以試題反應理論分析CES-D 量表不同調查方式的差異效果

Abstract

本研究主要是探討網路問卷與紙本問卷這兩種調查方式是否有所差異，問卷內容是採用國際間廣泛使用的CES-D 憂鬱量表。2400 位樣本分層隨機取自台灣中小學教師，其中共有1141位教師在20 題CES-D 量表上作答，包括535 位網路組及606 位紙本組教師。本研究藉由單參數試題反應理論軟體ConQuest 開發的多向度潛在回歸模組來比較網路問卷組與紙本問卷組在問卷內容上的差異情形。研究結果顯示，儘管網路問卷回收率稍低於紙本問卷，但並未導致兩組受試者在背景變項上有所差異。在問卷內容的比較上，雖然網路問卷組在問卷內容的潛在平均數顯著地低於紙本問卷組，但是並無實質上的重要差異，亦即問卷組別變項對CES-D 量表各因素的解釋力趨近於0。本研究認為仍有必要持續進行調查方式的比較研究，以確認網路問卷是否可取代紙本問卷，以減少研究及社會成本。

This study investigated whether differences exist between Intemet and paper questionnaires in survey mode, wherethelilternationa l1y- and widely-used “ Center for Epidemiological Studies Depression" (CES-D) scale was adopted as content of thequestionnaire. Of 2400 primary and secondary school teachers stratified and sampled randomly from Taiwan, 1141 teach ers responded to the20_item CES-D Scale, including 535 teachers in theInternet group, and 606 teachers in thepaper group. This research used multidimensional latent regression module deVeloped by ConQu est softwar e. Using one-parameter item response theory, we compared the differences between the Intemet and paper questionnaire groups. Theresuits of this study showed no significant differencebetweenthetwo groups in background variables; nonetheless; response ratein theInternet group was lower than that in the paper group. As for comparisons between content of questionnaires, although the latent means of the Internet group were significantly lower than those of the paper group, these differences had no substantial meanings. The group variable of survey mode explained closeto zero vanancein factors of the CES-D ScaleTo examine whether Intemet questionnaire survey can be used to substitute for paper survey in order to reduce research and administration cost, we encourage more comparisons between the two survey modes.