How best to split up a comprehensive literature review

Question

I've written (with a co-author) a comprehensive literature review on a particular subject. Here's the preprint (it needs some editing). We're very happy with the way it turned out, but the problem we are running into is that it's too long: 49 pages, 23000 words if you include the references and a big summary chart at the end. It's "only" 33 pages and 16000 words if you remove those. We've been rejected from two journals already just because of the word count; most journals in this field max out at 8000 words. So, we're looking to split this up into 2-3 smaller parts and submit them separately.

But this is turning out to be tricky because it's a literature review, which means that while we do summarize research literature on four separate topics, we are also synthesizing what the whole of the research says and looking for broad trends. So it's very hard to disentangle the parts of the paper and repackage them without losing context or repeating ourselves.

The paper is currently structured as follows.

1. Overview
2. Methods -- including our methods for finding the research, databases we searched and queries we used, our inclusion criteria, and overall characteristics of the research literature we found
3. Findings on specific research questions (there are four of these; in the preprint we have each question as a separate section but we've since combined them all into one big section)
4. Discussion -- Primary themes over all the research literature, limitations of the studies, current holes in the research literature
5. Conclusion

It would be easy enough to write up just the findings for the different research questions, since our first and second questions are closely related and the other two could be lumped together, making two papers. But what's the best way to deal with the methods and discussion?

• If I were to split this up into two papers, should I include all of our methods in the first part and then just refer to those methods in the second part? Or summarize the methods from part 1 when we write part 2? Or something else?
• Should the discussion be split up so that the discussion section in part 1 only deals with the findings for part 1, and similarly for part 2? Or, should I save all discussion for just part 2? Or something else?

I worry that if we put all the methods in part 1 and all the discussion in part 2 (or part 3, if there is one) that people reading part 2 won't have the info they need for part 1 and vice versa. But if we split up the methods and discussion, I worry that we'll lose the sense of the big picture, which I think we did a good job capturing in the Too Large version.

Thanks in advance for your thoughts.

Answer 1

I would try to do 4* topic papers (one on on each subtopic within section 3). Then do a wrapup. I.e. 5 LPUs.

In the four topic papers you can be a skimpier on the methods (or perhaps go full bore on methods in first one and skimpy on others, but "referring" to the discussion in first one.) The wrapup will "review" the previous 4 and make some general observations (cross-correlations, etc.). I would be brief on definitions (less than currently in there) for the 4 topic papers. Just basics with some footnotes (skip history entirely). You can go more in depth on history in the wrapup self review paper.

*If a couple naturally combine, fine, do 2 or 3. But 4 is fine too:

P.s. I'm not an expert in your topic and I really just glanced at it. But what I saw looked solid.

P.s.s. Chopping into LPUs is not sleazy. It's actually in some ways more efficient to process (at least for a reviewer!)