Factor structure of the Social Interaction Anxiety Scale and the Social Phobia Scale Short Forms

The study used frequentist confirmatory factor analyses (CFA) and Bayesian CFA (BCFA) (one-factor, two-factor, and bifactor models) for Peters, Sunderland, Andrews, Rapee, and Mattick's (2012) short form Social Anxiety Interaction Scale (SAIS) and Fergus, Valentiner, McGrath, Gier-Lonsway, and Kim's (2012) short form Social Phobia Scale (SPS) Short Forms. Participants (N = 200) were adults from the general community who completed the full version of SIAS and SPS measures. For the different models tested, CFA provided moderate support for the two-factor model for Peters et al.'s Short Forms. BCFA showed good support for the two-factor and bifactor models for Peters et al.'s Short Forms, with the bifactor model showing better fit. This bifactor model showed high internal consistency reliability and had a high amount of explained common variance for its general factor. The SIAS and SPS specific factors of the bifactor model showed almost negligible internal consistency reliabilities and explained common variances.