Files
Abstract
Since the development of modern cognitive assessments, psychologists have been instructed by assessment “experts” and taught in graduate training programs that significant variability among index scores, or scatter, renders the general intelligence score unsound for clinical interpretation and diagnostic decision-making. The existing empirical research, however, challenges the validity of this interpretive heuristic known as the variability hypothesis. To investigate the credibility of the variability hypothesis, the present study examined the impact of significant factor score variability on the structural and incremental validity of the WAIS-IV with a sample of 1,211 individuals referred for evaluation at a university-based clinic. The sample included those both with (n = 616) and without (n = 595) significant variability in their WAIS-IV index scores. Exploratory factor analysis with multiple factor extraction criteria supported the extraction of four factors consistent with the WAIS-IV theoretical structure for the Scattered Group. The higher-order general factor also accounted for the largest portions of total and common variance. Incremental validity analyses indicated that the FSIQ accounted for large and statistically significant effects across measures of academic achievement for the Scattered Group. Although the four first-order factors combined accounted for significant incremental predictive contributions, individual WAIS-IV factor scores contributed trivial amounts of achievement variance beyond the FSIQ. Implications for clinical practice and the interpretation of the WAIS-IV in the presence of significant scatter are discussed.