学术期刊评价中DEA方法的系统误差研究  

Research on Systematic Error of DEA Method in Evaluation of Academic Journals

作　　者：俞立平 杜维[1,2] 魏建良 Yu Liping;Du Wei;Wei Jianliang(School of Statistics and Mathematics,Zhejiang Gongshang University,Hangzhou 310018,China;Collaborative Innovation Center of Statistical Data Engineering Technology&Application,Zhejiang Gongshang University,Hangzhou 310018,China;School of E-business and Management,Zhejiang Gongshang University,Hangzhou 310018,China)

机构地区：[1]浙江工商大学统计与数学学院,浙江杭州310018 [2]浙江工商大学统计数据工程技术与应用协同创新中心,浙江杭州310018 [3]浙江工商大学管理工程与电子商务学院,浙江杭州310018

出　　处：《现代情报》2024年第9期154-162,共9页Journal of Modern Information

基　　金：浙江省自然科学基金重点项目“制造业从数量型创新向质量型创新转型机制研究”(项目编号:Z21G030004);浙江省软科学重点项目“‘科技新政50条’落实情况评估研究”(项目编号:2023C25027)。

摘　　要：[目的/意义]DEA分析在学术期刊评价中得到了广泛应用,但其系统误差问题却没有得到足够重视。因此,深入研究该问题的根源及其解决策略,可以更为精确地揭示学术期刊的真实水平。[方法/过程]本文将系统误差理论引入到DEA分析中,在理论分析的基础上,对学术期刊评价中DEA方法的系统误差的产生机制及其影响进行深度探究,并在此基础上提出有效预防和消减系统误差的方法。[结果/结论]研究发现,我国DEA学术期刊评价仍处于探索阶段,学术期刊评价中DEA方法的系统误差的产生原因包括评价目的确定的系统误差、投入产出变量选取的系统误差、评价方法适用性系统误差、模型设定误差与结果公布系统误差,评价指标选取与方法选择是学术期刊评价中DEA方法的系统误差产生的主要原因,并提出了预防和消除学术期刊DEA效率分析系统误差的解决方案:认真选择学术期刊效率评价投入产出指标,保证评价对象数量达到一定规模,结合评价目的选择评价方法,不宜滥用复杂的DEA模型,采用专家赋权法设定权重。科技评价效率分析同样存在系统误差问题,需因地制宜进行处理,对于具有公共物品属性的学术评价应适度采用DEA方法。[Purpose/Significance]DEA analysis has been widely used in the evaluation of academic journals,but its systematic error has not been paid attention to.Therefore,an in-depth study of the root causes of this issue and its solution strategies can more accurately reveal the true quality of academic journals.[Method/Process]In this paper,the systemat-ic error theory was introduced into DEA analysis.On the basis of theoretical analysis,the generation mechanism and influ-ence of systematic error of DEA method in academic journal evaluation were deeply explored,and effective methods to pre-vent and reduce systematic error were proposed.[Result/Conclusion]It is found that the evaluation of DEA academic jour-nals in China is still in the exploratory stage.The causes of systematic errors of DEA methods in the evaluation of academic journals include systematic errors in the determination of evaluation purpose,systematic errors in the selection of input-out-put variables,systematic errors in the applicability of evaluation methods,model setting errors and systematic errors in the publication of results.The selection of evaluation index and method is the main cause of systematic error of DEA method in academic journal evaluation.The paper also proposes a solution to prevent and eliminate systematic errors in DEA efficiency analysis of academic journals:carefully selecrts input and output indicators of efficiency evaluation of academic journals,ensures that the number of evaluation objects reaches a certain scale,and selects evaluation methods according to the evalu-ation purpose.It is not appropriate to abuse complex DEA models,and the expert empowerment method should be adopted to set weights.The efficiency analysis of science and technology evaluation also has the problem of systematic error,which needs to be dealt with according to local conditions.For academic evaluation with the attribute of public goods,DEA meth-od should be appropriately adopted.

关 键 词：学术期刊 DEA 效率分析 系统误差 学术评价 

分 类 号：G302[文化科学]