4

At the moment, I am aware the American Mathematical Monthly and Mathematics Magazine (from several years) and Proceedings of the American Mathematical Society (from one year, approximately) are using the double-blind review system. I find it very useful, especially for young researchers.

Where can I find the list of similar math journals which uses the double-blind review system?

Ps. Is there some reason why most math journals prefer single-blind review system?

Ps2. An interesting related thread here.

Paolo Leonetti
  • 267
  • 1
  • 6
  • Because basically all reviews used to be single-blind and change is hard... – Jon Custer Jan 04 '23 at 15:58
  • 1
    I know that most are single-blind, but I still do not get a reason to prefer this system to the double-blind one – Paolo Leonetti Jan 04 '23 at 16:07
  • 1
    A specific choice has to be made to change to double-blind, and then the journal processes have to be brought in line with that decision. Change is hard as I said. – Jon Custer Jan 04 '23 at 16:09
  • 7
    In many (most?) fields, double-blind is very laborious or even not feasible at all. It does not always work well in math, either, because the papers are on arXiv anyway. And change requires certain infrastructure in place. Consider a website engine someone would have written for journals in physics - you either could borrow it as-is and do single-blind, or invest into switching to double-blind. Required labor is not free, and sharing resources helps to reduce costs. – Lodinn Jan 04 '23 at 16:15
  • 1
    For researchers, it would be enough to not post it on arXiv on the very same day that the manuscript is finished (and it could be a mandatory rule for submissions). The benefit for young researchers would be clear, and this would avoid biases in several other cases.

    I understand that the change is not for free, but I don't get why almost no math journal adopts this system.

     – Paolo Leonetti Jan 04 '23 at 16:20
  • 7
    I would argue that at least in part this is because preprint publishing is generally encouraged in math, and forbidding preprints would be contrary to the late trend that emerged to fight the publisher mafia. And no, the suggested benefit for young researchers is also not immediately apparent - if you are alluding to possible reviewer biases with not taking these young researchers seriously, writing likely gives it away anyway. – Lodinn Jan 04 '23 at 16:31
  • 2
    Change is only useful if there is concrete evidence that change leads to better outcomes. I am not convinced that that would be so. – Wolfgang Bangerth Jan 04 '23 at 19:07
  • 9
    @Paolo Leonetti "The benefit for young researchers would be clear" The benefit to young researchers is not clear to me at all. On the contrary, young researchers with few publications under their belt benefit the most from being able to disseminate their new work more efficiently. The idea that there is some bias against young researchers seems entirely speculative and not at all congruent with my experience. (Is that what you are suggesting? You seem to take it for granted that it's clear why double blind is better for young researchers than single blind, but it really is not clear to me.) – Adam Přenosil Jan 04 '23 at 19:20
  • 1
    The side question was discussed in Why don't all disciplines follow a double-blind review system? – Anyon Jan 05 '23 at 08:00
  • 1
    @Lodinn What if preprint submission was anonymous during review, too? Would this solve the issues that you point out, at least in part? – Federico Poloni Jan 06 '23 at 08:18
  • 1
    @Lodinn The lower bias with double-blind peer review is a well studied effect in other fields, see e.g. https://arxiv.org/abs/2101.02701 . If anything, the speculative opinion is the presumption that this does not apply to mathematics. – Federico Poloni Jan 06 '23 at 08:26
  • 1
    @FedericoPoloni Yes, that would help somewhat. My biggest issue is that it is straight up not feasible in some fields: there is only one LHC in the world and groups working on it generally know who is assigned what. From the fields I worked in in any capacity, only bioinformatics seems like something with enough pieces of key hardware worldwide that M&M would not give the game away. The article you linked is still generally inconclusive even for a field without those problems, although I will accept it is surely better than idle speculation. – Lodinn Jan 06 '23 at 09:08
  • 2
    Theoretical fields and fields with accessible hardware at least can be feasibly double blinded, but I would have to speculate again to guess how prevalent the dependency on something fairly specific is in science (and that can happen in purely theoretical fields as well). All in all, if it came at no additional time cost for me as an author and a reviewer, I would happily roll with double blind, I just don't see enough value in it to spend time every submission scrubbing it clean. Maybe because it takes me an anomalously long time, I don't know. – Lodinn Jan 06 '23 at 09:19
  • @FedericoPoloni The problem is that there are too many preprints to read all of them. So you have to advertise your results, e.g. in the form of talks. This you cannot really do without breaking anonymity, as in small topics most potential reviewers attend the same conferences. So with double blind reviews I would have to delay talking about my results for the 1+ year or so it often takes to bring the result from preprint to publication in math. For a young researcher, having recognition of your work delayed by this long, at least to me, seems much more damaging than biased journal decisions. – mlk Jan 09 '23 at 14:34
  • 1
    @mlk I did not suggest refraining from giving talks about one's research. Even if authors can be de-anonymyzed in some cases, I do not think that this would make double-blind peer review useless. There are already other 'tells' such as writing style, topics and tools used. – Federico Poloni Jan 09 '23 at 14:40
  • @FedericoPoloni I am not saying it would be useless, I am just not sure that this would not reintroduce a lot of bias again, e.g. "I have not heard anyone talk about that result at the last conference were all my friends and anyone important was, therefore it cannot be a result that matters." In particular since people with larger names tend to give more talks to larger audiences and are thus more likely to be de-anonymised. – mlk Jan 09 '23 at 14:55
  • @FedericoPoloni Thank you! Indeed, I think it wouldn't be a bad idea if a nonnegligible subset of journals will adopt a double-blind system. The aim of the thread was to understand what is really the current situation, surely not a "shopping question" as it has been tagged. Anyway, it seems from the comments that there are no other ones. – Paolo Leonetti Jan 10 '23 at 07:51
  • @AdamPřenosil Honestly, I do not agree, or better, there is some dependence on the field. It happened (and it still happens) that new researchers "add" the advisor or another guy "with a good affiliation" in the list of authors just to have better chances to get accepted somewhere. This sounds completely unfair to me, and double-blind system would avoid these situations. – Paolo Leonetti Jan 10 '23 at 08:12

0 Answers0