Because good research needs good data

IDCC15 session A2: Curation Infrastructure, Education and Training

A report on session A2: 'Curation Infrastructure, Education and Training' at the International Digital Curation Conference, London, on Tuesday 10 February 2015.  Session A2 provided two analyses of the current priorities and workflows of researchers as they go about their research acti...

Laura Molloy | 18 February 2015

Session A2 provided two analyses of the current priorities and workflows of researchers as they go about their research activity. In both cases, the findings have potential use in improving institutional support for digital curation and research data management.

Our first paper was ‘Assessing perceived usability of the Data Curation Profiles Toolkit using the Technology Acceptance Model’, presented by two of the authors: Tao Zhang and D. Scott Brandt of Purdue University. The Data Curation Profiles Toolkit (DCPT) - demonstrated earlier in the conference - is a useful method for librarians to engage researchers in discussions about research data.  Information about a dataset across the lifecycle is captured via a structured interview. The directory contains many examples of the kind of report that results from the interview process, and there are examples from disparate disciplines such as history, genetics, linguistics and astronomy.

The team at Purdue has been assiduous in developing the Toolkit and seeking to identify any opportunities for improvement in uptake and use.  Whilst the Toolkit appears elegantly simple, it is in fact attempting to achieve a complex and important set of closely-related aims: to introduce and promulgate the topic of RDM in research-performing institutions; to perform advocacy in the librarian and researcher populations; to move towards a more cohesive way for librarians and researchers at a given institution to talk about RDM using a shared language, agreed through the interview contact; to increase librarians’ comfort in discussing research data issues and to improve confidence in doing so; to pull together a body of knowledge at the institution (and potentially beyond) about who is creating data, when, what form that data takes and its size and extent; to clarify the rights holders; to set out use of standards and tools; and to anticipate the RDM and digital preservation needs of the research. 

These are important points for institutions and research groups to know and record. In addition (and resonating with elements of the Oxford Brookes researcher interviews) there are questions that address the critical emotional aspects of research work. Here, this is approached via questions about the researcher’s hopes and aspirations for the use of the data, their attitude towards discoverability and the anticipated value of the data.

Tao Zhang undertook a formal and structured assessment of the DCPT using the Technology Acceptance Model (see slide 6 for an explanation and reference) to reveal factors affecting user perception of the tool, intention to use and any difficulties and areas where the Toolkit could be improved.  Unsurprisingly, the work found that the richness of information about each single project takes considerable time to achieve through the interview method and this may inhibit use of the Toolkit in some situations. However, users were generally aware of the value of the rich information yielded by the DCPT and were in favour of engaging with training and guidance to assist use.         

The slides for this talk are available here: ‘Assessing perceived usability of the Data Curation Profiles Toolkit using the Technology Acceptance Model’, and you can also read the full paper here.

Another piece of work that cloaked complexity in elegant simplicity was presented in this session. Daisy Abbott - formerly of the late, lamented AHDS Performing Arts and, more briefly, the DCC - is now Research Developer at the Digital Design Studio, Glasgow School of Art. Daisy has been investigating the role of the research supervisor in the digital curation of research data. Noting that there is something of an implied expectation that research supervisors can provide advice and guidance on RDM via the supervision process, she asks whether this is a realistic expectation. At this point in RDM maturity in UK HEIs, this is a timely and important inquiry and I was delighted to see Daisy’s characteristically clear and forthright approach being applied to the question. A survey was circulated, receiving 116 responses. Respondents were categorised into four groups: students in the early phase of their postgraduate work; students in the late phase of their postgraduate work; doctorate holders who don’t supervise; and research supervisors (it is implied that the last of these four are also holders of doctorates). The postgraduate degrees being undertaken here are generally PhDs but the questions and findings are likely to also be relevant to other types of postgraduate research degree.

Daisy found that – perhaps unsurprisingly – a majority of respondents believed research data management activities were important but that many had little or no prior RDM knowledge. The most familiar RDM activity identified was, again perhaps unsurprisingly, backing up data. 

Considering the relevant parties to be the student, the supervisor and the institution, Daisy anticipated that respondents would generally allocate RDM responsibility elsewhere. However, she was surprised to discover that respondents from both categories of students and from the supervisor group frequently indicated they considered themselves responsible for ensuring RDM was correctly performed for a given piece of postgraduate research. The supervisors broadly thought their students were mostly responsible for RDM, but supervisors gave themselves a larger share of responsibility than their students did.

There was a consistently low rate of recognition of the institutional responsibility for RDM, a finding that may either have discouraged or relieved institutional RDM staff, depending on their current capacity, but certainly raises the issue of how clearly drivers such as the RCUK principles are being understood in the PGR community and their supervisors. Institutional RDM staff may also be interested to note that the awareness of RDM guidance and support appeared to be higher the closer its source was to the respondent. For example, an institutional policy may have a better chance of being noticed than UK-wide legislation, which suggests to me the crucial role of RDM staff in translating wider policy into the local context. Daisy’s work also suggests that whether or not supervisors need to personally be competent in RDM practice, it is critical that they firstly understand there is a set of skills and knowledge to be apprehended, and secondly that they know where to direct students for guidance and support in RDM. There is more information in Daisy’s slides: 'Digital curation and doctoral research: Current practice'. 

For the full detail and context of both of these papers, please read them freely by visiting the International Journal of Digital Curation at:

What would be helpful for embedding good RDM practice into existing researcher workflows?  We’d love to hear your thoughts in the comments below.