Social competence, social anxiety, and self-evaluation: Influences on accuracy of social perception
Abstract
Male subjects completed a self-report measure of social anxiety, then participated in two brief videotaped interactions with female confederates. A behavioral measure of heterosocial competence was derived from these interactions. Subjects later returned to complete a social perception rating task, and did so under either high or low conditions of self-evaluation. Accuracy of subjects' ratings on this task was evaluated by examining discrepancies between subject and criterion judge ratings, in terms of both overall accuracy (D-square) and Cronbach's (1955) components of scatter error and pattern error. Multiple regression models predicting error components were derived using social anxiety, social competence, and self-evaluation condition as predictors. Low competent subjects were found to show greater scatter error in the high self-evaluation condition than in the low self-evaluation condition, as a function of both social competence and social anxiety. Pattern error showed a more complex relationship with the predictor variables. In general high social competence was related to lower pattern error scores. Though scatter error generally went down as a function of increasing anxiety, higher anxiety levels were, in the high self-evaluation condition, mostly associated with higher pattern error. Results are discussed in terms of consistency with past research and efforts to identify discriminable contributions of social anxiety and social competence to relative accuracy of social perception.
Degree
Ph.D.
Advisors
Conger, Purdue University.
Subject Area
Psychotherapy
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.