Peters et al.'s (2012) Social interaction anxiety scale and the social phobia scale short forms: Support for a bifactor model and measurement invariance across gender

Full text for this resource is not available from the Research Repository.

Gomez, Rapson ORCID: 0000-0001-7637-1551, Watson, Shaun ORCID: 0000-0002-8112-9687 and Stavropoulos, Vasileios ORCID: 0000-0001-6964-4662 (2019) Peters et al.'s (2012) Social interaction anxiety scale and the social phobia scale short forms: Support for a bifactor model and measurement invariance across gender. In: Advances in Psychology Research. Volume 138. Columbus, Alexandra M, ed. Advances in Psychology Research (138). Nova Science Publishing, New York, pp. 75-112.

Abstract

For the Peters, Sunderland, Andrews, Rapee, and Mattick's (2012) Social Phobia Scale (SPS) and Social Interaction Anxiety Scale (SIAS) Short Forms together (PSIAS/SPS-SF), this study examined support for a bifactor model, factor-based internal consistency reliabilities of the factors in this model, and measurement invariance across gender for the items in the bifactor model. Participants (N = 819, females = 615, males = 204) were adults (age ranging from 18 to 79 years, mean = 25.73 years, SD = 9.70) from the general community who completed the PSIAS/SPS-SF. Confirmatory factor analysis (CFA) of their ratings indicated support for the bifactor model. The general factor (but not the specific factors) in this model showed good internal consistency reliability. Multiple group-CFA (MG-CFA) showed that with the exception of the intercept for one SIAS item (item # 3) all other measurement parameters (factor loadings, intercepts, and error variances) were invariant across males and females. Additionally, MIMIC analysis, controlling for age, showed differential item functioning for only two item intercepts (items # 3 and 5). Thus there was good support for measurement invariance. The MG-CFA showed no difference across the gender groups for the all factor mean scores in the bifactor model. The MIMIC analysis showed higher latent mean scores for females for the general factor. The findings can be taken as support the use of the total (and not the separate) scores for Peters et al.'s SPS and SIAS short forms, and comparability of observed scores provided by ratings of males and females on the PSIAS/SPS-SF.

Item type Book Section
URI https://vuir.vu.edu.au/id/eprint/44102
Edition 1st
ISBN 9781536162448
Subjects Current > Division/Research > Institute for Health and Sport
Current > Division/Research > College of Health and Biomedicine
Keywords gender, Social Interaction Anxiety Scale, Social Phobia Scale
Download/View statistics View download statistics for this item

Search Google Scholar

Repository staff login