Date of Award

1993

Document Type

Dissertation

Degree Name

Doctor of Philosophy

School

School of Education

Program

Counseling Psychology, Ph.D.

First Advisor

Jimmy Kijai

Second Advisor

Wilfred G. A. Futcher

Third Advisor

Elsie Jackson

Abstract

Problem. The purpose of this study was to develop an instrument to screen children who may be at risk for Attention Deficit Disorder (ADD). This study utilized ten subtests/variables--letter sequencing, numerical operations, number sequencing, coding, listening concentration, color sequencing, object sequence, attention for sequencing, selective attention, and sentence repetition--to generate responses from two groups of students. The intent was to determine if the linear combination of variables or some subset of them could discriminate between the ADD and nonADD groups.

Method. The items in each subtest were carefully selected and subjected to critiques by eight psychologists, a psychometrician, and two medical doctors specializing in the diagnosis of ADD. The instrument was then subjected to a pilot study. The revised Carlisle-Attention Deficit Diagnostic System was later administered to 63 ADD children and 59 matched-paired "normal" children. Item analysis was performed and scale reliability indexes estimated. Analysis of variance, discriminant analysis and a series of t-tests were used to analyze the data.

Results. Because of its low reliability, the number sequencing subtest was omitted. The reliability coefficients of the subtests retained ranged from 0.61 to 0.96.

On each of the subtests, the group mean of the nonADD group was significantly higher than that of the ADD group. Gender differences were observed on the coding and numerical subtests. Significant interactions existed between (1) gender and test group and (2) age and test group on some subtests.

The discriminant functions identified the listening concentration, attention for sequencing, and coding subtests as contributing most to the group separation. The various discriminant functions correctly classified 76-100% of the subjects into their respective ADD/nonADD groups.

Conclusion. Based on the analyses, the following conclusions were deduced: (1) Nine C-ADDS subtests have moderate to high reliability and reasonable discriminative validity. (2) Various subtests on the C-ADDS were able to detect age group and gender differences. (3) Certain linear combinations of variables successfully separated the different subgroups. Thus, certain traits of the Attention Deficit Disorder can be assessed/measured with the use of C-ADDS-like items. (4) The C-ADDS, as a screening device, has useful diagnostic capabilities and represents an appropriate instrument for screening ADD subjects.

Subject Area

Attention-deficit hyperactivity disorder--Diagnosis

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

Share

COinS