Some authors of retracted studies persist in citing their retracted work, according to a new study in Science and Engineering Ethics that calls the trend “very concerning.”
From the abstract of the paper by Charisse Madlock-Brown and David Eichmann of the University of Iowa:
Our findings indicate new reasons for retractions have emerged in recent years, and more editors are penning retractions. The rates of increase for retraction varies by category, and there is statistically significant difference of average impact factor between many categories. 18 % of authors self-cite retracted work post retraction with only 10 % of those authors also citing the retraction notice. Further, there is a positive correlation between self-cites and after retraction citations.
The authors note that “In recent years, there has been a dramatic increase in the number of retractions written by editors:”
Ferric Fang, whose work on retractions will be familiar to Retraction Watch readers, said that was good news::
The increasing role of editors in retracting articles is a welcome trend, and Madlock-Brown and Eichmann document that it is a quite recent phenomenon. Not so long ago many editors used to feel that only authors could retract a publication, and this allowed the persistence of fraudulent work in the literature.
The authors say they’ve identified a new category of reasons for retraction, namely editorial mistakes for which the authors are blameless. But we’ve certainly seen a lot of those over the past few years, and Grant Steen noted them in a 2010 study that the new one cites.
The authors of the new paper conclude:
That many authors of retracted work continue to self-cite is very concerning, particularly given that many refrain from citing the retraction. Clearly, if the author needs to refer to their retracted work, the retraction should additionally be cited. We found a positive correlation between self-cites of retracted work and the number of citations to the work after retraction. This correlation suggests that authors may be able to influence the way their retracted work is viewed is viewed by referring to retracted work without citing the retraction and thereby maintaining the appearance of legitimacy. However, we can’t necessarily assume nefarious intent. Authors may self-cite while highlighting aspects of their work that remain valid. They may not consider self-citing without citing the retraction as inappropriate. It is further possible that not all listed co-authors are made aware of the retraction and cite it in good faith.
The phenomenon of self-cites to retracted work is a serious problem. Most (66 %) of the self-cites involve retractions due to errors and non-reproducibility. Boosting the overall citation counts with self-cites increases the perceived legitimacy of the work and can lead to wasted hours of research and resources.
They call for a retraction database, which we’ve said for a few years that we want to create (and have begun working on, with our categorization of posts). And also for publishers to use CrossMark, which we endorsed in late 2011.
Fang told us that “most of the observations seem confirmatory of earlier studies,” but that the title of the paper — “The (lack of) Impact of Retraction on Citation Networks” — might be a bit misleading:
It has already been shown in multiple studies that citations decline following a retraction, so retraction clearly does impact citation networks (e.g., Pfeifer & Snodgrass. JAMA, 1990; Lu et al. Sci Rep, 2013). The new study shows that some authors continue to cite papers that they have retracted. I’m not certain what to make of this. Perhaps this is inappropriate, but perhaps not. Most of these self-cited papers were retracted due to error or non-reproducibility of data. As the authors acknowledge, there may still be legitimate reasons for citing a retracted paper if some of the data remain valid, and one would have to check on a case-by-case basis to determine whether the citation was appropriate. Since this was not done, I am not persuaded that this is “very concerning,” as the authors suggest. If retracted articles are clearly marked as such and identified in databases, the research community should be reasonably safeguarded from mistakenly regarding the articles as valid.
Hat tip: Rolf Degen
Like Retraction Watch? Consider supporting our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, and sign up on our homepage for an email every time there’s a new post.
In my view, this is just one category of “inappropriate citations”, which do not all fall within the scope of RW of course. Numerous other examples exist, like citing a paper that doesn’t actually claim anything similar to the statement the authors wish to substantiate; anyone who has had a look at the contexts in which their own work is cited probably recognises this? I do. 🙂
If the field truly finds that it should be impossible to cite the original paper without citing the retraction too, then it should make it impossible to cite that paper. For instance by completely withdrawing the paper from the literature. Of course that has the disadvantage that no one would have access to the contents of that paper anymore (for instance to verify that it was rightfully retracted), but that can easily be corrected by quoting the entire original paper verbatim in the retraction notice. Thus the paper itself would become “meta-content”, as it probably should.
True, that doesn’t work for paper copies of the journal, but that practice is quickly becoming obsolete anyway. If the databases wouldn’t list the original paper anymore but the “retraction-notice-plus” instead, and publishers would check reference-lists of accepted papers against such databases, that would prevent this problem from occurring.
(Aside, it would also give you more value when you need to pay $$$ just to access the retraction notice, although that remains a bad practice nevertheless.)
Should Seralini et a be able to cite the retracted paper by Seralini et al? What if Seralini actually DOES bring a successful lawsuit to have the paper un-retracted? Should the un-retraction of the retraction be noted?
When an author disagrees with the reason for retraction and can point to editorials that disagree with the reason for attraction, should the editorials be noted as well in any citation?
There is also the possibility that an idea in a paper needs to be cited, while the paper itself was retracted because of bad data. If the retraction notice doesn’t explicitly mention that idea, is it appropriate to cite?
If a retracted paper was referenced in the reference list of a new paper, this shows two unequivocal things:
a) the dishonesty of the authors, their lack of responsibility towards science, and their arrogance towards the community;
b) the incompetence of the editors, editor-in-chief and journal/publisher for not having in place measures to force authors to NOT reference retracted papers or to, should they choose to reference them, also have statements that clearly state that the paper has been retracted and that the results are thus invalid.
Some time ago, I had already pointed out how I had discovered the posting of retracted papers/work on Academia (http://www.academia.edu/). THe uncontrolled world of open access, which is of course inherently an extremely positive thing, is being corrupted by individuals with no ethos, no scruples, and no remorse.
For both category a) and b) individuals, there is only ONE viable and immediate solution: public embarassament. This should involve an increase in the number of blogs that show clear cases of hipocrisy. The list of such blogs, which should be established, and maintained by academics, rather than wasting their time on editor boards for commercial publishers, should set up an International Organization of Science Justice. However, so as not to feed the hungry appetites of lawyers, the site needs to be based on academic and ethical principles, should exclude groups such as COPE, because I think their funding by publishers gives them a conflict of interest, and should strike the balance between truth-telling, and libel. One can easily show abuse without being abusive, if the facts are clearly, and carefully, portrayed.
I understand that my more radical approaches, in some respects, will have the rights advocates up in arms, but what this study indicates is that a publishing world that is unregulated, or regulated by dozens of different publishing “guidelines” serves only to protect the economic interests of the publishers. Science is being trashed, scientists are being trashed, and far too many scientists who actually do have a solid ethical base are afraid to speak up, or take action, because there are no structural mechanisms in place to defend them. Because the entire global publishing structure is so fallible, porous, susceptible to abuse, or mockery, why should a researcher whose paper was retracted, but who spent tens of thousands in budgets, possibly thousands of hours investing hard work, bow to “ethical” rules established by often publishers that demonstrate two-faced attitudes and policies towards ethics?
Although this news is not surprising, I am not pleased to read it, but feel that the problem also lies with the lack of suitable structure provided by publishers.
Such blanket statements fail to take into account the diverse reasons for retraction and the potentially valid reasons for citing a retracted article. One example is a 2008 paper by Forman et al. in Infection and Immunity, a journal for which I serve as the editor-in-chief. The paper described YadBC as a new virulence determinant of Yersinia pestis, the plague bacillus. The authors subsequently discovered that the mutant strain used in the bubonic plague infection model had acquired a second site mutation that was responsible for the effects on virulence. As this affected one of the central conclusions of the study, the authors retracted the 2008 article, but the retraction notice stated that the “rest of the paper was not affected” by the error. Last month the authors published a follow-up study of YadBC and its temperature-dependent phenotypes. The 2008 article was accordingly cited. Precisely how does this reflect poorly on the integrity of the authors or the editors? On the contrary, I think that reckless accusations reflect poorly on the individuals making them.
Ferric Fang, thank you for the critique. One is not a very representative sample size. Feel free to expand the list of valid mentions of retractions. It is obvious that the Forman et al. paper has used the referencing of a retracted paper in a valid, and honest way. Kudos to them, and to you for using your expertise on the editor board, to understand this fine line. But you seem to think that 100% of editors, and editors in chief, of many journals, may have, understand, or perceive, the same risks. In your case, Infection and Immunity, published by ASM, may have a great academic foundation because its editors, like you, understand these fine-scale issues. But in the plant sciences, particularly Elsevier and Springer journals with IF of about 0.5 to 3.0, supposedly “academic” and “peer reviewed” journals, dozens upon dozens of flaws, even when reported, are ignored. In such journals, where the editorial firewall is not aggressively ignorant, but naively short-sighted, I would disagree with your last statement. The only reason why I have not yet published (or been able to publish) such non-academic behaviour by these elistist editor board members, and the publishers who hold their backs, is because of the work load I have. Also, there is massive resistance to publish such case studies. So, after a couple of attempts, I have come to the conclusion that the only viable way forward is through self-publishing. Incidentally, I listed you as one potential reviewer of a 80-page post-publication peer review recently submitted to, and rejected by PLoS Biology, because they did not like my “structure”. For the record, I would be grateful to receive a confirmation that you were not involved with that peer review. As you can see, the potential for building bias towards, or against individuals, even editors, exists, even on blogs. Imagine my next paper on an ethics- or retraction-related paper suggests Ferric Fang as the “peer” reviewer, how this excahnge between us would not already have a base of bias. This is why the whole traditional “peer reviewer” system is totally flawed: because it is so susceptible to influence.
Another example is a retracted Sep 2010 article by Lo, et al. in PNAS (doi: 10.1073/pnas.1006901107), concerning murine leukemia virus-like sequences in the blood of patients with chronic fatigue syndrome. The authors retracted the article in Jan 2012 (doi: 10.1073/pnas.1119641109) due to their concern that the work could not be corroborated. In Sep 2012 the senior author (Alter) published a follow-up article in mBio (doi: 10.1128/mBio.00266-12) concluding that there is in fact no association between murine leukemia virus and chronic fatigue syndrome. That article self-cited the retracted PNAS article as a source of methodology and to explain the history behind the study. In my view the self-citation of the retracted article was completely appropriate, and this sequence of events is in fact an admirable example of researchers making every effort to ensure the integrity of the scientific record. My point is not that all self-citations of retracted articles are appropriate, but merely that each case should be evaluated on its own merits. I object to the assumption that every self-citation of a retracted article implies dishonesty or incompetence on the part of authors and editors. The two cited examples prove that this cannot be assumed.
I do not feel it is appropriate for you to ask whether I was an anonymous peer reviewer of your submission to PLoS Biology. However, since you have asked, I will assure you that I was not involved.
On the Beall blog, criticism has been leveled against your paper with Casadevall by Bill White:
http://scholarlyoa.com/2014/03/27/new-oa-publisher-the-council-for-innovative-research/#comments
Your paper (http://mbio.asm.org/content/5/2/e00064-14.full), published in March 2014, does not reference, or acknowledge, a paper published in SEE, in January 2014. In his comments, he states ““The disaster of the impact factor” where we can read the question:
“…Why such a biased tool continues to exist in Science?…”
http://link.springer.com/article/10.1007%2Fs11948-014-9517-0 but Casadevall and Fang do not cite the source paper! This is unethical! Other information is also overlapping with this paper without proper citation! Unethical and patronizing behavior from Casadevall and Fang?” Care to comment on the validity of this claim?
This is a scurrilous accusation. This is the first that I have heard of the article by Moustafa in Science and Engineering Ethics. First, I would point out that Moustafa’s paper was published on January 28th, and ours was submitted to mBio two weeks earlier, so there is no way that we could have been aware of the work. Second, Moustafa’s paper and ours have very little overlap, other than we are both critical of misuse of the impact factor. Moustafa focuses on damage caused by the impact factor, whereas our focus is on the underlying reason for its persistence as a surrogate measure of article quality in the scientific community. Third, the impact factor has been around for decades, and there are hundreds of articles written about it. An author is not unethical for restricting references to earlier publications that have directly informed his or her work. I don’t see what this has to do with the topic of this thread but thank you for demonstrating how easy it is to make casual irresponsible and libelous comments on blogs.
Nowhere did Ferric Fang claim anything about “100% of editors, and editors in chief, of many journals”. On the other hand, you (JATdS) made an assertion about 100% of cases where a retracted publication is cited: that it shows unequivocally that the editors are incompetent and the authors are dishonest. This sweeping indictment is clearly unwarranted. Fang merely said so, and provided a counterexample.
There’s a *lot* of variation between fields and sub-fields in scientific publishing on what the role of the editor is, what warrants citation, when retractions are in order, and whether retractions due to “we were totally wrong about one aspect of the paper due to this new data but a few of the ideas seem to hold” should still be cited. There’s a good argument for keeping “honest but wrong” papers *in* the literature, so no one wastes time and money barking up the wrong tree.
Some of this seems historic, some seems very political, a lot of it looks to be monetary, and a small but hopefully increasing fraction is common sense. There are darn good reasons for a lot of the variations in publishing standards- a synthetic organic chemistry lab is hardly the same thing as a molecular biology lab; and both are of course separate from how a computational theory group should be run and managed. The publications feed directly into how much funding the lab in question gets, so these are not trivial issues.
Trying to over-generalize- saying “because we do it this way here, that is the way it must be done for everyone”- might not be helpful. Some fields do have legitimate editorial power issues- in others, everything is hunky-dory. And supposedly “common” sense can in fact become quite rare……
yes, in this vein, it’s possible that someone published a paper outlining A) a hypothesis and B) experiments that test that hypothesis. imagine a geneticist published a paper first outlining a theory, never proposed before, that there are two copies of each gene. he then uses fraudulent data to “prove” his theory. this fake data was uncovered… and thus the paper was retracted. some one else thinks, hum, maybe there was something to that theory, tests it and proves it. it would A) be wrong to cite the original data, as it was fraudulent/retracted and B) wrong not to cite the original idea, even though the paper was retracted, because even though it wasn’t backed up by any evidence, it wasn’t your original idea. mentioning the retraction notice could be construed as unnecessary because you were citing an idea (that as long as it had not been plagiarised) exists independent of the retraction. thus the dichotomy of citing someones once printed ideas, which, really can never be fully “retracted” out of existence (and failing to cite them would be plagiarism) and citing data, which of course can be fully retracted/identified as having been fake.
The issue here seems to be citing the paper, but not citing the retraction. In the example by QAQ I would have though the solution is along the lines of : “In a study subsequently retracted (citation retraction) it was proposed that there are two copies of each gene (citation original paper)……”
Of course you have to cite the retraction as well as the original article! In Web of Science, retractions count as separate publications, so this is a no-brainer for boosting both your citation count and your h-index. Even better, a retraction counts as a citation to the original article, so every time you retract one of your papers, you get one citation for free!
(This posting might contain traces of sarcasm.)