China frequently makes news for being at the forefront of peer-review scandals like this one and this one. And data appears to bear that out, showing China contributed well over half of the papers retracted for compromised peer review from 2012 to 2016, according to data obtained by Quartz.
Peer review by scientists in the same field as someone trying to publish research is supposed to help journals and their readers make sense of how credible and important the work is. Problems with peer review taint that process, and can extend the gamut from reviews that are carried out by someone affiliated with the researcher to entirely made-up reviews.
Over the past five years, a total of 498 papers have been retracted over peer-review issues, according to the US blog Retraction Watch (papers can also be retracted for other reasons, but those papers aren’t included here). The blog used the nationalities of corresponding authors to reach its tally, since they are responsible for paper submissions. The breakdown by country below totals 502, reflecting papers counted twice because of corresponding authors with affiliations in multiple countries.
While the blog’s co-founder Ivan Oransky notes that the list isn’t comprehensive, the blog does track scientific and medical publishing very closely, including with a researcher who “is constantly scanning those databases,” making the numbers a good indicator. “I am confident that we capture most retractions for fake peer review,” said Oransky.
China has the worst performance on peer-review integrity even when compared with the top five countries (see chart below) for scientific publishing in the last five years, using Nature Index, which tracks publications in nearly 70 journals as an indicator of a nation’s high-quality research output. The index’s data covers a 12-month rolling window, with the latest numbers covering February 2016 through January 2017.
Part of the problem is that Chinese researchers, who have been publishing in ever greater numbers—last year Nature Index showed China published 9,721 research articles, up from 6,587 in 2012—often rely on third-party companies to help translate, format and submit their work to journals, and sometimes this outsourcing results in efforts to game the system.
Last month, over 100 papers from Chinese researchers published between 2012 to 2016 in Tumor Biology, a cancer-research journal, were linked to review fabrications. Meanwhile, Germany had one retraction over peer review in the past five years, and Japan, which published some 4,000 papers last year, hasn’t seen any papers retracted for these problems.
“Many scientists in Japan write fluently in English, and those that don’t have help from university translation services,” says Oransky, “There doesn’t seem to be the demand for such services that created a market for unscrupulous third-party companies.”
An earlier version of this story said 502 papers were retracted for the period studied. The 502 number includes papers counted twice for reasons explained above.