AHP和CPC之比較--以教育資料為例

Abstract

儘管層級分析法（AHP）已被廣泛應用於國內教育和其他領域建構指標及決策的研究中，但仍有一些分析方法的問題需要解決。近年來，論者針對該法提出不少改良的新方法，其中連續成對比較法（CPC）有簡化成對比較次數、運算簡易等優點，具參考應用的價值，值得注意。本研究自理念、分析流程、方法、應用現況等方面比較兩者，發現兩者有相似的理念和分析流程，但是分析方法截然不同，AHP較為完整且被廣泛採用，CPC則嫌簡略且無人應用。以實際教育資料考驗，發現兩者分析結果不相同，CPC在輸入值的邏輯一致性、輸出權重值及其次序的誤差等方面，均明顯優於AHP。

Although Analytic Hierarchy Process (AHP) has been widely used in the education and other areas studies of index and decision-making in Taiwan, it still has some methodological problems needed to be resolved. There are some new methods have been proposed recently to improve the AHP. The Chainwise Paired Comparisons (CPC) is the noticeable one, which reduced the frequency of paired comparison and operated simply. The study compared AHP with CPC in analytic ideas, procedure, operation, and application. It found that AHP was similar to CPC in analytic ideas and procedure but was different in operation. AHP has a complete analytic system and was widely used, but CPC was not. Using practical educational data to test, it showed that AHP ’s outcomes were different from CPC. In the consistency of input data and the item errors in weights and its orders, CPC were more excellent than AHP.