PSYCHOLOGICAL-TESTS AND CLINICAL DISCRIMINATIONS - BEGINNING TO ADDRESS THE BASE-RATE PROBLEM

Psychologists often use tests to make clinical discriminations. The ac curacy of those decisions depends on the discriminant validity of the tests on which they are based. However, traditional measures of discri minant validity do not reflect the accuracy of individual classificati ons. Although the effect of base rate on the ''curacy of clinical disc riminations has long been recognized, base rates have only recently be en considered in clinical assessment. A scheme for determining the pro babilities of valid test-based discriminations is discussed and relate d to several commonly used clinical tests. It is argued that such prob abilities must be considered in routine clinical practice, not just in the assessment of very low base rate disorders.