We selected a sample of patients with depression (N = 70) and used Bayesian analyses to examine the diagnostic predictive validity of the Wechsler Adult Intelligence Scale-Revised Verbal IQ (VIQ) minus Performance IQ (PIQ) difference score for depression. The patients showed average VIQ-PIQ differences in the expected direction (i.e., VIQ > PIQ). However, based on sensitivity, specificity, and positive predictive value statistics, we concluded that the VIQ-PIQ split has very low predictive validity in persons with depression.