Search

The Woodcock-Johnson (fourth edition; WJ IV; Schrank, McGrew, & Mather, 2014a) was recently redeveloped and retains its linkage to Cattell-Horn-Carroll theory (CHC). Independent reviews (e.g., Canivez, 2017) and investigations (Dombrowski, McGill, & Canivez, 2017) of the structure of the WJ IV full test battery and WJ IV Cognitive have suggested the need for additional factor analytic exploration. Accordingly, the present study used principal axis factoring (PAF) followed by the Schmid and Leiman (SL; Schmid & Leiman, 1957) procedure with the 2 school-aged correlation matrices from the normative sample to determine the degree to which the WJ IV total battery structure could be replicated. Although 7 factors emerged across the 9 to 19 age range, the pattern of subtests loadings did not fully cohere with the structure presented in the "Technical Manual," most notably for the academic fluency subtests. Also, the Fluid Reasoning (Gf) and Quantitative Reasoning (Gq) subtests coalesced to form a combined factor rather than 2 separate factors and the Long Term Retrieval (Gltr) subtests aligned with a variety of different factors. The results of this study indicated that the general intelligence factor variance far exceeded the variance attributed to the lower-order CHC factors. The combination of subtest migration and nominal total/common variance of the CHC lower-order factors suggests caution when interpreting the myriad CHC-related indices when making high stakes decisions. Implications for clinical practice are discussed.

The Wechsler Intelligence Scale for Children--Fourth Edition (WISC-IV) is one of the most frequently used intelligence tests in clinical assessments of children with learning difficulties. Construct validity studies of the WISC-IV have generally supported the higher order structure with four correlated first-order factors and one higher-order general intelligence factor, but recent studies have supported an alternate model in which general intelligence is conceptualized as a breadth factor rather than a superordinate factor (M. W. Watkins, 2010, Structure of the Wechsler Intelligence Scale for Children-Fourth Edition among a national sample of referred students, "Psychological Assessment," Vol. 22, pp. 782-787; M. W. Watkins, G. L. Canivez, T. James, K. & R. Good, in press, Construct validity of the WISC-IVUK with a large referred Irish sample, "International Journal of School and Educational Psychology"). WISC-IV core subtest data obtained from evaluations to assess learning difficulties in 345 children (224 boys, 121 girls) were examined. One through four, first order factor models "and" indirect versus direct hierarchical models were compared using confirmatory factor analyses. The correlated four-factor Wechsler model provided good fit to these data, but the direct hierarchical model showed statistically significant improvement over the indirect hierarchical model "and" correlated four-factor model. The direct hierarchical model was judged the best explanation of the WISC-IV factor structure, with the general factor accounting for 71.6% of the common variance while the first order factors accounted for 2.4-10.3% of the common variance. Thus, the results with the present sample of referred children were similar to those from other investigations (G. E. Gignac, 2005, Revisiting the factor structure of the WAIS-R: Insights through nested factor modeling, "Assessment," Vol. 12, pp. 320-329; G. E. Gignac, 2006, The WAIS-III as a nested factors model: A useful alternative to the more conventional oblique and higher-order models, "Journal of Individual Differences," Vol. 27, pp. 73-86; P. Golay, I. Reverte, J. Rossier, N. Favez, & T. Lecerf, 2012, Further insights on the French WISC-IV factor structure through Bayesian structural equation modeling. "Psychological Assessment," advance online publication; M. W. Watkins, 2010, Structure of the Wechsler Intelligence Scale for Children-Fourth Edition among a national sample of referred students, "Psychological Assessment," Vol. 22, pp. 782-787; M. W. Watkins, G. L. Canivez, T. James, K. & R. Good, in press, Construct validity of the WISC-IVUK with a large referred Irish sample, "International Journal of School and Educational Psychology") supporting primary interpretation of the Full Scale IQ rather than the factor index scores.

Orthogonal higher-order factor structure of the Cognitive Assessment System (CAS; Naglieri & Das, 1997a) for the 5-7 and 8-17 age groups in the CAS standardization sample is reported. Following the same procedure as recent studies of other prominent intelligence tests (Dombrowski, Watkins, & Brogan, 2009; Canivez, 2008; Canivez & Watkins, 2010a, 2010b; Nelson & Canivez, 2011; Nelson, Canivez, Lindstrom, & Hatt, 2007; Watkins, 2006; Watkins, Wilson, Kotz, Carbone, & Babula, 2006), three- and four-factor CAS exploratory factor extractions were analyzed with the Schmid and Leiman (1957) procedure using MacOrtho (Watkins, 2004) to assess the hierarchical factor structure by sequentially partitioning variance to the second- and first- order dimensions as recommended by Carroll (1993, 1995). Results showed that greater portions of total and common variance were accounted for by the second-order, global factor, but compared to other tests of intelligence CAS subtests measured less second-order variance and greater first-order Planning, Attention, Simultaneous, and Successive (PASS) factor variance. (Contains 4 tables.)

The factor structure of the Wechsler Adult Intelligence Scale-Fourth Edition (WAIS-IV; Wechsler, 2008a) with the adolescent participants (ages 16-19 years; N = 400) in the standardization sample was assessed using exploratory factor analysis, multiple factor extraction criteria, and higher-order exploratory factor analyses. Results from exploratory factor analyses were not included in the WAIS-IV Technical and Interpretation Manual (Wechsler, 2008b) and are necessary for determining convergence or divergence with the reported confirmatory factor analyses. As found with the total WAIS-IV standardization sample (Canivez & Watkins, in press), the present results with the adolescent subsample found all WAIS-IV subtests (10- and 15-subtest configurations) were properly associated with their four theoretically proposed first-order factors, but only one factor extraction criterion (standard error of scree) recommended extraction of four factors. Hierarchical exploratory analyses with the Schmid and Leiman (1957) procedure found that the second-order g factor accounted for major portions of total and common variance, while the four first-order factors accounted for small portions of total and common variance. It was concluded that the WAIS-IV provides strong measurement of general intelligence in adolescents and clinical interpretation should be primarily at that level. (Contains 3 tables and 2 figures.)

The Wechsler Abbreviated Scale of Intelligence (WASI; Psychological Corporation, 1999) and the Wide Range Intelligence Test (WRIT; Glutting, Adams, & Sheslow, 2000) are two well-normed brief measures of general intelligence with subtests purportedly assessing verbal-crystallized abilities and nonverbal-fluid-visual abilities. With a sample of 152 children, adolescents, and adults, the present study reports meaningful convergent validity coefficients and a latent factor structure consistent with the theoretical intellectual models both tests were constructed to reflect. Consideration of the hierarchical model of intelligence tests and issues regarding test interpretation are presented. (Contains 4 tables and 3 figures.)

Orthogonal higher-order factor structure of the Stanford-Binet Intelligence Scales-Fifth Edition (SB-5; Roid, 2003a) for child and adolescent samples is reported. Multiple criteria for factor extraction unanimously supported extraction of only one dimension and a unidimensional model. However, following results from DiStefano and Dombrowski (2006) and theoretical consideration, two factors were extracted and obliquely rotated and further subjected to the Schmid and Leiman (1957) procedure using MacOrtho (Watkins, 2004). Results showed that the largest portions of total and common variance were accounted for by the second-order, global ("g") factor and interpretation of the SB-5 should focus primarily, if not exclusively, on the general, Full Scale IQ. No evidence for a five-factor solution was found. (Contains 3 tables and 1 figure.)