I am wondering how can I split my research into two or more research article. I heard that there's a type of papers that called revisiting which basically relies on previous work with some enhancements.

The question is what's the structure for the revisiting paper and what's the structure for the revised paper and what are the differences between them.


Although you can split a paper in two, it is often not the best choice, known as salami publication.

There are, of course, papers revisiting a certain topic, sometimes tens or hundreds years after the first publication. Revisiting a topic is not the same, as simply splitting a paper in two. There is no specific structure for a revisiting paper - the term refers to its relation with previous research only.

In academic publishing, we call a revised paper a version of the manuscript which underwent a peer review process (i.e. in a journal), received feedback and was already corrected (revised) based on the feedback from peers. This term refers to the stage of the publication process and not to the structure or purpose of the paper.

| improve this answer | |
  • Note that the concept of "salami publication" is not universally acknowledged, but seems to be specific to empirical fields. The website you cited defines the concept as "a publication of two or more articles derived from a single study" -- in theoretical science, for example, there is no concept of "studies". – lighthouse keeper Sep 6 '19 at 9:11

Revisiting papers

These are papers that are examining a topic from the past (probably years), and then providing new insights, a review, or new data/experiments/whatever that puts the old stuff in a new light. As pretty much all scienctific papers are in the business of "new insights", one might ask the question why does such a thing like "revisiting" paper even exist. In some fields there are certain topics that have been neglected, for various reasons. Then it is appropriate to do something like this. Usually the people who write a "revisiting" paper are experts in the field with years of experience.

Revising a paper

Just an ordinary process of peer review, that (almsot) every paper goes through. You submit a paper to a journal, it goes to review, and comes back with a list of issues the reviewers had. You revise the paper in order to address the issues and make the paper suitable for publication.

Both of these things are unrelated to your wish to split your research into two papers or more.

Some of the criteria to decide whether to split you research are:

  1. Is the research in its current "one paper" form too long compared to other similar papers in similar journals?
  2. Are there several main ideas in the research that each one can be in a paper of its own?
  3. Is the research such that different aspects of it are relevant to researchers in different fields, such that it is appropriate for publication in different journals?

If you want to split your research just for the sake of having more papers, do not do it. It's called "salami slicing" (see the other answer here) and is usually frowned upon because it is a waste of everybody's time.

| improve this answer | |
  • Thanks a lot for the explanation however, I found some papers in my field, authors have re-implement old algorithms and re-publish it again with some enhancements, for instance, this article "Revising Max-Min for Scheduling in a Cloud Computing Context" link, So, that was my concern about the revising. – M Ala'anzy Sep 6 '19 at 12:41

In my research area, software engineering, the concept of "revising/revisiting article" is not standard terminology.

There is, however, a very similar concept: the publication of extended versions. An extended version extends a paper by adding some new material. The extended version is always published in a journal, whereas the original paper may have been published in conference or workshop proceedings (which are legitimate publication venues in my area).

Almost all journals accept submissions of extended versions. They explicitly specifiy the amount of required material for an extended version (typically between at least 25% to 40%) and the nature of the extension. Some journals may be fine with adding more details (like a more detailed proof to a theorem), whereas other journals will require substantial new contributions (like a significantly extended empirical evaluation).

| improve this answer | |

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.