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.