Confirmatory factor analysis of the Kaufman Adolescent and Adult Intelligence Test with young adolescents

This study examined the factor structure of the Kaufman Adolescent and Adult Intelligence Test (KAIT) in 375 11- to 14-year-olds (normative sample) and 60 sixth- and eighth-graders (cross-validation sample). Previous exploratory and confirmatory factor analyses of the KAIT have included optional subtests, which are often not administered in practice, and have not tested oblique versus orthogonal two-factor models. In this study, three factor models were tested via confirmatory factor analysis for the six core subtests of the KAIT: a one-factor general intelligence (g) model, an orthogonal fluid intelligence (Gf)-crystallized intelligence (Gc) model, and an oblique Gf-Gc model. The orthogonal Gf-Gc model fit poorly in both samples. The g model fit only the cross-validation sample. The oblique Gf-Gc model fit both samples and fit significantly better than the g model in both samples. Additional tests of factor structure, path coefficients, and covariance between Gf and Gc indicated that the data from both samples fit even the most restrictive oblique Gf-Gc model tested. KAIT users can be confident that the Fluid and Crystallized IQs from the KAIT are adequate representations of a robust and interpretable factor structure when only the core subtests are administered to young adolescents.