12

When quoting the number of citations for each paper you have published, different sources can be used, Google Scholar, ResearcherID, Scopus. Google Scholar covers a larger range of literature.

Is it acceptable to use Google Scholar? Or it is not a professional resource, and we necessarily should use ISI Web of Knowledge for counting citations?

2
  • 1
    In my limited experience, looking at my own publications: I find Google Scholar listed many citations that should not be included in a "professional" list. Are you publishing in some field that does not have a professional citation list?
    – GEdgar
    May 5, 2015 at 18:03
  • 5
    Many problems with google scholar. These include (1) imperfect merging of versions means you can get multiple citations from multiple versions of the same paper, (2) imperfect parsing of author lists means that I get "credit" for all of the citations to papers I've edited for one journal, despite trying repeatedly to remove them from my profile, and (3) the system is easily gamed.
    – Corvus
    May 5, 2015 at 18:32

2 Answers 2

11

None of the citation databases are particularly good. Google Scholar tends to err on the side of inclusiveness (thereby over-representing impact), while the more curated databases (e.g., ISI WoK) tend to err on the side of exclusiveness (thereby under-representing impact). This is particularly acute in some fields: computer science, for example, is notoriously under-represented in the curated databases to the degree that it has is own independent citation database, DBLP, which of course has its own different problems.

You can potentially use any of these databases reasonably to report citations in a reasonable and professional manner, as long as you are consistent and make it clear which database you are using, such that readers can adjust their expectations accordingly.

1
  • 1
    DBLP is not a citation database, but a publication database. They recently added links to third-party citation databases to their web interface (opencitations.net and crossref.org), but these seem to be vastly incomplete within computer science. Feb 15, 2020 at 9:39
3

Google Scholar is actually quite useful, insofar as it draws a broad net for publications, whereas many other citation databases fail to capture important papers. Unfortunately, Google Scholar also has some problems with double-counting, since it often treats different versions of the same articles as if they were different publications, which can artificially inflate citation count. A secondary problem with Google Scholar (which also applies to other citation databases) is that it gives "raw" citation counts (and derived metrics like H-index, etc.) rather than "author adjusted" metrics. This also artificially inflates citation count, especially for authors who do most of their work with a substantial number of co-authors.

One problem with Google Scholar is that its broad sources for papers means that it can be subject to manipulation by publishing mock papers that give no scholarly contribution but give citations to other works or to each other (see e.g., López-Cózar et al 2012). The more tightly curated citation databases do not have this problem since they only count publications in approved sources; typically these sources are established scholarly journals with effective peer review processes.

As the other answer here points out, none of the citation databases are very good. They all suffer from under- or over-coverage and most do not make a serious attempt to adjust metrics for authorship. Personally, I think Google Scholar is less bad than many of the others, since it at least captures all the important papers a person has written, rather than artificially excluding those that fall outside a narrow range of journals. However, you should be aware of its drawbacks in relation to other citation databases and ensure that you carefully inspect the individual publications when using it as an evaluative source.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .