Social anxiety in the general population: Introducing abbreviated versions of SIAS and SPS

Abstract

Social anxiety is characterized by the experience of stress, discomfort and fear in social situations, and is associated with substantial personal and societal burden. Two questionnaires exist that assess the aspects of social anxiety, i.e. social interaction anxiety (SIAS) and social phobia (SPS). There is no agreement in literature on the dimensionality of social anxiety. Further, the length of a questionnaire may negatively affect response rates and participation at follow-up occasions. To explore the structure of social anxiety in the general population, and to examine psychosocial and sociodemographic correlates. Our second aim was to construct abbreviated versions of SIAS and SPS that can be easily used and with minimal burden. A total of 1598 adults from the general Dutch population completed a survey asking information on social anxiety, mood and demographics. Exploratory and confirmatory factor analyses as well as reliability analysis with item-total statistics were performed. Confirmatory factor analysis revealed a 3-factor structure for social phobia, and a 2-factor structure for the SIAS, with the second factor containing both reversely scored items. The abbreviated versions of SPS (11 items) and SIAS (10 items) show excellent discriminant and construct validity (Cronbach's α=.90 and .92), while specificity analysis showed that gender, marital status and educational level (SIAS(10): p<.0005; SPS(11): p<.0005) are important determinants of social anxiety. In the general population, social interaction anxiety and social phobia are two aspects of a higher-order factor of social anxiety. Social anxiety is validly captured by the short versions of SPS and SIAS, reducing the questionnaire burden for participants in epidemiological and biobehavioral research.