Let's say that I wrote my research paper and then I published it as a preprint on the ArXiv, and I got citations for this preprint. Afterwards, my same paper has been peer reviewed by a journal then published, so, how will search engines count the citations? Will they do a sum or union or consider them as two different papers?


This obviously depends on the indexing service used. Typically, the records for different versions of the same paper are merged, and the service will provide the sum of the references.

For example, on google scholar, you can click on "all versions" to see all versions of a paper included in their record. For example here are the 222 versions of the LIGO GW150914 discovery paper, included in Google's record.

(I am not sure how the extremely rare case that a single work references multiple versions of the same paper is handled. But this should indeed be very rare.)

  • okay, thank you very much, so the Google Scholar method is the most used method right ?
    – Adelovi
    Nov 30 '20 at 17:00
  • I don't know of explicit examples of services that do it differently. That doesn't mean they don't exist.
    – mmeent
    Nov 30 '20 at 17:26
  • Note that Google Scholar sometimes does not merge different versions of the same paper automatically (e.g. when the title changed etc.)
    – user151413
    Dec 26 '20 at 21:53

Conventional "search engines", that is, services which index papers to count citations etc., such as Web of Science etc., will not count citations to preprints, nor citations in preprints. Similarly, they don't index all journals, but only certain ones (though, of course, the idea is that it covers the majority of journals, and in particular the relevant and established journals). Thus, all those citations are lost. (Though, of course, they might decide at some point to index additional journals, or the arxiv, and they might decide to do so retroactively.)

Google Scholar, as mentioned above, indexes both the arxiv and the journal version, and usually merges them (which means that citations which appear in a journal and a preprint of the same paper are usually only counted once). On the other hand, there is no official protocol as to what and how Google Scholar indexes things, and they might change this at any time, so in that sense this is less transparent (though it does not mean it gives worse citation data).

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.