earticle

영어

The purpose of this study was to synthesize journals that studied about causal factor of school violence for 11 years by using meta-analysis. There were many factors having effects on school violence, so that results of these factors need to be integrated by conducting meta-analysis. By using R package, the effect sizes of correlation coefficient of variables related to school violence were calculated and the relevance of several potential moderator variables was examined. This meta-analysis integrated the results of 18 journals with 18 variables under 5 categories. Each of 18 variables under 5 categories were divided by an individual, family, school, peer and community factor. The school factor had the largest effect size and the community factor had the fourth effect size among the 5 categories. In categories, the problem behavior variable had the largest effect size and domestic violence, a victim of school violence, peer conformity, harmful environment were also large in effect size in each of categories. Because some of variables have yielded heterogeneous results, we examined moderating effects using meta ANOVA. Moderator analysis using meta ANOVA showed the school type and sex configuration influenced on the effect size. This study confirmed that the school and community factors has a notable effect on school violence. Finally, the implications of these findings were discussed in terms of school violence prevention.