Within the research sector it is now widely recognised that open data is a good thing and beneficial to research and its subsequent impacts and uses. It improves the integrity of research by making it more transparent, open to scrutiny and therefore trustworthy, while supporting reuse, new advances, and contributing to economic growth.
But while funders and universities have their own policies, and in some cases mandates, in place for open research data, what hasn’t been so apparent is that these key stakeholders are working together towards shared goals and are aligned.
The Concordat on Open Research Data expresses these shared principles. Working on it collaboratively has been a significant step towards key stakeholders both being on the same page and also demonstrating their shared aspiration for open research data by default.
The publication of the concordat today will support our sector to achieve its vision of research data being accessible and re-useable. It makes clear that, while all parties and disciplines should act with a view to open by default, existing legal, ethical, disciplinary and regulatory frameworks need to be respected, as do due costs.
I’m going to use this blog to explain the thinking behind the concordat, some of its key principles and resultant activity.
Starting the journey
A decade ago open data was the preserve of a few pioneers. Their advocacy – and a growing acceptance of the need for better, more reproducible, research – have led to major funders, research institutes and publishers mandating data sharing. Whilst by no means universal, open research data is now a part of mainstream conversations across academia, though disciplines do vary in terminology, incentives and practice.
However, this variability makes it difficult for institutions and others to reliably support open data practice. So, many months ago the UK Government, via the Royal Society, brought together the UK Open Research Data Forum a group of organisations, each with an interest in research data, in a bid to move the open data agenda forward. The forum has a wide set of stakeholders, from funders, researchers, learned societies, publishers, libraries, standards bodies and organisations like Jisc and the Open Data Institute. It was decided to develop a concordat to articulate commonalities of aspiration and the shared nature of the endeavour.
Concordat working group
A working group then set to work on the concordat, including the Arts and Humanities Research Council (AHRC), Natural Environment Research Council (NERC), Research Councils UK (RCUK), Newcastle University, Universities UK (UUK), Higher Education Funding Council for England (HEFCE), University of Warwick, Research Information Network, Springer Nature, British Library, Wellcome Trust, University of Essex, The Russell Group and Jisc.
The thinking behind this was so: data policies have similar goals but vary substantially in the specifics of implementation. While we all agreed on the significant benefits of opening up research data, particularly with regards to reproducibility of research and demonstrating research integrity to help prevent scientific fraud, the way in which these goals are to be achieved urgently needed to be aligned.
Likewise, the known open research data issues – for example, ethical, privacy and commercial constraints stopping certain data from being made open, as well as the logistics of managing data over its lifetime due to the sheer wealth of data that would need to be made available – needed better co-ordination to help unify, so work to address open research data can be taken forward within a shared framework. A concordat, agreed to by all stakeholders, could help to overcome this by endorsing a set of shared principles that bring together some of the disciplinary discrepancies, and stakeholder concerns.
Setting out our stall
Working together our group put forward a set of key principles in summer 2015, to be considered by the research community. We wanted to hear from stakeholders about what they wanted to see in a concordat, feedback on the proposals as well as the value they thought it would have for the sector.
This exercise collected around 80 responses including higher education organisations, learned societies, publishers, infrastructure providers, and others engaged in research. Feedback was informative: all wished to achieve the goal of open research data, but stressed the need to respect some of the challenges relating to funding and established disciplinary practices.
Our resultant concordat, therefore, sets out ten clear and practical principles, not meant to mandate specific activities, but to instead outline what good practice looks like, in order to assist and inspire.
Happily, the concordat has been well - received, with HEFCE, RCUK, Wellcome Trust and UUK the first signatories – giving us good ground from which to build. These key components of the UK research landscape send a powerful message to their constituents – open data as good research practice can be achieved through their leadership without significant negative implications for researchers, research managers, institutions and sector infrastructure.
As a group we hope to announce the secondary signatories in the coming months and actively invite others to sign up by contacting firstname.lastname@example.org.
Devil is in the detail
I’m sure that you’ll want to take the time to properly read the concordat yourself, and I’d urge you to do so. I also wanted to offer some further insight into three of the principles around managing and sharing data, and how Jisc is helping. These are outlined below:
- Open access to research data carries a significant cost, which should be respected by all parties
- Good data management is fundamental to all stages of the research process and should be established at the outset
- Data curation is vital to make data useful for others and for long-term preservation of data
Repeatedly I hear about management of data being one of the biggest barriers to making it accessible and thereby implementing the open research data agenda. The three principals I mentioned take up this theme, drawing out the cost in terms of IT and administrative infrastructure, skill and obligation in making data open; the subsequent use of that infrastructure plus any other chosen repository to collect, store, manage and use data throughout its lifecycle; and to support the on-going preservation of said data, including in keeping within the current Engineering and Physical Sciences Research Council (EPSRC) funding requirements that this be held for ten years after last access.
Development of a research data shared service
To support these complex needs we at Jisc are undertaking a major activity, in partnership with 17 higher education organisations, to develop a research data shared service, which will make a significant contribution to help address some of the headaches posed.
The service we’re designing and procuring will support the research data sharing lifecycle, from deposit and access, through to publication and long-term preservation – and supporting the development of management information to ensure that the decisions about research data curation are made in an evidence-driven way.
Stay up to date
We’ll be sharing more information about our research data shared service on this blog shortly – watch this space.
You can also visit our research data project blog for more regular updates.