本研究以TIMSS 1999釋出之數據進行次級資料分析(secondary data analysis),針對台灣5,690位八年級學生的調查結果,以結構方程模式檢驗Marsh和Köller(2003, 2004)整合模式應用於解釋數學和科學成就與其自我概念關係之適用性。在橫斷性資料之條件下,本文將整合模式簡化為內外在參照架構模式(internal/external frame of reference model;簡稱I/E模式)與動機資源競爭模式(motivation-resource competition model;簡稱MR模式)。I/E模式認為學業成就對於同領域的學業自我概念有正向預測力,對不同領域的學業自我概念有抑制效果;MR模式則認為學業自我概念對於同領域的學業成就有正向預測力,而對不同領域的學業成就有抑制作用。為因應數學與科學學科特質,本研究提出MR的修訂模式,分析結果顯示模式路徑係數皆符合預期。
This study is based on a secondary analysis of data collection from 5,690 Taiwanese 8th graders in TIMSS 1999. Marsh and Köller's (2003, 2004) Unification model was examined by using a structural equation modeling technique. Due to limitation of cross-sectional data, the Unification model was further divided into an internal/external frame of reference model (I/E model) and a motivation-resource competition model (MR model). In support of the I/E model, math achievement had a positive effect on math self-concept but a negative effect on science self-concept. On the other hand, science achievement had a positive effect on science self-concept but a negative effect on math self-concept. Consistent with the revised MR model, self-concept in one domain (math or science) had a positive effect on achievement in the same subject, but a negative effect of achievement in another domain (science or math). The results indicated that I/E model and modified MR model could be generalized to math and science subject areas.