Data gathering and interpretation are central to the practices of many research disciplines. And, indeed, data is itself an important output of research, reflected in the emergence of data journals, data repositories, and global data-sharing initiatives and scientific committees.
The value of open data
The idea that research data should be openly accessible and reusable in the long term has been around since the 1950s. But it has been brought into sharp focus by the rise of the internet and the web, which promise to make it relatively quick and cheap to share data freely.
The possible rewards of open data are vast, perhaps even to a greater extent than with open access to research publications. Opening up data can spur economic growth, stimulate new research, enable innovative research practices, attract investment, and build trust in research.
And the insights to be found by looking at ‘big data’ produced by researchers working in the physical and social sciences, and from other walks of life (for example, customer or user data), are stimulating fresh demand for opening up more and more datasets.
On a separate note, open data has also been shown to have a positive relationship to the quality of published research. If we are to tackle the reproducibility crisis, weed out unhelpful publication biases and boost confidence in the integrity of our research, then it’s clear that open data will have an increasingly important role to play in scholarly communications.
Like many other public funders of research, therefore, we believe that the material that arises from the research we fund should be made as widely accessible and reusable as possible. We are keen to make progress on the open data agenda as part of this.
That is why we were delighted to work with our partners on the UK Open Research Data Forum on a draft Concordat on Open Research Data. The Concordat, developed by a wide affiliation of research funders, universities, publishers and others, seeks to establish some core principles for the effective management and sharing of research data.
At the heart of these principles is a commitment to the benefits of open data, but also a recognition of the challenges. The most obvious challenge is technical. Without the proper infrastructure and processes in place, much of our research data may well end up being poorly managed, hidden from view, or lost. But curating data can be time-consuming and expensive for researchers and institutions to manage, and these costs need to be respected.
Another challenge is that of consent. There are sometimes very good reasons why data cannot be made open. Good examples are the private patient information used in medical research and sensitive information provided to social or ethnographic studies; making these open for all to see would potentially be very damaging.
In contrast, in some disciplines, like genomics, open data practices are well-developed and embedded. Indeed, in some disciplines, not making your data openly accessible would be a serious misstep. There is doubtless great potential to draw on the experiences of researchers in the more pioneering disciplines as we start to turn open data into a reality across the academic spectrum.
The final challenge I want to mention is that of motivation. Managing and curating data is time-consuming, as is making it available in a form that is easily understood and processed by users and software. It’s not always clear what the benefit is to the individual researcher doing this work, and these benefits need to be developed, explained and embedded.
The community of funders, publishers and technical specialists have been helping with this by removing barriers and introducing incentives. A number of publishers now have mandatory data access policies for all published papers. Jisc continue to do considerable work to develop the technical and cultural infrastructure to support open data, including via the Digital Curation Centre. And funders, such as EPSRC, have introduced open data expectations for their funded research.
Via the REF, we fully recognise data as an equally valid form of research output, and, through our open access policy for the next REF, we plan to reward research environments that deliver open access to a wider set of outputs than just journal articles and conference papers. More broadly, though, we remain interested to hear views on what more the REF could do to incentivise institutions and research communities to deliver open data.
In this context, the Concordat on Research Data aims to bring fresh clarity and impetus to what has been a difficult and thorny area, helping funders, academics, publishers and universities to make progress sensitively and carefully. The challenges of making open data a reality are unique and very real, but the benefits of open data are surely a prize worth fighting for.
The draft Concordat on Open Research Data is available on the Research Councils UK website. Feedback and comments should be sent to OpenData@rcuk.ac.uk by 28 September 2015.