Nonlinear models of clinical judgment: Meehl's data revisited

Previous attempts to detect nonlinearity in clinical judgments have not succeeded because of a lack of good nonlinear models. Much research in this area was based on data collected by Paul Meehl, which include clinicians' judgments of mental disorder on the basis of Minnesota Multiphasic Personality Inventory profiles. In this article, Meehl's data are reanalyzed using several versions of the scatter model in which nonlinearity is represented by the within profile scatter(s) of the cues. The author finds that these versions give a better fit to the data than the linear model. He also finds systematic patterns of nonlinearity that lend themselves to psychological interpretation.