›› 2016, Vol. 28 ›› Issue (11): 85-94.

• 技术与创新管理 • 上一篇    下一篇

基于滞后非径向超效率DEA的高校科研效率评价研究

倪渊   

  1. 北京信息科技大学经济管理学院, 北京 100192
  • 收稿日期:2014-06-16 出版日期:2016-11-28 发布日期:2016-11-23
  • 作者简介:倪渊,北京信息科技大学经济管理学院讲师,博士。
  • 基金资助:

    国家自然科学基金项目(71171021/G0117);北京社科基金青年项目(15JGC1774);北京市教委社科计划一般项目(SM201611232002)。

Evaluating the Efficiency of Scientific Research in Higher Educational Institutions: Based on the Lagged Non-radial Super-efficiency DEA Model

Ni Yuan   

  1. School of Economics and Management, Beijing Information Science & Technology University, Beijing 100192
  • Received:2014-06-16 Online:2016-11-28 Published:2016-11-23

摘要:

数据包络分析(DEA)是高校科研效率评价领域重要的建模方法。然而,已有基于DEA的高校科研效率评价模型未能充分反映高校科研投入产出滞后性、评价结果排序性以及评价价值偏好等三方面的特征。为了弥补已有研究不足,本文提出了一种组合评价模型:滞后非径向超效率DEA模型。该模型将阿尔蒙多项式、非径向偏好DEA和超效率DEA有机融合,通过阿尔蒙多项式对投入产出数据的预处理来模拟高校科研系统投入产出的滞后性,借助非径向超效率DEA满足高校科研效率评价对于高区分度和偏好性的需求。最后,应用该模型对36所985工程高校科研效率进行评价,并采用比较验证方法来实证模型的可行性和有效性。结果显示:滞后的非径向超效率DEA具有较好的区分度,可以实现评价对象的全排序,且比标准DEA和非径向超效率DEA的评价结果更好地体现了高校科研投入产出的滞后效应。本文研究结论是对现有科研管理和DEA理论的有力补充,有助于指导高校发展和相应管理政策的制定。

关键词: DEA, 高校科研效率, 非径向, 超效率, 投入产出滞后

Abstract:

Data envelopment analysis (DEA) is the most significant method in evaluating the efficiency of scientific research in higher educational institutions. However, the existing models based on DEA fail to simultaneously consider three characteristics of scientific re-search system, i.e., the input-output lagging of scientific research, the ranking of evaluation results and the evaluators' preferences. To make up for this lack of the existing researches, we propose a combined evaluation model-lagged non-radial super-efficiency DEA model. This model integrates the Almon polynomial, the non-radial DEA and the super efficiency DEA, using the Almon polynomial to simulate the process of input-output lagging of scientific research and depending on the non-radial super-efficiency DEA to meet the high-efficiency evaluation discrimination and preference requirements. Finally, an application example is conducted based on thirty-six "985 program" universities, using multi-channels to collect data and comparative methods to verify the feasibility and effectiveness of the model. The re-sults show that:lagged non-radial super-efficiency DEA owning a good discrimination could rank all evaluation objects and its evaluating result is more accurate than the standard and non-radial super-efficiency DEA and better indicative of the lag effect of the input and out-put of scientific research. This conclusion contributes significantly to the existing scientific management theory and DEA theory, and can be used to guide the development of universities and make appropriate management policies.

Key words: DEA, efficiency of scientific research, non-radial, super-efficiency, input-output lagging