中文版感覺史量表發展暨信效度研究

Abstract

研究目的：感覺史量表已在多個國家被廣泛地用來評估孩童的感覺問題。本研究將翻譯感覺史量表為中文版本、檢驗其心理計量特性，並進一步使用中文版感覺史量表來探討正常發展孩童與自閉症孩童對一般感覺事件的反應特徵。 方法：翻譯感覺史量表為中文版本，並由三位專家進行內容效度審查。949位3-10歲正常發展孩童的家長有效填答中文版感覺史量表（平均年齡7.3歲，標準差2.42，482位男童，467位女童）。內部一致性取樣樣本為949位孩童，間隔三星期之再測信度取樣30位自閉症孩童（平均年齡5.80歲，標準差2.07，24位男童，6位女童）。區辨效度則是比較53位自閉症孩童與53位正常發展孩童在感覺史量表分數上差異，並事先配對其年齡（平均年齡5.83，標準差1.53）與性別（82男童，24女童）。建構效度則是使用驗證性因素分析來檢驗原版感覺史量表的九因素模式是否適用於台灣孩童。 結果：70％的中文版感覺史量表題目對正常發展孩童而言是不尋常的行為。中文版感覺史量表具備可接受∼好的內部一致性(Cronbach’s α=.62 -.90)、好的總分之再測信度(ICC=.79)，不佳∼好的分測驗分數及因素分數之再測信度(ICC=.44 - .90)。驗證性因素分析則支持原本的九因素模式(SRMR = 0.068, RMSEA = 0.063, NNFI = 0.94, and CFI =0.94)。中文版感覺史量表的所有分測驗及因素均能區辨正常發展孩童與自閉症孩童，約50％題目可區辨這兩個族群。 結論：本研究結果顯示，中文版感覺史量表是一個可信且有效的工具，適用於評量台灣孩童的感覺處理功能。

Objects: The Sensory Profile (SP) has been widely used in evaluating a child’s sensory processing in Western countries. In this study, we translated, culturally adapted the SP into Chinese, and examined the psychometric properties for its cross-culture use in Taiwan. Furthermore, the responses to common sensory events of typically-developing and autistic children were evaluated by using the Chinese version SP. Methods: The SP was translated and cross-culturally adapted into Chinese. Content validity was evaluated by 3 experts. Parents of 949 typically-developing children with the age of 3 to 10 years (mean age = 7.3 years, SD = 2.42, 482 males, 467 females) completed the adapted questionnaires. Internal consistency was determined using the data of 949 typically-developing children. Test-retest reliability was examined using 30 children with autism (mean age = 5.80 years, SD = 1.07, 24 males, 6 females) in a 3-week interval. Discriminate validity were examined by comparing the Chinese version SP scores between 53 children with autism and 53 children without autism matched for age (mean age = 5.83 years, SD = 1.53) and gender (82 males, 24 females).Construct validity was examined using confirmatory factor analysis to investigate if the 9-factor model of the original SP was applicable to Taiwanese data. Results: Seventy percents of items on the Chinese version SP were uncommon behaviors for typically-developing children. The Chinese version SP demonstrated acceptable to good internal consistency (Cronbach’s α=.62 -.90), good test-retest reliability for total scores (ICC=.79), and poor to good test-retest reliability for each section scores and each factor scores (ICC=.44 - .90). Confirmatory factor analysis of the Chinese version SP provided strong support for a nine-factor model (SRMR = 0.068, RMSEA = 0.063, NNFI = 0.94, and CFI =0.94). The performance of children with autism was significantly different from that of typically-developing children on all sections and factors. Nearly 50% of items contributed to the significant difference between groups. Conclusions: Findings from the studies suggest that the Chinese version SP is a reliable and valid instrument for evaluating children’s sensory processing functions in Taiwan.