Influences of perceived autonomy support on physical activity within the theory of planned behavior

Three studies tested the contribution of perceived autonomy support to the prediction of health-related intentions within the theory of planned behavior. Perceived autonomy support refers to the extent to which individuals perceive that significant others encourage choice and participation in decisionmaking, provide a meaningful rationale, minimize pressure, and acknowledge the individual’s feelings and perspectives. Findings from Studies 1 and 2 demonstrated that perceived autonomy support predicted intentions to participate in physical activity behavior directly and indirectly via attitudes. Perceived autonomy support predicted intention even after statistically controlling for the effects from past behavior, descriptive norms, and perceived social support. Study 3 found that persuasive communications influenced perceptions of autonomy support, attitudes, and intentions. Overall, the findings support the incorporation of perceived autonomy support into the theory of planned behavior. Copyright # 2006 John Wiley & Sons, Ltd. Despite the considerable amount of research aimed at understanding the low levels of adherence to health behaviors, social psychologists confess to having limited knowledge of the mechanisms responsible for compliance (Haynes, McDonald, Garg, & Montague, 2003). Thus, there is still a need for theoretically guided research that furthers researchers’ and practitioners’ grasp of the nature of adherence to health behaviors. The theory of planned behavior is a social cognitive model of decision-making that provides a useful framework for predicting and explaining health behavior (Ajzen, 1991). THEORY OF PLANNED BEHAVIOR According to the theory of planned behavior, optimal prediction of behavior can be attained from a person’s intention, which is an indicator of how hard people are willing to try, and how much effort s, School of Psychology, University of Plymouth, England, UK. s, Ltd. Received 1 November 2005 Accepted 17 August 2006 Autonomy support and planned behavior 935 people plan to exert toward performance of behavior (Ajzen, 1991). Intention is in turn function of attitudes (a person’s positive or negative evaluation of performing the behavior), subjective norms (an individual’s belief that significant others may exert pressure on them to execute the behavior), and perceived behavioral control (the extent to which people believe that they can control performance of social behavior). According to Ajzen, the construct of perceived behavioral control can also predict behavior directly when behavior is not under complete volitional control and when perceived behavioral control is accurate in a sense that it reflects actual control. Assumptions underlying the theory of planned behavior have been corroborated by numerous meta-analytic reviews across a number of different social behaviors (Armitage & Conner, 2001) and health behaviors (Hagger, Chatzisarantis, & Biddle, 2002). Indeed, tests of the theory of planned behavior have provided strong evidence for the overall predictive validity of intentions, and have shown that attitudes and perceived behavioral control explain a substantial proportion of the variance in intentions (Armitage & Conner; Hagger et al.). Notwithstanding this corroborating evidence, meta-analytic reviews have shown that the effect of subjective norms on intentions is small (Hagger et al.). NORMATIVE EXTENSIONS AND MODIFICATIONS OF THE THEORY OF PLANNED BEHAVIOR In response to the typically lesser role of subjective norms in the theory of planned behavior, some researchers have argued that the operational definition of subjective norms insufficiently captures social influences (Grube, Morgan, & McGree, 1986). Empirical evidence indicates that descriptive norms, a social influence construct that reflects perceptions of significant others’ behavior, predict intentions and/or behavior directly, and/or indirectly via perceived behavioral control (Nucifora, Gallois, & Kashima, 1993; Rivis & Sheeran, 2003). In addition, it has been suggested that descriptive norms should be incorporated as a main component of norms in the theory of planned behavior (Ajzen & Fishbein, 2005). Further, Grube et al. discovered a moderating effect of descriptive norms on the attitude-intention relationship. Their study revealed support for a contingent-consistency hypothesis by showing that the relationship between attitudes and intentions to smoke increased as descriptive norms increased. However, there are notable studies in which descriptive norms did not exert main or moderating effects on intentions and behavior (Conner & McMillan, 1999). Therefore, studies have not found consistently main or moderating effects from descriptive norms to intentions. The logic behind incorporating descriptive norms in the theory of planned behavior is based on learning theories that posit that modeling the perceived behavior of others is an important source of social influence (Grube et al., 1986). For example, children may ignore parental disapproval of dieting because they model the dieting regime of their friends, whose opinions they highly value. In these cases descriptive norms and subjective norms may exert independent effects on intentions. One limitation of the constructs of subjective norms and descriptive norms is that they do not differentiate between interpersonal influences from significant figures (i.e., parental influence) and group influences (i.e., peer influence). Like interpersonal influences, group influences do affect intentions and behavior, but in a different way. For example, social identity theory proposes that group norms and group identification influence intentions and behavior (Tajfel & Turner, 1986). In general, group norms indicate the perceived attitudes and behaviors of a behaviorally relevant group whereas group identification indicates the strength with which an individual identifies with a behaviorally relevant group. So far, research within the theory of planned behavior has supported an interaction between group norms and group identification such that the perceived behaviors and attitudes Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 37, 934–954 (2007)