An alternating treatments design was used to evaluate and compare the effects of two interdependent group contingencies on the academic performance, on-task behavior, and disruptive behavior of eighth-grade students in a social studies class. All students were enrolled in a self-contained alternative school for students with behavior problems. Delivering rewards contingent upon participants' average percent correct enhanced on-task behavior and percent correct on independent seatwork assignments; however, delivering rewards contingent upon participants' on-task behavior yielded more consistent and larger increases in percent correct and on-task behavior. Neither group contingency resulted in consistent or meaningful changes in disruptive behavior. Theoretical and applied implications related to direct and indirect effects of interdependent group rewards are discussed along with directions for future research.