Behavior Analysis, Physical Activity and Individuals with Disabilities: A Meta-Analysis

Background/Purpose: It is well known that physical activity can aide in the prevention of obesity, diabetes, and cardiovascular diseases. However, people with physical and/or intellectual disabilities may be at greater risk of acquiring such health problems than non-disabled individuals because they tend to be less active (Berktas et al., 2011). It is important, therefore, to identify teaching techniques that are effective with individuals with disabilities. Many studies based in Applied Behavior Analysis (ABA), a scientific perspective of human behavior that focuses on variables extrinsic to individuals as the main influences on their behavior, have improved the physical activity of people with disabilities as well as positively influenced social interactions within physical activity settings. Despite the successes of ABA-based interventions with individuals with disabilities reported within singular studies, the overall effect of these interventions is unknown. Therefore, the primary purpose of this study was to use meta-analytic techniques to examine this overall effect. The secondary purpose was to assess the influence moderator variables (gender, age, disability type, intervention type, skill type, and target behavior) had on the overall effect. 

Method: To identify studies for inclusion, several electronic databases were searched using related key words. Once studies were identified, they had to meet three criteria to be included in the meta-analysis: (1) utilized single subject design, (2) included an ABA-based intervention, and (3) included a dependent variable that measured behavior of individuals with disabilities in a physical activity setting. Once articles were identified for inclusion, moderator variable information was coded and numeric data were extracted to calculate effect sizes (ES).

Analysis/Results: The standardized mean difference formula (SMD) was used to calculate each ES_SMD (Olive & Smith, 2005). Using a random effects model, the overall weighted mean ES and moderator analyses were conducted using Comprehensive Meta-Analysis software. The ES_SMD was 1.743 (95% CI = 1.533, 1.953), suggesting ABA interventions had a very large influence on the behavior of individuals with disabilities in physical activity settings. Four moderator variables, gender, age, disability type, and skill type, had a significant influence (p < .05) on the overall effect.

Conclusions: The findings of this study suggest that ABA-based interventions are very effective in improving motor and social behaviors of individuals with disabilities, regardless of the type of ABA technique used. Adapted physical educators can be confident in the usefulness of ABA, but should also be mindful of individual differences when implementing these teaching techniques.