Do We Know What We Publish? Comparing Self-Reported Publication Data to Scopus and Web of Science

Abstract

Research productivity is a basic quantitative measure for universities, but obtaining a comprehensive view of a university’s output is difficult. Databases like Web of Science and Scopus can provide a list of publications associated with an institution, but they are limited in the formats and disciplines they cover. An institution can rely on its researchers to provide information on their own publications, which may avoid format and discipline limitations, but the data is often too messy or incomplete for research assessments. This study compares research productivity data obtained through Web of Science and Scopus to self-provided data from researchers at the University of Colorado Colorado Springs (UCCS). The study seeks to determine if certain disciplines are advantaged or disadvantaged by their representation in each dataset and to determine how complete researcher’s self-provided data is. It was determined that basing UCCS’s research assessment on researcher provided data led to more comprehensive results.

Keywords:

• Bibliometrics
• research assessment
• research productivity
• scholarly communications

Acknowledgements

Thank you to the librarians at Colorado College who snuck me into their library during a pandemic to access Web of Science.

Disclosure statement

The author reports there are no competing interests to declare.

Notes

1 The majority of these were University of Colorado hospitals and health centers located in Colorado Springs, but not affiliated with UCCS.

2 Unlike Scopus, which identifies conference proceedings by their content (“[o]riginal article reporting data presented at a conference or symposium” (Elsevier, 2020)), WoS delineates conference proceedings by their container, specifying, “Records covered in the two Conference Proceedings indexes…are identified as Proceedings Paper. However, the same records covered in the three indexes…are identified as Article when published in a journal. Proceedings papers will have a dual document type: Article; Proceedings paper.” (Clarivate, 2021). For the purposes of this study, records marked both “Proceedings Paper” and “Article; Proceedings paper” were counted as Conference Proceedings for WoS. However, the small number of conference proceedings found in WoS suggest either very limited coverage of this document type or that proceedings are being labeled as another type of document in WoS.