Abstract
The reliability, discriminant validity, and convergent validity of the positive behavior subscale (Part A) of the Social Performance Schedule (SPSS) were examined. College students completed the SPSS once and a subsample completed it again one month later. High skill (HS) and low skill (LS) subjects on Part A completed the Hopkins Symptom Check List (to assess discriminant validity), self-monitored their naturally occurring social interactions, listed the number of friends they had, and participated in an in vivo interaction. Also, two friends of each subject were asked to complete a peer rating form to assess subjects' social competence. The reliability (temporal and internal consistency) of Part A was adequate. HS and LS subjects did not differ on the Symptom Check List, supporting Part A's discriminant validity. MANOVAs and subsequent univariate ANOVAs indicated that the two skill groups differed significantly on most of the convergent validity criteria. Findings from a previous SPSS study (Miller & Funabiki, 1984), which found consistent HS-LS differences on a laboratory-based interaction, were replicated for global but not for specific measures of skill. Reasons for this discrepancy and future research directions were briefly discussed.
Get full access to this article
View all access options for this article.
