Beyond individual tests : the effects of children's and adolescents’ cognitive abilities on their achievement

Students’ performance across several tests, including both cognitive and achievement tests, is often analyzed together to better understand their learning. This analysis is guided by the assumption that there are specific relations between students’ cognitive abilities and their reading, writing, and math skills. The research supporting this assumption is limited because cognitive-achievement research findings are mostly based on a single test, the Woodcock-Johnson tests (McGrew & Wendling, 2010), and previous studies involve analyzing a single intelligence and achievement test in isolation. Thus, findings are limited to the specific tests that are included in those analyses, and are not necessarily generalizable across other tests. Research that incorporates multiple intelligence and achievement tests, cross-battery analyses, can better address questions about the broader influences of children’s cognitive abilities on their achievement. Such cross-battery research can extend psychologists’ understanding of how intelligence and achievement relate beyond the test-level to the construct level.
Six intelligence tests (KABC-II, WJ III, WISC-III, WISC-IV, WISC-V, and DAS-II) and three achievement tests (KTEA-II, WIAT-II, WIAT-III) were analyzed in a cross-battery cognitive-achievement analysis in the current study. Data were derived from seven of the tests’ standardization or linking samples; participants were 3,930 children and adolescents aged 6 to 16. In order to simultaneously analyze several tests a planned missingness approach and structural equation modeling were used. Six broad abilities (Gc, Gf, Gv, Gsm, Gs, and Glr) and g were modeled as latent variables; each broad ability latent variable was indicated by 7 – 14 subtests. Results suggest Gf and g were perfectly correlated and it was impossible to separate the two abilities statistically. The cognitive abilities were predictors of three achievement skills (basic reading, broad writing, and broad math), which were indicated by four to six subtests. Findings indicated Gc influenced all three academic skills; Gsm and Glr influenced basic reading and broad writing; Gs influenced broad writing and broad math; Gf exerted a significant effect on broad math; and Gv was not significantly related to any academic skill. Significant cognitive-achievement relations have implications for diagnostic decision-making regarding specific learning disabilities, assessment planning, and educational recommendations.