Are you sure you're really telling the truth?

Abstract

The core constructs of psychological assessment, reliability and validity, are shown to be incomplete and lacking whenever base rates for the conditions under evaluation skew markedly from 50/50. Under such circumstances, base rate information can exert a greater influence on diagnostic accuracy that is typically recognized among practitioners. The influence of base rates is so profound, that conclusions based on reliable and valid test data are often, more probably than not, wrong! This paper outlines our historical understanding of the “base rate fallacy”, and offers explanations for its persistence in the practice of diagnostic psychology. Recommendations for self-monitoring and policing of our profession are offered, in order that neuropsychology might improve its diagnostic accuracy at a rate more comparable to the progress made in similar fields in medicine.