Demystifying research data

20 July, 2012
Michael Day

Earlier this week, I attended a workshop entitled 'Demystifying research data: don't be scared, be prepared', an event that was organised by the JIBS User Group and RLUK (Research Libraries UK), and held at the Brunei Gallery, SOAS (School of Oriental and African Studies), University of London.

The event was intended to provide a general introduction to research data management (RDM) for subject/liaison librarians, and others responsible for providing support to academic departments.

The event preamble suggested that while most attendees would currently have little or no experience of RDM, they were likely to become increasingly involved in supporting RDM activities as institutions begin to develop their policies and services further.

Following a welcome and introduction by the event Chair, June Hedges of University College London, my first task was to give the opening presentation, a general introduction to research data management. My brief was to explain why RDM was important, to identify who needed to be involved from an institutional perspective, and to provide some background on the concept of data management life cycles. 

It will be for others to judge whether I succeeded in this (the presentation slides are available on Slideshare) but, in retrospect, I feel that it might have been useful if I had provided examples of the wide range of content types that can be considered under the general heading of research data.

In my introduction, I tried to explain some of the different motivations that were driving researchers, institutions and funding bodies to take RDM increasingly seriously, including the recent (and influential) EPSRC Policy Framework on Research Data.

Given the subtitle of the event, I tried to be encouraging, concluding that institutions needed to approach the RDM challenge incrementally; e.g., building first on areas where libraries had existing knowledge and skills.

I was followed by Liz Holliday, who introduced the University of the West of England's Managing Research Data project, a JISC-funded pilot that was working with seven projects in the health and life sciences.

Holliday argued that some aspects of RDM built upon the existing skills of librarians - e.g., with regard to things like metadata or in running training for researchers - while acknowledging the need for libraries to work collaboratively with researchers and other stakeholders within institutions.

Breakout groups before lunch began to consider where the library might best fit within RDM lifecycles.

In the reporting back, most of the groups mentioned libraries' existing expertise in things like resource description and metadata, collection development, and preservation.

Newer roles might include, for example, the identification and analysis of institutional data collections or the articulation of institutional RDM requirements (e.g. using DCC tools like DAF and CARDIO).

It was recognised, however, that success would be dependent upon good collaboration with researchers. 

Three more presentations followed after lunch, all of them introducing projects funded under JISC's Managing Research Data programme.

First, Rachel Proudfoot provided an outline of the University of Leeds's RoaDMaP project, where the university library has been working with other institutional stakeholders in order to develop a RDM policy and to assess institutional requirements with regard to different stages of the grant funding cycle.

On developing metadata infrastructures, Proudfoot asked whether it might be better to build on existing systems or to create new ones, noting that Leeds would be trialling Oxford's DataFlow.

She also acknowledged the that the findings of the Incremental project - a JISC-funded collaboration of the universities of Cambridge and Glasgow - had been very helpful in making Leeds feel much less scared about tackling RDM. 

In the following presentation, Carmen O'Dell and Barbara Sen of the University of Sheffield introduced the new RDMRose project, a collaboration of the universities of Sheffield, Leeds and York that will be developing training materials to support CPD as well as a dedicated RDM learning module for Sheffield's iSchool. 

The final presentation was an overview of the JISC DataPool project by Dorothy Byatt of the University of Southampton.

DataPool was a follow-up to Southampton's earlier Institutional Data Management Blueprint (IDMB) project, and was concerned with embedding good RDM practice within the institution, e.g. through the provision of guidelines and other support materials linked to the Web version of the university's Research Data Management Policy.

A second breakout session considered what new roles librarians might have in supporting RDM and the skills that would be needed to fulfil these.

The group feedback suggested that support staff like subject librarians would need a far more detailed understanding of researchers' practices and workflows, although it was not entirely clear how collaboration might actually work in practice (in part this was a matter of professional confidence).

In addition, there would definitely be a need for the deeper propagation of RDM knowledge and skills and there would be an ongoing need for additional training and CPD. 

It seems to me that the DCC may have a useful role in supporting this, as will projects like RDMRose.

The day was very encouraging from a number of perspectives.

The participants demonstrated a very keen interest in RDM and were extremely positive about the need for university libraries to roll their sleeves up and get involved with the development of policies and infrastructures.

The fact that the event itself had been oversubscribed perhaps also demonstrates a wider hunger for RDM knowledge from subject/liaison librarians in UK research libraries.