Socially competent children and youth use a variety of strategies to initiate and maintain social interactions, and manage conflict (Sheridan, 1995). In contrast, socially incompetent children engage in behaviors that can lead them to become increasingly withdrawn and in some instances aggressive (Coie, 1990). One factor that is critical to social skill learning is the degree to which students have opportunities to learn appropriate social skills. Physical education classes provide students opportunities to improve social behaviors. However, studies have shown students are not very supportive of their classmates during lessons (e.g., Giebink & McKenzie 1985; Vidoni, 2004). Group contingencies are a commonly used strategy to reducing misbehavior in schools but have been seldom been used to teach social behaviors. This study investigated the effects of a dependent group contingency on the supportive and unsupportive fair play behaviors of 6th grade students engaged in volleyball games as part of their regular physical education classes during 23 lessons. Six students, one male and one female from three classes taught by the same teacher, identified as demonstrating low supportive behaviors participated in this study. A multiple baseline design across classes was used to examine the data. The intervention consisted of: (a) an explanation of supportive behaviors explanation, (b) examples and non-examples were provided, (c) public posting with daily goals set and if they were achieved and (d) teacher prompts. Dependent variables consisted of supportive, and unsupportive behaviors. Subcategories of behaviors were: (a) physical, (b) gestural, and (c) verbal. An event-recording sheet was used to observe occurrences of behaviors during 10-minute games. Interobserver agreement for 50% classes was 97%. Baseline data showed few incidences of supportive behaviors among the six students. Following the social skill intervention students increased their rate of behaviors significantly. Follow-up data during a sand volleyball activity showed that although students were able to perform supportive behaviors in rates higher than baseline, students did not maintain the similar level as they had during the intervention. Results showed that dependent group-oriented contingency was effective in improving students’ supportive behaviors during volleyball games. This study extends the literature in two ways. First, it continues the line of inquiry examining the effects of group-oriented contingencies. Second, the findings in this study demonstrated fair play behaviors can be specifically defined and intervened upon.