综合评价结果的区分度度量新方法  

A New Method For Discriminability Measurement of Comprehensive Evaluation Results

作　　者：李兴奇[1] 高晓红[2] Li Xingqi;Gao Xiaohong(School of Management and Economics,Chuxiong Normal University,Chuxiong Yunnan 675000,China;School of Mathematics and Computer Science,Chuxiong Normal University,Chuxiong Yunnan 675000,China)

机构地区：[1]楚雄师范学院管理与经济学院,云南楚雄675000 [2]楚雄师范学院数学与计算机科学学院,云南楚雄675000

出　　处：《统计与决策》2022年第21期31-35,共5页Statistics & Decision

基　　金：国家自然科学基金资助项目(11261001);云南省应用基础研究计划青年项目(2017FD152)。

摘　　要：文章基于综合评价区分度的定义,分析了传统区分度度量方法的不足,并对其进行改进,提出了多种新的区分度度量方法,最后通过模拟实验来检验各种方法的有效性。结果表明:传统区分度度量方法未能反映任意两个评价值间的可区分性,不利于准确度量综合评价结果的区分度;两两平方差法和两两离差法虽考虑了所有评价值间的可区分性大小,但计算量太大,且会出现评价值的区分度相互线性补偿的问题;相邻平方差法和相邻离差法虽有效简化了两两平方差法和两两离差法的计算量,但同样会出现评价值的区分度相互线性补偿的问题;经过不断改进得到的相邻对数离差法既解决了区分度相互线性补偿的问题,又能减小计算量,方便比较,是一种效果较好的区分度度量方法。This paper is based on the definition of comprehensive evaluation discrimination to analyze the shortcomings of the traditional discriminability measurement methods,then improves them to propose a variety of new methods of discrimination measurement,and finally tests the effectiveness of these methods by simulation experiments.The results are shown as follows:The traditional methods of discriminability measurement cannot reflect the distinguishability between any two evaluation values,which is not conducive to accurately measuring the discriminability of comprehensive evaluation results.Although the distinguishability between all evaluation values is considered in the pairwise square variance method and pairwise deviation method,the calculation is too large,and the problem of mutual linear compensation among the discriminability of the evaluation values will appear.Although the adjacent square variance method and the adjacent deviation method effectively simplify the calculation amount of the pairwise square variance method and the pairwise deviation method,the problem of linear compensation between the discriminability of the evaluation values will also occur.After continuous improvement,the adjacent logarithmic deviation method not only solves the problem of mutual linear compensation of discriminability,but also simplifies calculation and facilitates comparison,which is a good discriminability measurement method.

关 键 词：综合评价结果 区分度 模拟实验 相邻对数离差法 

分 类 号：C812[社会学—统计学]