In April of this year, Corneel Vandelanotte realized something had gone wrong with a paper he had recently published.
First, there was a post about his paper by Nick Brown, a scientific sleuth, questioning the results, ethics, and authors behind the work. That was followed by a comment on PubPeer by Elisabeth Bik, another scientific sleuth.
“People started alerting me,” Vandelanotte, a public health researcher at Central Queensland University in Rockhampton, told Retraction Watch. “Hey, have you seen this blog by Nick Brown? And, and then yeah, okay, that was a bad day. Let me put it that way.”
Vandelanotte grew concerned. He asked the lead author on the paper to see the data. When the lead author refused to share them, saying they were inaccessible, Vandelanotte became convinced: He had been deceived.
The episode would lead to a retraction — and a lesson for Vandelanotte about being fooled by would-be collaborators.
In March 2019, Vandelanotte received an email asking for help revising an article from Faustin Etindele Sosso, a researcher who said he was studying mental health trends in African gamers. Vandelanotte often advises early-career researchers, he said, and was willing to help Sosso — someone who he assumed based on their exchanges might not have the resources or the language proficiency to get his paper published alone.
Over the course of the next several months, Vandelanotte looked at three versions of the paper, providing general recommendations and writing advice. He said the original manuscript was poorly written. He also said some of the analyses he’d chosen didn’t make sense to use, so he recommended redoing them. But never asked to see the data.
At one point, Vandelanotte said, he recommended contacting a statistician at his own university, who could help with the analysis. But Vandelanotte said Sosso ignored his suggestion and continued to ask for text edits.
On one version, Vandelanotte was listed as third author. He says he was surprised that there were 16 total authors, none of whom — with the exception Sasso — he’d ever had any correspondence with.
Before that it had been, according to Vandelanotte, “just me and him going back and forth.”
Vandelanotte said he started to wonder at this point why, with so many other authors, his help was necessary at all. He told Sosso to add an author contribution statement to break down the roles of the many authors.
Vandelanotte said the last version he saw “was the type of analysis that I would have expected to see, and the outcomes were magnificent.”
He didn’t see the final draft sent to peer review, but the peer review process was surprisingly easy, he said: “They were the smoothest easiest reviews ever.”
The study, “Insomnia, sleepiness, anxiety and depression among different types of gamers in African countries” was published in Scientific Reports on February 6, 2020.
Vandelanotte heard in mid-April that a critique of the article was in the works and the authors were given the opportunity to respond with a letter. This was the first time Sosso shared any of the data–creating a public link to access part of it as part of his effort to respond to the critique. But it wasn’t the whole data set.
Vandelanotte contacted Scientific Reports. The journal began cc’ing all authors and they went back and forth until the editor updated the piece with the following editor’s note: “Readers are alerted that the results of this paper are subject to criticisms that are being considered by editors. A further editorial response will follow the resolution of these issues when all parties have been given a chance to respond.”
Around the same time, Bik’s post appeared. She contacted Vandelanotte and recommended he try to get his hands on the data.
By this point, Vandelanotte was worried that Sosso might have misrepresented himself. (Sosso did not respond directly to requests for comment from Retraction Watch, instead consulting with a few of his co-authors about how to respond, according to emails reviewed by Retraction Watch.)
Sosso’s author bio lists an affiliation with the Center for Advanced Research in Sleep Medicine (CEAMS in French) at the University of Montreal. A press officer for the university told Retraction Watch that Sosso “is pursuing his doctorate under the supervision of a researcher who is working at the CEAMS,” but that the paper “was written prior to his doctorate.” The press officer declined to name the research adviser.
Vandelanotte also wondered whether Sosso had recommended fake peer reviewers, but a spokesperson for Scientific Reports denied that claim. “Our priority as a journal is to maintain the scientific record based on the validity of the research published,” the spokesperson wrote. “An independent post-publication peer review confirmed that there were numerous anomalies in the data which undermined the results and conclusions, and for this reason the editors concluded that a retraction was warranted.”
He continued: “The peer review process for any paper is confidential, but we can confirm that the peer reviewers were independently selected by the Editorial Board Member handling the paper and were not suggested by the authors.”
The study was retracted on June 4. The retraction notice described the flawed mathematical analyses but didn’t mention the allegations of false authors or data fabrication: “After publication serious concerns were raised that the results described in the Article, in particular the outcomes of statistical analyses, are internally inconsistent and that some results are mathematically impossible. Independent post-publication peer review has confirmed that there are numerous anomalies in the data which undermine the results and conclusions.”
Brown said the notice doesn’t tell the whole story. He found evidence of copy-pasting and statistically unbelievable data.
Vandelanotte wasn’t the only one fooled by Sosso. Ejaz Husain, of the Centre for Physiotherapy and Rehabilitation Sciences in New Delhi, shared an email from Sosso in which Sosso described himself as a Canadian researcher who, since 2015, had collected data from more than 2000 Canadian students and more than 100,000 participants in African countries.
“It appears that we’re all in the same boat, we’re all approached out of the blue, smooth-talked into being an author on this paper,” Vandelanotte said.
Sosso asked for Husain’s help, and Husain agreed. Husain wrote in an email to Retraction Watch that his role, along with two other authors, Pooja Bhati and Anam Aseem, was only to improve the writing in the article. He said he knew nothing of the other authors, reviewers or the manner in which data was obtained.
Daria Kuss, second author on the study, wrote in email that she had a similar experience — responding to Sosso’s request for help revising his article, and claiming she had not seen the data until after concerns were raised. However, her name was on the preprint of the article, posted in 2018. “The lead author has also taken sole responsibility for the data collection and analysis in an email sent to the co-authors on 1/5/2020,” said Kuss, a researcher at Nottingham Trent University in the UK.
Brown also looked at other papers Sosso had co-authored, and based on his research, thinks that several of the more suspicious names on those papers are fakes.
Brown told Retraction Watch that he contacted all the integrity offices he could find at the author’s given institutional affiliations, including Vandelanotte’s. He says only two acknowledged his emails.
A Central Queensland University investigation into the episode concluded that Vandelanotte’s “actions have been in breach of the Australian Code for the Responsible Conduct of Research,” according to a June 8 letter signed by Grant Stanley, the deputy vice chancellor of research. Vandelanotte, the letter said, “did not take reasonable steps to ensure the accuracy and integrity of the contributions of your co-authors, or take reasonable steps to ensure that you could be confident in the accuracy and integrity of the research before the paper was submitted for publication.”
While the university’s “assessment has found no evidence that this breach was deliberate or malicious…or that it should be investigated as research misconduct or serious misconduct,” Vandelanotte is required to “review the Code and prepare a presentation on authorship responsibilities and develop template materials and tip sheets about academic authorship.”
Brown called the article “an absolute microcosm of all of the problems in scientific publishing.”
He said that based on the edits he has seen, he thinks Vandelanotte was genuinely trying to improve the manuscript, but thinks he should have asked to see the data. “I feel some sympathy for the guy ‘cause he’s at least worked out what is going on and he probably feels pretty bad about it,” Brown said.
And feel bad about it Vandelanotte does. “The bottom-line of my story is this: an experienced fraud tricked me into being on one of his papers, abusing my goodwill, trust and reputation in the process,” Vandelanotte wrote to Retraction Watch “All I wanted to do was offer help to someone I thought was in need of it. I am paying a pretty hefty price for it now. I usually don’t fall for scams, and pay good attention…but they got me this time.”
“Please be aware this is how I got fooled,” he wrote. “Learn from my story and avoid trouble yourself.”
Update, 1600 UTC, 6/12/20: Paragraph beginning “Brown also looked at other papers…” corrected to note that Brown was referring to other work by Sosso, not the Scientific Reports article.
Like Retraction Watch? You can make a tax-deductible contribution to support our work, follow us on Twitter, like us on Facebook, add us to your RSS reader, or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at team@retractionwatch.com.
Fascinating, and remarkably quick and appropriate actions by publisher and at least one institution. But to the ‘data thugs’ if you’re reading, prey tell (ha), of the daily deluge of published articles, what was about this title that caught your eye for a deeper look?
It wasn’t the title that caught my eye. A friend (A) had a colleague (B) who had read the paper and not been able to make head or tail of it. So A put B in touch with me, and we went from there. It only took about two minutes with Table 3 to realise that the whole paper was completely borked.
The guy should have aimed higher. If he had claimed to be founder of a company making internationally-used software for monitoring gamer mental health, and to have access to a huge database of anonymised case-studies through that company, he could have got away with it for longer.
I just can’t see collaborating on a paper with someone who contacts you out of the blue and who you don’t know at all without trying to find out a lot more about them. Didn’t Vandelanotte wonder about what was going on?
I am with you on this one. Especially if it is an extremely poor manuscript to start with, as he states it was. To me this smacks of extreme professional negligence, by both Vandelanotte and Kuss.
Exactly.
I was wondering that too.
Of all the authors on the manuscript, it is the black man who is made the villain here, while the white and European authors quickly turn themselves into victims, and we all have to believe them that they really did not know that the manuscript was faulty. They are a bunch of opportunists who were too eager to put their names on the manuscript of a colored scholat and once caught red handed they know how to manouvre around in this system. Hence, it is all the black man’s fault. Retraction watch usually puts a mugshot of the (usually colored) perpetrator on these kinds of blogs, but I guess they did not have any so they put a photo of the white “victim”. I have a feeling that this comment wont be published, but we colored scholars are fed up, and Retraction Watch can soon find an analysis of their own blogs online.
Hi Jonathan,
I am a “coloured” person too.
So are the doctors who wrote the now-retracted NEJM & Lancet COVID-19 papers based on what looks like fake data.
Dishonest folk come in all colours and hues!
I feel this researcher (Corneel Vandelanotte) just did what he is used to doing- ie putting his name in other peoples’ paper.
It is just that he got caught this time.
Hope I did not offend you. Tq
So do you have proof that Vandelanotte “did what he is used to doing- ie putting his name in other people’s paper” that you would be willing to share with all of us or will you be retracting your potentially libelous statement as false. Thanks in advance.
Retraction watch (RW) needs to be more balanced in their writing. Based on this article, it is difficult to understand how anyone was ‘fooled by Sosso’. With the many faults outlined for this paper, I find it highly unlikely that blame lies only with one person. It seems researchers had the opportunity to put their names on a paper and did so. I fail to see how anyone can claim ‘they got me this time’ in this situation. If an author reviewed the manuscript at any time, then they had access to the same published data that Brown used in the analysis posted on his blog. At a minimum, if any other author made a substantial contribution to the writing to warrant authorship then one would not expect the outcome of the writing comparison shown on Brown’s blog. Why would a researcher feel confident about keeping their name on a submission where the analysis chosen ‘didn’t make sense’ and recommendations for correction were ‘ignored’? The clam is not seeing ‘the data until after concerns were raised’, why would you confirm authorship on a submitted manuscript and not at least review the contents? How can researchers make the excuse of (and RW published this excuse) ‘only to improve the writing in the article’ when according to Brown’s blog the text of the ‘poorly written (as described in this piece)’ manuscript hardly changed? RW does good work, but this piece is extremely unbalanced (to say the least).
I agree that the co-authors are not innocent victims here. Choosing to accept co-authorship with someone you don’t know, for a study in which you have no access to the data or can even verify the existence of the data, is a serious breach of ethics of authorship. However, the reason someone like Nick Brown (a premier data thug) is able to notice the errors in the published data alone is largely because he is a trained statistician with lots of experience in “sniffing out” problematic data. Someone with less (or no) expertise in statistics could very likely not realize that the analytic findings didn’t make sense.
That said, I completely agree with your larger point. None of the co-authors are blameless.
Hi – Nick Brown here. A couple of points in response to the above (apologies for the delay):
1. @Sophia “If an author reviewed the manuscript at any time, then they had access to the same published data that Brown used in the analysis posted on his blog.” From the correspondence that I have seen between the authors, none of them saw the data until April of this year. (One possible explanation of that would be if the data did not exist before April of this year, and were hastily assembled by the lead author in an attempt to cover his tracks.) Of course, that calls into question the statement in the “Author contributions” of the paper that four authors contributed to the analyses.
@Aaron: “However, the reason someone like Nick Brown … is able to notice the errors in the published data alone is largely because he is a trained statistician with lots of experience in “sniffing out” problematic data. Someone with less (or no) expertise in statistics could very likely not realize that the analytic findings didn’t make sense.” — I think that this somewhat overstates my qualifications and overestimates the difficulties in this case. I am not an especially well-trained scientist and I definitely do not identify as a statistician. But any social science researcher with more than a single graduate-level statistics course under their belt — which is arguably at, or below, the absolute minimum level of knowledge required to be a peer reviewer for any journal, let alone a “prestigious” one like Scientific Reports — ought to be able to see from the Abstract alone of the article that the results are highly implausible. And anyone who has ever written up the results of a linear regression analysis — a standard thing to do for most Masters students in the social and behavioural sciences — ought to have spotted that *every single line* of Table 3 contained multiple impossible numbers. It was the equivalent of reading last night’s soccer results and finding that every team scored at least 30 goals.
Maybe it’s because I’m a new scholar and I work in the humanities, but I don’t understand why the co-authors were co-authors in the first place. If this article is correct, Vandelanotte helped edit and proofread an article. They didn’t do the research or write up the results. I edit and proofread all the time but I would never expect my name to appear as a co-author. Is this normal?
It is “normal” in that undeserved authorship happens all the time in the sciences, but it is not appropriate and not how authorship is supposed to work. That said, it’s not entirely clear whether the improvements added by Vandelanotte were merely copy-edits, or if he also provided feedback that impacted the intellectual content of the paper. In the latter case, depending on how substantial those changes were, authorship could be appropriate. But for merely copy-edits, definitely not appropriate.
Like many here, I am weary about how lightly co-authors are getting off with it. Clearly, their culpability is not as severe as the lead author’s. But they still strike me as being guilty of something akin to lazyness: an easy authorship on a paper with minimal contribution. When I read “Vandelanotte often advises early-career researchers, he said, and was willing to help Sosso”, I can’t help but be a bit cynical about it. You provide a few comments and edits, and you don’t blink when you end up being a co-author? Really?
Appears like Sosso is at it again, ”fooling” two new co-authors:
”Influence of socioeconomic status on objective sleep measurement: A systematic review and meta-analysis of actigraphy studies”
https://www.sciencedirect.com/science/article/abs/pii/S2352721821001121#!
What is the likelihood of data fabrication by Sosso in this case (given his precedents)? I guess there is no mechanism to prevent this from happening again and again.
Well, it’d be nice if the journal at least checked his affiliation. I can’t seem to find the Redavi Institute. Can you?
The senior author tells us:
Good point about this being a review, but I wouldn’t want to have my name associated with him for sure. At this point, this guy is a serial offender who has published >35 papers (mostly low impact) with a mix of fabricated data, fake authors and/or fake affiliations. At the very least, a simple Google search on Sosso would have been warranted before agreeing to co-author an article.
How can Faustin Armel Etindele Sosso be stopped ?
Three of those previously published (and therefore reliable) papers were by Sosso.
This seems consistent with his self-citation pattern…
Also of interest here: https://steamtraen.blogspot.com/2020/04/the-mystery-of-missing-authors.html
In which Nick Brown surveys a circle of mysterious untraceable authors who churn out papers through parasitical journals… papers that exist only to cite the magisterial studies of Faustin Etindele Sosso.