11

Inspired by this question (including a comment by @JeffE that a paper should be short to be reviewed within a few hours) and some other questions on this website.

Almost everyone believes that peer-review of a paper takes about 1 day and peer-review of 20-30 papers per month means low-quality reviews.

My supervisor receives 50-100 peer-review requests every month (I know because I organize his mailbox) and accepts 20-30 papers submitted to high-impact journals (say impact factor 10) to review. He never asks us to review (contrary to many comments) and he does the review personally. I do not have exact statistics, but based on his schedule, he shouldn't spend more than 1 hour to review a paper.

My question is: if his peer-review is careless and of low-quality (as assumed as a fact by many here), then, why do editors frequently invite him to review? Note that he is one of many.

If the editors do not consider the quality of reviews and simply invite famous people frequently. Then, there should be a flaw in the quality of high-impact journals.

  • 6
    The time will vary according to discipline. For example, math papers take quite a bit of time to review. – Dave L Renfro Nov 23 '17 at 15:13
  • 6
    Some of my professors could read my submission in 5 minutes , ask me three questions and I had 3 weeks work - some really know their stuff... – Solar Mike Nov 23 '17 at 15:39
  • 2
    Some professors can fit three days in what you call a day. – Mark Nov 23 '17 at 17:04
  • 2
    Almost everyone believes that peer-review of a paper takes about 1 day — For a 5-page paper, maybe. See my answer on TCS.SE, especially point 6. – JeffE Nov 23 '17 at 20:31
  • 2
    @JeffE As usual, things are more field dependent than length dependent. Where there are description of experiments, tables of data etc. a fast reader can easily read a 10-page paper in a couple of hours and make annotations. Different is the case where one has to check calculations. But not all world is math or CS. – Massimo Ortolano Nov 23 '17 at 22:09
10

I think you're making a mistaken assumption that your supervisor's not able to generate useful reviews quickly. How long it takes to do a useful review depends on a few things: (i) the paper being reviewed, (ii) the kind of review being done, and (iii) the person doing the review.

Yes, if you're reviewing a long technical paper with lots of details it takes a long time to read through it. It takes even longer if you want to check all the details for yourself. It takes even longer if you're not familiar with these calculations.

I would assume that because your supervisor continues to get asked to do many reviews, the editors asking him find his reviews at least somewhat helpful. It's true that editors, at least in my field, have trouble finding enough good reviewers, but I can't imagine they would continue to ask someone who's review are consistently worthless.

I don't know what reviewing is like in your field (doing 20-30 reviews/month in my field would be unheard of) but here are some possibilities for how your supervisor could be doing many reviews quickly:

  • he's only selecting reviews he can do quickly and/or only reviews papers he wants to read anyway

  • since he does so many, he's probably gotten very efficient about doing reviews

  • he knows the field very well, so he can read the papers very quickly (and possibly assess it solely on the introduction, particular if he's going to recommend rejection)

  • he doesn't check low-level details, and primarily gives high-level reviews (which are still useful to editors, and depending on your area may be all that's expected)

  • he works really long hours

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.