Preprint servers gain prominence despite peer review concerns

This article was first published on University World News on 17 March 2022

While they have been around for the past five decades, preprint servers, repositories that allow access to original manuscripts to the public before they have undergone peer review, have gained more prominence over the past two years, owing to the sheer volume of research on COVID-19 that needed to be communicated.

In Africa, as elsewhere in the world, they have also been attracting debate and controversy due to their very nature of availing science before it is peer reviewed, including allowing users to cite it, which, to some in academia and the wider research community, remains unconventional and unacceptable.

Despite that, promoters believe preprints come with many advantages, and could be one of the interventions Africa needs to increase its scientific output or make its research more visible. One such advantage is that they shorten the period an article needs before it is published, since preprints make works available online as soon as they are submitted.

“Ordinarily, the publishing process, from submission to review and final publication in a journal, may take between three and 12 months and, in some cases, even longer when a paper is rejected,” says Joy Owango, a member of the advisory board of Africa Archives (AfricArXiv), Africa’s first public preprint server. The University of South Africa, or UNISA, also has a preprint server, but it is accessible only to its own researchers.

This, she says, makes delays a challenge, eventually affecting the visibility of science, especially in Africa, where research output is already low.

It makes preprints necessary because they are able to make academic publishing more ‘rapid’. By their very nature, she adds, preprint servers operate under the principles of open science and open access, which makes their content more discoverable.

Challenging a traditional mindset

While many universities in Africa have their own repositories, they also face infrastructural challenges that make the visibility of research a challenge, a problem that can be solved by preprints through indexing.

The preprint repository, besides publishing in indigenous African languages, supports universities to help make their scientific output, not only more visible, but also easily “findable and accessible and in a platform that is interoperable and more see-able” through indexing, she notes.

Besides universities, the organisation was conducting training with higher education entities, including African libraries associations, training them on, among other things, the importance of indexing, in promoting discoverability of research, Owango explained.

Some of the universities that had shown interest in working with AfricArXiv included the universities of Nairobi and Dar es Salaam in Kenya and Tanzania respectively, and other institutions in other regions were excited about the preprint approach.

“Even as preprints remain relatively new in Africa, academia is beginning to see their importance in that works get noticed even before they are presented in forums such as conferences,” Owango added.

She says that, since its launch in 2018, AfricaArXiv has received huge interest from librarians and heads of research in universities and, where it has faced resistance, it was mainly due to lack of awareness, blamed on the traditional academic publishing mindset, fear of the unknown and resistance to change.

Wherever preprints are embraced, they, for example, make it easy for staff to earn promotions as their work appears on the repositories even before it is formally published.

Upholding data sovereignty

Preprints, according to Jo Havemann, a co-founder of AfricArXiv, have the advantage of making data easy to discover from wherever it is searched for, including engines such as Google.

Discoverability of research is one of the challenges that faced African academic publishing, she noted, adding that one of the strengths of preprints was the ability to guard against plagiarism and theft.

“Preprints protect manuscripts and uphold research data and sovereignty through digital objects identifiers to indicate ownership of works,” she said.

They are about acceleration of science communication, upheld quality, published free of charge, while upholding data safety,” she added.

They facilitated research exchange, had a robust system of quality assurance and allowed feedback from the public while challenging traditional norms of academic publishing.

“Publishing in prestigious journals does not necessarily mean that the research work is good,” she declared.

‘A helpful cultural shift?’

Stephanie Dawson, the CEO of Science Open GmbH, Germany said: “In an academic landscape where gateways abound, preprints are a way for African researchers to publish their ideas early, fast and open access.”

Their voices can be heard even while the important process of peer review and the less important sorting by impact factors is under way. As a result, they were happy to work with AfricArXiv to increase the discoverability of African research, she said.

According to Catherine Ahearn, the head of content at the Knowledge Futures Group in the US, one way in which her organisation was promoting a more effective, equitable, and sustainable knowledge economy was by working with partners like AfricArXiv.

Preprints and the openness, greater collaboration, and faster time frames that often accompany them, plus the associated Publish, Review, Curate model mark a “helpful cultural shift”, to support African researchers.

A recent survey showed that more than 40 new preprint servers have been established in the past decade.

This includes disciplinary and regional preprint servers, such as AfricArXiv, SciELO in Latin America, RINarxiv in Indonesia and IndiaRxiv, says Luke Drury of the International Science Council (ISC).

“The ISC is deeply concerned about the extent to which the dominant commercial model of scientific publishing fails to serve the needs of modern science,” he said.

“Among other failings, journal pricing far exceeds the cost of production and lies beyond the reach of authors or readers, particularly in low- and middle-income countries,” he lamented.

He asserted that preprints perform a valuable service in accelerating the advancement of science, are here to stay, and in all likelihood will see that usage continues to grow everywhere, including in Africa, as they offer a “zero-cost method” of rapidly publishing and accessing the latest scientific research.

One of the disadvantages of using preprints is that they are sometimes considered to be less valuable forms of research output in funding or hiring decisions, Drury noted.

“In addition, preprint servers depend on third-party support and voluntary contributions, like all shared infrastructures that exist for the common good.

“However, they are comparatively cheap when compared to the costs of traditional journal subscriptions or article processing charges,” he told University World News.

You might also be interested in

The normalization of preprints

An ISC Occasional paper by Luke Drury that explores the explosive growth in the use of preprints and associated preprint servers by large sections of the scientific community. This ISC Occasional Paper addresses the history of the preprint, its advantages and potential disadvantages, and concludes with some recommendations for how the growing acceptance of preprint posting should be handled within academia, and the changes in cultural norms that this entails.

Most of the criticism levelled at them is that they have not been peer reviewed, but even without formal peer review, many preprints will have already been scrutinised by colleagues.

The servers operated some kind of gate-keeping to eliminate “inappropriate” content, and always made it clear to readers that preprints have not been peer reviewed, he added.

“There are certainly risks associated with preprints, but I think they are exaggerated and, in my view, excessive trust in peer review as currently organised by journals is even riskier,” he said.

“All forms of publication are potentially open to manipulation by bad actors, but it is hard to see how preprints could do anything like the harm that predatory publishers do at the moment,” he observed.

There was evidence to indicate that plagiarised and fraudulent papers are almost never posted as preprints, presumably because they attract early scrutiny from multiple peers.

“Traditional peer-review systems remain fallible; it is well known that some of the most damaging cases of fraud and misinformation have appeared in established journals as well,” he added.

Scientists who are eager and perhaps impatient for their findings to be published are increasingly posting their findings in preprints, according to an article by François van Schalkwyk of Stellenbosch University, South Africa, and Jonathan Dudek of Leiden University, The Netherlands.

However, scientific publications which are yet to undergo scrutiny by peers remain provisional, and are risky, especially when their claims are repeated in the news media in “undifferentiated or uncritical ways”.

This calls for caution and for the judicious use of science without compromising the benefits of its openness, they noted.

Preprints have had huge growth over the past 10 years as researchers see the need for researchers to be credited for their work in an efficient manner, with no barriers to entry globally, says Mark Hahnel of Digital Science UK.

Post-COVID, the world understands the need for “fast but good” publication of research and there was a need to foster this approach across Africa through preprints, he said.

Image by Susan Q Yin on Unsplash