![[IOP]](http://images.iop.org/cws/icons/themes/phw/colophon-small.png){kind=icon}
By Hamish Johnston
Not exactly, but they are near the top of the league table when it comes to publishing the same paper in two different journals — according to a preprint from sociologists Vincent Lariviere and Yves Gingras at the University of Quebec in Montreal.
The pair combed millions of papers published between 1980-2007 looking for articles with the exactly the same title, first author and number of references. They found nearly 5000 papers that had been published twice — or about 0.05%.
They then compared the abstracts of the duplicate pairs (when available) and found that more than 52% were identical — and the remaining 48% very similar.
So how did physicists make out?
Nearly 0.08% of papers were duplicates, putting physics in second place after “engineering and technology” with over 0.11%.
Lariviere and Gingras point out that the high number in engineering and technology could be related to the large numbers of conference proceedings published in this field. Interestingly, I had a chat about this with a few journals editors here at IOP Publishing and they told me that publishing a paper in a conference proceedings and then a journal seemed to be a common practice for engineers.
Does this duplication matter?
It does if hiring committees or funding bodies simply tote up a candidate’s publications. However, if it’s quality they are looking for, then duplicate papers appear to be rather poor — Lariviere and Gingras show that the average “impact factor” and number of citations of the duplicates is about 65% of the average value for a physics paper.
Are there legitimate reasons for publishing the same thing twice?
I suppose your could make a case if the research cuts across two different disciplines that rarely read each other’s journals.
But in an age when peer-reviewed publications are the currency of success, it does seem like counterfeiting.
These are very good observations. Regarding the peer reviewed journals, eliminating double publishing rests as the responsibility of the referees. It’s their duty to determine whether a submitted paper is just recycled and whether it has enough new results to justify another publication. Indeed, this can only work if the referees are well aware about what is happening in their field; in other words, in the quality of the referees. Another problem is the proliferation of journals. It may be sometimes difficult for a referee to be aware if the paper in question has been already published by some obscure regional journal which hardly anybody reads. I have often found that this can be solved by using search machines, such as Google Scholar, but I wouldn’t give a 100% guarantee.
Publication in conference proceedings is a different matter. It’s not uncommon that some of the main conferences only publish abstracts, and when they publish extended papers, they are limited in space. Furthermore, they are seldom refereed with the same rigour as archival journals. Therefore, it is a common practice to publish preliminary results in conference proceedings, and then submit the full work to a refereed journal. Where I come from, the evaluating committees simply separate them, and papers in refereed journals are given a greater weight. There’s also a tendency to classify them according to the impact factor of the journals in which the papers are published… but that’s yet another matter you may like to address separately. In the end, I think that what should matter most is the net contribution to the field by a given researcher or research group, independently of all this.
Does the lower impact factor for the duplicated papers take into account the fact that the impact of each paper is effectively cut in half – half the people reference one paper, the rest the other?
Repetition is mother of wisdom, isn’t it?
I was glad to learn today from IOP that:
“As part of our ongoing commitment to prevent plagiarism, IOP Publishing has joined the CrossCheck service (http://www.crossref.org/crosscheck.html). This initiative will allow us to compare a submitted paper’s content
against material published by most of the core publishers in the field as well as preprints and other content available on the web. We are sure you will agree that this is a positive step towards isolating cases of plagiarism prior to publication.”
Many conference proceedings are not refereed, and physicists publish in them as a courtesy to the conference organizers. Later they publish their paper in a refereed journal.