Diane J. Catellier, Dr. P.H.
University of North Carolina at Chapel Hill
Both individual and environmental factors contribute to an individual's propensity to engage in or refrain from health risk behaviors. What these factors are, how they interact, and for whom, are all questions of importance in understanding health risk behaviors and behavior change. Of particular interest are biological, physiological, psychological, genetic, and social/environmental factors that influence the initiation and maintenance of adequate exercise and good dietary practices that in turn influence weight outcomes and ultimately an individual?s vulnerability to poor health outcomes. Understanding the mechanisms of behavior change is necessary for developing effective interventions and to avoid wasting both time and resources on interventions that target factors that are not in the causal pathway leading to healthy lifestyle behaviors.
Interventions for behavior change are typically based on theoretical models (e.g., social cognitive theory, the ecological model) that postulate that interventions influence behavior by changing constructs, such as self-efficacy and social support, that are believed to be important for behavior change. To test whether the hypothesized causal pathways and mechanisms indeed influence behavior change, appropriate statistical methods must be used. Statistical procedures include structural equation modeling (including path analysis), traditional regression modeling, and hierarchical modeling. If the theoretical model involves latent constructs, structural equation modeling provides the basic data analysis strategy; if the model involves only measured variables, regression analysis or hierarchical modeling should be used. Regardless of the analytic method used, the criteria described by Baron and Kenny (1986) should be used to conclusively establish which factors are on the causal pathway.
- Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. J Pers Soc Psychol 1986;51:1173-82.