CHC study of WJ-R and DAS in preschool sample

Tusing, M. B., & Ford, L. (2004). Examining preschool cognitive abilities using a CHC framework. International Journal of Testing, 4(2), 91-114.


Abstract

• Although there has been a substantial growth in the number of published studies examining tests of cognitive abilities and using contemporary theories of cognitive abilities, to date none have done so with preschool cognitive tests. In this study the relation between cognitive ability measures for young children and Cattell–Horn–Carroll (CHC) theory is examined. Tests and subtests from the Differential Ability Scales: Upper Preschool Level and the Woodcock–Johnson Tests of Cognitive Ability–Revised with a sample of 158 children between 4 and 5 years of age were used in a series of joint factor analyses. Although a series of models were explored, the model representative of the CHC theory of cognitive abilities was best supported by the data. This provides evidence for a greater differentiation of young children’s cognitive abilities than are typically interpreted. Results are discussed with regard to understanding the link between contemporary theories of intelligence and young children’s cognitive abilities, as well as implications for intellectual assessment practices with young children.

Additional comments (by study authors) in discussion section

• "The notion that young children's cognitive abilities are best conceptualized as dichotomous is dismissed."

• "In this study, five broad ability factors were reliably identified: Gc intelligence, Glr, Gsm, Ga, and a fifth factor that we originally referred to as nonverbal ability."

• "Evidence for the differentiation of the Gy factor represented in Carroll's three-stratum theory into two distinct broad abilities (Glr and Gsm) and the identification of visual-memory as a narrow ability under Gv provide further support for McGrew and Flanagan's (McGrew, 1997; McGrew & Flanagan, 1998) integration of three-stratum theory and Gf--Gc theory into CHC theory."

• "Two broad ability factors thought to be represented by the subtests included in this study did not significantly distinguish themselves in this sample, namely, Gf and Gq. It is possible that these broad ability factors are not able to be distinguishedcfrom other broad abilities with samples of young children, as their loadingscon the broad ability factor with which they were identified were significant.cHowever, failure to identify the specific broad abilities of Gf and Gq is more likely due to a lack of additional tasks measuring the same abilities."