USF Libraries
USF Digital Collections

It is time to play!

MISSING IMAGE

Material Information

Title:
It is time to play! peer implemented pivotal response training with a child with autism during recess
Physical Description:
Book
Language:
English
Creator:
Sams, Leigh Anne
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Peer training
PRT
Teaching play skills
Peer mediated interventions
ASD
Dissertations, Academic -- Applied Behavior Analysis -- Masters -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: Children with autism, by diagnostic criteria, experience a lack of age appropriate play and social skills (American Psychiatric Association, 1994). This study evaluated three typically developing peer's ability to implement Pivotal Response Training strategies during recess with a child with autism in their third grade class. A concurrent multiple probe baseline across peers design was used to assess peers ability to implement Pivotal Response Training strategies with the target child and the effects of intervention on the play and communication behavior of the target child. Generalization measures were taken in an untrained environment. Measures of social validity in the form of peer interviews and teacher questionnaire were completed. Results displayed that peers were able to use Pivotal Response Training strategies during recess with a child with autism. During post-training/intervention and follow-up sessions the child with autism engaged in higher levels of communication to peers, and slightly lower levels of communication to self. The target child also experienced a decreasing trend in levels of solitary play, an increasing trend in levels of play with peer trainers and levels of parallel play remained near baseline levels. Peers were also able to generalize these skills to an untrained environment to an extent.
Thesis:
Thesis (M.A.)--University of South Florida, 2009.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Leigh Anne Sams.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 64 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002028840
oclc - 436453132
usfldc doi - E14-SFE0002819
usfldc handle - e14.2819
System ID:
SFS0027136:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200385Ka 4500
controlfield tag 001 002028840
005 20090914141751.0
007 cr bnu|||uuuuu
008 090914s2009 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002819
035
(OCoLC)436453132
040
FHM
c FHM
049
FHMM
090
BF176.7 (Online)
1 100
Sams, Leigh Anne.
0 245
It is time to play! :
b peer implemented pivotal response training with a child with autism during recess.
h [electronic resource] /
by Leigh Anne Sams.
260
[Tampa, Fla] :
University of South Florida,
2009.
500
Title from PDF of title page.
Document formatted into pages; contains 64 pages.
502
Thesis (M.A.)--University of South Florida, 2009.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
3 520
ABSTRACT: Children with autism, by diagnostic criteria, experience a lack of age appropriate play and social skills (American Psychiatric Association, 1994). This study evaluated three typically developing peer's ability to implement Pivotal Response Training strategies during recess with a child with autism in their third grade class. A concurrent multiple probe baseline across peers design was used to assess peers ability to implement Pivotal Response Training strategies with the target child and the effects of intervention on the play and communication behavior of the target child. Generalization measures were taken in an untrained environment. Measures of social validity in the form of peer interviews and teacher questionnaire were completed. Results displayed that peers were able to use Pivotal Response Training strategies during recess with a child with autism. During post-training/intervention and follow-up sessions the child with autism engaged in higher levels of communication to peers, and slightly lower levels of communication to self. The target child also experienced a decreasing trend in levels of solitary play, an increasing trend in levels of play with peer trainers and levels of parallel play remained near baseline levels. Peers were also able to generalize these skills to an untrained environment to an extent.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
590
Advisor: Lise Fox, Ph. D.
653
Peer training
PRT
Teaching play skills
Peer mediated interventions
ASD
690
Dissertations, Academic
z USF
x Applied Behavior Analysis
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.2819



PAGE 1

It is Time to Play! Peer Implemented Pivotal Response Training with a Child with Autism during Recess by Leigh Anne Sams A thesis submitted in partial fulfillment of the requirements for the degree of Masters of Applied Behavior Analysis Department of Child and Family Studies College of Behavioral and Community Sciences University of South Florida Major Professor: Lise Fox, Ph.D. Bobbie Vaughn, Ph.D. Debra Mowery, Ph.D. Data of Approval: March 20, 2009 Keywords: peer training, PRT, teaching play skills, peer mediated interventions, ASD Copyright 2009, Leigh Sams

PAGE 2

i Table of Contents List of Tables iii List of Figures iv Abstract v Chapter One: Introduction 1 Peer Mediated Interventions 2 Peer Training Approaches 4 Peer Outcomes 10 Pivotal Response Training (PRT) 11 PRT to Teach Social and Play skills 13 PRT Implemented by Peers 14 Conclusion 16 Chapter Two: Method 18 Purpose 18 Participants 19 Settings and Materials 20 Design 22 Procedures 22 Baseline 22 Peer PRT Training 23 Post-training/Intervention 25 Booster Training Session 26 Generalization 26 Follow-up 27 Measures 27 Dependent Variable 27 Inter-observer Agreement 30 Peer’s Target Behaviors 30 Eve’s Target Behaviors 31 Data Collectors 31 Fidelity of Peer Training 31 Social Validity 32 Peers 32 Teacher 33

PAGE 3

ii Chapter Three: Results 34 Peer’s use of PRT strategies 34 Eve’s Play Behaviors 37 Solitary Play 37 Parallel Play 38 Play with Peer Trainer 39 Eve’s Communication Behaviors 42 Communication to Self 42 Communication to Peers 43 Social Validity 45 Peers 45 Teacher 45 Chapter Four: Discussion 47 References 54 Appendices 60 Appendix A: Peer Data Collection Sheet 61 Appendix B: Target Child Data Collection Sheet 62 Appendix C: Peer Interview Ques tions 63 Appendix D: Teacher Questionnaire 64

PAGE 4

iii List of Tables Table 1 Eve’s play behaviors du ring generalization probes 41

PAGE 5

iv List of Figures Figure 1 Peers’ use of PRT strategi es 36 Figure 2 Eve’s play behaviors 40 Figure 3 Eve’s communication behaviors 44

PAGE 6

v It is Time to Play! Peer Implemented Pivotal Response Training with a Child with Autism during Recess Leigh Anne Sams ABSTRACT Children with autism, by diagnostic criteria, experience a lack of age appropriate play and social skills (American Psychiatric A ssociation, 1994). This study evaluated three typically developing peer’s ab ility to implement Pivotal Response Training strategies during recess with a child with autism in thei r third grade class. A concurrent multiple probe baseline across peers design was used to assess peers ability to implement Pivotal Response Training strategies with the target child and the effects of interven tion on the play and communication behavior of the targ et child. Generalization measures were taken in an untrained environment. Measures of social validity in the form of peer interviews and teacher questionnaire were co mpleted. Results displayed that peers were able to use Pivotal Response Training strate gies during recess with a child with autism. During post-training/intervention and follow-up se ssions the child with autism engaged in higher levels of communication to peers, a nd slightly lower levels of communication to self. The target child also experienced a decr easing trend in levels of solitary play, an increasing trend in levels of pl ay with peer trainers and leve ls of parallel play remained near baseline levels. Peers were also able to generalize these skills to an untrained environment to an extent.

PAGE 7

1 Chapter One Introduction A lack of appropriate play and social skills is one of the diagnostic core components of autism. The Diagnostic and Statistical Manua l of Mental Disorders (4th ed.; DSM-IV ; American Psychiatric Association, 1994, p.75) describes the social delays of children with autism as a “qualitative impair ment in social intera ction”. Features of autism include the failure to develop developm entally appropriate pe er relationships and social or emotional reciprocity, as well as impairments in communication, which includes “lack of spontaneous make-believe play or social imitative play appropriate to developmental level (American Psyc hiatric Association, 1994 p.75).” A critical component of inst ruction for students with autism is the promotion of social interaction skills. Many students with autism ar e taught within inclusive classrooms with their typically developing peers. When students with autism are included in classrooms with nondisabled stude nts, there are opportunities for learning social interaction skills. However, placemen t alone does not result in an increase in social interaction ability (Bass, & Mulick, 2007 ). Bass and Mulick ( 2007) warn that if specific supports are not put in place to aid in socialization when including students with autism in a classroom with typically deve loping children, the children with autism may experience social isolation. The authors al so explain that when peers are used in

PAGE 8

2 interventions with children with autism, then peers can provide the target child with more opportunities to practice thes e skills in a variety of play activities. One way to help children with autism impr ove social and play skills that has been utilized in many studies over the past few decad es is through peer mediated interventions. In peer mediated approaches, typically de veloping children are trained to deliver systematic procedures designed to promote ta rgeted skills by their peers with autism. Another way to help children with autism improve these skills is Pivotal Response Training or (PRT). Pivotal Response Training (PRT) is a naturalistic behavioral intervention that has been succe ssfully used to teach students with autism social and play skills (Koegel, & Frea, 1993; Stahmer, 1995; Thorp, Stahmer, & Schreibman, 1995). A promising approach for promoting the acquisiti on of play and social skills by students with autism is to combine these two inte rvention approaches (Harper, Symon, & Frea, 2008; Kuhn, Bodkin, Devlin, & Dogget, 2008; Pi erce, & Schreibman, 1995; Pierce, & Schreibman, 1997). Peer-Mediated Interventions Peer mediated interventions or peer tu toring refers to the use of typically developing peers to support, prompt, or syst ematically interact with peers who have disabilities. Peer tutoring has been used in school and therapy settings to improve the social skills of children with autism (Laushey, & Heflin, 2000; Morrison, Kamps, Garcia, & Parker, 2001; Owen-DeSchryver, Carr, Ca le, & Blakeley-Smith, 2008; Strain, 1983; Strain, Shores, & Timm, 1977). Training proced ures for peer tutors have included the use of scripts and textual cuing (Goldste in, & Cisar, 1992; Petu rsdottir, McComas, McMasters, & Horner, 2007; Thiemann, & Gold stein, 2004), teaching peers to implement

PAGE 9

3 systematic prompts and praise (Goldste in, Kaczmarek, Pennington, & Shafer, 1992; Strain, Kerr, & Ragland, 1979), training peers to implement naturalistic and incidental teaching (Kohler, & Strain, 1997; McGee, Al meida, Sulzer-Azaroff, & Feildman, 1992), using group oriented contingenc ies (Kohler et al., 1995) and the creation of integrated play groups and peer networks to support the social interactions of ch ildren with autism with typically developing peers (Kamps, Potueck, Lopez, Kravits, & Kemmerer, 1997). Research on the use of peer-mediated interv entions and the benefits noted for students with autism and their nondisa bled peers are described belo w. Following that discussion, a description of the use of p eer mediated delivery of a sp ecific prompting procedure, pivotal response training, is provided with an analysis of the resear ch that has supported the use of this approach for helping student s with autism in their development of peer social and play interactions. The use of peer-mediated interventions for th e instruction of play and social skills for students with autism is a pr eferred approach to instruction by adults of social skills for several reasons. First, a primar y objective of social skills training is to teach the child with autism to interact with typically deve loping peers. When using peers as the training agent, interaction with peers is immedi ately occurring. Second, the goal of teaching social skills is to have st udents with autism use those skil ls in natural contexts. By having peers teach within the natural context, there is no extra step from learning social skills with an adult and then transferring those skills to interacting with peers (DiSalvo, & Oswald, 2002). Finally, the use of peers for in struction allows students with autism to experience many more instructional trials than would be possible if the classroom teacher was the only intervention agent.

PAGE 10

4 Peer training approaches. Researchers have approached the training of peers in a variety of ways (Laushey, & Heflin, 2000; Mo rrison et al, 2001; Owe n-DeSchryver et al., 2008; Strain, 1983; Strain et al., 1977). In an early study by Strain et al. (1977) general instructions were provided to typically develo ping peers to get the target child to play with them, and stressed the importance of bei ng persistent. Each of the target children’s levels of social behavior incr eased after intervention, decrease d in a return to baseline and increased when intervention was implemented again. Because adults were not involved in the actual play session with the childre n, there was a presumption that the intervention would generalize to natural play routines. However, because data were not collected outside of the treatment sett ing, it is unknown if these beha viors generalized to other settings (Strain et al., 1977). Several years later Strain (1983) conducted a similar study on the use of peers to work with children with autism and measured ge neralization of these skills to integrated verses segregated setting. This study involved providing a typi cally developing peer structured information about wa ys to interact with childre n with autism such as some verbal play organizers, how to share play items, and how to provide physical assistance related to play. In this study, the peer was inst ructed to try his best to get the target child to play and to be persistent. All of the ta rget children’s levels of positive interaction increased during intervention, a nd generalization was greater in integrated as opposed to segregated settings. These results display th e importance of placing children with autism in an environment that will be supportive of their newly gained skills (Strain, 1983). Unlike the previous studies that only trained one peer to work with the target child, Laushey and Heflin (2000) trained an entire ki ndergarten class to sta y, play, and talk to a

PAGE 11

5 buddy. Buddy pairs were composed of two ch ildren in the class and included target children and their typically developing peers. Results showed that the target children had more appropriate social skills while in the buddy program than when they simply played in the same area as typically developing childr en. These results clearly display that in order to increase the social interactions between typically developing children and children with autism simply placing these chil dren in the same environment is not enough (Laushey, & Heflin, 2000). Morrison et al. (2001) also implemente d a treatment that involved teaching typically developing peers and target children techniques to improve social skills, such as requesting, commenting, and sharing. Peers mon itored their behavior and the behavior of the target children for occurren ces of these social skills and delivered rewards for target children’s appropriate behavior Generalization data were collected during lunch and recess. The intervention resulted in target children’s increased use of social skills, and while some generalization of skills occurred, generalization did not occur for all students. Additional training may have assisted in th e generalization of these skills to new environments (Morrison et al., 2001). More recently Owen-DeSchryver et al. (2008) provided training to typically developing peers on ways to increase social in teractions of students with autism in an inclusive school setting. Peer s received training on five centr al themes about interacting with children with autism during lunch a nd recess. These central themes involved teaching nondisabled peers when to play with the target child, what to talk about, what to play, how to help him or her learn to play, and what to do if he or she is not responding or engages in unusual behavior. Peer initiations increased for all of the participants after

PAGE 12

6 intervention, and they generali zed their skills to untrained peers. This study shows how the use of multiple peers can increase the like lihood of skills genera lizing to other people (Owen-DeSchryver et al., 2008). Some studies have been conducted that involved teaching typically developing children to use prompting and praising procedur es with children with autism to promote their use of play and social in teraction skills (Golds tein et al., 1992; Stra in et al., 1979). As in previous studies by Strain, peers in a study by Strain et al. (1979) were first instructed to attempt to get the target child to play with them and to be persistent. Peers were then provided training to prompt the ta rget child to play and to provide social reinforcement for appropriate play. The resu lts showed that both interventions increased the target child’s positive soci al behavior equally well, but that neither method resulted in generalization outside of the expe riment (Strain et al., 1979). An additional study that looked at prom pting and praising procedures was done by Goldstein et al. (1992). During training peers were taugh t strategies for providing mutual attention to the play activity, making comments on ongoing activities, and acknowledging their partner’s communicative behaviors. Following training, peers played with the target children. After th e intervention was implemented, the target children exhibited an increase in the freque ncy of total social behavior. Since no measures of generalization were taken, it is un known if these skills ge neralized outside of the training session (Gol dstein et al., 1992). Another approach to the use of peer mediat ed interventions is the use of scripts or textual cueing by typically developing students to promote the social behavior of their peers with autism (Goldstein, & Cisar, 1992; Petursdottir et al., 2007; Thiemann, &

PAGE 13

7 Goldstein, 2004). Goldstein and Cisar (1992) used sociodramatic scripts to teach children with autism and their typically deve loping peers’ appropr iate verbal and nonverbal behaviors to engage in during three different sociodrama tic themed play situations. The results of the experiment show that the ch ildren were able to learn the three scripts, but only after repeating the same highly st ructured play activity numerous times. Although some children engaged in social behavior s not directly related to the scripts it is not known if the experiment re sulted in any sign ificant changes in social behavior because measures of generalization were not taken outside of the training sessions (Goldstein, & Cisar, 1992). In their 2004 study, Thiemann and Goldstei n provided typically developing peers training in appropriate social skills to use with children with autism such as “look, wait and listen”, “answer questions”, “keep talking”, “say something nice”, and “start talking”. Visual supports were added to the interaction that included written phrases that related to the target skill and forms for m onitoring their progress. Only two of the target children improved their rates of interaction following pe er training alone, but once the textual cues were added to the intervention all of the children improved in the rates of the three targeted social-communication skills (T hiemann, & Goldstein, 2004). Although the experiment increased the children’s target ed social-communication skills, this only occurred when numerous textual prompts we re added, which is unnatural for typical children’s play. In their 2007 study Petursdottir et al. used Peer-Assisted Learning Strategies for Kindergartners (K-PALS), a program for be ginning readers that involves peers using scripted lines to offer assistance with readi ng partners, with and without the addition of

PAGE 14

8 play related stimuli. The study found that simply being involved in K-PALS did not increase the target child’s level of interac tion during a free-play se ssion after the activity. When play related stimuli were included in the session and carried over into free play sessions afterwards the target child’s level of interactions did increase and when that stimuli was removed from the K-PALS session interactions again decreased. These results again show that simply having ch ildren with autism interact with typically developing children does not automatically result in positive changes in behavior (Petursdottir et al., 2007). In an attempt to use more natural methods of instruction and to promote generalization of skills, researchers have exam ined the use of naturalistic or incidental teaching (Kohler, & Strain, 1997; McGee et al, 1992). Cowan and Allen (2007) explain that “naturalistic procedures typically take place through loosely structured sessions, which are initiated and paced by the child, take place in a variety of locations and positions, and employ a variety of stimuli” (Cowan, & Allen, 2007, p. 702). McGee et al. (1992) tau ght typically developing ch ildren to use incidental teaching with children with autism. Peers re ceived instructions to wait for the child to make a request, have the child label the item, and then give the item to the child and provide praise. An experimenter provided pi cture prompts of each step of the incidental teaching process while the peer interacted with the target child. Target children’s use of reciprocal interactions increased after intervention. Because multiple peers were used with each of the target children in the study, they had multiple opportunities at interacting throughout the day (McGee et al., 1992).

PAGE 15

9 Kohler and Strain (1997) conduc ted an experiment that in itially used naturalistic teaching strategies facilitated by the teacher al one and then included peers in naturalistic teaching interactions to teach skills from the Individual Education Plan (IEP) objectives of a preschooler with autism. After natura listic teaching with the addition of peers was implemented, the target children spent more time interacting with peers, and more time working on the target skills. Findings from this intervention show that the amount of time children with autism spent interacting w ith typically developi ng children increased during intervention, but no meas ures were taken on the quali ty of those interactions (Kohler, & Strain, 1997). Another approach to the use of typically developing peers to work with children with autism is the use of p eer networks (Kamps et al., 1997 ). This approach involves the development of a network of students who ha ve agreed to provide assistance to the student with autism. Kamps at al. (1997) conducted a study that involved training typically developing students to work with ch ildren with autism in a variety of areas throughout the day including the classroom, recess, and lunch. The intervention increased social interactions for all th ree target children across settings. By implementing intervention in numerous setti ngs throughout the day th e target children’s opportunities to interact with peers were greatly increased, and thus improving the probability of these skills generalizing to other settings (Kamps et al., 1997). A unique approach to the promotion of p eer social interactions that involved typically developing peers was the use of group-oriented con tingencies to increase the social interactions between children with au tism and their peers (K ohler et al., 1995). First, all of the children recei ved social skills training on su ggestions for play organizers,

PAGE 16

10 how to provide requests and offe rs for sharing, and how to provide requests and offers for assistance. Children then received prizes if all the members of thei r group engaged in the appropriate social skills. After classwide so cial skills training and the group contingency were implemented all of the subjects exhibite d an increase in social exchanges. When intervention was withdrawn each of the target children’s levels of social interactions decreased to baseline levels, thus the skill s learned during interven tion did not generalize outside of the training setti ng (Kohler et al., 1995). Peer outcomes. A possible concern that some teachers or families could have about peer mediated interventi ons is whether participation in the intervention might have a negative effect on the typically developing student. A few studies have documented the benefits of being involved in peer mediat ed interventions on the typically developing peers (Jones, 2007; Kamps et al., 1998). Kamps and colleagues (1998) measured the positive peer effects of participating in peer mediated interventions. Students were interviewed or completed questionnaires about their experience in peer-involved or peer mediated programs. The results indicated that most of the peers had positive attitudes towards the children with autism, explaining that they learned a lot from bei ng a part of the intervention. So me of the benefits students experienced included improvements in self concept, more tolerance of others, and friendship. The authors also st ated that peers are not only accepting of the children with autism but they are often excited to participate in social activiti es with them (Kamps et al., 1998). Jones (2007) also looked at th e impact of peer mediated interventions on typically developing peers. The author stated that when peers a nd children with autism work

PAGE 17

11 together it creates more effective incl usive schools by improving relationships and improving communication between students. During this intervention peers were instructed to try to play close to the childre n with autism and then to play with them. After the intervention was completed peers were asked questions about being a peer tutor. Each of them enjoyed being a peer tutor a nd half stated that th ey noticed a positive change in themselves as a result of being a pe er tutor. Parents and teachers also reported that they felt that peer tutoring was a good experience for the children (Jones, 2007). In summary, there is substantial evidence that peer mediated interventions are a promising intervention approach for promoting the social interactions of students with autism. In addition, there might be benef its to the typically developing student who participates in the interventi on. In examining over 20 years of research on peer mediated interventions, there is an in creasing emphasis on the design of interventions that are as natural as possible that will yield targeted effects on social interaction skills and the generalization of those skills to nontraining se ttings. In the following section of this paper, the use of Pivotal Response Trai ning (PRT) to promote the acquisition and generalization of social and communication skills is described. This is a naturalistic intervention that generates grea ter generalization of targeted skills to nontrained contexts. In addition, there is some evidence that this approach may be used as an effective peer mediated intervention. Pivotal Response Training Pivotal Response Training (PRT) is an inte rvention that is used with children with autism to address the development of “pivotal behaviors” or behavior s that appear to be central to many areas of functioning. It is believed that once improvements in these

PAGE 18

12 pivotal behaviors occur, then the child s hould also experience positive changes in many other areas of their life. PRT can be implem ented by parents, siblings, peers, teachers, and other individuals w ho interact with children with autism. PRT involves providing clear instructions and questions, interspersi ng maintenance task with new tasks, providing child choice, using direct and natural reinforc ers, and reinforcing at tempts (Koegel et al., 1989). Koegel and colleagues have identified three goals of PRT: 1) to teach the child to be responsive to the many learning opportunities and social interactions that occur in the natural e nvironment, 2) to decrease the need for constant vigilance by an intervention provi der, and 3) to decrease the number of services that remove the child from the natural environment (1999b, p178). Increasing motivation is an important co mponent of PRT. Koegel and colleagues (1999a) suggest several ways to increase the motivation of children with autism, such as providing the child choices in selecting items, topics of conversation or toys, and following the child’s lead dur ing interactions. Another advised method to increase motivation is to use natural reinforcers, or re inforers that are a direct consequence of the behavior, while interacting with the child. Interspersing maintenance tasks with new tasks allows the child to experience succe sses while teaching new skills and is an additional method that Koegel and colleague s advise using to increase motivation. A final technique that they recommend to increas e motivation is to reinforce child attempts at the target behavior. (K oegel et al., 1999a; Koegel, Koegel, Harrower, & Carter, 1999b). Sherer and Schreibman (2005) identify pot ential child characte ristics that would predict the effectiveness of PRT for children with autism. They explained that children

PAGE 19

13 typically respond better to treatment if they have a mode rate or high interest in toys, allow other people to be in close proximity to them, have low or moderate rates of nonverbal self-stimulatory behavior, and have moderate or high rates of verbal stimulatory behaviors. Sher er and Schreibman explain th at children who possessed these characteristics typically have greater impr ovements in language, play behaviors, and social behaviors after pa rticipating in PRT (2005). PRT to teach social and play skills. PRT strategies have been facilitated by adults in order to teach social and play skills to children with autism (Koegel, & Frea, 1993; Stahmer, 1995; Thorp et al., 1995). In a st udy by Koegel and Frea (1993) the authors modified pivotal social behaviors in several children with autism in an attempt to increase their social behaviors. The authors provi ded training on appropria te and inappropriate instances of the target social behavior. Afte r that the children recorded if their target social behavior occurred during an interval of time while th ey had a conversation with the therapist. If they succeeded in engaging in the appropriate behavior they were able to play a video game (Koegel, & Frea, 1993) It is unknown if these newly acquired behaviors generalized to peer s because generalization data were not collected. Using nondisabled peers for the target children to converse with, as opposed to adults, would have clearly added social va lidity to this experiment. Stahmer (1995) implemented an interventi on that taught symbolic play skills to children with autism utilizing PRT strate gies. An adult therapist implemented components of PRT during play sessions; such as using child preferred toys, modeling appropriate actions, reinforcing approximations, and varying play material. Each of the target children displayed an increase in sym bolic play skills after participating in the

PAGE 20

14 intervention (Stahmer, 1995). Although this intervention was successful at teaching symbolic play skills, this study was limited due to the fact that it did not include any measures of generalization with peers, the individuals who th e target children should be playing with. Thorp et al. (1995) examined using PR T to increase sociodramatic play in children with autism. The training of sociodr amatic play was similar to the training of symbolic play in the previous study, and bot h studies based their training on the PRT training manual (Koegel et al ., 1989). Role playing, pers istence, and make believe transformations increased for a ll the target children after in tervention. For this study, like the previous two studies it is unknown if play skills acquired while working with an adult generalized to play with children because data was not taken on this (Thorp et al., 1995). PRT implemented by peers. Several researchers have examined using peers to implement PRT strategies with children with autism (Harper, Symon, & Frea, 2008; Kuhn, Bodkin, Devlin, & Dogget, 2008; Pierce, & Schreibman, 1995; Pierce, & Schreibman, 1997). A benefit of peer implem ented PRT is that it “…may be a ‘looser’ intervention, it provides great er choice to peers in terms of interactions” (McConnel, 2002, p.364). A series of articles by Pierce and Sc hreibman evaluated using typically developing peers to implement PRT with chil dren with autism (Pierce, & Schreibman, 1995; Pierce, & Schreibman, 1997). In their seminal article on peer-implemented PRT Pierce and Schreibman provided training to peers on paying atte ntion, providing child choice, varying toys, modeling appropriate social behavior, reinforcing attempts, encouraging conversation, extending convers ations, turn taking, narrating play, and

PAGE 21

15 teaching responsively to multiple cues (1995). Th e results of this experiment showed that peers were able to effectively implemen t PRT with children with autism. After intervention both target children began to initiate play and co nversation with their peers. One limitation of this study was that it lacked measures to promot e generalization, such as using multiple peers (Pierce, & Schreibman, 1995). The limitation mentioned above was addr essed in Pierce and Schreibman’s follow up study by using three nondisabled peers to im plement PRT with each child with autism (1997). Peer training and inte rvention was implemented the same way as their previous study. After intervention both children displaye d an increase in their ability to initiate and maintain interactions with their peers dur ing play, and both were able to generalize these skills to an untrained peer. In both studies peers im plemented PRT with the target child in a classroom with no other children pr esent. The contrived settings that these interventions were implemented in are a shortcoming of the studies (Pierce, & Schreibman, 1997). Kuhn et al. (2008) taught peers in a special education class to facilitate PRT with children with autism. The researchers trained peers within the target child’s class that included children with mental retarda tion, specific leaning disabilities, and developmental delays. Peer training consiste d of intensive training sessions over several weeks on skills such as paying attention, prov iding child choice, reinforcing attempts, extending conversation, turn ta king, and narrating play. Both target children and peers experienced an increase in social skills af ter participating in the intervention. A Weakness of this study was that no measures of generalization were conducted, therefore it is unknown if these skills generalized to other setti ngs (Kuhn et al., 2008).

PAGE 22

16 A recent study conducted by Harper et al. (2008) addressed some limitations of previous studies by training multiple peers to implement PRT strategies with children with autism in the more natu ral setting of recess. Peers we re trained to gain attention, vary activities, narrate play, reinforce attempts, and take turns with the target children during recess. Both children experienced an increase in social interaction after participating in the study, and these gains maintained during generalization probes. A weak point of this study is that measures of generalization were not taken in any other enviroments than the training environment; th erefore it is unknown if peers used these skills in other locations in addition to the playground (Harpe r et al., 2008). Conclusion Harper et al. (2008) suggested that future research should examine the benefits to nondisabled peers for facilitating PRT with child ren with autism. Di splaying benefits to both peers and children with au tism would present a more socially valid intervention. Kamps and colleagues (1998) suggest that futu re research should be done to find better ways to measure social validity in peer mediated interventions. Strain and Schwartz (2001) warn that so cial skills cannot ade quately be taught in a contrived setting, they need to be taught in the environment that they will be used. Therefore future research should be conducte d that evaluates implementing interventions to improve the social and play skills of ch ildren with autism in more natural locations, such as at recess. Doing so will increase the child with autism’s opportunities to use their newly acquired social and play skills and may increase the probability that they are able to generalize these skills to ot her natural play environments.

PAGE 23

17 The biggest problem with having adults faci litate social and play skills training with children with autism is that it is less natural and it could lead to the child being dependent on adults for social interactions (Weiss, & Harris, 2001). Future research need to be conducted to find the more effective ways to facilitate play and social skills by the people who know the most about play, children.

PAGE 24

18 Chapter Two Method Purpose The purpose of the study was to evalua te peers ability to implement Pivotal Response Training (PRT) and examine the effect s of peer-implemented PRT on the social and communication behavior of a child with autism. This study was approved by the Institutional Review Board (IRB) at the University of South Florida and the IRB approved consent forms were used to enro ll participants. The study extended the literature on PRT in the following ways: 1) by conducting all traini ng sessions in the environment where the skills will be used; 2) by including measures of generalization in untrained enviroments; 3) by including soci al validity measures of the benefits of intervention to the target child as well as the peers and; 4) by implementing the intervention in a Waldorf school. Specifically, the study addressed the following research questions: 1. When provided with training on the use of Pivotal Respon se Training with a target child with autism, will third grade peers learn the techniques and apply them during recess, and will these skills generalize to an untrained environment? 2. When peers implement Pivotal Response Trai ning with a target child with autism, will that child exhibit increases in play and communication behaviors?

PAGE 25

19 Participants The participants in this study included an eight year old Caucasian girl diagnosed with mild to moderate autism, and three ty pically developing classmates who attended a private school. The target child, Eve, was nominated for the study due to her delays in social skills and lack of interac tion with her third grade peers. Eve was diagnosed with an autism spectrum disorder at age four by a developmental pediatrician not associated with this study. She was in itially enrolled in a public school special education class for students with varyin g disabilities from prekindergarten through first grade where she received speech therapy and physical therapy. At the age of four Eve began receiving speech, occupational, and physical therapies from a private therapy center and c ontinued to do so during the time that this study was conducted. At age seven, Eve was en rolled at a private community Waldorf school where she was fully included. During the study, Eve was in a third grade class of 14 students taught by 1 primary teacher and seve ral specials teachers for music, Spanish, German, and handwork. She also received ad ditional academic support from a classroom assistant. At the beginning of this study the assistant worked with Eve the entire school day, but during the course of the study the am ount of time the assistant worked with Eve was reduced to half the school day. Eve spoke in simple sentences, and used language primarily to gain access to items and activities. Eve was working slightly below grade level in the areas of math and handwriting but was above grade level in he r reading ability. She engaged in some immediate and delayed echolalia. Socially, Ev e rarely initiated interactions with her

PAGE 26

20 peers. During recess she typically wandere d around alone, dug in the dirt, and climbed trees. Three typically developing peers, Jamie, Ashley, and Lily, who are classmates of Eve, were selected to par ticipate in the study. Peers we re selected by the criteria suggested by Odom and Strain (1984). Peer s eligible to participate in the study were ones that a) attended school regularly, b) interacted with peers in an age appropriate manner, c) were capable of complying with th e researchers instructi ons, d) were able to imitate the behavior of the trainer, and e) had the ability to concentrat e on the task at hand for the entire intervention session (Odem, & Strain, 1984). Peers were recruited by asking the teacher which students displayed an interest in interacting with Eve. The teacher nominated three female classmates who were Caucasian females between the ages of eight and nine years old, and had b een in Eve’s class for a period ranging from several months to a year a nd several months prior the beginning of the study. Once identified, the researcher aske d peers if they wanted to be part of a study to help Eve learn to play. All children expressed an in terest in participating in the study. The researcher then provided an IRB approved c onsent form to the peers’ parents by the teacher to explain the study and request permi ssion for the children to participate. Signed consent forms were obtained for all particip ating children prior to the initiation of baseline data collection. Setting and Materials The study took place in a private commun ity school that used the Waldorf curriculum. The Waldorf curriculum is base d on the pedagogical principles of Rudolf Steiner and provides an education to meet th e needs of each child’s capabilities (Suncoast

PAGE 27

21 Waldorf School, 2008). In Wal dorf programs, children are pr ovided with ample time to play and engage in movement activities through the day. Play materials that are commonly used in a Waldorf program are b ean bags, wooden toys, scarves, and cloth dolls. During outdoor play, children are encour aged to engage in pretend play, organize their own games, and can climb trees and fr eely explore the environment. Children in Waldorf schools typically are educated with the same peers and teacher from first through eighth grade (Why Waldorf Works, 2009). The study took place outdoors and within th e third grade classroom on the school grounds. Training, baseline, post-training/in tervention, and follow-up sessions were conducted in a large meadow behind the sc hool (59.94 x 26.42 m) where children played during recess. The meadow contained two r ope swings, shovels, a wagon, and trees that the children were able to climb. Due to th e large size of the mead ow, the experiment was conducted in a restricted area of the meadow th at was designated for the play procedures. Generalization probes occurred in the thir d grade classroom (8.38 x 4.57m) within the class of fourteen students. The classroom contained desks, chairs, books, art materials, and Waldorf materials such as bean ba gs, wooden toys, and bees wax. A training manual was used to train peers. This manual was adapted from similar training materials used by Pierce and Schreibman (n.d.) to train typically-developing peers to use PRT procedures with child ren who have autism. The manual included written and pictorial explanations of PRT st rategies that were written at a level of understanding that was appropriate for early elementary age children. A video camera was used during baseline, intervention, genera lization, and follow-up sessions to record the behavior of the target child and peers. A video camera was also used to record the

PAGE 28

22 instructions and interactions that occurre d with each peer du ring training sessions. Several materials and toys were used during baseline, training, posttraining/ intervention and follow-up sessions, including: gnome dolls, wooden animals, tree blocks, and balls. These are materials that were in the classr oom and also available for use during outdoor play. Design A concurrent multiple probe baseline across peers design was used to assess each peer’s ability to implement PR T strategies with a child with autism and the benefits of intervention to the target child (Kazdin, 1982). Baseline data on the types of interactions that each peer had with the target child we re gathered unt il levels of these behaviors stabilized. Once baseline levels were stable or showed a decreasing trend, peers received individual training on one PRT t echnique at a time until all six strategies were trained. After the first peer completed training a nd displayed improvement over baseline during post-training/intervention sessions, the next pe er was trained and so on until all of the peers were trained. Data on the peer’s use of the PRT intervention and behavior of the target child in interaction with the peer were gathered after pe ers were trained. Procedures Baseline Before peers were trained on PRT strategies ba seline data were collected during recess. During baseline each peer was asked one at a time to play in a designated area of the meadow, but the peer was not given any prompt s to play with the target child. Data collection began when a set of toys were pr esented to the children and the researcher announced “It is time to play” and ended af ter seven minutes. The toy set included:

PAGE 29

23 gnome dolls, wooden animals, tree blocks, a nd balls. Data were collected on peer interactions and the target child’s play a nd communication behaviors until the data were stable or a decreasing trend was evident indicat ing that social inter actions were becoming less frequent over time (see Appendix A for p eer data collection sh eet; see Appendix B for target child data collection sheet). Peer PRT training Peer training was based on the manual “Kids Helping Kids: Teaching Typical Children to Enhance the Play and Social Skill s of Their Friends with Autism and Other PDD’s” by Pierce and Schreibman (n.d.) with permission from the authors. The manual was adapted from its original version in the following ways: the strategy “use developmentally appropriate language ‘eas y sentences’” was removed as an actual strategy, and instead this con cept was stressed through all peer training sessions. The strategy “ increasing obs ervational learning ‘tell what you are doing’” was also removed as an actual strategy and in stead narration of play was st ressed while teaching “modeling appropriate and complex play skills ‘show good play’”. The adapted manual was used to provide peers with training on the implem entation of the following PRT strategies: 1.) Orient attention “Paying attention”gain Eve’s attention before attempting to give a direction. 2.) Enhance motivation by offeri ng choices “Give choices”provide Eve with choices between items and activities. 3.) Model appropriate and complex play skills “Show good playing”comment on and narrate play (e.g., saying “playing cars is fun” while racing toy cars).

PAGE 30

24 4.) Encourage conversation “Ask your friend to talk”get Eve to ask for desired items and activities (e.g., requiring Eve to say “I want the ball” before giving her the ball). 5.) Teach taking turns “Take turns”take turns with desire d items and activities. 6.) Reinforce appropriate social behavior “Good, nice try”providing praise to Eve for engaging in socially appropria te behavior (e.g., while playing catch, saying “nice throw”). Peer training occurred in the meadow during four, 30-minute sessions over the course of two weeks. During the first se ssion the PRT strategy “o rient attention” was taught. During the second session the st rategies “enhance motivation by offering choices” and “model appropriate and complex play skills” we re taught. During the third session the PRT strategies “encourage conversati on” and “teach taking turns” were taught. During the final training session the strategy “reinforce approp riate social behavior” was taught and all six PRT strategies were reviewed. The same set of toys that were available during baseline, post-training/intervention, and follow-up sessions were used during training. Peers were presented with a ma nual that included written and pictorial explanations of all six PRT strategies. E ach strategy was taught one at a time; the researcher first explained the strategy to th e peer and then modeled it with the peer, having the peer pretend to be the target chil d while the researcher pretended to be the peer to model the incorrect and correct use of each strategy. The peer was then prompted to demonstrate the incorrect and correct us e of the strategy with the researcher, and feedback was provided on the peer’s perfor mance in the form of praise for correct behaviors and corrective feedback for problem areas. At the end of the session the peer

PAGE 31

25 was asked questions about each of the stra tegies to make sure that she thoroughly understood the concepts and an opportunity was provided to the peer to ask any questions that they may have. On the final day of tr aining, after all of the PRT strategies were taught, a role play checklist was used to co llect data on peers’ use of PRT strategies during three role plays. Peer s were trained until they met the training criteria of being able to perform all six strategies properly duri ng three separate role plays. The role play checklist displayed that Jamie, Ashley, a nd Lily were able to use each of the PRT strategies during three co nsecutive role plays. Jamie was the first peer to participate in PRT training; once she received training on all PRT strategies to criteria she was then able to use the strategies with Eve during re cess (i.e., post-training/in tervention sessions). These same procedures were repeated with As hley, and then Lily in a multiple baseline format. Post-training / Intervention Once a peer was able to perform all si x PRT strategies properly during three separate role plays, the child was asked to us e those skills and play with the target child during recess. Post-training /intervention sessions began wh en the researcher announced “It is time to play.” Peers then used PR T strategies while playing with Eve for seven minutes. Each session consisted of one of the peers playing with Eve. For posttraining/intervention sessions, similar to ba seline sessions, the child ren were required to play in a designated area of the meadow. Da ta on intervention sessions were collected two to three times a week during recess, and la sted seven minutes for each peer. Due to the length of recess, Eve onl y participated in one or two post-training/intervention sessions with peers per recess.

PAGE 32

26 Booster training session If a peer did not show a marked impr ovement over baseline during the first three post-training/intervention sessions, a booste r training session wa s provided. During booster training the peer receiv ed an additional 30 minute tr aining session that consisted of reviewing and role playing all of the PRT strategies and the peer received feedback on their performance. If a peer was required to participate in booster training, on subsequent post-training sessions they al so received post-session praise and corrective feedback on their performance until the peer moved into follow-up. Due to the fact that Lily’s use of PRT strategies during her fi rst three post-training sessions decreased to baseline levels she received a booster training session, post-session feedback, and the opportunity to earn reinforc ing items for using PRT strategies during post-training sessions. During th e booster training session Lily was informed that if she used the PRT strategies while playing with Eve she would receive a small item. Items included markers, erasers, lip gloss, and playing cards. After each post-training session Lily received feedback about her performance and was able to choose an item if she had used the PRT strategies during the session. When Lily moved into follow-up sessions she did receiving post-session feedback or the opportunity to earn reinforcers. Generalization Generalization probes were conducted in the untrained environment of the classroom during indoor free play to see if p eers generalized their us e of PRT strategies to other environments. Several generaliza tion probes were taken during baseline, posttraining/intervention, and follow-up. For measures of generalization peers were not told “It is time to play” and the basket of toys were not introduced. Each peer was asked one

PAGE 33

27 at a time to play with the ta rget child. Data collection bega n after indoor free play started and peers were asked to play with the ta rget child and the se ssion lasted for seven minutes. During generalization probes no pos t-session feedback was provided, even if the peer had participated in booster trai ning and was receiving post-session feedback during post-training/intervention sessions at the time the generalization probe was conducted. Follow-up Follow-up sessions occurred af ter a peer met the criteri on of using PRT strategies for at least 50% of intervals during a post-training/intervention session for three consecutive sessions. Follow-up sessions were identic al to baseline and posttraining/intervention sessions for peers who did not receive a booste r training session. For peers who received booster training, pos t-session feedback was no longer provided during follow-up sessions. Measures Dependent variables Baseline, post-training/in tervention, generalization, a nd follow-up sessions were video recorded and analyzed to assess each peer’s ability to impl ement PRT strategies with the target child and the play and comm unication behavior of the target child. The main dependent variable was the pee r’s ability to use each PRT strategy: paying attention, offering choices, modeli ng appropriate and complex play skills, encouraging conversations, taki ng turns, and reinforcing appr opriate social behavior. Seven minute recess sessions were analy zed in 15 s interv als, using partial interval recording. The number of intervals that a peer us ed any of the PRT strategies

PAGE 34

28 was divided by the total number of interv als, 28, and multiplied by 100 to produce a percentage of intervals that the behavior occu rred. For example; if the peer used any of the PRT strategies during 15 of the 28 interval s then she did so for 54% of the session. Video taped seven minutes sessions we re analyzed for occurrences of the following behaviors. 1.) Orient attention “Paying attention” occurred when the peer gained Eve’s attention before attempting to give a direction. 2.) Enhance motivation by offeri ng choices “Give choices” occurred when the peer provided Eve with choices between items and activities. 3.) Modeling appropriate and complex pl ay skills “Show good playing” occurred when the peer commented on and narrated her own play (e.g., saying “playing cars is fun” while racing toy cars). 4.) Encouraging conversation “Ask your friend to talk” occurred when the peer required Eve to ask for desired items and activities (e.g., requiring Eve to say “I want the ball” before giving her the ball). 5.) Teaching taking turns “Take turns” occurred when the peer took turns with desired items and activities. 6.) Reinforcing appropriate social behavior “Good, nice try” occurred when the peer provided praise to Eve for engagi ng in socially appropriate behavior (e.g., while playing catch, saying “nice throw”). In addition to measuring the peer’s us e of PRT strategies, the study included measures of Eve’s play and communication beha viors. Definitions of solitary play and play with peer trainers were derived from Wolfberg and Schuler’s (1993) definition of

PAGE 35

29 isolate play and play with a common focus. The definition of parallel play was derived from Rubin’s (1989) definition of parallel play. The seven minute videotaped sessions we re analyzed in 15 s intervals, using partial interval recording to determine the pe rcentage of intervals that the target child engaged in play and communication behavior. Th ese sessions were analyzed to determine the occurrences of the following behaviors as defined by the operational definitions below: Solitary play Eve plays independently. During solitary play she may appear to be oblivious or unaware of others (e.g., Eve sits away from peers and digs a hole in the dirt). Parallel PlayEve plays beside or in the comp any of the peer, but does not play with the peer. She uses toys that are simila r to those that the peer trainer is using. Eve may seem to be somewhat aware of and attentive to the peer, but does not interact with her (e.g., Eve and the peer bot h play with blocks near each other but Eve does not speak to, or engage with the peer). Play with peer trainers Eve engages in activities involving the peer. This includes turn taking, giving and receivi ng assistance, and active sharing of materials. Playing with peers involves a common focus or attention on the play (e.g., Eve manipulates a gnome doll and hands it to the peer). Communication behaviors were measured accordi ng to the operational definitions below: Communication to peer utterances and vocalizati ons emitted by Eve that are directed to a peer. This includes singing songs with th e peer trainer, answering questions posed by the peer trainer, and any other utterance or vocalization not

PAGE 36

30 directed to self in the form of vocal s timulation (e.g., Eve says “Ball please” when asked by a peer “What do you want?”, or she sings the same song as the peer.) Communication to self utterances and vocalizations emitted by Eve that are directed to herself in the form of vocal stimulation. This includes singing songs to herself, repeating lines from tele vision shows and movies, and mumbling to herself (e.g. Eve makes the sound “eeeeee”, or she mumbles songs lyrics to herself.) Inter-observer agreement Inter-observer agreement (IOA) was calcula ted for 80% of baseline, post-training/ intervention, generalization, and follow up sessions. To conduct IOA two observers independently viewed videos of peers interac ting with Eve for occurrences of the target behaviors for peers and Eve. An agreement of the occurrence of a target behavior was defined as both observers recording that a target behavior occurred within the same interval. A disagreement ove r the occurrence of a target behavior was defined as one observer recording the occurrence of a targ et behavior and the other observer not recording an occurrence of a target behavior within the same interval. Inter-observer reliability was calculated by dividing the number of agreements by the number of agreements plus disagreements and multiplying by 100 Peer’s target behaviors. IOA was calculated for the occurrence of each of the PRT skills. IOA for orient attention wa s an average of 89.63% (range 64 to 100%) enhance motivation by offering choices was a mean of 99.05% (range 89 to 100%), model appropriate and complex play skills was an average of 97% (range 75 to 100%), encouraging conversation was a mean of 97.73% (range 79 to 100%), teaching taking

PAGE 37

31 turns was an average of 97.85% (range 82 to 100%), and reinforcing appropriate social behavior was a mean of 99.43% (range 89 to 100%). IOA for the nonoccurrence of any of the peers’ target behavior was an average of 88.65% (range 71 to 100%). Eve’s target behaviors. IOA for the occurrence of solitary play was an average of 82.85% (range 61 to 100%), parallel play was a mean of 76.87% (range 57 to 100%), and play with peer trainers was an averag e of 86.33% (range 61 to 100%). IOA for the occurrence of communication to self was a mean of 86.56% (range 61 to 100%), and communication to peer trainer was an average of 96.38% (range 82 to 100%). Data collectors Data collectors were trained prior to coding any of the research study videos by the researcher. Training involved reviewing the operational definitions for each measure and the procedures that would be used to c ode the videos and ensu re independent scoring during the session. The research er then explained the data collection sheets and how to record data on peers and Ev e’s target behaviors. Video segments were developed to be used for training data collectors on the dependent measures. In these video segments, a 10 year old male child was asked to play with his twin three year old sisters and use the PRT strategies. These two segments were shown to the data collectors to prepare them to reliably record data on the peer’s target behaviors. Fidelity of peer training To asses the degree to which the compone nts of training were implemented with integrity, data were collected on training fidelity. All of the peer training sessions were video recorded. These video recordings were reviewed by observers who completed a

PAGE 38

32 check list of the relevant components of trai ning. Observers used th e checklist to see if the training session included: an introduction of the PRT component, explanation of the PRT component, modeling the PRT component role playing the PRT component with the peer, providing the peer feedback on th eir performance, asking the peer questions about the PRT component, and asking the peer if they have any questions. The checklist conformed that the researcher used all n ecessary components of training during each training session with all three peers. IOA for measures of training fidelity was conducted for 25% of the training sessions. IOA was measured by having two observers independently view the video recorded sessions and record if each component of the session occurred during training. An agreement of the occurrence of a traini ng component was defined as both observers recording that a training com ponent occurred. A disagreement over the occurrence of a training component was defined as one observe r recording the occurr ence of a training component and the other observe r not recording an occurren ce of a training component. Inter-observer reliability was calculated by dividing the number of agreements by the number of agreements plus disagreements and multiplying by 100. IOA for measures of training fidelity was 100%. Social Validity Peers. Social validity was measured through interviews with peers about their experiences with participating in the in tervention after the study was completed. Interviews were administered by the teacher to each peer individually in order to avoid the possibility of a peer influencing the re sponding of the other peers (see Appendix C for

PAGE 39

33 peer interview questions). The interview wa s based on the peer que stionnaire presented in Jones (2007). Teacher. After the study was completed, the t eacher was given a questionnaire to complete that posed questions about her views of the interven tion procedures and outcomes. The teacher answered questions about her perception of the interactions between the target child and each of the nondi sabled peers (see Appendix D for teacher questionnaire).

PAGE 40

34 Chapter 3 Results Figure 1 presents data on the use of PRT strategies by peers within play sessions. During baseline sessions peers only used a few of the PRT strategies on a few occasions, after receiving training on the strategies they increased the frequency of, as well as the variety of, strategies used. During follow-up sessions one peer’s use of PRT strategies maintained above criterion level and another pe er’s use of the strate gies decreased below post-training/intervention sessions but remained above baseline sessions. The data show evidence that peers generalized their use of PRT strategies to the untrained environment of the classroom to an extent. Peers’ use of PRT strategies Jamie was the first peer to participate in PRT traini ng. In baseline sessions Jamie’s use of PRT strategies occurred for an average of 5.5% of intervals (range 0 to 18%) across four sessions. In post-training sessions she used PRT strategies for an average of 45% (range 25 to 57%) of inte rvals across seven sessions. During follow-up sessions she used the strategies for a mean of 32% (range 21 to 46%) of intervals across three sessions. During generalization probes in the classroom Jamie used PRT strategies an average of 2% of inte rvals during baseline, 29% of intervals during post-training, and a mean of 45% intervals during follow-up.

PAGE 41

35 Ashley was the next peer to participate in PRT training. She used PRT strategies for an average of 15.3% (range 4 to 25%) of intervals across four sessions in baseline. During post-training sessions with Eve she us ed PRT strategies for an average of 59.7% (range 54 to 71%) of intervals across three sessions. When Ashley moved into follow-up sessions she used PRT strategies for a mean of 64.33% (range 54 to 71%) of intervals across six sessions. During generalization probe s Ashley used PRT strategies an average of 2% of intervals during baseline and a mean of 41% of intervals during follow-up. Due to the fact that Ashley moved from post-training to follow-up in only three sessions, time constraints did not allow for a generalization probe to be cond ucted during post-training. Lily was the final peer to participate in PRT training. During baseline sessions Lily used the strategies for a mean of 16.5% (range 7 to 32%) of intervals across six sessions. In post-training se ssions she used PRT strategi es for an average of 44.5% (range 7 to 71%) of intervals across six sessions. In one follow-up session Lily used the PRT strategies for 82% of intervals. Lily used PRT strategies during generalization probes an average of 4% of interval s during baseline, and a mean of 18% of intervals during post-training. Due to time constraints no generaliz ation probes were conducted between Lily and Eve during post-training af ter Lily received a boos ter training session or during follow-up.

PAGE 42

36 Figure 1. Percentage of intervals that Peer s (Jamie: top, Ashley: middle, Lily: bottom) implemented PRT with the Eve.

PAGE 43

37 Eve’s Play Behaviors Figure 2 depicts the play behaviors of the target child during baseline, intervention, and follow-up sessions with peers. Table 1 displays th e play behaviors of the target child duri ng generalization probes. During baseline sessions with all three peers Eve engaged in high levels of solitary play and low levels of play with peer trainers. With two of the three peers Eve experienced a decreasing trend in levels of solitary play and an increasing trend in levels of play w ith peer trainers as sessions progressed. The positive effects of intervention were also seen when peers used PRT strategies in the classroom to an extent. Solitary play Eve engaged in solitary play during ba seline sessions with Jamie for an average of 58.75% (range 36 to 71%) of intervals, during interv ention sessions for a mean of 58.71% (range 25 to 86%) of intervals, a nd during follow-up sessions for an average of 21.67% (range 0 to 36%) of intervals. In generaliza tion probes Eve engaged in solitary play with Jamie for an average of 100% of baseline intervals, 93% of intervals during intervention and a mean of 51.5% of intervals in follow-up. During baseline sessions with Ashley, Eve e ngaged in solitary play for an average of 41.25%, (range 29 to 71%) of intervals, an average of 44% (range 14 to 64%) of intervals while in intervention session, and in follow-up sessions for a mean of 22.5% (range 7 to 32%) of intervals. While participatin g in generalization probes with Ashley, Eve engaged in solitary pl ay for an average of 66% of intervals during baseline and a mean of 25% of intervals during follow-up.

PAGE 44

38 Eve engaged in solitary play with Lily during baseline se ssions for an average of 59.67% (range 11 to 96%) of intervals, duri ng intervention sessions for a mean of 25.5% (range 0 to 64%) of intervals, and in one follow-up session for 4% of intervals. During generalization probes with Lily Eve enga ged in solitary play for a mean of 75% of intervals during baseline and an average of 89% of in tervals during intervention. Parallel play Eve engaged in parallel play with Ja mie for an average of 31.25% (range 7 to 43%) of intervals, for a mean of 38.14% (range 18 to 57%) of intervals during intervention, and for an average of 30% (range 4 to 54%) of intervals during follow-up sessions. In generalization probe s Eve engaged in parallel play with Jamie for an average of 2% of intervals during baseline, 14% of in tervals during interven tion, and a mean of 32% of intervals during follow-up. During baseline sessions Eve engaged in para llel play with Ashley for an average of 41.75% (range 7 to 75%) of intervals, in interven tion sessions she did so for an average of 43% (range 11 to 79%) of intervals, and during follow-up for a mean of 41% (range 18 to 71%) of intervals. While participati ng in generalization probes Eve engaged in parallel play with Ashley for a mean of 27% of intervals du ring baseline, and an average of 41% of intervals during follow-up. Eve engaged in parallel play and with L ily during baseline se ssion for a mean of 25.67% (range 0 to 50%) of intervals, for an average of 33.67% (range 14 to 64%) of intervals in intervention sessions, and for 50% of intervals while in one follow-up session. During generalization probes Eve engaged in para llel play with Lily for an average of 7% of intervals during baseline and a mean of 10.5% of intervals during intervention.

PAGE 45

39 Play with peer trainer Eve engaged in play with peer trainers with Jamie for an average of 20.75% (range 0 to 36%) of intervals in baseline session, during intervention sessions she did so for a mean of 22.57% (range 0 to 54%) of intervals, and while in follow-up sessions for an average of 59.33% (range 39 to 96%) of intervals. In generalization probes with Jamie, Eve engaged in play with peer tr ainers for a mean of 0% of intervals during baseline, 0% of intervals during interv ention, and an average 30% of intervals during follow-up. While in baseline sessions Eve engaged in pl ay with peer trainers with Ashley for a mean of 22.75% (range 7 to 68%) of intervals, for an average of 27.7% (range 11 to 43%) of intervals in intervention sessi ons, and for an average of 50% (range 39 to 61%) of intervals while in follow-up sessions. Du ring generalization probes with Ashley, Eve engaged in play with peer trainers for an average of 21.5% of in tervals during baseline and a mean of 52% of intervals during follow-up. Eve engaged in play with peer trainers in baseline sessions with Lily for an average of 31% (range 7 to 79%) of intervals, during intervention for a mean of 53.5% (range 11 to 89%) of intervals, and in one follow-up session for 54% of intervals. During generalization probes with Ashley, Eve engaged in play with peer trainers for an average of 23% of intervals during baseline and a mean of 14% of intervals during intervention.

PAGE 46

40 Figure 2. Percentage of intervals that Eve e ngaged in play behaviors with peers (Jamie: top, Ashley: middle, Lily: bottom).

PAGE 47

41 Table 1: Eve’s play behavior during generalization probes Table 1. Percentage of intervals that Eve e ngaged in play behavi ors with peers during generalization probes (Jamie: top, Ashley: middle, Lily: bottom). Peer Phase Session # Solitary play Parallel play Play with peer 5 100 0 0 Baseline 6 100 4 0 Intervention 11 93 14 0 18 57 39 21 Jamie Follow-up 22 46 25 39 5 43 54 25 Baseline 10 89 0 18 18 46 68 18 Ashley Follow-up 26 4 14 86 5 100 0 0 Baseline 10 50 14 46 22 96 0 14 Lily Intervention 23 82 21 14

PAGE 48

42 Eve’s communication behaviors Figure 3 provides data on Eve’s communi cation behavior with each peer during baseline, intervention, and follow-up sessions During baseline Eve engaged in low levels of communication to peers and high levels of communication to self. During intervention and follow-up sessions Eve engage d in higher levels of communication to peers and slightly lower leve ls of communication to self. Communication to self During baseline sessions with Jaime, Ev e engaged in communication to self for a mean of 38.5% (range 18 to 79%) of intervals, during in tervention sessions for an average of 28.71% (range 4 to 54%) of intervals, and during follow-up for an average of 11.67% (range 0 to 21%) of intervals. During genera lization probes with Jamie, Eve engaged in communication to self for a mean of 25% of intervals dur ing baseline, 46% of intervals during intervention, and an averag e of 30.5% of intervals during follow-up. Eve engaged in communication to self duri ng baseline sessions w ith Ashley for an average of 39.25% (range 21 to 61%) of intervals, while participating in intervention session for an average of 18% (range 4 to 43%) of intervals, and for a mean of 7.83% (range 0 to 18%) of intervals in follow-up session s. During generalization probes with Ashley, Eve engaged in communication to self for an average of 59% of intervals during baseline and a mean of 48% of intervals during follow-up. During baseline sessions with Lily, Eve e ngaged in communication to self for an average of 33.67% (range 7 to 46%) of intervals, for a m ean of 6.83% (range 0 to 11%) of intervals in intervention sessions, a nd for 11% of intervals during one follow-up session. While participating in generalization probes with Lily, Eve engaged in

PAGE 49

43 communication to self for an average of 30% of intervals during baseline and for a mean of 41% of intervals during intervention. Communication to peer Eve engaged in communication to Jamie dur ing baseline sessions an average of 0 % of intervals, during intervention sessions for an average of 12.71% (range 4 to 25%) of intervals, and for a mean of 23.67% (ra nge 21 to 29%) of intervals in follow-up sessions. During generalization probes Eve engaged in communication to Jamie for an average of 0% of intervals during baseline, 14% of intervals du ring intervention, and a mean of 27% of intervals during follow-up sessions. While participating in baseline sessio ns, Eve engaged in communication to Ashley for an average of 5.5% (range 4 to 7 %) of intervals, during intervention sessions for an average of 38.3% (range 25 to 54%) of intervals, and for a mean of 29.83% (range 11 to 39%) of intervals while in follow-up sessions. Du ring generalization probes Eve engaged in communication to Ashley fo r an average of 2% of intervals during baseline and a mean of 11% of intervals during follow-up. During baseline sessions Eve engaged with communication to Lily for a mean of 2% (range 0 to 4%) of intervals, while in intervention for an aver age of 17.83% (range 0 to 54%) of intervals, and during one follo w-up session for 39% of intervals. While participating in generalization probes Eve engaged in communication to Lily for an average of 5.5% of intervals during baseline and a mean of 12.5% of intervals during intervention.

PAGE 50

44 Figure 3. Percentage of intervals that Eve engaged in communication behaviors with peers (Jamie: top, Ashley: middle, Lily: bottom).

PAGE 51

45 Social Validity Peers After the study was completed the classroom teacher administered an interview to Jamie and Ashley about their participation in the study. When asked how much they liked PRT, both peers said “I liked it a lot.” When asked if there was anything that they really liked about PRT, Jamie reported that she liked when th e researcher pretended to be the target child during role pl ays in training sessions and Ashl ey stated that she enjoyed getting to spend time with Eve. When asked how hard it was to do PRT, both peers said “It was ok, not too hard.” When asked what they thought was the hardest part of PRT, the peers reported that it was hard to play with Eve when she was “grumpy” or “did not listen”. When asked if there were any part of PRT that they did not like, both peers said no. When asked what they learned from doing PRT, they said they learned more about Eve and how to get her attention. When aske d if and how they thought PRT helped them, Jamie replied that it helped he r learn more about Eve, and As hley said that it helped her see what Eve likes. When asked if they t hought that other kids should learn to do PRT, both peers said “yes.” An interview was not c onducted with Lily due to time constraints. Teacher The classroom teacher filled out a questionnaire about the use of PRT strategies by students in her class. When asked how mu ch of an improvement in interactions she had seen between Eve and Jamie she reporte d she had seen “a large improvement.” When asked how much of an improvement in interactions she had seen between Eve and Ashley she stated that she had also seen “a large improvement.” When asked how much of an improvement in communication she had seen between Eve and the peers after peers

PAGE 52

46 were trained to use PRT strategies she reported that she had observed “a large improvement.” When asked if she would sugge st that other teachers have their typically developing students trained to use PRT strategies to work with children with autism she said “yes.” The classroom teacher was not as ked about the level of interactions that she observed between Lily and Eve due to time constraints.

PAGE 53

47 Chapter 4 Discussion Results from this study provide evidence that elementary school aged students can learn PRT strategies and use them within free play settings with a child with autism. The data presented in this study shows that th e training procedures alone were effective enough to teach one peer PRT strategies, the training procedures in addition to one booster training session were necessary to t each another peer PRT strategies, and the training procedure in addition to the booster training session and the opportunity to earn reinforcing items was necessary to teach anothe r peer the PRT strategies. Interviews with peers who had completed the study revealed that they enjoyed learning and using PRT strategies with a child with autism. During sessions that peers us ed PRT strategies the target child experienced an increasing trend in play with peer trai ners, a decreasing trend in levels of solitary play and levels of parallel play remained similar to baseline levels with two of the three peers. After peers began using PRT stra tegies, the child with autism displayed a slight decrease in ve rbal behavior to self and an increase in speech to peers. While participating in baseline sessions, Ashley and Lily showed some use of PRT strategies with Eve. It is important to note that during baseline sessions peers generally engaged in the PRT strategy “teaching taking turns” by throwing a ball back and forth, and “orienting attention” by saying Eve’s name be fore delivering a direction.

PAGE 54

48 On only one occasion during baseline sessions did a peer use any other PRT strategy, this occurred when Ashley used “enhance motivation by offering choices”. During post-training sessions prior to receiving a boos ter training session Jamie used PRT strategies during an average of 35.7% of intervals, but after Jamie received the booster training session her use of PRT strate gies increased to an average of 52.75% of intervals. She reached criterion to move in to follow-up sessions within four post-training sessions. Ashley did not require a booster tr aining session and reached criterion to move into follow-up in only three sessions. As hley used PRT strategies during a higher percentage of intervals than the other peers, using PRT strategies for as high as 71% of intervals during some session. She also us ed multiple PRT strategies during a single interval more often than th e other peers. For example Ashley frequently used PRT strategies that easily worked together, su ch as first using th e PRT strategy “orient attention,” then “enhance motivation by offering choices,” followed by “encourage conversation,” and then “reinf orcing appropriate social beha vior,” all within a 15 s interval. During post-training sessions before Lily received a booste r training session she use PRT strategies during an average of 23.67% of intervals, afte r she received the booster training she use PRT stra tegies during post-training sessions for an average of 65.33% of intervals and she reached criterion to move into follow-up in three sessions. During each of the three post-training sessions that Lily had an opportunity to earn a reinforcing item, she did so. Because no ge neralization probes were conducted between Lily and Eve during post-training after Lily received a booster trai ning session or during follow-up, it is unknown if she would have used PRT strategies for a higher percentage of intervals during generalization probes as she did during recess session s. While all peers

PAGE 55

49 also used PRT strategies during generalizati on probes in the classroom with Eve, Jamie and Ashley only used the st rategies above criterion leve l once during follow-up. An important factor to consider when evaluating data collected during baseline sessions is that the structure of baseline sessions created a pl ay situation that was slightly contrived compared to typical recesses, re sulting in Eve and her peer’s level of interactions to be higher duri ng baseline sessions compared to pre-baseline recesses. Eve’s teacher and classroom assistant reporte d that during recesses prior to baseline sessions, Eve typically played independently and dug in the dirt or climbed trees alone for the entire recess period. Once baseline sess ions began, while peers were not directly instructed to interact with Eve, they were asked to play in the same area as her and provided with a set of toys. These were toys that children could access independently if desired, but were not readily accessible (i.e. provided by an adult fo r the purpose of play) during recess, with the excepti on of a ball. Eve’s increase d proximity to peers during baseline sessions within the de signated play space resulted in higher levels of interactions than typical recesses. The average percentage of intervals that Eve engaged in solitary play remained the same from baseline to intervention sessi ons with Jamie and incr eased slightly with Ashley. The average percentage of intervals th at Eve engaged in soli tary play with Lily decreased slightly during intervention sessi ons before Lily receive d the booster training session but decreased much more afterwards. The average percentage of intervals that Eve engaged in solitary play decreased duri ng follow-up sessions with all three peers. The researcher noted that peer s behavior differed during inte rvals recorded as solitary play during baseline and posttraining/intervention sessions. In baseline sessions while

PAGE 56

50 Eve engaged in solitary play, peers typically also engaged in solitary play. During posttraining/intervention sessions when Eve enga ged in solitary pla y, peers were usually attempting to use the PRT strategy “model appr opriate and complex play skills.” Data collectors reported that as intervention sessi ons progressed Eve began to engage in more appropriate play with toys duri ng intervals recorded as parall el play and play with peer trainers, and also began imitati ng the play of her peers. For example, peers sometimes used one gnome doll to repres ent a mother and another sma ller gnome doll to represent a baby, by putting the smaller gnome doll in th e larger gnome doll’s arms, and without prompting from peers Eve also be gan engaging in this behavior. During sessions with both Jamie and Ashley the average percentage of intervals that Eve engaged in parallel pl ay increased during intervention sessions, but then returned to baseline levels during follow-up sessions. With both Jamie and Ashley the mean percentage of intervals that Eve engaged in play with peer trainers increased slightly during intervention sessions, but increased more during follow-up sessions. During sessions with Lily, the average percentage of intervals that Eve enga ged in parallel play and play with peer trainers increased dur ing intervention sessions and remained high during one follow-up session. It is noteworthy that the type of pl ay activities differed during intervals that Eve engage d in play with peer trainers during baseline sessions and intervention sessions. For example, while in baseline sessions, play with peer trainers were typically recorded when Eve and the peer threw a ball back and forth or when the peer pushed Eve on a swing. After peers receiv ed training on PRT strategies play with peer trainers were often recorded when Eve and the peer appropriately played with the same toys, and took turns with materials.

PAGE 57

51 The most notable area of Eve’s target be haviors that changed after peers began using PRT strategies during recess were her co mmunication behaviors. With all peers the average percentage of intervals that Eve engaged in communication to self decreased slightly during intervention sessions compared to baseline sessions. During follow-up sessions with Jamie, Ashley, and Lily this slight decreasing trend continued. While in intervention sessions with all pe ers the average percentage of intervals that Eve engaged in communication to peers increased. During fo llow-up sessions with Jamie and Lily this increase continued. Eve’s communication to peers decreased during follow-up sessions with Ashley compared to intervention sessions, but remained well above baseline sessions. During intervention sessions Eve e ngaged in communication to peers in order to gain access to desired items and activiti es, to answer questi ons, and to make novel demands. For instance, during a session with Ashley, Eve told her peer to “make happy face” and then smiled at Ashley. Similar levels of Eve’s target behavior s occurred in generalization probes. Although Eve engaged in higher levels of co mmunication to self during generalization probes compared to recess sessions. A potential explanation for this event is that because peers were trained to use PRT strategies with a specific set of toys, and only some of these toys were available in the classroo m, they may have experienced difficulties generalizing these strategies to other toys available in the classroom, and thus had problems engaging Eve in play. Future stud ies should include a wider variety of toys during training sessions. Due to the fact that generalization m easures were not conducted between peers and novel children with autism it is unknown if peers in the pres ent study would have

PAGE 58

52 generalized the use of PRT strategies to play interactions with other children and thus is a limitation of this study. Future research s hould include measures of generalization that involve novel children with autism. In this study, two peers received both a booster training session and post-session feedback when they failed to improve over baseline sessions. The booster training session alone or the post-session feedback al one might have been enough support to help peers use PRT strategies more effectively w ith Eve during recess, a nd therefore the other component of training may be unnecessary. Future research s hould be conducted to evaluate the benefits of in cluding post-session feedback alone compared to a booster training session alone to analyze the be nefits of both training components. Future research should be conducted to examine utilizing typically developing peers to facilitate PRT strategies with child ren with autism in other locations, such as afterschool programs, in home play dates, and summer camps. Children are typically provided extended durations of free play time in these locations and therefore peers would potentially have more opport unities to use PRT strategies with a child with autism than peers in the present study. Another potential area of research could evaluate teaching PRT strategies to siblings of childre n with autism and meas uring their abilities to use those strategies at ho me and in the community with their sibling with autism. This study provides data that indicate the PRT training procedure is effective in changing the behavior of peers toward th eir classmate with autism during free play situations. Moreover, these changes in peer interactions resulted in changes in the communication and peer play behavior of a ch ild with autism. While this experiment was conducted in only one classroom, that uses a program curriculum that is not

PAGE 59

53 commonly implemented in elementary schools, the results of the study suggest that PRT might be an effective procedure for promo ting peer interactions within inclusive programs. It is also encouraging to note that in this study, changes in peer interactions during free play resulted in positive changes in the communication behavior of the child with autism. In future studies, it would be of interest to examine the effects of teaching all students in the classroom the use of PRT strategies to promote the social and communication behavior of their classmates with disabilities.

PAGE 60

54 References American Psychiatric Association. (1994). Di agnostic and statistical manual of mental disorders (4th ed.). Washington, D.C: American Psychiatric Press. Bass, J.D., & Mulick, J.A. (2007). Social pl ay skill enhancement of children with autism using peers and sibli ngs as therapists. Psychology in the school 44, 727-735. Cowan, R., J., & Allen, D.D. (2007). Using na turalistic procedures to enhance learning in individuals with autism: A focus on generalized teaching within the school setting. Psychology in the Schools 44, 701-715. Cowan, R., J., & Allen, D.D. (2007). Using na turalistic procedures to enhance learning in individuals with autism: A focus on generalized teaching within the school setting. Psychology in the Schools 44, 701-715. DiSalvo, C.A., & Oswald, D.P. (2002). Peer-med iated interventions to increase the social interaction of children with autism: Consideration of peer expectancies Focus on Autism and Other Developmental Disabilities, 17, 198 207. Goldstein, H., & Cisar, C. (1992). Promoti ng interaction during sociodramatic play: Teaching scripts to typical preschooler s and classmates with disabilities. Journal of Applied Behavior Analysis 25, 265-280. Goldstein, H., Kaczmarek, L., Pennington, R ., & Shafer, K. (1992). Peer-mediated intervention: Attending to, commenting on, and acknowledging the behavior of preschoolers with autism. Journal on Applied Behavior Analysis 25, 289-305.

PAGE 61

55 Harper, C.B., Symon, J.B.G., & Frea, W.D. (2008). Recess is time-in: Using peers to improve social skills of children with autism. Journal of Autism and Developmental Disorders 38, 815-826. Jones, V. (2007). ‘I felt like I did someth ing good’-the impact on mainstream pupils of a peer tutoring programme for children with autism. British Journal of Special Education 34, 3-9. Kazdin, A. E. (1982). Single-case research designs: Methods for clinical and applied settings. Oxford, NY: Oxford University Press. Kamps, D., Gonzalez-Lopez, A., Potucek, J., Kravits, T., Kemmerer, K., & GarrisonHarrell, L. (1998). What do peers think? So cial validity of integrated programs. Education & Treatment of Children 21, 107-134 Kamps, D.M., Potueck, J., Lopez, A.G., Krav its, T., & Kemmerer, K. (1997). The use of peer networks across multiple settings to improve social interactions for students with autism. Journal of Behavioral Education 3, 335-357. Koegel, R.L., & Frea, W.D. (1993). Treatments of social behavior in autism through the modification of pivot al social skills. Journal of Applied Behavior Analysis 26, 369-377. Koegel, R.L., Koegel, L.K., & Carter, M.C. (1999a). Pivotal teaching interactions for children with autism. School Psychology Review 28, 576-594. Koegel, L.K., Koegel, R.L., Harrower, J.K ., & Carter, C.M. (1999b). Pivotal responses intervention I: Overview of approach. Journal of the Association for Persons with Sever Handicaps 24, 186-198.

PAGE 62

56 Koegel, R.L., Schreibman, L., Good, A., Cern iglia, L., Murphy, C., & Koegel, L.K. (1989). How to teach pivotal behaviors to children with autism: A training manual. Santa Barbara: Universi ty of California. Kohler, F.W., & Strain, P.S. (1997). Merg ing naturalistic teac hing and peer-based strategies to address the IEP objectives of preschoolers with autism: An examination of structural a nd child behavior outcomes. Focus on Autism & Other Developmental Disabilities 12, 196-207. Kohler, F.W., Strain, P.S., Hoyson, M., Davi s, L., Donina, W.M., & Rapp, N. (1995). Using group-oriented continge ncy to increase social interactions between children with autism and their peers. Behavior Modification 19, 10-32. Kuhn, L.R., Bodkin, A.E., Devlin, S.D., & Dogge t, R.A. (2008). Using pivotal response training with peers in special education to facilitate play in two children with autism. Education and Training in Developmental Disabilities 43, 37-45. Laushey, K., & Heflin, L.J. (2000). Enhancing so cial skills of kinde rgarten children with autism through the training of multiple peers as tutors. Journal of Autism and Developmental Disorders 30, 183-193. McConnel, S.R. (2002). Interventions to faci litate social inter action for young children with autism: Review of available resear ch and recommendations for educational intervention and future research. Journal of Autism and Developmental Disorders 32, 51-372. McGee, G.G., Almeida, C., Sulzer-Azaro ff, B., & Feildman, R.S. (1992). Promoting reciprocal interactions vi a peer incidental teaching. Journal of Applied Behavior Analysis 25, 117-126.

PAGE 63

57 Mitchell, D. (n.d.). Why waldorf works; everything you need to know about waldorf education. Retrieved January 24, 2009 from http://www.whywaldorfworks.org/ 02_W_Education/grades1_8.asp Morrison, L., Kamps, D., Garcia, J., & Parker D. (2001). Peer mediation and monitoring strategies to improve initiations and so cial skills for students with autism. Journal of Positive Behavior Interventions 3, 237-250. Odem, S.L., & Strain, P.S. (1984). Peer-med iated approaches to prompting children’s social interaction: A review. American Journal of Orthopschiarty 54, 544-557. Owen-DeSchryver, Carr, E.G., Cale, S.I., & Blakeley-Smith, A. (2008). Promoting social interactions betwee n students with autism sp ectrum disorders and their peers in inclusive school settings. Focus on Autism and Other Developmental Disabilities 23, 15-28. Petursdottir, A, McComas, J., McMasters, K., & Horner, K. (2007). The effects of scripted peer tutoring and programming comm on stimuli on social interactions of a student with autism spectrum disorder. Journal of Applied Behavior Analysis, 40, 353-357. Pierce, K., & Schreibman, L. (1995). Increas ing complex social behaviors in children with autism: effects of peer-imple mented pivotal response training. Journal of Applied Behavior Analysis 28, 285-295. Pierce, K., & Schreibman, L. (1997). Multiple peer use of pivotal response training to increase social behaviors of classmates with autism: Results from trained and untrained peers. Journal of Applied Behavior Analysis 30, 157-160.

PAGE 64

58 Pierce, K., & Schreibman L. (n.d.) Kids helping kids: Teaching typical children to enhance the play and social skills of their friends w ith pervasive developmental disorders. Rubin, K.H. (1989). The play observation scal e. Waterloo, Ontario: University of Waterloo Sherer, M.R., & Schreibman, L. (2005). Indi vidual behavior profiles and predictors of treatment effectiveness for children with autism. Journal of Consulting and Psychology 73, 525-538. Stahmer, A.C. (1995). Teaching symbolic play skills to children with autism using pivotal response training. Journal of Autism and Developmental Disorders 25, 123-141. Strain, P.S. (1983). Generalization of autistic children’s social beha vior change: Effects of developmentally integrat ed and segregated settings. Analysis and Intervention in Developmental Disabilities 3, 23-34. Strain, P.S., Kerr, M.M., & Ragland, E.U. (1979). Effects of peer-mediated socialinitiations and prompting/reinforcement procedures on the social behavior of autistic children. Journal of Autism and Developmental Disorders 9, 41-54. Strain, P., & Schwartz, J. (2001). ABA a nd the development of meaningful social relations for young children with autism. Focus of Autism and Developmental Disabilities 16, 120-128. Strain, P.S., Shores, R.E.,& Timm, M.A. (1977) Effects of peer social initiations on the behavior of withdraw n preschool children. Journal on Applied Behavior Analysis 10, 289-298

PAGE 65

59 Suncoast Waldorf School (2008). Retrieved September 13, 2008 from http://www.suncoastwaldorf.org/ Thiemann, K.S., & Goldstein, H. (2004). Effects of peer training and written text cueing on social communication of school-aged ch ildren with pervasive developmental disorder. Journal of Speech, Language and Hearing Research 47, 126-144. Thorp, D.M., Stahmer, A.C., & Schreibman, L. (1995). Effects of sociodramatic play training on children with autism. Journal of Autism and Developmental Disorders 25, 265-282. Weiss, M.J., & Harris, S.L. (2001). Teachi ng social skills to people with autism. Behavior Modification 25, 785-802. Wolfberg, P.J., & Schuler, A.L. (1993). Integr ated play groups: A model for promoting the social and cognitive dimensions of play in children with autism Journal of Autism and Developmental Disorders 23, 467-489.

PAGE 66

60 Appendices

PAGE 67

61 Appendix A: Peer Data Collection Sheet Session date:_______ Peer initials:_____ Data recorder name:_____________ 15 s 30s 45 s 60s 1 O C M Ta Tu R+ none O C M Ta Tu R+ none O C M Ta Tu R+ none O C M Ta Tu R+ none 2 O C M Ta Tu R+ none O C M Ta Tu R+ none O C M Ta Tu R+ none O C M Ta Tu R+ none 3 O C M Ta Tu R+ none O C M Ta Tu R+ none O C M Ta Tu R+ none O C M Ta Tu R+ none 4 O C M Ta Tu R+ none O C M Ta Tu R+ none O C M Ta Tu R+ none O C M Ta Tu R+ none 5 O C M Ta Tu R+ none O C M Ta Tu R+ none O C M Ta Tu R+ none O C M Ta Tu R+ none 6 O C M Ta Tu R+ none O C M Ta Tu R+ none O C M Ta Tu R+ none O C M Ta Tu R+ none 7 O C M Ta Tu R+ none O C M Ta Tu R+ none O C M Ta Tu R+ none O C M Ta Tu R+ none 1.) O = Orient attention “Paying attention”: occurs when the peer gains the target child attention before attempting to give a direction. 2.) C = Enhance motivation by offering choices “Give choices” occurs when the peer provides the target child with choices between items and activities. 3.) M= Modeling appropriate and complex pl ay skills “Show good playing” occurs when the peer comments on and na rrates his or her ow n play (e.g., saying “playing cars is fun” while racing toy cars). 4.) Ta= Encouraging conversation “Ask your friend to talk” occurs when the peer gets the target child to ask for desire d items and activities (e.g., requiring the target child to say “I want the ba ll” before giving her the ball). 5.) Tu= Teaching taking turns “Take turns” occurs when the peer takes turns with desired items and activities. 6.) R+= Reinforcing appropriate soci al behavior “Good, nice try” occurs when the peer provides praise to the target child for engaging in socially appropriate behavior (e.g., while playing catch, saying “nice throw”). 7.) None= no occurrence of any of the six PRT strategies _______% of intervals that peer used a PRT strategy

PAGE 68

62 Appendix B: Target Child Data Collection Sheet Session date: _____ Peer initial :_____ Data recorder name:__________________ Play: ____% of intervals w ith solitary play (S) ____% of intervals with parallel play (P) ____% of intervals with pl ay with peer trainer (T) Communication: ____% of intervals with co mmunication to self (Cs) ____% of intervals with communication to peer (Cp) 15s 30s 45s 60s 1 S P T Cs Cp S P T Cs Cp S P T Cs Cp S P T Cs Cp 2 S P T Cs Cp S P T Cs Cp S P T Cs Cp S P T Cs Cp 3 S P T Cs Cp S P T Cs Cp S P T Cs Cp S P T Cs Cp 4 S P T Cs Cp S P T Cs Cp S P T Cs Cp S P T Cs Cp 5 S P T Cs Cp S P T Cs Cp S P T Cs Cp S P T Cs Cp 6 S P T Cs Cp S P T Cs Cp S P T Cs Cp S P T Cs Cp 7 S P T Cs Cp S P T Cs Cp S P T Cs Cp S P T Cs Cp

PAGE 69

63 Appendix C: Peer Interview Qu estions 1.) How much did you like “It is time to play”? a.) I liked it a lot. b.) I liked it. c.) I liked it a little bit d.) I did not like it. 2.) Is there anything that you really liked about “It is time to play”? If so, then what? 3.) How hard was it to do “It is time to play”? a.) It was really hard. b.) It was hard. c.) It was ok, not too hard. d.) It was not hard at all, it was easy. 4.) What do you think the hardest part wa s about “It is time to play”? 5.) Were there any parts of “It is time to pl ay” that you did not like? If so, what? 6.) What did you learn from doing “It is time to play”? 7.) Has doing “It is time to play” helped you in any way? If so, then how? 8.) Do you think that other kids should l earn to do “It is time to play”?

PAGE 70

64 Appendix D: Teacher Questionnaire 1.) How much of an improvement in interac tions have you seen between the target child and peer 1 after peer 1 began using PRT? a.) A large improvement b.) Some improvement c.) No improvement d.) Less interactions then before 2.) How much of an improvement in interac tions have you seen between the target child and peer 2 after peer 2 began using PRT? a.) A large improvement b.) Some improvement c.) No improvement d.) Less interactions then before 3.) How much of an improvement in comm unication have you seen in the target child after peers started using PRT? a.) A large improvement b.) Some improvement c.) No improvement d.) Less communication then before 4.) Would you suggest that other teachers have their typically developing students trained on PRT strategies to work with children with autism? a.) Yes b.) No