Because good research needs good data

An arts perspective: day two and three - the sixth DCC Roadshow on data management

John Murtagh from the University of the Arts London presents an arts perspective on research data management. All images in this blog come from the publication, Drawing skirts: new papercuts by Charlotte Hodes.

John Murtagh | 12 December 2011

I’m John Murtagh from the University of the Arts London, part of the JISC Managing Research Data Programme (02) 2011-13 and in collaboration with Goldsmiths (University of London), University for the Creative Arts and Glasgow School of Art in the JISC Kaptur project. Kaptur aims to discover, create and pilot a sectoral model of best practice in the management of research data in the visual arts.

I attended the recent DCC Roadshow in order to find out about effective research data management (RDM), best practice and to understand more about how the DCC can help with RDM.

Understandably the majority of presentations and discussions surrounded the generation of data in hard sciences (Nathan Cunningham of the British Antarctic Survey’s (BAS) Polar Data Centre, David Shotton of the Managing, publishing and citing data, Image Bioinformatics Group, University of Oxford, Research data management at Leicester - Jonathan Tedds, University of Leicester)

My colleague on the KAPTUR project, Tahani Nadim from Goldsmiths in her account of Day One of the DCC Roadshow noted that “data sets are becoming the new instruments of science and establishing new ways of working”.

But was our research data in the visual arts really about statistics, laboratory results, and the ability to cite these so-called data sets?

What about the practice-based research which is the foremost method of research in the creative arts?

A documentary about the making of an exhibition by a researcher at the London College of Fashion, University of the Arts London, offers a very different point of view of research methods and results.

Image from the documentary about the making of Charlotte Hodes' latest exhibition, Drawing skirts: new papercuts.

Image from the documentary about the making of Charlotte Hodes' latest exhibition, Drawing skirts: new papercuts.

At the DCC Roadshow a further viewpoint was offered by Dr Anne Alexander from University of Cambridge was based around the research of social media outlets such as Facebook, YouTube and Twitter during the Egyptian Revolution of 2011. With technological advancements in communication her presentation raised the question of how research in the arts and humanities had changed. How is this research “data” contextualised and indeed processed? She asked the rhetorical question of who actually owns this data, if for example feedback and comments were created via Facebook?

Indeed a browse of my own institution’s research repository (UAL Research Online) returned only four items described as ‘datasets’. Was the term ‘research data’ even a credible term that we could use in the creative arts?

An earlier JISC project called Curating Artistic Research Output (CAiRO) for example stated in its User Needs Report that "terminology is important (...) and phrases must (where possible) be drawn from within the arts rather than library science or similar.” The JISC Incremental Project underscored the need for jargon-free terminology, even stating that “most researchers don’t know what ‘digital curation’ is and humanities researchers don’t think of their manuscripts as ‘data’.” Careful terminology and perhaps the creation of a new phraseology would perhaps be needed.

Day Two of DCC Roadshow show was entitled: “The Research Data Challenge: Developing an Institutional Response” and involved a useful mixture of presentations, group work exercises and discussion to allow time for networking and collaboration.

One of the group exercises involved the identification of drivers and enablers for research data management. As a driver, it was interesting to note the emergence of Freedom of Information (FOI) requests for access to research data funded by universities, colleges, or publicly-funded research institutions as a factor. Despite exemptions to disclose (including information accessible by other means, information containing personal data, information subject to a duty of confidentiality, and information the release of which would prejudice legitimate commercial interests) it does raise the issue of access to data researchers needing to be prepared to provide and at short notice. A JISC advisory publication on this very subject, “Freedom of Information and research data: Questions and answers” discusses the implications.

A further driver to research data management was the need to ensure that research data was accessible for future researchers. At my own institution, which hosts The Stanley Kubrick Archive, I have had direct experience of how popular the resource is for researchers who wish to study the archives of the director of such films such as A Clockwork Orange and Dr Strangelove. This material (from location research and art department photographs to scripts in various stages with annotations, handwritten and machine typed) it is not unreasonable to suggest is the ‘research data’ of today will be the special collections of the future – something which special collections of libraries and universities should be mindful of in RDM.

Other issues that were raised during the roadshow were the training of researchers in data management. An observation was made from the floor which stated that self-funded PhD students, particularly those in the arts, were very different to the research PhD students who were funded by research funder bodies. How? Self-funded (read: arts and humanities) PhD students are not currently being instructed in research data management. Surely, Kaptur was well-placed to investigate and develop a best practice for data management in this sector?

Advice was at hand on the third day of the Roadshow: “Managing research data: tips and tools”. Led by the same team behind the JISC Incremental project, DataTrain – was a project which developed data management training modules for post-graduate courses in Archaeology and Social Anthropology at the University of Cambridge. Although these modules were not in the Creative Arts, any training in data management, especially for those subjects that have strong elements of the humanities as well for future researchers is welcome.

Most usefully of all is the module content produced by the DataTrain project is available for free under Creative Commons licence BY-NC-SA so this work could well be built upon for best practice by the JISC Kaptur project. It is also worth noting that Kaptur will also be looking at the training module provided by Project CAiRO as well.

Lastly, the DCC outlined throughout the Roadshow their entire suite of services on offer for RDM. This included CARDIO (assessing data management infrastructure at research group or departmental level), Data Asset Framework (identifying what data exists and how it is being managed and shared) but also DMPonline enabling institutions to build and edit Data Management Plans (DMP) according to the requirements stipulated by the major UK funders.

As KAPTUR is funded by JISC it was only left to me to recommend via Twitter (#UALKaptur) that we in fact practice what we preach (or are looking to preach!) and use the DMPonline in our project. The results will appear in due course on the Kaptur website over the next 16 months.

John Murtagh