I'm wondering if the growing catalogue of preprint repositories have known 'impact factors' similar to those of popular print journals. If so, can anyone highlight them?

2 Answers 2


Even if there are services computing them, the impact factor for a preprint repository would not be useful.

Impact factors represent how many citations an article gets on average within a certain period of time. As such, they are supposed to be a rough metric for the selectiveness of a journal. If a journal is highly selective, then only highly impactful articles should (in theory) make it in, driving the impact factor of the journal up.

Now a preprint repository is explicitly designed to not be selective - Every paper gets in (possibly under some conditions that are hardly comparable to peer-review). As such, the impact factor of such a repository would not measure anything of interest.

  • Thanks Jeromy. But how would you really tell if your article is making any impact? I mean, how would you grade the preprint outlets in terms of their effectiveness? Commented Nov 18, 2019 at 11:12

The whole point of the preprint repositories is to provide an outlet for new results that bypasses the peer review system without offering an alternative to it. Since preprints are not peer-reviewed, they should not be cited (otherwise the whole peer review system would be broken, and we might as well cite Wikipedia). As they are not cited, they cannot contribute to an impact factor.

The most a preprint can receive in terms of "impact" is exposure via social media, and most preprint archives offer such metrics (bioarxiv for instance). This can be nice to look at, but ultimately it really means very little about the traditional impact of the paper, the same way that a newspaper writing about some research does not increase its credibility or scientific importance. It can also give a false sense of achievement.

From my experience the media hype around preprints is not related to the repository but rather the authors sharing it via twitter and it being picked up by sufficiently connected users. Again, this doesn't say anything about the impact of the paper but rather how "viral" it is.

Why are you publishing a preprint? What would be your motivation to choose one repository over another? I can't see any reason not to go with whatever repository is respected in your field. If you are publishing a preprint to avoid scooping, or out of sheer altruism, it shouldn't matter so long as the information is publicly available. If you want publicity, it also doesn't matter (as explained above).

  • 1
    I completely disagree with "Since preprints are not peer-reviewed, they should not be cited". It's important for people citing articles to know whether they are peer-reviewed or not and to take that into account when citing, but that doesn't mean preprints shouldn't be cited, and it's normal to do so.
    – Bryan Krause
    Commented Nov 18, 2019 at 23:21
  • Ok, so let's start citing preprints, and citing papers that cite preprints, and soon the whole peer review system collapses. This isn't my opinion, this is just policy of most journals. You can mention a preprint, but it shouldn't be used as supporting evidence. A similar issue is met by citations of retracted papers, only that it is harder to solve in retrospect. The problem is that there is no tiered system in references that can separate ones that have been retracted or preprints from peer-reviewed papers, so it all gets added to the same stats.
    – Caharpuka
    Commented Nov 19, 2019 at 0:53
  • "This isn't my opinion, this is just policy of most journals" - can you support this statement somehow? I am not aware of any journal that bans citing of preprints. Peer review isn't quite as magical as you seem to think it is - lots of bad stuff makes it through peer review, and lots of good stuff hasn't been peer reviewed yet. As an author, if I am citing a preprint I am going to do my own "peer review" in deciding that the work is worth citing.
    – Bryan Krause
    Commented Nov 19, 2019 at 1:04
  • 1
    Since preprints are not peer-reviewed, they should not be cited --- This view is probably a bit field dependent, as it is fairly common in mathematics to include preprints in references (now, 20 years ago, 40 years ago, and earlier). See, for example, this google scholar search for "Borel set" (a rather common mathematical term in many areas of mathematics) and "preprint". (continued) Commented Nov 19, 2019 at 6:57
  • 1
    In earlier years (and often even now) one would usually see things like "unpublished manuscript", "in press", "submitted", "to appear", etc. rather than the actual word "preprint", so the search I gave using "preprint" shows only the tip of the iceberg, especially for papers more than 20 or 30 years ago (which are still often very relevant in mathematics, by the way). Commented Nov 19, 2019 at 7:01

You must log in to answer this question.

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