想法行为混淆量表在中国大学生中的修订

Validation of the Thought-Action Fusion Scale among Chinese college students

目的：在大学生群体中修订想法行为混淆量表（Thought-Action Fusion Scale,TAFS）,并考察其信效度.方法：采用方便取样方法,对样本1的378名大学生施测TAFS、贝克抑郁问卷、状态特质焦虑问卷、Frost完美主义问卷、多元个人传统性量表,以检验量表的效度,并随机选取其中42名学生,在初测4周后进行了重测以检验重测信度;对样本2的273名大学生施测TAFS、白熊压抑量表、帕多瓦量表,以验证强迫观念认知模型的跨文化有效性.结果：对样本1的探索性因素分析获得TAF量表的3个因素（道德、可能性-自我和可能性-他人）,3个因素可解释总变异的51%（各条目在所属因子的因素负荷为0.418～0.847）;对样本2的验证性因素分析显示,三因素模型拟合良好（χ2/df =2.25,CFI=0.92,TLI=0.90,RMSEA=0.07,SRMR=0.06）.TAFS总分的内部一致性系数为0.86,重测信度为0.67;3个因素的内部一致性系数分别为0.84、0.83和 0.86.效标效度分析表明TAFS总分和TAF可能性分量表均与抑郁（r=0.19、0.27）、完美主义（r=0.25、0.27）、迷信信念（r=0.25、0.26）呈正相关,均P＜0.05;TAF可能性分量表与特质焦虑呈正相关（r=0.22,P＜0.05）;TAF道德分量表与以上量表均无相关（P〉0.05）.结构方程模型的结果证实了以下模型：想法行为混淆导致思维压抑,而思维压抑导致强迫症状（χ2/df=1.285,CFI=0.988,GFI=0.990,TFL=0.982,RMSEA=0.035）.结论：想法行为混淆量表的中文修订版具有良好的信度和效度,可以用来测量对强迫症症状的发展及维持起作用的两种认知偏差.

Objective： To examine the psychometric properties of the Chinese version of Shafran et al. ＇s （1996） Thought-Action Fusion Scale （TAFS） . Methods： In order to examine the validity of the TAFS, a convenience sample of 378 college students completed the Beck Depression Inventory, State-Trait Anxiety Inventory, Frost Multidimensional Perfectionism Scale, a measure of Chinese superstition and the TAFS. In order to test the cross-cultural validity of the cognitive model of obsessions, a second sample of 273 students completed the White Bear Suppression Inventory, Padua Inventory, and the TAFS. Results： CFA results provided support for a 3-factor model （χ2/df = 2.25, CFI = 0.92, TLI = 0.90, RMSEA = 0.07, SRMR = 0.06） . Internal consistency and test-retest coefficients for the TAFS-total were 0. 86 and 0. 67, respectively. Criterion-related validity results could be summarized as follows. TAFS-total scores and TAF-Likelihood scores showed positive significant correlations with scores of depression （ r = 0. 19, 0. 27 ）, perfectionism { r = 0. 25, 0. 27 ）, and superstition （ r = 0. 25, 0. 26 ） . TAF-Likelihood scores also showed a positive significant correlation with anxiety scores （ r = 0. 22 ）, and TAF-Morality scores revealed very low correlation coefficients with the constructs mentioned above. Furthermore, SEM results confirmed a model wherein thought suppression mediates the relationship between TAF and obsessivecompulsive symptoms （χ2/df = 1.29, CFI = 0.99, GFI = 0.99, TFL = 0.98, RMSEA = 0.04） .Conclusion： The present findings provide support for the validity and reliability of the Chinese version of the TAFS. Theoretical, clinical and cross-cultural implications are discussed. Limitations and directions for future research are identified as well.