An article was submitted to Journal A for publication. According to the journal’s policy, the article was scanned using anti-plagiarism detection software, which gave a 17% similarity result. As the journal allows up to 20% similarity, the article was sent for peer review to two reviewers. One of the reviewers noted that the article had been published in a similar form in a conference proceedings. The reviewer sent us the article downloaded from the conference website highlighting the overlapping paragraphs, which was nearly 80%.
The journal wrote to the authors asking for an explanation. The authors stated that they had submitted an abstract of their work for participation in the international conference. The abstract was accepted on the condition that they submitted the whole article. The author submitted the article but on the condition that it would not be published in the conference proceedings as the authors wanted to have it published in an indexed journal with an impact factor. Two emails were sent on this subject, copies of which the author provided to the journal.
The authors had no knowledge of the publication of the article as the conference organisers did not inform the authors. After the response from the journal, the authors wrote to the conference organisers asking them to retract the article from their website. The conference organisers are not replying to the emails from the authors.
Questions for COPE Council
- Is this a case of duplicate submission/self plagiarism with the authors being unaware?
- Should the journal reject the article and close the file?
- Should some action be taken against the authors although they are claiming to be unaware of the previous publication?
- What further action should be taken by the journal?
Advice on this case is from a small number of COPE Council Members. Most cases on the COPE website are presented to the COPE Forum where advice is offered by a wider group of COPE Members and COPE Council Members. Advice on individual cases is not formal COPE guidance.
This would appear to be a case of duplicate submission, and whether the authors knew it or not is irrelevant. It seems the authors did not have a reply from the conference organiser about their initial request for non-publication. If the conference did not reply and they continued with the abstract submission, the conference policy still holds. Perhaps the journal can clarify if the authors received a reply from the conference. If not, they should not retract the paper from the conference proceedings, and it cannot be republished in the journal. The authors will still have a published paper in the literature and should just learn from their mistake. If the conference agreed to the initial request and then went against their word, then that may be a mistake that can be appealed against. The journal can check that its policy requires authors to state previous presentations.
If the research is so good and the authors admit to naivety about publishing in the conference proceedings, the journal could consider asking the authors to resubmit their paper after paraphrasing and adding additional original content. The conference paper may be preliminary data only and could be developed substantially. Some journals allow this practice after authors have cleared copyright with the conference (or can show they kept the copyright) and reference the first publication. The journal could consider it as a new submission and with full referencing to the prior conference paper.
More important than having 'original' content published, which is dependent upon each journal's policy on the matter, is that the full context is transparent to readers and systematic reviewers so that data are not included twice in a meta-analysis. Hence journals need to decide a policy and make it clear in their guidelines. Whether the journal should reject depends very much on the journal policy. But the journal should expect full disclosure and should frown upon finding a prior publication that was not informed in the cover letter of the first submission. In some specialties, conference proceedings are published as abstracts only, and so republication of that material is permitted. If a specialty has a situation where conferences will publish the full text, then the journal may wish to use this case to articulate a policy going forward. It also depends on the practices of that research community. In some disciplines, publication of whole papers via conferences is the norm; whole papers are submitted, and if accepted, become part of the publication record.
In general, it is not a good idea to use hard cut-offs on plagiarism detection software, but rather to look at the areas of overlap for substantive or important duplications. Plenty of articles with 30%+ overlap are fine or easily fixed, and some severe examples of plagiarism have been caught with much less than 20%.