摘要
[研究目的]针对同一组对象使用不同方法进行评价存在评价结果不一致的缺陷,提出了一种基于荟萃评价模型进行综合评价的方法。[研究方法]以高校发明人专利数据为例,分别采用H指数、专利迹、技术循环时间和TOPSIS等方法进行事前异质性检验后,通过荟萃方法获得了发明人专利质量的加权平均效应,最后运用兼容度与差异度对评价模型进行事后一致性检验。[研究结论]研究表明,采用荟萃分析可以消除单一方法产生的误用,为发明人专利质量综合评价提供了新的思路和方法。
[Research purpose]Meta-analysis is proposed to address the shortcomings that the conclusions by different methods may differ greatly and there are inconsistent evaluation effects. [Research method]Taking the patent data of university inventors as an example, the weighted average effect of inventor patent quality was obtained by meta-analysis after heterogeneity tests using H-index, patent traces, technology cycle time and TOPSIS, and finally the ex post consistency test of the evaluation model was conducted based on compatibility and variability. [Research conclusion]This study shows that meta-analysis can eliminate the random errors and systematic bias generated by a single method, and provides a new idea and method for the evaluation of inventor patent quality.
作者
杨毓丽
彭博
吴华蕾
Yang Yuli;Peng Bo;Wu Hualei(Dalian University of Technolgy Library,Dalian 116024)
出处
《情报杂志》
CSSCI
北大核心
2022年第11期192-198,109,共8页
Journal of Intelligence
