少年网络攻击行为评定量表的信效度多少

目的:编制我国少年网络攻击行为评定量表(AOABS),并检验其信效度。方法:方便抽取689名中学生和大学生进行问卷调查,回收有效问卷597份。通过项目分析筛选题目,将整体样本随机分半后分别进行探索性因素分析(n=297)和验证性因素分析(n=300)以检验量表的结构效度,并检验量表的信度。结果:少年网络攻击行为评定量表分为工具性攻击(instrumental aggression)和反应性攻击(reactive aggression)两个分量表,每个分量表内部又分为外显攻击和关系攻击两部分。自编量表的两个分量表经探索性因素分析所得的题目载荷在0.52～0.75之间,累积方差贡献率分别为52.75%和53.24%;两个分量表经验证性因素分析所得模型均符合心理测量学要求(RMSEA0.9,NNFI>0.9,CFI>0.9)。两个分量表的Cronbachα系数分别为0.83和0.86。结论:少年网络攻击行为评定量表具有良好的结构效度和内部一致性信度,可以用于测量我国少年在网络上的攻击行为。; Objective： To develop an Adolescent Online Aggressive Behavior Scale （AOABS） and test its re- liability and validity. Methods： Totally 689 students from junior high schools, senior high schools and universities were selected by convenience sampling method, and were tested with the self-made Chinese version of Adolescent Online Aggressive Behavior Scale. The scale was revised through item analysis. Then the construct validity of the revised scale was tested through exploratory and confirmatory factor analysis, and the reliability was tested. Results： AOABS consists of two subscales, including instrumental aggression （15 items） and reactive aggression （16 i- tems）, and each subscale was divided by two factors of overt aggression and relational aggression. The loadings of i- tems in both subscales were ranged from 0. 52 to 0.75. The two factors in both subscales could explain 52. 8% and 53.2% of the total variance. The indicators of confirmatory factor analysis with both subscales were in requirements of psychometric （ RMSEA 〈 0.08, NFI 〉 0. 9, NNFI 〉 0.9, CFI 〉 0. 9）. The Cronbach α coefficients of both sub- scales were 0. 83 and 0. 86 respectively. Conclusion： It suggests that the AOABS could be used as an effective and reliable tool for assessing online aggression behavior of Chinese adolescents.