Thursday, August 11, 2011

How Many Times?

How many times can a paper be rejected before you give up submitting it to journals?

.. a reader wonders.

We need data. How many times have you (re)submitted a (rejected) manuscript before you gave up on publishing the paper, at least in a form that mostly resembled that in which it was submitted?

Do you give up after the first rejection, particularly if the rejection is quite emphatic, or do you keep going no matter what?

Most likely, the results will vary, even for a particular author, depending on how you feel about a particular paper or project. The number may also vary for individuals at different career stages.

And of course 'giving up' is a nebulous concept. A particular paper might be shelved, but parts of it may be resurrected in another paper. That might be giving up on the paper in its original form, but some key elements of the paper may yet live.

You decide how to define the various relevant terms: rejection, giving up, paper, you etc.

What's your typical number? your highest number? Zero, 1, 2, 3, more than 3?

I have no problem revising and resubmitting a rejected manuscript to another journal. I will typically revise and resubmit until a paper is accepted somewhere, although it is rare for this to take more than 2 submissions. That doesn't mean I wouldn't re-submit more times. I think my max resubmits has been 3.

The variability in interpretation of the question renders the following poll entirely useless, but let's not let that stop us from getting data:

How many times would you (re)submit a manuscript before giving up?
more than 3
I would never give up free polls


Margaret said...

Each rejection will involve a rework to incorporate all the assessors comments and extra data if requested. After that there is usually a major rearrangement of data. So far haven't had any go past one reject, but would keep trying. We need the publications for our careers, no doubt about it.

Carmelo Fruciano said...

I guess that correlated aspects would be how fast one descends the "impact factor ladder" and the starting and ending (lowest acceptable IF, if any) points...

Also, while certain rejections may be emphatic, in most cases they come with useful suggestions so that every time one tries it's not the same paper as the previous good are the suggestions given by referees and how open is the author to accept them might also play a role on how "patient" one is.

Wavefunction said...

That's a good question and I would say it depends on the paper. If it's novel or important work that I have been doing for a long time and really want to see published, then even multiple rejections won't dull my enthusiasm. If it's relatively trivial work then I may give up after two or three times at most. In the past I have had a paper accepted after three submissions so I do know I am willing to try at least thrice.

Anonymous said...

Until you run out of matching journals worth publishing at, right?

Liz said...

I voted "never give up" but in reality, I have not ever submitted more than 3 times before acceptance. My feeling is that if it takes more than 3 tries, you are not being realistic with journal choice.

By the time a manuscript of mine is being submitted, it is (in my eyes), a complete, relevant, and polished story. It also likely represents year(s) of a student's or postdoc's blood, sweat, and tears. I don't think I would give up on it at that point. Hopefully, if the story is so flawed that it will not be accepted anywhere, this would be discovered long before the experiments were finalized and the manuscript written.

Notorious Ph.D. said...

Believe it or not, an article of mine that just came out was only accepted on the *fourth* submission to the same journal. By the end, I was resubmitting out of pure meanness.

Bear in mind that, in my field, articles are single-authored, and usually 7,000-10,000 words, so that may have an effect on how you interpret my example. Also, this was for a flagship journal in my field -- I wouldn't have done this for Journal of Podunk Studies.

Yet in my poll answer, I selected "two," mainly because I've learned that if they don't take your article on the second submission, it's going to be more productive in most cases to just move the hell on to the next journal down the line.

Anonymous said...

My record is 5, although this was for a conference paper (which in CS is like journal papers in other fields). If you're on the edge of several different CS subfields, I've found sometimes you have to be willing to submit and resubmit to different conferences before you get a set of reviewers that doesn't automatically say "this is an X paper, not a Y paper". I've also just had a journal paper accepted after going 3 rounds of R&R (with 4 different sets of reviewers!). Maybe I'm just uber-persistent!

Anonymous said...

I'm in astrophysics. To first order, our papers, even in our 1st journal, are never rejected. I rejected one recently as a referee but I wouldn't be surprised if the authors just asked for a new ref and got one.

As you can imagine, the S/N in our journals has been dropping lately.

MamaRox said...

I'm currently on submission #3 of a paper, which is my maximum thus far.

Journal #1 editor said "try Journal #2". Then, Journal #2, editor said, "this paper might be better suited for a journal like Journal #1." Of course!

So, after some reworking... and soul searching... I'm going back to Journal #1, with a slightly different focus and a much stronger cover letter.

I voted for 3, but I would go further with this paper. I think it is among the more interesting and relevant work I've done, but it happens to be on the edge of several fields. Plus, I'm highly motivated on behalf of the student co-author, for whom this will be a first publication.

Also, it was done without much funding, so it just more "feels" fringe. I'm not sure if that's part of the reason? I'm not sure at what point I'd give up.

yolio said...

This has to vary a lot by field. My number one complaint about my research field is that it is rejection rates are so high that it is typical to need to resubmit a manuscript 2-3 times, at least. Unfortunately, this mostly is not a reflection of ms quality, but rather a nebulous concept known as "fit." Scientists are wasting a lot of time and energy shopping their papers around from journal to journal trying to find the one that will bite. Journal self-descriptions are largely unhelpful here, because what a journal really values varies depending on who the editors are at the moment this changes constantly. The bottom line is, typical rejection rates at mid-tier journals are super high, and too much of the review process ends up focused on fit rather than quality. This also does not seem to have had a positive impact on the overall quality of what gets published.

mathgirl said...

The maximum number of times I ever submitted a paper is 2.5.

It works like this: the first submission resulted in total rejection. The second submission (to another journal) resulted in half the paper being accepted (a very precise and concrete half). Finally, the second half was submitted to a third journal which accepted it.

Alex said...

Back when I was a postdoc, my boss got the idea that we should submit a paper to a really prestigious place. We knew it didn't have a chance, but we did as we were ordered. It got summarily rejected without even going out for review. Then he ordered us to send it somewhere even more prestigious. It got rejected. He ordered us to send it to another prestigious place. Again, summarily rejected within a day or two. He started setting his sights lower, but it still got rejected very quickly. The paper was a nice little piece of work, but it was very incremental.

Eventually it got into exactly the sort of place it was destined for. Personally, I think we could have gone one step higher, but we spent literally every other day reformatting the paper for another journal. In that situation, it's hard to maintain the mental focus to do a few more experiments and calculations. We literally wasted a few weeks that could have been spent improving the paper.

All of my other papers have been accepted at the first or second journal they were sent to. Sometimes it took a few rounds of revision at the second journal to satisfy a reviewer, but it got in.

Anonymous said...

I didn't answer the poll, because it depends completely on the paper. My first rejected paper was rejected on the grounds that the basic physical premise of the paper was incorrect. I was persuaded that the referee was correct, and so I totally abandoned the paper. In retrospect, I could probably have modified the premise and still been able to pull out a publishable revision. But it was a pretty short paper and the referee did convince me that it was fundamentally flawed. However, on another rejection, the referees were totally off-base and unfair. I resubmitted that paper 3 times (the last time to a different journal, where it sailed through) and it has earned a respectable citation count.

Mary said...

George Akerlof has a rather famous story about an article he had submit to four journals before one finally accepted it, back when he was a first year assistant professor. The first two journal rejections said it was "trivial." The third journal said it was "incorrect." The fourth journal said yes. That was in 1970.

After publication, the paper quickly became one of the most widely cited and most influential papers in economics and spawned an important line of economic inquiry, for which he ultimately shared the 2001 Nobel Memorial Prize in Economics.

His Nobel memoir recounts the story of his difficulties in getting his article published.

Anonymous said...

When a colleague saw this post/poll, he said "How could anyone submit a paper 3 or more times?". So I reminded him of two of our recent papers that took 3 submissions before being accepted for publication. Maybe some people just blank it out and don't remember how many tries it took. In this case, he didn't remember the 3 tries because the first 2 rejections were 'unfair' (acc to us), so he didn't think of them as rejections in the same way as other people's (just a hypothesis).

Anonymous said...

I'm a computer scientist like acdalal, so most of the papers that I submit are to conferences. My personal record is two, but I often think about this problem from the other side of the equation, as a reviewer. In many of the conferences I review for, I see the same low-quality papers submitted repeatedly, usually without change even after receiving well-deserved scathing reviews; there is one paper that I've seen at least half a dozen times. So there certainly are people who will never give up, but it makes me wonder if these people have a realistic assessment of their own work.

GMP said...

My maximum is 3; I have a collaborator who thinks everything we produce is pure gold and always wants to send papers to Nature or Nature Progeny first, where most of it gets rejected without review. Next comes a very prestigious society level journal, where the criteria are also really high. There are several papers of ours where you could readily see that it's decent work for a reputable but not fancy-shmancy society-level journal, where the paper should have been sent to begin with and where it did end up published smoothly with minor revisions, but it was a 3rd journal for us.

When we believe the work is really groundbreaking, I think it's worth fighting tooth and nail though to get it into a very high-IF journal. But one should be realistic about the quality and hotness of one's work; for the sake of students and postdocs who need publications, when the paper is solid but but not super hot, just send it where it belongs and don't waste time trying to overshoot.

Anonymous said...

I've gone 5 times. Unlike Alex, it was my (PhD student then) wish to try as high as I could and my boss went along with me trying. It was a valuable learning experience. The paper definitely had weaknesses that were pointed out correctly, it improved and changed focus somewhat with every new version and ended up where it belonged. Most difficult was getting to the point where I agreed that the figure of Nice Data that were Hard To Get was just in the way of the flow of the paper and to get rid of it.

Anonymous said...

I've submitted a paper to 6 different journals before it got into journal no 6. Admittedly we were aiming high in terms of journals, but it was a novel & controversial paper.

I've only once submitted a paper but never published it. The reviewer found a possible technical flaw, we did some extra expts and the flaw could explain the data. So that paper sank.

Anonymous said...

I've gone to 4 times in two cases. In one case the paper was pretty damn cool and I thought it was worth giving Nature/Science/PNAS (in that order) a shot. Ultimately it went to something not too far below PNAS so honestly it was a good paper and worth a try - they usually say no but you never know unless you try. The other was a cool data set that got rejected 2x without review b/c the editors thought the data was so logical it must be trivial. Hadn't actually been empirically shown before but yes it was the intuitive finding. It published lower than I would like but is quickly acquiring cites - so I guess it was more useful than they thought.

I would give up more quickly if the reviews were really critical but in both cases the papers were rejected because they weren't seen as being cutting edge enough (& I didn't agree with that assessment based on the literature).

Anonymous said...

I'm one of the few people who said no retries. I tend to take rejection very hard, and I often can't look at a paper again after it has been rejected. This has undoubtedly depressed my publication rate (which is already pretty low because of the difficulty I have getting things written up).

I have once contested a rejection, when one of the reviewers was clearly too stupid to hold an honest job. My co-author and I pointed out all the errors in the review to the editor, who threw out the review and called in another reviewer, who accepted the paper.

Mostly, though, if I get a paper rejected, I just get depressed for a month or two. (Luckily for my sanity, I've rarely had a paper rejected.)

MZ said...

A related query that I'd be curious to get FSP's take on: how often do people contest a flat out rejection? I've been an editor for 4 different journals and it has been getting much more common to have authors not be able to take "no" for an answer. My admittedly un-random sample suggests that it is overwhelmingly male authors who do this. I must confess that I have only protested a rejection twice in my (20 years plus) scientific career. Once it worked, once it didn't.

Anonymous said...

Gas-station - my PhD supervisor always said if you aren't getting 50% of your papers rejected, you aren't aiming high enough. So with my 6 submissions (up-thread), I reckon I'm making him proud!

Anonymous said...

"Gas-station—my PhD supervisor always said if you aren't getting 50% of your papers rejected, you aren't aiming high enough"

I've never agreed with that philosophy, just like I've never bought into the idea that if you never miss a plane you are getting to the airport too soon.

I prefer having an honest assessment of my work and placing it in the appropriate venue. Perhaps I would have a higher salary if I spun my work as being grander than it is, but I'd feel like a cheat.

Alex said...

Perhaps I would have a higher salary if I spun my work as being grander than it is, but I'd feel like a cheat.

You get salary increases based on the prestige of where you publish? Just one question: Is your school hiring? Because I teach at a place where salary increases are currently non-existent, and if they ever do exist they are at fixed rates for either across-the-board cost of living increases, or fixed rates for promotion.

Merit raises sound really good to me right now. What's saddest is that my colleagues actually seem to think that merit raises a bad idea.

Anonymous said...

"You get salary increases based on the prestige of where you publish? Just one question: Is your school hiring?"

No, we're a public university and we've not been hiring for a couple of years—next year is likely to be worse. They promised us no more "furloughs" (that should be "pay cuts" since there was no reduction in required work), but the state just cut the budget again and it is expected to be cut every 6 months for the forseeable future until we are a wholly private university.