Investigation of the validity of the AAMD adaptive behavior scale



Journal Title

Journal ISSN

Volume Title



The purpose of this study was to provide validity data on the Adaptive Behavior Scale (ABS) of the American Association of Mental Deficiency. Twenty mildly and twenty moderately retarded adults from the Cullen Residence Hall served as subjects. Objective criterion tests were developed on ten subtests from the ABS In order to objectively determine each subject's level of functioning In these areas. Four examiners, trained In standardized administration and scoring of the tests, obtained scores which were compared with ABS ratings given by four staff counselors at the residence. Each counselor also completed an "Opportunity to Observe" (OTO) scale on their subjects. Both sets of scores were compared to obtain "difference" scores and OTO scores which were analyzed by rater, task, level of retardation, and "OTO." Chi square analyses were done on the number, degree (number of difference levels), and direction (overestimates, underestimates) of misclassifications for each of the four variables. No significant differences were found among raters for number, degree, or direction of misclassification. Mild and moderate categories did not differ greatly In number and degree of misclassifications; however, there was a strong tendency to misclassify mildly and moderately retarded subjects by direction, with the mildly retarded being underestimated and the moderately retarded being overestimated. No significant differences were found among opportunities to observe for number or degree of misclassifications, but a significant difference was found for direction of misclassification; as the opportunity to observe increased, so did the tendency to overestimate. A Significant difference was found among tasks for all types of misclassifications. Of 400 total observations, 55.8% were misclassified, with 40.8% (of errors) being misclassified two levels or more. These results raise serious questions regarding the usefulness of the scales, both as a screening instrument and as a device for making specific behavioral statements about individuals.