中文版儿童期创伤问卷在精神障碍患者中信度、效度分析

Reliability and Validity of Chinese Version of Childhood Trauma Questionnaire in Patients with Mental Disorders

目的：研究在精神障碍患者中,儿童期创伤问卷的信度、效度,为探索精神障碍患者的童年创伤经历提供更加准确的测量工具。方法：在芜湖市某一精神科医院,采用整群抽样的方法,由病区医生协助研究者组织患者,采用统一的指导语,运用儿童期创伤问卷对病区患者进行测评。选择芜湖市某一大专,由辅导老师组织并采用统一的指导语施测。采用Epidata 3.1进行数据的录入,采用SPSS 17.0中的检验对两组被试的人口学变量进行分析,用可靠性检验问卷的信度,用独立样本t检验探索问卷的区分效度,用相关分析检验问卷的结构效度,采用AMOS 22进行验证性因子分析。结果：在精神障碍患者组中,儿童期创伤问卷总量表的Cronbachα系数为0.824,分半信度为0.812,除躯体忽视这一维度的信度不理想（Cronbachα=0.556,split-half=0.562）外,其余4个维度的信度均在0.7以上。儿童期创伤问卷各维度与总量表之间的相关在0.587-0.862,有统计学意义（P〈0.01）,问卷各维度之间的相关性显著,高分组与低分组在量表总分及各维度上均有显著差异。验证性因子分析显示,在一阶验证性因子分析模型中,CMIN/DF〈3,说明该模型的拟合度可接受,GFI、IFI、CFI均接近于0.9,RMSEA〈0.08,表示模型可勉强接受。在二阶验证性因子分析模型中,CMIN/DF〈2,优于一阶验证性因子分析模型,GFI、FI、CFI、RMSEA各项指标均优于一阶验证性因子分析模型。结论：中文版儿童期创伤问卷可以应用于对精神障碍患者童年经历的调查。

Objective：To research reliability and validity of Chinese version of childhood trauma questionnaire in patients with mental disorders,to provide a more accurate measurement tool for exploring the childhood trauma experience of patients with mental disorders.Methods：In a psychiatric hospital in Wuhu,a cluster sampling method was adopted.The patients in the ward were organized by the ward doctors,and the patients were evaluated by using the unified guidance language and the childhood trauma questionnaire.We selected a college in Wuhu,which was organized by the tutor and we adopt a unified guide to test.Epidata 3.1 was used for data entry.The demographic variables of the two groups were analyzed by test in SPSS17.0 reliability test was used for the reliability of the questionnaire,independent sample t test was used to explore the discriminant validity of questionnaire,construct validity of the questionnaire was analyzed by correlation analysis,SPSS17.0 analysis data and AMOS22 were used for confirmatory factor analysis.Results：In a group of patients with mental disorders,the total number of childhood trauma questionnaire Cronbach alpha coefficient was 0.824,the split half reliability was 0.812,the reliability of emotional abuse and physical abuse,sexual abuse,emotional neglect were above 0.7,only physical neglect was not so well（Cronbachα=0.556,split-half=0.562）.The correlation between each dimension and total scale were in the range of 0.587-0.862,with statistical significance（P〈0.01）.The high group and the low group had significant differences on scale total score and the dimensions.Confirmatory factor analysis indicated that the fitting degree of the first-order confirmatory analysis model is acceptable（CMIN/DF3,GFI,IFI,CFI are close to 0.9,RMSEA0.08）.A second-order confirmatory factor analysis model was better than the first-order confirmatory factor analysis model（CMIN/DF2）,with better GFI,FI,CFI,RMSEA indicators.Conclusion：The Chinese version of childhood trauma questionnaire in patients with mental disorders has good reliability and validity.