Discussing the findings of their recent independent review of the ESRC’s research data policy, Paul Allanson, Angela Daly, Alistair Geddes, Maeve Malone, Niamh Nic Daeid, and Lucille Tetley-Brown, outline how the research funder can develop its policies in response to a changing data landscape.

How can research data policies in social science keep up with developments in open data, data practices and the legal and policy agenda in ways which are meaningful and useful? This is the question we aim to answer in our review of the ESRC’s research data policy, commissioned by the ESRC.

The ESRC’s research data policy governs data produced by ESRC-funded research, containing notably the requirement that such ESRC-funded data is deposited as a default with UK Data Service at the end of the grant, and this data is then made available, ideally on an open basis. We consider that this requirement serves a dual purpose of recognising open social science data as a public good funded by the public purse and facilitating the reproducibility and replicability of research.

The ESRC’s research data policy was last revised in 2018. Since that time, there have been several developments, including the introduction of the General Data Protection Regulation (GDPR), new and innovative uses of data in social sciences, such as synthetic data, and new data policies from other UKRI research councils, notably the MRC in 2023 and EPSRC in 2022.

If the ESRC’s policy is to continue to serve the twin goals of open social science data for the public good, and the reproducibility and replicability of research, we consider that certain changes are needed.

Our recommendations are based on the work we did to review the ESRC research data policy. Specifically, we carried out a scoping review and engaged with stakeholders involved in ESRC-funded research data via an online survey whose views we explored further in focus groups. We also analysed a sample of completed data management plans (DMPs) which are required as part of the process for applying for ESRC funding.

Our main findings comprise:

• Many of the people we spoke to considered that the current ESRC research data policy has got various things right, especially the requirement of data deposit as a default, and also found DMPs to be a useful tool.
• However, there is also the need for recognising how diverse and varied social science data is across methods and disciplines.
• There is also uncertainty around how the research community can best produce open research data and under-acknowledgement of data that cannot be made fully open, as well as insufficient financial support for preparing data for sharing.

We make a series of recommendations designed to strengthen the ESRC research data policy and associated practices across four key areas:

• Aims and guiding principles for an updated ESRC research data policy
• Means for implementing policy developments
• Scope for aligning policies across UKRI and ESRC
• Points requiring further work and consideration

Alongside other recommendations, we advocate that the ESRC keeps the data deposit by default mandate, while better acknowledging the ‘tiers of sensitivity’ of data. This would address concerns that were raised by some participants that the policy lacked relevance for data that could not be made fully open due to privacy and confidentiality motives. This would also help move beyond perceptions of the open-closed binary for data, which in practice is more of a spectrum between openness and secured. It would also encourage researchers to think more broadly and creatively about what data or data-related materials can be shared.

We also recommend that data management plans (DMPs) are retained as part of the application process, and should be longer than the 500-word text box that is currently part of the UKRI Funding Service. We offer some detailed recommendations on how DMPs could be enhanced to become ‘living’ documents to cover the entire data lifecycle and thereby be a more useful tool for researchers and data management and support staff. These documents could be built on the current Digital Curation Centre’s DMPonline tool, in line with the Research Data Alliance DMP Common Standard.

There is also a need for further cross-disciplinary research into the legal and ethical aspects of combining datasets. In particular we suggest the ESRC launches a consultation on the definition of ‘data’ and how and where to deposit software code associated with research data, the latter of which echoes a similar recommendation in the recent SSI report from 2023.

Achieving maximum coherence among UKRI research council data policies, and recognising that social science research and data are not just funded by the ESRC, but also by other UKRI research councils is also important. Therefore, beyond our immediate recommendations, we consider whether in future there should be a move towards different research policies for different kinds of data or contexts, rather than different policies from different research councils.

It was a pleasure to undertake this review, and we thank all those with whom we engaged and who helped us understand better the social science research data landscape across the different ESRC disciples, methods and forms of data production and use. Overall, there is strong interest from ESRC constituents in making research data in its many and updated forms available for use by others in appropriate ways in order to realise its value as a public good and as a foundation for open (social) science. We hope our report, findings and recommendations in seeking to achieve these goals in changing times will prove useful recompense for all who assisted us.

This post draws on the authors’ post, Doing ESRC Data Better: A study for the Economic and Social Research Council (ESRC) — Discovery – the University of Dundee Research Portal.

The content generated on this blog is for information purposes only. This Article gives the views and opinions of the authors and does not reflect the views and opinions of the Impact of Social Science blog (the blog), nor of the London School of Economics and Political Science. Please review our comments policy if you have any concerns on posting a comment below.

Image Credit: Anton Grabolle, Better Images of AI, AI Architecture (CC-BY 4.0).