I have research work that has been conducted and is at the stage where it is worth disseminating to the research community. It is somewhat cross-discipline in nature, and involves some technical contributions and is also of interest to a set of end users who are completely non-technical.

The problem is that, having attempted previous submissions, it has become painfully clear that neither community seems to appreciate quite the same issues as the other. Thus, a single paper aimed at addressing all the issues from both perspectives ends up being perhaps lacklustre, and apparently unappealing from both sides. This is partly due to constraints on paper length, which prevent adequate detail for all aspects of the paper, and also the fact that fundamentally large portions of the paper end up being targeted towards an audience that is not present for the given journal.

Therefore, I was wondering if I could adopt the approach whereby a high-level paper is written targeting the end users of the development and submitted to an appropriate journal, and a second paper which covers the technical aspects in detail without trying to cover the aspects necessary for the end-user audience. Considering that both papers will present the same results, is this an acceptable approach to publication?

  • I tried to create the tag "multiple-publication" - apparently it is not such a big issue but perhaps somebody would like to create this tag and add it to this question? – user1207217 Oct 27 '14 at 16:40
  • 1
    I don't believe that the tag is necessary. Using multidisciplinary and publications would adequately describe the situation as it is, and adding a third tag for a scenario where both are present seems redundant. – Compass Oct 27 '14 at 16:50
  • I think there are venues that deal with usability. I'm assuming that you have done some research that has other researchers (with a different background) as the audience. I understand you have two different types of results, depending on the perspective and the (base) state of the art that you consider. In that case there should be no problem in splitting them into two papers. Otherwise, I don't understand your question. – Trylks Oct 27 '14 at 18:35

This is actually something my group also deals with extensively. A lot of the work we do is focused on improving existing computational methods to improve their efficiency or extend their range. A lot of the time, the actual work done is of very little interest to the end-user audience, but quite significant from a computer science perspective. So a lot of our recent work has been divided in exactly the manner you propose: we present the basics of the method and some end-user applications in papers geared toward the application community, and specialized papers for the methodology that are directed toward the CS community.

The main thing to remember is that the publications should stand independently of one another as much as possible. Some overlap will of course be inevitable here, but you should strive to make the "stories" they tell as distinct as possible.

  • 4
    I've had much the same experience. This is very common in multidisciplinary work. It's important to make sure that the publications reference one another and clearly distinguish their relationships, so that there is no possibility of being perceived as duplicate publishing. – jakebeal Oct 27 '14 at 17:20

I think it is a good decision to target your manuscript for a single audience if your experience has shown that presenting it in a broader fashion misses the mark. It is not unusual to write two different papers that advance the same prior work and present new results.

However, this shouldn't simply be a matter of repackaging the results of the first paper, it necessarily entails analysis and additional work to produce a substantially new result.

Another option that can work well is to divide your article into separate, directly related articles (Part I and Part II) published in the same journal. In my experience this ends up being a lot of work, but it relaxes the length constraints and allows you to section and adjust the presentation of the material in a unique way.

On the other hand, the answer to this question as it is stated is, unequivically, no.

Considering that both papers will present the same results, is this an acceptable approach to publication?

Quality journals nearly always stipulate that they will not accept work that has previously been published elsewhere--a practice known as redundant or duplicate publication. Unfortunately, it is not always possible for reviewers and editors to be aware of multiple simultaneous submissions and there seem to be a number examples of people abusing the system with few consequences.

What constitutes original work may vary to some degree in different fields, but the point of any paper is to communicate something noteworthy that isn't otherwise available. "Results" need not be defined narrowly as new measurement data, but every article should be an honest attempt to contribute something original to the field.

For instance, good review papers are very much original contributions in the sense that they synthesize results and highlight connections that might otherwise go unnoticed. But if you set out with the premise to publish the same results in more than one article you are creating more noise for all of us trying to keep track of the literature. You may not always be called on it, but the practice is something that good reviewers and editors notice.

  • 1
    I feel that the second part of this answer is too black-and-white. No paper - certainly no good paper - is only about results; it's always about the story as well as the results, and in some cases even more about the story. I have myself published multiple papers with the same formal results, but different explanations, different language, and different motivations; and I strongly believe that, provided this is done well, it is not only acceptable but even the right thing to do. – Mark Peletier Oct 28 '14 at 23:24
  • @MarkPeletier my response to your comment was too long, please see the last two paragraphs of my answer. It may be that our disagreement is largely semantic; if you argue that the two articles are different that is another matter, but if you tell me they are the same work with nothing new then in my view it is very black-and-white situation. – dionys Oct 29 '14 at 8:54
  • I think we agree - of course, if the two papers are `the same work with nothing new' then there is little to be said. – Mark Peletier Oct 30 '14 at 15:08
  • But I do think your link to <a href="publicationethics.org/case/… page</a> is a red herring - in that case there was obvious intent to mislead, while we're talking now about the situation where everything is in the open. – Mark Peletier Oct 30 '14 at 15:10

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.