University–industry R&D linkage metrics: validity and applicability in world university rankings

详细信息    查看全文

• 作者：Robert J. W. Tijssen ; Alfredo Yegros-Yegros ; Jos J. Winnink
• 关键词：University rankings ; Research performance ; Indicators ; Measurement
• 刊名：Scientometrics
• 出版年：2016
• 出版时间：November 2016
• 年：2016
• 卷：109
• 期：2
• 页码：677-696
• 全文大小：466 KB
• 刊物主题：Information Storage and Retrieval; Library Science; Interdisciplinary Studies;
• 出版者：Springer Netherlands
• ISSN：1588-2861
• 卷排序：109

文摘

In September 2015 Thomson Reuters published its Ranking of Innovative Universities (RIU). Covering 100 large research-intensive universities worldwide, Stanford University came in first, MIT was second and Harvard in third position. But how meaningful is this outcome? In this paper we will take a critical view from a methodological perspective. We focus our attention on the various types of metrics available, whether or not data redundancies are addressed, and if metrics should be assembled into a single composite overall score or not. We address these issues in some detail by emphasizing one metric in particular: university–industry co-authored publications (UICs). We compare the RIU with three variants of our own University–Industry R&D Linkage Index, which we derived from the bibliometric analysis of 750 research universities worldwide. Our findings highlight conceptual and methodological problems with UIC-based data, as well as computational weaknesses such university ranking systems. Avoiding choices between size-dependent or independent metrics, and between single-metrics and multi-metrics systems, we recommend an alternative ‘scoreboard’ approach: (1) without weighing systems of metrics and composite scores; (2) computational procedures and information sources are made more transparent; (3) size-dependent metrics are kept separate from size-independent metrics; (4) UIC metrics are selected according to the type of proximity relationship between universities and industry.