Monday, October 13, 2008

Dissertation dish: K-ABC II, SB5, WJ III CHC factor analysis studies

Two new CHC-based dissertations I stumbled across this weekend.


A joint-confirmatory factor analysis using the Woodcock-Johnson III Tests of Cognitive Ability and the Stanford-Binet Intelligence Scales: Fifth Edition with high-achieving children by Williams, Tasha H., Ph.D., Ball State University, 2005, 206 pages; AAT 3176652


  • Abstract: A considerable about of research has concentrated on studying the performance of high achieving children on measures of intellectual functioning. Findings have indicated high achieving children display differences in performance patterns as well as in the cognitive constructs measured when compared to their average peers. The conceptualization of intelligence has evolved over time and contemporary theories of intelligence have described cognitive ability as consisting of multiple constructs which are often interrelated. Currently, one of the most comprehensive and empirically supported theories of intelligence is the Cattell-Horn-Carroll (CHC) theory (Cattell, 1941; Horn, 1968; Carroll, 1993). The multidimensional and hierarchical CHC theory has served as the foundation for the development and recent revisions of cognitive ability measures such as the Woodcock-Johnson Tests of Cognitive Ability - Third Edition (WJ-III COG; McGrew & Woodcock, 2001) and the Stanford-Binet Intelligence Scales - Fifth Edition (SBS; Roid, 2003b). The purpose of this study was to explore the construct validity of the WJ-III COG and SBS with a sample of high achieving children. Individual confirmatory factor analyses were conducted using the WJ-III COG and SBS. Additionally, a joint confirmatory factor analysis was conducted using both the WJ-III COG and SBS. The results indicated an alternative six-factor WJ-III COG and four-factor SBS models provided the best fit to the data of a high achieving sample, supporting previous research suggesting high achieving children display differences in cognitive constructs when compared with their average counterparts. The joint-confirmatory factor analysis indicated the best measures for the CHC factors measured by both the WJ-III COG and SB5 to help guide cross-battery assessments with high functioning children. Clinical applications of the findings are discussed.

The validity of intelligence tests using the Cattell-Horn-Carroll Model of intelligence with a preschool population by Morgan, Kimberly Elaine, Ph.D., Ball State University, 2008, 219 pages; AAT 3303357

  • Abstract: Individual differences in human intellectual abilities and the measurement of those differences have been of great interest to the field of school psychology. As such, different theoretical perspectives and corresponding test batteries have evolved over the years as a way to explain and measure these abilities. A growing interest in the field of school psychology has been to use more than one intelligence test in a "cross-battery" assessment in hopes of measuring a wider range (or a more in-depth but selective range) of cognitive abilities. Additionally, interest in assessing intelligence began to focus on preschool-aged children because of initiatives to intervene early with at-risk children. The purpose of this study was to examine the Stanford-Binet Intelligence Scales, Fifth Edition (SB-V) and Kaufman Assessment Battery for Children, Second Edition (KABC-II) in relation to the Cattell-Hom-Carroll (CHC) theory of intelligence using a population of 200 preschool children. Confirmatory factor analyses (CFAs) were conducted with these two tests individually as well as in conjunction with one another. Different variations of the CHC model were examined to determine which provided the best representation of the underlying CHC constructs measured by these tests. Results of the CFAs with the SB-V revealed that it was best interpreted from a two-stratum model, although results with the KABC-II indicated that the three-stratum CHC model was the best overall design. Finally, results from the joint CFA did not provide support for a cross-battery assessment with these two particular tests.

Technorati Tags: , , , , , , , , , , , , , ,

No comments: