本研究藉由文獻回顧歸納重要-表現度分析(IPA)在設計上的限制及運用上的優缺點，並比較許多改良式IPA方法的內容，及評估直接與間接量測法之差異。在IPA運用過程，劃分象限中心點方式會明顯影響屬性最後的象限配置，進而決定資源的規劃策略。因此，本文選擇四種劃分象限中心點之方式，包含重要-表現度缺口分析(IPGA)、數據中心、刻度中心及對角線方法，並以實際統計數據評估四種方式在屬性象限配置上之精準度。最後利用IPGA方法計算相對重要度與相對表現度之數值，提升量測服務屬性的客觀性及準確度，並計算改善項目之量化排序，提供管理者更精細的評估結果與解讀空間，可優化資源分配的決策品質。本研究以某大學體育課程教學服務品質為分析個案，利用IPA及IPGA方法，評估學生對課程的期待(重要度)與體驗(表現度)結果。研究發現IPGA方法對屬性象限之分布預測最具準確性，其次是對角線結合數據中心劃分象限法，刻度中心劃分象限法之精準度最低。IPA缺口分析結果顯示學生對課程重要度與表現度之間是有明顯落差(負值缺口)，顯示學生對體育課程的期望水準高於實際體驗的品質，存在需要改善的空間。IPGA方法發現座落於第Ⅱ象限(優先改善)之問項共21項，其餘9項位置落於第Ⅲ象限(次要改善)，第Ⅰ象限及第Ⅳ象限並無問項存在，所有問項T檢定結果均達到顯著值水準。利用IPGA公式，可計算改善項目座標與原點之距離，距離越大者，改善排序越優先，學校可藉此規劃資源分配策略，以提升教學服務品質。

This study summarizes the constraints and operational advantages and disadvantages of the Important-performance analysis (IPA) by literature review, and compares the content of the improved IPA methods and the differeance of direct and indirect rating methods. In the IPA application process, dividing the quadrant crosshair mode will significantly affect the last quadrant configuration of the attribute, and then determine the resource planning strategy. Therefore, this paper chooses the four methods of dividing the crosshair-point of the quadrant, including the Importance-performance and gap analysis (IPGA), data-centered, scale-centered and diagonal-line methods, to evaluate their performance with actual statistics for improving the precision of attribute configuration in IPA quadrant. Finally, the IPGA method is used to calculate the relative importance and relative performance values, improve the objectivity and accuracy of the measurement on service attributes, as well as calculate the quantitative priority of the improvement actions. The findings of this study provide managers with more detailed evaluation results and interpretation space, and optimize the quality of the decision-making of resources allocation. This study uses the IPA and IPGA methods to evaluate students' expectations (importance) and experience (performance) results based on a case study of the university’s physical curriculum in Taiwan. The study found that the IPGA method has the most accurate prediction of the distribution of attribute quadrants, followed by the diagonal-line combined with data-centered method, and the scale-centered method has the lowest accuracy. The results of the IPA gap analysis show that there is a significant gap between the students' importance and performance (negative value gap), indicating that students' expectations of physical curriculum are higher than the actual experience quality, and there is space for improvement. The IPGA analysis found that there were 21 questions in the second quadrant (concentrate here), the remaining nine were in the third quadrant (low priority), and no item existed in the first and the fourth quadrant. All questions reached a significant level of t-test. Using the IPGA formula, the distance between attribute coordinates and original point can be calculated, thus the greater the distance, the higher the priority is. The school can use this result to plan resource allocation strategies for improving the quality of teaching services.