Today I received my first rejected paper.

The committee of 2 people left some good points but at the end they didn't get the point of my theme. It took me 3 months to write this paper for the Requirements Engineering Conference in Lisbon, Portugal.

I talked to 2 PhD students and they said that this takes time and passion. They told me that the average of rejected papers/journals per capita is 7. Do you agree with this average?

  • 1
    @Prof.SantaClaus, do you think 7 tries is expected if one has an advisor who provides some guidance on how to do the work, write the paper, and select the forum? I grant that a "newbie" going it completely alone will hit major hurdles. – Fred Douglis Jun 16 '17 at 21:02
  • 1
    @FredDouglis like you said there are many factors. Taking the 'advisor' factor into account, I would question his/her 'average rejection'. – Prof. Santa Claus Jun 16 '17 at 21:15
  • 5
    If your papers are always accepted the first place you sent them, then you are aiming too low. – GEdgar Jun 17 '17 at 1:24
  • 1
    This is probably somewhat field dependent. While I agree heartily with @GEdgar's comment [and he is also a mathematician], 7 seems like an unpleasantly large number to me. The amount of time one would spend waiting for a math paper to be refereed 7 times is anywhere from 2 to 10 years: that seems like too long. I can't think of a paper I've submitted more than four or five times, and having to submit that many times felt quite grueling. – Pete L. Clark Jun 17 '17 at 2:01
  • 1
    @PeteL.Clark I wouldn't say that implies there are 7 tiers. In my mind, the decision on a paper is like a random variable, and it can make sense to sense to try at few different journals (nearly independent events) where you estimate you acceptance probability, say, at 50%. (That said, I completely agree 7 seems really high, at least in math) – Kimball Jun 17 '17 at 6:12

This varies depending on a huge number of factors. The venues you select to submit your work, the quality of the work, whether the work is described well or penalized because it is written badly, misses important related work, or many other factors. And it no doubt depends on field as well.

I would however say that the idea that you submit 7 times for each time you get something published means you're setting your sights too high. In Computer Science, for instance, probably most conferences have an acceptance ratio of 4:1 or 5:1, perhaps 6:1, at worst. But then many others have much high acceptance ratios, so I would think the mean acceptance rate is 4:1 or 3:1. In my own experience most papers have gotten in either the first time or the second. One took 3, and another time I stopped trying to publish the work, but the average would be around 1.5-2 at most.

Whatever the ratio is in your field, by all means don't get too stressed over your first rejection. It's part of the process! And work with your advisor to find appropriate targets that are realistic.

  • 3
    @maxmin Without your advisor's help, it's virtually impossible to publish in good conferences. I advise you to seek proper guidance before re submission. Otherwise, it's a waste of time. – Prof. Santa Claus Jun 17 '17 at 4:31
  • Tks @lighthouse keeper ... – maxmin Jun 17 '17 at 16:04

I generally agree with Fred Douglis, and 7 resubmissions seems high.

However, it is important to note that the link between acceptance ratio of journals and the number of tries that are needed is not straightforward for many reasons. The review process is not random, the resubmissions are not independent trials, and you are supposed to improve your paper between submissions based on the reviews. So for example if someone keeps submitting to journals that are out of scope, is not skilled in writing, or refuses to consider reviewer feedback, they might need to submit as much as 7 times before getting a paper accepted.

  • 1
    While I generally agree to this answer, concerning the statement "The review process is not random" there is some evidence that the review process may actually include a non-negligible amount of randomness, especially for papers that are neither outstandingly good or bad. – lighthouse keeper Jun 17 '17 at 11:52
  • 1
    @lighthousekeeper I agree - I did not mean that the process is not stochastic. I meant that there are factors such as the quality of writing, appropriateness of the selected journal/conference and the significance of the results that can affect the probability of acceptance. – Bitwise Jun 18 '17 at 5:57

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.