Fourth Floor Studio

September 13, 2007

Giving up?

Filed under: science — chris @ 6:14 pm

How many times do incarnations of a paper need to be rejected (outright!) before you lose hope in it? This one has been doing the rounds for >3 years in 5+ versions.

Advertisements

10 Comments »

  1. It is a depressing experience. I think basically, editors instruct their referees to reject if at all possible these days.

    They’ve also developed an annoying style where the first paragraph reads like outright rejection (“we cannot accept your paper for publication”) and then they explain in the second paragraph “should you wish to resubmit at a later date”. I’m told that this is because people get annoyed when they are rejected after revisions. The editor can say “well, we sort of rejected it the first time” after the second, final rejection.

    Anyway, it’s all very depressing and I feel your pain. I’ve just spent a month doing the most tedious extra work suggested by referees after a “non-rejection rejection” of the type described above. I genuinely believe that I’ve done good work and thought more about the issues involved than previous authors, but I’m bracing for more rejection.

    It’s also hard to “get closure” and move onto the next phase of your work when people won’t accept your basic premise for some unknown reason. Let’s face it, peer review is rubbish. It’s just that noone has a better idea.

    Comment by nsaunders — September 14, 2007 @ 12:45 am

  2. […] Giving up? – on endless paper rejection […]

    Pingback by Malaise « What You’re Doing Is Rather Desperate — September 14, 2007 @ 1:19 am

  3. Depending on some collaborators, it takes one rejection to bury a paper. Happened to me a couple of times.

    Comment by nuin — September 14, 2007 @ 11:37 am

  4. I would not say that peer review is rubbish. It has helped in the past to really improve my work. Part of it is not useful but it does iron out some of the ladder thing of trying several different journals. I would much rather have an open auction. Here is the paper who wants to take it ?

    Comment by Pedro Beltrao — September 14, 2007 @ 11:50 am

  5. I should have said “some peer review is rubbish”. At its best, it is helpful and constructive. The problem is that an awful lot of peer review is unhelpful, lazy, sloppily-written, does nothing but expose the ignorance of the reviewer and is aimed solely at sinking your paper.

    Comment by nsaunders — September 14, 2007 @ 8:49 pm

  6. I have all this to look forward too! I’m currently working out kinks in my first paper, trying to cover any and all possible objections a reviewer might have.
    .
    On the other hand, I’ve seen some papers marked up for rejection, and rightly so. All the author had done was extend an algorithm, adding a layer of complexity and then failed to carry out any comparisons to the previous version, or any other algorithms designed to do the same job!

    Comment by riadsala — September 17, 2007 @ 3:27 pm

  7. As referees seem compelled to ask for at least some changes, there is a school of thought that says you should either build in some fixable flaws into your manuscripts, or leave out the last confirmatory experiments they are likely to ask for (which you can execute while the paper is in review).
    You can then write a response agreeing “in principle” with each comment, fixing the ones you intended to and ignoring the rest. That way you appear to give in completely whilst writing the paper as you had originally intended.

    Comment by chris — September 26, 2007 @ 1:17 am

  8. Can I give an editor’s point of view? There are two types of rejection, closed reject and open reject.

    1. Closed reject.
    This means that we’re not interested. Nothing the authors can do will persuade us to publish this, short of a formal appeal. It is either too flawed, or too small an advance.

    2. Open reject.
    The paper has too many problems at the moment, and we’d be stringing the authors along if we did invite revisions. We ask authors to make their revisions within three weeks, and at most three months – if the revisions are likely to take longer than that, we’d prefer to close the file for now.

    Sometimes an open reject is best as it is a sign to the author that we really do want them to fix the problems before we want to see it again – asking for revisions can result in an author trying their luck and making only cosmetic changes, which just wastes their time and that of the editors and reviewers when it later gets rejected outright.

    Rejection frees up the reviewers from the expectation that they will re-review – we may well invite them to review if the work is resubmitted, but they can much more easily decline if busy or not inclined to look again.

    If an article is repeatedly rejected from several journals, then there’s something wrong. Either you’re coming up against the same biased reviewers by sheer bad luck (or in a small field) or else there is something about the paper that doesn’t sit right with the field. To avoid repeated rejections, then you need to take the reviewers’ criticisms seriously, even if you’re not going to submit again to that same journal. Reviewers and editors rarely reject arbitrarily, and they always give reasons. Even if you don’t agree with the comments, acknowledging and countering criticism within the text of the manuscript is probably a good idea.

    We don’t want to have to reject papers! We want to publish as much sound science as possible. To be devil’s advocate, think who is to blame for a rejection – is it the peer reviewers and editors, or is it the authors who failed to meet the standards for publication?

    (I work on the BMC-series with BioMed Central, but – STANDARD DISCLAIMER – these are my own thoughts, not those of the journals)

    Comment by Matt Hodgkinson — November 2, 2007 @ 9:32 pm

  9. “As referees seem compelled to ask for at least some changes, there is a school of thought that says you should either build in some fixable flaws into your manuscripts, or leave out the last confirmatory experiments they are likely to ask for (which you can execute while the paper is in review).
    You can then write a response agreeing “in principle” with each comment, fixing the ones you intended to and ignoring the rest. That way you appear to give in completely whilst writing the paper as you had originally intended”.

    Sigh. Reporting science research isn’t and shouldn’t be a game. Submitting a paper with flaws or with supporting work left out just increases the chances that your work will be rejected. If you add in extra work after the initial review, that greatly increases the likelihood that the work will be sent for re-review, taking up more of the reviewers’ time and delaying publication. Should these flaws not be noticed by the reviewers, would you remember to fix them? If you don’t get picked up on the need for the confirmatory work, does that then allow you salami slice, leaving them to a subsequent paper?

    STANDARD DISCLAIMER – these my thoughts, not those of BioMed Central.

    Comment by Matt Hodgkinson — November 2, 2007 @ 9:47 pm

  10. Matt – thanks for chiming in.

    First off, I agree that if a paper is rejected (even just once) that suggests more work of some form needs to be done. I have also found reviews to be mostly helpful and constructive – although the subset of nasty/peremtory/off-topic ones always sit badly.

    I also agree that publishing shouldn’t be a game. However, playing becomes more advantageous as the stakes rise (an ESS, if you like), until you can’t afford *not* to play (if say, your promotion is on the line).

    I’ve also found that sometimes one is held hostage to reviewer whimsy (such as stylistic disagreements), particularly at the more competitive journals where every submission is at best a crap-shoot. There are so many other papers vying to get in that even minor frowns from reviewers jeopardise what little chance you have of getting accepted. In this context, it’s almost impossible not to play games to improve your odds.

    On the other hand, as an ever more frequent reviewer, I find myself classifying submissions into two piles*: “no way”, and “should be perfect”. I generally write shorter reviews for the former than the latter, for which I argue points to maximise clarity and content. My point being that I have never, ever accepted a paper as is. Out of 30+ reviews. Adding the 15-20 submissions I’ve made as an author, and the only time I’ve encountered a perfect acceptance was for an invited review for a special issue (which arguably doesn’t count).

    Which brings me to the nature of the game: given that a reviewer is likely to ask for at least some modifications, and that different reviewers will probably want different things, it is most efficient to submit the least polished paper acceptable and get the chosen reviewers to tell you how they want it customised this time around. It’s risky, because you could just get an outright rejection, but if you think the material is journal-appropriate it’s a tempting strategy. It’s also time-efficient if your field is brisk, and potentially gets you that coveted early submission date.

    * There’s also the third pile: the interim submission, a manuscript obviously in draft form, lacking some data, where the authors are trying to gauge referee response and/or garner ideas more than anything else. These are the shortest reviews, generally along the lines of “finish the damned thing before submitting!”.

    Comment by chris — November 11, 2007 @ 6:04 pm


RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Blog at WordPress.com.

%d bloggers like this: