Problematic behaviors exhibited by young children have become a topic of concern. Head Start staff members are reporting an increase in the number of children displaying challenging and disruptive behaviors (Piotrkowski, Collins, Knitzer, & Robinson, 1994). Despite the increase in classroom behavior problems, teachers reported deficiencies in managing these behaviors. In fact, behavior management strategies comprised 3 of the top 5 areas in which Head Start staff requested additional training (Buscemi, Bennett, Thomas, & DeLuca, 1995). Therefore, it is important to provide teachers with effective strategies to manage classroom behaviors. Comparing Response Cost and Token Economy Procedures Research has supported the effectiveness of both response cost and token economy procedures in decreasing disruptive behavior in academic settings. However, both procedures employ different strategies to manage classroom behavioral problems. Thus, it is important to examine research comparing the relative effectiveness of these two procedures. Iwata and Bailey (1974) used a reversal design to compare reward and cost token systems with special education children in elementary school. The children’s behavior was observed every morning for three months during math class. Results indicated that both programs were equally effective in decreasing off-task behavior and the violation of the teacher’s rules. Mean percent off-task behavior returned to a level similar to baseline during the reversal phase. The number of mathematic problems completed more than doubled for the group earning tokens while the cost group only showed a small increase in problems completed. The teacher provided more statements of approval to students who were in the reward-cost group than to children who were in the cost-reward group. Sullivan and O’Leary (1990) also compared the effectiveness of both response cost and token economy procedures. Children were observed daily for 40 minutes during reading and language and math class. Both programs were highly and equally effective in reducing the amount of off-task behavior. More specifically, average percent on-task behavior in the baseline condition was approximately 60%, but then increased to approximately 85% when both programs were in effect. However, rates of on-task behavior differed between the two programs when the programs were faded from use. When the response cost program was faded, improvements in on-task behavior were maintained for all students participating in this program. However, half of the children who participated in the reward program did not maintain treatment effects when the program was faded from the classroom.