Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Donald A. Williamson


Two interdependent group-oriented contingency (g-o-c) systems to improve academic behavior were compared to a no treatment control group. Additionally, the two interdependent group-oriented contingency systems were compared to each other regarding their effects on several collateral behaviors (classroom behavior, social interactions and peer sociometric ratings). Six fourth and fifth grade classes (126 students) were randomly assigned to three experimental conditions. The first treatment group (positive g-o-c) was an interdependent group-oriented contingency system which reinforced overall science performance at or better than a level of 90%. The second treatment group (negative g-o-c) was also an interdependent group-oriented contingency system which discouraged students from allowing their classroom science average to fall below 90%. The third experimental condition was a no treatment control group. Dependent variables were measured during Baseline and Phase-1 and Phase-2 treatments. The study was conducted over a six-week period (each phase spanned two weeks). The results of this study showed that although the academic performance for the interdependent g-o-cs was higher than the control group during treatment, this pattern of performance was also present during Baseline. Classroom behavior was not affected by the treatments; yet social interactions and peer sociometric ratings differed. During the treatment phases, the treatment groups demonstrated improved positive and neutral social interactions. The treatment groups also received lower sociometric ratings during Phase-1, but this pattern did not continue into Phase-2. Teacher and student acceptance ratings supported the use of interdependent group-oriented contingencies to improve academic performance. Student acceptance ratings were more favorable and social interactions were more often positive under the positive g-o-c conditions. The acceptance data also highlighted subtle perceptual differences between students and teachers relative to the two treatments. The major findings were: (1) academic performance was confounded by Baseline group differences; (2) classroom behavior was not affected by the treatments; (3) under the group-oriented contingencies positive and neutral peer social interactions improved; (4) peer sociometric ratings did not permanently decrease as a function of the group-oriented contingencies; and (5) teachers and students rated the interventions acceptable, with student responses and social behavior reflecting a small advantage for the positive g-o-c.