LEARNING DISABLED COLLEGE STUDENTS: THE ASSESSMENT OF ACADEMIC APTITUDE

A 2 x 2 quasi-experimental design with two levels each of ability group (learning disabled, nonlearning disabled) and testing condition (timed, untimed) was employed to assess subjects' test performance. Two dependent variables were employed in the study: the American College Test (ACT) and the Nelson-Denny Reading Test (1981 edition).

Undergraduate college students (N = 96), ages 18-23, participated as subjects, including 48 learning disabled (LD) and 48 nonlearning disabled (nonLD) students. All subjects had WAIS-R Full Scale IQs of 90 or higher, reported a history without severe emotional problems or severe hearing or vision deficits, and reported English as their first language. Subjects with WAIS-R Verbal, Performance, or Full Scale IQs 15 points (or more) higher than Wide Range Achievement Test standard scores were included in the LD group. LD and nonLD subjects were matched on WAIS-R Full Scale IQ (109.9, 109.7), educational level (13.9, 13.6), sex (25 males, 23 females in each group), and race (47 Caucasians, 1 Hispanic in each group). LD subjects were older than nonLD subjects (19.98, 18.94).

Subjects took the ACT and Nelson-Denny under timed (standardized) condition or untimed condition (without time limits). Obtained results were analyzed by an analysis of variance procedure and by Dunn's Multiple Comparison Procedure. Additional analyses included Cochran's Homogeneity Test, power tests, and omega squared estimates of treatment magnitude.

Results indicated that testing condition had a significant effect on test performance, but primarily for LD subjects. In the timed testing condition, LD subjects' ACT and Nelson-Denny raw scores were significantly lower than those of nonLD subjects, which is consistent with previous findings. However, there was no difference between the two ability groups' test scores in the untimed testing condition. For LD subjects, both timed and untimed ACT total raw scores correlated significantly with college GPAs (.69 and .53, respectively). For nonLD subjects, only timed ACT total raw scores correlated significantly with college GPAs. It was concluded that LD college students' academic aptitude may be more accurately assessed under untimed testing conditions than under timed testing conditions.