Psychometric properties and three proposed subscales of a self-report version of the Liebowitz Social Anxiety Scale translated into Hebrew

We investigated the overall test-retest reliability and other psychometric properties of a self-report version of the Liebowitz Social Anxiety Scale (LSAS) translated into Hebrew. We also evaluated the utility of three new subscales that were identified by nonparametric analysis (multidimensional scaling; MDS). Two hundred and seven patients who sought treatment for social anxiety or panic disorder were evaluated. All patients completed the self-administered version of the LSAS. A subsample completed the LSAS a second time prior to the beginning of treatment. The results indicate that the self-report format of the LSAS translated into Hebrew demonstrates high test-retest reliability, internal consistency, and discriminant validity. Additionally, some evidence for convergent and divergent validity was noted, and treatment sensitivity was high. MDS analysis followed by the investigation of common underlying facets for items related in two-dimensional space identified three subgroups: 1) the Group Performance/Interaction ("Group") subscale that consists of group performance and group interaction items; 2) the Dyadic Interaction ("Dyadic") subscale that consists of Dyadic interaction items; and 3) the Public Activities ("Public") subscale that consists of individual activities carried out in public. The three new subscales identified by MDS appear to provide clinically relevant information that relates to both demographic and treatment outcome variables and warrant further study. (C) 2002 Wiley-Liss, Inc.