Skip to Main Content
Skip Nav Destination

Study quantifies the growing traction of open access

18 April 2019

Spurred by the policies of funding agencies, academics are getting much better at posting their peer-reviewed research in freely available online repositories.

Open access buttons
Credit: h_pampel, CC BY-SA 2.0

Five years ago, the UK funding body Research England, then known as the Higher Education Funding Council for England, announced an ambitious policy designed to speed up the transition to open-access publishing. To become eligible for a slice of billions of pounds of government money distributed to UK universities, academics would have to post their research on free-to-access websites such as preprint servers and institutional repositories within three months of acceptance by a journal. Though the policy took effect in 2016, it has been difficult to judge its efficacy.

Now an analysis shows that researchers in the UK are indeed posting their papers online earlier, as are their colleagues all over the world. The time researchers are taking to post papers online shrunk by an average of 472 days per country between 2013 and 2017, finds a study published on 17 April and to be presented at the ACM/IEEE Joint Conference on Digital Libraries in June. Though the authors can’t definitively say what’s behind the trend, they suggest that the Research England policy and other funding eligibility requirements recently announced worldwide are pushing academics to rapidly make their work freely available.

Drahomira Herrmannova, Nancy Pontika, and Petr Knoth at the Open University in Milton Keynes, UK, evaluated the time it took for academics to deposit some 800 000 papers—including more than 90 000 from the fields of physics and astronomy—in repositories beginning in 2013. They compared the date on which each study was posted to a repository with the journal publication date. The bibliometric data came from CORE, a website that aggregates data about scholarly publications from online repositories, including preprint servers such as arXiv and institutional servers like White Rose Research Online, a collection of work from the UK’s Universities of Leeds, Sheffield, and York.

The authors found that in 2017, researchers took an average of 135 days following publication to deposit their papers in an online repository. Since 2016, UK-based scientists have been posting their papers online more quickly than those in the other four nations with the highest number of papers in the dataset: the US, the Netherlands, Italy, and Switzerland. All five countries have rapidly decreasing deposition times, the study suggests, with Italy making the largest improvement, from an average of 706 days in 2013 to 48 days in 2018.

In 2014 “there was a big change in behavior, and suddenly the deposit time lag starts going down,” says Herrmannova. The shift is less pronounced for physicists and astronomers, but that’s because many of them were already religiously posting to arXiv.

Depositing work in open-access repositories
Academics from five countries with high research output have been taking less and less time to post their journal papers in open-access repositories. Source: D. Herrmannova, N. Pontika, P. Knoth, Open University (2019); created with Flourish

The authors speculate that researchers have been motivated to more quickly post their work by new funding policies. By following the 2016 Research England policy, UK researchers make themselves eligible for Research Excellence Framework 2021, a system designed to evaluate the quality of research to guide government funding. In Italy, legislation passed in 2013 requires that all research receiving at least 50% of its funding from public sources be made freely available. Other policy shifts toward open access are reflected in the US Public Access Plan, which was introduced in 2013, and Europe’s Plan S, due to take effect in 2020.

It’s surprising that open access didn’t take over sooner, considering the financial burdens placed on universities by the cost of journal subscriptions, says Tom McLeish, a polymer physicist at the University of York, who was not involved in the work. In February the University of California system announced that it had canceled its subscription contract with Elsevier.

Daniel Himmelstein, a data scientist at the University of Pennsylvania, says it’s possible the new study has a large selection bias, since it doesn’t consider papers that weren’t deposited at all.

The authors found many repositories where more than 90% of the material had been posted within three months of publication in a journal, but they also found many sites with less than 50% compliance with the Research England policy. (Because of data limitations, the authors defined compliance as posting within three months of the publication date rather than the acceptance date.) It’s up to institutions to remind their staff to make sure their studies are compliant, Knoth says. “This is not a game of medicine versus physics or mathematics,” he notes. “This is a game of institutional policies and practices.”

The study’s finding that institutional culture seems to affect how quickly researchers deposit papers is a novel one, says Vincent Larivière, an information scientist at the University of Montreal. Last year, Larivière and Cassidy Sugimoto of Indiana University Bloomington reported that up to one-third of the 1.3 million papers subject to open-access mandates from 12 major funders were not free to read. Funder policies that allowed depositing to be delayed had lower compliance rates than those that required immediate deposit.

Editor’s note, 7 May: The article and graph were corrected to state that Switzerland, not China, was among the five countries analyzed in the study.

Close Modal

or Create an Account

Close Modal
Close Modal