Inference Patterns in Theoretical Social Psychology: Looking Back as We Move Forward

Abstract

An important hallmark of modern social psychology is the central importance of theory and hypothesis testing. The inference structure, however, underlying such tests are rarely acknowledged and may have important implications for the field. In this study, we examine the inference patterns utilized over 20 years of published work in a leading journal of social psychology. We found that verification, in contrast to falsification and crucial tests of alternative hypotheses, was the dominant inference structure utilized. Crucial tests were conducted least of all. These trends were discussed in light of larger issues from the philosophy of science, as well as its potential implications for the field. Without doubt the current trend in social psychology is towards the objective, not speculative, study of social behavior. Watchwords are experimentalism, automatic computation, statistical reliability, replicability. Noteworthy scientific gains results form this ‘hard nosed’ approach (Allport, 1968, p. 68). One of the hallmarks of modern social psychology is the central importance of theory and hypothesis testing. As a result, a critical but often underappreciated issue for the field concerns the inferential structures utilized to test these propositions. This is important as inference patterns shape every aspect of the scientific process. It obviously influences how we interpret results, but less appreciated is its influence on how researchers design ⁄conduct studies and build theoretical frameworks. For instance, a verification perspective may encourage the testing of a dominant theoretical model. The lack of alternative perspectives, in turn, may place pressures on such a model to undergo ad hoc changes in light of novel or discrepant data (Ladyman, 2002). To our knowledge, no published article has systematically reviewed general inference patterns in social psychology. Although we suspect the results of such a review are unlikely to surprise social psychologists who specialize in philosophy of science and ⁄or scientific inference (because these individuals may have informally tracked these patterns), our aim for this review is to bring these patterns and their implications to the attention of a more general social psychological audience. In addition, given the increased technological, methodological, and statistical demands of modern social psychology (Cacioppo, Berntson, & Nusbaum, 2008; MacKinnon, 2008; Reis & Judd, 2000), we suspect that recently trained social psychologists may not have had as much exposure to these philosophical issues. We thus examined trends in inference patterns utilized over the past 20 years in a flagship journal of social psychology and discuss these patterns in light of larger issues from the philosophy of science and social psychology more generally. One potential issue to consider is that in some cases it is possible that inference strategies better reflect what social psychologists present in articles rather than the inference