A Demonstration of the Utility of Formalizing Theory: Mapping, Evaluating, and Testing the Theory of Devaluation of Alternatives
Abstract
Devaluation of alternatives is often identified as a key relationship maintenance strategy that is used to combat the threat of attractive alternatives. However, since the theory was proposed by Johnson and Rusbult (1989), few efforts have been dedicated to evaluating the progression of the theory and assessing the quantity and quality of the existing evidence that claims to corroborate it. The present research fulfilled this need by formalizing the theory of devaluation of alternatives using a metatheoretical approach that involved categorization, deconstruction, reconstruction via theory mapping, and evaluation (Chapter 2). P-curve meta-analyses were used to determine whether p-hacking or selective reporting could be detected in a selection of the literature (Chapter 3). These processes revealed many of the theoretical assumptions were understudied and lacked sufficient corroboration, and evidence of selective reporting was not detected. Guided by these findings, the assumption that attractive alternatives are threatening was selected and tested using a novel experimental manipulation (Chapter 4). Following the completion of a compatibility questionnaire, participants were shown fabricated compatibility results with their current partners and a real alternative partner of their choosing. Compatibility with the alternative was manipulated to be high, average, or low, with higher compatibility representing stronger threat. Quantitative and qualitative results indicated that participants across conditions did not significantly differ in their experiences of negative affect or threat. While participants reported experiencing negative emotions, disappointment was experienced to a greater extent than anxiety or insecurity. These results indicate a failure to corroborate the primary assumption of devaluation of alternatives. In addition, across conditions participants expressed devaluation by criticizing the questionnaire/algorithm in a free response, while quantitative measures demonstrated devaluation in all conditions except the low-threat condition. These results indicate conflicting evidence for when devaluation is elicited. This research demonstrated that theories that are thought to be well-founded may not be. Social psychology is a theory-rich discipline; however, our empirical pursuits would be much improved by dedicating effort towards formalizing and evaluating the current state and verisimilitude of our existing theories to best determine how much has been accomplished and where to focus future research efforts.