Doctoral Dissertations

Date of Award

5-2000

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Major

Education

Major Professor

R. Steve McCallum

Committee Members

Donald Dessart, Donald Dickinson, Thomas George

Abstract

The purposes of this research were threefold. The first goal was to develop and apply a normative typology using multivariate profile analysis of subtest scores of the Universal Nonverbal Intelligence Test (Bracken &McCallum, 1998; UNIT), taken from the national standardization sample; the second goal was to develop and apply a typology using multivariate subtestprofile analysis of a subsample of identified children with Learning Disabilities from the national standardization sample; and the third goal was to provide practitioners a description of user friendly strategies needed to compare meaningful subtest profiles of individual examinees with commonly occuring normative profiles, as identified in goal one above.To accomplish these goals, multistage cluster analyses were applied to the standardization sample for the UNIT Extended Battery, comprised of all six subtests; the UNIT Standard Battery, comprised of four subtests; and theUNIT Learning Disabled subsample, comprised of the four Standard Batterysubtests. The results of these analyses yielded a seven profile cluster solution for the Extended Battery, a six profile cluster solution for the Standard Battery,and a four profile cluster solution for the Learning Disabled subsample. The Psychometric properties of the respective analyses were impressive with extremely tight profile clusters that were separated extremely well from eachother.VDemographic prevalence trends of the resultant clusters are similar to other studies, but help to describe the cluster composition. Additionally, the results lend support to the UNIT'S underlying factor structure. To fulfill the third goal of this research, user-friendly methods of determining whether ornot clinical profiles are unique when compared to the profiles identified in the standardization sample are discussed.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS