Development of a rating scale to assess social skill deficits in mentally retarded adults

In the current study, two experiments were conducted to establish the reliability of a scale developed with persons of normal intelligence for the assessment of social skill deficits and excesses of the mentally retarded. The experiment was divided into two studies. The first was aimed at establishing which of the 100 items could be accurately rated by having pairs of staff independently score 22 mild and moderately retarded adults who were outpatients at a local community mental health center. A correlation of r = .30 was established as a minimum cut off for reliability. With this criterion, 57 items were selected from the scale for further testing. In the second study, 207 different patients from the same outpatient clinic, from the mental retardation center of a nearby state psychiatric hospital and from Champaign, Illinois, were assessed on the 57-item Social Performance Survey Schedule. A factor analysis was then run and four subscales of the assessment instrument were established. Implications of the findings are discussed.