Behavior Analysis and Social Behavior During Physical Activity: A Meta-Analysis

Background/Purpose: Disruptive behaviors cause problems at all levels of teaching by wasting learning time (Oruc, 2011) and increasing teacher stress (Repp, Deitz, & Deitz, 1976). Effective strategies for managing behavior are needed for successful learning environments. While research has found that techniques based in Applied Behavior Analysis (ABA), a branch of psychology concerned with developing technology for improving behavior (Cooper, Heron, & Heward, 2007), are effective in many physical activity (PA) settings for improving social behaviors, the overall effect of these interventions is unknown. The purpose of this study is to use meta-analysis techniques to examine the overall effect of ABA-based interventions on social behaviors in PA settings and the influence of 11 moderator variables.

Method: Six databases, reference lists of included studies, and websites of several known ABA journals were searched using related keywords in one or more combination to collect studies. From the 45 studies collected, 29 met the inclusion criteria for the analysis. From these, 250 effect sizes were calculated using the standardized mean difference formula (ES_SMD). The mean ES and the 95% CI were calculated using a random effects model. Analyses of the 11 moderating variables were also conducted.

Analysis/Results: The overall ES_SMD was 1.50 (95% CI = 1.30, 1.70), indicating ABA-based interventions had a very large, positive effect on improving social behaviors within PA settings. Moderator analysis revealed that that eight of the moderators significantly (p < 0.05) impacted the overall mean effect size: gender (Qb = 6.26, df = 2), target behavior (Qb = 3.89, df = 1), age (Qb = 10.97, df = 3), social behavior (Qb = 9.78, df = 4), management strategy (Qb = 69.49, df = 5), setting (Qb = 15.12, df = 3, 0.01), motor skill (Qb = 23.30, df = 4), and type of reinforcer (Qb = 14.20, df = 4). Three moderators, intervention level (Qb = 0.28, df = 2, p = 0.87), developmental level (Qb = 1.39, df = 1, p = 0.24), and group size (Qb = 6.63, df = 3, p = 0.09), did not influence the overall mean ES_SMD.

Conclusions: Results indicate that ABA-based interventions have a very strong, positive effect on improving social behaviors in PA settings regardless of the age, developmental level of participants, setting of its implementation, or social behavior targeted. ABA-based interventions can be effective solutions for instructors looking to improve the behavior of their students.