It is very common for researchers to start a large scale study which generates volumes of data and results. I wonder how to decide on how to divide the whole work in to different pieces and publish them as separate papers. To be brief, all the work can be a single paper or the work can be divided in six to twelve papers.
My guess is that this is a very discipline-specific question, so it's difficult to make a recommendation in the form of, say, the number of pages that make up the Minimum Publishable Unit.
I think that each paper should make a new and substantive contribution to the body of knowledge, i.e. something that cannot be easily extrapolated from previous papers drawing on the same study. So for example, if I had conducted a study describing the dialect of people in a region, it would likely be inappropriate to publish separate papers for each individual village, or age group which contained similar findings. Such 'salami publishing' would be frowned upon by many journal editors, and is borderline malpractice. You may want to browse the COPE case repository for examples of publications that were deemed redundant. ( http://publicationethics.org/cases )
On the other hand, there are cases when cramming too much data in a single paper would cause you to lose focus. For example, you may have some insights that are interesting, and theoretically significant, but adding them to an article would force you to remove important corroborative evidence for your main claims. Sometimes you can get that from the reviewer's comments. In these cases, it is probably acceptable to cut these insights out and develop them into a new article.
In summary, the criteria seem to be that each article should be internally coherent (i.e. tell one story), and novel (i.e. tell a new story). I realise that this is probably too general to be of practical help. :/
- Badges
- Science topic
- Similar topics
- Education
- Higher Education
- Academic Writing