Category: disputed data

Kaggle removes problematic stroke dataset for copyright infringement 

Images from the now-deleted dataset on Kaggle purportedly showed people who had a stroke, but the folder included stock images and photos of politicians and celebrities, as seen in this screenshot.

The online data repository Kaggle has removed a problematic dataset for violating intellectual property rights months after it was flagged by sleuths for containing celebrity photos and lacking information on data provenance or confirmed medical diagnoses. 

Researchers had used the images to train machine learning algorithms to detect stroke, but the dataset instead contained images of people with Bell’s palsy — as well as pictures of celebrities from movies and on the red carpet, as we reported in May. Adrian Barnett, a medical statistician at Queensland University of Technology in Australia, flagged the dataset after discovering two others he and his colleagues investigated in a February preprint that showed signs of being fabricated. 

Using a reverse image search, Barnett matched four of the photos in the dataset, called “Facial droop and facial paralysis image,” to a paper published years earlier in the Journal of the Canadian Dental Association (JCDA). Only after prompting the journal to complain to Kaggle did the company take down the dataset. 

Continue reading Kaggle removes problematic stroke dataset for copyright infringement 

Sex pay ban paper earns a retraction after a long and winding road for an unhappy author

In March 2024, Riccardo Ciacci, an economist at the Universidad Pontificia Comillas in Spain, published a paper claiming Sweden’s ban on buying sex had increased reported rapes by as much as 62%. The finding gained attention on social media, and quickly drew criticism from others in the field. 

In particular, a group of three economists took their concerns to social media and to the journal editors, and eventually published a critique of Ciacci’s work. They claimed his analysis reports a statistical relationship not relevant to the finding described in the paper. They concluded there was no large or statistically significant finding. 

What followed was a year-long effort to fix the paper, and then ultimately, a decision to retract it. Ciacci, who was not accused of misconduct, said the retraction, which he disagrees with, has cost him a promotion and funding for future research. He also alleges he experienced an onslaught of harassment on social media. In the end, Ciacci maintains the retraction was unjustified, and critics say it came far too late. 

Continue reading Sex pay ban paper earns a retraction after a long and winding road for an unhappy author

How the media hypes “research that is absurd on its face”

Aaron Brown says his new book, Wrong Number: How to Extract Truth from a Blizzard of Quantitative Disinformation, “isn’t an exposé of fraud—Retraction Watch covers that ground. It’s about legitimate-looking research that is absurd on its face.” 

Published this month by Wiley, Brown uses dozens of case studies to show “why widely reported and influential studies in top journals are not just wrong, but obviously and egregiously illogical or contrary to simple fact. My focus is less on the policy and statistical errors than on why no one seems to care,” he says.

Brown is a risk manager working in hedge fund management. He also teaches statistics at New York University and the University of California San Diego and writes columns for Reason and Bloomberg, among other outlets. We asked him to tell us more about how he thinks about the nexus of science, journalism and the publish-or-perish system that also pushes researchers to engage with non-experts to promote their work.

Continue reading How the media hypes “research that is absurd on its face”

Lawsuit fails to block retraction of paper claiming to link heart-related deaths to COVID-19 vaccines

Greg J. Marchand in a photo from his research institute’s website.

A Taylor & Francis journal has retracted a widely-read paper linking cardiac-related mortality to COVID-19 vaccines after an unsuccessful legal attempt by the lead author to block the withdrawal. That author says he is considering further legal action against the publisher.

The article, “Risk of all-cause and cardiac-related mortality after vaccination against COVID-19: A meta-analysis of self-controlled case series studies,” drew swift criticism when it was published in Human Vaccines & Immunotherapeutics in August 2023. At the time, critics and sleuths were quick to challenge the data and methods used in the paper, which now has more than 143,000 views on the Taylor & Francis website and has been cited 15 times, including by two letters to the editor of the journal and a response from the authors, according to Clarivate’s Web of Science. 

The retraction notice, posted online January 16, states the retraction resulted from concerns that arose about the methodology of the study and the integrity and availability of the data. The authors provided a full response to the queries; however, the publisher determined the validity of the findings remained in question, the notice states. It continues:

Continue reading Lawsuit fails to block retraction of paper claiming to link heart-related deaths to COVID-19 vaccines

Flawed climate science paper “exposed potential weaknesses” in the peer review process

Before we present this new post, a question: Do you enjoy reading Retraction Watch? If so, we could really use your help. Would you consider a tax-deductible donation to support our work? Thanks in advance.

How did a deeply flawed paper, which contradicts mainstream science on climate change, pass peer review?

That is what three editorial board members tried to figure out after the journal, Global and Planetary Change, faced heavy criticism for publishing the controversial paper last year. The board members published their findings earlier this month in a commentary.

Martin Grosjean, the corresponding author on the editorial, told Retraction Watch that the editors and publisher, Elsevier, share the same interest: Continue reading Flawed climate science paper “exposed potential weaknesses” in the peer review process

Journal silently fixes emergency care paper — after misleading press coverage

Last month, a colleague of emergency medicine doctor Daniel Waxman sent him some newly reported findings that took him by surprise. Waxman knew from the title of a press release about the recent paper — “Nearly Half of U.S. Medical Care Comes From Emergency Rooms” — that something was wrong.

Immediately I said, that’s not true. It’s just crazy.

Waxman quickly realized the mistake: The data were based only on care provided in hospitals — much of which, not surprisingly, originates from emergency departments (EDs). But the title of the paper, the abstract, and other places in the text do not specify that. What’s more,  the press release about the study says the findings relate to “all medical care.” The journal has since changed the paper, including the title, to make that distinction clear, but not provided any editorial notice indicating the text had been updated. Meanwhile, the press release and news stories about the original study continue to report the “surprising” original findings.

Continue reading Journal silently fixes emergency care paper — after misleading press coverage

A journal printed a sharp critique of a paper it had published. If only it had checked with the authors first.

In 2016, researchers published a paper showing that an RNA molecule may be overactive in breast tumor tissue. But after reading the paper, three biologists believed the data supported the opposite conclusion.

What happened after that is a tale of misunderstandings and unnecessarily bruised feelings. We’ve seen plenty of cases where researchers ignore criticism, which at first glance seemed to be the case here. But upon closer inspection, it wasn’t. Continue reading A journal printed a sharp critique of a paper it had published. If only it had checked with the authors first.

Board member resigns from journal over handling of paper accused of plagiarism

A biologist is crying foul at a journal’s decision to correct (and not retract) a paper he claims plagiarized his work — and one of his colleagues has resigned from the journal’s editorial board as a result.

The 2016 paper, published by Scientific Reports, is an application of a previously published algorithm designed to better identify regulatory sequences in DNA. The three authors, based at the Shenzhen campus of the Harbin Institute of Technology, used the technique to identify recombination spots in DNA. They called it SVM-gkm.

On April 2, Michael Beer of Johns Hopkins University in Baltimore notified the editors of Scientific Reports that he believed the paper had plagiarized his work. Despite Beer’s efforts, the journal ultimately decided to issue a correction notice, which cites “errors” and the authors’ failure to credit Beer’s work. That isn’t good enough for Beer — nor one of his colleagues at Johns Hopkins, who resigned from the journal’s editorial board saying “the recent affair with Mike Beer’s work being plagiarized did not impress me.”

Continue reading Board member resigns from journal over handling of paper accused of plagiarism

Journal republishes withdrawn paper on emergency care prices, amid controversy

The Annals of Emergency Medicine has republished a controversial paper it withdrew earlier this year which compared the cost of emergency care at different types of facilities.

Because the paper drew heavy criticism when it was originally released, the journal has published a revised version, along with several editorials and discussions between the authors and critics. One point of contention: The analysis stems from data provided by an insurance company — Blue Cross Blue Shield — which it declined to share.

The paper — originally published in February —  caught national attention (and raised concerns among some emergency care providers) when it reported the cost of treatment in emergency departments can be significantly higher than at urgent care centers, even for the same conditions. The journal withdrew the paper in spring, and re-published it Tuesday, with minor changes.

First author Vivian Ho at Rice University told us she made “slight changes”  to some headings, phrases, and the appendix, but:

Continue reading Journal republishes withdrawn paper on emergency care prices, amid controversy

Released FDA docs reveal details of agency’s (failed) attempt to retract paper

Earlier this year, a raging controversy regarding a new drug spilled into the pages of a leading medical journal: the head of the U.S. Food and Drug Administration and another official publicly called for the retraction or correction of a peer-reviewed article about the drug. They didn’t get their wish. Now, documents released by the FDA via a lawsuit shed light on the attempt — and show how tricky it can be to correct the official record.

The controversy surrounds the approval of eteplirsen, a drug approved last September to treat Duchenne Muscular Dystrophy, a rare but invariably fatal disease that strikes (mostly) young boys. Eteplirsen was approved over the objections of the FDA team that reviewed the drug, which determined that there was insufficient evidence to approve the drug.

But the controversy didn’t end within the walls of the FDA complex. Ellis Unger, who led the review team, believed that one of the principal studies of the drug, published in the Annals of Neurology, was “misleading” because it was based on “unreliable data.” So in early November, Unger, joined by the then-head of the FDA, Robert Califf, took the extremely rare move of writing to the editor of the journal to “urge that the paper be corrected or retracted….”

According to the documents, the authors of the Annals article did, in fact, agree to correct the article.

On November 9, 2016, Clifford Saper, Annals‘ editor in chief, wrote to Califf and Unger:

Continue reading Released FDA docs reveal details of agency’s (failed) attempt to retract paper