Unique Associations Between Peer Relations and Social Anxiety in Early Adolescence

Abstract This study examined the unique associations between feelings of social anxiety and multiple dimensions of peer relations (positive peer nominations, peer- and self-reported peer victimization, and self-reported friendship quality) among 383 sixth- and seventh-grade students. Hierarchical regression analysis provided evidence for the unique contribution made by peer relations to social anxiety above that made by adolescents' individual vulnerabilities (i.e., teacher ratings of social behavior, self-reported social appraisals assessed by hypothetical vignettes). Two subgroups of socially anxious adolescents—those with and without peer problems—were distinguished by their social behavior but not their social appraisals. Notes 1A copy of the measure can be obtained from the first author. 2The percentage of adolescents who met criteria for inclusion in the two peer-problem groups are as follows: For the anxious-peer problems group, 26% of adolescents were included based on peer-reported victimization only, 28% based on low positive peer nominations, and 46% based on both peer problems; for the peer problems-only group, 31% of adolescents met inclusion criteria based on peer-reported victimization only, 47% based on low positive peer nominations received only, and 22% based on both. ∗p < .05; ∗∗p < .01; ∗∗∗p < .001; † p < .10. Note. F(11, 371) = 16.85; p < .001 on overall regression analysis. ∗p < .05; ∗∗∗p < .001; † p < .10. 3The regression analysis was also repeated twice with peer-reported victimization only and self-reported victimization only entered into Step 3 instead of the victimization composite score. Nearly identical results were found with social appraisals, positive peer nominations, and peer victimization each explaining significant unique variance in social anxiety, and one significant interaction (Gender × Victimization). For the equation including peer-reported victimization, the marginal relationship between friendship quality and social anxiety became significant. Note. Values in the table represent the group means. Standard deviations are shown in parentheses. All F values are significant at p < .001. Different subscripts indicate significant mean differences at p < .05, using Tukey post hoc analyses. a n = 40; b n = 39; c n = 217; d n = 87. 4The higher level of social anxiety among the anxious-peer problems group in comparison to the anxious-only group was noted. Thus, all analyses were also conducted with just the high anxious subgroups controlling for social anxiety. Results were the same. Of importance, the anxious-peer problems group displayed less prosocial behavior than the anxious-only group, even when level of social anxiety was controlled.