科研基金项目立项评估方法的研究与改进

Research and Improvement of the Evaluation Methodfor Scientific Research Fund Project Selection

以我国某一基础研究资助部门的基金项目立项评估为讨论对象,根据自身工作的实际体验,分析了目前立项评估中存在的问题,包括:1专家判断映射失真;2不完全信息难以描述;3量化信息过于粗糙.随后在对处理不确定性的有效工具——证据理论的基本概念进行论述的基础上,引入证据理论对现有评估方法加以改进,提出了一种科研基金项目立项评估的改进方法,阐述了改进方法的原理,列举了其实施步骤.在最后的讨论中,从有效性、精确性、简单性等方面论证了改进方法的合理性,并给出了关于非共识项目的定量描述和分析说明.

Taking scientific research fund project selection as the discussion object, analyse the existed problems in project selection during the current time according to the authors' work experience. Such problems include mapping distortion of expert judgment, description difficulty of incomplete information and too rough measure of quantitative information. Based on the fundamental concepts of evidence theory, an effective tool to deal with uncertainty, some improvement of the current evaluation method is made by introducing evidence theory and a modified method of scientific research fund project selection is put forward. The principle of the proposed method is set forth and its implementation steps are listed. Finally, the rationality of the modified method is verified from three aspects and some description and analysis on the projects with no common recognition are made.