Last summer, I covered the saga of Francesca Gino of Harvard Business School, who was blatantly accused of fabricating data in at least four of her published studies. She was caught when some Internet data sleuths — who investigate research misconduct in their spare time — I found inconsistencies In its paper data and investigated further.
They eventually raised their concerns with Harvard University, which eventually investigated the matter Request to withdraw the relevant papers. (Gino foot Harvard sued and the bloggers, accusing them of colluding to smear its reputation.)
I kept thinking about Gino's case as I read Oddly similar story Scandal at Harvard-affiliated Dana-Farber Cancer Institute, a leading cancer research hospital in Boston.
Dana Farber was shocked in January of this year by A Blog post Written by Sholto David, a molecular biologist and internet data sleuth, in which he presents evidence of widespread data manipulation in cancer research published by leading researchers including the institute's CEO and COO. David It said She contacted the institute over concerns about 57 research papers, 38 of which the institute had “primary responsibility for potential errors in the data.” The institute has requested the withdrawal of 6 of them and has begun corrections for 31 of them.
To be clear, these data manipulations were not subtle. (David's somewhat flowery post announcing the evidence called it “pathetically amateurish and excessive.”) Many of the cases he identified involved reusing the same images over and over again in different shapes, with different labels, and with the shapes rotated or rotated clumsily . Extended in Photoshop or similar image editor. The data collection plots on different days are vaguely identical. Test results are clearly copied and pasted.
It raises the question: Assuming there was some misconduct behind the copied and pasted images, how dare people commit such blatant fraud, so openly, for so long? How much grant money has been secured on the basis of fabricated data, and how much has the crucial fight against cancer been undermined by published inaccuracies in this research?
And perhaps more importantly, is this just the tip of the iceberg?
Anatomy of a cancer data scandal
For many years, biomedical researchers have been aware that the field faces the problem of fake images in papers. In a 2016 paper, Dutch microbiologist Elisabeth Beck wrote Scanned More than 20,000 biomedical papers looked for evidence of such manipulation, and found that 3.8% of the papers had signs of it, “with at least half showing features suggestive of intentional manipulation.” Worse still, the problem appears to be on the rise. “The prevalence of research papers containing problematic images has risen significantly over the past decade,” Beck found.
Ha A scale to describe manipulation It examines three types of fake photos – cases where the same photo is used twice, with different captions (which could be an innocent mistake), and cases where the same photo is used twice but in one instance it has been intentionally cropped (which seems less likely to be a fake photo). It is an innocent mistake (an innocent mistake), and cases where the image has something else pasted onto it (which seems unlikely to be an innocent mistake).
So, biomedical scientists were well aware that this field had a problem. Some of the specific manipulations highlighted in David's blog post were well known among scientists A hot topic of discussion on the PubPeer discussion forum. But although the concerns were well known, it seems it took David's position Quick withdrawal and internal investigation.
Mistakes have consequences
It is troubling that cases like Gino and Dana Farber's required external data checks to emerge. Being a data detective is extremely unrewarding, and even risky. David Currently unemployed and doing data processing reporting work In his spare time between gigs, he told The Guardian.
Many data investigators have been threatened with lawsuits to expose data fraud. “A lot of important science is done not by big institutions that question things, but by independent people like that,” defamation lawyer Ken White told me last summer. The problem is that there is no institutional process for reviewing research unless someone else highlights the problems, and most scientists do not want to jeopardize their careers to do this frustrating and not-so-beautiful work.
It is also worrying that the counterfeiting was so blatant. We're not talking about complex data processing here, but rather about cases where scientists have badly edited images of their experimental results. “We are only seeing the small tip of the fraudulent iceberg – image data duplication, the last resort of a failed world after all other tricks have failed to deliver the desired result,” David wrote in his original blog post. In a culture where the experimental results of Photoshop are repeated, this is unlikely to be the only form of manipulation.
There is another common thread between Gino's failure and Dana-Farber's failure: Harvard. Between the Gino case, the resignation of Harvard President Claudine Guy, and now allegedly bogus cancer research, Harvard's reputation for academic excellence has undoubtedly taken a major hit.
But the discovery of these challenges at America's most prestigious university also helped draw public attention to an issue that desperately needs attention. Maybe embarrassing Harvard will bring about change.
A version of this story originally appeared on The future is perfect the news. Register here!
“Infuriatingly humble alcohol fanatic. Unapologetic beer practitioner. Analyst.”