An Instrument for Measuring the Behavioral Dimension of Social Anxiety

Abstract

A review of the literature suggested a need for a behavioral measure of characteristic social anxiety, and Rehm and Marston's (1968) role-playing technique was adapted as an analog. This technique requires subjects to place themselves imaginatively in a series of anxiety-producing social situations which are described on a tape recording; their task is to respond to a line of dialog spoken by someone in the situation. Two equivalent test forms were prepared, each of which sampled a wide range of interpersonal situations relevant to both male and female students. Subjects' responses were videotaped and scored using a check list of behavioral anxiety indicators derived from Paul's (1966) list. Interrater reliabilities for the check list score were high ( r = .92), and test-retest reliability over 11 wk. was acceptable ( r = .57). The check list scores were unrelated to Marlowe-Crowne social desirability scores and weakly though positively related to self-report anxiety measures. Limitations of the instrument were pointed out, and suggestions for further validation studies and possible modifications were noted.