Supporting universities to implement effective research data management

By the end of this month research organisations in receipt of funding from the Engineering and Physical Sciences Research Council (EPSRC) must be able to comply with the policy framework on research data management (RDM) that the funder has set out, writes Verena Weigert of Jisc.

And, as we now know, after this deadline PVCs for Research will be asked to fill in a short self-assessment survey to report on how their institutions are doing. Then, from the autumn term, the research funder will be ‘dipstick’ checking by randomly sampling the availability of data that underpins research in papers published on or after 1 May 2015.

 

Not surprisingly, then, many research institutions are focusing a lot of effort on getting ready for the 1 May deadline. The EPSRC has made it clear that it is the responsibility of qualifying institutions to support their academics to deal with research data effectively. Institutions need to promote best practice, provide support services and tools and systems that will enable researchers to meet the EPSRC’s required standards.

 

And while the policy relates specifically to data from EPSRC-funded research, it is having a much wider impact. It has stimulated institutions to look again at the research data lifecycle and to develop RDM services in a more holistic way - not specifically for compliance with the EPSRC’s expectations.

 

Challenges

 

Some people have found the EPSRC guidelines themselves to be disconcertingly non-prescriptive but the flexibility is deliberate, allowing for the different kinds and volumes of research projects that institutions are involved with, as well as the many different types of data that their projects generate. The main challenge for institutions is perhaps the lack of certainty about how much it will all cost and how to sustain resources in the long term. Institutions, for example, need to find a solution that will ensure that EPSRC-funded research data is kept for at least a decade from the last date of access by a third party. The difficulty here is to estimate how much data there will be and how more storage to budget for.

 

It should also be mentioned that not all data can be kept forever; it is up to researchers to decide what to keep and what to dispose of but they may need support to make this decision. In the case of published research, a decision must be taken that preserves whatever people would need if they were seeking to check the underpinning evidence used to support the findings. If the research data does not support published research findings then researchers should make their judgement about what to archive based on the likely future value of the data.

 

There’s a strong emphasis on getting the metadata right too, to enable other researchers to find the data. After all, the main purpose of the policy is to support independent validation of research and reuse of the data, to support good research practice and to generate maximum public benefit from the initial investment.

 

Progress

 

From speaking to a number of RDM support staff in the last few months we found that universities are rising to the challenge and that numerous solutions are being developed. Many would agree that the EPSRC’s expectations have really pushed HEIs over the past four years to develop and implement RDM policies and strategies. Some welcome this top-down approach, which has certainly helped to make the case for investment and also encouraged institutions to see research data as an asset that must be managed and sensibly exploited. While the EPSRC policy has served as an important catalyst, dealing with research data is - of course - not only about complying with policy. More importantly it is, at heart, part of the research process itself and RDM services should principally reflect the needs and motivations of the research community.

 

Guide

 

While universities have been working towards the EPSRC’s 1 May deadline, there is an expectation that service development and refinement will continue after this date. Depending on their organisational structure and existing infrastructure, institutions may have also focused on particular areas to start with. Jisc has published ‘Steps that HEIs can take to meet the EPSRC research data policy’, designed to offer a set of practical steps, provide concrete examples of what is working well for other universities, and signpost readers to useful resources. It covers several topics including policy, strategy and sustainability, support services to improve RDM capabilities and skills and also the kinds of technical infrastructure/services that universities will need to support data preservation and sharing.

 

Alongside these resources we have produced a report, 'Directions for Research Data Management in UK Universities' with colleagues at the Association of Research Managers and Administrators (ARMA), Research Libraries UK (RLUK), the Russell Universities Group of IT Directors (RUGIT), the Society of College, National and University Libraries (SCONUL) and the Universities and Colleges Information Systems Association (UCISA). It explores several strategies to ensure that, over the next five years, effective RDM becomes fully embedded within the research practices of the UK’s research institutions. One of these strategies is incentivisation and it describes why meaningful incentives for researchers are currently lacking before going on to show how they could play a defining role.

 

Incentives

 

Universities are working hard to develop RDM services and robust data archiving solutions but even so, demand from researchers is limited. Although ‘sticks’ such as funder mandates are increasingly a fact of life for researchers many are still reluctant to offer up research data in ways in which it can be cited and re-used, especially in some academic fields. The need for compliance alone will not result in researchers embracing RDM willingly. They will also need meaningful rewards either through mechanisms such as the REF, or through career progression within their institution. Ultimately, effective RDM should make it easier for researchers to do their work.

 

Longer term, universities need to find sustainable and efficient RDM solutions and through Research at Risk we are developing shared infrastructure services and advice to meet these needs.

 

Verena Weigert is a senior technology manager at Jisc