were developed by Hanako 2005TM and

digital photographs were retrieved from freeinternet sites, with some modifications, usingPhotoshop 6.0TM. The illustrations wereplaced on the lower part of each panel forTaro and of a page for Kenji.

Data Collection

Taro. For Taro, Voice TrekTM, an inte-grated circuit (IC) recorder, was utilized torecord his persistent and aggressive verbal be-haviors. The recorder was set by his teachersin a closet, which was invisible to the studentbut allowed the recorder to catch conversa-tions effectively. The recorder was activatedsoon after Taro entered his classroom eachmorning. A 6-minute block was randomly se-lected from a 15-minute record in each ses-sion to be analyzed by the primary investi-gator, using a 10-second partial intervalrecording procedure. This observation wasconducted during only the weeks when Tarowas not in the chair on Friday.

TABLE 1

Social Stories

The Social Story for TaroWho is in the chair of Friday?I am Taro. I am a boy who loves Bob Sap. I am good at being in chair. I want to be in the chair of Friday.

But sometimes I am not always in the chair of Friday because the chair comes around in turn. Wheneverybody complies with the turn, everybody feels good because everybody can be fairly in the chair. The chair ofFriday this week is (student name). (student name) is good at being in chair, too. So, it is OK. I willcomply with the turn.

The Social Story for KenjiSitting in a cool wayThere are cool people and non-cool people in the world. Whether the person is cool or non-cool depends

on how the person does something. I am one of cool people because I write a lunch menu of the dayneatly on the blackboard and I never take a break during morning exercise. When I ate lunch at a hotel,I did in a good manner. So, many people in the hotel thought I am cool. Many people around me watch me dosomething and think I am cool. As for sitting, there are a cool way and a non-cool way. Sitting in a cool waymeans placing hip all the way, making body upright, and putting heels near the sofa. The points areshown in the following figure (line drawing is shown). I will take care of the way of sitting.

The Social Story for TatsunoriWhy do I have to wash my hands?I am Tatsunori, a boy who loves Giants. After toileting, people usually wash their hands. There are germs in

a bathroom. When people go to a bathroom, germs may be put on their hands. But, germs are invisible.If people wash their hands, most of the germs on their hands are gone. So, people wash their hands. Iwill wash my hands with soap after toileting. Ms. (teacher’s name) thinks the person is wise when the personwashes hands after toileting. Mr. (teacher’s name) thinks the person is cool when the person washes hands aftertoileting. Ms. (teacher’s name) thinks the person is lovely when the person washes hands after toileting.

Note. The italic parts are perspective sentences that are withdrawn in SS without PS and not withdrawn in SSwith PS.


Kenji. For Kenji, all observation sessionswere video-recorded. The camera was set in abasket on a teacher’s desk, which was invisibleto the student but allowed the camera to viewhis behaviors effectively. The camera was acti-vated immediately after Kenji sat on the sofaduring a break. The video was reviewed by thefirst author, using a 10-second partial intervalrecording procedure to determine the level ofinappropriate sitting behaviors. Each observa-tion was conducted for 10 minutes, for 5 daysa week, except when Kenji was absent fromschool.

Inter-Rater Reliability

The data recorded for Taro by the IC recorderin the baseline and Intervention 1 conditionswere accidentally deleted prior to a reliabilitytest was implemented. Therefore, no inter-rater reliability was computed for the two con-ditions. A graduate student served as a secondcoder. The primary investigator explained thedefinition of each target behavior and showedexamples and non-examples of the target be-havior until the second coder felt ready tocode. After the primary investigator coded thedata recorded by the IC recorder and video-tapes, the second coder independently codedthe data for 25.2% for Taro and 33.3% forKenji of all sessions across conditions with theexception of the baseline condition and theIntervention 1 condition of Taro, as men-tioned above. Percentages of inter-rater reli-ability were computed by dividing the numberof agreements by the sum of agreements plusdisagreements and multiplying by 100%. Reli-abilities ranged from 83.3% to 97.2% for Taro(M � 91.6%) and 93.3% to 100% for Kenji(M � 97.8%) and, therefore, were acceptable.

Experimental Design

An ABCA design was used for Taro, and anABC for Kenji, with an SS without PS in the Bcondition and an SS with PS in the C condi-tion. Some studies suggest that Social Storiesmay result in irreversible learning (Kuoch &Mirenda, 2003; Scattone et al., 2002), which inturn prevents us from using a withdrawalphase to reverse data to a baseline level. Im-plementing C followed by B may not allow usto detect the effect of including a perspective

sentence in a Social Story because understand-ing a social situation described by the SocialStory that includes a perspective sentence mayremain intact even after withdrawing the per-spective sentence from the Social Story.Therefore, to be able to detect the impact ofadding perspective sentence as clearly as pos-sible, we decided to implement B followed byC. Due to time constraints (i.e., the schoolyear ended), a withdrawal phase for Kenji wasnot implemented.

Procedure

The teachers for each participant were askedto deal with target behaviors as they did priorto this study. Anecdotal records taken weeklyby the primary investigator about the teach-ers’ behaviors indicated that both students’teachers behaved in the same ways contingenton the occurrence of the target behaviors re-gardless of conditions. That is, Taro’s teachersprovided verbal reprimands when his verbalbehavior escalated. Kenji’s teachers did notinteract with him even when he exhibited in-appropriate sitting. In addition, both teachersimplemented Social Story interventions asplanned.

Baseline. No instructions related to SocialStories were provided to the participants.

Intervention 1. This condition was identi-cal to baseline with the exception that a SocialStory was read. That is, an SS without PS wasread by one of the teachers for each partici-pant.

Before implementing Intervention 1, theprimary investigator explained the teachers,with modeling, the following four points re-lated to effective use of the Social Story: (a)the Social Story sessions should be imple-mented in a calm and safe atmosphere, (b) ateacher should sit side-by-side, but a little be-hind the student’s shoulder, (c) when thestudent asked a question about the story, theteacher could answer the question, (d) ac-cording to the progress of the student’s read-ing, prompts for reading should be faded out.The primary investigator then asked theteachers to engage in Social Story instructionsto determine if they implemented the proce-dure correctly. After making sure the teacherscorrectly implemented the procedure, the pri-


mary investigator asked them to initiate thefirst session.

In the first of the four sessions in Interven-tion, Taro’s teacher read the SS without PS forhim. Taro, then, took a turn to read the SocialStory out loud. His teacher provided control-ling prompt (i.e., modeling) when he read thestory incorrectly. The teacher for Kenji readthe SS without PS with giving examples andexplaining the meaning of the each sentence.Starting with the second session, both stu-dents read the SS without PS out indepen-dently. A formal reading time for SS withoutPS was provided once each day, just prior toinitiating observation of the target behavior.During the rest of the day, the Social Storybook was placed on a shelf to be accessible tothe student.

Intervention 2. This condition was identi-cal to Intervention 1 with the exception thatperspective sentences were added to a storyused in Intervention 1. The teachers providedstudents an SS with PS and asked them to readthe story. Because the content of the SS withPS was identical to that of the SS without PS,with the exception of a perspective sentence,Taro read the story with a few prompts at thebeginning of the intervention phase, followed

by no prompts from the second session. ForKenji, his teacher asked him to read the SSwith PS, explaining the meaning of the addedperspective sentence the first session of theintervention phase. He did not have any ques-tions, and he read the story independentlyfrom the second session.

Baseline 2. Taro’s did not provide him anyopportunities to access the Social Story book(SS with PS) throughout the day. No Baseline2 was implemented for Kenji due to the ter-mination of the semester.

Results

Figure 1 shows the change of Taro’s persistentand aggressive verbal behaviors across Base-line, Intervention 1, Intervention 2, and Base-line 2. As illustrated, during the first baselinephase, the percentage of his inappropriateverbal behaviors was high, for an average of49.1%, ranging from 19.4% to 86.1%. Follow-ing the implementation of SS without PS, hisinappropriate behaviors showed a downwardtrend, for an average of 10.6%, ranging from0.0% to 16.7%. It should be noted, however,that the difference in level between Baselineand Intervention 1 was not substantial. After

Figure 1. Percentage of 10-s intervals with inappropriate verbal behaviors during baseline (A), Social Storywithout perspective sentences (B), Social Story with perspective sentences (C), and withdrawal (A)for Taro.


initiating SS with PS, the level of his inappro-priate verbal behaviors remained low fromSession 10 to Session 13 (M � 6.6%), butrebounded in Sessions 14 and 15 (M �16.7%), resulting in an average of 9.7%, witha range from 0% to 16.7%. Immediately afterwithdrawing the SS with PS in Baseline 2, noinappropriate verbal behaviors occurred.

Figure 2 presents the results of observationsof Kenji’s inappropriate sitting across Base-line, Intervention 1, and Intervention 2. Asillustrated, during the baseline phase, he fre-quently exhibited inappropriate sitting, for anaverage of 72.3%, ranging from 52.5% to96.7%. Immediately after sitting on the sofa,he initiated repetitive behaviors, followed byputting his leg on the sofa or lying down. Incontrast, his inappropriate sitting behaviorsdisappeared immediately following introduc-tion of the SS without PS (M � 0.84%). Intro-ducing the SS with PS was immediately fol-lowed by high percentage and low percentageof the target behavior alternately (M � 23.9%,range � 0% – 73.5%).

Discussion

In Study 1, an ABCA or ABC pre-experimentaldesign was used for two students with ASD to

examine the effects of adding a perspectivesentence on reducing problem behaviors. Vi-sual inspection indicated that a Social Storywas effective in improving adaptive behaviorseven though the story did not include anyperspective sentences. Adding a perspectivesentence did not have any additional impacton improving the target behaviors.

“When everybody complies with the turn, every-body feels good because everybody can be fairly in thechair” was added as a perspective sentence toTaro’s SS without PS. “So, many people in thehotel thought I am cool. Many people around mewatched me do something and think I am cool”were added as perspective sentences to Kenji’sSS without PS. This was done based on thebelief that information was missing for eachparticipant to understand why he needed tobehave in a certain way and that consequentlyhis inappropriate behaviors would improvewhen he obtained the information. However,the results were different from our expecta-tion. One factor that may need to be addressedis motivation. In other words, it should be dis-cussed whether or not the information con-veyed by the perspective sentences can elevatethe motivation of participants to the level thatthey want to change the behaviors.

Figure 2. Percentage of 10-s intervals with inappropriate sitting behaviors during baseline (A), Social Storywithout perspective sentences (B), and Social Story with perspective sentences (C) for Kenji.


To discuss the motivational issue of perspec-tive sentences, introducing a framework ofestablishing operation may be helpful. Thetheory of establishing operation (Michael,2000) explains that some antecedent events orstimuli contribute to changing behaviors be-cause they change the value or meaning of theconsequence of emitting the behaviors. Add-ing a perspective sentence to a Social Storywould have impact on improving target behav-iors of individuals with ASD and related dis-abilities if the perspective described by thesentence had the capacity to elevate the valueof the consequence of engaging in appropri-ate behaviors. That is, if Taro obtained infor-mation about a contingency, “everybody feelsgood,,” by reading the story that did not have areinforcement value to him, he would notchange his behaviors. Similarly, if Kenji ob-tained information about a contingency,“many people in the hotel thought I am cool. Manypeople around me watched me do something andthink I am cool,” by reading the story, that wasnot a reinforcer to him, he would not changehis behaviors.

The fact that adding perspective sentencesdid not have any impact on the participants’target behaviors may be due partly to the lackof power of the sentence to enhance the valueof the consequence. Based on this theory,when developing perspective sentences, itshould be taken into consideration (a) notonly what perspective is missing to explainwhy the individual needs to behave in a cer-tain way but (b) also what perspective wouldenhance the value of the consequence contin-gent on engaging in the target behavior. Spe-cifically, we hypothesized that the perspectiveselected is a strong determinator of the effec-tiveness of a perspective sentence. For exam-ple, “Everybody” or “Many people,” which werethe terms used in this study, apparently werenot deemed specific or relevant enough to thestudent to affect behavior.

In addition to the nature of the perspectivesentences used in this study, drawbacks in theresearch design should be noted. We changedconditions prior to confirming the trend orstability of the data. We were forced to do sobecause the study did not start until the schoolyear was almost over. As a result, we were notable to clarify the functional relationship be-tween the two types of Social Story and de-

creased inappropriate behaviors. Second, andrelated to the first point, according to theanecdotal reports from the teachers of bothparticipants, students’ daily routines totallychanged at the time when SS with PS wasinitiated, specifically, prior to Session 10 forTaro and prior to Session 7 for Kenji. Becauseboth students were in their graduation year,they were required to participate in practicefor the graduation ceremony every day duringIntervention 2. This irregular schedule re-duced the length of recess and time for otheractivities that Taro was interested in, such ascooking, snack eating, and walking. For Kenji,the irregular schedule increased non-pre-ferred activities (i.e., a practice for the gradu-ation ceremony). This, in turn, likely contrib-uted to increases in the target behaviors. It isunknown, however, why withdrawing SocialStory instruction, which was implementedwhile the irregular schedule continued, re-sulted in complete suppression of Taro’s per-sistent and aggressive verbal behaviors.

These limitations and emerging hypothesesprompted us to conduct Study 2, which exam-ined the effects of the following: (a) adding aperspective sentence that was likely to en-hance the value of the consequence contin-gent on engaging in a target behavior, (b)keeping the same condition until the trend orstability was confirmed, and (c) conductingthe experiment while a daily routine re-mained stable. Study 2 is described in detailbelow.

Study 2

Method

Participants

Tatsunori, an 11-year-old boy, participated inthis study. Although he was enrolled in thesame special school as the students in Study 1,the possibility of disrupting his daily routinewas minimal because he was not in his gradu-ation year. Due to family issues, he lived in achild-care facility. He was diagnosed with at-tention deficit-hyperactivity disorder (ADHD)and mild mental retardation (IQ 68, tested byTanaka-Binet Intelligence Test) by a qualifiedpediatrician. Although he was capable of en-


gaging in complicated verbal communication,his reading level was equivalent to first grade.He often exhibited aggressive behaviors in re-sponse to directions or reprimands providedby his teachers. At the time of this study, hetook carbamazepine in the morning and atnight.

Target Behavior

Tatsunori’s target behavior was described as“Washing hands with water and soap after toi-leting without any prompts.” The selectionprocess of the target behavior was identical tothat of Study 1. That is, the target behavior wasselected based on the interview with Tatsuno-ri’s teachers and direct observations. Tatsuno-ri’s teachers provided verbal prompts when-ever they realized he did not wash hands aftertoileting. However, this procedure did notmake a positive change of this behavior.

Setting

The observation of Tatsunori’s target behav-ior was conducted at the bathroom nearest tohis homeroom. The sink where he was sup-posed to wash his hands was located closest tothe entrance. Typically, one or two other stu-dents were using the bathroom with a teacherwhen Tatsunori used the bathroom.

Materials

As in Study 1, two types of Social Story weredeveloped: one without PS and one with PS.The SS without PS consisted of six panels,whereas the SS with PS consisted of nine pan-els. The perspective sentences added to the SSwithout PS were “Ms. [teacher’s name] thinks theperson is wise when the person washes hands aftertoileting. Mr. [teacher’s name] thinks the person iscool when the person washes hands after toileting.Ms. [teacher’s name] thinks the person is lovelywhen the person washes hands after toileting.” Inthese sentences, the names of teachers whomTatsunori appeared to like and adjectives usedfrequently by these teachers were inserted.(The Table 1 shows the text part of bothSocial Stories.) The primary investigator wrotethe Social Stories, with feedback from Tat-sunori’s teachers and the second author. Tat-sunori’s teachers confirmed that the level in

complexity and length of the story matchedhis level of reading.

Data Collection

Tatsunori’s teachers took responsibility indata collection. Whenever he went to thebathroom and one of his teachers was avail-able, observation was conducted. Because theteacher-to-student ratio was 2:6, the teacherswere not always able to observe his hand-wash-ing behaviors. Contingent on his toileting, theteachers classified his hand-washing behaviorinto one of four categories: 3 � wash handswith water and soup, 2 � wash hands withwater only, 1 � wash hands with water andsoup with verbal prompts, 0 � did not washhands. One of the teachers (hereafter re-ferred to as main teacher) recorded the dataand sent them to the primary investigator.

Inter-Rater Reliability

When collecting functional assessment data,the primary investigator attempted to observeTatsunori wash his hands in a bathroom. How-ever, Tatsunori was extremely sensitive to beobserved by the investigator, leading to ver-bally aggressive behaviors and more difficultyengaging in hand-washing. When one teacherfollowed him to the bathroom, the other wasleft alone to take care of the remaining stu-dents in the classroom. This kept him verybusy and prevented another teacher fromserving as a rater. Therefore, we had to totallyrely on the data observed by one of the teach-ers.

Procedure

Tatsunori’s teachers were asked to deal withtarget behaviors as they had done prior to thestudy. Anecdotal records taken weekly by theprimary investigator about teachers’ behaviorsindicated that they behaved in the same waycontingent on the occurrence of the targetbehaviors regardless of condition. In addition,the records indicated that both teachers im-plemented Social Story interventions asplanned.

Baseline. In baseline, no Social Story inter-ventions were implemented.


Baseline Dash. This condition was identicalto baseline with the exception that a perma-nent visual step poster was added. In Session15, the main teacher voluntarily developed apermanent visual support poster and posted iton the wall over the sink. The poster delin-eated each step of hand-washing, accompa-nied by words and illustrations. We classifiedthis phase as Baseline Dash because addingthe visual cue was not in our original plan. Inaddition, the main teacher thought that usinga visual cue would be a natural and universalsupport and planned to use the poster contin-uously regardless of the Social Story interven-tions.

Intervention 1. This condition was identi-cal to Baseline Dash with the exception that aSocial Story was read. In this condition, themain teacher was asked to introduce an SSwithout PS. In the first session, the mainteacher provided Tatsunori with an SS withoutPS and asked him to read the story. The mainteacher found that Tatsunori independentlyread the story. When the teacher asked him ifhe understood the content of the story, hesaid, “I have no question.” Therefore, fromthe second session on, the main teacher sat alittle behind the student’s shoulder and hadno interaction with him, except for saying “itis time to read” at the beginning of the ses-sion. Although one Social Story session wasscheduled daily, the story book was placed on

a shelf that allowed Tatsunori to use it at anytime.

Baseline Dash 2. In this phase, the mainteacher stopped providing Tatsunori with ses-sions to read SS without PS. The teacher alsoeliminated the Social Story book from theshelf to prevent him from accessing the bookat any time.

Intervention 2. This condition was identi-cal to Intervention 1, except that the teachersread a Social Story from Intervention 1 towhich a perspective sentence had been added.

Baseline Dash 3. After confirming the pos-itive effect of SS with PS on Tatsunori’s targetbehavior, the main teacher withdrew the So-cial Story intervention, returning to the con-dition identical to Baseline Dash 1 and 2.

Experimental Design

A reversal design (AA’BA’CA’) was employed,with a baseline in the A Condition, a perma-nent visual step poster in the A’ Condition, anSS without PS � a permanent visual stepposter in the B Condition, and an SS withPS � a permanent visual step poster in the CCondition, to examine the influence of a per-spective sentence on the occurrence of thetarget behavior.

Figure 3. Score of hand-washing tasks during baseline (A), permanent visual step poster (A’), Social Storywithout perspective sentences (B), permanent visual step poster (A’), Social Story with perspectivesentences (C), and permanent visual step poster (A’) for Tatsunori.


Results

The results of the Social Story intervention arepresented in Figure 3. As illustrated, in thefirst baseline condition, Tatsunori neverwashed his hands with water and soap withoutprompts (M � 1.1, Range � 0 - 2). Whenintroducing a permanent visual step support,he independently washed his hands for twoconsecutive times. However, his behavior wasnot stable (M � 1.3, range � 0 - 3), and he didnot engage in hand-washing even with theteacher’s prompt in the last two sessions.

Initiating the SS without PS appeared tohave some impact on improving his hand-washing. Certainly, the average of his correctbehavior was improved (M � 1.9, Range � 0 -3); still the last two sessions showed a down-ward trend. After withdrawing the SS withoutPS, Tatsunori’s behavior was stable, averaging2.2, with a range from 2 to 3. His behavior wasmuch more improved immediately after theSS with PS was initiated. In almost all trials, heindependently washed his hands with soap,averaging 2.8, and ranging from 0 to 3. Thiseffect was maintained even after withdrawingthe SS with PS (M � 2.7, Range � 1 - 3).

Discussion

In Study 2, we examined the effect of addinga perspective sentence on the target behaviorof a student with ADHD, with consideration ofthe following three things. First, we created aperspective sentence that described a conse-quence that immediately followed the occur-rence of a target behavior and was likely toenhance the value of the consequence contin-gent on engaging in the target behavior. Sec-ond, we included sufficient sessions for eachcondition to clarify the trend or stability of thedata. Third, we implemented the experimentwhen the daily routine was not disrupted byschool events.

A visual inspection revealed that althoughthe Social Story without a perspective sen-tence contributed to improving Tatsunori’shand-washing, his target behavior did not con-stantly occur until a perspective sentence wasadded. The fact that positive and stable behav-ior change occurred immediately after an SSwith PS was introduced seems to prove thatadding a perspective sentence was responsible

for the behavior change. Anecdotal reportsfrom the main teacher showed that Tatsunorisaid “Shut up! You always tell me to washhands” in the last session of Intervention 1.However, in the first session in Intervention 2,he voluntarily informed his teacher that hehad engaged in hand-washing. This episodesuggests that he recognized that engaging inhand-washing was followed by positive feelingof his teacher toward him, which may havefunctioned as a reinforcer. This, in turn, sug-gests that we successfully selected a perspec-tive sentence that had capacity to enhance thevalue of consequence of hand-washing.

General Discussion

Previous research has used Social Storiesthemselves as the independent variable, butfailed to determine the effect of the compo-nents of the story. Considering that SocialStories include several types of sentences (e.g.,descriptive, perspective, directive), the strat-egy itself can be called “multi-component.”Unless we demonstrate empirical evidence ofwhat is important among the components tobetter change of a target behavior, we are notable to develop better Social Stories.

In this study, we conducted a preliminarycomponent analysis to narrow the gap be-tween the Social Story practices and existingempirical evidence. Specifically, we examinedthe effect of adding perspective sentences(i.e., sentences describing the thoughts andfeelings of others) on improving the adaptivebehaviors of students with ASD and relateddisabilities. The results in Study 1 suggest thatSocial Stories may have positive impact onreducing problem behaviors even if they donot include perspective sentences. The resultsof Study 2, using more rigorous design, cor-roborate the suggestion in Study 1. However,Study 2 presented one more suggestion: add-ing a perspective sentence boosted the im-provement of the target behavior if some pa-rameters of the perspective sentence werechanged.

Specifically, in Study 2, the perspective sen-tence was chosen by taking into considerationthat the sentence not only described why Tat-sunori needed to engage in hand-washing butalso that the sentence was likely to raise thevalue of a consequence contingent on engag-


ing in the target behavior. To raise the valuesof the consequence, we identified the personswho were relevant to or had established rap-port with Tatsunori, and then described thefeelings they might experience when he en-gaged in appropriate behaviors, using thephrases that they typically used. Retrospec-tively, we changed three parameters in theperspective sentence from Study 1 to Study 2:a specificity parameter, a time parameter, anda likeability parameter.

A specificity parameter has a continuum, rang-ing from a specific perspective to a generalperspective. Specific perspective sentences in-clude the thoughts and feelings of an individ-ual specific and relevant to the student. Forexample, Scattone et al. (2002) used “Ms Annwill be happy if I do not holler” (p. 542) intheir perspective sentences. Kuoch andMirenda (2003) used “(Interventionist) will bevery happy to see everyone playing games andhaving fun” “Mom will be happy if Henry eatsthe food” (p. 227).

In contrast, general perspective sentencesdescribe the perspectives of people in general(e.g., friends, we, people, everybody) who are notspecific or relevant to the student. For exam-ple, Thiemann and Goldstein (2001) used“Friends like to show each other what they aredoing,” “This means they want to show mesomething, and they like it if I look” in a SocialStory (p. 432). Similarly, Ivey, Heflin, and Al-berto (2004) used “Usually people like to havea party to celebrate” in their Social Story (p.170). Between the two loci of the continuum,“my friends” “my teachers” or “my neighbors”are placed. For example, Brownell (2002) in-cluded “If I say things that I heard on TV, myfriends might not know what I’m talkingabout” (p. 128).

With respect to a time parameter, two types ofperspectives may be used: non-contingent per-spective and contingent perspective. A non-contingent perspective uses others’ perspectivesthat might exist regardless of the occurrenceof a target behavior. For example, Thiemannand Goldstein (2001) used “Friends like play-ing with different toys and games,” “Friendslike to show each other what they are doing”(p. 432) in their perspective sentences. Theseperspectives may exist regardless of the occur-rence of contingent responses, securing atten-tion, initiating comments, and initiating re-

quests, which were selected as targetbehaviors. Similarly, a Social Story used byLorimer et al. (2002) included “Adults like totalk” (p. 56) as a perspective sentence. Theseperspectives may exist regardless of the occur-rence of interrupting vocalization or tantrum,which was targeted to be reduced.

In contrast, contingent perspectives describeothers’ perspectives that emerge immediatelyfollowing the occurrence or non-occurrenceof target behaviors. For example, Adams,Gouvousis, Vanlue, and Waldron (2004) used“Mom and Dad are sad when I get upset,”“When I use my quiet voice, Mom and Dad arehappy” for a perspective sentence. In thisstudy, crying, screaming, falling, and hittingwere selected as target behaviors. Similarly,Kuoch and Mirenda (2003) used “It makespeople very sad when Andrew doesn’t share”in their perspective sentences. For this partic-ipant, aggression, yelling, and crying were se-lected as target behaviors, which often fol-lowed sharing objects.

Finally, a likeability parameter shows a contin-uum of a reinforcement value of the personwhose perspective is described in a SocialStory, ranging from least to most valuable.Logically, the perspectives of people who arenot specific to the individual are likely to be ofless valuable to him or her. However, evenamong persons specific to the individual, thevalue of perspective varies, depending onwhose perspective it is.

For Taro, we used a perspective sentence(“When everybody complies with the turn, everybodyfeels good because everybody can be fairly in thechair”) that described the perspective of peo-ple who were not highly specific (specificityparameter) and not highly valuable (likeabil-ity parameter), and that emerged contingenton the occurrence of his target behaviors(contingency parameter). For Kenji, we usedperspective sentences (“So, many people in thehotel thought I am cool. Many people around mewatch me do something and think I am cool”) thatdescribe the perspective of people who werehighly general (specificity parameter) and,therefore, less valuable (likeablity parameter),and that existed regardless of the occurrenceof his target behaviors (contingent parame-ter).

In Study 2, we used a perspective sentence(“Ms. [teacher’s name] thinks the person is wise


when the person washes hands after toileting. Mr.[teacher’s name] thinks the person is cool when theperson washes hands after toileting. Ms. [teacher’sname] thinks the person is lovely when the personwashes hands after toileting”) that included aperspective of others who were specific (spec-ificity parameter) and valuable (likeability pa-rameter) to Tatsunori and that emerged con-tingent on engaging in his target behaviors(contingency parameter).

Theoretically, the perspectives of otherswho are specific and valuable to the studentare more likely to be meaningful. Further-more, describing others’ perspectives that oc-cur contingent on a target behavior are muchmore helpful in understanding the meaningof the behavior being exhibited (Gray, 2004).Therefore, using a perspective sentence de-scribing a specific, valuable, and contingentperspective is more likely to be effective inimproving target behaviors. However, the neg-ative side needs to be noted, too. For example,using the perspective of a person specific tothe individual may hinder generalization ofbehavior change. Therefore, if the goal of theintervention is to enable the individual to usea skill in a variety of situations, using a per-spective of a person less specific to the indi-vidual may be appropriate.

It cannot be emphasized enough that wedid not demonstrate the effectiveness of theparameter change in this study. In addition tochanging the parameters of Social Stories, wechanged participants from Study 1 to Study 2.Taro and Kenji in Study 1 were diagnosed withor scored in a range of autism, whereas Tat-sunori in Study 2 was diagnosed with ADHD.Adding a perspective sentence may havechanged Tatsunori’s hand-washing, not be-cause the parameters of the perspective sen-tences were changed to be specific, valuable,and contingent but because he was more mo-tivated by others’ perspectives than Taro andKenji, who was diagnosed with or consideredto have autism. Many more participants withautism spectrum disorders and related disabil-ities need to be studied to determine if usinga sentence including more specific, valuable,and contingent perspective is effective in im-proving any type of adaptive behaviors in anysituations. Not only component analyses (e.g.,the effect of perspective sentences) but alsoparametric analyses (e.g., the effect of speci-

ficity, contingency, and likeability parameters)warrant future research.

References

Adams, L., Gouvousis, A., Vanlue, M., & Waldron, C.(2004). Social Story intervention: Improving com-munication skills in a child with an autism spec-trum disorder. Focus on Autism and Other Develop-mental Disabilities, 19, 87–94.

Asahide-gakuen-kyoiku-kenkyu-sho, & Nippon-shinri-tekisei-kenkyu-sho. (1980). Shinpan S � Msyakai-seikatsu-nouryoku-kensa [Revised versionof social maturity scale]. Tokyo, Japan: Nippon-bunka-kagaku-sya.

Bledsoe, R., Myles, B. S., & Simpson, R. L. (2003).Use of a Social Story intervention to improvemealtime skills of an adolescent with Aspergersyndrome. Autism, 7, 289–295.

Brownell, M. D. (2002). Musical adapted Social Sto-ries to modify behaviors in students with autism:Four case studies. Journal of Music Therapy, 39,117–144.

Gray, C. (2004). The new Social Story book. Arlington,TX: Future Horizon.

Hagiwara, T., & Myles, B. S. (1999). A multimediaSocial Story intervention: Teaching skills to chil-dren with autism. Focus on Autism and Other Devel-opmental Disabilities, 14, 82–95.

Ivey, M. L., Heflin, J., & Alberto, P. (2004). The useof Social Stories to promote independent behav-iors in novel events for children with PDD-NOS.Focus on Autism and Other Developmental Disabilities,19, 164–176.

Kuoch, H., & Mirenda, P. (2003). Social Story inter-ventions for young children with autism spectrumdisorders. Focus on Autism and Other DevelopmentalDisabilities, 18, 219–227.

Kuttler, S., Myles, B. S., & Carlson, J. K. (1999). Theuse of Social Stories to reduce precursors to tan-trum behavior in a student with autism. Focus onAutism and Other Developmental Disabilities, 13,176–182.

Lorimer, P. A., Simpson, R. L., Myles, B. S., & Ganz,J. B. (2002). The use of Social Stories as a preven-tative behavioral intervention in a home settingwith a child with autism. Journal of Positive BehaviorInterventions, 4, 53–60.

Michael, J. (2000). Implications and refinements ofthe establishing operation concept. Journal of Ap-plied Behavior Analysis, 33, 401–410.

Myles, B. S., Trautman, M., & Schelvan, R. (2004).The hidden curriculum: Practical solutions for under-standing unstated rules in social situations. ShawneeMission, KS: Autism Asperger Publishing Co.

Sansosti, F. J., Powell-Smith, K. A., & Kincaid D.(2004). A research systhesis of social story inter-


ventions for children with autism spectrum disor-ders. Focus on Autism and Other Developmental Dis-abilities, 19, 194–204.

Scattone, D., Wilczynski, S. M., Edwards, R. P., &Rabian, B. (2002). Decreasing disruptive behav-iors of children with autism using Social Stories.Journal of Autism and Developmental Disorders, 32,535–543.

Thiemann, K. S., & Goldstein, H. (2001). Social Sto-ries, written text cues, and video feedback: Effectson social communication of children with autism.Journal of Applied Behavior Analysis, 34, 425–446.

Received: 20 May 2006Initial Acceptance: 22 July 2006Final Acceptance: 22 December 2006


Effectiveness of the Picture Exchange Communication Systemas a Functional Communication Intervention for Individualswith Autism Spectrum Disorders: A Practice-Based Research

Synthesis

Kai-Chien TienUniversity of Kansas

Abstract: This research synthesis verifies the effectiveness of the Picture Exchange Communication System(PECS) for improving the functional communication skills of individuals with autism spectrum disorders(ASD). The research synthesis was focused on the degree to which variations in PECS training are associatedwith variations in functional communication outcomes (Dunst, Trivette & Cutspec, 2002). The communi-cation consequences of PECS were examined in 13 studies, which included 125 participants with ASD who hadbeen identified as having limited or no functional communication skills. Claims that PECS is an effectiveintervention for improving functional communication skills appeared to be supported by the available researchevidence.

The purpose of this practice-based researchsynthesis was to verify the effectiveness of thePicture Exchange Communication System(PECS) for improving the functional commu-nication skills of individuals with autism spec-trum disorders (ASD). The synthesis focusedon the degree to which variations in the PECStraining are associated with variations in func-tional communication consequences. In gen-eral terms, a practice-based research synthesisdiffers from more traditional meta-analyses bysystematically examining and unpacking thecharacteristics of practices that are related todifferences in outcomes or consequences.Specifically, this type of analysis focuses moreon understanding how the same or similarcharacteristics exert the same or similar ob-servable effects and not solely on statistical orobservation-based relationships between oramong these variables. The reader is referredto Dunst et al. (2002) for a detailed explana-tion of this framework.

Individuals diagnosed with ASD share sig-nificant deficits in communication (American

Psychiatric Association, 2000). In order to ad-dress the communication challenges of chil-dren with ASD, behavior analysts, speech-lan-guage pathologists, and special educatorsincreasingly have turned to augmentative andalternative communication (ACC) (Frea, Ar-nold, & Vittimberga, 2001). The Picture Ex-change Communication System (PECS) is oneof such augmentative communication systemdesigned to increase functional communica-tion skills and potentially provide a bridge tospeech acquisition.

The Picture Exchange Communication Sys-tem been supported by a small number of casestudies and a large body of anecdotal litera-ture (Charlop-Christy, Carpenter, Le, Le-Blanc, & Kellet, 2002; Magiati & Howlin 2003;Mirenda, 2001; Mirenda & Erickson, 2000).Thus few published experimental studies havespecifically investigated the effectiveness ofPECS for children with ASD or other develop-mental disabilities. Furthermore, the majorityof the anecdotal literatures reviewing the ef-fectiveness of PECS were contributed by itsdevelopers, Andy Bondy and Lori Frost.

Bondy and Frost’s first published article(1993) on PECS detailed its use with childrenand adults with developmental disabilities.Specifically, they described the procedures

Correspondence concerning this article shouldbe addressed to Kai-Chien Tien, University of Kan-sas, 4227 Wimbledon Dr., Lawrence, KS 66047.


Functional Communication Intervention / 61

used to train school-based staff in Peru to usethe system over a five-month period. Althoughno formal data were collected, the school re-ported that of the 74 students who receivedthe PECS training, at the end of a 3-monthperiod, 28 were working on mastering Phase I,28 were working on mastering Phases II, and18 were working on mastering Phase III of thePECS training protocol.

In another study, Bondy and Frost (1994)reported outcomes for 85 noncommunicativepreschool children with ASD aged 5 years oryounger. While children’s cognitive abilitieswere not assessed, they were estimated as rang-ing from near-normal to profoundly disabled.Over 95% learned to use two or more pictureswithin the exchange format; almost alllearned at least one picture within one monthof starting the PECS. For the 66 children whoused PECS for more than a year, 41 were ableto use speech independently, whereas the re-maining 25 children were using a combina-tion of pictures or symbols and speech. Allchildren mastered using pictures or symbolsto communicate, although not all reached thehighest levels of PECS. Bondy and Frost(1994) also presented anecdotal data from anumber of single-case and small-group stud-ies. Most indicated encouraging results interms of increased spontaneous communica-tion and speech and, in some case, decreasedbehavioral problems.

The impact of PECS on problem behaviorswas also examined in several studies beyondthat of Bondy and Frost. For example, PECSwas reported as an effective intervention for a4-year-old with autism to decrease aggressivebehavior in a general education preschoolclassroom (Frea et al., 2001). Dooley, Wilczen-ski, and Torem (2001) reported a dramaticdecrease in problem behaviors and increase incompliance during transitions following PECSon a 3-year-old boy with a diagnosis of perva-sive developmental disorders (PDD).

Description of PECS Practice

Developed in 1985 by Andrew Bondy and LoriFrost, PECS originally was primarily used forpreschool-age children with ASD and othersocial communicative disorders who displayedno functional or socially acceptable speech(Frost & Bondy, 2002, pp. 46). The rationale

behind PECS is that the exchange of a picturefor a reinforcing item parallels the communi-cative exchange that takes place in typical con-versation (Bondy & Frost, 1993, 1994). ThePECS training is based on research and prac-tice in the principles of applied behavior anal-ysis. Thus, distinct teaching strategies, rein-forcement strategies, error correctionstrategies and generalization strategies are es-sential for teaching each skill (see Frost &Bondy, pp. 46-47).

PECS is different from other communica-tion systems in three main ways: (a) it does notrequire prerequisite skills; (b) it was designedto address the lack of motivation for socialreinforcement; and (c) it immediately teachesinitiating, instead of teaching responding be-fore initiating (Bondy & Frost, 1994). ThePECS training consists of six phases, which willbe described in detail in the following.

Phase I–“How” to Communicate. In thisphase, the terminal objective is that upon see-ing a “highly preferred” item, the child willpick up a picture of the item, reach toward thecommunicative partner, and release the pic-ture into the trainer’s hand (Frost & Bondy,2002, pp. 67). One trainer entices the childwith an object that is highly desired. As thechild reaches for the desired object, the sec-ond trainer, the facilitator, physically assiststhe child in picking up a picture for the de-sired object. The first trainer immediatelygives the child a reward along with an appro-priate comment, such as “Oh, you wantM&M!” when he/she receives the picture.

Phase II – Distance and Persistence. In thisstage, the exchange continues with attemptsto increase the child’s independence. Thus,the terminal objective is that the child goes tohis communication book where his picture isstored, pulls the picture off, goes to thetrainer, gets the trainer’s attention, and re-leases he picture into he trainer’s hand (Frost& Bondy, 2002, pp. 93).The child now is en-couraged to use greater spontaneity and per-sistence, and to generalize the skill he ac-quired. The facilitator is still available for as-needed assistance. Thus, the child learns toremove the picture from a display board forthe exchange and must engage in more phys-ical movement than in Phrase I in order toaccomplish the exchange. However, the child


is still encountering only one symbol on aboard at any one time.

Phase III – Picture Discrimination. The ter-minal object for this phase is that the childrequests desired items by going to a commu-nication book, selecting the appropriate pic-tures from an array, and going to a communi-cation partner and giving him/her the picture(Frost & Bondy, 2002, pp.123). In this stagethe child is asked to discriminate between sev-eral items on a board, choosing which item hewants, or which activities he wants to try. Thechild begins by answering forms of the ques-tion “What do you want?” but these are fadedquickly so the child will make choices sponta-neously as well as in response to a question. Asthe child becomes more comfortable makingdiscriminations, a third item may be added,and so on.

Phase IV – Sentence Structure. The terminalobjective is that the child requests present andnon-present items using a multi-word phraseby going to the book, picking up a picture/symbol of “I want,” putting it on a sentencestrip, picking out the picture of what shewants, putting it on the sentence strip, remov-ing the strip from the communication board,and finally approaching the communicativepartner and giving the sentence strip to him(Frost & Bondy, 2002, pp.159). Thus, thechild is taught to combine the object picturewith the carrier phrase “I want” on a sentencestrip and to give the strip to the adult orcommunication partner. The two pictures areattached to a sentence strip and the entirestrip is exchanged with the communicativepartner in return for the pictured item.

Phase V – Responding to “What do you want?”In this stage the child learns to respond to thequestion “What do you want?” by exchangingthe sentence strip. Thus, this phase extendsthe sentence structure begun in Phase IV. Useof the questioning phrase is deliberately de-layed until this phase because the exchangebehavior should be automatic by that point inthe programming sequence (Frost & Bondy,2002, pp. 209). Adjectives and other wordsmay be added to the child’s repertoire to helpher further refine her requests.

Phase VI – Commenting. In this finial stage,the child learns to respond to the questions“What do you want?” “What do you see?”“What do you have?” This phase makes a fun-

damental shift in the child’s communicationas well as the expected outcome from theteachers or peers. That is, it is designed tointroduce the child to commenting behavior,while the previous stages focused on request-ing behavior. Through the use of pictures for“I see,” “I hear,” “I smell,” etc., the child istaught to comment on elements of his/herenvironment.

Search Strategy

Search Terms

Relevant studies were identified by using thekeywords “PECS” and “Picture ExchangeCommunication System.” The term “autism”was used to further restrict the search. Fur-ther, an author search was conducted using“Andy Bondy” and “Lori Frost.”

Sources

A computer-assisted bibliographic search wasconducted. The Psychological Abstracts (Psy-cINFO), Educational Resources InformationCenter (ERIC) database, Expanded AcademicASAP, Wilson OmniFile, MEDLINE, Disserta-tion Abstract Online, Center for InternationalRehabilitation Research Information and Ex-change (CIRRIE), and REHABDATA were theprimary information databases searched forrelevant studies. An online search of the In-ternet via the Google search engine was alsoconducted. The reference lists of all acquiredsources were also reviewed. In addition, handsearches were completed for journal articles,book chapters, and books to locate additionalstudies of PECS that may have been omittedfrom the bibliographic search findings. Fi-nally, repeated sweeps of various sources weremade until no further studies could be lo-cated.

Selection Criteria

Studies were included in the research synthe-sis if they met all the following criteria: (a) thefocus of the study was to establish the effec-tiveness of PECS for improving functionalcommunication skills; (b) the PECS trainingwas described in sufficient detail to ascertainthat the intervention applied in a review study


was the same as the intervention describedunder Description of the Practice; (c) individ-uals involved in the study were diagnosed withASD; (d) communication consequences werethe major outcome measured; and (e) articleswere written in English.

Exclusion criteria. It was necessary to ex-clude on study (Cummings & Williams, 2000)that appeared to have met all the inclusioncriteria during the initial phase of the searchprocess. Close inspection of the study revealedthat the PECS training was only one compo-nent of the treatment so as to warrant itsexclusion.

Search Results

Eleven articles, including 13 studies and 125participants, met the selection criteria andwere included in the research synthesis. Table1 shows selected characteristics of the partici-pants. Table 2 lists the research designs usedin the studies, dependent measures, and thecharacteristics of the intervention.

Participants

The 125 participants who participated in thestudies all exhibited limited or no functionalcommunication skills (see Table 1). Agesranged from one to twelve years old at thebaseline assessment. Participants’ gender wasreported in 10 of the studies (77%). The vastmajority (65%) of the participants were re-ported as males (female � 36, male � 68).Across all the studies, participants’ ethnicitywas only reported in three studies (Charlop-Christy et al., 2002; Ganz & Simpson, 2004;Tincani, 2004).

Participants’ language age was reported infive studies (36%); however, different meth-ods were used for assessment and reporting.Eight studies (57%) did not report partici-pants’ language age, but provided descrip-tions of their speech abilities (Adkins & Axel-rod, 2002; Ganz & Simpson, 2004; Heneker &Page, 2003; Liddle, 2001; Schwartz, Garfinkle,& Bauer, 1998; Tincani, 2004). In the twostudies (Charlop-Christy et al., 2002; Jones,2005) that specifically reported expressive andreceptive language ages, the participants’ ex-pressive language ages ranged from 1.2 to 1.8years, with a mean age of 1.4; the participants’

receptive language ages ranged from 1.8 to1.9 years, with a mean age of 1.9 years.

Participants’ developmental age was re-ported in three studies (Anderson, 2002; Ganz& Simpson, 2004; Kravits, Kamps, Kemmerer,& Potucek, 2002), but different methods wereused for assessment and reporting. One study(Tincani, 2004) reported participants’ stan-dardized intelligence scores on the Develop-mental Profile-II. Nine studies (64%) did notreport any IQ, developmental ages, or otherrelated information on participants; however,one study (Schwartz et al., 1998) indicatedthat the participants were identified as havingcognitive delays.

Research Designs

Table 2 summarizes the research design em-ployed by the studies included in this synthe-sis. Twelve studies (92%) used single-partici-pant designs. One study used a retrospectiveanalysis of archival data to examine pre-/post-intervention outcomes (Schwartz et al., 1998).

Among the 12 studies employing single-par-ticipant designs, four types of research designswere employed. First of all, an AB or a varia-tion of the design was used in four studies(Ganz & Simpson, 2004; Heneker & Page,2003; Magiati & Howlin, 2003). Second, twostudies employed multiple-baseline designacross participants (Charlop-Christy et al.,2002; Jones, 2005) while one study used mul-tiple-baseline designed across settings (Kravitset al., 2002). Third, two studies used a chang-ing-criterion design to eliminate the need towithdraw the intervention and include severalinterventions subphases (Ganz & Simpson;Liddle, 2001). Last, an alternating-treatmentsdesign was employed in three studies compar-ing the effectiveness of PECS and sign lan-guage training (Adkins & Axelrod, 2002;Anderson, 2002; Tincani, 2004).

Three of the 12 single-participant studies(25%) reported follow-up data after post treat-ment (Charlop-Christy et al., 2002; Jones,2005; Schwartz et al., 1998). Length of timebetween post-treatment and follow-up rangedfrom 1 month to 12 months. Two studies con-ducted by Heneker and Page (2003) reportedfollow-up results but did not provide data.

In addition, all 12 single-participant studiesemployed outcome measures that require ob-


TABLE 1

Characteristic of Participants

Study Na

Gender

Ethnicity

ChronologicalAge

(years:months)

Language Age(years:months)/

Other Info.

DevelopmentalAge/IQ/

Other Info. DiagnosisM F

Adkins & Axelrod(2002) 1 1 0 N/A* 7

No functionallanguage N/A PDDb and ADHDc

Anderson (2002) 6 4 2 N/A 2–4

0:8–1:6(CDId

Production)0:8–1:4(CDI

Comprehension)

1:3–1:7Years

(Bayley)

a. Autismb. PDD

Charlop-Christyet al. (2002) 3 3 0

a. Ethiopian-American

b. Chinese-American

c. Korean-American

3:8–12

1:2(MCDIe Expressive)

1:9(PPVTf Receptive)

0:10–1:6(VABSg

CommunicationDomain)

N/A Autism

Ganz & Simpson(2004) 3 2 1

a. Asianb. African-

Americanc. Caucasian

3:9–7:2 No functionallanguage

0:7 Years(VABS)

for 1 child.2:4 Years(Battelle)

for1 child,N/A for1 child

a. Autismb. Developmental

delay with autisticcharacteristics

c. Developmentaldelay and speech/language impairment

Heneker & Page(2003) N/A N/A N/A 1–3 No functional

language N/A Autism

N/A N/A N/A 4–5 Limited functionallanguage

N/A Autism

Jones (2005) 5 3 2 N/A 5:1–8:20:10–1:10

(VABS AdaptiveCommunication)

N/A Autism

1:9–1:11(PPVT Receptive)

1:4–1:10(PLSh Expressive)

Kravits et al.(2002) 1 0 1 N/A 6

27% rank(WPPSI-Ri Verbal

Behavior)

2:8 Years(VABS) Autism

2–2:5 Years(PEP-Rj)

Liddle (2001) 21 N/A N/A N/ALimited or no

functionallanguage

N/Aa. Autismb. Severe learning

disabilities

Magiati &Howlin(2003) 34 29 5 N/A 5–10:6

0:8–2:8(VABS

CommunicationDomain)

N/A Autism

31 22 9 N/A 3–6

Limited or nofunctional

communicationskills

Cognitivedelay

a. Autismb. PDD-NOSk

c. Other develomentaldisabilities

Schwartz et al.(1998) 18 3 15 N/A 3:3–5:11 Nonverbal N/A

a. Autismb. Otherdevelopmentaldisabilities


servational coding. Interrater reliability datawere only presented in eight of the studies(67%), 5 studies (Adkins & Axelrod, 2002;Heneker & Page, 2003; Liddle, 2001; Magiati &Howlin, 2003) did not report reliability data.

Characteristics of Application of PECS

PECS was delivered in a variety of naturalsettings, such as homes and schools, across the13 studies. With two exceptions, interventionagents were reported (85%). PECS was imple-mented specifically by experimenters ortrained personnel in three studies (Adkins &Axelrod, 2002; Charlop-Christy et al., 2002;Ganz & Simpson, 2004), while the remainingeight studies utilized teachers or parents asthe intervention deliverers.

Selected characteristics of the PECS interven-tion implemented in each study are also pre-sented in Table 2. As illustrated, nine studiesindicated how many phases of the PECS trainingwere conducted; the remaining studies did not(Adkins & Axelrod, 2002; Heneker & Page,2003; Schwartz et al., 1998). In five of the studiesthat indicated PECS phases, the participants re-ceived the entire six phases of the PECS train-ing. The remaining four studies (Anderson,2002; Ganz & Simpson, 2004; Kravits et al., 2002;

Tincani, 2004) reported that the participantsreceived three or four phases of the training.

Treatment fidelity. Information regardingthe fidelity of treatment implementation wasreported in all the studies. However, fidelity datawere only reported in three studies; the remain-ing studies (77%) just stated that the treatmentwas implemented according to the proceduresdescribed in The Picture Exchange CommunicationSystem Training Manual (Adkins & Axelrod,2002; Charlop-Christy et al., 2002; Ganz & Simp-son, 2004; Kravits et al., 2002; Schwartz et al.,1998) or that the implementers were formallytrained to use PECS (Heneker & Page, 2003;Liddle; 2001; Magiati & Howlin, 2003). In thethree studies that reported fidelity data, two in-dependent observers coded sessions or sessionvideotapes to establish the implements’ degreeof adherence to the treatment manual content(Anderson, 2002; Jones, 2005; Tincani, 2004).For those three studies, the mean interobserveragreement for treatment fidelity was 93.86%(range from 91.675 to 96.8%).

Outcomes

Participants’ communication outcomes weremeasured in all the studies using a range ofdata collection methods and a range of re-

TABLE 1—(Continued)

Study Na

Gender

Ethnicity

ChronologicalAge

(years:months)

Language Age(years:months)/

Other Info.

DevelopmentalAge/IQ/

Other Info. DiagnosisM F

Tincani(2004) 2 1 1

a. African-American

b. Asian-American

5:10–6:8 No functionallanguage

54IQ(Develop-

mentalProfileII)for1child,

N/Afor1child

Autism

Total 125 68 36 1–12

a Number of participants.b Pervasive developmental disorder.c Attention deficit hyperactivity disorder.d MacArthur Communicative Development Inventory.e Minnesota Child developmental Inventory.f Peabody Picture Vocabulary Test.g Vineland Adaptive Behavior Scales.h Preschool Language Scale.i Wechsler Preschool and Primary Scale of Intelligence.j Psychoeducational Profile-Revised.k Pervasive developmental disorder-not otherwise specified.* N/A � Information not available in the article.


spondents. The majority of communicationoutcomes were measured through two individ-ual observation reports or through observa-

tional coding of sessions videotapes by twoindependent observers. Outcomes assessed in-cluded (a) observer reports of mastery rate of

TABLE 2

Characteristics of the Research Designs and Interventions

StudyResearchDesigna Dependent Measures

NaturalIntervention

SettingIntervention

AgentPECSPhase

AverageLength ofFollow-up Fidelityb Reliability

Adkins &Axelrod(2002)

(S)Alternativetreatments

a. Correct responses

Y Experimenters N/A* N/A M N/Ab. Mastery ratec. Most preferred

respondingtechnique

Anderson(2002)

(S)Alternativetreatments

a. Correct responses

N N/A I–III N/A M, O Y

b. Mastery ratec. Frequency of

initiationd. Behaviore. Eye contactf. Vocalization.

Charlop-Christyet al. (2002)

(S)Multiplebaseline

a. Spoken language

Y Therapists I–VI 10months M Y

b. Social-communicativebehavior

c. Problem behavior

Ganz &Simpson(2004)

(S)Changingcriterion

a. PECS acquisition

Y Experimenters I–IV N/A M Yb. Intelligible wordsc. Non-word

vocalization

Heneker &Page (2003)

(S)O1XO2O3

a. Amount ofcommunication

Y Teachers N/A 10months C N/A

b. Function ofcommunication

c. Methods ofcommunication

d. Level of adultsupport

(S)O1XO2O3

a. Amount ofcommunication

Y Teachers N/A 6months C N/A

b. Function ofcommunication

c. Methods ofcommunication

d. Level of adultsupport

Jones (2005)(S)

Multiplebaseline

a. Spontaneousrequests

Y N/A I–VI 1 month C, O Yb. Time delayc. Generalization

Kravits et al.(2002)

(S)Multiplebaseline

The frequency ofspontaneouslanguage

YTeachers

I–III N/A M YMother

Liddle (2001)(S)

Changingcriterion

PECS acquisition Y

Teachers

I–VI N/A C N/ASpeech andlanguagetherapist


PECS acquisition, (b) frequency of spontane-ous requests initiated by participants, (c)method and function of the participants’ com-munication, and (d) number of pictures andspoken words used by the participants follow-ing introduction of PECS. In addition, partic-ipants’ behavior outcomes were measured inthree studies, Anderson (2002), Charlop-Christy et al. (2002), and Magiati and Howlin(2003). The behavior outcomes examined in-cluded frequency of problem behaviors partic-ipants engaged in and the incidence of frus-tration showed by participants.

Synthesis Findings

Table 3 summarizes the findings of the syn-thesis regarding the communication out-comes of PECS reported across studies. Thesummary includes a description of the influ-ence of the PECS training on functional com-munication outcomes as reported in eachstudy. In addition, it contains informationabout the degree to which change in commu-nication status was demonstrated as a directresult of the PECS training. As illustrated,

there was little variation across the 13 studiesregarding the specificity of documenting ap-propriate implementation of PECS.

For purposes of the synthesis, studies thatlacked data demonstrating implementers’mastery of the PECS skills were categorized ashaving Low Specificity (N � 0; 0 %). Studiesthat provided evidence of implementers’ mas-tery of PECS skills but did not report anytreatment-fidelity procedures were catego-rized as having Moderate Specificity (N � 9;69%). Finally, studies that provided data re-garding implementers’ skill mastery and thetreatment fidelity procedures used were clas-sified as High Specificity studies (N � 4; 31%).The studies with High Specificity provided thestrongest evidence that change in communi-cation outcomes were a direct consequence ofthe PECS training.

Results

Communication consequence. Across thestudies, participants who received the PECStraining experienced positive gains in func-tional communication skills. Thus, communi-


StudyResearchDesigna Dependent Measures

NaturalIntervention

SettingIntervention

AgentPECSPhase

AverageLength ofFollow-up Fidelityb Reliability

Magiati &Howlin(2003)

(S)O1XO2O3

PECS acquisition Y Teachers I–VI 6 months C N/A

Schwartz et al.(1998)

Qualitativeresearch PECS acquisition Y Teachers I–VI N/A M N/A

(S)XO1O2O3O4

a. Spontaneousspeech

Y Teachers N/A 12months M Y

b. Communicativeformsc. Communicativefunctionsd. Communicativeacts

Tincani (2004)(S)Alternativetreatments

a. Motor imitationY Experimenters I–III N/A O Yb. Mands

c. Word vocalization

* N/A � Information not available in the article.a (S) � Single subject, O � Different assessment phases of a study, X � PECS intervention phase of study.b M � Implementer used intervention manual containing strategies of PECS, C � Implementer demon-

strated mastery of strategies to acceptable criterion, O � Objective observer coded observations for adherenceto PECS during sessions.

c Y � Reliability data reported.


TABLE 3

Major Findings

Study Communication Consequences Other ConsequencesReport of

GeneralizationaRelation to

PECSb

Adkins & Axelrod(2002)

The use of PECS produceda better acquisition rate,more spontaneoususage, and a highergeneralization rate thanthe use of sign languagefor the child with PDD.

The child appeared to be ableto achieve correspondenceto the presented object.

Y M

PECS was a more effectivemethod for the childwith PDD, and itgeneralized underdifferent conditions.

PECS was the preferredresponse method usedby the child.

The child began to imitatesome sounds and one-syllable words

Anderson (2002)

For the majority of thechildren, rates ofacquisition in the PECScondition wereuniformly faster thanrates of acquisition inthe sign languagecondition.

Five of the six childrendemonstrated more eyecontact in the sign languagecondition than in the PECScondition.

Y H

All of the childrenmastered more items inthe PECS condition thanin the sign languagecondition.

Several children beganvocalizing during treatmentin both the sign languageand PECS conditions.However, the three childrenwho vocalized duringcorrect responding did sosignificantly more often inthe sign language conditionthan in the PECS condition.

All the childrendemonstrated responsegeneralization in thePECS condition; fourout of the six childrendemonstrated skillgeneralization in thesign language condition.

Tantrum and avoidancebehaviors decreased for allthe children in both PECSand sign languageconditions.

Three of the childrenappeared to behaviorallyprefer PECS; the otherthree children appearedto behaviorally prefersign language.

Three of the six participantsdemonstrated significantincreases in positive affectin the sign languagecondition, while only onesignificant increase wasfound in the PECScondition.

All the children initiatedwith PECS more oftenthan with sign languagewhen both modalitieswere available.

Two children demonstratedmore self-stimulation in thesign language condition,while two children engagedin significantly more self-stimulated in the PECScondition.

The three children whoappeared to prefer signlanguage initiated thegreatest number of signtrials during the freechoices without PECSprobes.





PECSb

All the childrendemonstratedapproximately 30.7 morespontaneous initiationsper hour if they wereusing sign language thanif they were using PECS.

Charlop-Christyet al. (2002)

All three children masteredPECS use within arelatively short time.

A 70% or greater reduction wasobserved for 10 of 12problem behaviors, and fourproblem behaviors wereeliminated.

N M

All children showedincreases in spontaneous/imitation speech andmean length of utterance.

All children had collateralgain in social-communicative behavior,such as joint attentionand eye contact, followingthe PECS training.

Ganz &Simpson(2004)

All three children madeprogress in mastery of thePECS system anddemonstrated increases inaverage intelligible wordsspoken per trial.

Y M

All participants began PhaseI without word utterancesor speaking in one-wordutterances and endedphase IV speaking three-to four-word-phases.

All three children beganusing longer phrases andspeaking with morecomplex syntax by theend of the PECS training.

One child showed adecrease in non-wordvocalizations, while onechild showed an increase.The third one appearedrelatively stable.

Heneker &Page (2003)

The children were usingmore sophisticated formsof communication andneed less prompting to doso.

The children did not show anincrease in spontaneouslygaining the adult’s attention.

Y M

Requesting was the mainfunction ofcommunication.

The children appeared to showless frustration and were ableto accept the fact that theycould not always have whatthey had asked for.

The children were observedto use symbols as themain methods ofcommunication.

The children appeared tohave learned theimportance of needingsomebody’s attentionbefore communicatingwith them.





PECSb

The children werespontaneously attempting toexchange the symbols in 95%of cases.

Requesting was the mainfunction of communication.

The children were observed tomove towards using moreformal methods ofcommunication across allcontexts.

The children showed anincrease in spontaneouslygaining the adult’s attention.

N M

Spontaneous exchangeoccurred on average 96% ofthe time for all interactionsthat involved a symbol.

Jones (2005)

The PECS training was highlysuccessful for three of the fivechildren. They learned to usetarget utterances, made gainsin length and variability ofspontaneous utterances,generalized those gains acrossitems and people, andmaintained those gains.

Two children increased theirability to imitate targetutterances, while one childdecreased her ability toimitate.

Y H

One of the five childrenshowed only minimal change(no generalization and nochange in variability) inspontaneous speech. Theremaining one did not showgains in spontaneity, but didshow important gains in theprerequisite skill of imitation.

Four out of the five childrenshowed an increase in theirmean length of utterances fortraining items after the PECStraining.

Three children showed anincrease in word variationduring the PECS and time-delay sessions. One childshowed no increases in newword use after thepreliminary increases seen inbaseline, while one childintroduced new words duringeach of the procedural phasesof the study and followup.

Four of the five children beganto use target utterancesspontaneously during PECSand time-delay sessions. Onlyone maintained this abilityduring followup, and onlyone generalized this abilityacross communicativepartners.

One child significantlyincreased his ability toexpressively label attributes;the remaining four childrenremained level throughPECS.





PECSb

Kravits et al.(2002)

The child demonstratedsuccessful use of PECS.

Y M

The child’s spontaneouslanguage, which includesverbalizations and iconuse, increased with theintervention.

The child demonstratedsignificantly moreinitiations andverbalizations duringintervention sessions thanduring baseline session.

The child did notsignificantly increase therange of spokenvocabulary duringintervention.

The duration of the child’speer interactionssignificantly increased.

Liddle (2001)

Twenty of the childrenlearned to use PECS torequest items, and onechild failed to achievephase I.

Y M

Eleven out of 20 childrenwho learned to use PECSlearned to use sentencestrips to request items.The remaining ninechildren improved intheir ability to interactwith others by being ableto initiate requesting.

Nine out of 21 childrenwere been observed tohave increased theirattempts at spokenlanguage.

Magiati &Howlin(2003)

The children showedsignificant improvementsin their use of PECS, withthe level of PECS,frequency of the PECSuse, and extent of PECSvocabulary all increasingover time.

There was a significantreduction in the children’stotal score on the RimlandAutism treatment EvaluationCheck-list (from a mean of74.9, SD � 20.98 to 65.1, SD� 20.89, t � 3.91, p � .001),indicating an overallimprovement in problembehaviors.

Y M

The children were found toshow an improvement intheir overall level ofcommunication. Therewere increases in thechildren’s use of theother forms ofcommunication, such assigns/gestures, spokenwords and phrases.

Changes in the less speechgroup were slow to occur,but became marked overtime. For children withhigher level of speechlarge gains occurredinitially, but these tendedto be plateau out.


cative behavior change was documented in allstudies. The most commonly reported com-munication consequences of PECS included:(a) successful use of PECS as a communica-tion tool (reported in 100% studies); (b) anincrease in overall level of communicationand language (reported in 62% studies); (c)an increase in spontaneous language/speech/imitation (reported in 46% studies);(d) an increase in initiations of communica-tion (reported in 31% studies); and (e) anincrease in mean length of utterance (re-

ported in 23% studies). Furthermore, thestudies that included a follow-up assessment in-dicated maintenances of the gains identified atthe time of posttesting (Charlop-Christy et al.,2002; Heneker & Page, 2003; Jones, 2005; Ma-giati & Howlin, 2003; Schwartz et al., 1998).

The studies that compared PECS againstsign language training demonstrated that(a) rates of acquisition in PECS were fasterthan rates of acquisition in sign language,(b) PECS was the preferred method of com-munication for most participants compared




PECSb

Schwartz et al.(1998)

14 months on average (range �3-28 months) after thebeginning of PECS training,children were using the PECSsystem in a functional mannerto communicate with adults andpeers. Y M

The children mastered thefundamental PECS protocolwithin 11 months, and learnedto exchange with peers in anadditional 3 months.

Children who learned PECS usethe system across settings. Forty-four percent of the childrenacquired unprompted, non-echolalic spokencommunication, and allchildren demonstrated manysuccessful communicativeinteractions across trained anduntrained functions andsettings.

Y M

Children who received training inone communicative functiondemonstrated increased use ofdifferent untrainedcommunicative functions.

Tincani (2004)

One child with weak hand-motorimitation skills learned PECSmore rapidly than signlanguage. On the other hand,another child with moderateimitation skills learned signlanguage more rapidly thanPECS. Y H

Sign language training producedmore vocalization for bothchildren; however, a proceduralmodification to the PECSsystem increased one child’svocalization to a level similar tothat in sign language training.

a Y � Yes, N � No.b H � High specificity, M � moderate specificity, L � Low specificity (see text for a more detailed description

of specificity ratings).


to sing language, and (c) PECS was associ-ated with significantly greater improvementsfor the participants without hand-motor im-itation relative to the sign language training(Adkins & Axelrod, 2002; Anderson, 2002;Tincani, 2004).

Other consequences. Positive behavioralchange was documented in three of the stud-ies (Anderson, 2002; Charlop-Christy et al.,2002; Magiati & Howlin, 2003). The mostcommonly reported behavior consequencewere significant reductions in problem behav-iors.

Rival Explanations

A number of rival explanations might explainthe positive findings reported in the studiesreviewed in this synthesis. However, many ofthese rival explanations can be refuted as aconsequence of the generally high quality ofthe research designs.

First, the fact that studies typically employedobservational coding as the primary methodof data collection could have resulted in ob-server or rater bias. However, every study em-ployed two independent observational data re-corders and sections of different interventionphases (e.g., baseline, intervention). More-over, with strong reliability data reported foreight of the 13 (62%) studies available, con-cerns that measurement variations in mater-nal interactions were a result of observer biasare minimized.

Second, the positive changes of communi-cation consequences may have been emergedas part of maturation. However, this possibilityis mitigated by the fact that many behavioral,social, and communicative deficits exhibitedby children with autism spectrum disorders donot spontaneously remit over time if un-treated (American Psychiatric Association,2000). In addition, 12 out of 13 studies used asingle-subject design, and single-subject de-sign methodology establishes the casual rela-tionship between treatment and outcomes byas series of intrasubject or intersubject repli-cations of treatment effect (National ResearchCouncil, 2001). Furthermore, in single-sub-ject, multiple-baseline designs, participantsserve as their own control group. The pres-ence of a control group can serve to separate

the effects of maturation and those of treat-ments.

Third, the fact that the experimenter him/herself implemented the treatment in severalof the studies could lead to problem with ex-perimenter bias; that is, the experimentermight influence the participant’s responses.However, this concern is mitigated somewhatby the fact that in most of the studies reviewed(62%), the person implementing PECS wassomeone other than the experimenter.

In summary, a number of common threatsto internal validity were addressed within theresearch designs of these studies. Therefore,despite such potential threats to the general-izability of the practice, the synthesis findingssupport the effectiveness of the PECS training.

Conclusion

The primary focus of this synthesis of the lit-erature was to summarize findings regardingthe effectiveness of PECS for enhancing thefunctional communication skills of childrenwith ASD. In brief, the evidence for the inter-vention’s effectiveness was provided by studiesthat (a) assessed the level of adherence to astandardized treatment protocol (i.e., treat-ment fidelity); (b) utilized appropriate andwell-executed research designs; (c) used mea-sures with well-established reliability to assessoutcomes; (d) replicated finding across partic-ipants; and (e) employed a follow-up compo-nent to demonstrate the stability of treatmenteffects. Taken as a whole, therefore, results ofthe studies reviewed provide evidence for theeffectiveness of PECS; specifically, PECS is ef-fective in enhancing functional communica-tion skills of individuals with ASD. Therefore,PECS is recommended as an evidence-basedintervention for this purpose.

Nevertheless, several points should be con-sidered regarding recommending of PECS asan evidence-based intervention. First, most ofthe participants of the studies included in thissynthesis were male (65%), which most likelyis a reflection of the differential prevalencerates of autism across genders. The pattern ofgains exhibited by the participants was similarfor males and females, which indicates thatthe PECS training had the same effect regard-less of gender.

Second, children in the studies were diag-


nosed as having ASD. Therefore, PECS canonly be recommended as an evidence-basedintervention for individual with ASD, ratherthan for individuals with other diagnoses. Fur-ther research involving individuals with othertypes of diagnoses will be needed to deter-mine whether or not PECS is effective as afunctional communication intervention forother populations.

Implications for Practice

For practitioners working the children withASD, there are two primary implications forpractice that can be derived from this researchsynthesis. First, PECS training can easily beincorporated into an individual’s usual rou-tine without requiring large-scale changes toclass or home routines. Thus, similar pro-grams may successfully be implemented byteachers and parents. Second, the value ofPECS may lie not only in its ability to enhancecommunication skills initially, but also to fa-cilitate easy maintenance and application tonew situations. This is especially critical whenconsidering that other interventions some-times require constructed environments and,therefore, are not likely to generalize outsideof specially designed environments.

In summary, the evidence reviewed in thissynthesis supports claims the PECS is effectivein enhancing functional communicationalskills of children with ASD. In addition, theimplications derived from this synthesis sug-gested that PECS can be easily integrated intoan individual’s usual routing and that theskills acquired from PECS training can bemaintained and generalized across differentsituations. Insofar, PECS is recommended asan evidence-based intervention for enhancingfunctional communication skills of individualswith ASD. However, further research involvingindividuals with other types of disabilities isrecommended.

References

Adkins, T., & Axelrod, S. (2002). Topography-versusselection-based responding Comparison ofmands acquisition in each modality. The BehaviorAnalyst Today, 2, 259–266.

American Psychiatric Association. (2000). Diagnostic

and statistical manual of mental disorders (Rev. ed.).Washington, DC: Author.

Anderson, A. E. (2002). Augmentative communica-tion and autism: A comparison of sign languageand the Picture Exchange Communication Sys-tem. (Doctoral dissertation, University of Califor-nia, San Diego, 2001). Dissertation Abstracts In-ternational, 62(9-B), 4269.

Bondy, A., & Frost, L. (1993). Mands across thewater: A report on the application of the Picture-Exchange Communication System in Peru. TheBehavior Analyst, 16, 123–128.

Bondy, A., & Frost, L. (1994). The Picture ExchangeCommunication System. Focus on Autistic Behavior,9(3), 1–19.

Charlop-Christy, M., Carpenter, M., Le, L., LeBlanc,L., & Kellet, K. (2002). Using the Picture Ex-change Communication System (PECS) with chil-dren with autism: Assessment of PECS acquisi-tion, speech, social communicative behavior, andproblem behavior. Journal of Applied Behavior Anal-ysis, 35, 213–231.

Cummings, A., & Williams, W. (2000). Visual iden-tity matching and vocal imitation training withchildren with autism: A surprising finding. Journalon Developmental Disabilities, 7, 109–122.

Dooley, P., Wilczenski, F. L., & Torem, C. (2001).Using an activity schedule to smooth school tran-sition. Journal of Positive Behavior Analysis, 35, 213–231.

Dunst, C. J., Trivette, C. M., & Cutspec, P. A. (2002).Toward an operational definition of evidence-based practice. Centerscope, 1(1), 1–10.

Frea, W. D., Arnold, C. L., & Vittimberga, G. L.(2001). A demonstration of the effects of aug-mentative communication on the extreme aggres-sive behavior of a child with Autism within anintegrated preschool setting. Journal of Positive Be-havior Intervention, 3, 194–198.

Frost, L., & Bondy, A. (2002). PECS: The PictureExchange Communication System training manual.Newark, DE: Pyramid Educational Products Inc.

Ganz, J. B., & Simpson, R. L. (2004). Effects oncommunicative requesting and speech develop-ment of the Picture Exchange CommunicationSystem in children with characteristics of Autism.Journal of Autism and Developmental Disorders, 34,395–490.

Heneker, S., & Page, L. M. (2003). Functional com-munication: The impact of PECS. Speech & Lan-guage Therapy in Practice, Autumn, 12–14.

Jones, C. M. (2005). Using Picture Exchange Com-munication System and time delay to enhance thespontaneous speech of children with Autism.(Doctoral dissertation, Claremont GraduationUniversity, 2004). Dissertation Abstracts Interna-tional, 65(8-B), 4270.

Kravits, T. R., Kamps, D. M., Kemmerer, K., & Po-


tucek, J. (2002). Brief reports: Increasing commu-nication skills for an elementary-aged studentwith Autism using the Picture Exchange Commu-nication System. Journal of Autism and Developmen-tal Disorders, 32, 225–230.

Liddle, K. (2001). Implementing the Picture Ex-change Communication System (PECS). Interna-tional Journal of Language and Communication Dis-orders, 36, 391–395.

Magiati, I., & Howlin, P. (2003). A pilot evaluationstudy of the Picture Exchange CommunicationSystem for children with Autistic Spectrum Disor-der. Autism: The International Journal of Researchand Practice, 7, 297–320.

Mirenda, P. (2001). Autism, augmentative commu-nication, and assistive technology: What do wereally know? Focus on Autism and Other Developmen-tal Disabilities, 16, 141–145.

Mirenda, P., & Erickson, K. A. (2000). Augmenta-tive communication and literacy. In A. M. Weth-

erby & B. M. Priznang (Eds.), Autism SpectrumDisorder: A transactional approach (pp. 333–369).Baltimore: Paul H. Brookes Publishing Co.

National Research Council. (2001). Educating Chil-dren with Autism. Washington, DC: National Acad-emies Press.

Schwartz, I. S., Garfinkle, A. N., & Bauer, J. (1998).The Picture Exchange Communication System:Communicative outcomes for young childrenwith disabilities. Topics in Early Childhood SpecialEducation, 18, 144–159.

Tincani, M. (2004). Comparing the Picture Ex-change Communication System and sign lan-guage training for children with Autism. Focus onAutism and Other Developmental Disabilities, 19, 152–164.

Received: 3 May 2006Initial Acceptance: 28 June 2006Final Acceptance: 19 December 2006


Preschool Teacher Perceptions of Assistive Technology andProfessional Development Responses

Julia B. Stoner, Howard P. Parette, Emily H. Watts, and Brian W. WojcikIllinois State University

Tina FogalBloomington District 87 Schools

Abstract: This study investigated the perspectives of teachers in an early childhood center concerning theirthoughts and feelings about the implementation of a school wide assistive technology (AT) program designed toenhance emergent literacy skills for children identified as being at-risk or having special needs. Qualitativemethodology was used to gain perspectives of all participants. Semi-structured interviews, observations in theclassroom, and a self-assessment of AT knowledge and practice were used to collect data. Data were analyzedusing a multiple coding approach resulting in identification of four major themes: (a) perceptions of technology,(b) perceived challenges to implementing technology, (c) perceptions of AT and literacy and (d) self-reported ATuse. Discussion and recommendations focus on best practices of implementing school wide AT in early childhoodsettings.

Young children with disabilities have increas-ingly been placed in child care and preschoolsettings since the passage of the Individualswith Disabilities Education Act of 1997 (IDEA97; National Early Childhood Technical Assis-tance Center, 2003). To serve these childrenappropriately in such settings, an individualeducation program (IEP) is developed foreach child in which assistive technology (AT)devices [20 U. S. C. § 1401(1)] and servicesmust be considered [20 U.S.C. 1401(2)]. Bothdevices and services can result in a range ofdevelopmental benefits for young childrenwith disabilities (Drasgow, Yell, & Robinson,2001; Mistreet, Lane, & Ruffino, 2005; Na-tional Association for the Education of YoungChildren [NAEYC], 1996; Reed & Bowser,2005; Wiekle & Hadadian, 2003). However,the potential of AT to benefit these children iscontingent on “thoughtful integration into

the early childhood curriculum” (NAEYC, p.3).

While IDEA 97 provides a clear require-ment for ‘thoughtful’ consideration of AT indeveloping young children’s service plans, theNo Child Left Behind Act of 2001 (NCLB)also provides additional demands for teachersby (a) requiring that children with disabilitiesbe taught to the same standards as all children,and (b) holding schools accountable for stu-dent achievement. However, “taking full ad-vantage of their rights to a high quality edu-cation requires support to learn in ways thatmeet their educational needs” (National Cen-ter for Technology Innovation, 2005, p. 3). ATaffords many young children with disabilitieswith the necessary learning supports to learnand achieve, especially with regard to develop-ing emergent literacy skills that provide thefoundation for later success when enteringpublic schools (Parette, Stoner, Watts, &Wojcik, 2006).

Use of technology to develop emergent lit-eracy skills has been examined by numerousauthorities (Beck, 2002; Pierce & Porter, 1996;Smedley et al., 1997; Scooter & Boss, 2002;Wright & Shade, 1994). In typical classrooms,early childhood teachers plan and implementan array of emergent literacy activities for chil-

This project was supported by a grant from theIllinois Children’s Healthcare Foundation to thefirst three authors. Correspondence concerning thisarticle should be addressed to Howard P. Parette,Department of Special Education, Illinois State Uni-versity, Box 5910, Normal, IL 61790-5910. Email:[email protected]


Preschool Teacher Perceptions / 77

dren that emphasize oral language, experi-ences with print, storybook reading, and writ-ing for different purposes (Beck; Donovan,Milewicz, & Smolkin, 2003; Van Scooter &Boss). Sadly, AT is infrequently used in manyclassrooms to support emerging literacy activ-ities, and its potential to help develop impor-tant skills needed for later learning and suc-cess may be minimized. This is compoundedby the fact that most early childhood educa-tion preservice programs still do not prepareteachers to be able to effectively consider ATduring IEP development processes (Derer,Posgrove, & Reith, 1996; Lahm, 2003; Lesar,1998; Mistreet et al., 2005; Parette, Peterson-Karlan, & Wojcik, 2005), nor do they rapidlyintegrate AT into their curricula (Zorfass &Rivero, 2005). The net result has been that (a)relatively few children receive AT in early in-tervention programs (RESNA Technical Assis-tance Project, as cited in Long, Huang, Wood-bridge, Woolverton, & Minkel, 2003); (b) fewprofessional publications discuss AT usagewith young children (Edyburn, 2001, 2002,2003); and (c) little is known about effectiveAT emergent literacy integration practiceswith early childhood populations.

Thus, it seems appropriate to ask the ques-tions, “Why are teachers not considering andusing AT for young children with disabilities?”and “What are the concerns that teachers haveabout AT?” Ashton (2005) noted that if nega-tive perceptions or attitudes towards AT existamong practicing professionals, it is virtuallyan insurmountable task to change their mind-sets. Ashton further observed that “Forcingteachers to learn something they perceive asunnecessary will prove a fruitless endeavor”(p. 236).

Exploring Early Childhood Teacher Needs

One project designed to address this need isthe Making A Difference Using Assistive Tech-nology (MDAT) Project, funded by the IllinoisChildren’s Healthcare Foundation in 2005(Parette, Stoner, & Watts, 2005). The goal ofthis project was to develop an AT toolkit toenhance emerging literacy skills with pre-school children who were at-risk or who haddisabilities. However, the project involved thepotential for substantive organization changeswith regard to curricula approaches and pro-

fessional development strategies used inschools. Research has indicated that when anorganization is facing change there should bea systematic and deliberate process in place toensure success and it is vital to involve thoseindividuals affected by change in the initialplanning process (Kotter, 1999). Similarly, at-titudes of education professionals toward ATand its implementation in school program-ming has been reported to be a major chal-lenge nationwide (SEAT Center, NationalCenter for Technology Innovation, and Kan-sas University, 2005).

Consequently, before the MDAT Projectcould be implemented and an AT toolkit de-veloped, it was imperative to examine teacherperceptions regarding technology and literacyto assist project staff in decision-making re-garding effective professional developmentand subsequent intervention approaches. Thispreliminary work prior to project implemen-tation was guided by the following researchquestions:

1. How do teachers of preschool children de-scribe their perceptions AT?

2. How do preschool teachers describe theirconcerns about implementation of aschool-wide AT program?

3. How do preschool teachers perceive AT forassisting with literacy skill development?

4. What are the preschool teachers’ experi-ences and self-reported uses of AT?

Method

Participants

Participants were nine teachers at a self-con-tained preschool facility in a Midwestern city.Four of the teachers were teaching childrenwith identified disabilities and the remainingfive taught children who were identified asbeing ‘at-risk.’ All teachers held state teachingcertificates (see Table 1).

Research Design

The study employed qualitative strategies de-scribed by numerous researchers (e.g., Bog-dan & Biklen, 1998; Leedy & Ormrod, 2001;Millan & Wergin, 2002). Specifically, qualita-tive research allowed the researchers to deeply


explore the perspectives and gain insight intothe feelings, emotions, and thought processesof the participants (Creswell, 2002; Strauss &Corbin, 1998). The study of perspectives ofteachers who were designated to participate inthe implementation of a school-wide AT pro-gram lends itself to qualitative methodologyprecisely because it is a phenomenon aboutwhich little is known. Additionally, under-standing the perspectives of teachers as usersof the AT was deemed essential prior to theimplementation of the AT program.

Interview questions were developed to ad-dress the research questions (Kvale, 1996) andsemi-structured interviews lasting approxi-mately a half-hour were conducted with allparticipants. Semi-structured interviews al-lowed the researchers to ask for clarificationor additional information. (See Table 2 for alist of interview questions.) All interviews wereaudio-taped and transcribed verbatim to en-sure accuracy.

The Early Language and Literacy Class-

room Observation (ELLCO) Toolkit, (Smith,Dickinson, Sangeorge, & Anastasopoulos,2002) was used to assess environmental vari-ables related to language development andliteracy in each of the teachers’ classrooms(i.e., morning classes for the 3-year-old stu-dents and afternoon classes for the 4-year-oldstudents). The ELLCO has three distinctparts: the (a) Literacy Environment Checklist,(b) Classroom Observation and Teacher In-terview, and (c) Literacy Activities RatingScale.

Two researchers observed each classroomto determine the extent of the diversity ofreading and writing materials and classroomlayout. Next, interactions between teachersand students were observed during readingand writing instruction, use of technology,oral language use, and assessment strategies.Finally, student-teacher and student-studentinteractions were observed for the number oftimes and length of time for book reading andwriting during classroom activities. Inter-rater

TABLE 1

Teacher Demographics1

Teacher Education Level Yrs Experience Classroom

Barbara B.A. 10 At-riskCarole B.A. 17 At-riskDonna M.A. 12 Special needsEllen B.A. 30 Special needsJane B.A. 1 At-riskKaren B.A. 5 At-riskSara B.A. 18 Special needsToni M.A. 35 Special needsTeri B.A. 3 At-risk

1 All teachers hold state teaching certificates

TABLE 2

Interview Questions

1. Describe your classroom.2. Describe the literacy activities in your classroom.3. Can you tell me about your experiences with assistive technology?4. Describe your involvement with AT.5. Tell me about your feelings using assistive technology with your students.6. What additional information would you like concerning assistive technology?7. How do you think assistive technology can help your students?


reliability was conducted between the re-searchers on all three parts of the ELLCO.

Additionally, an Assistive Technology Self-Assessment Survey was delivered to partici-pants in person, prior to classroom observa-tions and interviews (see Table 3). Thepaper-pencil survey was developed using rec-ommended best practices from the Divisionfor Early Childhood (DEC) (Sandall,McLean, & Smith, 2000) for the applicationof technology in settings for young childrenwho attend early intervention and earlychildhood special education programs. Con-tent of the survey included the definition ofassistive technology according to IDEA and16 questions related to DEC’s recom-mended technology practices. The format ofresponse options followed a five-point com-mon rating scale focusing on agreement anddisagreement (Fink, 1995).

Data Analysis

This study employed collective case studymethodology (Stake, 2000), involving thestudy of more than one case in order to “in-vestigate a phenomenon, population, or gen-eral condition” (p. 437). This approach as-sumes that investigating a number of cases willlead to better comprehension and better the-orizing. Cross-case analysis was used to analyzeeach individual participant responses as awhole entity. A comparative analysis of all par-ticipant responses was then conducted whichallowed researchers to see processes and out-comes across many participants, thereby de-veloping a deeper understanding of theemerging phenomena through more power-ful descriptions and explanations (Miles &Huberman, 1994).

After completion of the interviews, datawere analyzed using a line by line multiplecoding approach (Barbour, 2001). The re-searchers then met frequently as a group anddeveloped categories based on their individ-ual line-by-line coding. Disagreements aboutthe categories were discussed, categories wererefined, expanded, and/or deleted as needed,and concordance was reached (Barbour). Theconstant comparative method by which re-searchers continually returned to the data foranalysis was used as an overall methodologicalframework (Charmaz, 2000). Three members

of the research team (i.e., three faculty mem-bers in a Midwestern university’s Departmentof Special Education) analyzed the data.NVivo© 2.0, a data management software pro-gram, was used to manage the data (Richards,2002).

Confirmability

Confirmability of the findings was achievedthrough three approaches: (a) triangulation(Creswell, 2002) of incidences that occurredacross cases and confirmed through observa-tions in the classroom, results of the ELLCO,and responses to an Assistive Technology Self-Assessment Survey; (b) respondent validation(Creswell, 2002), i.e., confirmation of graphicand textual findings presented to participantsregarding the researchers’ understanding ofobservations; and (c) member checks (Janesick,2000), or allowing participants and the schoolprincipal the opportunity to review andquotes used in this report. All participantsconfirmed the findings.

Findings

Four major themes emerged from the dataanalysis: (a) perceptions of technology, (b)perceived challenges to implementing tech-nology, (c) perceptions of AT and literacy,and (d) self-reported AT use. Each of thesethemes is discussed in the following sections,with participant quotes supporting the find-ings (see Table 2).

Theme 1: AT Use

The use of AT was assessed by two means:observations and self-reports. Prior to thestructured interviews conducted with teach-ers, observations were made by two of theresearchers using the ELLCO. An AssistiveTechnology Self-Assessment Survey was also com-pleted by each of the participants.

Observed AT use. Limited use of AT wasobserved in the classrooms identified as “at-risk” and, while there was more use of AT inthe classrooms which had students with dis-abilities, not all classrooms used AT at thesame level. The ELLCO instrument was usedas an observation tool because it has a sectionthat focuses on the use of AT in preschool


TABLE 3

Assistive Technology Self-Assessment

Code Number:Date:

Directions: Read the definition of AT and circle one number for each item.

Assistive technology (AT) is defined as “any item, piece of equipment, or product system, whether acquiredcommercially off the shelf, modified, or customized, that is used to increase, maintain, or improvefunctional capabilities of children with disabilities” (IDEA, 1997).

Items:1. I am confident in using assistive technology (AT) as a vehicle for more effectively serving children and families.

1 2 3 4 5Not at all Great Extent

2. I am confident in my ability to consider assistive technology applications to increase children’s ability to function and participatein diverse and less restrictive environments.1 2 3 4 5Not at all Great Extent

3. I am confident in my ability to consider chronological age-appropriateness for children when selecting types of assistivetechnology in assessment and intervention.1 2 3 4 5Not at all Great Extent

4. I am confident in my ability to consider developmentally appropriate practices for children when selecting types of assistivetechnology in assessment and intervention.1 2 3 4 5Not at all Great Extent

5. I am confident in my ability to match assistive technology tools/devices to individual children’s capabilities and limitations.1 2 3 4 5Not at all Great Extent

6. I am confident in my ability to consider the least intrusive, least intensive, yet effective low-tech tools/devices in making decisionsabout assistive technology for individual children.1 2 3 4 5Not at all Great Extent

7. I am confident in my ability to use assistive technology to facilitate the assessment process of children.1 2 3 4 5Not at all Great Extent

8. I am knowledgeable of sources for funding assistive technology.1 2 3 4 5Not at all Great Extent

9. I am confident in my ability to select and use assistive technology based on families’ preferences.1 2 3 4 5Not at all Great Extent

10. I am confident in my ability to provide assistance to individual families in the use, maintenance, and generalization of assistivetechnology to facilitate child development.


11. I am confident in my ability to provide children access to assistive technology across situations and settings where instructionand interaction can take place.


12. I am confident in my ability to be responsive to the culture, language, and economics of the family when making decisionsconcerning assistive technology applications.


13. I am confident in my ability to identify and evaluate educational software to meet the needs of children.1 2 3 4 5Not at all Great Extent

14. I am confident in my ability to integrate or embed assistive technology within children’s school activities.1 2 3 4 5Not at all Great Extent

15. I am confident in my ability to verify proper implementation of mechanical and electrical safety practices in the assembly andintegration of the technology to meet the needs of children.


16. I understand the legislative mandates and governmental regulations and their implications for technology in special education.1 2 3 4 5Not at all Great Extent


classrooms. For each the three “at-risk” class-rooms, the presence and use of technologywere scored at a ‘basic’ level within the gen-eral classroom environment. Indicators of ‘ba-sic’ technology use included (a) computersaccessible to children with use being limitedto unmonitored game-playing; (b) infrequentinstructional use of overhead projectors, au-diotapes, or digital cameras; and (c) lack of arange of technology used for a variety of pur-poses. Across the four classrooms that hadstudents with disabilities, the availability anduse of technology varied from a ‘basic’ level toan ‘exemplary’ level. Exemplary use was char-acterized by daily use of a range of technolo-gies (e.g., switches, adaptive keyboard, com-puter, picture schedules, and communicationboards) for a variety of purposes.

Self-reported AT use. There was a range ofAT use reported by teachers. Teachers in thefour classrooms with children having disabili-ties had the most extensive experiences withAT; two had children with severe disabilitiesand already had some AT devices. Two of thefive teachers—Jane and Terri—who were inat-risk classrooms, had graduated in the pre-vious two years and each had an AT course intheir teacher certification curriculum. Thethree other teachers--Karen, Barbara, and Car-ole--in the at-risk classrooms reported limitedexperience and use of AT. Teachers wereforthright in assessing their AT experiences.For example, Karen stated that her experi-ence was very minimal:

I am familiar with the computers andthings, and the touch screens. I know thatthere are certain kinds of mouse that helpthe children use that easier. But I haven’treally worked with much of that, just haveseen it.

Overall, the Assistive Technology Self-AssessmentSurvey revealed that all teachers indicated abasic level of confidence in their abilities inmaking decisions concerning low-tech/devicesfor individual children and were least confi-dent in two areas: understanding AT legisla-tion or regulations in special education andsources of funding for assistive technology.Comparison of the teachers of “at-risk stu-dents” and teachers of students with disabili-ties indicated that both groups of teachers did

not differ significantly in their AT self-assess-ment, except in one area. Teachers of stu-dents with disabilities were more confident intheir abilities to provide assistance to familieswhen using AT to facilitate child developmentthan teachers of “at-risk” students.

Overall, observations of classroom activitiesconfirmed the teachers’ self-reported limiteduse of AT. However, all teachers, regardless oftheir AT experience, stated that they were (a)excited about the school-wide project, (b) will-ing to learn, and (c) desired to use their newknowledge to benefit their students. As notedby Carole: “You know, I hear of all of thesenew things that are going on, and I know thatthe horizon is expanding. I am willing to doanything that would promote that.”

Theme 2: Perceptions of Technology

All teachers identified AT in terms of comput-ers, software, augmentative and alternativecommunication (AAC), or items such as digi-tal cameras or tape recorders. For example,Carole defined AT as,

My first vision would be a non-verbal childtouching something and saying “I wantthat” or picking up a card to say ‘eat.’ So,when you say assistive technology, it meansto me, you are helping those who are non-verbal to communicate.

The majority of the teachers defined AT interms of devices or objects that children coulduse to assist them in completing specific tasks.

Two teachers with previous AT experiencesidentified a wider variety of tools. Ellen de-scribed her experience as,

We do use communication books at snack/breakfast, beginning PECS. We use ‘I needa break’ cards, sabotaging something sothey have to ask for it. I use Big Mac switchesfor participation. In the morning we say ‘yo’for attendance, and if they can’t make thatverbalization, we use a recording. We use GoTalk, minimally at the moment for recogniz-ing classmates. I use Tech Speak, and againwith kids that are more non-verbal, morephysical disabilities, to be able to answerquestions about colors, shapes, classmates,letters, etc.


Integration or supplemental perspective. Dur-ing the interviews teachers specifically de-scribed how they used or could use AT andhow AT affected or could affect their students’learning. From their responses emerged twobroad categories of perspectives. There weretwo teachers who described AT as somethingintegrated in the curriculum. These two teach-ers--Sara and Ellen--worked in classrooms withchildren with disabilities and perceived AT asbeing an integral and vital component of theircurriculum. A statement by Ellen revealed hercomfort level and recognition of the value ofAT:

Right now I am real comfortable with lowtech, use of pictures, switches. I feel likewith the students I teach, I want them toparticipate, be challenged, and do some-thing, and that is my best option. I couldn’tlive without AT. I couldn’t be comfortableas a teacher and not use some of thosethings.

Similarly, Sara described the benefit of AT forher students: “It fosters independence, pridebecause they can participate, and it makesthem active learners.” These teachers, becauseof their knowledge, their experience, andtheir student’s needs, appeared to integrateAT actively into their curriculum.

The remaining seven teachers perceived ATas a supplement rather than an integratedaspect of their curriculum. This was illustratedby Terri’s response to a request to describehow she uses AT in the classroom, “This willbe short! Basic use of a computer and digitalcamera. I haven’t used the computer as muchbecause we had so many pictures already, butI am very limited in assistive technology.” Theresearch team observed that these teachershad worked hard to meet the needs of theirstudents and foster learning. AT had beendemonstrated to them at various points intheir careers, although they had no inserviceprofessional development experiences in AT,and had no support for the limited amount oftechnology in their classrooms. All teachersrecognized their limitations, stated their de-sire to learn, and spoke of the challengesawaiting them with regard to using AT in theirclassrooms.

Theme 3: Perceived Challenges to AT Use

All teachers spoke of challenges regarding theimplementation of school-wide AT. Thesechallenges were categorized into three pri-mary sub-themes: (a) support, (b) time, and(c) student population characteristics. Each ofthese sub-themes is discussed in the followingsections.

Support. Teachers noted their intenseneed of AT support, especially technical sup-port and professional development. Technicalsupport was defined as supports related to useof computers, the ability to problem solvetechnical difficulties, and the need for guid-ance in choosing appropriate AT. There wasconsensus on the desire for a new operatingplatform for their computers. Teachers spokefrequently of the difficulty they had with thecurrent operating system in the Macintosh(Macs) computers that were in their class-rooms. The primary difficulty was not in theperformance of the Macs, but in their ownlevel of knowledge as to how to use them. Allteachers had a Windows operating system ontheir home computers, used that operatingsystem primarily, and stated that they had lit-tle time to learn to use the Macs proficiently.Carole flatly stated, “We are not too happywith the Macs. I just don’t think they are aseasy. I just don’t care for them.” And whenprobed with a question about her use of Win-dows she added, “Yes, and that could be thedifference, that I am used to working withWindows.”

Additional technical support needs werethe need for a network color printer thatwould consistently print, in color, and couldbe accessed directly from the classroom.When equipment did not function properly,teachers were frustrated since they had spentlimited free time to produce a product. Ellenspoke of a particular instance that illustratedher frustration:

All that time you just spent, and then youhave to redo it [Boardmaker pictures] onanother computer. And I told you aboutanother time, getting my breakfast cardsdone; I paid for it, went to Kinkos, lami-nated it, put Velcro on it, all that kind ofstuff. Then I bring it back here, and thepictures came off. All the time you spent wasjust gone.


Needed support for professional developmentin specific areas was identified by all teachers.One particular need was professional develop-ment focusing on the technical skill of oper-ating AT. Terri spoke of her willingness toreceive training in this area: “I would be verypositive towards it if somebody trained us, I’mvery open to it, I just need to learn about it.”Additionally, teachers identified a need to beinformed or guided towards purchasing ATthat would benefit their students. For exam-ple, Toni commented, “I guess I would like toknow what there is to help the kids. High orlow tech, I don’t even know what to ask for.”

Characteristics of student population. Allteachers discussed challenges with the studentpopulations they served. Sara was only atschool in the morning when the initial inter-views were conducted; however, she is cur-rently at the school full time. The structure ofthe school day, having both morning and af-ternoon groups, coupled with varying abilitylevels among these groups, provided substan-tive challenges. Additional challenging factorsthat the teachers identified were (a) poverty;(b) high mobility among the children; (c)English as a second language; and (d) differ-ences in maturity between the younger, morn-ing group (M age � 3.5 yrs) and the older,afternoon group (M age � 4.5 yrs). The highmobility rate was noted to be both particularlychallenging and disturbing. Carole said,

Yeah, especially for the children that need itthe most, they’re the ones that are here forthree weeks and then take off. We open thedoor for the homeless, or if we find out theyare in the Mission. They come for two daysand then we’ll never see them again be-cause they’ve moved, and that’s tough. Wehad a little girl that was here for 6 days, andwas here for our class picture, and then shewas gone. And then you know you look atthat, and you say ‘Where’s she at?’ I hopeshe is safe and warm.”

Teachers cared deeply about their students.They recognized that some of their studentshad no opportunity to experience technologyother than what was provided in the class-room, and even then such opportunities wereof short duration.

Instructional goals for the morning and af-

ternoon groups differed also, with teachersidentifying the afternoon groups as more ac-ademically oriented while the morning groupsrequired a significant amount of social in-struction. Donna describes her morning classas needing instruction in social skills: “Wehave just gotten a couple of really interestingstudents. In the morning class, we’ve got sucha variety, the range is incredible. There aresome self-help things that we really need topush.” Teachers wanted AT that would benefitboth their morning classes, with a focus onsocial skills, and their afternoon classes, whichfocused more on academic skills.

Time. A prevalent theme that emergedwas the issue of time constraints. All teachersdiscussed their concern that AT would costthem time--a limited commodity in preschoolsettings. Time concerns included time (a) re-quired to learn AT, (b) required to incorpo-rate the AT into lessons, and (c) needed toteach the children how to use AT. Observa-tions of the classrooms confirmed that thetypical day was filled with activities, with chil-dren being engaged in a range of centers,activities, and therapy sessions (for childrenwith disabilities). Carole described a typicalclass:

From the moment they walk in here in themorning or in the afternoon, for the twohours and 40 minutes they are in here,there is not one time when we sit down.There’s just constant motion, constant tran-sitions, not only for us but for the kids.

An additional concern related to time, was thetime required to assist children with AT, spe-cifically with computer use. Karen describedthis issue as, “If things [computer softwareprograms] get too difficult then me or myassistant have to be over there to show them.You know, I don’t mind helping them, but Ihave 20 other children.” Observations con-firmed that 20 3- and 4-year-olds required con-sistent attention by both teacher and parapro-fessional in each of the classrooms.

Ellen spoke of the time required to makethe products using AT. Ellen was one of theteachers that used AT consistently in her class-room and knew first-hand of the time re-quired to produce products for use with 10children with disabilities. She explains herconcern,


I don’t know if there are any answers to that[time required to produce products]. If youwant to use that stuff, you do have to makeit, and I know there is time involved. [Iwould like] Any ease in that process, andmaking it practical or efficient in the class-room.

Teachers’ perceptions of challenges to imple-menting AT technology in their classroomwere grounded in the knowledge of theirclassroom students, their classroom schedules,and their need for additional time.

Theme 4: Perceptions of AT and Literacy

Teachers described the literacy in the curric-ulum and consistently made references toreading. Reading was the primary focus of allliteracy activities; activities that incorporatedwriting were not mentioned. Most of the writ-ing in the classrooms was focused on students’learning to write their names. This perceptionof literacy being comprised almost exclusivelyby reading activities was confirmed by obser-vations in the classroom and the observationalstructure provided by the ELLCO. Terri’s de-scription of her classroom literacy activities istypical of other teachers’ responses:

Oh yeah, we use the large picture books,and books on tape, we do a lot of that.Yesterday, we were doing some role playwith it, we have a lot of fun with it, andexpand a lot of the stories. We try to usepictures and different kinds of things withthat also.

Sara, one of the teachers with the most ATexperience, described her inclusion of ATduring literacy activities:

When we are in calendar we have thoseBoardmaker® pictures and we make thesentences, “Today is _____.” Differentthings with the Intellikeys�. We are begin-ning to make the letters of our names andrecognizing our names, without picturesnow, which I’m very excited. So much of it[AT] wraps into every little activity it’s hardto separate it out.

However, most of the teachers did notuse AT during reading, or during writing.The exception was if one of the children

was receiving occupational therapy services;then the occupational therapist may haverecommended a low tech AT device to assistwith the technical aspects of writing.

The overall finding was that teachers per-ceived literacy, at this young age, as primarilyreading activities and writing was minimallyaddressed. The incorporation of AT duringreading or writing activities was also mini-mal. More incorporation occurred in thetwo classes where teachers (Ellen and Sara)had more experience and knowledge of AT.

Discussion

Implementing AT into a literacy curriculumin any preschool program requires substantialorganizational change. Consistent with the re-search on successful organizational changes,one of the primary stakeholders in this pre-school setting--the teachers--were interviewedto understand their perspectives about AT, ATand literacy, and to understand their per-ceived challenges regarding implementationof the project. Such understanding was criticalbefore developing any intervention approach-es—particularly professional developmentsupport strategies. The following discussion isorganized around each of the four majorthemes.

Assistive Technology Use

Findings regarding lack of familiarity with ATand infrequent usage is consistent with previ-ous studies (Derer et al., 1996; Lesar, 1998;Parette, 1997; Scott, 1997). Involving teachersin AT planning processes, whether for a singledevice or a school-wide AT program, has beenstrongly recommended in the literature(Carey & Sale, 1994; Copley & Ziviani, 2004;Riemer-Reiss & Wacker, 2000; Todis &Walker, 1993). Prior to any professional devel-opment approach designed to increaseteacher familiarity with AT, it was critical toask teachers about their concerns, knowledge,and perceptions of AT. Interviews were thefirst step in enhancing teacher involvementwith the MDAT project.

The teacher participants in this studyreadily admitted their lack of training and lackof knowledge regarding types of AT, whilealso noting their willingness to learn. Kotter


(1999) observed that the first step to effectivechange in organizations is creation of a senseof ‘urgency’ for the change. All teachers werewell aware of the impact of the NCLB studentachievement emphasis on their profession.They were also aware that emerging literacyskills provided the foundation for children’ssuccess on entering school. This knowledge,supported by administrative support for theMDAT grant (i.e., expressed and fiscal sup-port by the school administration), appearedto drive their own commitments to participatein the project.

Perceptions of Technology

Consistent with previous research (e.g., Hut-inger, Johanson, & Stoneburner, 1996) partic-ipants in this study, with the exception of twoof the special educators, viewed AT as supple-mental to the curriculum, rather than beingintegrated. Approaching AT use from this par-adigm is limiting to both teachers and stu-dents. The participants most commonlyviewed AT narrowly, specifically in terms ofcomputer and educational software. Only twospecial educators saw AT from an integrationperspective. It was determined that all teach-ers must learn, experience, and receive sup-port when implementing AT as an integral partof the curriculum. Changing this perspectiveof current teacher paradigm is viewed as amajor challenge confronting this project, andmost preschool programs currently. Strategiesto deal with such challenges include directcontact in classrooms by intervention staff reg-ularly, and engaging students and teachers inhands-on activities that model integrative AT.

Perceived Challenges to AT in the Classroom

The anticipated challenges identified byteachers were support, time, and studentneeds. Interestingly, even with their limitedAT experience and use, the teachers clearlyarticulated challenges that are consistentlyidentified in the literature. Support chal-lenges, specifically technical support and pro-fessional development, have been docu-mented as barriers to the use of AT (e.g.,Beukelman & Mirenda, 1998; Carey & Sale,1994; Copley & Ziviani, 2004; Riemer-Reiss &Wacker, 2000).

The challenge of time focused on findingtime to learn and implement the AT into theclassroom. In earlier studies involving in-structional technology, researchers con-cluded that something on the order of 3 to 5years is required for teachers to become re-ally adept at incorporating technology intotheir teaching practice (Sheingold, 1991,1992). Currently, time to learn and effec-tively use technology in classrooms remainsa concern for many teachers (Valmont,2003). More recent publications have em-phasized essential conditions for effectiveuse of technology in classrooms, including(a) a shared vision for integration; (b) stan-dards and curricula support; (c) requiredpolicies (e.g., use of the Internet, legal use,equity); (d) access to hardware, software,and other resources; (e) trained personnel;(f) technical assistance; and (g) appropriateteaching and assessment approaches(Roblyer, 2006; Whitehead, Jensen, & Bos-chee, 2003). However, less is known aboutthe time commitment required for teachersto become comfortable with integrating ATinto their curricula (see, e.g., Randle & Har-ris, 2004). Ideally teachers should have timeto learn the AT, experiment with it in theclassroom, and discuss any difficulties withsupport staff (MacArthur, 2001).

The reality of teaching in an early child-hood setting with children identified beingat-risk or having disabilities is that time is a‘precious commodity’ and adequate time--both for professional development in learningto use devices, planning for integration, andimplementation of devices in the curriculum--are critical elements for successful technologyintegration (Speck & Knipe, 2005). Chal-lenges of time to the project staff continue tobe of prime concern.

When the MDAT project was first imple-mented students did not attend school onFriday, and that day was designated for pro-fessional development several times a month.Within the first year of project implementa-tion, students began attending school five daysa week. The original professional develop-ment schedule was significantly affected andadaptations, consisting of hands-on activitygroups in the classroom conducted by projectstaff and graduate students, early morning ab-breviated training sessions, half day Institute


sessions, and the establishment of usergroups, was immediately implemented. How-ever, time for professional development con-tinues to be a challenge for all involved.

Characteristics of students in the preschoolclassrooms also presented unique challenges.It was found that access to AT in the homesvaried, and the mobility rate of the studentswas high, i.e., children across classrooms weretransitioning in and out of the program.Teachers also repeatedly reported that deal-ing with behavioral concerns often took pre-cedence in all activities occurring in the class-rooms. One particular AT strategy that canassist teachers with behavioral concerns in theclassroom would be to incorporate visual strat-egies to promote appropriate behavior.Teachers in this project have been instructedin the use of Boardmaker™ and Writing withSymbols™ to create visual schedules for class-room routines and sequences in specific class-room activities. This and other strategies willbe systematically infused into the AT profes-sional development approaches to ensure thatteachers develop the necessary skill sets to useAT effectively in the curriculum.

AT and Literacy

The philosophy of emergent literacy views in-dividuals who are either “chronologically orcognitively young learners” as capable oflearning literacy (Pierce & Porter, 1996, p.142). Skills that emerge during early literacylearning include an understanding of thefunction and concept of print. Understandingthe function of print can be facilitated byactivities such as reading a story, making lists,or scribbling and drawing (Justice & Pullen,2003; Sulzby & Teale, 1991). Similarly theseactivities can also facilitate concepts of printsuch as turning the page from right to left,reading from left to right and recognizingprinted vocabulary (Pierce & Porter). Addi-tionally, early literacy intervention is criticalbecause young children who exhibit difficul-ties with emergent literacy skills rarely attainthe literacy level of peers who have adequateemergent literacy skills (Juel, 1988). The stu-dents in the targeted preschool were identi-fied as at-risk or had identified disabilities,which underscored the importance of facili-tating emergent literacy skills.

Summary and Recommendations

The MDAT project has been implementedwith the goal of using AT to foster emergingliteracy skills. However, the first step of thisproject was to investigate the teachers’ use andperspectives of AT, address their needs of pro-fessional development and minimize theirchallenges to making AT integral to their cur-riculum. Supported by the knowledge ofteachers’ needs and perceived challenges theproject staff has provided teachers with an ATtoolkit, implemented professional develop-ment for AT within the toolkit, and modeleduse of AT during group activities. The follow-ing recommendations are based on initial ex-periences with this project that should haveapplicability to programs nationally.

Strategy 1. Gain perspectives of stakeholders.All too often, top-down approaches are usedin public school settings (Kolderie, 1990) thatare sometimes characterized by such artifactsas lack of teacher input regarding selectionand development of curricula, budgeting, anddecisions regarding professional developmentneeds. However, for professional develop-ment to be successful, teachers need opportu-nities to discuss their beliefs about technologyand its relationship to pedagogy (MacArthur,2001). This project assumed that teacher in-put was critical to developing effective ap-proaches for the integration of AT into thepreschool curricula. For example, had the re-search team not explored teacher perceptionsprior to developing intervention strategiesand an AT toolkit to facilitate writing literacyskills, a decision may have been made to usecurrently available computers (Macs). How-ever, interviews with teachers revealed thatpreferences for operating platforms should beincorporated into the AT toolkit to maximizetheir use. The change of the platform hasincreased teacher involvement with theproject, increased their use of AT, and facili-tated their enthusiasm for the changes thisproject has required. Similarly, at every phaseof project decision-making regarding curric-ula approaches, ideas were presented to teach-ers for feedback, and teacher suggestions weresystematically considered.

Strategy 2. Ensure flexibility in professional de-velopment approaches. With the loss of a des-ignated professional development day, project


staff had to make adjustments in key strategiesto meet the identified needs of training.Training sessions were scheduled beforeschool, which were minimally successful sincethey were of short duration and offered noopportunity to have hands-on experience withthe technology. Moving into the classroomand modeling activities for the teachers hasbeen received well. The activities embed ATand writing assessments and teachers are pro-vided with a time to observe the implementa-tion of AT.

Strategy 3. Use stipends to support teachers.Even though the U.S. Department of Educa-tion has recommended that 30% of a schooldistrict’s budget be allocated to support pro-fessional development, only about 6% is allot-ted for such activities (International ReadingAssociation, 2001). Zabala and Carl (2005)argue that education agencies have obliga-tions to (a) include AT in the technologyplanning and budgeting processes, and (b)provide continuous AT learning opportunitiesfor teachers. Though it may be difficult tosecure internal funding in school systems tosupport teachers for professional develop-ment, numerous approaches have been pro-moted in recent years to expand the profes-sional growth of teachers, including (a)school/university partnerships (Bauer &Anderson, 2001; Maring, Boxie, & Wiseman,2000; Wojcik, Peterson-Karlan, Watts, & Pa-rette, 2004); (b) use of online professionaldevelopment resources [cf. Infinitec AssistiveTechnology Coalition, 2006; North CentralProfessional Development Laboratory, n.d.;Special Education Assistive Technology[SEAT], 2004; (c) cybermentoring (Boxie &Maring, 2001); (d) and increased access toprofessional organization training (cf. Centerfor Applied Special Technology, n.d.; DonJohnston, n.d.; Intellitools™, n.d.), and list-servs (Quality Indicators for Assistive Technol-ogy [QIAT], n.d.). Securing small mini-grants(Parette, Murdick, & Gartin, 1996) from localcommunity groups, and grants from state andnational private foundations may also be help-ful in securing resources to support profes-sional development activities of teachers.While some on-line professional developmentresources are free (e.g., QIAT), others requiremembership fees or payment for participation

(e.g., Infinitec Assistive Technology Coali-tion).

Regardless of costs for professional develop-ment, teacher participation in any AT curric-ula integration effort can be enhanced by pro-viding teacher stipends (Hirsch, 2006;Reichardt, 2001). As noted by the NationalStaff Development Council (Hirsch), schoolsmust increase their capacity for teachers tolearn and use technology by using stipends forteacher leaders who serve as mentors for newteachers, team leaders for learning teams, andtrainers. One strategy that holds particularpromise is the use of AT User Groups. Gener-ally, a user group is defined as “a set of peoplewho have similar interests, goals, or concerns.The members have regular meetings wherethey can share their ideas” (Whatis.com,2006). Creating such groups requires a com-mitment on the part of (a) one or more indi-viduals having expertise with specific AT ap-plications to serve as a facilitator, and (b) asmall group of 6-8 teachers who express inter-est in developing new AT integration skills.User groups would meet at regularly sched-uled times, such as after school, for 1-2 hours,and allow teachers the opportunity to shareideas with one another regarding AT solutionsused in their classrooms. It would also allowfor more intensive instruction and creativeproblem-solving in collaboration with one ormore facilitators having more advanced skills.Integral to the creation and maintenance ofthese groups is a stipend to support participa-tion. In this project, the research team willemploy a user group approach currently usedin a collaborative project—the Heart of Illi-nois Low Incidence Association (HILIA)—which involves a cadre of teachers having ATinterests, and who represent Illinois State Uni-versity, and five school districts in Central Illi-nois (SEAT Center, 2006). In this project, adecision was made to offer interested teachersa small stipend to participate in minimum of 6out of 8 scheduled user group sessions thatwould be held in scheduled after-school ses-sions. Since teachers have other demandsplaced on their time (e.g., families) and giventhe lack of time available during typical schooldays for professional development, monetaryincentives are both appropriate and necessary.They ensure teacher buy-in, and also supportthe development of critical AT competencies


that would be difficult to develop outside of aformal, ongoing professional developmentprogram.

References

Ashton, T. M. (2005). Students with learning dis-abilities using assistive technology in the inclusiveclassroom. In D. L. Edyburn, K. Higgins, & R.Boone (Eds.), Handbook of special education technol-ogy research and practice (pp. 229–238). WhitefishBay, WI: Knowledge by Design.

Barbour, R. S. (2001). Checklists for improving ri-gour in qualitative research: A case of the tailwagging the dog? British Medical Journal, 322,1115–1118.

Bauer, J. F., & Anderson, R. S. (2001). A construc-tive stretch: Preservice teachers meet preteensin a technology-based literacy project. ReadingOnline, 5(5). Retrieved February 20, 2006, fromhttp://www.readingonline.org/articles/art_index.asp?HREF�/articles/bauer/index.html

Beck, J. (2002). Emerging literacy through assistivetechnology. Teaching Exceptional Children, 35(2),44–48.

Beukelman, D., & Mirenda, P. (1998). Augmentativeand alternative communication: Management of severecommunication disorders in children and adults (2nd

ed.). Baltimore: Brookes.Bogdan, R., & Biklen, S. (1998). Qualitative research

in education: An introduction to theory and methods.Boston: Allyn and Bacon.

Boxie, P., & Maring, G. H. (2001). Cybermentoring:The relationship between preservice teachers’ useof online literacy strategies and student achieve-ment. Reading Online, 4(10). Retrieved February20, 2006, from http://www.readingonline.org/articles/art_index.asp?HREF�/articles/boxie/index.html

Carey, D. M., & Sale, P. (1994). Practical consider-ations in the use of technology to facilitate theinclusion of students with severe disabilities. Tech-nology and Disability 3, 77–86.

Center for Applied Special Technology. (n.d.). Pro-fessional development. Retrieved February 20, 2006,from http://www.cast.org/pd/index.html

Charmaz, K. (2000). Grounded theory: Objectivistand constructivist methods. In N. K. Denzin &Y. S. Lincoln (Eds.), Handbook of qualitative research(2nd ed., pp. 509–536). Thousand Oaks, CA:Sage.

Copley, J., & Ziviani, J. (2004). Barriers to the use ofassistive technology for children with multipledisabilities. Occupational Therapy International, 11,229–243.

Creswell, J. (2002). Educational research: Planning,conducting, and evaluating quantitative and qualita-

tive research. Upper Saddle River, NJ: Merrill Pren-tice Hall.

Derer, K., Posgrove, L., & Reith, H. (1996). Surveyof assistive technology applications in schools andrecommendations for practice. Journal of SpecialEducation Technology, 8, 62–80.

Don Johnston. (n.d.). Professional services. Re-trieved February 20, 2006, from http://www.donjohnston.com/proservices/prodevfrm.htm

Donovan, C. A., Milewicz, E., & Smolkin, L. B.(2003). More than a single text: Nurturing chil-dren’s interest in reading and writing for multiplepurposes. Young Children, 58(2), 30–36.

Drasgow, E., Yell, M. L., & Robinson, T. W. (2001).Developing legally correct and educationally ap-propriate IEPs. Remedial and Special Education, 22,359–373.

Edyburn, D. L. (2001). 2000 in review: A synthesis ofthe special education technology literature. Jour-nal of Special Education Technology, 16(2), 5–25.



Fink, A. (Ed.). (1995). How to ask survey questions.Thousand Oaks, CA: Sage.

Hirsch, S. (2006). Dollars and sense. Retrieved Feb-ruary 20, 2006, from the National Staff Develop-ment Council website at http://www.nsdc.org/library/publications/jsd/hirsh243.cfm

Hutinger, P., Johanson, J., & Stoneburner, R.(1996). Assistive technology applications in edu-cational programs of children with multiple dis-abilities: A case study report on the state of thepractice. Journal of Special Education Technology,13(1), 16–35.

Individuals with Disabilities Education Act Amend-ments, 20 U.S.C. § 1400 et seq (1997).

Infinitec Assistive Technology Coalition. (2006).Infinite potential through technology. Retrieved Febru-

ary 20, 2006, from http://coalitionconnection.org/includes/index.cfm

Intellitools™. (n.d.). Intellitools™ professional develop-ment. Retrieved February 20, 2006, from http://www.intellitools.com/

International Reading Association. (2001). Integrat-ing literacy and technology in the curriculum: A posi-tion statement of the International Reading Association.Newark, DE: Author.

Janesick, V. (2000). The choreography of qualitativeresearch design. In N. K. Denzin & Y. S. Lincoln(Eds.), Handbook of qualitative research (pp. 379–399). Thousand Oaks, CA: Sage.

Juel, C. (1988). Learning to read and write: A lon-gitudinal study of 54 children from first through


fourth grades. Journal of Educational Psychology, 80,437–447.

Justice L. M., & Pullen, P. C. (2003). Promisinginterventions for promoting emergent literacyskills: three evidence based approaches. Topics inEarly Childhood Special Education, 23(3), 99–113.

Kolderie, T. (1990). Beyond choice to public schools. With-drawing the exclusive franchise in public education. Re-trieved February 20, 2006, from http://www.ppionline.org/ndol/print.cfm?contentid�1692

Kotter, J. (1999). Change leadership. Executive Excel-lence, 16(4), 16–18.

Kvale, S. (1996). Interviews: An introduction to quali-tative research interviewing. Thousand Oaks, CA:Sage.

Lahm, E. A. (2003). Assistive technology specialists.Remedial and Special Education, 24, 141–153.

Leedy, P. D., & Ormrod, J. E. (2001). Practical re-search: Planning and design (7th ed.). Upper Sad-dle River, NJ: Merrill Prentice Hall.

Lesar, S. (1998). Use of assistive technology withyoung children with disabilities: Current statusand training needs. Journal of Early Intervention,21, 146–159.

Long, T., Huang, L., Woodbridge, M., Woolverton,M., & Minkel, J. (2003). Integrating assistive tech-nology into an outcome-driven model of servicedelivery. Infants and Young Children, 16, 272–283.

MacArthur, C. A. (2001). Technology implementa-tion in special education. In J. Woodward & L.Cuban (Eds.), Technology, curriculum and profes-sional development. Adapting schools to meet the needsof students with disabilities (pp. 115–120). Thou-sand Oaks, CA: Corwin.

Maring, G. H., Boxie, P., & Wiseman, B. J. (2000).School-university partnerships through on-line pat-tern books. Reading Online, 4(5). Retrieved February20, 2006, from http://www.readingonline.org/articles/art_index.asp?HREF�/articles/maring/index.html

McMillan, J. H., & Wergin, J. F. (2002). Understand-ing and evaluating educational research (2nd ed.).Columbus, OH: Merrill.

Miles, M., & Huberman, A. (1994). Qualitative dataanalysis. Thousand Oaks, CA: Sage.

Mistreet, S. G., Lane, S. J., & Ruffino, A. G. (2005).Growing and learning through technology: Birthto five. In D. Edyburn, K. Higgins, & R. Boone(Eds.), Handbook of special education technology re-search and practice (pp. 273–307). Whitefish Bay,WI: Knowledge by Design.

National Association for the Education of YoungChildren. (1996). Technology and young children—Ages 3 through 8 (Position Statement). Washing-ton, DC: Author.

National Center for Technology Innovation.(2005). Moving towards solutions. Assistive and learn-

ing technology for all students. Washington, DC: Au-thor.

National Early Childhood Technical Assistance Cen-ter. (2003). Assistive technology overview. RetrievedOctober 24, 2005, from http://www.nectac.org/topics/atech/overview.asp?text�1

No Child Left Behind Act, 20 U.S.C. 6301 et seq.(2001).

North Central Professional Development Labora-tory. (n.d.). Professional development. RetrievedFebruary 20, 2006, from http://www.ncrel.org/info/pd/

Parette, H. (1997). Assistive technology devices andservices. Education and Training in Mental Retarda-tion and Developmental Disabilities, 32, 267–280.

Parette, H. P., Murdick, N. L., & Gartin, B. (1996).Mini-grant to the rescue! Using community re-sources to obtain assistive technology devices forchildren with disabilities. Teaching Exceptional Chil-dren, 28(2), 20–23.

Parette, H. P., Peterson-Karlan, G. R., & Wojcik,B. W. (2005). The state of assistive technologyservices nationally and implications for future de-velopment. Assistive Technology Outcomes and Bene-fits, 2(1), 13–24.

Parette, H. P., Stoner, J. B., & Watts, E. H. (2005).Making a difference with assistive technology. Grantfunded by the Illinois Children’s HealthcareFoundation.

Parette, P., Stoner, J., Watts, E., & Wojcik, B. W.(2006, January). Using AT toolkits to develop earlywriting skills with preschoolers. Paper presented tothe Technology, Reading, and Learning Difficul-ties World Conference, San Francisco, CA.

Pierce, P., & Porter, P. B. (1996). Helping personswith disabilities to become literate using assistivetechnology: Practice and policy suggestions. Focuson Autism and Other Developmental Disabilities, 11,142–148.

Quality Indicators for Assistive Technology. (n.d.).Welcome. Retrieved February 20, 2006, fromhttp://sweb.uky.edu/�jszaba0/QIAT.html

Randle, K., & Harris, S. (2004, March). Providingeffective technology training through web based solu-tions. Paper presented at the Center on Disabili-ties Technology and Persons with DisabilitiesConference 2004, Northridge, CA.

Reed, P., & Bowser, G. (2005). Assistive technologyand the IEP. In D. Edyburn, K. Higgins, & R.Boone (Eds.), Handbook of special education technol-ogy research and practice (pp. 61–77). Whitefish Bay,WI: Knowledge by Design.

Reichardt, R. (2001). Toward a comprehensive ap-proach to teacher quality. Policy brief. Aurora, CO:Mid-Continent Research for Education andLearning.

Richards, L. (2002). NVivo. [Computer software].Bundoora Victoria, Australia: QSR International.


Riemer-Reiss, M. L., & Wacker, R. R. (2000). Factorsassociated with assistive technology discontinu-ance among individuals with disabilities. Journal ofRehabilitation, 66(3), 44–50.

Roblyer, M. D. (2006). Integrating educational technol-ogy into teaching (4th ed.). Upper Saddle River, NJ:Merrill Prentice Hall.

Sandall, S., McLean, M., & Smith, B. J. (2000). DECrecommended practices in early intervention/early child-hood special education. Longmont, CO: Sopris West.

Scott, S. B. (1997). Comparison of service deliverymodels influencing teachers’ use of assistive tech-nology for students with severe disabilities. Occu-pational Therapy in Health Care, 11(1), 61–74.

SEAT Center. (2006). SEAT partners. RetrievedApril 25, 2006, from http://www.seat.ilstu.org/aboutus/partners.shtml

SEAT Center, National Center for TechnologyInnovation, and University of Kansas. (2005).Assistive technology & educational Progress-. . .Charting a new direction. Executive summary.Retrieved February 17, 2006, from http://www.nationaltechcenter.org/default.asp

Sheingold, K. (1991). Restructuring for learningwith technology: The potential for synergy. PhiDelta Kappan, 73(1), 17–27.

Sheingold, K. (1992). Technology integration andteachers’ professional development. In Learningtechnologies essential for education change (pp. 41–51). Washington, DC: Council of Chief StateSchool Officers.

Smedley, C. E., Heiple, V. S., Baker, S., Dunn, N.,Parette, H. P., & Hendricks, M. (1997). Keyboardkids: Using computers to teach young children. LittleRock, AR: Arkansas Easter Seal Society.

Smith, M. W., Dickinson, D. K., Sangeorge, A., &Anastasopoulos, L. (2002). Early language and lit-eracy classroom observation (ELLCO) toolkit, researchedition. Baltimore: Brookes.

Special Education Assistive Technology Center.(2004). AT2. Preparing all teachers for assistive tech-nology. Retrieved February 20, 2006, from http://www.coe.ilstu.edu/seat/at2/

Speck, M., & Knipe, C. (2005). Why can’t we get itright? Designing high quality professional developmentfor standards-based schools. Thousand Oaks, CA:Corwin.

Stake, R. (2000). Case studies. In N. K. Denzin &

Y. S. Lincoln (Eds.), Handbook of qualitative research(pp. 435–454). Thousand Oaks, CA: Sage.

Strauss, A., & Corbin, J. (1998). Basics of qualitativeresearch: Techniques and procedures for developinggrounded theory. Thousand Oaks, CA: Sage.

Sulzby, E., & Teale, W. (1991). Emergent literacy. InR. Bart, M. Kamil, P. Mosenthal, & D. Pearson(Eds.), Handbook of reading research: Vol. 2 (pp.727–758). New York: Longman.

Todis, B., & Walker, H. M. (1993). User perspectiveson assistive technology in educational settings.Focus on Exceptional Children, 26(3), 1–16.

Valmont, W. J. (2003). Technology for literacy andlearning. Boston: Houghton Mifflin.

Van Scooter, J., & Boss, S. (2002). Learners, language,and technology: Making connections that support liter-acy. Portland, OR: Northwest Regional Educa-tional Laboratory.

Whatis.com. (2006). User group. Retrieved February20, 2006, from http://whatis.techtarget.com/definition/0,,sid9_gci881409,00.html

Whitehead, B. M., Jensen, D. F. N., & Boschee, F.(2003). Planning for technology. A guide for schooladministrators, technology coordinators, and curricu-lum leaders. Thousand Oaks, CA: Corwin.

Wiekle, B., & Hadadian, A. (2003). Can assistivetechnology help us not to leave any child behind?Preventing School Failure, 47, 181–186.

Wojcik, B. W., Peterson-Karlan, G., Watts, E. H., &Parette, P. (2004). Assistive technology outcomesin a teacher education curriculum. Assistive Tech-nology Outcomes and Benefits, 1, 21–32.

Wright, J. L., & Shade, D. D. (Eds.). (1994). Youngchildren: Active learners in a technological age. Wash-ington, DC: NAEYC.

Zabala, J. S., & Carl, D. F. (2005). Quality indicatorsfor assistive technology in schools. In D. Edyburn,K. Higgins, & R. Boone (Eds.), Handbook of specialeducation technology research and practice (pp. 179–207). Whitefish Bay, WI: Knowledge by Design,Inc.

Zorfass, J., & Rivero, H. K. (2005). Collaboration iskey: How a community of practice promotes tech-nology integration. Journal of Special EducationTechnology, 20(3), 51–67.

Received: 20 May 2006Initial Acceptance: 20 July 2006Final Acceptance: 1 October 2006


Effects of the STAR Intervention Program on Interactionsbetween Campers with and without Disabilities during

Inclusive Summer Day Camp Activities

Christina M. Boyd, Jeffrey L. Fraiman, Kelly A. Hawkins, Jennifer M. Labin,Mary Beth Sutter, and Meghan R. Wahl

University of Maryland

Abstract: The purpose of this study was to examine the effects of a peer intervention program designed to increaseinteractions between children with and without disabilities in an inclusive summer camp. A multiple probesingle subject design was used to determine the effects of the STAR intervention on six dyads of campers agedfive through ten over two week sessions. Each dyad consisted of one camper with a mild to moderate disabilityand one camper without a disability. The results showed an overall increase in the number of interactions anddemonstrated that the STAR program was effective in increasing interactions between campers with and withoutdisabilities. Factors contributing to the success of the intervention are discussed as well as limitations.

Many studies have investigated the dynamicsof social interaction between young childrenwith and without disabilities. In particular, theuse of peer training has been examined as anintervention to increase social interactions inpreschool and elementary school settings(Garfinkle & Schwartz, 2002; Goldstein, Kacz-marek, Pennington, & Shafer, 1992; Hundert& Houghton, 1992; Odom, Chandler, Os-trosky, McConnell, & Reaney, 1992). English,Goldstein, Kaczmarek, and Shafer (1996) de-veloped a peer skills training program thattaught children to “stay,” “play,” and “talk”with a peer with a disability. English, Gold-stein, Shafer, and Kaczmarek (1997), Gold-stein and English (1997), and Laushey andHeflin (2000) investigated the effectiveness ofthe “stay-play-talk” training procedure in pre-school and kindergarten children. In all threestudies, children without disabilities receivedtraining sessions during which they were

taught how to interact with a child with adisability. The results showed a substantial in-crease in the interactions within the dyad afterthe “stay-play-talk” procedure was taught.

Gonzalez-Lopez and Kamps (1997) pro-vided information about disabilities to typi-cally developing peers. In addition, theytaught social skills to children with autism andtheir peers, which increased the incidence ofpositive interactions between the children.Sasso, Mundschenk, Melloy, and Casey (1998)examined the effects of multiple variables onthe social behavior of children with autismand other developmental disabilities. Resultsof the research suggested that dyads pro-moted social interaction on the part of thechild with a disability better than triads. Fur-thermore, the study showed that social behav-iors were more prevalent during peer-initiatedfree play than when peers were instructed toteach a child with a disability how to play aspecific game. Despite these findings, few re-search studies have been published regardingattempts to increase social interaction be-tween children with and without disabilitiesoutside of the school setting. Though pre-school and elementary schools are critical are-nas for social development in young children,they are not the only settings where childrencome together, interact, and make friends.

Authors are members of the University of Mary-land Gemstone Program, a multidisciplinary four-year research program for selected honors students.Dr. Francey Kohl, the faculty mentor, assisted inpreparing the manuscript. Correspondence con-cerning this article should be addressed to Dr.Francey Kohl, 1308 Benjamin Building, Departmentof Special Education, University of Maryland, Col-lege Park, MD 20742.



One widely overlooked arena for social inter-action is community recreation programs.

There is a demonstrated benefit of socialinteraction during leisure activities for chil-dren with disabilities (Bedini, 2000). Studieshave examined methods to improve inclusiverecreation in areas such as staff training andadministrative or structural modification(Herbert, 2000; Schleien, Germ, & McAvoy,1996). In addition, several studies have dem-onstrated the benefit of inclusive recreationon the social acceptance of children with dis-abilities (Anderson, Schleien, McAvoy, Lais, &Seligmann, 1997; Devine, 2004; Sable, 1995).However, these studies did not address im-proving interactions between children withand without disabilities, which is an exten-sively studied area in educational settings. Dueto the lack of intervention studies in leisuresettings, research is needed to determine theimpact of peer training at recreation sites.Therefore, the purpose of this investigationwas to examine effects of a peer interventionprocedure designed to increase interactionsbetween children with and without disabilitiesduring an inclusive summer day camp.

Method

Participant Selection

The study was conducted in an inclusive sum-mer day camp organized by a suburban com-munity recreation program in the mid-Atlan-tic region of the United States. Theinvestigation took place during regularlyscheduled camp hours (9am-5pm) at indoor

and outdoor camp facilities. The camp pro-gram included activities such as organizedteam sports, games, arts and crafts, and swim-ming. The ratio of camp staff to children wasapproximately 1 to 5. The ratio of childrenwith disabilities to children without disabilitieswas approximately 1 to 10. There were fourtwo-week camp sessions in which approxi-mately 20 campers attended per session. Thestudy was implemented during two camp ses-sions, the first and third of the sequence. Dur-ing each camp session, campers at three dif-ferent sites were observed. Participants weresummer day campers between five and tenyears of age and consisted of a combination ofboys and girls. Table 1 displays participantcharacteristics, including age and gender.

Children with disabilities. Recreation coor-dinators were asked to mail an informationletter and a permission form to parents of allcampers with disclosed disabilities, maintain-ing confidentiality until there was an agree-ment by parents for their child to participatein the research. Three campers with disabili-ties participated in each camp session, for atotal of six campers with disabilities over allcamp sites. All children with disabilities forwhom a parental permission form was re-turned were potential participants. Severity ofthe disabilities ranged from mild to moderate.

Children without disabilities. On the firstday of camp, campers without disabilities andtheir parents were approached. The researchwas explained and parents were asked if theywould allow their child to participate. If par-ents agreed to have their child participate,information letters and permission forms

TABLE 1

Characteristics of Campers With and Without Disabilities

Campers Without Disabilities Campers With Disabilities

CampSession Gender Age Gender Age Disability (Verbal Ability)

1-A Male 7 Female 5 Emotional Disorder (Verbal)1-B Female 7 Male 10 Autism (Verbal)1-C Male 5 Male 5 Autism (Language Delay)2-A Male 8 Male 8 Visual Impairment (Verbal)2-B Female 7 Male 10 Autism (Verbal)2-C Male 7 Male 7 Autism (Limited language usage)

STAR Intervention Program / 93

were provided and parents were asked to re-turn them by the second day of camp. Oncepermission was granted, three campers with-out disabilities from each camp session wereselected based on counselor recommenda-tions or random selection from those whovolunteered by returning their permissionform.

Procedure

Experimental design. A multiple probe sin-gle subject experimental design was used.During the second day of camp, baseline ob-servations of interactions between camperswith and without disabilities began. Camper Awas observed for a minimum of three baselinesessions at the same time that Campers B andC were observed for one observation. Whenbaseline data were stable for Camper A, theintervention procedures were implemented.The intervention continued with Camper Auntil a noticeable increase in the number ofinteractions was observed and then Camper Bwas observed for three additional baseline ob-servation sessions and Camper C was observedfor one additional baseline session. WhenCamper B had a stable baseline, the interven-tion procedure was implemented. Likewise,intervention continued until Camper Bshowed a noticeable increase in the numberof interactions. Intervention observationswere continued for Campers A and B, whileCamper C was observed for an additionalthree baseline observations before the inter-vention began.

Dependent variable and data collection. Thedependent variable was the percentage of in-tervals in which an interaction between camp-ers with and without disabilities occurred. Theprimary focus of each observation interval wasthe camper without a disability, namelyCamper A, B, or C. Interactions were definedas social behaviors occurring between camp-ers with and without a disability including: (a)non-verbal communication (e.g., sustainingeye contact for three seconds, smiling, wav-ing); (b) talking to each other; (c) directingan activity (e.g., explaining the directions foran art project); (d) sharing materials; (e) par-ticipating in an activity together (e.g., assem-bling a puzzle); (f) prompting a skill (e.g.,gesturing to cut materials, modeling how to

open a jar of paint); (g) physical assistance(e.g., taking the hand of camper with a dis-ability to connect the dots); and (h) physicalor verbal encouragement (e.g., patting on theback, giving a high five, saying “good job”).

Data were collected during 5 minute obser-vation sessions. Within the 5 minutes, therewere a total of twenty 10 second intervals forobserving that were separated by 5 secondintervals for recording. The observer wasprompted via an earphone with a tape re-corded signal to start observing, to stop ob-serving, or to record. A partial interval record-ing system was used in which an interval wasmarked indicating that an interaction was ob-served regardless of the frequency of interac-tions or length of time for each interaction. Atthe end of each observation session, the per-centage of intervals in which interactions oc-curred was determined. Data collection pro-cedures were identical for all baseline andintervention conditions across all campersand camp sessions.

Independent variable. After baseline datawere completed for the campers without dis-abilities, the first part of the peer interventionwas implemented. Using the “stay-play-talk”training by English et al. (1996) as a basis, theSTAR intervention was developed for thisstudy to target elementary school aged chil-dren in recreation settings. The interventionconsisted of four behaviors that were modeledand taught to the campers without disabilitiesto increase interactions with campers with dis-abilities: “S” stood for stay, “T” for talk, “A” forassist, and “R” for reward. Campers withoutdisabilities were asked after their baseline ob-servations were completed if they would par-ticipate in the STAR intervention. Before thetraining session began, the campers withoutdisabilities were each read an assent form andparticipation in the research was agreed to bythe campers.

At the time of the STAR intervention, camp-ers without disabilities were separated fromthe group for no longer than 20 minutes. Thesession began with a series of questions posedto the campers: (a) What is a disability? (b)What are different types of disabilities? and(c) What are possible similarities and differ-ences between their camp experience and theway campers with disabilities may experiencethe camp? If the campers did not respond,


appropriate answers were provided. Thecampers without disabilities were told thatduring an activity such as arts and crafts, theywould be given a button illustrating the STARacronym, which would be used as a reminderof how to interact with campers with disabili-ties. The campers were partnered with camp-ers with disabilities and asked to be a “STAR.”After showing the campers the button duringthe intervention, the acronym was explained.The explanation included examples as well asopportunities for the campers to participatein role playing the four behaviors of “STAR.”The scenarios, which were used in the roleplay, were chosen based partly on the disabil-ities of their assigned campers. After the STARintervention was completed, the campers re-turned to the group. Each day, when the tar-geted activity occurred (e.g., arts and crafts),the second part of the STAR intervention wasimplemented. The campers without disabili-ties were partnered with the assigned camperswith disabilities, given the STAR button, andreminded by one of the researchers to stay,talk, assist, and reward the campers with dis-abilities.

Interobserver reliability. Six observers (allauthors) acted as the primary and secondarydata collectors. Prior to the implementationof the study, all observers became familiarwith the definition of the dependent variableand what qualified as an interaction betweencampers. Data collectors were paired off andgiven a tape recorder with a double jack fortwo separate ear phones. Observers practiceddata collection and calculations on a video ofchildren interacting until a reliability agree-ment of 90% or above was reached. A point bypoint formula was used to calculate reliability:number of agreements of interaction intervalsdivided by the number of agreements plusdisagreements of interaction intervals multi-plied by 100. During camp observations, twoobservers entered the room with the tape re-corder and double jack earphones, sat as faraway as possible from one another and lis-tened for the observe, record, or stop promptswith separate headsets.

Procedural reliability. Procedural reliabilitywas calculated for two different aspects of thestudy: (a) during all training sessions in whichthe campers without disabilities were taughtthe STAR intervention and (b) before daily

observation sessions when a reminder of theSTAR procedures was given along with theSTAR button. Procedural reliability was ob-served and recorded by one of the authors. Achecklist of steps that was to be followed forthe STAR intervention was used to insure alltraining steps were completed. During proce-dural reliability, the researcher observed andrecorded the number of steps completed cor-rectly. To calculate reliability, the followingformula was used: number of steps completeddivided by total number of steps multiplied by100.

Results

Results of the STAR intervention program onthe percentage of interactions between camp-ers with and without disabilities in Camp Ses-sion 1 are presented in Figure 1. The baselinecondition for Camper 1-A consisted of threeobservation periods with no interactions oc-curring with the camper with a disability. Theintervention condition for Camper 1-A con-sisted of 12 observation periods with an aver-age of 33.8% interactions with the camperwith a disability, ranging from 5-55%. Thebaseline condition for Camper 1-B consistedof four observation periods with an average of3.8% interactions with the designated camperwith a disability and a range of 0-15%. Whenthe intervention condition was introduced forCamper 1-B over 12 observation periods, anaverage of 20.4% interactions occurred, rang-ing from 0-45%. The baseline condition forCamper 1-C consisted of five observation pe-riods during which no interactions occurredwith the camper with a disability. The inter-vention condition for Camper 1-C consisted ofthree observation periods with an average of35% interactions with a range of 10-75%.

Results of the STAR intervention programon the percentage of interactions betweencampers with and without disabilities overeach observation period in Camp Session 2are presented in Figure 2. The baseline con-dition for Camper 2-A consisted of three ob-servation periods with an average of 1.7% in-teractions with the camper with a disability,ranging from 0-5%. When the interventioncondition for Camper 2-A was in effect for 16observation periods, the average percentageof interactions was 31.6% with a range of 10-


Figure 1. Effects of STAR intervention program on percentage of interactions between campers with andwithout disabilities during camp Session 1.


Figure 2. Effects of STAR intervention program on percentage of interactions between campers with andwithout disabilities during camp Session 2.


55%. The baseline condition for Camper 2-Bconsisted of four observation periods with anaverage of 2.5% of interactions occurring withthe camper with a disability, ranging from0-10%. Once the intervention condition wasintroduced for Camper 2-B over 12 observa-tion periods, the average percentage of inter-actions was 16.7% with a range of 0%-60%.The baseline condition for Camper 2-C con-sisted of five observation periods with an aver-age of 1% of interactions occurring with thedesignated camper with a disability, rangingfrom 0-5%. The intervention condition forCamper 2-C consisted of 12 observation peri-ods with an average of 20.8% interactions witha range of 0-65%.

Results of this investigation showed that af-ter the intervention training was given to eachcamper without a disability, the average per-centage of interactions between campers withand without disabilities increased across all sixcampers. The mean percentage of increasedinteractions for all six campers from baselineto intervention conditions was 26.4%.

Interobserver reliability. During the baselineconditions for Camp Session 1 (N � 12)across all three campers, reliability measureswere taken on 75% of all sessions and themean reliability calculation was 100%. Duringintervention sessions for Camp Session 1 (N �27) across all three campers, reliability mea-sures were taken on 44% of all sessions, with amean of 97.5% agreement and a range of90-100%. Therefore, in Session 1 across allconditions and campers, reliability was takenon 54% of all observations with a mean of99.2% agreement and a range of 90-100%.

During baseline conditions for Camp Ses-sion 2 (N � 12) across all three campers,reliability measures were taken on 75% of allsessions and the mean reliability calculationwas 100%. During intervention conditions forCamp Session 2 (N � 40) across all threecampers, reliability measures were taken on27.5% of all sessions, with a mean of 95.8%agreement and a range of 90-100%. There-fore, in Camp Session 2 across all conditionsand campers, reliability was taken on 36.5% ofall observations with a mean of 98% rangingfrom 90-100%. Overall, for Camp Sessions 1and 2 across all baseline conditions, interven-tion conditions, and six campers, interob-server reliability was obtained on 44% of all

observations with a mean of 98.6% agreementand a range of 90-100%.

Procedural reliability. On 100% of all STARintervention training sessions across bothcamp sessions with six campers, proceduralreliability was taken by a secondary observer. Achecklist of 20 critical steps necessary to im-plement the STAR intervention program wasgenerated. During the intervention trainingsessions, the secondary observer indicatedwhether the trainer implemented the STARintervention consistently. The number ofsteps implemented during the interventiontraining divided by the total number of steps(N � 20) multiplied by 100 yielded proce-dural reliability results. For Camp Sessions 1and 2, the mean procedural reliability was100%.

On 38.9% of all daily reminder sessionsacross both camp sessions, procedural reliabil-ity was taken by a secondary observer. A check-list of five critical steps necessary to implementthe daily reminders was generated. During thedaily reminder sessions, the second observerindicated whether the trainer implementedeach step. The number of steps implementedduring intervention conditions divided by thetotal number of steps (N � 5) multiplied by100 yielded procedural reliability results. ForCamp Sessions 1 and 2, the mean proceduralreliability for the reminder procedures was100%.

Discussion

The STAR program was shown to be an effec-tive intervention to increase interactions be-tween campers with and without disabilities ineach of the six pairs in an inclusive summerday camp. These results were similar to thosefound by English et al. (1997), Goldstein andEnglish (1997), and Laushey and Heflin(2000) in school settings. It is believed thatthe increase in interactions between camperswith and without disabilities during the twoweek camp sessions was a direct result of acombination of variables. First, the initialtraining session provided useful informationon disabilities by emphasizing similaritiesacross all campers and helped to make thecampers feel more comfortable around theirpeers with disabilities. The importance oftraining for children without disabilities re-


garding how to effectively interact with chil-dren with disabilities has been echoed by sev-eral other studies (Goldstein & English;Gonzalez-Lopez & Kamps, 1997). Secondly,the STAR procedures (specifically, “stay, talk,assist, and reward”) were effective becausecampers without disabilities were given spe-cific ways to initiate and sustain interactionswith campers with disabilities. Finally, thedaily reminders were beneficial and necessarybecause they prompted the campers to assistcampers with disabilities in specific camp ac-tivities.

However, there were sudden drops in inter-actions for certain observation sessions. Inmost of these cases, camp circumstances be-yond the researchers’ control led to the de-creased frequency of interactions. Decreasesin interactions occurred when observed activ-ities ended prematurely. The remainder ofthe observation session was then conductedwhile transitioning to another activity and in-teractions were not as likely to occur. In in-stances where the frequency of interactionsdropped to zero, the primary reason was thatthe camper without a disability became tooengrossed in his or her own assigned activityand failed to interact with the camper with adisability.

The atmosphere surrounding the camp alsoaffected the frequency of interactions betweencampers. On multiple occasions the observa-tion periods were affected by changes in thedaily schedule that resulted in all or part ofthe observation occurring during activitieswhere interactions were made difficult by theconstraints of the activity. Not all observationswere able to occur during the passive activityof arts and crafts as originally planned and itwas found that physical, outdoor activities of-ten produced fewer interactions betweencampers. Some observation periods were alsoconducted during unstructured activities be-cause field trips occupied the majority of theday, which also decreased the frequency ofinteractions observed. Another factor wasdaily absences by either camper, interruptingthe flow of the intervention. Also, counselors,particularly the counselors in training (CIT)who were adolescent volunteers, on occasioninterfered with the campers during observa-tion sessions. Most of the campers with disabil-ities were assigned a CIT by the recreation

department who would sometimes monopo-lize the attention of his or her camper, whichobstructed the camper without a disabilityfrom assisting his or her partner.

Despite the success of the intervention, theprogram could be improved by refining theSTAR procedures. It was found that the train-ing for some of the campers without disabili-ties appeared to have been too advanced fortheir age, which ranged from five to eightyears. The concept of a disability was not al-ways understood by the younger campers,which affected the implementation of theSTAR procedures. This was especially true ofsituations where campers had a disability thatwas not visible. Some of the campers had milddisabilities, such as an emotional disorder,which were not discernable to other campers.Additionally, some campers without disabili-ties would focus on only one or two compo-nents of the STAR intervention, such as stayand talk. One possible way to remedy theproblem would be to use picture symbols forstay, talk, assist, and reward displayed on anindex card that could be placed in front of thecamper. The modeling and role-playing com-ponent of the intervention was also essential.Adding more practice to the role-playing com-ponent may help the camper without disabil-ities to better understand the expectations ofthe STAR procedure. This could be accom-plished by including a guided approach thatwould start with modeling, then a promptedrole-playing situation, and finally lead to inde-pendent role-playing by the camper.

Although the daily reminders were criticalto the success of the STAR program, theycould also be improved in three ways. The firstrecommendation is to implement a role-play-ing component similar to the initial interven-tion into the daily routine by having the camp-ers act out specific behaviors of the STARinterventions before receiving the buttoneach day. This may enhance younger camp-ers’ understanding of the procedures and fur-ther increase their interactions. The secondrecommendation would be to allow the campstaff to give the daily reminders. This familiar-ity may help the campers feel more comfort-able in asking questions about the campers’disability, how to assist, and what is expectedof them, as well as facilitate generalization ofthe STAR behaviors across the entire day. The


last recommendation would be to provideconstructive feedback at each daily remindersession to the campers without disabilitiesabout their interactions the previous campday.

Limitations

There were several factors both foreseen andunforeseen that produced limitations to theresults of the STAR intervention. As is oftenthe case with inclusive recreation programs,there were very few participants with disabili-ties at each camp site. For this study therewere only one or two campers with disabilitieswho had permission to participate in the studyat each camp site and these campers had milddisabilities that were not apparent to the othercampers. It is unclear if the results of the studywould have been different if the campers withdisabilities had impairments easily recogniz-able to the other campers (e.g., a child in awheelchair).

An additional limitation to the researchcomes as a result of the design of the inter-vention itself. During the baseline condition,the campers with and without disabilities wereseated away from each other, essentially elim-inating any opportunity for interaction be-tween the two. During the intervention obser-vations, the campers with and withoutdisabilities were seated next to each other.Simply placing the campers next to each othermay have increased the incidence of interac-tion between them. The campers without dis-abilities were always free to move during theintervention; however, most stayed in the seatthey were asked to sit in. Had the participantpool been larger, it may have been possible totake baseline data and complete the interven-tion with a camper who was initially sittingnext to the camper with a disability.

Future Directions

The positive results of the STAR interventionhave far-reaching directions for future re-search. It is important to replicate across dif-ferent variables so the STAR intervention maybe used by a variety of recreation programs inthe future. It is also important to replicate theprocedures using campers with a variety ofdisabilities and a range of severity. This should

include campers with physical, emotional, andcognitive disabilities. Procedures should bereplicated using different recreation settingsand activities, such as team sports, games,swimming, and dance. A final factor to con-sider for future research is the age and gendersimilarities of the pair of campers. Genderand age differences in the dyads may haveaffected the results of this investigation, butfuture replications with pairs of the same gen-der or age versus pairs of differing gender andage will give further insight into facilitatinginteractions in camp settings. Future studiescould also explore different interventionsbased on the age of the child, such as the useof pictures for younger children versus ab-stract presentations for older children. Futureresearch will strengthen the effectiveness ofthe STAR intervention and will expand theopportunity for many recreation programs touse this method to increase friendshipsamong campers with and without disabilities.

References

Anderson, L., Schleien, S. J., McAvoy, L., Lais, G., &Seligmann, D. (1997). Creating positive changethrough an integrated outdoor adventure pro-gram. Therapeutic Recreation Journal, 31, 214–229.

Bedini, L. A. (2000). “Just sit down so we can talk”:Perceived stigma and community recreation pur-suits of people with disabilities. Therapeutic Recre-ation Journal. 34, 55–68.

Devine, M. A. (2004). “Being a ‘doer’ instead of a‘viewer’”: The role of inclusive leisure contests indetermining social acceptance for people withdisabilities. Journal of Leisure Research, 36, 137–159.

English, K., Goldstein, H., Kaczmarek, L., & Shafer,K. (1996). “Buddy skills” for preschoolers. Teach-ing Exceptional Children, 28(3), 62–66.

English, K., Goldstein, H., Shafer, K., & Kaczmarek,L. (1997). Promoting interactions among pre-schoolers with and without disabilities: Effects of abuddy skills-training program. Exceptional Chil-dren, 63, 229–243.

Garfinkle, A. N., & Schwartz, I. S. (2002). Peerimitation: Increasing social interactions in chil-dren with autism and other developmental dis-abilities in inclusive preschool classrooms. Topicsin Early Childhood Special Education, 22, 26–38.

Goldstein, H., & English, K. (1997). Interactionamong preschoolers with and without disabilities:Effects of across-the-day peer intervention. Journalof Speech, Language & Hearing Research, 40, 33–49.

Goldstein, H., Kaczmarek, L., Pennington, R., &


Shafer K. (1992). Peer-mediating intervention:Attending to, commenting on, and acknowledg-ing the behavior of preschoolers with autism. Jour-nal of Applied Behavior Analysis, 25, 289–305.

Gonzalez-Lopez, A., & Kamps, D. M. (1997). Socialskills training to increase social interactions be-tween children with autism and their typicalpeers. Focus on Autism and Other Developmental Dis-abilities, 12, 2–15.

Herbert, J. T. (2000). Therapeutic adventure staffattitudes and preferences for working with per-sons with disabilities. Therapeutic Recreation Jour-nal, 34, 211–226.

Hundert, J., & Houghton, A. (1992). Promotingsocial interaction of children with disabilities inintegrated preschools: A failure to generalize. Ex-ceptional Children, 48, 311–320.

Laushey, K. M., & Heflin, L. J. (2000). Enhancingsocial skills of kindergarten children with autismthrough the training of multiple peers as tutors.Journal of Autism and Developmental Disabilities, 30,183–193.

Odom, S. L., Chandler, L. K., Ostrosky, M., McCo-nnell, S. R., & Reaney, S. (1992). Fading teacher

prompts from peer-initiated interventions foryoung children with disabilities. Journal of AppliedBehavior Analysis, 25, 307–317.

Sable, J. R. (1995). Efficacy of physical integration,disability awareness, and adventure programmingon adolescents’ acceptance of individuals withdisabilities. Therapeutic Recreation Journal, 29, 206–227.

Sasso, G. M., Mundschenk, N. A., Melloy, K. J., &Casey, S. D. (1998). A comparison of the effects oforganismic and setting variables on the socialinteraction behavior of children with develop-mental disabilities and autism. Focus on Autism andOther Developmental Disabilities, 13, 2–17.

Schleien, S. J., Germ, P. A., & McAvoy, L. H. (1996).Inclusive community leisure services: Recom-mended professional practices and barriers en-countered. Therapeutic Recreation Journal, 30, 260–273.

Received: 23 August 2006Initial Acceptance: 17 October 2006Final Acceptance: 9 January 2007


Use of a Handheld Prompting System to TransitionIndependently Through Vocational Tasks for Students with

Moderate and Severe Intellectual Disabilities

David F. CihakUniversity of Tennessee

Kelby Kessler and Paul A. AlbertoGeorgia State University

Abstract: The use of a handheld prompting system by four students with moderate to severe intellectualdisabilities to independently transition between an ordered chain of tasks was examined in a communityvocational setting Effectiveness of the handheld prompting system was assessed using a multiple-probe designacross participants. Analysis of the data revealed that students successfully used the handheld system to increaseindependent transitions from task to task. Independent transitioning was maintained at a 100% level for upto nine weeks.

The need to shift stimulus control from ateacher to some other stimulus in the naturalenvironment provides a foundation for thedevelopment of self-management procedures(Cooper, Heron, & Heward, 1987; Kazdin,1994; Repp, 1983). Self-management proce-dures frequently limit discriminative stimuliallowing the individual to focus attention on atarget behavior (Gifford, Rusch, Martin, &White, 1984). One self-management proce-dure that promotes a shift in stimulus controlby limiting discriminative stimuli is the use ofhandheld prompting systems (Cihak, Kessler,& Alberto, in press; Davies, Stock, & Weh-meyer, 2002a, 2002b, 2004; Ferguson, Myles-Smith, & Hagiwara, 2005; Furniss et al., 2001;Riffel et al., 2005).

Workers with moderate and severe intellec-tual disabilities often are asked to changefrom one task to the next without help. Thosewho are unable to do so may experience de-creased success (Lagomarcino, Hughes, &Rusch, 1989; Mank & Horner, 1988), main-tain excessive dependence on job coaches andexhibit limited behavioral maintenance(Rusch, 1986), engage in off-task behaviorwhile waiting for external delivered prompts

(Browder & Shapiro, 1985; MacDuff, Krantz,McClannahan, 1993), and maintain low pro-ductivity rates (Agran, Fodor-Davis, & Moore,1986). The probability of those undesirableoutcomes was decreased by instruction in spe-cific self-management skills (Certo, Mezzullo,& Hunter, 1985; Rusch, Martin, Lagomarcino,& White, 1987). Thus, the use of handheldprompting systems by workers with moderateand severe intellectual disabilities may resultin employee autonomy and decrease depen-dence on coworkers, teachers, job coaches,and peers (Cihak et al., in press; Davies et al.,2002a, b, 2004; Ferguson et al., 2005; Furnisset al., 2001; Riffel et al., 2005).

Handheld computers (e.g., PDA, PocketPC) are promising technology devices becausethey are portable, inexpensive, reliable, easyto maintain, program, use, and are sociallydesirable. One way to distinguish betweenhandheld computer technologies is betweencommercially available devices and those thatare custom made for an individual person(Cook & Hussey, 2002). The term commer-cially available refers to devices that are mass-produced. These include commercial devicesdesigned for the general population. Increas-ingly, commercial products are being de-signed according to the principles of universaldesign. Universal design is the design of tech-nologies to be usable by all people, to thegreatest extent possible, without the need foradaptation or specialized design. In this ap-

Correspondence concerning this article shouldbe addressed to David Cihak, College of Education,Health, & Human Sciences, A412 Claxton Complex,Knoxville, TN 37996-3442.



proach, features are built into the product(e.g., various display options-visual, auditory;alternatives to reading text-icons, pictures),which makes a product more accessible toindividuals with disabilities. This is much lessexpensive than adapting a product after pro-duction in order to meet the needs of anindividual with a disability.

If commercially available devices cannotmeet an individual’s needs, it may be modi-fied. However, when modification or commer-cial devices are not appropriate, it is necessaryto design one specifically for the task-at-hand.This approach results in a custom device.Since custom products are not mass-pro-duced, a custom device costs are much higherbecause it is a special product or a “one of akind” and the costs of development must berecovered from the smaller production.

Using a custom-made handheld device andsoftware, Davis et al. (2002a, 2002b, 2004)reported that participants with intellectual dis-abilities were more independent, requiredfewer external prompts, were more produc-tive, and made fewer errors. Custom-madehandheld computers and software have beenused successfully across various disability char-acteristics including, developmental delay, au-tism, and Prader Willi Syndrome (Riffel et al.,2005). Moreover, custom-made devices havesupported individuals with special needs incareer and leisure activities (Furniss et al.,2001). Furniss et al. concluded that the use ofa custom-handheld prompting system wasmore effective than static picture prompts in abooklet, it was easily used in real work settings,and that students with severe disabilities pre-ferred the handheld device to the picturebooklet.

Using a commercially produced handhelddevice and software, Ferguson et al. (2005)successfully decreased adult reliance to com-plete tasks at home and school for one ado-lescent with Asperger’s Syndrome. Cihak et al.(in press) also used a commercially producedhandheld device to successfully teach studentswith moderate intellectual disabilities to oper-ate the use of the handheld prompting systemand to generalize skills across increasing com-plex vocational tasks in the work setting with-out additional training. Although the litera-ture has noted that handheld computersefficacy for task acquisition, generalization,

and maintenance, Davies et al. (2004) notedthat further research was needed to assess theeffectiveness of handheld computers as aprompting system across a variety of tasks,domains, and ecologically valid work and em-ployment settings.

The purpose of this study was to determinethe effectiveness of a commercially-producedhandheld computer, as a prompting system tofacilitate the independent transitions fromtask to task in a community-based vocationalinstructional site for students with moderateand severe intellectual disabilities.

Method

Participants

Four students, Aaron, Bill, Cate, and Dougwere selected to participate based on the fol-lowing: (a) willingness to participate, (b) levelof cognitive functioning within the moderateto severe intellectual disability range, (c) cur-rent participation in a high school programwith regularly scheduled community-based in-struction, (d) no sensory deficits, (e) parentalpermission, and (f) the student’s verbal agree-ment to participate. Aaron and Doug were 16years olds with a full-scale IQ of 36 and 40,respectfully. Bill and Cate were 17 years oldwith a full-scale IQ of 48 and 50, respectfully.IQ’s were assessed using the Wechsler Intelli-gence for Children (Wechsler, 1991) forAaron, Bill, and Doug. The Standford Binet(Thorndike, Hagen, & Sattler, 1986) was usedto assess Cate. Students were able to indepen-dently complete individual tasks; however,each was dependent on an external source fortask transitions. According to teacher reports,after the completion of individual task, if stu-dents were not prompted immediately to be-gin the next task, each student would usuallyengage in some form of off-task behavior.

Settings

Pretraining instruction using the handheldcomputer occurred in each student’s schoolresource classroom. Baseline, handheldprompting, and maintenance phases occurredduring community-based instruction (CBI) inthree community settings; grocery store, de-partment store and restaurant. Stores and res-

Handheld Prompting System / 103

taurant were selected because of the conve-nient location to the students’ neighborhoodschool. Community instruction for Aaron andBill was at a grocery store, Doug at a restau-rant, and Cate at a department store.

Materials

A Kodak DX3600 Zoom digital camera wasused to digitally photograph each task ana-lyzed step. Digital photos were then down-loaded into an Axium X30 handheld com-puter that was used to deliver the picture andauditory prompts. The Axium X30 was se-lected since it was the least expensive devicethat allowed capabilities of photo display andto record narration. Picture Perfect softwarewas used to develop the picture and auditoryprompts task sequence. The Picture Perfectsoftware was selected due to its relative inex-pensive price and capabilities of creating arelatively easy picture and auditory promptingsystem. Students placed the handheld com-puter in a pack fastened to their waist with asmall headphone that attached around theear. The headphone wire was worn under theuniform to reduce interference during taskengagement. Students advanced the prompt-ing system by pressing an arrow hardware but-ton.

Tasks

Ten vocational tasks were identified for eachstudent at each work site. Table 1 displays the

list of tasks each student completed and tran-sitioned to and from. Tasks were randomlyordered so that no two individual tasks wouldfollow one another from session to session.

Data Collection

Event recording was used to record the num-ber of times a student independently transi-tioned to a separate vocational task within aprescribed chain of tasks. Ten tasks were iden-tified for each student. Since each studentbegan their workday in the employees’ break-room, a total of 10 transitions were available.An independent transition was defined as thestudent’s ability to move from task to taskwithout relying on an individual to direct himor her. A student’s independent transition toa task was recorded as correct when the stu-dent completed the first task, physicallymoved to the second task area, and completedthe first step of the next task. These transitionsoccurred after correct completion of each in-dividual ordered task. A transition was consid-ered assisted if the student asked for direc-tions on what was next, waited for someone toprovide assistance (a verbal reminder to use athe handheld prompting system was providedafter 10s), or received unanticipated instruc-tions from a coworker (such as gestures). If astudent, after receiving a handheld prompt,moved to an unprompted task out of se-quence, he or she was provided with a verbalreminder of the next task and the responsewas recorded as assisted. The total number of

TABLE 1

Tasks assigned to each student

Students

Aaron Bill Cate Doug

Tasks 1) Gathering carts, 1) Gathering carts, 1) Straightening mushrooms, 1) Preparing coleslaw,2) Stocking milk, 2) Stocking milk, 2) Stocking bananas, 2) Preparing broccoli,3) Vacuuming, 3) Vacuuming, 3) Stocking pineapples, 3) Skewering shrimp,4) Preparing rolls, 4) Preparing rolls, 4) Cleaning fitting room, 4) Preparing tea,5) Putting back returns, 5) Putting back returns, 5) Cleaning registers, 5) Rolling silverware,6) Stocking cans, 6) Stocking cans, 6) Clothes processing, 6) Bussing tables,7) Making cookies, 7) Making cookies, 7) Sweeping, 7) Setting tables,8) Cleaning registers, 8) Cleaning registers, 8) Cleaning windows, 8) Sweeping,9) Cleaning windows, 9) Cleaning windows, 9) Stocking picture frames, 9) Cleaning windows,10) Sweeping 10) Sweeping 10) Dusting shelves 10) Taking out trash


independent transitions were totaled and di-vided by 10 to compute a percentage of inde-pendent transitions for each student.

Experimental Design

A multiple-probe design across participants(Barlow & Hersen, 1984) was used to deter-mine the efficacy of the handheld promptingsystem and student’s independent transitions.The multiple-probe design allowed sequentialapplication, comparison effectiveness, and anopportunity to replicate the effects of thehandheld prompting system across students.The study included three phases; baseline, ac-quisition of handheld prompting instruction,and maintenance phases. Prior to baseline apre-training period occurred during whichthe students were taught to operate the hand-held prompting system, to select the differenttasks to perform, to press the hardware buttonto advance to the next step of the task analysisand to follow the recorded directions.

Tasks were ordered in a semi-random fash-ion on a session-by-session basis to ensure thatno two individual tasks would follow one ses-sion to the next ensuring different task-to-tasktransitions for each session. Before the imple-mentation of the intervention phase, thenames all 10 tasks were written on a piece ofpaper and randomly drawn. If a task wasdrawn, that followed the same task from theprevious session, it was returned to the draw-ing and another task was selected. The ran-domization of task ordering allowed differenttransitions to occur on a session-by-session ba-sis, which reduced the likelihood of practiceeffects and students remembering what tran-sition was next.

Experimental Procedures

Pretraining. Similar to Cihak et al. (inpress), prior to baseline, students’ partici-pated in a pretraining period. For the firstphase of pretraining, students were instructedhow to operate the handheld computer. Theywere instructed to physically turn on the de-vice, to wear the headphones, to select color-cued icons representing the different tasks orjobs to perform, and to select a color-cuedhardware button to advance to the next stepof the task analysis.

In the second phase, students were requiredto turn-on the device and select a prerecordedicon, which caused a popped-up window tooccur with a familiar picture plus auditoryprompt, and to follow each instruction. Stu-dents also were required to press the hardwarebutton to display the next direction. The pre-recorded device instructed students to com-plete a two-step task familiar to students andnormally associated with their morning class-room routine. Students were instructed to“close the door and hang up your coat,” or “sitdown at the table and pick-up your pencil.”Each student was required to reach a criterionof 100% accuracy for two consecutive sessions.

Baseline. During baseline, the number ofindependent transition students made be-tween assigned tasks within a vocational sitewas recorded. If a student was unable to inde-pendently move to the next ordered task andsimply waited for assistance, relied on acowoker or peers to direct him or her to thenext task, or asked a teacher or supervisor fordirectional assistance, assistance was providedand the transition was recorded as “assisted.”Assistance for all transitions was provided byan external source only after 10-s interval ofno response. Data were collected until a stablebaseline was achieved for a minimum of fivesessions.

Handheld prompting procedures. All studentsstarted their workday in the employees’ break-room. Students were provided with the hand-held prompting system and headphones andinstructed to turn on the device. After thedevice was activated, a pop-up window with anicon of the targeted task was displayed. Afterthe student pressed the icon, a picture andauditory prompt of the first step of the taskwas displayed. Students then pressed the col-or-cued hardware button to advance to thenext step of the task. After the student pro-gressed through the entire chained task, apicture and auditory prompt cued the studentto transition to the next task. This processcontinued until all 10 tasks were complete. Ifa student did not transition, requested assis-tance, and/or 10-s interval elapsed with noresponse, the student was prompted to watchthe handheld device screen. Criterion forcompletion of this phase was 100% indepen-dent transitions for three consecutive sessions.


Maintenance procedures. Follow-up probeswere collected nine weeks after the studentmeet acquisition criterion. Follow-up probesoccurred in the community setting where thestudent was initially trained. Follow-up probeswere collected to determine if the initial in-structional affected the student’s performanceover time.

Reliability

Interobserver reliability data and proceduralreliability data were collected simultaneouslyby the primary investigator and the classroomteacher. Interobserver and procedural reli-ability data were collected during 33% of base-line and each concurrent phase. Observersindependently and simultaneously recordedthe number of steps the student performedindependently or the required prompt andresponse time. Interobserver agreement wascalculated by dividing the number of agree-ments of student responses by the number ofagreements plus disagreements and multiply-ing by 100. Interobserver reliability rangedfrom 96 to 100%, with a mean of 99% agree-ment. The mean interobserver reliabilityagreement for each student across conditionswas Aaron, 97%; Bill, 100%; Cate, 100%, andDoug, 97%.

Procedural integrity measures check the in-vestigator’s performance by using the correctprompting hierarchy and response time. Theclassroom teacher was trained using an item-ized checklist that listed the task-analyzedsteps of each task and the level of prompt. Theteacher was considered successfully trained af-ter completing 100% of the checklist for threeconsecutive trials. The procedural agreementlevel was calculated by dividing the number ofobserved teacher behaviors by the number ofplanned teacher behaviors and multiplying by100 (Billingsley, White, & Munson, 1980).Procedural reliability ranged from 97-100%,with a mean of 99%. The mean proceduralreliability agreement for each student acrossconditions was: Aaron, 99%; Bill, 100%; Cate,96% and Doug 100%

Results

Number of independent transitions made bystudents during baseline, handheld prompt-

ing intervention, and maintenance phases inthe vocational setting is presented in Figure 1.During baseline, students demonstrated lim-ited ability to independently transition be-tween specific work tasks. Number of indepen-dent task transitions made by students duringthis phase was zero. When students used thehandheld prompting system during interven-tion, increases in the number of independenttransitions were apparent. Number of inde-pendent transitions ranged from 40% to100% with a mean of 86%. Aaron demon-strated no independent transitions duringbaseline and increased independent transi-tioning to a mean of 82% (range � 40-100%)during intervention. Bill demonstrated no in-dependent transitions during baseline and in-creased independent transitioning to a meanof 84% (range � 50-100%) during interven-tion. Cate demonstrated no independent tran-sitions during baseline and increased inde-pendent transitioning to a mean of 80%(range � 50-100%) during intervention.Doug demonstrated no independent transi-tions during baseline and increased indepen-dent transitioning to a mean of 94% (range �80-100%) during intervention. Number of in-structional sessions to reach criteria rangedfrom nine to five with a mean of six. More-over, all students maintained transitioningnine weeks later with 100% independence.

Discussion

The purpose of this study was to determinethe effectiveness of a commercially-producedhandheld computer, as a prompting system tofacilitate independent transitions from task totask in a community-based vocational instruc-tional site for students with moderate and se-vere intellectual disabilities. Analysis of thedata indicated a functional relationship intask transitions performance between baselineand intervention replicated across partici-pants. The current investigation contributesto a growing body of research examining strat-egies to increase vocational skills of studentswith intellectual disabilities. This study repli-cates findings from previous studies whichfound that students with moderate to severeintellectual disabilities can learn to effectivelyuse handheld prompting systems, increase stu-dent autonomy at the workplace, and de-


crease dependency on coworkers, teachers,job coaches, and peers (Cihak et al., in press;Davies et al., 2002a, 2002b, 2004; Ferguson et

al., 2005; Furniss et al., 2001; Riffel et al.,2005).

Prior to the handheld prompting interven-

Figure 1. Number of independent tasks transitions across students.


tion, all students relied on specialized servicesand personnel to assist in meeting the stu-dents’ vocational needs. Unfortunately, natu-ral support options in the workplace are re-duced greatly for individuals who requireextensive and pervasive services or who areunable to work independently. Nisbet (1992)referred to natural supports as reliance onpersons within typical environments. Beforeplacement of a worker with disabilities, theworkplace should be examined carefully todetermine the availability of natural supportsand social interactions so that supported em-ployment services can build upon what is al-ready in place. That is, supervisors and co-workers, rather than job coacher or vocationalcounselors, would be responsible for provid-ing some services to assist in successful jobplacement and job retention. However, theconcept of natural supports also includes as-suring employers of their minimal involve-ment (Nisbet & Hagner, 1988). Althoughbusinesses have begun to recognize the needto support all workers, how much support isminimal?

Natural supports are successful for individ-uals with disabilities who require minimal as-sistance. However, natural supports becomeproblematic if the expectation is that the em-ployer and/or coworker must consistently di-rect and supervise the worker. Issues includ-ing (a) training in strategies and techniques,and (b) employers and coworkers’ skill levelto reliably implement intervention plans be-come heighten when workers require exten-sive and pervasive assistances. Additionally, ifthe student requires extensive supports andservices, do employers or coworkers find theindividual with disability an interference withtheir performance?

This study extends the research literature ofhandheld prompting systems by enabling stu-dents to use the handheld prompting systemto independently transition from task to taskin a vocational setting. Workers with moderateand severe intellectual disabilities often areasked to change from one task to the nextwithout help. Those who are unable to do somay experience decreased success resulting inlow productivity rates and dependency on jobcoaches. These prompts were effective forteaching students to manage their own taskchange behaviors. The use of the handheld

prompting system served as the stimulus con-trol for the desired behavior of independenttransitioning. With this methodology, teach-ers and professionals can increase the level ofindependence, self-sufficiency and the qualitylife of students with disabilities.

A second purpose of this study was to in-crease the probability of long-term mainte-nance of skills addressed by the use of thehandheld prompting system. Since the profes-sional literature recognizes that maintenanceis a difficult skill for students with moderateand severe intellectual disabilities to learn, it isincumbent upon teachers and professionals toaddress these concerns in instructional pro-gram planning and teaching strategies. Theuse of a handheld computer prompting sys-tem is one way of enhancing this instructionalcomponent.

A third purpose was to use a commerciallyavailable handheld computer to enhance skillacquisition and maintenance. The advantagesof using commercially available products in-clude lower costs and the availability supportsand technical assistances. Moreover, when theuniversal design approach is applied, accessi-bility and usability of handheld computers in-creases. More individuals with disabilities canthen utilize this promising technology, whichpromotes greater independence.

Several limitations of this study may haveaffected the results and interpretations. First,students were familiar with the vocational site;they could easily navigate to other areas of thestore or restaurant. Novel vocational sites mayrequire additionally prompts and directionalcues, which may produce differentiated out-comes. Second, the study was conducted withstudents who had extensive CBI experiences.Students with less extensive community expe-riences may require more intensive instruc-tion to acquire, generalize, and maintain tar-geted skills. Third, all students demonstratedno resistant behaviors toward wearing the de-vice and were extremely motivated using thehandheld computer. Students who are lessmotivated or resistant to using the handheldcomputer may perform differently.

Future research is needed to verify the re-sults of this study and to investigate generali-zation across novel work settings. Additionally,different type of transitions (e.g., place toplace, preferred to nonpreferred tasks) and a


type of skills (e.g., domestic, leisure, and com-munity) needs to be investigated. Future re-search also should attempt to replicate theseresults across natural support instructors (e.g.,job coach, coworker, and parent), studentcharacteristics, and the inclusion of self-evalu-ation steps.

References

Agran, M., Fodor-Davis, J., & Moore, S. (1986). Theeffects of self-instructional training on job-tasksequencing: Suggesting a problem-solving strat-egy. Education and Training of the Mentally Retarded,21, 273–281.

Barlow, D. H., & Hersen, M. (1984). Single caseexperimental designs: Strategies for studying behaviorchange (2nd ed.). Needham Heights, MA: Allyn &Bacon.

Billingsley, F. F., White, O. R., & Munson, R. (1980).Procedure reliability: A rational and an example.Behavioral Assessments, 2, 229–241.

Browder, D. M., & Shapiro, E. S. (1985). Applica-tions of self-management to individuals with se-vere handicaps: A review. Journal of the Associationfor Persons with Severe Handicaps, 10, 200–208.

Certo, N., Mezzullo, K., & Hunter, D. (1985). Theeffect of total task chain training on the acquisi-tion of busperson job skills at a full service com-munity restaurant. Education and Training of theMentally Retarded, 20, 148–156.

Cihak, D. F., Kessler, K., & Alberto, P. A. (in press).Generalized use of handheld prompting systems.Research in Developmental Disabilities.

Cook, A. M., & Hussey, S. M. (2002). Assistive tech-nologies principles and practices. St. Louis, MO:Mosby.

Cooper, J. O., Heron, T. E., & Heward, W. L.(1987). Applied behavior analysis. Columbus, OH:Merrill Publishing Co.

Davies, D. K., Stock, S., & Wehmeyer, M. L. (2002a).Enhancing independent task performance for in-dividuals with mental retardation through use ofa handheld self-directed visual and audio prompt-ing system. Education and Training in Mental Retar-dation and Developmental Disabilities, 37, 209–218.

Davies, D. K., Stock, S., & Wehmeyer, M. L. (2002b).Enhancing independent time-management skillsof individuals with mental retardation using apalmtop personal computer. Mental Retardation,40, 358–365.

Davies, D. K., Stock, S., & Wehmeyer, M. L. (2004).A palmtop computer-based intelligent aid for in-dividuals with intellectual disabilities to increaseindependent decision making. Research and Prac-tice for Persons with Severe Disabilities, 28, 182–193.

Ferguson, H., Myles-Smith, B., & Hagiwara, T.

(2005). Using a personal digital assistant to en-hance the independence of an adolescent withasperger syndrome. Education and Training inMental Retardation and Developmental Disabilities,40, 60–67.

Furniss, F., Lancioni, G., Rocha, N., Cunha, B.,Seedhouse, P., Morato, P., et al. (2001). VICAID:Development and evaluation of palmtop-basedjob aid for workers with severe developmentaldisabilities. British Journal of Educational Technol-ogy, 32, 277–287.

Gifford, J., Rusch, F., Martin, J., & White, D. (1984).Autonomy and adaptability: A proposed technol-ogy for maintaining work behavior. In N. Ellis &N. Bray (Eds.), International review of research onmental retardation: Vol. 12. (pp. 285–314). NewYork: Academic Press.

Kazdin, A. E. (1994). Behavior modification in appliedsettings. Pacific Grove, CA: Brooks/Cole Publish-ing Co.

Lagomarcino, T. R., Hughes, C., & Rusch, F. R.(1989). Utilizing self-management to teach inde-pendence on the job. Education and Training of theMentally Retarded, 24, 139–148.

MacDuff, G. S., Krantz, P. J., & McClannahan, L. E.(1993). Teaching children with autism to usephotographic activity schedules: Maintenanceand generalization of complex response chains.Journal of Applied Behavior Analysis, 26, 89–97.

Mank, D., & Horner, R. H. (1988). Instructionalprogramming in vocational education. In R. Gay-lord-Ross (Ed.), Vocational education for persons withhandicaps (pp. 142–173). Mountain View, CA:Mayfield.

Nisbet, J. (1992). Natural supports in school, at work,and in the community for people with severe disabilities.Baltimore: Paul H. Brooks.

Nisbet, J., & Hagnet, D. (1988). Natural supports inthe workplace: A reexamination of supported em-ployment. Journal of the Association for Persons withSevere Handicaps, 13, 260–267.

Repp, A. C. (1983). Teaching the mentally retarded.Englewood Cliffs, NJ: Prentice-Hall, Inc.

Riffel, L., Wehmeyer, M. L., Turnball, A. P., Latti-more, J., Davies, D., Stock, S., et al. (2005). Pro-moting independent performance of transition-related tasks using a palmtop pc-based self-directed visual and auditory prompting system.Journal of Special Education Technology, 20, 5–14.

Rusch, F. R. (1986). Developing a long-term fol-low-up program. In F. R. Rusch (Ed.), Competitiveemployment issues and strategies (pp. 225–232). Bal-timore: Paul H. Brookes.

Rusch, F. R., Martin, J. E., Lagomarcino, T. R., &White, D. M. (1987). Teaching task sequencingvia verbal mediation. Education and Training inMental Retardation, 22, 229–234.


Thorndike, R. L., Hagen, E. P., & Sattler, J. M.(1986). The Standford-Binet Intelligence Scale, FourthEdition: Guide for administering and scoring. Chi-cago: Riverside.

Wechsler, D. (1991). Wechsler Intelligence Scale forChildren, Third Edition Manual. San Antonio, TX:

Psychological Corporation Harcourt Brace Jo-vanovich.

Received: 3 May 2006Initial Acceptance: 1 July 2006Final Acceptance: 10 October 2006


Peer-Implemented Time Delay Procedures on the Acquisitionof Chained Tasks by Students with Moderate

and Severe Disabilities

Janet Read GodseyChristian County Schools, Hopkinsville, Kentucky

John W. Schuster, Amy Shearer Lingo, Belva C. Collins,and Harold L. Kleinert

University of Kentucky

Abstract: This study evaluated the effectiveness of and reliability of peer tutors implementing a constant timedelay procedure when teaching four high school students with moderate and severe disabilities to prepare foodsusing picture recipes. We used a multiple probe design across subjects to determine the effectiveness of the peertutor implemented constant time delay procedure on the acquisition of chained food preparation tasks. Dataindicate that the tutors were effective in teaching the four students to prepare food using picture recipes. Peertutors also implemented the procedure with a high degree of reliability. In addition, all students maintained thetasks at high levels and responded with 100% accuracy during the final maintenance assessment. Discussionincludes a comparison of reliability data with other studies utilizing teacher-implemented constant time delayand chained tasks.

Finding the personnel required to providehigh quality, individualized instruction for stu-dents with moderate and severe disabilities isone of the most difficult obstacles classroomteachers face. Because the intellectual andphysical capabilities of many students withmoderate and severe disabilities require re-peated, systematic, individualized instruction,teachers often find that that students mayhave to do a lot of waiting and “seatwork”activities while teacher time is spent for indi-vidualized and small group instruction else-where. Teachers who have many students re-quiring this type of instruction find it nearlyimpossible to provide students with sufficient

opportunities to respond during instructionalsessions, thereby limiting the student’s oppor-tunities in building acquisition skills to flu-ency, maintenance, and generalization levels(Kamps, Locke, Delquadri, & Hall, 1989).

Although placement of students with mod-erate and severe disabilities into general edu-cation classrooms has produced mixed reac-tions from educators and given the ever-growing number of peers without disabilitiesin high school special education classrooms incredit generating peer programs, it does pro-vide an opportunity to utilize instructional as-sistance in the form of peer tutors. Classroomteachers may be reluctant to use this instruc-tional arrangement for numerous reasons.First, teachers may wonder if peer-deliveredinstruction is as reliable or efficient as teacherand paraprofessional instruction. Second,they also may expect that peer tutors will lackthe discipline to perform the procedures sys-tematically. Finally, teachers may have con-cerns that peer tutors, even after extensivetraining, may not generalize their use of sys-tematic procedures when assigned to teachdifferent skills or with different students.

This study was the first author’s thesis and wascompleted as part of the requirements for a Masterof Science degree in the Department of SpecialEducation and Rehabilitation Counseling at theUniversity of Kentucky. Correspondence concern-ing this article should be addressed to John W.Schuster, 229 Taylor Education Building, Depart-ment of Special Education and RehabilitationCounseling, University of Kentucky, Lexington, KY40506-0001. E-mail: [email protected]


Peer-Implemented Time Delay / 111

Research has supported the use of peerswhen teaching skills to students with moder-ate and severe disabilities (e.g., Koury & Brow-der, 1986; Miracle, Collins, Schuster, &Grisham-Brown, 2001; Romer, Busse, Fewell,& Vadasy, 1985; Tekin-Iftar, 2003; Tekin &Kircaali-Iftar, 2002; Werts, Caldwell, & Wolery,1996). For example, Kamps et al. (1989) useda multiple probe across behavior design toteach discrete tasks to two elementary schoolstudents with autism. Two fifth graders with-out disabilities served as peer tutors. Tutorsused verbal reinforcement, instructive feed-back, and model and verbal prompts to effec-tively teach skills such as identification ofcoins, naming of opposites, and reading com-prehension.

Peer tutors also can successfully teachchained tasks to students with disabilities.Werts et al. (1996) taught elementary-agedpeer tutors to implement an observationallearning strategy to teach students with dis-abilities a variety of chained tasks. Target skillsincluded sequencing numbers using tiles,playing a cassette tape, sharpening a pencil,using a calculator to compute simple additionproblems, and using a computer game. Theresearch found that peer tutors were success-ful in teaching the target skill and used thetraining procedure with a high degree of pro-cedural reliability indicating that peer tutorscan provide highly effective instruction of sim-ple chained tasks.

In addition to research finding that peertutors can effectively teach students with dis-abilities, studies also have addressed whetherpeers can deliver instruction as effectively andefficiently as teachers (Romer et al., 1985; Mir-acle et al., 2001). Romer et al. compared theefficiency of peer tutor instruction to teacherinstruction on the performance of vocationalskills by students classified as deaf-blind withsevere mental retardation. Results indicatedthat acquisition measures of targeted skills bystudents were similar for both peer tutor in-struction and teacher instruction. Studentswith disabilities, on the average, took fewersessions to reach criterion with peer tutor in-struction but required more assistance thanfor tasks taught by the teacher. Results indi-cated that there was little difference betweenpeer instruction and teacher instruction. Ad-ditionally, Miracle et al. compared teacher-

deliver instruction to peer-delivered instruc-tion when teaching high school students withmoderate disabilities to read sight words withthe constant time delay procedure. Anadapted alternating treatments design wasused to compare the two interventions. Re-sults indicated that peer-delivered and teach-er-delivered instruction was effective. Theteacher-delivered instruction occurred at aquicker pace (i.e., less time per session) butthe differences were minimal.

Although research has demonstrated thatpeers can implement a variety of instructionalstrategies with discrete, and to a much lesserextent, chained tasks, there is little researchinvolving peer tutors using systematic re-sponse prompting strategies (Wolery, Ault, &Doyle, 1992) like the system of least prompts,simultaneous prompting, and constant timedelay. Tekin-Iftar (2003) researched the effec-tiveness of peer tutors in delivery of a simul-taneous prompting procedure to four stu-dents, aged 10 through 13, with a variety ofdisabilities. The author found that peer tutorssuccessfully (a) taught students with disabili-ties to expressively identify community signsand (b) embedded instructive feedback (i.e.,definitions of the signs) in the consequentevent. These results also occurred whenTekin-Iftar and Kircaali-Iftar (2002) had peersibling tutors implement both constant timedelay and simultaneous prompting to teachchildren with mental retardation to expres-sively name animals. The authors found thatthe tutors implemented both procedures reli-ably.

Constant time delay has been the subject ofnumerous studies and has been shown to beeffective, easy to use, and more efficient thanother near errorless strategies such as the sys-tem of least prompts (Schuster et al., 1998).Researchers have taught peer tutors to imple-ment constant time delay with students withdisabilities. Peer tutors without disabilitieshave successfully used constant time delay toteach basic sight word reading to high schoolstudents with moderate disabilities (Miracle etal., 2001), generalized reading of cooking la-bels to high schoolers with mental disabilities(Collins, Branson, & Hall, 1995), and sightword identification to elementary-aged stu-dents with severe disabilities (Wolery, Werts,Snyder, & Caldwell, 1994). All of these studies


show that peer tutors can reliably implementconstant time delay procedures with discretetasks.

The research studies described have sup-ported the effectiveness of peer tutors as reli-able implementers of systematic instruction.While few studies used peer tutors to teachchained tasks and several studies used con-stant time delay, no study could be found thathas examined the effectiveness of peer tutorsimplementing constant time delay to teachchained tasks. This investigation was designedto answer the following research questions (1)Is a triadic instructional arrangement with in-structor modeling and role-play effective inteaching high school-age peer tutors to imple-ment the constant time delay procedure? (2)Will high school-age peer tutors without dis-abilities reliably implement the constant timedelay procedure in teaching chained tasksacross participants and over time? and (3) Willa peer tutor implemented constant time delayprocedure be effective in teaching highschool students with moderate and severe dis-abilities the acquisition and maintenance ofchained cooking tasks?

Method

Participants

Students. Four students (all male), rangingin age from 15 to 20 years, enrolled in a publichigh school participated. All attended at leastone general education class, lunch, and as-semblies with students who did not have dis-abilities.

Jake, a 16-year-old student who on the Wech-sler Intelligence Scale for Children – WISC – III(Wechsler, 1991), obtained an IQ of 40, wasdiagnosed with moderate mental retardation.Jake had a sight word vocabulary of 100-125words, could read and respond to many envi-ronmental signs, proper nouns, and wordsfrom computer pull-down menus, He was un-able to read simple directions. Jake wrote andverbalized his personal information (e.g.,name, guardian’s name, etc.). He preparedsimple snacks (e.g., cold cereal, microwavepopcorn). He attended general educationclass for weightlifting and attended a socialskills class and math class for students withmild mental disabilities. He went out of the

building twice a week for vocational trainingat a local grocery store. Jake’s IndividualizedEducation Plan (IEP) included objectives forusing tools and equipment in a safe manner,following pictorial directions, behavior self-management, employability skills, and func-tional mathematics.

Louis, a 20-year-old student who obtainedan IQ of 40 on the Wechsler Adult IntelligenceScale (Wechsler, 1997), was diagnosed withmoderate mental retardation. He was in hisfifth year of enrollment at the high school andattended general education classes in lifetimesports physical education, art, and food ser-vices technology. Louis had a sight wordvocabulary of 150-200 words. He read andresponded appropriately to many environ-mental signs, proper nouns, and commonlyused words drawn from vocational experi-ences, and could prepare several simple foodsfrom memory (such a cold cereal and micro-wave popcorn). He did need supervision tomonitor proportions used (e.g., overflowingthe cereal bowl with milk, etc.). He receivedvocational training in the community at thelibrary and at a local community serviceagency 2-3 times a week. Louis was unable tocomprehend written directions, and his IEPobjectives were in the areas of food prepara-tion, employability skills, and purchasing,

Charlie, a 15-year-old male with an IQ of 40on the WISC – III (Wechsler, 1991), was diag-nosed with moderate mental retardation,Down syndrome, and a severe speech deficit.He attended a general education class in phys-ical education. Charlie went out of the build-ing two times per week for community-basedinstruction in safety and shopping instruction.Charlie had a sight vocabulary of 25-30 words.He read and responded appropriately tomany environmental signs, product names,and names of family members, teachers, andfellow students. His speech often was difficultto understand, and during the time of thestudy the teacher and speech pathologist weretrying several voice-output devices to supple-ment speech. Charlie could follow verbal/model directions for making several simplesnacks such as cold cereal and crackers withspreadables. Charlie was unable to read sim-ple written directions, and his IEP includedobjectives in the areas of food preparation,following pictured directions, and self-moni-


toring of behavior. He received speech ther-apy at school 2 times per week.

Jonah, a 17-year-old male with a KaufmanBrief Intelligence Test (K-BIT) (Kaufman & Kauf-man, 1990) composite score of 40, was diag-nosed with moderate mental retardation, amoderate hearing impairment in one ear, anda severe visual impairment in the opposite eye.He attended general education classes inphysical education and food services technol-ogy, and also was out of the building 1-2 timesper week for community-based instruction insafety and shopping skills. Jonah had a sightword vocabulary of 25-50 words. He read andresponded appropriately to many environ-mental signs, product names, and names offamily, teachers, and fellow students. Hisspeech often was difficult to understand, andat the time of the study was in the process ofobtaining a dynamic-screen, voice-output aug-mentative communication device to supple-ment his speech. Jonah followed verbal/model directions for making several simplesnacks such as cold cereal and crackers withspreadables. He followed 1-2 step instructionswithout redirection if given adequate process-ing time. Jonah was unable to read simplewritten directions, and his IEP included ob-jectives in the areas of following pictureddirections, communication, and self-care.He received weekly speech therapy and oc-cupational therapy at school and services forvision impairment on a consultative basis.

Peer tutors. Eleven peer tutors, 2 males and9 females, ranging in age from 16 to 18 years(high school juniors and seniors), and en-rolled in a peer tutoring class for high schoolcredit were tutors during the study. Five of thepeer tutors were enrolled in advanced levelclasses and were on a pre-college curriculum.The other seven were enrolled in standardlevel classes and were on a vocational educa-tion track for either industry or business re-lated fields. All tutors who participated metprerequisite skills including availability for af-ter school training sessions, faculty perfor-mance ratings, and interest in the researchstudy. Faculty performance rating forms weredistributed to faculty members and they ratedstudents’ personal characteristics and workethic on a scale of 1 to 5 (5 was the highestrating). Students must have received an over-

all 3.5 from at least five faculty members inorder to be eligible for participation.

Setting

All experimental sessions occurred in a kitch-en/living area adjacent to the students’ self-contained classroom. Because the kitchenarea was a separate room with a door thatcould be closed and because only one studentat a time was permitted to be in the kitchen toprepare a snack, additional control for distrac-tions did not occur.

Materials/Equipment

The teacher and/or peer tutors used the fol-lowing materials: (a) digital camera for creat-ing the pictorial recipes, (b) dishes and uten-sils, (c) food stored in cabinets, a refrigeratoror freezer, (d) pictorial recipe cards with writ-ten directions beneath each photo with pho-tos in sequence according to the task analysisand held in order by a large metal ring, (e)data sheets, and (f) reliability forms for thedependent and independent measures.

Skill Selection

All participating students had food prepara-tion or the following of pictorial directions asobjectives included on their IEP. In addition,the teacher interviewed the students’ parents,former teachers, and the students about thefoods they would prefer to cook. After deter-mining that the students had no prior experi-ence in preparing the target recipes, theteacher screened the students for their abilityto follow a verbal/model prompt and to mo-torically imitate specific skills required to pre-pare the target recipes. Target recipes for Jakeand Louis included making a milkshake in theblender (27 critical steps) and making agrilled cheese sandwich (32 critical steps).Target skills for Charlie and Jonah includedmaking a waffle in the toaster (27 criticalsteps) and making juice from a frozen concen-trate (25 critical steps). In addition, each taskanalysis included a “turning the page” step inbetween each critical step for students to turnthe pages of the picture recipe in order to seethe next step.


General Procedures

The dependent variable was the percent ofcorrect steps completed independently on thechained food preparation tasks. Peer tutorsimplementing the constant time delay proce-dure was the independent variable. Initially,peer tutor training occurred followed by skillselection of the target recipes. Then singleopportunity baseline sessions occurred beforetraining started (Schuster, Gast, Wolery, &Guiltinan, 1988). After the teacher collectedstable baseline data, Jake began training onmaking a milkshake. When he reached or ap-proached criterion, Jake started training onmaking a grilled cheese sandwich at the sametime that Louis began training on making amilkshake. When Jake and Louis reached orapproached criterion, Louis started trainingon preparing a grilled cheese sandwich whileCharlie began training on toasting a waffle.When Charlie reached or approached crite-rion on making a waffle, he began training onmaking juice while Jonah began training onmaking a waffle. Then Jonah was trained tomake juice. Intermittent maintenance probesessions occurred on the average of once every15 sessions for acquired tasks.

Peer Tutor Training

Prior to implementing the constant time delayprocedure, the teacher trained the peer tutorsthrough direct instruction, modeling, and tri-adic role-play with peer tutors acting the inthe roles of data collector, prompter, and stu-dent (trainee). The teacher trained the peertutors during two 90-minute after school ses-sions. During training sessions, the teachertaught the peer tutors how to implement theconstant time delay procedure and to recordstudent responses. At the conclusion of thetraining sessions each peer tutor had to dem-onstrate performance of data collection andimplementation of constant time delay stepswith at least 90% accuracy to participate. Inaddition to performance criteria, the teacherrequired the peer tutors to correctly answer atleast 90% of questions on a written exam be-fore they were allowed to participate in imple-menting the constant time delay procedurewith the students. Twelve students initiallycompleted training; however, one tutor failed

to meet the criteria for inclusion. The 11 peertutors randomly served as both prompter anddata collector throughout the study andworked with a variety of students (i.e., peertutors were not relegated to one role andassigned to one student).

Baseline Sessions

The teacher conducted a single opportunitybaseline session on all eight skills (two skillsper student) at the beginning of the study.Intermittent probe sessions occurred at leastonce every five sessions on untrained tasks forJake and Louis and every 10 sessions for Char-lie and Jonah to help control for repeatedtesting and maturation effects. Three consec-utive baseline sessions occurred immediatelyprior to intervention on each task. Theteacher used single opportunity probe ses-sions to help control for repeated testing andmaturation effects as well as to reduce thecosts associated with duplicate materials.

The teacher conducted all baseline sessions.During these sessions, the teacher gave stu-dents an attentional cue, a task request tomake the food, and waited for a student’sresponse. If the student initiated a correctresponse within 5 s and completed the re-sponse within 20 s, the teacher recorded acorrect response, provided verbal praise, andwaited 5 s for the student to initiate the nextstep in the task analysis. The sessions contin-ued until a student committed an error (i.e.,did not initiate a response within 5 s, or initi-ated a response within 5 s but completed thewrong step, completed the step incorrectly, ortook too long to complete the step). If thisoccurred, the student was praised for attempt-ing the task and the session was terminated. Inaddition, at the end of the session, studentswere allowed to eat a snack (previously pre-pared by the teacher or student).

Training Procedures

After stable baseline data, training occurred.One peer tutor acted as the prompter and theother peer tutor as the data collector. Foreach task for Jake, Louis, and Charlie, two 0-ssessions occurred. For Jonah, three 0-s ses-sions occurred. Thereafter, all sessions used a5-s constant time delay procedure.


During 0-s sessions, the prompter gave anattentional cue, stated a task request (e.g.,“Make ______”), immediately provided thecontrolling prompt (i.e., a verbal/modelprompt of the step to be completed), andwaited 5 s for a student response. If a studentinitiated a response, the prompter allowed thestudent 20 s to complete the step. After theprompter provided consequences for eachstep the tutor waited 5 s for the student toinitiate the next step in the task analysis.

During 5-s sessions, the prompter gave anattentional cue, stated a task request (e.g.,“Make the ______”), and waited 5 s for a stu-dent response. If no response occurred after5 s, the prompter provided the controllingprompt and waited 5 s for the student to im-itate the prompt. If the student initiated aresponse within 5 s, the prompter allowed thestudent 20 s to complete the response. Afterthe prompter provided the consequences, thetutor waited 5 s for the student to initiate thenext step in the task analysis.

The data collector recorded five possibleresponses during training sessions. The datacollector recorded a correct response beforethe prompt (i.e., correct anticipation) when astudent initiated a response before the con-trolling prompt and completed the stepwithin 20 s. All correct anticipations resultedin specific verbal praise. The data collectorrecorded an incorrect response before theprompt (i.e., non-wait error) when a studentinitiated a response before the prompt butcompleted the step (a) out of sequence (se-quence error), (b) incorrectly (topographicalerror), or (c) with too long of a duration(duration error). These errors resulted in theprompter interrupting the student, remind-ing the student to wait, and delivering thecontrolling prompt. When a student initiateda response within 5 s after the delivery of thecontrolling prompt and completed the stepwithin 20 s, the data collector recorded a cor-rect after the prompt (i.e., correct wait).These responses resulted in the trainer deliv-ering verbal praise. The data collector re-corded an incorrect after the prompt (i.e.,wait error) when a student initiated a re-sponse within 5 s of the prompter deliveringthe controlling prompt and completed thestep (a) out of sequence (sequence error), (b)incorrectly (topographical error), or (c) with

too long of a duration (duration error). Whenthese errors occurred, the prompter redeliv-ered the controlling prompt and waited for astudent response. If the student still did notcomplete a correct response, the prompternotified the teacher. When a student did notinitiate a response within 5 s of the controllingprompt the data collector recorded a no re-sponse, and the prompter delivered an addi-tional attentional cue and prompt while com-pleting the step for the student.

At the end of each session, the student atethe prepared food. Training sessions contin-ued until each student reached three sessionsat 100% correct anticipations.

Maintenance

After criterion was reached on each task, theteacher conducted maintenance sessions atleast once every 15 sessions until all studentsreached criterion on all cooking tasks. Thetrainer conducted these sessions like 5-s delaysessions.

Experimental Design

We used a multiple probe across subjects andbehaviors design to demonstrate experimen-tal control. Experimental control was demon-strated when a change occurred in the depen-dent variable when, and only when, theindependent variable was applied and thischange occurred across tiers in a time-laggedmanner.

Reliability

Peer tutor training. A general educationteacher collected procedural reliability dataon the teacher’s implementation of the peertutor training procedure during 100% of thetraining sessions. The teacher calculated thesedata by dividing the number of trainer behav-iors observed by the number of trainer behav-ior planned and multiplying by 100 (Billings-ley, White, & Munson, 1980). Some of thetrainer behaviors measured included provid-ing definitions and examples of key terms,passing out materials, demonstrating the con-stant time delay procedure, having studentsrole play, providing written examinations, andproviding feedback.


Skill training. The teacher and one peertutor collected both dependent and indepen-dent variable reliability data. They collectedthese data during 25% of all baseline sessionsand 26% of all intervention sessions. Thepoint-by-point method was used to calculatedependent variable reliability data (i.e., divid-ing the number of agreements by the numberof agreements and disagreements and multi-plying by 100). Procedural reliability data (i.e.,independent variable reliability data) werecalculated according to formula cited by Bill-ingsley et al., 1980. Some of the behaviorsmeasured, when appropriate, included deliv-ering the attentional cue, stating the task re-quest, waiting the appropriate delay interval,providing the controlling prompt, waiting theresponse interval, and providing correct con-sequences.

Results

Reliability Data

Reliability data of the instructor training,modeling and triadic role-play procedure fortraining the peers tutors indicated 100% ac-curacy during both peer tutoring training ses-sions.

During baseline sessions, student responsereliability data (i.e., dependent variable) re-sulted in a 100% agreement. During interven-tion sessions when student data collectors re-corded student responses, student responsereliability resulted in a mean agreement of98% (range � 94-100%).

During baseline sessions conducted by theteacher, procedural reliability data resulted inan overall mean accuracy of 99.5% (range �98-100%) in following the required interven-tion behaviors. Procedural reliability data dur-ing intervention sessions indicated that theprompters followed the planned behaviorswith an overall mean accuracy of 93.1%(range � 84-99%).

Student Acquisition Data

Student acquisition data are displayed in Ta-ble 1 and Figures 1 and 2. Data are graphedusing separate symbols to represent total tasksteps (the steps of turning the page betweeneach picture card are included) with open

squares and critical steps (turning page stepsomitted) by open triangles. All studentsachieved criterion on all tasks. Overall, stu-dents required a total of 69 instructional ses-sions to reach criterion on all tasks. Individualstudents required between 6 and 12 sessionsper task to reach criterion (an average of 8.6sessions). Across all tasks, students committeda total of 40 errors through criterion. Individ-ually, students committed between 3 and 11errors per task for an overall error percentageof 1.04%. Error percentage was calculated bydividing the number of errors committed bythe product of the total number of interven-tion sessions per skill by the total number ofsteps in each task analysis each student per-formed (Schuster et al., 1988). Of the 40 totalerrors, 39 (97.5%) were non-wait errors (i.e.,errors before the prompt). The remaining er-ror was a wait error (i.e., an error after theprompt) committed by Jake when acquiringthe grilled cheese task.

Student Maintenance Data

Maintenance data indicate that Jake and Char-lie maintained all tasks with 100% accuracythroughout the length of the study. Louis’performance decreased to 94% of total tasksteps and 96% of critical steps during thesecond maintenance probe session but re-turned to 100% accuracy during the finalmaintenance probe session. The teacher didnot collect maintenance data on Jonah due tothe end of the school year.

TABLE 1

Number of Sessions Through Criterion

Student SkillNumber of SessionsThrough Criterion

Jake Milkshake 9Grilled Cheese 7

Louis Milkshake 12Grilled Cheese 9

Charlie Waffle 10Juice 6

Jonah Waffle 9Juice 7

Total 69


Discussion

The first research question this study soughtto answer concerned the effectiveness of apeer tutor-training program consisting of in-structor modeling and triadic role-play. Highprocedural reliability data for these trainingsessions occurred. Eleven of the 12 peer tutorsselected to participate in this study met boththe written and performance criteria withinthe allotted time. Both dependent and inde-pendent variable reliability data collected dur-ing intervention sessions supports the effec-

tiveness of the peer tutor training procedure.The fact that peer tutors were counterbal-anced across students and tasks suggests thatthe peer tutor training procedure was effec-tive in programming for peer tutor generali-zation across different students and differenttasks within the cooking curricular area. Thesuccess of this program contributes to thepeer tutoring literature through replicating,in part, the peer tutor training proceduresused by Collins et al. (1995). However, whilethe Collins et al. study addressed peer tutoring

Figure 1. Percent of correct responses for Jake and Louis during baseline, intervention, and maintenancesessions.


with discrete tasks, results of this study extendthe expectations of peer tutors’ abilities tochained task instruction, increasing their ver-satility in the classroom.

The second research question addressedthe degree of reliability with which peer tutorscould implement constant time delay in teach-ing chained tasks. Dependent variable reliabil-ity data indicate that overall peer tutors col-lected student response data with 98%

agreement (range � 94-100%) with the reli-ability observer. Peer tutors as a whole wereable to implement the response promptingstrategy with 93% accuracy (range � 84-99%).Wolery, Bailey, and Sugai (1988) state reliabil-ity above 90% is desirable, above 80% is ac-ceptable. Nine out of the 11 peer tutors per-formed the steps of the constant time delayprocedure with 90% or better accuracy.

Results of the peer tutoring reliability data

Figure 2. Percent of correct responses for Charlie and Jonah during baseline, intervention, and maintenancesessions.


in this study contribute to the peer tutoringliterature due to the similarity to studies uti-lizing teacher-implemented constant time de-lay. Numerous studies (e.g., Braham, Collins,Schuster, & Kleinert, 1999; Fiscus, Schuster,Morse, & Collins, 2002; Graves, Collins, Schus-ter, & Kleinert, 2005; Griffen, Wolery, &Schuster, 1992; Hall, Schuster, Wolery, Gast, &Doyle, 1992; Miller & Test, 1989; Miracle etal., 2001; Schuster et al., 1988) using constanttime delay with chained tasks have shown thatteacher-delivered instruction with this proce-dure has resulted in similar levels of accept-able and desirable procedural reliability data.

There were some interesting findings whenusing peer tutors to deliver the instruction.First, most of the errors made by peer tutorsinvolved not providing descriptive verbalpraise after each correct response. Whengiven feedback about this omission, manyquestioned the need for a continuous rein-forcement schedule. The peer tutors com-plained that it “felt unnatural” to praise stu-dents for every step. This could be ofsignificant interest in future research, consid-ering that failure to provide continuous verbalpraise did not appear to adversely affect stu-dent acquisition. Second, having peer tutorswork in pairs as prompter and data collectorresulted some disagreement at times. For ex-ample, peer tutors would disagree on theamount of butter that was spread on the breadand therefore whether the response was cor-rect or not. Other similar differences of opin-ion also occurred that required the teacher tointervene.

The third research question addressed theeffectiveness of peer-tutor implemented con-stant time delay on the acquisition and main-tenance of chained cooking tasks by studentswith moderate and severe disabilities. Dataindicate that students only acquired the skillswhen the time delay procedure was imple-mented. Intervention data indicate that stu-dents reached criterion quickly with no over-lap between baseline and interventionconditions. A low occurrence of errors withnoted absences of wait and no response errorsmay support the peer tutors’ effectiveness inproviding clear, consistent, verbal/modelprompts. Maintenance data show that stu-dents retained the skills up to 22 sessions aftermeeting criterion. The noticeable lack of vari-

ability in student response data supports theinterpretation that student acquisition was notadversely affected by the interchanging ofpeer tutor roles as data collector andprompter and across students.

Several other findings about the interven-tion are noteworthy. First, each student, therewas a slight increase in baseline data percent-ages in the second skill after the first zerosecond session occurred in the first skill. Thiswas most likely due to the fact that the first fewsteps of the pictorial recipes were identical(e.g., select recipe book, wash hands, etc.). Ineach of these instances, the baseline data be-came stable prior to intervention. Second, itshould be noted that each student learnedtheir second task in fewer sessions than theirfirst. This could be the result of (a) the secondtask being easier than the first (though this isdoubtful given that two different sets of skillswere targeted), or (b) students were “learningto learn” a picture recipe with constant timedelay. Third, the overall error percentage inthis study (i.e., 1.04%) is lower than reportedin the constant time delay chained task litera-ture (Schuster et al., 1998). This is especiallynoteworthy given that peer tutors deliveredthe instruction.

When teaching tasks that contain consum-ables, costs should be considered. Groceryitems for this study amounted to $74.03. Stu-dents purchased groceries during community-based instruction trips using funds from anannual community-based budget provided bythe school district. The district supplied thecamera, disks, and ink cartridges as each spe-cial education classroom in the district re-ceived these materials to assist in developingstudent alternate portfolios and instructionalmaterials. The district also provided the com-puter, printer, paper, and laminating film.The teacher spent $84.00 of her personalfunds on various fast food gift certificate asincentives for peer tutors to remain as partic-ipants throughout the study.

Results of this study are significant for class-room teachers in that they demonstrate thevalue and capability of peer tutors in deliver-ing quality one-on-one instruction to studentswith moderate and severe disabilities. Becausestudents with these disabilities often areplaced in the same classroom, time and per-sonnel to meet the individual needs of such a


heterogeneous group of students is difficult. Ifpeer tutors can provide the same or similarquality of instruction as classroom teachersand instructional assistants for particulartasks, classroom staff may take on more ofsupervisory role, while students receive moretime for one-on-one instruction. Freeing uptime for classroom teachers may allow them tomore closely monitor the effectiveness of in-structional procedures and to modify instruc-tion in a timelier manner. Furthermore, peertutor instruction may be preferable to stu-dents with disabilities because it may be per-ceived a more enjoyable and less stigmatizingthan instruction from an adult, especially if itis paired with opportunities to develop socialrelationships. If the teacher promotes devel-opment of social relationships and empha-sizes the role of peer tutors as same-age learn-ing partners rather than “mini teachers”,students may perceive the instruction frompeer tutors favorably. It may be helpful forfuture research to address such social validityissues.

This study contributes to the literature inseveral ways. First, it contributes to the timedelay literature because no study has ad-dressed teaching chained tasks through theuse of peer tutors. Second, it contributes tothe peer tutoring literature because we couldfind no study that addressed peer tutors teach-ing chained tasks with a systematic responseprompting strategy with high school students.However, the study was limited in externalvalidity because of the small number of partic-ipants. It also may be difficult to replicate thestudy in settings where peer tutors are notgraded for performance (as in this studywhere peers were enrolled in a credit gener-ating class) or in settings where peer tutorscannot be trained at the same time duringlarge blocks of time. Further research shouldtrain peer tutors to use other systematic in-structional procedures with chained tasks,train students to perform chained tasks fromother instructional domains, or to use re-sponse prompting strategies in communitybased settings. In future studies it may be wiseto include generalization measures for bothpeer tutors and students with respect to usingother materials in other settings. Future stud-ies also may address whether peer tutors couldreliably code student errors (i.e., duration,

sequence, or topographical) or implementthe procedure alone, acting as both data col-lector and prompter.

References

Billingsley, F. F., White, O. R., & Munson, R. (1980).Procedural reliability: A rationale and an exam-ple. Behavioral Assessment, 2, 229–241.

Braham, R. S., Collins, B. C., Schuster, J. W., &Kleinert, H. (1999). Teaching community skills tostudents with moderate disabilities: Comparingcombined techniques of classroom simulation,videotape modeling, and community-based in-struction. Education and Training in Mental Retar-dation and Developmental Disabilities, 34, 170–181.

Collins, B. C., Branson, T. A., & Hall, M. G. (1995).Teaching generalized reading of cooking productlabels to adolescents with mental disabilitiesthrough the use of key words taught by peertutors. Education and Training in Mental Retarda-tion and Developmental Disabilities, 30, 65–75.

Fiscus, R. S., Schuster, J. W., Morse, T. E., & Collins,B. C. (2002). Teaching elementary students withcognitive disabilities food preparation skills whileembedding instructive feedback in the promptand consequent event. Education and Training inMental Retardation and Developmental Disabilities,37, 55–69.

Graves, T. B., Collins, B. C., Schuster, J. W., & Klein-ert, K. (2005). Using video prompting to teachcooking skills to secondary students with moder-ate disabilities. Education and Training in Develop-mental Disabilities, 40, 34–46.

Griffen, A. K., Wolery, M., & Schuster, J. W. (1992).Triadic instruction of chained for preparationresponses: Acquisition and observational learn-ing. Journal of Applied Behavior Analysis, 25, 193–204.

Hall, M. G., Schuster, J. W., Wolery, M., Gast, D. L.,& Doyle, P. M. (1992). Teaching chained tasks ina divided half format. Journal of Behavioral Educa-tion, 3, 257–280.

Kamps, D., Locke, P., Delquadri, J., & Hall, R. V.(1989). Increasing academic skills of studentswith autism using fifth grade peers as tutors. Ed-ucation and Treatment of Children, 12, 38–51.

Kaufman, A., & Kaufman, N. (1990). Kaufman briefintelligence test. Circle Pines, MN: American Guid-ance Service.

Koury, M., & Browder, D. M. (1986). The use ofdelay to teach sight words to peer tutors classifiedas moderately mentally retarded. Education andTraining of the Mentally Retarded, 21, 252–258.

Miller, U. C., & Test, D. W. (1989). A comparison ofconstant time delay and most-to-least promptingin teaching laundry skills to students with moder-


ate retardation. Education and Training in MentalRetardation, 24, 363–370.

Miracle, S. A., Collins, B. C., Schuster, J. W., &Grisham-Brown, J. (2001). Peer- versus teacher-delivered instruction: Effects on acquisition andmaintenance. Education and Training in MentalRetardation and Developmental Disabilities, 36, 373–385.

Romer, L. T., Busse, D. G., Fewell, R. R., & Vadasy,P. F. (1985). The relative effectiveness of specialeducation teachers and peer tutors. Education ofthe Visually Handicapped, 17, 99–115.

Schuster, J. W., Gast, D. L., Wolery, M., & Guiltinan,S. (1988). The effectiveness of a constant timedelay procedure to teach chained responses toadolescents with mental retardation. Journal ofApplied Behavior Analysis, 21, 169–178.

Schuster, J. W., Morse, T. E., Ault, M. J., Doyle,P. M., Wolery, M., & Crawford, M. (1998). Con-stant time delay with chained tasks: A review ofthe literature. Education and Treatment of Children,21, 74–106.

Tekin-Iftar, E. (2003). Effectiveness of peer deliv-ered simultaneous prompting on teaching com-munity signs to students with developmental dis-abilities. Education and Training in DevelopmentalDisabilities, 38, 77–94.

Tekin, E., & Kircaali-Iftar, G. (2002). Comparison ofthe effectiveness and efficiency of two responseprompting procedures delivered by sibling tutors.

Education and Training in Mental Retardation andDevelopmental Disabilities, 37, 283–299.

Wechsler, D. (1991). Wechsler intelligence scale for chil-dren – third edition. San Antonio, TX: The Psycho-logical Corporation.

Wechsler, D. (1997). Wechsler adult intelligence scale –third edition. San Antonio, TX: The PsychologicalCorporation.

Werts, M. G., Caldwell, N. K., & Wolery, M. (1996).Peer modeling of response chains: Observationallearning by students with disabilities. Journal ofApplied Behavior Analysis, 29, 53–66.

Wolery, M., Ault, M. J., Doyle, P. M. (1992). Teachingstudents with moderate and severe disabilities: Use ofresponse prompting strategies. White Plains, NY:Longman.

Wolery, M., Bailey, D. B., & Sugai, G. M. (1988).Effective teaching: Principles and procedures of appliedbehavior analysis with exceptional children. Boston,MA: Allyn and Bacon, Inc.

Wolery, M., Werts, M. G., Snyder, E. D., & Caldwell,N. K. (1994). Efficacy of constant time delay im-plemented by peer tutors in general educationclassrooms. Journal of Behavioral Education, 4, 415–436.

Received: 28 July 2006Initial Acceptance: 27 September 2006Final Acceptance: 7 January 2007


States’ Eligibility Guidelines for Mental Retardation: AnUpdate and Consideration of Part Scores and

Unreliability of IQs

Renee Bergeron, Randy G. Floyd, and Elizabeth I. ShandsThe University of Memphis

Abstract: Mental retardation (MR) has traditionally been defined as a disorder in intellectual andadaptive functioning beginning in the developmental period. Guided by a federal definition of MRdescribed in the Individuals with Disabilities Education Act, it is the responsibility of each of the UnitedStates to describe eligibility guidelines for special education services. The purpose of this study was toexamine eligibility guidelines for MR for the 50 states and the District of Columbia. This study examinedthe terms used to describe MR, the use of classification levels, the cutoff scores, and the adaptive behaviorconsiderations for each state. In addition, this study examined guidelines for consideration of intelligencetest part scores and consideration of the unreliability of IQs through consideration of the standard errorof measurement (SEM) or an IQ range. As found in previous studies, results revealed great variation inthe specific eligibility guidelines for MR from state to state. The greatest variation appeared to be across theadaptive behavior considerations. Approximately 20% of states (10) recommend consideration of intelli-gence test part scores, and approximately 39% of states (20) recommend attention to unreliability of IQsthrough consideration of the SEM or an IQ range.

Individuals with mental retardation (MR)have been described in literature and histori-cal documents for many centuries. Sincearound 1900, definitions of MR in the UnitedStates have included three general aspects:deficits in intellectual functioning, impairedfunctioning in the daily environment, and on-set during the developmental period (Sheer-enberger, 1983). Although these three criteriahave been included in nearly all recent defi-nitions of MR proposed by professional orga-nizations (e.g., American Association on Men-tal Retardation [AAMR], 2002; AmericanPsychiatric Association [APA], 2000), the spe-cific criteria within each domain have beenmore variable across organizations and overtime.

Deficits in intellectual functioning are usu-ally defined by poor performance on norm-referenced intelligence tests via IQs. IQs are

often considered reflections of general intel-ligence, which represents intelligence as a sin-gle, global factor (Jensen, 1998). Research hasestablished the predictive validity of IQs onvarious outcomes, such as academic achieve-ment and adaptation to environmental de-mands (Brody, 1997; Neisser et al., 1996). Theuse of IQs to determine deficient intellectualfunctioning has been included in most defini-tions of MR across professional groups sincethe American Association on Mental Retarda-tion (AAMR) first specified the use of intelli-gence tests in its 1959 definition of MR. Mostcurrent definitions adopted by professionalorganizations set the upper IQ cutoff for MRat two or more standard deviations below thepopulation mean (i.e., IQs of 70 or below;AAMR, 2002; APA, 2000). Moreover, becausesome degree of measurement error is inher-ent in obtained IQs, many professional orga-nizations now include an IQ range (e.g., IQsbelow 70 to 75) or specify that the standarderror of measurement (SEM) be consideredrather than a strict IQ cutoff criterion.

Whereas IQs have long been used to satisfythe intellectual deficit criterion for MR, there

Correspondence concerning this article shouldbe addressed to Randy G. Floyd, The University ofMemphis, Department of Psychology, Memphis, TN38152. Email: [email protected]


Eligibility Guidelines / 123

has been less agreement across professionalorganizations as to how adaptive behavior def-icits should be evidenced. Adaptive behaviorgenerally refers to the capacity to meet one’sdaily functional needs based on the individu-al’s age and the culture in which the individ-ual lives. To determine deficient adaptivefunctioning, some definitions specify the useof global adaptive behavior composite scores,others specify the use of scores reflectingadaptive domains (e.g., conceptual, social, andpractical; AAMR, 2002), and others specify theuse of scores reflecting adaptive behavior skillareas (e.g., self-care and community participa-tion; APA, 2000). However, some definitionsdo not define requirements for deficientadaptive functioning. To best evaluate an in-dividual’s adaptive functioning from an eco-logical perspective, most professional organi-zations specify that the individual’s adaptivefunctioning be assessed in multiple settings toascertain the pervasiveness of deficient func-tioning.

Mental Retardation in the Educational Setting

The Individuals with Disabilities Act (IDEA;P.L. 108-446, 2004) guarantees a free and ap-propriate public education to all students.Consistent with the educational focus of im-pact on academic functioning, IDEA definesMR as “significantly subaverage general intel-lectual functioning, existing concurrently withdeficits in adaptive behavior and manifestedduring the developmental period, that ad-versely affects a child’s educational perfor-mance” (National Archives and Records Ad-ministration, 2005, p. 35836). Although thefederal definition provides a general frame-work for determining eligibility for special ed-ucation services under the MR category, statesare permitted to refine eligibility guidelinesand to operationalize deficient functioning.

Purpose of the Study

Previous studies examining differences acrossstates’ eligibility criteria for MR under IDEAhave found significant variations in terms(e.g., mental retardation and mental disability),use of classification levels (e.g., mild, moderate,and severe), IQ cutoff scores, and adaptive be-

havior consideration and specified criterion(Denning, Chamberlain, & Polloway, 2000;Frakenberger, 1984; Frakenberger & Fronza-glio, 1991; Utley, Lowitzer, & Baumeister,1987). This study was designed to provide anupdated examination of states’ guidelines andto investigate two issues that have not receivedmuch attention in previous surveys.

Despite the historical reliance on IQs in MRdefinitions, researchers and advocates oftenassert that the over-reliance on a single scoreignores the complexity of abilities (Daniel,1997; Horn & Noll, 1997). Research and the-ories focused on cognitive abilities indicatethat important information about an individ-ual’s specific cognitive abilities may be over-looked if the focus is on only a single score, anIQ. Based on this reasoning, test authors andpublishers have drawn increasingly on theo-ries describing specific cognitive abilities dur-ing test development to develop composite-based part scores representing these abilities(Alfonso, Flanagan, & Radwan, 2005). Recentguidelines for the assessment and diagnosis ofMR have also placed greater emphasis on partscores. For example, the Social Security Ad-ministration (SSA) released new guidelines in2002 for disability determination for MR thatallow certain part scores to be used in place ofthe IQ in the diagnosis of MR when there isreason to doubt the validity of the IQ (Na-tional Research Council [NRC], 2002). Withthe increasing emphasis on theory-based testdevelopment and interpretation, as well as theintroduction of the SSA guidelines for MReligibility that allow use of part scores in cer-tain situations, this study investigates the prev-alence of intelligence test part score consider-ation.

It is well known that some degree of mea-surement error is inherent in obtained IQs,and many professional organizations specifythat a range of scores (e.g., 70 to 75) or theSEM be considered in determining deficientintellectual functioning. However, the federaldefinition of MR under IDEA provides noindication that such measurement error beconsidered in determining eligibility for MR.Therefore, this study examines the use ofscore ranges or the SEM in determining defi-cient intellectual functioning across states.


Method

Procedure

The first author developed a spreadsheet torecord data from states’ eligibility guidelinesbased on a review of similar studies (e.g., Den-ning et al., 2000; Utley et al., 1987). Itemsincluded the following: state, year guidelineswere last published, term used, whether levelsof MR were specified, the IQ threshold, thepractices for identifying adaptive behavior def-icits, and whether adaptive behavior require-ments should be evident across multiple set-tings. In addition, two items were added todetermine whether states included consider-ation of intelligence test part scores and con-sideration of either the SEM or a score rangearound IQs.

Through an iterative process, the first andthird authors initially obtained eligibilityguidelines from the Department of Educationwebsites for the 50 states and the District ofColumbia. (Hereafter, for the sake of simplic-ity, the District of Columbia is referred to as astate.) Next, all state Departments of Educa-tion were contacted via telephone, e-mail, orboth to confirm that the guidelines posted ontheir websites were currently used for estab-lishing eligibility for MR. Guidelines were con-firmed for 48 states during the period fromSeptember to December 2005, and all guide-lines (including California, Maine, and Texas)were confirmed by May 2006. Data from theguidelines were then entered into the spread-sheet by the first author. To ensure accuracyin data recording, the third author first re-viewed guidelines from a random selection of20 states (39%) and independently coded thedata. Across all items, there was 97.4% agree-ment. Although these estimates of inter-rateragreement indicate a high level of consistencyin coding, most of the disagreements betweencoders occurred with the three items devotedto adaptive behavior assessment. For thesethree items, there was only 95.0% percentagreement. After criteria for these items werefurther developed, the first author againcoded data for these three items for all 51states, and the third author independentlycoded another random selection of 20 states.Percentage agreement across these threeitems was 98.3%. Across both rounds of cod-

ing, disagreements were evaluated and re-solved by consensus.

Results

Table 1 summarizes the eligibility criteria forMR for the 51 states that were obtained byreviewing the official documents includingthe states’ guidelines. Overall, 53% of statesuse the term mental retardation to describe thecondition, 12% use the term mental disability,and 6% use the term intellectual disability.Other terms used by two or fewer states in-clude cognitive delay, cognitive disability, cognitiveimpairment, cognitively impaired, developmentalcognitive disability, intellectual impairment, learn-ing impairment/delay in learning, mental handi-cap, mentally disabled, mentally handicapped, andsignificant limited intellectual capacity. Of the 51states, 18 differentiated MR according to levelof impairment or degree of severity based onIQs. Most states used the terms mild, moderate,and severe/profound, and three used the termseducable MR, trainable MR, and severe/profound.

Intellectual Deficit Criterion

To satisfy the intellectual deficit criterion, themajority of states (59%) use an IQ cutoff of atleast two SDs below the normative mean (orstandard scores of 70 or below). Approxi-mately 6% of states require an IQs to be belowtwo SDs (or standard scores below 70), andone state uses an IQ cutoff of at least one anda half SDs below the normative mean (orscores approximately 78 and below). How-ever, 22% of states’ guidelines contained onlythe federal definition of MR without specificeligibility criteria and one state (Iowa) uses anoncategorical approach and does not pro-vide eligibility criteria specific to MR. As notedin Table 1, two states’ guidelines include ex-ceptions to their IQ cutoff criterion. Nebraskaguidelines specify that students may be eligi-ble for special education services for MRbased on either (a) an IQ � 2 SDs below thenormative mean with commensurate (notspecified) deficits in adaptive functioning or(b) an IQ � 80 with significant deficits in oneor more adaptive skill or achievement areas(defined as standard scores � 2 SDs below thenormative mean). Wisconsin guidelines re-quire students initially being considered for


TA

BL

E1

Elig

ibili

tyG

uide

lines

for

Men

tal

Ret

arda

tion

acro

ssSt

ates

Stat

eYe

arT

erm

Lev

els

IQSc

ore

rang

eIQ

part

scor

esA

BSc

ore

AB

crite

rion

Mul

tiple

AB

setti

ngs

Ala

bam

a20

05M

enta

lR

etar

dati

onN

�2

SDN

NE

ith

erco

mpo

site

or2

dom

ain

s/sk

illar

eas

�2

SDM

easu

red

in2

dist

inct

sett

ings

(cri

teri

onm

etin

1)A

lask

a20

05M

enta

lR

etar

dati

onN

�2

SDN

NN

SN

SN

SA

rizo

na

2003

Men

tal

Ret

arda

tion

Y�

2SD

NN

NS

�2

SDN

SA

rkan

sas

2004

Men

tal

Ret

arda

tion

N�

70–7

5Y

ND

omai

ns/

skill

area

(2)

NS

Mea

sure

din

1se

ttin

g;op

tion

ally

mea

sure

din

mul

tipl

ese

ttin

gsC

alif

orn

ia20

05M

enta

lR

etar

dati

onN

FED

NN

NS

NS

NS

Col

orad

o20

05Si

gnifi

can

tL

imit

edIn

telle

ctua

lC

apac

ity

N�

2SD

NN

NS

NS

NS

Con

nec

ticu

t20

00In

telle

ctua

lD

isab

ility

N�

2SD

YY

Dom

ain

s/sk

illar

ea(m

ajor

ity)

�1.

5SD

Mea

sure

din

2di

stin

ctse

ttin

gsD

elaw

are

2004

Men

tal

Dis

abili

tyY

�70

YN

Dom

ain

s/sk

illar

ea(2

)N

SN

SD

istr

ict

ofC

olum

bia

1997

Men

tal

Ret

arda

tion

YFE

DN

NN

SN

SN

SFl

orid

a20

05M

enta

llyH

andi

capp

edY

�2

SDY

YN

SN

SN

SG

eorg

ia20

02In

telle

ctua

lD

isab

ility

Y�

701

YN

NS

NS

Mea

sure

din

2di

stin

ctse

ttin

gsH

awai

i20

00M

enta

lR

etar

dati

onN

�2

SDN

ND

omai

ns/

skill

area

(2)

NS

NS

Idah

o20

05C

ogn

itiv

eIm

pair

men

tN

�70

YN

Dom

ain

s/sk

illar

ea(2

)N

SN

SIl

linoi

s20

03M

enta

lR

etar

dati

onN

FED

NN

NS

NS

NS

Indi

ana

2002

Men

tal

Dis

abili

tyY

�2

SDN

NN

SN

SN

SIo

wa

2005

N/A

--

--

NA

NA

NA

Kan

sas

2001

Men

tal

Ret

arda

tion

N�

2SD

NN

Dom

ain

s/sk

illar

ea(2

)N

SN

SK

entu

cky

2000

Men

tal

Dis

abili

tyY

�2

SDN

NN

S�

2SD

NS

Lou

isia

na

2004

Men

tal

Dis

abili

tyY

�2

SDN

NB

oth

com

posi

tean

ddo

mai

ns/

skill

area

(2)

NS

NS

Mai

ne

2003

Men

tal

Ret

arda

tion

NFE

DN

NN

SN

SN

SM

aryl

and

2005

Men

tal

Ret

arda

tion

NFE

DN

NN

SN

SN

S


TA

BL

E1—

(Con

tinue

d)

Stat

eYe

arT

erm

Lev

els

IQSc

ore

rang

eIQ

part

scor

esA

BSc

ore

AB

crite

rion

Mul

tiple

AB

setti

ngs

Mas

sach

uset

tsIn

telle

ctua

lIm

pair

men

tN

NS

NN

NS

NS

Con

side

red

inm

ulti

ple

sett

ings

Mic

hig

an20

02C

ogn

itiv

eIm

pair

men

tY

�2

SDN

2N

NS

NS

NS

Min

nes

ota

2005

Dev

elop

men

tal

Cog

nit

ive

Dis

abili

tyY

�2

SDY

YB

oth

com

posi

tean

ddo

mai

ns/

skill

area

s(4

)

Com

posi

te�

15th

%ti

le;

docu

men

tati

onof

nee

dsin

4of

7do

mai

ns

Mea

sure

din

2di

stin

ctse

ttin

gs

Mis

siss

ippi

2003

Men

tal

Ret

arda

tion

Y�

2SD

YY

Bot

hco

mpo

site

and

dom

ain

s/sk

illar

eas

Vine

land

:C

ompo

site

�

2SD

,al

ldo

mai

ns

�

85;

AB

S-Sc

hool

:2:

Part

On

eFa

ctor

Con

side

red

inm

ulti

ple

sett

ings

Scor

es�

2SD

:m

ust

just

ify

scor

esab

ove

this

Mis

sour

i20

05M

enta

lR

etar

dati

onN

�2

SDN

NN

SN

SN

SM

onta

na

2004

Cog

nit

ive

Del

ayN

�2

SDY

NN

SN

SN

SN

ebra

ska

2004

Men

tal

Han

dica

pY

�2

SDor

�

80

NY

Dom

ain

s/sk

illar

ea(1

)�

2SD

(if

usin

gIQ

�

80cr

iter

ion

)O

ptio

nal

lym

easu

red

inat

leas

t2

dist

inct

sett

ings

Nev

ada

2000

Men

tal

Ret

arda

tion

Y�

2SD

NY

Dom

ain

s/sk

illar

ea(2

)N

SN

SN

ewH

amps

hir

e20

02M

enta

lR

etar

dati

onN

FED

NN

NS

NS

NS

New

Jers

ey20

03C

ogn

itiv

ely

Impa

ired

Y�

2SD

NN

NS

NS

Con

side

red

inm

ulti

ple

sett

ings

New

Mex

ico

2005

Men

tal

Ret

arda

tion

N�

70Y

NE

ith

erco

mpo

site

ordo

mai

ns/

skill

area

(1)

�2

SDN

S

New

York

2004

Men

tal

Ret

arda

tion

NFE

DN

NN

SN

SN

SN

orth

Car

olin

a20

04M

enta

llyD

isab

led

Y�

70Y

NN

SN

SN

SN

orth

Dak

ota

1999

Men

tal

Ret

arda

tion

Y�

70Y

NN

SN

SN

SO

hio

2002

Cog

nit

ive

Dis

abili

tyN

�70

YN

Dom

ain

s/sk

illar

ea(2

)N

SN

S


TA

BL

E1—

(Con

tinue

d)

Stat

eYe

arT

erm

Lev

els

IQSc

ore

rang

eIQ

part

scor

esA

BSc

ore

AB

crite

rion

Mul

tiple

AB

setti

ngs

Okl

ahom

a20

02M

enta

lR

etar

dati

onY

�2

SDY

YN

SN

SC

onsi

dere

din

mul

tipl

ese

ttin

gsO

rego

n20

04M

enta

lR

etar

dati

onN

�2

SDN

NN

SN

SN

SPe

nn

sylv

ania

2001

Men

tal

Ret

arda

tion

NFE

DN

NN

SN

SO

ptio

nal

lyco

nsi

dere

din

mul

tipl

ese

ttin

gsR

hod

eIs

lan

d20

00M

enta

lR

etar

dati

onN

FED

NN

NS

NS

NS

Sout

hC

arol

ina

2002

Men

tal

Dis

abili

tyY

�2

SDon bo

thV

/NV

YY

Com

posi

te�

2SD

Mea

sure

din

1se

ttin

g

Sout

hD

akot

a20

04M

enta

lR

etar

dati

onN

�2

SDY

NN

SN

SN

ST

enn

esse

e20

03M

enta

lR

etar

dati

onN

�2

SDY

NC

ompo

site

�2

SDM

easu

red

inm

ulti

ple

sett

ings

Tex

as20

02M

enta

lR

etar

dati

onN

�2

SDN

NN

SN

SN

SU

tah

2003

Inte

llect

ual

Dis

abili

tyN

�2

SDN

2Y

NS

NS

Mea

sure

din

mul

tipl

ese

ttin

gsV

erm

ont

2003

Lea

rnin

gIm

pair

men

t/D

elay

inL

earn

ing

N�

1.5

SDN

NN

SN

SN

S

Vir

gin

ia20

02M

enta

lR

etar

dati

onN

FED

NN

NS

NS

NS

Was

hin

gton

2002

Men

tal

Ret

arda

tion

NFE

DN

NN

SN

SN

SW

est

Vir

gin

ia20

05M

enta

llyIm

pair

edN

�70

YN

Dom

ain

s/sk

illar

ea(2

)N

SN

SW

isco

nsi

n20

02C

ogn

itiv

eD

isab

ility

N�

2SD

or�

1SD

YN

Dom

ain

s/sk

illar

ea(2

)�

2SD

NS

Wyo

min

g19

99M

enta

lD

isab

ility

N�

2SD

YY

NS

NS

NS

Not

e:A

B�

adap

tive

beh

avio

r,FE

D�

fede

ral

defi

nit

ion

only

,N

S�

not

spec

ified

.1

Un

der

Geo

rgia

guid

elin

es,

mor

eth

anon

efo

rmal

mea

sure

ofin

telli

gen

ceis

requ

ired

.2

Acc

ordi

ng

toM

ich

igan

guid

elin

es,I

Qs

ator

belo

wap

prox

imat

ely

2SD

sati

sfy

the

inte

llect

uald

efici

ency

crit

erio

n.S

imila

rly,

Uta

hgu

idel

ines

stat

edth

atIQ

ssh

ould

“gen

eral

ly”

bele

ssth

anor

equa

lto

2SD

belo

wth

em

ean

.How

ever

,bec

ause

the

guid

elin

esdo

not

defi

ne

“app

roxi

mat

ely”

or“g

ener

ally

”th

esc

ore

ran

ges

wer

eco

ded

asN

.


eligibility to exhibit an IQ two or more SDsbelow the normative mean, but the guidelinesallow continued eligibility for students who,upon reevaluation, exhibit IQs between oneand two SDs below the normative mean if thestudent is expected to exhibit intellectual def-icits indefinitely. Approximately 39% of stateguidelines specify consideration of measure-ment error using either an IQ range (e.g., 70to 75) or the SEM (and associated ranges)surrounding IQs in determining the intellec-tual deficit criterion.

Most states specify the use of a comprehen-sive intelligence test battery, which typicallyyield both an IQ and part scores, but approx-imately 20% of states (10 states) mention theuse of part scores in the MR definition orrequire school-based multidisciplinary teamsto consider part scores in eligibility determi-nation. Most states with provisions for the useof part scores do not allow the use of a partscore in isolation to satisfy the intellectual def-icit criterion but require consideration of partscores in eligibility decisions. For example,some guidelines require that profiles of partscores be within the deficient range, somespecify a cutoff criterion for part scores, andsome require examination of part scores andfurther investigation of a student’s abilities ifthere are discrepancies between part scores.However, only one state (South Carolina) re-quires normative deficiencies to be evidencedin part scores rather than an IQ.

Deficient Adaptive Functioning Criterion

Procedures for determining adaptive behaviordeficiencies vary greatly across states. The ma-jority of states (63%) do not specify whethercomposite scores reflecting global adaptivefunctioning or scores reflecting adaptive be-havior domains or skill areas be used to estab-lish deficient functioning. Among stateswhose guidelines specify the necessary scores,two states specify the use of adaptive behaviorcomposites reflecting overall adaptive func-tioning, 11 states specify the use of scores foradaptive behavior domains or skill areas, threestates require that both composites and do-mains or skill areas be used, and two statesallow for either composite scores or domainsor skill areas to be used. Most states (77%) donot include a specific score criterion for defi-

cient functioning, 16% require scores of least2 SDs below the normative mean, and 2%require scores of at least 1.5 SDs below thenormative mean. In contrast, 2% of states re-quire deficient adaptive functioning to bebased on separate criteria for the adaptivebehavior composite and the adaptive behaviordomains or skill areas. Although the diagnos-tic criteria outlined by most professionalgroups specify that adaptive skill deficits beevidenced across settings, only six states(12%) require adaptive functioning to bemeasured in multiple settings, and only fouradditional states specify that adaptive func-tioning be considered in multiple settings.(We coded the following two instances as in-dicating that adaptive behaviors were consid-ered in multiple settings: descriptions that in-dicated only considerations of or informalobservations of behaviors in non-school set-tings and descriptions of ratings completed bya single rater to describe behaviors in schooland other settings.) In contrast, two states(4%) specify procedures for measuring adap-tive functioning in one setting, and 37 states(73%) do not address the settings in whichadaptive functioning be measured.

Discussion

This study provided an updated summary ofstates’ guidelines and investigated two issuesthat have not received much attention in pre-vious reviews of state guidelines. In a manneralmost identical to Denning et al (2000), wefound the term mental retardation is used byapproximately half the states to describe thecondition and that mental handicap is used byapproximately 12% of states. In contrast,fewer states than reported in Denning et al.now require differentiating levels of MR ac-cording to degree of impairment or severitybased on IQs. We found that only 18 states inour study require these levels, whereas Den-ning et al. reported 27 states. Perhaps thisdecreased prevalence was influenced by criti-cisms, such as that by Wehmeyer (2003), not-ing that the focus on labels related to levels ofimpairment lowers expectations by encourag-ing educational placements and interventionsbased on perceived global impairment ratherthan on individual needs.

States require consideration of IQs with an


upper cutoff ranging from a low of 69 to ahigh of 80 for initial evaluations and a high of85 for re-evaluations. Approximately 39% ofstate guidelines specify consideration of mea-surement error using either an IQ range orthe SEM (and associated ranges) surroundingIQs in determining the intellectual deficit cri-terion. This percentage is only slightly higherthan that reported by Utley et al. (1987),which was 36%.

Effects of recent theory-based test develop-ment and interpretation and the SSA guide-lines (NRC, 2002) focusing on composite-based part scores do not appear to be farreaching as of yet. For example, fewer thanone-fifth of states make reference to partscores, and only one state requires normativedeficiencies to be evidenced in part scoresrather than in IQs. Furthermore, when statesguidelines made reference to part scores, theywere most often to those scores that are notbased on recent theories of intelligence (e.g.,Verbal IQ and Performance IQ; Alfonso et al.,2005; Kaufman & Lichtenberger, 2005; Wech-sler, 2003).

Across the eligibility guidelines, there waslittle consistency (a) in the descriptions ofwhich adaptive behavior scores (i.e., compos-ites or domains/skill areas) were required foridentification and (b) in the criteria used tojudged adaptive behaviors as deficits (e.g., �70). If there was any consistency in these ar-eas, it was revealed in the majority of stateguidelines omitting descriptions of whichadaptive behavior scores are required and inthe majority of state guidelines failing to spec-ify the criterion indicating adaptive behaviordeficits. However, in general, a few morestates (five more) now appear to list specificpractices recommended for adaptive behaviorassessment than they did about eight years ago(cf. Denning et al., 2000). It is required thatthose making judgments about special educa-tion eligibility for children suspected of hav-ing MR consider foremost educational impair-ment because of the risk of identifying “6-hourretarded children” whose deficits are not ap-parent across settings (President’s Committeeon Mental Retardation, 1969). Thus, it wasquite unexpected that well less than one-tenthof the states require adaptive functioning tobe measured in multiple settings. Despite thisserious limitation across the states, perhaps it

is beneficial to acknowledge that progress inthis area appears to have been made acrossalmost two decades. Our review revealed thatall states but one (Nebraska) require the pres-ence of adaptive behavior skill deficits foridentification of mental retardation, whereasUtley et al. (1987) and Frakenberger andFronzaglio (1991) revealed that only approx-imately two-thirds of states require the pres-ence of adaptive behavior skill deficits.

Advantage and Limitations

The accessibility of information from theWorld Wide Web allowed us direct access tothe eligibility guidelines for MR from mostevery state. Therefore, unlike previous re-search, state department administrators werenot surveyed by paper-and-pencil methods.With careful contact and follow-up by phoneto such individuals to ensure that we had ac-cessed the most recent versions of these guide-lines, limitations of our approach to data col-lection are minimal. As with all such research,we anticipate that there have been changes tothe eligibility guidelines since we obtained theinformation summarized in this manuscript.Similarly, it is possible that our coding of theinformation found in the state guidelines doesnot match perfectly with the manner in whichthe guidelines are interpreted by those withinstates. For example, it is likely that our inter-pretation of the wording in the guidelinesoften led us to conclude that specific criteriawere not specified well enough to code assomething other than “not specified,” whenthose using the guidelines may have inter-preted the wording in the guidelines differ-ently. Thus, despite relatively high levels ofinter-rater agreement in our study, the codingmay not reflect actual practices in the field.

Conclusion

Psychologists and other professionals involvedin assessment of children with or expected tohave MR should be not only (a) well informedabout their state’s and neighboring states’ el-igibility criteria but also (b) knowledgeableabout best practices in the use and interpre-tation of intelligence tests and adaptive behav-ior assessment instruments. Our results revealthat some notable variations exist in the eligi-


bility guidelines for MR from state to state.Variations include the terms used to describethis exceptionality, the criterion used to iden-tify an intellectual deficit, and the scores andcriteria used to identify adaptive behavior de-ficiencies. Despite the finding that few statesrequire that adaptive behavior deficits be ap-parent in more than one setting, psychologistsand other professionals should continue tofollow best practices by ensuring that suchdeficits are considered from an ecological per-spective through assessment in multiple set-tings. Similarly, despite what is known aboutinconsistencies in measurement due to ran-dom influences on test performance, fewerthan half of states recommend attention tounreliability of IQs though use of SEM andassociated score ranges. Although some psy-chologists and other professionals may placeemphasis on part scores that are theoreticallybased and reliable measures from recentlypublished intelligence tests, less than a quar-ter of states recommend consideration ofthem during eligibility. We urge those con-ducting such assessments and policy makers toconsider best practices and advancements intheory and measurement of intelligence andadaptive behaviors during upcoming revisionsto their state guidelines for MR.

References

Alfonso, V. C., Flanagan, D. P., & Radwan, S. (2005).The impact of the Cattell–Horn–Carroll theoryon test development and interpretation of cogni-tive and academic abilities. In D. P. Flanagan & P.Harrison (Eds.), Contemporary intellectual assess-ment (2nd ed., pp., 185–202). New York: GuilfordPress.

American Association on Mental Retardation.(2002). Mental retardation: Definition, classification,and systems of supports (10th ed.). Washington, DC:Author.

American Psychiatric Association. (2000). Diagnosticand statistical manual of mental disorders (4th ed.,text revision). Washington, DC: Author.

Brody, N. (1997). Intelligence, schooling, and soci-ety. American Psychologist, 52, 1046–1050.

Daniel, M. H. (1997). Intelligence testing. AmericanPsychologist, 52, 1038–1045.

Denning, C. B., Chamberlain, J. A., & Polloway,E. A. (2000). An evaluation of state guidelines formental retardation: Focus on definition and clas-

sification practices. Education and Training in Men-tal Retardation and Developmental Disabilities, 35,226–232.

Frakenberger, W. (1984). A survey of state guide-lines for identification of mental retardation.Mental Retardation, 22, 17–20.

Frakenberger, W., & Fronzaglio, K. (1991). States’definitions and procedures for identifying chil-dren with mental retardation: Comparison overnine years. Mental Retardation, 29, 315–321.

Horn, J. L., & Noll, J. (1997). Human cognitivecapabilities: Gf–Gc theory. In D. P. Flanagan, J. L.Genshaft, & P. L. Harrison (Eds.), Contemporaryintellectual assessment: Theories, tests, and issues (pp.53–91). New York: Guilford Press.

Individuals with Disabilities Education Act, Pub. L.108–446 (2004).

Jensen, A. R. (1998). The g factor. Westport, CT:Preager.

Kaufman, A. S., & Lichtenberger, E. O. (2005).Assessing adolescent and adult intelligence (3rd. ed.).

Needham Heights, MA: Allyn & Bacon.National Archives and Records Administration.

(2005). Assistance to states for the education ofchildren with disabilities, Federal Register, 70(118). Washington, DC: Government Printing Of-fice.

National Research Council. (2002). Mental retarda-tion: Determining eligibility for Social Security benefits.Washington, DC: National Academy Press.

Neisser, U., Boodoo, G., Bouchard, T. J., Boykin,A. W., Brody, N., Ceci, S. J., et al. (1996). Intelli-gence: Knowns and unknowns. American Psycholo-gist, 51, 77–101.

President’s Committee on Mental Retardation.(1969). The six-hour retarded child. Washington,DC: U.S. Government Printing Office.

Sheerenberger, R. C. (1983). A history of mental re-tardation. Baltimore, MD: P. H. Brookes.

Utley, C. A., Lowitzer, A. C., & Baumeister, A. A.(1987). A comparison of the AAMD’s definition,eligibility criteria, and classification schemes withstate departments of education guidelines. Educa-tion and Training in Mental Retardation, 22, 35–43.

Wechsler, D. (2003). The Wechsler Intelligence Scale forChildren–Fourth Edition. Antonio, TX: Psychologi-cal Corporation.

Wehmeyer, M. L. (2003). Defining mental retarda-tion and ensuring access to general curriculum.Education and Training in Developmental Disabilities,38, 271–282.

Received: 23 August 2006Initial Acceptance: 11 October 2006Final Acceptance: 20 January 2007


Date post:	08-Mar-2018
Category:	Documents
Upload:	lamthuy
View:	217 times
Download:	2 times

Education and Training in Developmental Disabilities and Training in Developmental Disabilities The...

Documents