中文版兒時創傷量表——簡版之因素結構與信度分析研究

Reliability and Factor Structure of the Chinese Version of Childhood Trauma Questionnaire-short Form in in Patients with Substance Use Disorder

目的：簡式兒時創傷問卷是目前最廣泛被用來評估多面向兒時創傷經驗的問卷，已有多 種語言版本的簡式兒童創傷問卷經過信效度分析，本研究目的為針對此中文化問卷進行問卷 內容的信度分析以及結構分析。方法：以 160 位物質濫用個案為樣本，進行簡式兒時創傷問 卷測量，並在首次施測二至三周後，進行再次問卷測量。針對問卷資料進行信效度分析，以 Cronbach α 數值，檢測內部信度，以 intraclass correlation coeffi cient 數值，檢測再測信度。 並以驗證性因素分析與探索式因素分析分別進行問卷的因素結構分析。結果：本研究共招 募 160 位物質濫用病人。問卷分析結果顯示中文版兒時創傷量表簡版有良好的內在一致性 （Cronbach’s α 數值在 0.574-0.895 之間）與再測信度（intraclass correlation coeffi cient 數值在 0.674-0.852 之間）。驗證性分析顯示在身體忽略之次分項上，有較低的因素負荷量。探索式 因素分析驗證了原先問卷的五個層面的次分項，另探索性分析顯示某些題目有交叉負荷量。 結論：本研究中的中文版兒時創傷量表簡版在物質濫用個案具有良好之信效度，可用以評估 兒時創傷經驗。另外中文問卷結果與先前文獻一致，顯示身體忽略此次分項在信效度上相對 而言較為不穩定。

Objectives: Childhood Trauma Questionnaire-Short Form (CTQ-SF), a 28- item retrospective self-report questionnaire, is the most widely used measure to assess multiple dimensions of childhood trauma exposure (CTE). The CTQ-SF has been translated into different languages, but the Chinese version have not yet been studied. In this study, we intended to examine the reliability and factor structure of the Chinese version of the CTQ-SF (C-CTQ-SF). Methods: We administered the questionnaire of the C-CTQ-SF to 160 patients with substance use disorder. The internal consistency and test–retest reliability of the C-CTQ-SF at 2–3 weeks after initial assessment were evaluated using Cronbach's α coeffi cients and intraclass correlation coeffi cients (ICCs). We also did confi rmatory factor analysis (CFA) and exploratory factor analysis (EFA), to test the factor structure of the C-CTQ-SF. Results: The C-CTQ-SF was found to have fair to adequate internal consistency (Cronbach's α ranging from 0.574 to 0.895) and test–retest reliability (ICCs ranging from 0.674 to 0.852). The results of CFA showed relatively low factor loading for the physical neglect subscale. Subsequent EFA yielded a stable fi ve-factor structure, and some items had cross-loadings with more than one factor. Conclusion: Our results indicated that C-CTQ-SF has adequate reliability and a fi ve-factor solution, supporting the feasibility of using the C-CTQ-SF to assess CTE in patients with substance use disorders. Consisting with previous reports, the physical neglect subscale did not emerge as a stable factor.