Ethnic differences in social anxiety between individuals of Asian heritage and European heritage: A meta-analytic review.

Abstract

Differences between Asian-heritage and European-heritage individuals have been found in both the prevalence and expression of social anxiety. Previous research suggests higher social anxiety among AH individuals, but both the methodology and the findings have been mixed and the effect sizes varied widely. The purpose of this meta-analytic study was to examine whether the AH/EH differences in social anxiety were robust across various moderators, and how these differences may be accounted by ethnic differences in self-construal and acculturative stress. The meta-analysis revealed a moderate mean effect size of d = 0.36, CI [0.27, 0.44]. Moderator analyses demonstrated that these effects were robust across moderators. Results of meta-regression analyses indicated that independent self-construal significantly contributed to social anxiety. Directions for future research include exploration of measurement issues, and a need to develop a theoretical model that includes factors that mediate the ethnic differences in independent self-construal and differences in social anxiety.

Comparing Social Anxiety in Japanese and European Americans: Integrating Measurement Invariance and Prototype Analyses

“…Additionally, the latent mean differences based on the partial scalar invariant measures were consistent with results of previous meta-analyses (e.g., Krieg & Xu, 2015) which showed higher social anxiety reported by individuals of Asian heritage than those of European heritage, providing some support for the validity of prior cultural comparisons. Furthermore, Study 2 showed that the majority of frequently mentioned descriptors by both cultural groups in response to the social anxiety vignette, such as 'symptoms of nervousness' and 'anxious', were negative attributes that are related to core features of social anxiety delineated in DSM, as well as the symptoms and experiences of social anxiety measured by SPS and SIAS.…”

“…The resulting latent mean difference was .28 (p = .01), indicating that Japanese scored higher than European Americans in social anxiety when equivalent measures were used. This is similar to the overall effect that Krieg and Xu (2015) found in their meta-analysis (d = .25).…”

“…Study 1 investigated measurement invariance of the Social Phobia Scale (SPS) and Social Interaction Anxiety Scale (SIAS) as well as their short forms, the SPS-6 and SIAS-6. These two popular social anxiety measures were chosen because they had been used to compare social anxiety between Japanese and European Americans more than once (Krieg & Xu, 2015). Study 1 involved three steps.…”

“…Given that there has been strong arguments made for using latent mean differences as oppose to mean differences of raw scores from manifest variables (e.g., Little, 1997), we estimated latent mean differences in social anxiety between European Americans and Japanese on the measures that were shown to be invariant. These differences were then compared to the overall effect size found in Krieg and Xu's (2015) meta-analysis.…”

Random‐effects models for meta‐analytic structural equation modeling: review, issues, and illustrations

“…In the 131 MASEM studies located, 123 of them used this approach. Seven studies (Fan, Jackson, Yang, Tang, & Zhang, 2010;Krieg & Xu, 2015;Montazemi, Pittaway, Saremi, & Wei, 2012;Murayama & Elliot, 2012;Simons, Leroy, Collewaert, & Masschelein, 2014;Trivette, Dunst, & Hamby, 2010;Yufik & Simms, 2010) used the two-stage SEM (TSSEM) approach, which will be introduced later. The remaining study (Grace et al, 2007) is the only study that used an approach similar to the parameter-based MASEM presented in the present paper.…”