14

Currently, the process of selecting reviewers for journal papers is based on take-it-or-leave-it offers: the journal editor invites me to review a particular paper, and I can either accept or decline. Often, I get papers that are quite far from my current interests, so I either decline, or accept just in order to help the editor. On the same time, there are some newly-submitted papers (that I see in arXiv) that I will be much more interested to read and review, but I have no way to tell this to the editor, as I do not even know which journal these papers are submitted to.

In contrast, in conferences (particularly in computer science), the process is much more efficient, as the PC members can bid on papers, and usually receive papers they are interested in.

My question: is it possible to design a similar mechanism for reviewing journal papers? E.g., a central authority to which all papers will be submitted, so that researchers will be able to indicate papers they would like to review, before the editor chooses the reviewers? I know this sounds quite vague, but maybe someone has already thought of a similar idea?

1
  • 7
    In some journals I know of, the panel of referees are urged to indicate the topics they are interested in reviewing.
    – GEdgar
    Jan 28 at 9:48

4 Answers 4

11

One very simple approach would be to contact the authors of preprints in your field and tell them you would be interested in reviewing. Then when they submit their manuscript to a journal, they can indicate you as a potential reviewer - either in the appropriate field in the submission form, or informally in their cover letter.

Of course, then the editor will have to decide whether to go with this suggestion or not: Do editors really follow suggested reviewers? But then again, this should really be a feature of any process that does what you are asking for.

And you could always read the preprint and offer suggestions to the authors quite apart from any submission to a journal. With the drawback that you have less of a lever to encourage the authors to actually make the changes you suggest.

3
  • 6
    If I suggest myself as a reviewer to the authors, doesn't it negate the principle of anonymous reviewing? Usually the authors should not know who reviews the paper - only the editor should know. Jan 28 at 10:52
  • 1
    Certainly. But in many fields, authors can suggest reviewers. Whether reviewer 1 then is one of those the authors suggested is, in a blind review, in principle not known to authors. Jan 28 at 11:32
  • 4
    If I find a preprint in my field that I find interesting and where I immediately think that I have something interesting to say about it (content-wise, not style/form-related), I'll often just write a quick email to the first author with my comments. Usually, they respond quickly and are very happy about the feedback (and that their papers seems to get visibility, of course). Whether they end up actually incorporating the feedback is a different question.
    – Eike P.
    Jan 29 at 13:07
7

This is an alternate version of "review" that is typical in the Software Patterns community in (mostly) computer science. It isn't blind, and the "shepherd" (not reviewer) works with the authors over a period of weeks to improve their submission. Shepherds are usually experienced at it and don't normally earn an authorship position from their work to improve the paper, though there was discussion about that some time ago.

The program committee publishes a list of papers with full authorship visible and potential shepherds offer to work on one or more papers (with the authors). If a "bid" is accepted by the committee then shepherding begins and a paper may go through several versions between authors and shepherds, generally a minimum of three. Both shepherds and authors are fully visible to one another and a shepherd is normally given an acknowledgement.

Ultimately the shepherd makes a recommendation on whether the paper should be accepted or rejected, and rejects usually come from authors unwilling to work with the shepherd on the contents. A paper that gets no bids from potential shepherds will likely be rejected, though someone might be assigned to work with the authors in any case to improve the paper for the future. There is also the potential for a workshop at the conference for these "not ready for prime time" papers, again with the goal of improving them.

This works in this, relatively small, community, in which experienced shepherds are generally known to conference committees. Shepherds make their availability known to the conference committee through an email campaign asking for shepherds. I don't know how it would work in a wider realm. I don't believe that in this community it has led to abuses.

Note, also, that these "pattern papers" are about the practice of things like software development and are not theoretical. Their intent is to spread the knowledge of "good practice" along with the evidence that it is, indeed, good. One of the "required" elements of a good patterns paper is the "known uses" of the practices described, so such things aren't especially "new". That is quite different from advances in scientific or other academic theory. That likely limits the applicability of such a practice.


I'm an experienced shepherd as well as an author in this community, so have seen the process from both sides.

7

SciPost, a transparently run non-profit diamond open-access publisher, has a Reports needed page on their website. There, everyone can see current submissions in need of refereeing, and volunteer contributing their own reports. The journals also use invited reviewers, so volunteer reports are more of a bonus than a replacement for the typical system. I tend to think it has to work this way, because otherwise you'd have to rely on people actively looking for more reviewing work. Since the journals publish referee reports (anonymous or signed, as chosen by the referee) one can see that some papers indeed receive contributed reports.

4

It's not really possible. As you mention you need some kind of central authority, and the only way that is going to work is if all publishers in the world unite into a single mega-publisher. Which would of course not be sustainable, because someone will inevitably complain about how the mega-publisher is a monopoly, has too much control, etc.

For individual publishers, though, this is doable. The mechanism is similar to the one for conferences that you describe: you need to volunteer as a reviewer, look at all the available manuscripts, and then apply to review. Right now the only major publisher I'm aware of that runs this kind of program is MDPI. If you're not already aware, I should warn that MDPI is highly controversial to the point that I will not review for MDPI on a personal level, but your mileage might vary.

Caveat: from the other side of the table, my experience is that most volunteer reviewers are patently unqualified to review the papers they apply for. I don't know why they apply anyway. Maybe because most of them are based in countries that receive few reviewer invitations, and they're looking to contribute (see this, page 15).

You must log in to answer this question.

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