Phyllis Wise, the chancellor of the University of Illinois and an obstetrics researcher, has called for a massive correction of a 2006 paper in Neuroscience for work she appears to have tried to pass off as having been previously unpublished — but which wasn’t.
The article, “Estrogen therapy: Does it help or hurt the adult and aging brain? Insights derived from animal models,” has been cited 47 times, according to Thomson Scientific’s Web of Knowledge.
And it had caught also the attention of readers on PubPeer, who noted that:
This paper (Neuro06) is a duplicate publication, presenting as new results already published in the paper (Endo01):
Neuroprotective Effects of Estradiol in Middle-Aged Female Rats, Dena B. Dubal, Phyllis M. Wise, Endocrinology Volume 142 Issue 1 pp43-48 (2001)
http://press.endocrine.org/doi/abs/10.1210/endo.142.1.7911It includes self-plagiarism, the misuse of images and the removal of co-authors, as well as the presentation of recycled scientific results in a manner that raises questions as to their veracity. In detail:
Self-Plagiarism: about 80% of the text of this paper is taken verbatim from Endo01: abstract, introduction, methods and materials, results and discussion. Endo01 is not cited. The only significant differences are the parts about bcl-2, ERα and ERβ, which do not appear in Endo01, but have been copied almost verbatim from another paper (JNeuro99):
“Estradiol modulates bcl-2 in cerebral ischemia: a potential role for estrogen receptors”, D B Dubal, P J Shughrue, M E Wilson, I Merchenthaler, P M Wise, J. Neurosci., Volume 1 Issue19(15), pp 6385-6393; (1999)
http://www.jneurosci.org/content/19/15/6385.fullWhile JNeuro99 is cited, at numerous points in the paper it suggests that the findings regarding bcl-2, ERα and ERβ expression are from this study rather than previous work (e.g. the first paragraph of the Discussion).
Misuse Of Images: All of the Figures in this paper have appeared previously:
1. Figure 1 of Neuro06 is identical to Figure 2 of Endo01
2. Figure 2 of Neuro06 is identical to Figure 3 of Endo01
3. Figure 3 of Neuro06 is identical to Figure 3 of JNeuro99
4. Figure 4 of Neuro06 is identical to Figure 5 of JNeuro99Endo01 is nowhere cited, while the Figures taken from JNeuro99 are never indicated as having appeared previously, let alone permission having been obtained. In most instances, information in the original Figures’ legends has been excluded upon reproduction, yielding less information (e.g. p-values have been removed).
Moreover, Figures from the plagiarized papers have been removed and references to them have been replaced by the phrase “data not shown”. For example, this phrase appears at the end of the first paragraph on page 832, replacing a reference to Figure 1 in Endo01, and a similar phrase at the end of the second last paragraph on the same page replaces references to Figure 4 in JNeuro99.
Removal Of Co-Authors: The paper under consideration is a single-author paper, while the paper it is mainly copied from, Endo01, has two co-authors, with different first author. The other source for the material, JNeuro99 lists 5 co-authors, of which the current author is listed as senior author.
Presentation Of Recycled Scientific Results: the presentation of the different results from two papers is done in a manner that is misleading. According to the Materials and Methods section of JNeuro99 and Endo01, only young mice were included in the bcl-2, ERα and ERβ expression experiments, a fact that is not mentioned when this data is presented as new in Neuro06.
In summary, such inconsistency, together with the suppression of information and misleading statements indicated above, raises serious questions about all aspects of this work. This is entirely in keeping with the following previous and subsequent publications by the same author:
https://pubpeer.com/publications/2EE674599778410770DFA3BA3A02F8#fb13702
https://pubpeer.com/publications/72872078379F852062CEC8ADA0939A#fb14353
Neuroscience evidently agrees. According to the corrigendum:
The author wishes to correct a number of serious errors in the writing of her publication, Neuroscience (2006) 138: 831–835. The paper was intended as a review of the previous work from her lab, submitted to a Special Issue of Neuroscience as part of the report of a Conference. However, the paper is written in a way that misleads the readers to think that it is an original article. The author wishes to correct that impression with the following changes in the text of the published paper:
Add to the reference list a previous paper from the author’s laboratory that was left out inadvertently: Dubal DB, Wise PM (2001) Neuroprotective effects of estradiol in middle-aged female rats. Endocrinology 142: 43–48.
Add a sentence to the legend of Figure 1: Reproduced with permission from Dubal and Wise (2001).
Add a sentence to the legend of Figure 2: Reproduced with permission from Dubal and Wise (2001).
Add a sentence to the legend of Figure 3: Reproduced with permission from Dubal et al. (1999).
Add a sentence to the legend of Figure 4: Reproduced with permission from Dubal et al. (1999).
The author regrets the extent to which the text in Wise (2006) was recycled from the prior papers.
Wise told Retraction Watch she agrees with the correction, and that there are no plans to correct any other papers.
So psychological: self-plagiarism. Lets petition for an NIH study on this.
Given the relatively high number of times this paper was cited, this case, as well as others like it, could really help us determine the degree to which self-plagiarism of data can distort the scientific record. Although such a study would, ideally, be carried out by someone who has more than a passing familiarity with the relevant literature, all is really needed here is to track down those papers that have cited this review plus the earlier papers on which this review was based, and assess whether or not the citing authors interpreted each paper as an independent contribution and whether the combined literature was interpreted to show a conclusion that is much more inflated than what the effects, at the time of the citation, actually showed.
Sorry for the very long sentence.
🙂
Retraction Watch wrote: “…for work she appears to have tried to pass off as having been previously unpublished…”
Is it fair to impute motive without citing evidence of this? Or at least a comment from the author?
Given that the correction notice, with which Wise said she agrees, states that “the paper is written in a way that misleads the readers to think that it is an original article.”, I think this is not such a bold statement from Retraction Watch.
Editorializing is the issue. She did not admit to intentionally misleading her readers” and did not comment on what her motives were.
Yeah, well, part of becoming a scholar means: analyzing and interpreting material. You can say the author is “editorializing” if it makes you happy to say so. Or you could say this: the author is deriving the same meaning from these materials that any reasonably skilled analyst would derive from them.
Sorry, your comment makes no sense. Your point is?
Her CV is online. The article is listed in her CV. So, it’s not like she didn’t know this second publication existed.
Where do you see a statement about “motive” in what RW wrote?
Anyway, let’s not play lawyer, this is not a court of law. Everyone is trying to get as many publications on their CV as possible, so what happened is understandable, and not as serious as data manipulation.
Reblogged this on The Academe Blog and commented:
Now this, from Retraction Watch, is fascinating, especially as it concerns University of Illinois Chancellor Phyllis Wise:
@Miguel Roig I would argue that the parallel citations do not distort the record in this instance, since it is a case of plagiarism, not data fabrication. It does, however, distort the record of citations and increase the noise of impact factors. However, where the two works have been cited in the same breath by other authors, it does beg questions as to whether they read the two papers side by side. Identifying the number of uncritical citations to both (or either) works would highlight sloppy thinking.
This does raise an interesting question regarding ethics and the enforcement of rules. Can we assume that a University of Illinois student who is found to have plagiarised work in a major way is allowed to keep the full marks and issue a “mega correction”. I hope so, otherwise the university is applying different standards to its students and staff. Any real information on that? Perhaps RW could enquire?
Feniglab, admittedly, I have not read any of these papers. My comments were based solely on the RW post, especially the PubPeer comment stating: “Presentation Of Recycled Scientific Results: the presentation of the different results from two papers is done in a manner that is misleading. According to the Materials and Methods section of JNeuro99 and Endo01, only young mice were included in the bcl-2, ERα and ERβ expression experiments, a fact that is not mentioned when this data is presented as new in Neuro06”. If, indeed, no earlier published data were presented as new in the Wise corrected paper, I withdraw my earlier comment as being applicable to this case.
You are the author, I believe, of the white paper on self-plagarism and plagarism which is referenced at ORI. What is your opinion of the history of the “self-plagarism” concept. I find scanty references to it prior to the middle of the 2000-2010 decade. Any thoughts of the history of the concept? This is an area in which you have done some research and thinking.
I believe that you, Statistical Observer, are generally correct about the notion of self-plagiarism having become much more visible as issue of concern within the past decade or so. Frankly, I haven’t looked at this issue from a historical perspective, but a recent editorial on the subject gives you some perspective on its growth in popularity in line with your own experiences: “More and more publications are appearing about issues of self-plagiarism, and much debate has ensued about the “scourge of self-plagiarism” (Green, 2005). In 2005, Green noted that a Google search of the keyword “self-plagiarism” resulted in 8,000 hits; in 2010, Brown-Syed found 38,000 hits; and in 2013, I conducted the same Google search and found 82,500 hits”(1, pg. 3). I think part of the problem here is one of terminology. The terms duplication, duplicate publication, redundancy, etc. have had a longer, more visible historical presence in the biomedical literature than the term self-plagiarism. I suspect that the growth in publications on these and related matters probably began, slowly at first, in the 1980s and coincided with congressional hearings on research misconduct that eventually gave rise to ORI (2)
Reference
1. Callaham, J. L. (2014). Creation of a Moral Panic? Self-plagiarism in the academy.
Human Resource Development Review, 13, 3-10.
2. Historial Background, http://ori.hhs.gov/historical-background.
Many institutions and professors do indeed let students use portions of prior work in later courses–they just expect students to ask permission first.
Indeed, but when the student does not indicate the source, then the student is done for plagiarism, because they are passing off someone else’s work as their own. So are there two rules or only one?
professors do this as well – they don’t indicate the source i their lectures either.
Reblogged this on This Got My Attention and commented:
Oops! When the chancellor isn’t stifling free speech and promoting her version of “civility” she’s correcting the record about her research.
http://www.dailyillini.com/news/article_cf55ebce-2f97-11e4-b672-0017a43b2370.html
This was the recent case where Salaita was officially hired and even given a class schedule only to have the offer revoked by Wise due to pressure from donors?
It seems she is willing to do what she must to remain in good standing, including being in favor of ‘recycling.’
I had heard that both Salaita and his wife had left their jobs in anticipation of the move and were thus left without healthcare and employment as a result of the chancellor’s sudden decision. Regardless of whether one agreed with the tweets that the board was contesting, IMO, it still seemed like an unfair deal.
I may be mistaken, but I heard a few academics had even led several boycotts against the institution as a result? It’s interesting timing that Wise’s oversight should surface now.
Here’s an article from the LA times that also has more info, though it may be a bit biased depending on whose side one agrees with. Although, most articles on the topic are if you google the affair:
http://www.latimes.com/business/hiltzik/la-fi-mh-the-salaita-case-20140915-column.html
One needs to remember that chancellors, such as Wyse, are academics turned politicians. I wish we could expect more of them. Thanks for commenting!
A corrigendum, not a retraction? Even if the paper is admitted not to be a self-plagiarism, by none less than the author herself? Passing it off as a review of one’s own work, since when does something like that exist? What were the Neuroscience editors thinking? Or is just the case of getting high enough (chancellor of University) to rise above rules of scientific decency?
She became chancellor in 2011. This article, published in 2006, was probably accepted in 2005. At that time, she was a Dean of a college. So the notion that the editors would be swayed by a “lofty position” is not accurate. I also think that the notion of “scientific decency” is subjective; what do you mean by it?
Nick, the allegations came up recently, as seen on PubPeer. Thus, when Neuroscience was dealing with it, Wise was already Chancellor for 3 years, also thanks to her “creative” approach to science publishing. As for scientific decency, see my comments below.
While I am not defending Dr. Wise, I will note that the current emphasis on NO self-plagarism is new. I for one never heard the term before about 2009. Thus, while the notion of self-plagarism is current, the careful monitoring of self-plagarism in 2006 was not occurring. This is retrospective use of current standards in past behavior. Retract it, possibly, but it’s important not to apply current standards to past behavior.
I am quite sure that re-publishing one’s earlier data elsewhere and pretending this is novel and original research was never ever accepted.
But that is not what is alleged here. This is not duplicate publication. This is substantial reuse. The two are not the same in any sense.
It doesn’t matter if it is a 1.1 copy from a single previous paper or a pot-pourri of data from several earlier papers, like here. It is still a clear case of misconduct, now and back then. Actually the second case is much more insidious, as it is difficult to uncover and to deal with (i.e, retract).
First of all, nothing is being “alleged”. The facts are clear.
Second, can you elaborate on the difference between duplicate publication and “substantial reuse” without citation of the work being reused?
Indeed, Leonid. in terms of its detrimental impact on the scientific record, I don’t really see a difference between data fabrication and passing off as new data that have already been published, regardless of what one might call the latter (e.g., duplication, self-plagiarism). The recycled data which, again, are being presented as new and, therefore, independent from the old data, simply do not exist.
I did a quick search for “self-plagarism”. There is a 2006 discussion shown at ORI. I found another reference (which I can’t find now) which indicates that the APA style manual V 5 has no discussion of “self-plagarism”, but there are references in V 6. So, sometime in this past decade, the notion was gaining credence, and it is now a clear issue.
The 2006 “discussion at ORI” of self-plagiarism – http://ori.hhs.gov/images/ddblock/plagiarism.pdf – was written, not by ORI staff, but by Dr. Miguel Roig.
In the 1993 proceedings of the ORI / AAAS conference on plagiarism – http://ori.hhs.gov/documents/aaas.pdf and ….aaas2 and aaas3 – which I created Edward Huth talked about self-plagiarism, and mentioned that I had made clear in my remarks – for ORI there is, by definition, no such thing as “self-plagiarism.” “Plagiarism” involves only the use of someone else’s words or ideas without giving appropriate credit to them. That is only duplicate publication or reuse one’s own work, whether cited or not (that is not “plagiarism” by others).
I disagree. Republishing already-published material has been considered unethical at least since I’ve been publishing (1981). I don’t much like the term ‘self-plagiarism’, but it’s a no-no.
Is this misconduct under the HHS definition? That’s why I raised the question of the wording of this ” she appears to have tried to pass off as having been previously unpublished.” Was it knowing? Intention?
Theresa, I think it depends. According to Dahlberg (1), “ORI often receives allegations of plagiarism that involve efforts by scientists to publish the same data in more than one journal article. Assuming that the duplicated figures represent the same experiment and are labeled the same in both cases (if not, possible falsification of data makes the allegation significantly more serious), this so-called ‘selfplagiarism’ does not meet the PHS research misconduct standard” (p. 4).
Reference
1. Dahlberg, J. (2007). ORI Retains Its Working Definition of Plagiarism under New Regulation. ORI Newsletter 15, 4.
Well, yes, an investigation would reveal certain facts.
Former ORI Director David Wright was proposing a new definition prior to his departure.
http://www.youtube.com/watch?v=4LEN9W-zvsk
Here is a question on double-publication or self-plagiarism or whatever we chose to call it. I heard stories of German doctors publishing a paper first in an international English language journal, and then again in a different, German language medical magazine. Thus, one could easily boost one’s publication list and no anti-plagiarism software will ever detect this. How common is this phenomenon of publishing in different languages actually?
Also I hear the same about Chinese researchers. Many persons can read German, but the number who can read Chinese is much smaller. I would imagine the problem to be larger there.
“Many persons can read German, but the number who can read Chinese is much smaller.”
Really?
I suspect that anyone who was actually in a position of needing to review such a CV would notice it, though.
One thing that I haven’t kept up on is the state of journals that are explicitly in translation. JETP is still around,* but there used to be substantial portfolios devoted to just this task. I’m inclined to wonder whether the practice that you describe is closer to filling a gap that has long been well recognized. As I recall, 25 years ago, reading proficiency in one of Russian, German, or French was still a prerequisite for a Ph.D. in the physical sciences.
*One test of whether a journal’s production office is really paying attention, rather than going through the motions – something that I suspect the Scholarly Kitchen glosses right over for a reason – is whether these are conformed in the references.
^ That is, a prerequisite in the U.S. Apologies for the latent imperialism.
Criticize Israel at home on your personal social media and they toss you out on the street.
Republish articles and they make you chancellor.
Sort of makes sense for the university representing the most corrupt state (statistically) in the country.
“I suspect that anyone who was actually in a position of needing to review such a CV would notice it, though.”
Hi Narad, in medicine, the number of papers authored and co-authored by a single doctor can easily go into many dozens and hundreds. It is unlikely someone is going to check and read, only count.
I agree that it is fairly common and we only see the tip of the iceberg (if we bother at all) as not all of these non-english journals are indexed in pubmed. In my experience this “habit” effects mostly case reports, reviews, and papers with clinical relevance. Original research articles are less affected, partially as these would be easier to track down, but mainly because it’s simply not worth that much effort.
That wasn’t the only time that she apparently self-plagiarized. See:https://pubpeer.com/publications/2EE674599778410770DFA3BA3A02F8
So Phyllis Wise, the chancellor of the University of Illinois, told RW that she agrees with the correction. The correction states:
(1): “The paper is written in a way that misleads the readers to think that it is an original article.” It seems to me that “misleading” is the crucial term in this sentence. It seems to me that such a behaviour (“misleading”) is not allowed for students of the University of Illinois. Anyone any idea what would happen with a student of UoI with such a behaviour?
(2): “The author regrets the extent to which the text in Wise (2006) was recycled from the prior papers.” It seems to me that “recycling texts from pior papers” is not allowed for students of the UoI. Anyone any idea what would happen with a student at UoI with such a behaviour?
Below are two very similare cases. Both highly cited papers, but only with a minor corrigendum. And still, the duplicated paper continued being cited even after corrigendum publication. Have a look at those:
https://pubpeer.com/publications/EDB78A82EFD2364AFED7492DED25B8
https://pubpeer.com/publications/B5FC3A74A47B0EE9DF724B9407F413
She used pictures and illustrations from the previous publication and labeled them as “source unknown”. I think this is a clear indication that she knew what she was doing was wrong and purposely tried to hide the original source.
The 2006 paper states in the Experimental Methods section, “Detailed methods are described in our original peer-reviewed papers (Dubal et al., 1998, 1999) and are only described briefly below.” This kind of thing happens when a journal asks for a special “summary of one’s own work” review. But the line between a review and a recycling, duplicate publication is blurred. Normally one has to sign a statement saying the work has not been published elsewhere.