/ 16 November 2018

The business of fake science

(John McCann/M&G)
(John McCann/M&G)

Predatory journals are scamming academics around the globe.

Most scientific journals charge academics fees to publish their work, which is then subjected to editorial and peer review, but predatory journals charge fees with little or no quality oversight. This means, for example, that when an academic writes about a health product, the companies marketing the product can claim scientific credibility for it, although this has not been verified by the academic’s peers.

It’s a global problem. Investigations worldwide have highlighted the problem of predatory journals and fake science; this year British newspaper The Guardian and German broadcaster Norddeutscher Rundfunk reported that 175 000 scientific articles published by the top five predatory journals bypassed basic checks and balances for scientific research.

The newspaper and the broadcaster reported that over the past five years more than 10 000 scientists from British and German universities have published articles in predatory journals. Süddeutsche Zeitung’s Katrin Langhan says worldwide 500 000 researchers have been targeted by these unscrupulous journals.

In South Africa, the department of higher education, which funds academic research, is reviewing its accreditation processes after a study last year showed that publications it supported have appeared on a list of predatory journals.

Research, by Johann Mouton and Astrid Valentine, in a paper titled The Extent of South African Authored Articles in Predatory Journals, shows that, of 276 academic journals supported by the department, 47 are “probably predatory”.

This means that between R100-million and R300-million in research grants over a 10-year period ended up supporting academics to publish in fake journals, Mouton and Valentine conclude. They based this on an average subsidy of R100 000 for research to become a published article.

The department says it is addressing the issue. It has commissioned “a study on the quality of research outputs, including predatory journals, [and] will share the findings once the work has been concluded”, says Fhumulani Maanda, who is responsible for research support and policy development in the department.

Susan Veldsman, of the Academy of Science of South Africa, says discussions, workshops and a conference have taken place with the goal of producing a declaration on ethical publishing practices.

Fake science publishing also means unscrupulous politicians and other people can use predatory journals to support their political ideologies, Langhans said at the African Investigative Journalism Conference at the University of the Witwatersrand earlier this month. She cited the example of politicians using this “research” to deny climate change.

Predatory journals have also been used to promote hoax remedies. Langhans’s team of 60 researchers found that “self-made scientists” used particular journals to promote “a useless treatment called GCMAF” to fight cancer.

The “wonder drug”, which has not been clinically tested, claims to use a “vitamin D binding protein derived macrophage activating factor” that occurs “naturally in the body”. It has been used to scam cancer patients for more than 20 years, according to online fact-checking website Snopes.

GCMAF began circulating in the public domain following an article published in various journals in 2008 and 2009. It was later denounced by a Belgian nonprofit organisation, leading to the publications retracting the article in 2014, but the “cancer cure research” nonetheless resurfaced again in 2015.

To illustrate the lack of sound editorial practices and scientific review, Langhans and her team used SciGEN, a web-based computer programme, which randomly generates computer science research papers, complete with graphs, figures and citations. Langhans “authored” a piece along with Albert Einstein and submitted it to a predatory journal this year. The article was accepted.

In 2013, science journalist John Bohannon exposed the lack of editorial protocol and peer review when he published a paper by Ocorrafoo Cobange on a made-up wonder drug and submitted it to 304 journals. The article was accepted by more than half of them. The journals were selected from a list of 923 that have shown evidence of being predatory, according to United States librarian Jeffrey Beall.

Between 2005 and 2014, 4 246 South African papers were published in the 47 “probably predatory” journals. Mouton and Valentine arrived at this figure after they refined Beall’s classification of predatory journals, ruling out some journals that Beall considered predatory based on poor publishing processes.

Mouton and Valentine say that publishing in journals is important for university rankings, research metrics and academic careers. “In academia today, there is huge pressure to publish, and publish fast.” They say, in the past 10 years, 3.4% of articles by South Africans were published in predatory journals; 2.5% of these appeared in journals with strong evidence of being predatory; and 0.09% of them in journals with weak evidence of being predatory.

Most of the articles published in predatory journals were linked to the social sciences and humanities, followed by economics and management.

Mouton says academics should not be criticised for publishing in these journals. Writing in the South African Journal of Science, he says, because the journals were listed with the department of higher education, the academics had done due diligence.

The publishers of predatory journals sometimes also promote equally dodgy conferences, which are designed to appear legitimate, sometimes announcing renowned scholars as speakers, who never appear because they were not invited or involved in the process. In some cases, delegates arrive to find that a conference for which they paid fees to attend is nonexistent.

The conferences applaud scholars for their research and ask them to present keynote addresses, Dr Mad-hukar Pai, the Canada research chair in epidemiology and global health at Montreal’s McGill University, says in a Huffington Post article.

Typically, Pai says, the conferences have little or nothing to do with the scholars’ area of expertise. If scholars fall for the conference and attend, they end up paying exorbitant registration fees combined with hotel, travel and food expenses.

Scholars have criticised Beall’s classifications, arguing that some journals may not intend to defraud academics, as predatory journals do, but that their editorial and review practices are poor. Beall’s list has also been accused of being biased, favouring Western subscription-based publications and being against open-access publishing, says Wits University’s Denise Nicholson.

The drawing up of Beall’s long list did not involve any direct communication with the editorial boards of the journals and threw undue suspicion on startup publishers, she says

Open-access publishing makes peer-reviewed publications available online. There are two types: gratis and libre. The former allows for free and permanent access to research. The latter also allows free access, with the additional benefit of republishing and reusing some of the articles.

Beall’s list, which was taken down after he was threatened personally, gave advice on how to avoid being scammed by predatory publishers and what to look out for, such as over-complimentary emails, badly written invitations and unrealistic turnaround times for the publication of articles.

In the year since Mouton and Valentine made their research public, South African institutions have made efforts to help academics to discern which publications are predatory and which are not, but this is not easy.

“There is no one-stop shop or comprehensive resource or guidelines available at this stage for scholarly authors to consult before publishing,” says Nicholson, although scholars can use resources such as Scopus, the Web of Science and the Directory of Open Access Journals.

“To avoid falling into the trap of predatory journals, it is wise for scholarly authors to research publishers carefully before submitting their manuscripts for publication,” she says.

“It is recommended that they check available whitelists and tools, their own reputable institutional journal collections and electronic databases, internationally accepted journals indices, as well as best practice guidelines produced by reputable international publishers.”

The number of published academic papers has increased in recent years since the introduction of grants and the ranking of researchers by the National Research Foundation.

This was noted by the 2015 South African Science, Technology and Innovation Indicators, which said 4 000 articles were published in 2005 compared with almost 11 000 articles published in 2014.

To date, there have been few to no prosecutions. One of the big name cases The Guardian reported has been India-based Omics, which came under fire from the United States Federal Trade Commission in 2016. The trade commission has charged the publication with “deceptive publishing practices”.

Omics allegedly charged between $30 and $1 800 to publish an article.

Currently, South African publishers and universities send their publications to the department in about June to be added to the department’s accredited list of journals. A panel assesses the publications and either places them on their list for the next year or rejects them.

Criteria include that a journal has to have been published without interruption for three years and articles must be peer-reviewed. The department also requires details about the publication’s editorial board and its institutional affiliations. Once a journal has been removed from the department’s list of approved publications because it has not upheld the criteria required, it has to wait for a minimum of two years before it can reapply.

According to the department’s research outputs for 2015, it samples journals from time to time but individuals and institutions can notify it, with their reasons for why a publication should be removed from the department’s list. It then reviews the publication.