Reasoned Action As an Approach to Understanding and Predicting Health Message Outcomes
The reasoned action approach is a behavioral theory that has been developed since the 1960s in a sequence of reformulations. It comprises the theory of reasoned action; the theory of planned behavior; the integrative model of behavioral prediction; and its current formulation, the reasoned action approach to explaining and changing behavior. Applied to health messages, reasoned action theory proposes a behavioral process that can be described in terms of four parts. First, together with a multitude of other potential sources, health messages are a source of beliefs about outcomes of a particular health behavior, about the extent of social support for performing that behavior from specific other people, and about factors that may hamper or facilitate engaging in the behavior. Second, these beliefs inform attitude toward performing the behavior, perceptions of normative influence, and perceptions of control with respect to performing the behavior. Third, attitude, perceived norms, and perceived control inform the intention to perform the behavior. Fourth, people will act on their intention if they have the required skills to do so and if there are no environmental obstacles that impede behavioral performance. The theory’s conceptual perspective on beliefs as the foundation of behavior offers a theoretical understanding of the role of health messages in behavior change. The theory also can be used as a practical tool for identifying those beliefs that may be most promising to address in health messages, which makes the theory useful for those designing health message interventions. Reasoned action theory is one of the most widely used theories in health behavior research and health intervention design, yet is not without its critics. Some critiques appear to be misconceptions, such as the incorrect contention that reasoned action theory is a theory of rational, deliberative decision making. Others are justified, such as the concern that the theory does not generate testable hypotheses about when which variable is most likely to predict a particular behavior.