國中生學習行動控制模式之驗證及行動控制變項與學習適應之關係

The Verification of Action Control Model and the Relationship among Action Control Variables and Learning Achievement

本研究的目的在:（l）建構一個國中生學習行動控制模式，並蒐集實際的觀察資料來驗證此一模式的適配度;（2）根據觀察資料來建立羅吉式迴歸預測模式，以預測國中生的學習成就與教師評定的學習適應。受試者是臺灣北部及中部地區I3所國中30個班級共780名學生及他們的班級導師。受試學生接受行動導向量表、行動控制策略覺察量表、行動控制策略使用量表、和情意反應量表的測量，班導師則接受教師評定量表的施測。
研究結果顯示:（l）本研究根據理論與實徵研究結果所提出的「國中生學習行動控制模式」在整體適配度考驗上雖然卡方值達顯著水準，但是其他七項重要的適配度考驗指標（如CFL, AGFL, NFI, NNFI, CFI, IFI, RFI等）都顯示模式與觀察資料可以適配，內在結構適配度也顯示理想的模式內在品質，潛在變項之間的效果考驗則支持本研究的假定。（2）本研究根據實微資料所建立的二個羅吉式迴歸預測模式都能有效預測學習成就和教師評定的學習適應這二個效標變項:預測變項對二個效標變項的正確區別率分別是67.95%和69.10%，而在預測變項的相對預測力上，觀察資料也顯示支持理論文獻的結果。

The purposes of this study were: (1) to verify the goodness of fit between empirically observed data and Action Control Model proposed by the author, and (2) to establish the logistic regression model with observed data. The participants were 780 students and 30 teachers from 13 junior high schools in Taiwan middle and northern area. The instruments used in this study include: Action Orientation Inventory, Awareness of Action Control Strategies Inventory, Action Control Strategies Inventory, Affective Response Inventory, and Teacher Rating Scale. The collected data were analyzed with LISREL and logistic regression analysis. The results of this study were as follow:.(1) Although the value of cit-square was significant at .05 level, other overall and internal structure model fit indices indicated that the Action Control Model fitted the observed data well. (2) Two logistic regression models proposed by the author could effectively predict students' learning achievement and teacher rating scores. The correct classification rate, 67.95% or 69.10%, was significantly higher than that expected by chance alone.