台灣夫妻情感互動觀察系統的跨文化轉譯研究

Cross-Cultural Identified Affect Coding System for Observing Marital Affective Interaction in Taiwan

美國學者Gottman 發展特定情感編碼系統（SPAFF），提出夫妻情感表露是預測婚姻穩定性的理論，此系統是夫妻觀察性研究中，具有最佳的構念和同時效度者。但華人文化不鼓勵表達，多迂迴式溝通，使得情感表露的研究相形困難；迄今，夫妻情感表露尚未建立客觀研究方法以反映夫妻互動，故有必要深入探討。本研究目的在於翻譯與建構本土化的特定情感編碼系統，以應用於台灣的夫妻與伴侶研究。在方法上透過跨文化轉譯、專家會議與20對夫妻的小樣本研究，以專家效度、評分者信度與效標關聯效度評估特定情感編碼系統中文版（SPAFF-C-24）的信效度。結果顯示SPAFF-C-24 在10 位專家效度評定中，24 項編碼的內容效度指標（CVI）分布為 .79~ .96，具備良好內容效度。在評分者信度上，兩位編碼員在20 對夫妻的平均一致性分別為85.6%、85.2%，而平均穩定性係數分別為 .53、.57，整體編碼的一致性達85.4%、穩定度達 .55，顯示有可接受的編碼穩定性。在效標關聯效度上，SPAFF-C-24 之負向情感與中性情感和「婚姻親密情感簡效量表」、「婚姻穩定量表」與「夫妻互動量表」各分層面有顯著相關，但正向情感與上述量表各分層面僅有部分相關。整體而言，SPAFF-C-24 可以應用於台灣華人夫妻的情感表露，唯正向情感部分編碼、本土化新增編碼之適當性仍有待未來大樣本研究進一步確認。

The Specific Affect Coding System (SPAFF), developed by Gottman, has been used to predict the stability of a marriage by observing interaction and emotional expression between couples. In recent observational studies of couples, the SPAFF has been recognized as the best evidence of construct and criterion validity. However, habitual behaviors derived from traditional Chinese culture, such as discouraging expression, indirect communication, and emotion suppression, may impede exploration in emotional expression. Therefore, it is necessary to explore the construct of emotional expression and reflect the essence of couple interaction through observational approaches. The aims of this study were to develop a culturally validated Chinese version of the SPAFF (SPAFF-C-24) and examine its appropriateness for use in marital and couple studies in Taiwan. Using a methodology that combines translational study, expert panel review, and small-scale study with 20 couple dyads, we established the reliability and validity of the SPAFF-C-24 through expert validity, inter-rater reliability, and criteria-related validity. The SPAFF-C-24 was reviewed by 10 experts, who found that the Content Validity Index (CVI) ranged from .79~ .96, indicating a reliable instrument with good content validity. Throughout the coder training with 20 couples in a small-scale study, the mean consistency between two coders with the agreement score in husbands and wives were .82 and .70, respectively. The mean interrater stability with Kappa values in husbands and wives were .53 and .57, respectively. Acceptable coding stability was thus endorsed with the overall agreement of 85.4% and its stability with a mean Kappa value of .55 among coders. In the criteria-related validity, the negative/neutral emotional aspects of the SPAFF-C-24 were significantly correlated with the scores in the Brief Version of Chinese Marital Affection Scale, Marital Quality Scale, and Couple Interaction Scale. Nevertheless, only partial correlations were identified between positive emotional aspects of the SPAFF-C-24 and those aforementioned subscales. In conclusion, the newly derived SPAFF-C-24 could be used to observe affective expressiveness among Chinese couples in Taiwan. However, the appropriateness of added codes and positive affects in the SPAFF-C-24 requires further validation through a large-scale study.