A psychological study of psychometric differences in Graduate Record Examinations General Test scores between learning disabled and non-learning disabled adults

Although the concept of learning disability (LD) remains controversial and the symptomology multifaceted--with various neurological and physiological theories of dysfunction or damage competing for explanatory preeminence--perhaps a most striking attribute of learning-disabled persons rests in their inability to read either quickly or with normal retention. Standardized aptitude tests, so crucial in admissions decisions, clearly represent extended exercises in rapid and accurate reading and are consequently assumed to present a special obstacle to individuals with reading disabilities. Educational Testing Service has attempted to rectify an inherent disadvantage by permitting learning-disabled candidates extra time in which to complete such tests as the Graduate Record Examinations General Test (GRE).

While offering more time to those with identified reading problems seems appropriate, scant empirical support exists for the practice. The purpose of this study was to assess the impact of expanded time limits on GRE scores. A subject pool ($N$ = 126), comprised of 72 female and 54 male undergraduate, graduate and non-students ranging in age from 20 to 56 was selected and assigned to one of two ability groups, LD or non-LD, on the basis of their scores on two selection instruments. Each subject took the GRE and Nelson-Denny Reading Test (dependent measures) both under standarized administration and under untimed conditions.

A quasi-experimental mixed-model design with one between-subjects variable and one within-subject variable was used. Data analysis used an analysis of variance procedure. Results indicated no significant difference in GRE scores between ability groups. However, testing conditions had a significant effect on test performance for both ability groups. Further data analyses were performed on the sample in which a mean-split was used to control for the IQ variable. Results indicated that for subjects with IQs of 117 and below timed scores of the LD group were significantly lower than those of the control group on but one of the dependent measures' scores. Test scores increased significantly for both groups between timed and untimed testing conditions. It was concluded that IQ may be a major factor affecting GRE performance for learning disabled adults.