Abductive inferences to psychological variables: Steiger's question and best explanations of psychopathy


  • Abductive inference often involves inference to the best explanation. A focus on the bestness of explanations facilitates a comparative analysis of how abductive inference would differ if approached with four contrasting sets of assumptions about how scientific inference works: positivism, realism, and two kinds of pragmatism. As a thought experiment, one can imagine a situation in which competing models of psychopathy differ in parsimony and fit to the data, but produce a tie when considering both virtues in combination. The thought experiment demonstrates that Steiger's (1990) question about how best to combine competing virtues in scientific inference applies to abductive inference and that the answers depend upon other assumptions about how science works. The comparative analysis helps focus some of the issues that require clarification before abductive inference can enter the Pantheon of standard research methods in psychology. More constructively, the analysis also demonstrates that one need not accept scientific realism to accept and use abductive inference. © 2008 Wiley Periodicals, Inc.