A Criteria-based Assessment of the Coverage of Scopus and Web of Science

作     者：Dag W.Aksnes Gunnar Sivertsen 

作者机构：Nordic Institute for Studies in Innovation Research and Education 

出 版 物：《Journal of Data and Information Science》 (数据与情报科学学报（英文版）)

年 卷 期：2019年第4卷第1期

页      面：1-21页

核心收录：

学科分类：05[文学] 

基　　金：Norges Forskningsråd 

主　　题：Scopus Web of Science Norwegian Science Index Research evaluation Evaluation criteria Scientific publications 

摘      要：Purpose: The purpose of this study is to assess the coverage of the scientific literature in Scopus and Web of Science from the perspective of research ***/methodology/approach: The academic communities of Norway have agreed on certain criteria for what should be included as original research publications in research evaluation and funding contexts. These criteria have been applied since 2004 in a comprehensive bibliographic database called the Norwegian Science Index(NSI). The relative coverages of Scopus and Web of Science are compared with regard to publication type, field of research and ***: Our results show that Scopus covers 72 percent of the total Norwegian scientific and scholarly publication output in 2015 and 2016, while the corresponding figure for Web of Science Core Collection is 69 percent. The coverages are most comprehensive in medicine and health(89 and 87 percent) and in the natural sciences and technology(85 and 84 percent). The social sciences(48 percent in Scopus and 40 percent in Web of Science Core Collection) and particularly the humanities(27 and 23 percent) are much less covered in the two international data sources. Research limitation: Comparing with data from only one country is a limitation of the study, but the criteria used to define a country s scientific output as well as the identification of patterns of field-dependent partial representations in Scopus and Web of Science should be recognizable and useful also for other countries. Originality/value: The novelty of this study is the criteria-based approach to studying coverage problems in the two data sources.