Date of Award


Degree Type


Degree Name

Doctor of Philosophy


School Psychology

Major Professor

Christopher H. Skinner

Committee Members

David F. Cihak, Merilee McCurdy, Tara C. Moore


Group contingencies are often used in the classroom to decrease disruptive behavior and improve both academic performance and on-task behavior. However, research is mixed regarding the most appropriate target for intervention: academic performance, on-task behavior, or disruptive behavior. The current study was designed to extend research on group contingency target behaviors by evaluating and comparing two interdependent group contingencies: an academic performance contingency and an on-task behavior contingency. Dependent variables included percent correct, percent on-task, and disruptive behavior. The participants were seven, eighth-grade students with Emotional Behavior Disorder. Data were collected in the social studies classroom. An adapted alternating treatments design was used with on-task behavior and academic performance contingencies serving as the treatments. Acceptability data were collected to determine if students and the teacher preferred one contingency over another.

Data were collected during independent seatwork, and participants were told whether it was an on-task contingency day or an academic performance contingency day. If the class average percent correct or percent on-task met a randomly selected criterion, all class members earned access to a randomly selected group reward. Visual analysis and effect size data showed that the on-task contingency produced large increases in class average percent correct and percent on-task. The academic performance contingency had small effects on class average percent correct and percent on-task. Neither contingency affected disruptive behavior.

Survey and interview data reveal that the classroom teacher and the students found both interventions acceptable. Neither contingency was preferred over the other. Theoretical and applied implications of the findings, study limitations, and directions for future research are discussed.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."