Earlier this week, in a story by Richard van Noorden, Nature revealed the hidden workings of a scheme referred to as “citation stacking” that has landed a number of journals in trouble. The story begins:
Mauricio Rocha-e-Silva thought that he had spotted an easy way to raise the profiles of Brazilian journals. From 2009, he and several other editors published articles containing hundreds of references to papers in each others’ journals — in order, he says, to elevate the journals’ impact factors.
As Nature reports, Rocha-e-Silva was apparently frustrated that Brazilian government agencies were relying heavily on impact factor to evaluate graduate programs. That meant few scientists were willing to publish in Brazilian journals, which had lower impact factors. Rocha-e-Silva describes some of these frustrations in an impassioned 2009 editorial (in Portuguese).
The citation stacking plan was discovered, however, by Thomson Reuters, which determines impact factors, and fourteen journals — including the one Rocha-e-Silva edited until he was fired following the incident — have been punished with suspensions of their impact factors for a year.
It’s a fascinating story, well worth the read. And van Noorden — whose work on retractions is also worth a read — told us about two retractions for such citation stacking.
The two retracted articles both appeared in Progress in Biochemistry and Biophysics in 2012 and included a lot of references to Science China Life Sciences. Both of those journals, as it turns out, were sanctioned by Thomson Scientific for citation manipulation in 2011, but not in 2012.
Here’s the notice:
The following two papers published in Progress in Biochemistry and Biophysics have some improper citations. Here, at the request of the authors, we declare to retract these two papers:
[1] Science China: life science in reform in retrospect on progresses in 2010～2011. Progress in Biochemistry and Biophysics, 2012, 39(11): 1066-1072
[2] Science China Life Sciences in 2011: a retrospect. Progress in Biochemistry and Biophysics, 2012, 39(12): 1181-1189
Editorial Office of Progress in Biochemistry and Biophysics 2013-07-02
One of the articles still seems to be available online.
This isn’t the first case of retraction for citation manipulation. Last year, we reported on two others, from the Scientific World Journal. Those cases, and the whole notion of “citation cartels,” were discovered by Phil Davis.
Interesting to see this Brazilian citation scheme highlighted in Nature. What about Nature publishing group themselves? Quick check done: Could the below articles which abundantly cite a rather new Nature publishing group journal not also be considered a citation scheme to boost the IPF? Fair enough, I’d say, since it is a competitive business and as long as the article content warrants scholarly publication I don’t have a big problem with it. It could also be just coincidence, obviously. Nevertheless, it would be interesting to know how Thomson’s algorithm works and where they set the cut-off for excessive “citation stacking”.
http://www.nature.com/cddis/journal/v3/n2/full/cddis20122a.html
http://www.nature.com/cddis/journal/v2/n9/full/cddis201189a.html
http://www.sciencedirect.com/science/article/pii/S0006291X11016846
http://www.sciencedirect.com/science/article/pii/S0006291X11017256
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3227447/
I went trough 3 of these articles you’ve mentioned: Nothing justify citing the same journals so many times. Some of the subjects just can’t be the key reference to be used in a couple of paragraphs….
Two of the papers above are Elsevier journals. Are you somehow suggesting collusion between Nature Publishing Group and Elsevier in the same way in which Rocha-e-Silva, editor of Clinics, colluded with other Brazilian journal editors to manipulate the Impact Factor of their journal? Some editors and/or journals in my field request, as part of the “peer” review, to include references from their own journal, in other words, if you do NOT include references from that journal, the paper can or will be rejected. Not only is this citation stacking and citation manipulation, it is blackmail. I will of course not list the specific journal or the editor-in-chief enforcing this rule, as I am now trying to gather more proof.
I’ve seen the “cite this work” issue come up multiple times in peer review for many journals across a few disciplines. In most cases, we were referred to papers that we had overlooked in our searches and that were indeed directly relevant to the “story at hand”. (Further going to show that’s is quite difficult to read *everything* these days, even in narrow fields.) I did encounter one case of the referee inappropriately requesting a citation; and I brought it up directly in my response to the referees to explain why I felt such a citation would not be relevant to the scientific narrative at hand. I think some of current state may in part be due to a bit of confusion across nations, cultures, and scientific disciplines on standards for citation, and what constitutes a proper bibliography for a scientific paper……
I had a look at the 2 nature group papers. They seem to be reviews of what have been recently published in this specific journal, so I guess it makes sense they would refer many (only) papers from this journal. At least it’s clear from the title and the abstract. Not sure one can call that a scheme.
I agree. It also appears that apart from one of the papers the senor authors are all editors of the new NPG journal or its partner journal. Probably an approach to guarantee an acceptable starting IPF.
Many journals use ‘tricks’ to raise impact factors besides publishing ‘guidelines,…’ For example, The International J of Cardiology obliges authors to refer an editiorial from the Ed in Chief about ‘Ethics in Publishing!’ in each article. Hence, the citation rate is automatically huge
That’s very ironic to require something so unethical…. as referring to an editorial on ethics in publishing. This editorial has been cited more than 2000 times!
Talking about tricks. Various journals publish papers as e-pub or preview papers online first before publishing them officially in an issue. This can be argued to offer peers the opportunity to get an early view on what is coming. However, some journals keep all papers online for more than a year, before they are assigned to an issue number. If I am not mistaken (but willing to be corrected), that means that the two/five-year period that counts for the calculation of IFs starts only then. Conveniently, that is when a lot of other papers may already be citing those new papers because they have been out for that long already. Thus, the journal avoids the first year in which few citations can appear because it takes a while for those citing papers to be published as well. Thus rasing IFs. I suspect that is a trick as well, or is it?
I think this post would deserve a tag for Brazil…
It is amazing to see that the passionate efforts of a scientist to save Brazilian journals have actually pulled them a bit lower down the gut. A true example of that Hell is full of Good Intentions…
I don’t get how this is in anyway unethical. Journals like science and nature routinely limit the number of citations that a paper can contain or charge the authors. This is blatantly unethical and forces authors to cite reviews rather than original research and only the most important papers (which are often in their journals anyways), and often times in the big reviews are in their own journals.
Since when have we become such a slave society to impact factors that they are something that are UNETHICAL to manipulate? And hell, journals like nature/science DO manipulate them directly when they reject papers with sound science simply because they are not “interesting enough to a broad audience” (read: will not be cited as much or won’t make a catchy NYT headline), often times in favor of science that’s less sound/comprehensive?! And if you look at the cadre of names publishing in top journals over and over, you know that it’s an old boys club that promotes each other’s science that backs each other’s trendy ideas and theories.
Not citing a source that you drew your work from is unethical. Anything else? Meh… an article writen purely to cite papers from a certain journal is probably a crappy article, but nothing more… and as scientists, why should we waste our time on this?
Quite true…
It is interesting to read the comments by the editor Mauricio Rocha-e-Silva on Nature website. (If one can understand them…) In the end the argument is exactly that; 1st world journals do it, so why do 3rd world journals get sanctions on it? It is a good point indeed,