Longitudinal Relationships Between Social Anxiety, Peer Victimisation, and Perceived Support Among Children

Social anxiety, peer victimisation, and social support are key issues during mid-childhood that provide important influence in the later adolescent years. While extensive research has evaluated these constructs in isolation and during adolescence, almost no longitudinal studies have evaluated relationships between them during mid-childhood. The current study obtained self-report measures of social anxiety, peer victimisation, and social support from a close friend, from 7846 students from grades three and four (M age = 9.01 yr), on three occasions over two years. Path analyses examined mediating and moderating relationships, as well as moderation by binary gender. There was minimal moderation by gender on any of the relationships between variables. Over time, there was a stronger positive prediction from social anxiety to later victimisation (β's = 0.09-0.12) than from victimisation to later social anxiety (β's = 0.04-0.07). Social anxiety negatively predicted later social support (β's = -0.05 to -0.09), but social support did not consistently predict later social anxiety (β's = -0.01 to -0.02). Interestingly, the results did not support a buffering effect of social support on peer victimisation; however, these longitudinal data did support several indirect paths. The results highlight the cyclical relationship between social anxiety and peer victimisation but point to social anxiety as a unidirectional predictor of social support from a close friend during the middle to late childhood years. Early intervention for social anxiety may produce positive downstream effects on both peer victimisation and social support.