题名

應用DEA評估行政機關施政計畫績效

并列篇名

Application of DEA for Evaluating the Performance of Public Program

DOI

10.7049/JCHCTC.201210.0011

作者

邱吉鶴(Chi-Ho Chiou)

关键词

資料包絡分析法 ; 施政計畫 ; 績效評估 ; 投入導向 ; 產出導向 ; data envelopment analysis(DEA) ; Public Program ; Performance evaluation ; input-oriented ; output-oriented

期刊名称

健行學報

卷期/出版年月

32卷4期(2012 / 10 / 26)

页次

11 - 27

内容语文

繁體中文

中文摘要

本文試圖解決行政機關應用傳統績效指標法之指標不夠客觀、評估過程冗長、評估結果不易橫向比較及無法即時回饋等問題,採用資料包絡分析法(DEA)評估行政機關施政計畫績效。研究結果發現,DEA方法之相對效率分析,可以快速量化比較不同個案計畫績效,並顯示不同個案計畫其無效率的來源;效率強度分析,可以明顯區分不同個案計畫效率高低;差額變數分析,可以指出不同個案計畫投入與產出項的效率值;因此,DEA方法確實可以解決上述相關部分問題。但是,DEA方法的應用必須考慮:計畫執行過程可能受到不可控制環境因素影響、執行機關是否提供充分正確資料(訊)、評估人員專業能力、及選擇投入與產出評估指標是否適當等,所產生效率值偏差問題。本文研究結果將可提供未來相關研究及行政機關實務應用參考。

英文摘要

In this study, due to the problems of subjectivism, over-long process of evaluation, difficulty for horizontal comparison of the result from evaluation, and incapability to respond immediately; therefore this article adopts application of DEA in order to solve problems above. The result examine that, based on relative efficiency analysis of DEA method, the comparison of different cases plan performance could be quickly quantily, the inefficiency source of different plan could be investigated explicitly; efficiency of intensity, the level of plan efficiency of different cases can be clearly differentiated; slack variable analysis, which can be pointed out that differentcase plan value of the efficiency of input and output, the application of DEA does solve the problems above. However, it should also be taken into consideration about the efficiency deviations generated by uncontrollable environmental effects under the executing process of the project, and the possibilities of uncompleted and incorrect information provided by the executing association, professional level of the evaluators, as well as the appropriation of indexes for evaluating. The results of this article can be the reference for future research, and the practical application for administrative organization.

主题分类 人文學 > 人文學綜合
工程學 > 工程學綜合
社會科學 > 社會科學綜合