Note: This blog was written by PLOS’ Iain Hrynaszkiewicz, Publisher, Open Research
Open science should be boosted in 2020 as the number of journals with research data policies increases as a result of collective action by publishers, who are being encouraged to adopt a new common framework for journal data policies. PLOS is the first publisher to align the policies of all its journals with this new framework.
To enable reproducible research we need transparent and accessible research outputs. And, as an enabler of reproducibility, transparency is what we – journals and publishers – can reasonably ask for, or even demand. Asking for transparency can begin not with new or sophisticated technology, but with a relatively low tech solution: policy. Introducing a policy on research data sharing is a logical first step to raising awareness of an issue and beginning to create change.
Policies have been in place for sharing some specific kinds of research data (such as protein structural data and genetic sequences) for a quarter of a century or more. Since 2014, PLOS has led the way with a mandatory data availability policy for all our articles, and since 2016 most larger publishers – Springer Nature, Elsevier, Wiley, Taylor & Francis – have introduced tiered data policy initiatives. Coupled with discipline-specific policy initiatives, such as in Earth sciences, this means thousands of journals now have a data policy of some kind.
Making a statement about data
An important tool, used by PLOS and others, for introducing a consistent data policy is a data availability statement in every published article. These statements indicate if, how and where the data supporting claims made in an article are available. Many journal and publisher research data policies still make data sharing and data availability statements optional rather than mandatory, but we welcome this steady progress on open research policies in the scholarly publishing community.
Requiring a new section in every article published incurs costs, which at PLOS we see as a worthwhile investment in open research. It takes time, training and resources for editors, authors, peer reviewers and editorial office staff, so mandating these statements is understandably a consideration for other publishers of thousands of articles per year.
There is growing recognition from funders, academic societies, editorial groups such as the ICMJE, that data availability statements are a practical, achievable and meaningful improvement to support transparency in research.
Wider adoption in 2020
The growth rate of journal research data policies is likely to increase with the launch, in January 2020, of the STM Association’s STM research data year, which has an objective to increase the number of journals with data policies and articles with data availability statements. Progress on policy implementation will be tracked across thousands of journals (STM is a publishing industry association with more than 120 organisational members). This feels like a turning point in open research policy.
The STM Association is recommending the use of a common policy framework for journal research data policy to promote consistent approaches to journal research data policies, at its wide variety of members.
The policy framework – published last week in a peer-reviewed journal after being available as a preprint – is an output of an initiative, begun in 2016, within the Research Data Alliance organisation. The framework includes 14 features, or common elements, of journal research data policies – including data citation, data repositories, and data peer review – and reusable policy text for journal editors and publishers to implement on their journals.
In 2019, we compared PLOS’ data availability policy to this framework and, as a first step, updated some of the language, such as to give explicit support for sharing Data Management Plans (DMPs) – a document increasingly required in funding agency data policies. In doing so, PLOS continues to lead the way, by being the first publisher, to our knowledge, to align its entire journal portfolio with this new framework. As well increasing data sharing, another anticipated benefit of harmonising policy is reducing the burden on researchers and support staff with different or conflicting requirements between journals, and funders. The framework also provides future opportunities to review data policy language to ensure requirements are easily understood.
We’re looking forward to collaborating with publishers and researchers to continue driving up standards for transparency and openness in research in 2020.
Image Credit: MyGene2.org; video frame