本研究主要在探索影響開發中國家的教育品質分析。本研究運用迴歸分析與線性結構模型檢定影響因素與教育品質之問因果關係。本研究以公元二千年「世界教育報告書」中1997年資料,選定7項教育品質指標與7個影響教育品質因素進行研究。本研究建構3個迴歸模型與3個結構方程模型,共提出6個虛無假設,以考驗潛在變項之間的因果關係。結果顯示,迴歸模型檢驗出最重要的影響教育品質因素;而結構方程模型,即以潛在變項因果模型整體檢定指標,即卡方考驗未達顯著,且AGFI、GFI、CFI、IFI值都接近1等判定標準。表示三個結構方程模型的經濟、人口、教育量與教育品質潛在變項之間適配度良好。也就是說,1997年開發中國家,影響教育品質因素影響教育品質潛在變項因果關係存在,但以模型內在評估指標衡量三個結構方程模型,發現仍有若干變項有誤差存在。據研究結論提出相關建議。
The main purpose of this study is to explore the causal relationships among the latent variables of population, economic, educational quantity, and educational quality in developing countries. The research employed the multiregression analysis and LISREL to test the causal relationships of those variables. It used seven variables of the educational quality, and six influencing variables of the educational quality for analysis. Those data are collected from World Education Report (2000), and one year (1997) data is selected for analysis. The research constructed three regression models and three structural models. Also it provided six null hypotheses to test the casual relationships among those variables. The results showed that there are some important factors in regression models, and the x(superscript 2) values of three causal models in structural equations are not significant. In addition, the criteria values of AGFI, GFI, NFl, CFI, and IFI are also near 1.0 in structural equations. That is, those models are well fitted in this study, and rejected three null hypotheses. It meant that the casual relationships among the latent variables of population, economic, educational quantity, and educational quality are well fitted. But according to other criteria, there are some error terms in the research models. From that finding, it provided some suggestions.