This study examined the effects of a dependent group contingency to increase positive verbal interactions among nine middle school-aged males in a self-contained classroom. Prior to implementing the group contingency, the students received social skills instruction. A withdrawal experimental design was then used to evaluate the effects of the dependent group contingency. Students' positive verbal interactions increased and negative verbal interactions decreased with the implementation of the dependent group contingency. When the group contingency was faded, positive and negative interaction levels were maintained. The results are discussed in terms of developing positive interaction skills among youth with emotional disturbance and the application of dependent group contingencies with these students. (Contains 2 tables and 9 figures.)