摘要: 目的：探究初中生的自我概念和同伴关系对校园欺凌行为的影响，以期从多角度为预防和治理校园欺凌提供有针对性的参考建议。方法：采取方便取样，以儿童自我概念量表(Children’s Self-Concept Scale PHCSS)，Olweus儿童欺凌问卷，儿童青少年同伴关系量表对深圳市某中学初一初二年级450名学生进行团体施测，回收问卷436份，其中有效问卷405份，问卷有效率92.9%。运用SPSS26.0对数据进行独立样本T检验、方差分析、相关分析和多元逐步回归分析。结果：1) 组间差异分析：a) 校园欺凌量表中欺凌分量表得分组间差异：男生组得分显著高于女生组(P < 0.05)；初二组得分显著高于初一组(P < 0.05)；父母婚姻关系不和谐组得分显著高于父母婚姻和谐组(P < 0.01)；朋友数量少于3个组得分显著高于朋友数量多于3个组(P < 0.01)；学习成绩差组受欺凌得分最高，成绩好组次之，成绩中等组受欺凌分数最低(P < 0.01)；b) 校园欺凌量表中欺凌分量表得分组间差异：男生组得分显著高于女生组(P < 0.05)；初二组显著高于初一组(P < 0.05)；朋友数量少于3个校园欺凌量表中欺凌分量表得分组间差异个组得分显著高于朋友数量多于3个组(P < 0.05)；2) 相关分析：a) 自我概念的各个因子及总分与受欺凌呈现显著负相关(P < 0.01)，同伴关系与受欺凌呈现显著正相关，自我概念各个因子及总分与同伴关系呈现显著负相关(P < 0.01)；b) 自我概念中合群因子、行为因子、幸福与满足感因子、智力与学校情况因子、躯体外貌与属性因子(P < 0.05)和自我概念总分(P < 0.01)均与欺凌存在显著负相关，同伴关系与欺凌存在显著正相关(P < 0.01)；3) 多元逐步回归分析：a) 自我概念中的合群、智力与学校情况、行为三个因子和同伴关系、性别、学习成绩对受欺凌有显著预测作用( R
2= 0.304,
= 0.294)，其中合群因子对受欺凌的预测贡献最大；b) 自我概念中的合群因子、行为因子、以及人口学特征中的性别对欺凌有显著预测作用(
R2= 0.132,
= 0.124)，其中行为因子对欺凌预测贡献最大。结论：1) 初中男生校园欺凌行为频次高于女生，初二年级学生校园欺凌行为频次显著高于初一年级，父母婚姻关系不和谐的初中生具有更高频次的校园欺凌行为，朋友数量少的初中生也更容易成为欺凌或被欺凌的对象。2) 自我概念越积极的初中生，校园欺凌发生频次越高，同伴关系越差的初中生校园欺凌的频次越高，同伴关系越好的初中生自我概念越积极。3) 合群自我概念对受欺凌的影响最大，行为自我概念对欺凌的影响最大。
Abstract:
Objective: To explore the influence of junior high school students’ self-concept and peer relationship on school bullying behavior, in order to provide targeted reference suggestions for prevention and management of school bullying from multiple perspectives. Method: Using Children’s Self-Concept Scale (PHCSS), Olweus Children’s Bullying Questionnaire, and Children and Adolescents Peer Relationship Scale, 450 students in grades 1 and 2 of a middle school in Shenzhen were tested. 436 questionnaires were collected and there were 405 valid questionnaires, with an effective rate of 92.9%. SPSS26.0 was used to conduct independent sample T-test, variance analysis, correlation analysis and multiple stepwise regression analysis. Results: 1) Analysis of differences between groups: a) The score of bullying subscale in the school bullying scale was different between groups: the score of the male group was significantly higher than that of the female group (P < 0.05); the score of junior 2 group was significantly higher than that of junior 1 group (P < 0.05); the score of parents’ marital disharmony group was significantly higher than that of parents’ marital harmony group (P < 0.01). The score of group with fewer than 3 friends was significantly higher than that of group with more than 3 friends (P < 0.01). The group with poor academic performance had the highest score of bullying, followed by the group with good academic performance, and the group with medium academic performance had the lowest score of bullying (P < 0.01). b) The subscale of bullying in the school bullying scale was different between groups: the score of the male group was significantly higher than that of the female group (P < 0.05); the second group was signifi-cantly higher than the first group (P < 0.05); groups with fewer than 3 friends scored significantly higher than groups with more than 3 friends on the bullying subscale of the bullying scale (P < 0.05); 2) Correlation analysis: a) Each factor and total score of self-concept were negatively correlated with bullying (P < 0.01), peer relationship was positively correlated with bullying, and each factor and total score of self-concept was negatively correlated with peer relationship (P < 0.01); b) In self-concept, social factors, behavioral factors, happiness and satisfaction factors, intelligence and school situation factors, physical appearance and attribute factors (P < 0.05) and total self-concept score (P < 0.01) were significantly negatively correlated with bullying, and peer relationship was significantly positively correlated with bullying (P < 0.01); 3) Multiple stepwise regression analysis: a) The three factors of self-concept, social, intelligence and school situation, behavior, peer relationship, gender and academic performance, had significant predictive effects on bullying ( R
2= 0.304,
= 0.294), and the social factor contributed the most to the prediction of bullying; b) Social factors, behavioral factors and gender in self-concept were significant predictors of bullying ( R
2= 0.132,
= 0.124), and behavioral factors contributed the most to the prediction of bullying. Conclusions: 1) The frequency of bullying behavior of junior high school boys is higher than that of girls, and the frequency of bullying behavior of junior high school students in grade two is significantly higher than that of junior high school students in grade one. Junior high school students whose parents have disharmonious marital relationship have higher frequency of bullying behavior, and junior high school students with fewer friends are more likely to be bullied or bullied. 2) The more positive the self-concept of junior high school students, the higher the frequency of school bullying, the worse the peer relationship of junior high school students, the higher the frequency of school bullying, the better the peer relationship of junior high school students, the more positive the self-concept. 3) Gregarious self-concept had the greatest influence on bullying, and behavioral self-concept had the greatest influence on bullying.