Open access vs paid-subscription journals: Flawed, gift-wrapped or just plain science?

Open access is getting more and more popular, but in some cases the free availability might carry a price. From www.plos.org

Scientific journals offering some or all of their articles as open access – i.e. freely available on the internet for anyone – is getting increasingly popular. Our own Physiologia Plantarum has given the option of open access for some time and many other journals also give authors an alternative to the traditional paid-subscription model. This is mainly because many funding bodies or institutions now require or strongly encourage scientists to publish in open access journals in order to get the best value for money by maximising exposure of the funded research.

But the free availability of open access articles might carry a price. First of all, there are expenses connected to publishing that the publisher would like to cover and since the far majority of publishers offering open access are private companies they might even want to generate a profit. So consequently most of them charge the authors a fee for open access publishing. This has led to concern that some – but definitely not all – publishers might be tempted to increase turnover by accepting papers of such low quality that they rightfully should have been rejected. But on the other hand some backers of open access are now shooting back. They accuse the most prestigious paid-subscription journals for a preference to publish ‘sexy’ articles that appeal to a large audience and generate media buzz, which in turn drive subscription sales.

The Directory of Open Access Journals (DOAJ) lists an increasing number of journals offering open access while not compromising on quality. From poeticeconomics.blogspot.dk

Around the turn of the millenium, a number of new and somewhat unusual scientific journals saw the light of day. The journals were exclusively online and their articles could be accessed for free by anyone. Soon to be dubbed ‘open access’ this kind of journals has become increasingly popular, the number of titles is steadily growing and they exist in numerous versions. Some, like PLOS Biology or American Journal of Plant Sciences, are ‘fully open access’ and all accepted papers are free to read for everybody from the day of publication. Others, like PNAS and our own Physiologia Plantarum, are so-called ‘hybrid open access’, i.e. paid-subscription journals that offer authors the option to publish under open access. And yet again some, like The Plant Cell and Plant Physiology, are so-called ‘delayed open acces’ where all articles become open access after a certain embargo period of variable length. For these particular journals – that also operate on a hybrid open access basis – the embargo period is 12 months.

Having an article published as open access will typically set back the authors in the range of $1000-$3000 but many journals offer a reduced fee or can even waive it entirely for those with low budgets. While traditional journals have an inherent interest in accepting only articles describing science of the highest quality – since that will give them a good reputation and allow them to sell more subscriptions – the situation is more complex for journals offering open access. Most of the journals offering full or partial open access are highly respectable journals that carefully evaluate the quality of submitted manuscripts through a thorough peer-review process. For these journals – Physiologia Plantarum being one of them – the option to pay for open access does not make any difference in how the manuscript is handled or evaluated and it is only regarded as a special service for the authors that the publisher need to be compensated for. But there are rotten apples in the basket.

Many Indian-based publishers accepted the flawed paper. Click here to enlarge. From scicomm.scimagdev.org

Reporting for Science, John Bohannon last year published a remarkable study. As a test, he had intentionally written an obviously flawed scientific article describing a novel – but purely imaginary – cancer drug and submitted it for publication in 304 open access journals. It turned out that more than half of them, 157, accepted the flawed article for publication – and accordingly also accepted to cash in the publication fee. 48% of those journals accepting the paper seemed to have given it no peer review at all, commenting exclusively on the bogus manuscript’s layout, formatting and language. Perhaps even more frightening was the fact that 10% of them actually identified the scientific flaws in their peer review process but still ended up by accepting it, eventhough the flaws were never corrected. Since John Bohannon’s paper is probably not the only low quality or even flawed manuscript that has been submitted to open access journals, we must expect that an unknown proportion of the scientific results we read about in some of these journals are simply not reliable.

More than one-third of the journals that accepted the bogus paper were based in India, but also journals from USA, Europe and Nigeria were happy to cash in on the scientific flaws. Remarkably, some of the journals that accepted the manuscript was published by renowned companies Elsevier, Wolters Kluwer and Sage. This can make it difficult for both readers and authors to discriminate between serious journals that strive to publish only the best research and those that, boldly spoken, will publish anything for a fee. However, it is important to keep in mind that these so called ‘predatory journals’ are not representative of all journals offering open access. Probably no one would question the high standard of hybrid open access journals like PNAS and Physiologia Plantarum, where the thorough peer reviewing process does not discriminate between how the papers are published.

PLOS ONE has grown to become the largest scientific journal in the world. Data from en.wikipedia.org/wiki/plos_one. Graphic by Gorm Palmgren

In fact, some of the most prestigious journals are fully open access. These include the whole bunch of PLOS journals, with the flagship PLOS ONE taking a most particular stance. Since being launched only 8 years ago it has grown to become the largest scientific journal in the world with 31,500 papers published last year – that is equivalent to publishing a copy of Nature every 4 hours. Behind the numbers lie a stark contrast to Nature and other highly acknowledged paid-subscription journals. While these will pre-select among submitted manuscripts and only sent around 10% out for peer review, PLOS ONE will peer review all manuscripts solely on the basis of scientific quality and publish all that passes. In other words, there is no subjective judgement on the paper’s relevance or importance. While this can be seen as a very democratic practice that honors all interests, publishing a huge number of articles with little interest or relevance to most readers might also lead to confusion and a blurry picture of the scientific state-of-the-art. Furthermore, one can argue that since this particular business model is probably very profitable there might be other arguments than pure idealism for choosing it. Anyway, the practice is very different from almost all other journals that make pre-selection, and according to Nobel Prize laureate Randy Schekman this specifically compromises Nature, Science and Cell, which he calls ‘luxury journals’.

In a comment in The Guardian last year, he argues that this pre-selection – which is necessary because of the limited number of pages – is skewed. Instead of deciding entirely on the basis of scientific quality, the editors tend to prefer papers “that will make waves because they explore sexy subjects or make challenging claims” – subsequently generating more citations, a higher impact factor, more subscriptions and a higher profit. Papers selected on such a basis will not necessary be good science, but they will be regarded as such since they are published in a ‘luxury journal’. Randy Schekman goes on to argue that “This influences the science that scientists do. It builds bubbles in fashionable fields where researchers can make the bold claims these journals want, while discouraging other important work, such as replication studies.”

eLife is backed by research institutions and has no publishing fee – at least for now. From elife.elifesciences.org

Despite having paved his own way to the Nobel Prize with 46 papers published in Nature, Science and Cell, Randy Schekman now encourage all scientists to boycott these ‘luxury journals’. And as an alternative, high-end publication he has launched the open access journal eLife, which is a unique breed in many ways. Unlike most other open access journals, eLife carries no publishing fee, since it is backed by £15 million from the Howard Hughes Medical Institute, the Max Planck Society and the Wellcome Trust, but according to the journals webpage they might introduce a publication fee at a later time. Like the ‘luxury journals’ it will pre-select among the submitted manuscripts, but rather than to do so on a subjective basis that ultimately aims at selling subscriptions, it will be as objective as possible and only publish “research of the very highest standard and significance” as judged by an editorial board of some of “the world’s most talented biomedical scientists”.

It is an open question whether the editors at eLife will be better and/or more objective in the pre-selection than their colleagues at other journals. Editors at Nature, Science, Cell and other journals – whether open access or not – are typically scientists holding a PhD, and even if they are no longer active researchers but work as professional editors, it is probably not fair to accuse them for being ignorant to scientific facts and quality. And even if professional editors might be slightly skewed towards ‘blockbuster’ papers, active high-profile scientists in the role as editors might tend to prefer articles that back their own scientific theories or interests. However, eLife strives to keep the peer-reviewing process transparent by giving authors the choice of having the editorial decision letter and their own response published alongside the paper. This practice, however, concerns only papers that are finally published, so the public will never know the reason why a manuscript was rejected during pre-selection. Since the PLOS series of specialized journals (not PLOS ONE), which also practice pre-selection, do not adhere to this kind of transparency, Randy Schekman hopes that it will enable eLife to compete better with the ‘luxury journals’.

Be critical to what you read! From commons.wikimedia.org

So where does this leave honest scientists when they want to publish their results or read about the progress of others? Can you trust the research published in open access journals or is it at risk of being flawed? And are the exciting discoveries you read about in Nature and Science just gift-wrapped trivialities? Of course the picture is not so black and white. Most open access journals are fully trustworthy and those operating on a hybrid basis – like Physiologia Plantarum – are probably even more so because they are still subscription driven and must care about their reputation. And clearly even sexy or provocative papers can be of high scientific quality. But one should keep in mind that both publishers and authors can have personal and hidden incentives for publishing: publishers might be skewed towards profits rather than pure science, and authors feeling the pressure from funding bodies might be tempted to expand their publication lists by submitting questionable research to those journals willing to publish it.

Since it is impossible to proof-read the massive amount of information available on the internet, we have learned to be critical, check the sources and pay attention to potential conflicts of interest. Probably we have to get used to do the same when we browse the scientific literature: not believe blindly in everything we read but keep a list of those journals we trust in and be skeptical to the rest.

By Gorm Palmgren, science writer, PhD, www.palmgren.dk