社会排斥与攻击的关系：一项元分析

s a kind of negative social behavior, aggression not only brings harm to others, but also disturbs social harmony and order and threatens public security. Numerous studies have discussed the inducing factors of aggression, among which social exclusion has been widely concerned as a strong predictor. However, the studies about the association between social exclusion and aggression have shown mixed results. Theoretically, there are two main arguments about the relationship between social exclusion?and aggression. The general aggression model suggests that social exclusion triggers negative emotions, hostile cognition, or a high level of physiological arousal, which leads to aggression. While the emotional numbness hypothesis?argued that social exclusion causes individuals to be a state of physiological or emotional numbness, which avoids the occurrence of aggressive behaviour. Empirically, the effect sizes of this relationship reported in the existing literature were far from consistent,?with r?values ranging from -0.02 to 0.74. Therefore, this meta-analysis was conducted to clarify the strength of the relationship between social exclusion and aggression and reveal possible moderators.?Through literature retrieval, 92 studies (99 independent?effect sizes, a total of 65564 subjects) met the inclusion criteria were selected. A random-effects model was selected to conduct the meta-analysis in Comprehensive Meta-Analysis 3.3 software, aiming at testing our hypotheses. The results of heterogeneity test illustrated that there was significant heterogeneity among 99 independent effects, indicating that the random effect model is suitable for subsequent meta-analysis. Based on the funnel chart and Egger’s regression test of intercept, no significant publication bias was found in the included studies.The main effect analysis indicated a significant positive correlation between social exclusion and aggression (r?= 0.38, 95% CI [0.34, 0.41]); The moderation analyses revealed that the relationship between social exclusion and aggression was moderated by age, research methods, aggression types (reactive aggression vs. proactive aggression), social exclusion and aggression measurement tools, and research design (cross-sectional study vs. longitudinal study), but not by individualism index and gender.The results of this meta-analysis can clarify the controversy of the correlation between social exclusion and aggression, indicating that excluded individuals are often accompanied by the increase of aggression level. Future research are encouraged to strengthen the prevention and intervention of aggressive behavior caused by social exclusion among low-age groups, especially preschool children and primary school students.