Publication ethics is the set of standards that keep the scholarly record honest and trustworthy. It governs how authorship is assigned, how originality and data integrity are maintained, how conflicts of interest are disclosed, and how errors and misconduct are corrected once published. For any researcher submitting work, these are not abstract principles: most journals now require explicit ethics statements at submission, and a breach, even an unintentional one, can derail a paper or a career.
This guide maps the two frameworks that dominate the field, the COPE guidelines and the ICMJE recommendations, and walks through the core issues every author should understand.
Science is cumulative. Each study builds on the published record, so the value of that record depends entirely on its trustworthiness. When a fabricated dataset, a hidden conflict, or a stolen passage enters the literature, it corrupts everything built on top of it. Research integrity is the collective effort to prevent that corruption, and publication ethics is its operational rulebook at the point of publishing.
Two bodies set the standards most journals follow. The Committee on Publication Ethics (COPE) publishes broad guidance and practical flowcharts that editors use to handle disputes and misconduct across all disciplines. The International Committee of Medical Journal Editors (ICMJE) issues recommendations focused on medical journals, including the authorship criteria and disclosure rules that have become a de facto standard well beyond medicine.
Authorship is the most common source of ethical dispute. The ICMJE defines an author through four criteria, all of which must be met: substantial contribution to the work, drafting or critically revising it, approving the final version, and agreeing to be accountable for it. Two failure modes recur. Gift or honorary authorship lists someone who did not genuinely contribute, often a senior figure. Ghost authorship omits someone who did, often a professional writer or an industry contributor. Both distort the record of who is responsible for the work. Our dedicated guide to authorship criteria covers the rules and the common disputes in detail. The single best protection is to agree author roles and order at the start of a project, before results, and therefore incentives, arrive.
Plagiarism and Originality
Plagiarism, presenting others' words or ideas as your own, is the clearest integrity breach. It ranges from copied passages to unattributed paraphrase to self-plagiarism, the reuse of your own previously published text without disclosure. Journals screen submissions with similarity-detection software, and a high overlap score triggers scrutiny before review. Originality also means not submitting the same manuscript to two journals at once, which wastes reviewer effort and can produce duplicate publication.
Data fabrication (inventing data) and falsification (manipulating or selectively reporting it) are the gravest forms of misconduct because they corrupt the evidence itself. Beyond outright fraud, ethical reporting requires transparency: pre-registration where appropriate, honest description of methods, and availability of data so that others can verify findings. For evidence syntheses, this transparency is built into the method; our guidance on conducting a rigorous work with our systematic review methodologists and reporting it against the PRISMA abstract checklist reflects how integrity is operationalised in practice.
A conflict of interest is any relationship, financial or personal, that could bias the work or its interpretation. The ethical requirement is not to have no conflicts, which is often impossible, but to disclose them. A funded study or an industry consultancy is not misconduct; concealing it is. Most journals require a conflict-of-interest statement at submission, and the ICMJE provides a standard disclosure form widely used across medical publishing.
When Things Go Wrong: Corrections and Retractions
Even careful work contains errors. Publication ethics provides a graded response:
- A correction (or erratum) fixes a specific mistake while the article remains valid.
- An expression of concern flags an article whose reliability is in doubt while an investigation proceeds.
- A retraction withdraws an article whose findings cannot be trusted, due to serious error or misconduct, while keeping it visible with a clear retraction notice so the record remains transparent.
Retraction is not erasure. The article stays accessible, labelled, so the literature that cited it can be re-evaluated. COPE's retraction guidelines define when each response is appropriate.
For authors, the practical upshot is that ethics is now part of the submission package, not an afterthought. Expect to provide authorship confirmation, a conflict-of-interest disclosure, ethics-approval details for studies involving humans or animals, and a data-availability statement. Getting these right signals a careful, credible team. Our medical writing from PhD writers and manuscript editing from PhD editors services prepare these statements to journal and ICMJE standards as part of readying a manuscript, and our guide to choosing a target journal explains how requirements vary across titles.
Sound publication ethics is ultimately about accountability: being clear about who did the work, where the data came from, what interests are in play, and how errors will be fixed. Journals reward that clarity, and the scholarly record depends on it.