Normalizing Google Scholar data for use in research evaluation

John Mingers* (Corresponding Author), Martin Meyer

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

32 Citations (Scopus)
3 Downloads (Pure)

Abstract

Using bibliometric data for the evaluation of the research of institutions and individuals is becoming increasingly common. Bibliometric evaluations across disciplines require that the data be normalized to the field because the fields are very different in their citation processes. Generally, the major bibliographic databases such as Web of Science (WoS) and Scopus are used for this but they have the disadvantage of limited coverage in the social science and humanities. Coverage in Google Scholar (GS) is much better but GS has less reliable data and fewer bibliometric tools. This paper tests a method for GS normalization developed by Bornmann et al. (J Assoc Inf Sci Technol 67:2778?2789, 2016) on an alternative set of data involving journal papers, book chapters and conference papers. The results show that GS normalization is possible although at the moment it requires extensive manual involvement in generating and validating the data. A comparison of the normalized results for journal papers with WoS data shows a high degree of convergent validity.
Original languageEnglish
Pages (from-to)1111-1121
Number of pages11
JournalScientometrics
Volume112
Issue number2
Early online date22 May 2017
DOIs
Publication statusPublished - Aug 2017

Keywords

  • Google Scholar
  • Normalization
  • Research evaluation

Fingerprint

Dive into the research topics of 'Normalizing Google Scholar data for use in research evaluation'. Together they form a unique fingerprint.

Cite this