More and more articles are being retracted due to manipulated peer review

Brussels (Chrono): There has been a significant increase in retractions of scientific articles due to manipulated peer reviews.

It shows the numbers from the retreat hour.

Since it started in 2010, Retraction Watch has mapped thousands of retreats and built a Database With over 47,000 entries. They were withdrawn for various reasons. Having someone withdrawn due to fake or questionable reviews is nothing new.

“This is an old story,” Ivan Oransky, co-founder of Retraction Watch, tells Khrono.

However, something happened.

And so it increased

The figure below shows the number of articles retracted due to faulty or questionable peer review. The numbers are from the Retraction Watch database, and show the year articles were retracted, not the year they were published.

But this is only part of the picture.

There has been a significant increase in the number of retracted articles. From this point of view, it is not surprising that there has also been an increase in articles being retracted due to peer reviews. But figures from the Retraction Watch database show this is becoming a more common reason for retracting articles.

Of the 5,447 articles retracted in 2022, a total of 2,748 articles were retracted due to false or questionable peer review, as the figure below shows.

As I said, these numbers show the year things were retracted, not the date they were published, but the numbers on the number of articles retracted due to peer reviews based on when they were published show a similar evolution.

“There are more and more of them, and they are responsible for a large percentage of the increase in withdrawals,” Oransky says. He adds that retractions due to manipulated peer review can often be linked to so-called “paper mills,” or article mills, that produce fake research articles.

Nine articles were withdrawn in two days

Last week, Khronow was able to talk about a researcher in Bergen who pulled nine articles in two days. The articles were published from February 2022 to January 2023 and were withdrawn in July this year. Apart from one article, all nine articles are about the coronavirus. In five of them, the University of Bergen is listed, and in the other four universities in Thailand and Japan respectively are listed as affiliated.

In justifying the retractions, publisher Wiley wrote that they were “withdrawn due to compromised peer review following an investigation conducted in accordance with COPE guidelines and the Wiley review process, and in agreement with the journal’s editor-in-chief.”

They offer no further explanation to Khrono. The articles’ corresponding author wrote in an email to Khrono that they had not received an answer about what kind of manipulation the publisher believes was involved.

Without knowing what it’s like in this case, Oransky says writers like to use the fact that publishers won’t say what actually happened to make it seem like there’s no problem.

I wrote about manipulated ratings

As Khronow wrote last week, regarding the case in Bergen, there are several known ways to cheat peer reviews. Editor-in-Chief Ari Breen noted, among other things, that some journals allow authors themselves to suggest colleagues. It allows you to provide false information, for example in the name of a non-existent colleague, with an email address that you control yourself. Researchers can also agree among themselves to evaluate each other’s articles.

Such a manipulation was written by Oransky and two of his colleagues at Retraction Watch already in 2014, i.e Peer-reviewed article about fraudwhich was published in the journal Nature.

They pointed to several examples. In one, a researcher in South Korea admitted that he wrote many of the peer reviews of his articles himself. It was addressed after receiving a large number of reviews quickly, often within 24 hours. As an author, he was asked to provide suggestions on colleagues, and they wrote that he came up with names of real researchers and pseudonyms, often using fake email addresses that went directly to him or his colleagues.

In another case, from 2013, a 14-month SAGE investigation revealed a group of researchers who cited and reviewed each other at a high rate. 60 subjects were drawn.

In most of the 60 articles, only peers nominated by the authors were used in the original evaluation.

Oransky and his co-authors point out that many of them believe that journals should not allow authors themselves to suggest peers.

However, such suggestions are not always followed. One of two editors in chief at Osteoporosis International, published by Springer, said in a 2014 article that they allow authors to recommend up to two of their peers, but are happy to use the information to exclude them. The same editor refers to a case in which an author suggested a colleague with the same first name as himself, but a different surname. It turns out that the author herself suggested only the last name she had before her marriage.

Announce the suspension using special numbers

It is not uncommon for several articles to be retracted at the same time due to peer review manipulation. This often happens, Oransky explains, because it is the result of an investigation by a journal, a publisher, or sometimes a university.

It also does not necessarily say anything about its size, but rather about the date it was found.

“If you set up a speedometer and catch ten people on the same day driving too fast on the highway, it doesn’t say anything about how often that happens, whether those are the worst drivers or whether they’re just the ones who get caught. Oransky points out that he is not saying anything about what will happen tomorrow.

As for Wiley, the publisher announced earlier this year that it would temporarily stop publishing special editions of Hindawi, which the main publisher acquired in 2021. This cost them several million US dollars in lost revenue.

There has been a significant growth in such special editions from publishers, who profit from publishing large volumes.

Oransky points out that many instances of false peer review occur in such private cases. Retraction Watch wrote earlier this year about how Wiley and Hindawi identified about 1,200 articles that had been subjected to compromised peer review and were scheduled to be retracted.

The nine articles written by the Bergen researcher were not published in Hindawi’s journals. Khrono asked Wiley if they were in any way connected to the withdrawal from Hindawi magazines that was announced earlier this year. They reject this. “These withdrawals are not related to withdrawals in Hindawi’s wallet,” a Wiley spokesperson wrote in an email to Khrono.