Background: Many educators assume that students develop appropriate social skills as a by-product of participation in physical education and sports (Hellison 2003). However, it has been demonstrated that appropriate social behaviours improve when interventions are implemented (Balderson and Sharpe 2005). It is known that students with disabilities, especially those with emotional or behavioural disorders, typically do not exhibit appropriate social behaviors (Gresham, Sugai, and Horner 2001). Purpose: The purpose of this study was to examine the influence of social skills instruction on the sport-and game-related behaviours of students with emotional or behavioural disorders. The study specifically addressed the instruction's influence on appropriate and inappropriate behaviours within both physical education and recess (break time) settings. Participants and setting: Six students (two females and four males) ages 10 through 17 and attending alternative education school programs designed to serve children and youth with emotional or behavioural disabilities participated in the study. Research design and intervention: A multiple baseline across participants design was used. A validated Appropriate Sport and Game Behaviours Curriculum modeled on the one used by Moore, Cartledge, and Heckman (1995) was implemented. The curriculum consisted of 15 lesson plans focusing on appropriate behaviours. Examples of appropriate behaviours consisted of respecting one's own equipment and that of others, congratulating the winner, avoiding blaming teammates, following rules, working cooperatively, and avoiding criticizing the loser. Instructional sessions took place three times per week and were between 20 and 25 minutes in duration. Data collection and analysis: Data were gathered from videotaped sessions for both physical education and recess settings. Observers were trained to code the dependent variables. In addition, interobserver reliability data were collected during randomly selected sessions for all participants across both settings. Procedural integrity was established to ensure that the intervention was implemented as intended. Visual analysis of behaviours was performed using graphical representation of data. Data were subjected to a trend analysis. Mean increases and decreases of appropriate and inappropriate behaviours were presented. Results and recommendations: In general, results were mixed. Five of the six participants (83%) were able to increase appropriate sports/games behaviours and decrease inappropriate behaviours in a physical education (acquisition) setting, and three of the six participants (50%) in a recess (generalisation) setting. While appropriate behaviours were above and inappropriate behaviours were below baseline levels, only one participant (17%) was able to maintain intervention levels. It is recommended that the intervention be extended over a longer period of time.