摘要目的:开发大学生主动性攻击问卷(college students' proactive aggression questionnaire，CPAQ)并对其信效度进行检验。方法:2023年3月至2024年4月，抽取1 552名大学生(样本1：618人；样本2：783人；样本3：151人，间隔6个月后进行重测；样本4：49人：从样本3中随机抽取)开展问卷调查，采用SPSS 24.0和Mplus 8.3，通过项目分析、探索性因素分析、验证性因素分析、内部一致性信度、重测信度、相容效度、区分效度和实证效度检验信度和效度。结果:项目分析结果显示CPAQ中的33个条目与主动性攻击总分呈正相关( r＝0.25～0.78， P<0.05)，各条目在高低分组间的得分均差异有统计学意义(均 P<0.001)。探索性因素分析保留了28个条目，分别归属主动性攻击的3个维度(工具性动机、道德抑制、道德推脱)，验证性因素分析结果显示该问卷的三维度模型拟合良好( χ2/ df＝3.25，CFI＝0.96，TLI＝0.95，RMSEA＝0.05，SRMR＝0.06)。CPAQ总分及其分维度的Cronbach's α系数分别为0.93、0.95、0.89和0.75，重测信度分别为0.64、0.63、0.67和0.60(均 P<0.01)。相容效度检验中，工具性动机维度与反应性-主动性攻击问卷的主动性攻击维度得分呈正相关( r＝0.50， P<0.05)、道德抑制维度与共情、内疚得分呈负相关( r＝－0.30，－0.39， P<0.05)，道德推脱维度与道德推脱量表得分呈正相关( r＝0.58， P<0.05)。区分效度检验中，CPAQ得分与反应性-主动性攻击问卷的反应性攻击维度得分相关性不显著( r＝0.08， P＝0.16)。效标关联效度检验中，CPAQ得分与各个效标问卷得分的相关系数为0.70～0.83。实证效度检验中，主动性攻击行为在不同CPAQ得分组间差异有统计学意义[高分组：(1.16±0.18)分，低分组：(1.04±0.08)分， t＝2.60， P<0.05]。 结论:CPAQ具有良好的区分度、信度和效度，可以作为测量大学生群体主动性攻击行为的工具。

abstractsObjective:To develop college students' proactive aggression questionnaire (CPAQ) and test its reliability and validity.Methods:From March 2023 to April 2024, a questionnaire survey was conducted among 1 552 college students(618 individuals in Sample 1, 783 individuals in Sample 2, 151 individuals in Sample 3 (retested after a 6-month interval), and 49 individuals in Sample 4 (randomly selected from Sample 3)).The reliability and validity of CPAQ were assessed using SPSS 24.0 and Mplus 8.3 software through item analysis, exploratory factor analysis, confirmatory factor analysis, internal consistency reliability, test-retest reliability, compatibility validity, discriminant validity, and empirical validity.Results:The results of the project analysis showed that the 33 items in the CPAQ were significantly correlated with the total score( r=0.25-0.78, P<0.05), and item scores between the high and low group were significantly different (all P<0.001). Exploratory factor analysis retained 28 items, which belonged to three dimensions of proactive aggression (instrumental motivation, moral inhibition, and moral disengagement). The results of confirmatory factor analysis showed that the three-dimensional model of the questionnaire was well fitted ( χ2/ df=3.25, CFI=0.96, TLI=0.95, RMSEA=0.05, SRMR=0.06). The Cronbach's α coefficients for the total score of CPAQ and its sub dimensions were 0.93, 0.95, 0.89 and 0.75, respectively(all P<0.05). The test-retest reliability was 0.64, 0.63, 0.67 and 0.60, respectively(all P<0.01). In the compatibility validity test, the instrumental motivation dimension score was positively correlated with the proactive aggression dimension score of reactive-proactive aggression questionnaire(RPQ)( r=0.50, P<0.05), while moral inhibition dimension was negatively correlated with empathy and guilt scores( r=-0.30, -0.39, P<0.05). The dimension of moral disengagement was positively correlated with the score of the moral disengagement scale( r=0.58, P<0.05). In the discriminant validity test, the correlation of CAPQ and RPQ reactive aggression dimension score was not significant( r=0.08, P=0.16). In the criterion related validity test, the correlation coefficients between the score of CAPQ and each calibration questionnaire and scale were 0.70-0.83. In the empirical validity test, there was a significant difference in the different CPAQ score groups for proactive aggressive behavior(high group: (1.16±0.18), low group: (1.04±0.08), t=2.60, P<0.05). Conclusion:CPAQ has good discrimination, reliability, and validity, which could be used as a tool to measure proactive aggressive behavior among college students.