An expos¨¦ which found that a Wiley journal had likely published articles written by Russian-linked paper mills and vetted by fake reviewers has highlighted the case for making open peer review mandatory, one of the academic sleuths behind the investigation has said.

In a study?that analysed six papers ¡°with serious flaws¡± published last year in the?Journal of Community Psychology, Anna Abalkina, a research fellow at Free University of Berlin, and Dorothy Bishop, who recently retired as professor of development neuropsychology at the University of Oxford, state that ¡°none would be published if proper peer review and editorial scrutiny had taken place¡±.

Of the six papers, most included citations to predatory journals and lacked any new empirical data, while some presented data in a ¡°disorganised way¡­making it difficult to work out what was done¡±, explains the investigation??which adds that there was ¡°no indication of any competence in experimental design or data analysis¡± in many of the data-focused papers.

Further concerns about the papers, which were initially flagged owing to ¡°suspicious email domain names¡± that did not correspond to the authors¡¯ country of residence, were raised when the peer review comments related to each paper available on?Publons?were analysed.

ÁñÁ«ÊÓƵ

¡°There is very strong evidence to suggest that this was fake peer review,¡± Dr Abalkina told?Times Higher Education. ¡°Most of the comments were very formulaic ¨C things like: ¡®Please do English language editing¡¯ or asking for them to change the text formatting or to update their citations,¡± she added.

¡°Many of the review comments written by those with anglicised names also used language that a native English speaker wouldn¡¯t use.¡±

ÁñÁ«ÊÓƵ

One example was the repeated use of the phrase ¡°methodological arsenal¡±, she said. ¡°That¡¯s a direct translation from Russia and I¡¯ve seen it in thousands of Russian-language papers but it¡¯s not an expression, I¡¯m told, that is used in English.¡±

Further investigations into the credentials of peer reviewers were also illuminating. ¡°These people did not exist ¨C you cannot find them on their university¡¯s website and they have no publications at all,¡± she said. Follow-up calls to genuine academics in their departments confirmed they had never worked there, said Dr Abalkina, whose research has uncovered hundreds of papers?likely to have been written on behalf of authors by paid-for services in Russia and India.

The dates on which reviewers submitted their comments also raised further suspicions that peer review had been corrupted, she continued. In nearly all cases, both reviewers submitted their first reviews on exactly the same day, and also submitted their second reviews on the same day (or a day later) despite supposedly operating independently.

Those patterns and other red flags, allied with the papers¡¯ low-quality content, suggest these papers came from a ¡°paper mill that targets psychology journals with weak editorial practices¡±, the pair write.

ÁñÁ«ÊÓƵ

The journal¡¯s editor-in-chief, Michael Blank, professor of psychiatry at the University of Pennsylvania, did not respond to?THE¡¯s request for comment, but a Wiley spokesman said the company was ¡°committed to upholding research integrity and publishing ethics in the journals we publish¡± and was ¡°investigating the concerns raised by their preprint in line with guidance set out by the Committee on Publication Ethics¡±.

¡°We thank Dr Abalkina and Dr Bishop for their excellent work and commitment to securing trust and integrity in scholarly research,¡± he said.

Dr Abalkina said the case demonstrated why open peer review, in which reviewers¡¯ identities and comments are freely available, must become the industry standard as these problems would not be able to surface without this kind of transparency.

¡°Before this case I didn¡¯t see confidential peer review as a problem, but I understand now that open peer review can be a very powerful tool for spotting problems,¡± she said.

ÁñÁ«ÊÓƵ

jack.grove@timeshighereducation.com