Doctoral Dissertations

Date of Award

12-1993

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Major

Education

Major Professor

Teresa A. Hutchens

Committee Members

William Poppen, Michael Hannum, Richard Saudargas

Abstract

This Study investigated the use of the LuriaNebraska Neuropsychological Battery-Children's Revision (LNNB-CR) as compared to the Simple Difference and Regression discrepancy methods (IQ - academic achievement) in differentiating Learning Disabled and Non-Learning Disabled students. Thirty Learning Disabled and 30 Non-Learning Disabled students were matched on chronological age, grade, gender, socioeconomic status and intelligence scores. The Wechsler Intelligence Scale for Children-Third Edition (WISC-III), Wechsler Individual Achievement Test (WIAT), and the LNNB-CR were administered. Results of t tests using the WIAT indicated that significant differences were evidenced between LD and Non-LD subject groups in six of seven achievement areas. The academic area evidencing the most discrepant score between LD and Non-LD groups was Written Expression. Results of t tests on the 12 clinical scales of the LNNB-CR indicated group differences on six clinical scales. The LNNB-CR scale that contributed most significantly to group differences was Writing. Results of the MANOVA indicated a significant statistical difference between the LD and Non-LD groups on the collapsed, 12 clinical scales of the LNNB-CR. Subsequent discriminant function analysis, with the 12 clinical scales generated an 86.7% correct classification rate with a significantly reduced rate of false positives as compared to the discrepancy methods. The standardized discriminant function coefficients indicated that the LNNB-CR Writing scale was the major contributor to between group differences. Analysis of the LNNB-CR partial battery, with academic scales removed, indicated no significant difference between LD and Non-LD subjects. Academic scales may account for significance in differentiating between groups. However, using the LNNB-CR profile analysis, based on two or more clinical scale elevations, an 85% rate of correct classification was obtained. Even excluding academic scales, the number of clinical scale elevations was significant between LD and Non-LD groups t(58) = 2.25, p < .05. The Simple Difference and Regression methods yielded overall correct classification rates of 71.6% and 76.7%, respectively. Findings indicated that compared with either the Simple Difference or Regression discrepancy methods, the LNNBCR discriminant function (all scales) and profile analysis with two or more clinical scale elevations most accurately discriminated LD from Non-LD students. The diagnostic utility and implications of these findings are discussed.

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

Share

COinS