Retraction Watch

Guest post: Genomics has a spreadsheet problem

Surveys show spreadsheets are the most widely used analytical tool in academic research. But they are prone to errors. During the height of the COVID-19 pandemic, Public Health England lost 16,000 test results after using an old Excel file format to handle data. Whether this mix-up hampered local infection control is anybody’s guess, but it certainly could have.

The software is causing trouble for other lines of research, too, as our team and others have shown: Genomics studies that rely on Excel spreadsheets to manage data turn out to be riddled with erroneous gene names, or gene-name errors, and the issue is affecting more and more journals.

Nature pulls study that found climate fears were overblown

It was that rarest of things: a sliver of good news about climate change.

According to calculations published last year in Nature, our planet was keeping pace, and then some, with rising emissions from tropical forest clearance by gobbling up more and more atmospheric carbon.

“What we can mainly prove is that the worst nightmare scenarios of an impaired carbon sink have not yet materialised and that the news is not quite as bad,” Guido van der Werf, a professor at Vrije Universiteit Amsterdam, said in a press release at the time.

His coauthor Dave van Wees added:

Exclusive: Probe suggests new retraction awaiting embattled Korean heart doctor

A prominent physician-scientist in South Korea may soon be facing his fourth retraction. Last month, Hui-Nam Pak of Yonsei University was found guilty of duplicate publication, a form of academic misconduct, according to a report from the school’s committee on research integrity Retraction Watch has obtained.

Pak, a cardiologist, has had dozens of papers flagged on PubPeer. As we reported in February, journals pulled two of his papers the previous month after a whistleblower pointed out problems with the articles. One was retracted for “a number of issues related to scientific misconduct,” while the other was a duplicate publication. A third paper by Pak was retracted years earlier after mistakenly being published twice by the same journal.

Our February story triggered a flood of comments, many of them malicious. Some likened whistleblowing to “academic vandalism.” Others asserted that “whistleblowers deserve strong legal penalties” and that “immoral whistleblowers” seemed bent on ruining Pak’s “outstanding career.” Many comments were rejected for not adhering to our commenting policies, in particular making unsubstantiated claims.

The September 12 report from Yonsei University (in Korean) explains that an “informant” reported two of Pak’s papers to the school’s Research Ethics Integrity Committee on March 7.

Swedish beauty study that sparked ‘storm of criticism’ is cleared

The economist behind a controversial study showing attractive female students got lower grades after classes moved online during the pandemic has been acquitted of research misconduct, according to a report from his former institution.

But the researcher, Adrian Mehic, did not get off without reproof: Even if the work kept to the letter of the law, it may still have had “unethical consequences,” Erik Renström, vice-chancellor of Lund University in Sweden, wrote in the June 8 report (in Swedish).

In the study, a jury made up largely of final-year high schoolers rated the looks of university students based on pictures taken from social media accounts. The ratings were then linked to other publicly available data about the students, including academic performance. The findings, published in the journal Economics Letters in August 2022, made headlines across the globe.

But the students had not consented to the research, nor were they informed about it. The revelation unleashed “a storm of criticism at the university,” according to local media.

Weekend reads: Crossref acquires the Retraction Watch Database; Italy minister’s papers scrutinized; Carlo Croce goes to court yet again

Would you consider a donation to support Weekend Reads, and our daily work?

The week at Retraction Watch featured:

Our list of retracted or withdrawn COVID-19 papers is up to well over 350. There are now nearly 43,000 retractions in The Retraction Watch Database — which is now part of Crossref. The Retraction Watch Hijacked Journal Checker now contains 200 titles. And have you seen our leaderboard of authors with the most retractions lately — or our list of top 10 most highly cited retracted papers?

Here’s what was happening elsewhere (some of these items may be paywalled, metered access, or require free registration to read):

Turmoil at Sage journal as retractions mount

In the midst of a tumultuous year, the journal Concurrent Engineering: Research and Applications, a Sage title, is retracting 21 papers after an investigation identified signs of “compromised” peer review.

Clarivate delisted the journal from its Web of Science index in March for failing to meet editorial quality criteria. Founding editor Biren Prasad, who managed the journal since 1992, also retired earlier this year, and the publisher took over management of peer review.

The journal’s online presence also needs attention: Neither of the associate editors listed on its website have been involved for many years, both told Retraction Watch – and one has threatened to sue the journal if she isn’t removed.

Overturning a dubious retraction proves difficult for education professor

For the past eight years, an education researcher in Spain has been waging an unsuccessful battle – including legal action – to quash a retraction she argues should never have happened.

Her paper, about the use of digital tools in early childhood education, was pulled by Computers & Education just months after it was published in 2015. According to the retraction notice, the article was submitted for publication while it was still under review at another journal, violating editorial policies against duplicate submission.

But according to the researcher, Elena Ramírez Orellana of the University of Salamanca, her paper had already been rejected by the first journal before it was sent to the other. She had the documents to prove it, she said, but that didn’t matter.

The Retraction Watch Database becomes completely open – and RW becomes far more sustainable

Today is a very big day for Retraction Watch and The Center For Scientific Integrity, our parent non-profit. Bear with me while I explain, starting with some history.

When Adam Marcus and I launched Retraction Watch in 2010, we envisioned it as a journalism blog that would break stories no one else was covering, and examine whether scientific correction mechanisms were robust. And for some time, that’s just what it was. Our traffic and visibility grew quite quickly, but the team didn’t. It was years before we even had an intern.

Things changed in 2014 and 2015. Three philanthropies – the MacArthur Foundation, the Arnold Foundation (now Arnold Ventures), and the Helmsley Trust – approached us with some version of “We think what you’re doing is important. How can we help?”

Weekend reads: ChatGPT in papers; a Russia-based paper mill; getting scooped becomes an opportunity

Would you consider a donation to support Weekend Reads, and our daily work?

The week at Retraction Watch featured:

Our list of retracted or withdrawn COVID-19 papers is up to well over 350. There are now well over 42,000 retractions in The Retraction Watch Database — which powers retraction alerts in Edifix, EndNote, LibKey, Papers, and Zotero. The Retraction Watch Hijacked Journal Checker now contains 200 titles. And have you seen our leaderboard of authors with the most retractions lately — or our list of top 10 most highly cited retracted papers?

Here’s what was happening elsewhere (some of these items may be paywalled, metered access, or require free registration to read):

Frontiers retracts nearly 40 papers linked to ‘authorship-for-sale’

The publisher Frontiers has retracted nearly 40 papers across multiple journals linked to “the unethical practice of buying or selling authorship on research papers,” according to a press release posted to a company website Monday.

The release also states Frontiers is adopting new policies to prevent the sale of authorships on papers it publishes.

The publisher’s old policy simply stated that “Requests to modify the author list after submission should be made to the editorial office using the authorship change form.”

Now, such requests “will only be granted under exceptional circumstances and after in-depth assessment by the Frontiers’ research integrity unit,” according to the release. The publisher will also keep track of the requests “to identify suspicious patterns and trends.”

Share this:

Share this:

Share this:

Share this:

Share this:

Share this:

Share this:

Share this:

Share this:

Share this: