基于DEA超效率验证的项目同行评议结果有效性研究

The Validity of Peer Review Results of DEA Based Super Efficiency Projects

[目的/意义]本文以项目评审为例探索专家知识为基础的同行评议判断的有效性。[方法/过程]本文采用数据包络分析测度项目的研发效率,评价模型为经超效率修正的CCR模型,并将该效率指标作为同行评议有效性的验证标准。数据源为国家林业局实施完毕的126项“948”项目,涵盖13项投入与产出评价指标。[结果/结论]实验发现:项目同行评议结果与实际研发效率(DEA超效率)之间呈现适度的正相关性,Spearman相关系数ρ=0.250 (Sig.<0.01)。主观评审犯Ⅰ类错误与Ⅱ类错误的比重分别为17.5%与18.2%,有效同行评议决策比重占64.3%。证实同行评议的确存在一定的系统误差但不可避免。

[Purpose/Significance]This article uses some projects’ peer review results as an example to explore the effectiveness of peer review based on expert knowledge. [Method/Process]It uses data envelopment analysis to measure the R&D efficiency of the projects.The evaluation model is a CCR model modified by super-efficiency, and the efficiency index is used as a verification standard for the effectiveness of peer review results. The data source is 126 "948 projects" implemented by the State Forestry Administration, covering 13 input and output evaluation indicators. [Results/Conclusions]The experiment found that there was a moderate positive correlation between the project peer review results and the actual R&D efficiency(DEA super efficiency), and the Spearman correlation coefficient ρ=0.250(Sig.<0.01). The percentages of type I errors and type II errors made by subjective review were 17.5% and 18.2%, respectively,and effective peer review decisions accounted for 64.3%. It is confirmed that there are certain systematic errors in peer review, but they are inevitable.