48

I just received a major revision decision (1st round) for a manuscript I submitted to a reputable journal. After looking into one of the reviewer's comments I found a major error in my analysis. The conclusions of my research are almost completely different as a result. I also don't think the reviewer realized the severity of the error.

I would like to withdraw my manuscript from the journal because of the major error. However, I don't know how to approach my co-authors or the journal about this issue. I've always been able to complete major revisions for other papers I've written, and so I've never had to deal with this type of problem before. I also realize that the editors and referees have already put a lot of time into reviewing my paper.

I would like to ask this forum for any suggestions about withdrawing a paper once it's in review. I don't know how common it is to do this, and how it is generally perceived in academia.

Improve this question
7
  • 28
    Is this a "everything's ruined, nothing can be salvaged, toss the whole thing in the bin" kind of major error, or a "everything changes, but we still have a paper" kind of major error?
    – zibadawa timmy
    Mar 20, 2018 at 2:24
  • 10
    I would say an "everything changes, but we still have a paper" but only after doing an extensive amount of additional work (i.e., include another data set and more simulations, rewrite intro and discussion). It really may not be worthwhile.
    – anonymous47
    Mar 20, 2018 at 2:33
  • 10
    But if you say "it really may not be worthwhile", wouldn't that then equal the "everything's ruined, nothing can be salvaged, toss the whole thing in the bin" category (if perhaps we insert "with reasonable effort" after "salvaged")?
    – Thomas
    Mar 20, 2018 at 7:22
  • 15
    @Thomas: It seems to me that the OP is saying that (they believe that) they still have a potentially publishable result, it's just a completely different result from what they originally thought they had and so would take a lot of effort to (re-)write up. That's very different from realizing, say, that your experimental results were just random statistical noise or that the axioms from which you'd proved a bunch of remarkable results were actually self-contradictory.
    – Ilmari Karonen
    Mar 20, 2018 at 10:48
  • 1
    @mathreadler "major revision" usually means that substantial parts of the paper have to be changed/improved/extended based on the remarks of the reviewers. If I am a reviewer and I recommend a major revision, I don't expect to get a paper with completely different conclusions. OP should notify that editor about the problem and resubmit the paper as a new paper. No harm done.
    – trunklop
    Mar 21, 2018 at 20:01

4 Answers 4

Reset to default
54

There are two parts to your question:

  1. How to broach the matter with your colleagues.
  2. How to withdraw your paper from review.

The trickier issue is how to broach the subject with your colleagues. If you feel there is absolutely no way for you to finish the revisions on schedule, then you owe it to your co-authors to tell them of the problem. However, this should be a dialogue, not a unilateral decision. They may see the issue as being less onerous timewise, and that it may be possible to finish the work within a reasonable period of time. Or they may agree with you that withdrawal is the best option. In either case, it should be a team decision, and once that's made, you can proceed with notifying the journal.

Notifying the journal is the simpler issue to handle. Most journals usually set a timeline for submitting major revisions. Just send the journal a note stating exactly what you wrote above. There is nothing wrong with independently withdrawing a paper because of significant errors. (Better that you do so now than need to retract it because of errors later on!)

Improve this answer
7
  • 52
    If you want to withdraw the paper, don't just let the deadline pass. Do it explicitly so the editor and journal staff don't waste their time sending "Hey, your revision is overdue -- do you need more time?" mails.
    – David Richerby
    Mar 20, 2018 at 10:47
  • 7
    I had a similar thing happen -- the journal was happy to accept my later, substansally improved paper, sent it to the same reviewers (I believe), and it was accepted. Remember journals are usually run by fellow academics, they will understand problems arising,
    – Chris Jefferson
    Mar 20, 2018 at 20:39
  • 6
    If you feel there is absolutely no way for you to finish the revisions on schedule, then you owe it to your co-authors to tell them of the problem. - If there is a major error to fix, you owe it to your co-authors to discuss this with them regardless of anything else. That should be the first thing you do.
    – Kimball
    Mar 20, 2018 at 23:24
  • 2
    Perhaps the "independently withdrawing" phrase is being read differently from what you meant, by some people. I think you mean that the authors (plural) withdraw their paper instead of having it rejected. But I think @StephenG interpreted it as "independent of your co-authors". Add an "after that" at the start of the last paragraph, and/or some other phrase that makes it clear you only do this if / when the group of authors decides to give up (at least temporarily) on this publishing deadline.
    – Peter Cordes
    Mar 21, 2018 at 14:13
  • 1
    @PeterCordes If that's the case, perhaps changing "independently" to "proactively" is another way to clarify the meaning.
    – David Z
    Mar 22, 2018 at 4:22
21

The whole point of the peer-review process is to avoid that any sub-standard garbage gets published. That is the reason your "reputable journal" is reputable in the first place.

Here, the review process pointed at a major issue in your work. That's great because that's why your paper gets reviewed in the first place. (I would even argue that whether or not the reviewer actually realized the gravity of the implication of their review is secondary.)

Now, you're currently in the following situation:

  • Presumably, you have already invested a lot of work to get to the current point in the research for this paper.
  • To conclude this part of your research, you've now realized you need to spend a whole lot of additional work.

Naturally, the question then arises: are you either (1) willing to spend the additional work or (2) completely scratch all the work you've already done?

The option that's missing here is: (3) pretend nothing happened and publish the current paper after addressing the outlined revisions. Clearly, you do not intend to go for that, so the decision is between (1) and (2). In both cases, you will have to tell your co-authors and the journal that the paper will need more additional work than anticipated.

But I don't quite see the problem with that. The journal did their job, so if you honestly inform them about what happened, I can't imagine that it will lead to any bad consequences for you. It is in their interest too, after all, that only high-quality results get published. So, why not thank the reviewers and inform the journal that you will (1) withdraw the paper completely, or (2) resubmit at some point in the future but probably not within the specified time frame for the revisions they asked for?

Also: I would probably discuss your newly gained insights with my co-authors as soon as possible. Not only so they are informed but also because maybe there's a different angle that you're missing and that would put the whole situation into a different light again? One of your co-authors might have some relevant insights that might change how you will move ahead.

Improve this answer
3
  • 1
    Thank you for the very thoughtful response. It seems that the sensible thing to do would be to discuss the error and any future plans for the manuscript with my coauthors. I'd be willing to do the extra work if my coauthors also think it's worthwhile.
    – anonymous47
    Mar 20, 2018 at 8:10
  • 2
    If you keep the journal editor informed then if you end up resubmitting to the same journal they may try and use the same reviewers (which may speed up the process).
    – SPPearce
    Mar 20, 2018 at 9:55
  • 3
    @anonymous47 I would suggest also to be gracious to the editor when you inform them of your decision. Their selection of reviewers has saved you some substantial potential embarrassment, and the "system" worked.
    – Bryan Krause ♦
    Mar 20, 2018 at 23:25
11

I'd say the problem is that you're trying to tackle this and make decisions alone.

Contact your co-authors and explain the situation just like you explained it to us. Don't tell them "we should do XYZ", just explain what's already happened. Let them make suggestions first, then make your own suggestion (or support somebody else's).

Your discussion with your co-authors will dictate how to inform the journal, but - the swifter the better. The only thing reviewers or editors may be angry about is delaying this kind of notification.

Improve this answer
2
  • " then make your own suggestion if at all" this reads like you're implying that OP shouldn't suggest a course of action, or that it would at least be a bad idea? Why?
    – Cubic
    Mar 21, 2018 at 13:59
  • @Cubic: Edited to clarify what I mean.
    – einpoklum
    Mar 21, 2018 at 14:02
2

It is possible to ask extra time on major revision and it certainly is alright to modify the paper in a major way. Talk with your co-authors on how to proceed. If you agree on to fix the paper, explain the editor that the modifications required to solve the problem pointed out by the reviewer would take longer than the allotted time. You might receive extra time to fix the issue, or you might need to resubmit later on. If the editor decides to give you the time you need, no work will be wasted. Just make sure you act swiftly.

Even if you do not get the chance, you should not feel sorry about the time invested in reviewing the paper. The system is worked and the mistake is found. The reviewers do reviews willingly for multiple reasons, one of which is to ensure their field is not littered with papers having wrong results.

Improve this answer

You must log in to answer this question.

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