A Meta-Analytical Test of Perceived Behavioral Control Interactions in the Theory of Planned Behavior
Fink, Edward L.
MetadataShow full item record
This study used meta-analytic procedures to test for interaction effects among the components of the theory of planned behavior (TPB). The central hypothesis examined was that attitudes and subjective norms should perform less well in explaining intentions when perceptions of behavioral control are low. A traditional meta-analysis of nine studies that directly examined perceived behavioral control (PBC) interactions was conducted. A second meta-analysis--the main focus of this dissertation--was conducted that tested for two- and three-way interactions in which the presence of PBC interactions was investigated in 121 studies, which provided 154 data sets with 44,424 participants. In addition to testing for two-way PBC interactions, this meta-analysis also examined whether the presence of PBC interactions depended on other variables. Specifically, three-way interactions with type of behavior classification (i.e., public versus private, familiar versus unfamiliar) and type of PBC operationalization (e.g., self-efficacy, perceived difficulty, perceived control, or some combination of the three) were explored. Results indicated that attitude by PBC interactions exist but that the effects vary depending on the type of PBC operationalization and behavior context. In addition, meta-analytic structural equation modeling was used to examine whether the association between PBC and intention is mediated by attitude and subjective norms; however, no evidence for this relationship was found. Finally, results from an auxiliary analysis revealed that the attitude by PBC interaction on intention had statistically significant nonlinear effects in addition to a linear effect. In contrast, the norm by PBC interaction did not have statistically significant linear or nonlinear effects. The discussion highlights the effects of different meta-analytic techniques, the need for future investigation using experimental designs, the implications of these findings for further theory development, and practical implications for health communication researchers. In sum, through the use of a multi-faceted approach to quantitatively review attitude by perceived control and norm by perceived control interactions in the TPB, this study helped to address inconclusive results with regard to the existence and type of PBC interactions.