There is a strong incentive in academia to publish often, and a few authors respond by making one piece of work look like several. Publishing the same findings twice, or carving a single study into the thinnest possible papers, might pad a CV — but it pollutes the literature, wastes reviewers' time, and can inflate a finding by making it appear in multiple places. Editors and ethics bodies such as COPE treat the serious cases as misconduct. Here is how to stay on the right side of the line.
Duplicate publication
Duplicate (or dual) publication means publishing substantially the same paper — same data, same conclusions — in more than one journal without disclosure. It is a problem because it double-counts a single result, breaches the copyright you granted the first journal, and misleads anyone doing a systematic review who may count the finding twice.
Redundant (overlapping) publication
Redundant publication is subtler: the papers are not identical, but they overlap so heavily that the second adds little. Reusing the same dataset for a paper that reaches essentially the same conclusion, or republishing with a new title and minor reanalysis, falls here. The test editors apply is simple: does the new paper tell the reader something genuinely new?
Salami slicing
Salami slicing is dividing one study that should be a single paper into several — the "least publishable unit" approach. A trial with three outcomes becomes three papers; a survey is split by question. It is tempting because each slice counts separately, but it fragments the evidence, forces readers to reassemble the study, and can hide that the analyses came from one underpowered sample.
What is genuinely allowed
Not all reuse is misconduct. These are accepted, with disclosure:
- Conference to journal — expanding an abstract or proceedings paper into a full article, citing the earlier version.
- Secondary publication — republishing in another language or for a different audience, with both editors' consent, a clear citation to the primary paper, and a note that it is secondary.
- Separate, distinct outputs — a genuinely different question answered with the same dataset (for example, a methods paper and a clinical-findings paper) can be two papers, provided each stands alone and cites the other.
The rule that keeps you safe
Disclose every related manuscript — published, in press, or under review — to the editor in your cover letter, and cite your own prior work. Transparency turns a potential misconduct case into a routine editorial decision.
If you are unsure whether two papers overlap too much, that uncertainty is itself the answer: tell the editor and let them judge. Honesty about overlap is never the thing that gets a paper rejected; hiding it is.