As an Open-Access Megajournal Cedes Some Ground, a Movement Gathers Steam

The world’s largest scientific journal, the open-access giant PLOS ONE, is feeling some pullback. Last year the free site published 10 percent fewer papers than it did two years ago. Its impact factor — a measure that uses citations to track its influence — has been on a five-year slide.

Rather than signaling a failure of the open-access movement, however, the declines are looking like the byproduct of a broader victory in a hard-fought campaign. More and more, major publishers are creating their own open-access journals, with articles freely available to anyone. And in many other cases they’re offering hybrid models that let authors pay for open access. An increasingly common version of author-paid open access is the “megajournal,” copying the PLOS ONE innovation of publishing a large volume of papers online across various disciplines.

In short, PLOS ONE — now consistently publishing around 30,000 articles a year — has attracted much more company in its mission to build huge stocks of freely available scientific research. “Since PLOS ONE’s tremendous success, everyone and their grandmother has created a megajournal,” said David J. Solomon, an emeritus professor of medicine at Michigan State University who studies open-access economics.

After years of traditional journals battling the open-access movement, said another analyst, Jevin D. West, an assistant professor of information studies at the University of Washington, “look at all the major publishers — they’re all playing now.”

There’s Scientific Reports, from the Nature Publishing Group, which has grown from about 200 articles in 2011 to 11,000 last year; BMJ’s BMJ Open, which increased during that period from about 100 to 1,000; and the American Institute of Physics’ AIP Advances, which jumped from about 250 to more than 500. And more are coming. Just in the past year, the publishing giant Elsevier introduced Heliyon, and the American Chemical Society announced ACS Omega, both open-access multidisciplinary megajournals.

Amid all that, PLOS ONE, the genre’s originator and overwhelming heavyweight, went from just under 7,000 articles published in 2010 to nearly 32,000 in 2013 before settling down to about 28,000 last year.

PLOS, the Public Library of Science, was founded in 2001 by a group of academic advocates of open access — the idea that universities would save money, and science would progress more quickly, if researchers and their funders paid journals in advance to review and publish their articles rather than pay later for subscriptions. PLOS ONE, its now-flagship journal, was created in 2006 with the additional idea of accepting submissions based only on their scientific quality, without regard to medical field, and without any attempt to predict whether an article would eventually prove important.

Officials with the library say the end of PLOS ONE’s rapid growth is not indicative of an underlying problem. Instead, it is due largely to the increased competition in open-access publishing and the finite supply of scientist-authors, especially at a time of tighter research budgets, said Elizabeth Marincola, chief executive officer of PLOS. “This was a very predictable evolution in the nature of PLOS ONE’s growth, and other megajournals experience the same thing,” Ms. Marincola said. “The number of papers being published by pre-existing OA publishers are softening simply because there are more OA choices.”

‘An Art and a Craft’

The slight slide in PLOS ONE’s impact factor — the average number of citations to a journal’s published articles in the previous two years — may be somewhat tougher to assess. After reaching 4.4 in 2010, it’s now down around 3.2 — still above all but a couple of dozen medical journals.

One cause could be simple mathematics. Articles acquire citations over time, so an impact factor tends to decline during periods of fast growth at a journal. Another reason, Ms. Marincola said, is PLOS ONE’s mission to provide a low-cost platform for sharing science as widely as possible.

Rather than focus on a journalwide impact factor, she said, the journal has other ways of judging its impact. PLOS ONE had more than 5,000 articles in 2013 that accrued 10 or more citations — five times as many as Nature, Ms. Marincola said.

The validity of impact factor and similar metrics, of course, is a matter of extensive debate, with advocates and skeptics spread across the open-access battlefield. Doubters include Björn Brembs, a professor of neurogenetics in the Institute of Zoology at Regensburg University, in Germany, who published findings this past week showing journals with higher impact factors also were more likely to have retractions. Mr. Brembs also cited cases of publishers’ improving their scores by negotiating with owners of databases over which articles would be included in the calculation.

Nevertheless, traditional publishers often still embrace citation-based measurements, saying they demonstrate the superiority of journals that customers willingly pay to purchase over open-access alternatives.

“There’s fierce competition between different publishing houses to attract and to work with the very best scientists in a field who have that real cutting-edge knowledge of where the discipline is at, where it’s come from, where it’s going, what advances the state of the art,” said Alicia Wise, director of access and policy at Elsevier, one of the world’s biggest publishers of scientific journals. “There’s an art and a craft in that — that’s what experienced publishers and editors bring to a title.”

Many leaders in open-access publishing, including PLOS, also see value in citation-based measurements, since many of their publications can score as well as traditional journals. But a megajournal such as PLOS ONE, Ms. Marincola said, is also based on the notion that it’s foolhardy to try predicting what science will prove most useful in the future. The best approach, she argued, is to publish as much sound research as possible, and then let the entire scientific community decide what is most helpful.

That continues to be a tough sell for some faculty members who have been conditioned to view impact factor as the chief definition of academic value, said Lorraine J. Haricombe, vice provost and director of libraries at the University of Texas at Austin. “Unless it’s out there in the community for others to review, use, and repurpose, they have no idea what the impact of that may be a generation later or more,” said Ms. Haricombe, a founder of the Coalition of Open Access Policy Institutions in North America, an international advocacy group for institutions with open-access policies.

Rooting Out Predators

For now, though, citation-based metrics can help weed out low-quality and sometimes “predatory” websites that have arisen since the early days of the open-access era. Those publishers try to collect authors’ submission fees without subjecting articles to even minimal levels of peer review, said Mr. West, of the University of Washington.

One of the more notorious examples of the problem occurred in 2013, when Science magazine published the results of a yearlong exposé of open-access journals. The investigators submitted a fraudulent cancer-research article to more than 300 open-access journals; about half of them agreed to publish it.

Mr. West has developed a statistical method of identifying such publishers. It’s simply a ratio of the fee each journal charges to its Eigenfactor — a weighted variant of impact factor — intended to show how much value each journal provides. PLOS ONE and other PLOS journals are among the global leaders in Mr. West’s measurement.

If anything, Mr. Solomon said, PLOS spent more than it could afford on editing and review in its early days, and it struggled to find the right pricing structure. It settled on a rate of $1,350 for PLOS ONE, which raised enough money over the past few years to make the entire PLOS family of journals profitable, he said. But after six years at that rate, PLOS ONE increased its per-article charge to $1,495 last October to cover growing expenses, said a PLOS spokesman, David Knutson.

Without statistical help of the type offered by Mr. West, identifying reputable open-access journals can sometimes pose a challenge, Mr. Solomon said. In most cases, he said, scientists understand the quality journals in their field, and the predatory alternatives simply look amateur. But “they are getting a bit more sophisticated,” he acknowledged.

In the meantime, even traditional publishers that have been cautious or even skeptical toward open access are increasingly finding reason for cheer. In addition to the booming success of its open-access Scientific Reports, the publisher Springer Nature has seen open-access submissions to the journal Nature Communications grow by nearly 200 percent since it adopted the format exclusively, in 2014.

Springer Nature published more than 56,000 articles last year in its fully open-access journals, representing more than 60 percent of the content on Nature.com, Sam Burridge, managing director for open research, said in a written statement. “This growth,” Ms. Burridge said, “is fantastic.”

Correction (1/13/2016, 10:55 a.m.): This article originally said that Scientific Reports is published by the journal Nature. It is actually published by the Nature Publishing Group. In addition, the article originally said that Mr. Solomon cited PLOS ONE in discussing the early pricing structure. He actually meant the entire PLOS organization. The article has been updated to reflect these corrections.

Paul Basken covers university research and its intersection with government policy. He can be found on Twitter @pbasken, or reached by email at paul.basken@chronicle.com.

Questions or concerns about this article? Email us or submit a letter to the editor.