The Validity of WISC-V Profiles of Strengths and Weaknesses

Dutch WISC-V standardization data plus a simulation study

Dutch school-age children; the normative standardization sample rather than a referred clinical sample

Factor-analytic decomposition of index difference scores plus a simulation that asks the decisive question - when a child's index score differs significantly from their own overall performance, how often does that correspond to a real difference on the underlying broad ability? A measurement study, so C is the ceiling this archive applies. Limits: one country's standardization of one battery, cognitive rather than achievement subtests, and a simulation whose answer depends on the factor model assumed.

The modern replication of the profile-analysis result, on a current instrument and a normative sample. Broad factors explained little of the variance in index scores, and in simulation a statistically significant discrepancy between an index score and overall performance corresponded to a genuine discrepancy on the underlying broad factor in only 40-74% of cases. Statistical significance of a profile difference is therefore not evidence that the difference is real in the sense a parent or teacher means - between a quarter and three-fifths of significant profile findings do not reflect the ability they are read as reflecting. This is the same conclusion McDermott et al. (1990, 1992) and Watkins & Canivez (2004) reached about the Wechsler scales a generation earlier, reached again on the fifth edition.

Not applicable. A claim about what index difference scores measure, not about the origins of ability.

Converges with the older ipsative-assessment literature (McDermott et al. 1990, 1992) and with Watkins & Canivez's chance-level test-retest replication of subtest strengths and weaknesses. Profile invalidity is one of the best-replicated negative findings in educational measurement, now spanning the WISC-R, WISC-III and WISC-V.