以TIMSS 資料檢視能力信念與任務價值對臺灣八年級學生數學成就之影響
No Thumbnail Available
Date
2013-09-??
Authors
陳敏瑜
游錦雲
Min-Yu Chen
Ching-Yun Yu
Journal Title
Journal ISSN
Volume Title
Publisher
國立臺灣師範大學
National Taiwan Normal University
National Taiwan Normal University
Abstract
本研究使用國際數學與科學成就趨勢調查（Trends in International Mathematics and ScienceStudy, TIMSS）2007年臺灣八年級學生的資料，以期望價值理論為架構，先進行能力信念與價值相關構面及題項的信度與效度分析，接續探討這些構面對數學成就之影響，並以多群組結構方程模型分析男、女生模型之差異。研究發現，數學能力信念、實用與內在價值三構面及其對應的題項都有良好的信度與效度，三構面中以能力信念的預測力最高，能解釋數學成就約三成六的變異量。男、女生模型在因素負荷量、題項截距、路徑係數及因素變異數／共變異數上皆具跨性別不變性，不過，在構面平均數上，男生的數學能力信念、實用與內在價值的平均數都顯著較女生高，且以在能力信念的差異最大。最後依據結果提出實務應用及未來研究上的建議。
Based on expectancy-value theory, we applied trends in mathematics and science study (TIMSS) data to investigate the reliability and validity of items relating to ability beliefs and task values, examine their effects on mathematical achievements, and test gender invariance in the proposed models by using multiple-group structural equation modeling. The results supported a three-factor solution reflecting ability beliefs, utility values, and intrinsic values. These factors and corresponding items all possessed strong reliability and validity. Among the three factors, ability beliefs exerted the strongest effect on mathematical achievements, explaining 36% of the variance of mathematical achievements. Gender invariance evidence was exhibited in the factor loadings, item intercepts, path coefficients, and factor variance/covariance. However, comparisons of latent factor means suggested that boys had significantly high mean scores regarding ability beliefs, utility values, and intrinsic values. Finally, suggestions for practical applications and future research are provided.
Based on expectancy-value theory, we applied trends in mathematics and science study (TIMSS) data to investigate the reliability and validity of items relating to ability beliefs and task values, examine their effects on mathematical achievements, and test gender invariance in the proposed models by using multiple-group structural equation modeling. The results supported a three-factor solution reflecting ability beliefs, utility values, and intrinsic values. These factors and corresponding items all possessed strong reliability and validity. Among the three factors, ability beliefs exerted the strongest effect on mathematical achievements, explaining 36% of the variance of mathematical achievements. Gender invariance evidence was exhibited in the factor loadings, item intercepts, path coefficients, and factor variance/covariance. However, comparisons of latent factor means suggested that boys had significantly high mean scores regarding ability beliefs, utility values, and intrinsic values. Finally, suggestions for practical applications and future research are provided.