The meaning of malingering data: Further applications of Bayes' theorem

A previous Behavioral Sciences and the Law article (Mossman & Hart, 1996) a sserted that information from malingering tests is best conceptualized usin g Bayes' theorem, and that courts therefore deserve Bayesian interpretation s when mental health professionals present evidence about malingering. Moss man and Hart gave several examples of estimated Bayesian posterior probabil ities, but they did not systematically address how one constructs confidenc e intervals for these estimates. This article explains how the usually impe rfect nature of humanly created diagnostic tests mandates Bayesian interpre tations of test results, and describes methods for generating confidence in tervals for posterior probabilities. Sample calculations show that Bayesian reasoning is quite feasible and would not require investigators to expend unusual efforts when constructing and validating malingering instruments. B ayesian interpretations most accurately capture what malingering tests do: provide information that alters one's beliefs about the likelihood of malin gering. Copyright (C) 2000 John Wiley & Sons, Ltd.