Date of Award


Document Type


Degree Name

Doctor of Education (Ed.D.)



First Advisor

Robert MacMillian

First Committee Member

Judith Van Hoorn

Second Committee Member

Kenneth L. Beauchamp

Third Committee Member

Robert D. Morrow

Fourth Committee Member

Maria Hefferman


The purpose of the study was to compare the use of peer administered contingent reinforcement for social interaction with the use of play materials that facilitate interaction. The subsequent effects on the social interaction of mainstreamed preschool handicapped children with their nonhandicapped peers was examined. Three nonhandicapped kindergarten children were paired with three moderately mentally handicapped preschool children and trained to initiate play. A single-subject, alternating treatment design with a withdrawal phase was used to compare the effects of the two treatments. Generalization immediately following each treatment was also examined as was maintenance over time. Observers used an interval recording procedure and showed a mean interobserver reliability rating of 95%. All observations were conducted in an outdoor playground setting. The use of play materials that facilitate interaction (Treatment C) was shown to be a significantly more effective method for increasing social interaction than was the use of peer administered contingent reinforcement (Treatment B). The mean child-child interaction total for Treatment C was 71% while the mean child-child interaction total for Treatment B was 27%. The t value at a.10 level of probability was $-7.74$. Generalization immediately following treatment was greater after Treatment C (mean 24%) than after Treatment B (mean 7%). The t value of $-1.98$ did not, however, show a significant difference in generalization between the two treatments. Treatment C was implemented as the only treatment upon completion of the alternating treatment phase. Relatively little generalization occurred during the withdrawal phase (mean 17%) and the treatment effects were not maintained over time (mean 7%). A supplemental analysis of the relationship between play attempts by the peer "helper" and the number of actual interactions showed that, while there were a greater number of play attempts during Treatment C than during Treatment B, the difference was not large enough to account for the success of Treatment C. The outcome of the study helps to ascertain that the use of trained nonhandicapped peer "helpers" coupled with the use of play materials that facilitate interaction can be an effective means of increasing social interaction between young handicapped and nonhandicapped children.



Included in

Education Commons