Abstract
A three-type psychoeducational discrepancy model was described and related to the kinds of discrepancy information provided by the Woodcock-Johnson Psycho-Educational Battery. The following rationale was presented for using the Scholastic Aptitude clusters rather than the Broad Cognitive Ability cluster in the evaluation of aptitude-achievement discrepancies: The aptitude clusters are not confounded with achievement; they can reduce testing time; they provide differential expectancy information based on norms; and they provide higher clinical validity. The inappropriateness of comparing the norm-referenced discrepancies obtained from the Woodcock-Johnson with the results of procedures that do not correct for regression error was explained.