NECTAR and the Data Asset Framework – first thoughts

KeepIt course module 1, Southampton, 19 January 2010
Tools this module: DAF, AIDA
Tags Find out more about: this module KeepIt course 1, the full KeepIt course
Presentations referred to in this blog entry Using DAF as a Data Scoping Tool, The DAF at Southampton (Slideshare)
Presentations and tutorial exercises course 1 (source files)

Three years ago, when we first conceived the principles which would underpin NECTAR, it was agreed that datasets were a bridge too far.  Instead, we would focus on the research outputs that were already in the public domain.  This was a good criterion to use – it satisfied the research community’s desire for a ‘quality’ benchmark without putting NECTAR administrators or fellow researchers in the difficult position of having to sit in judgement on colleagues’ work. So NECTAR was able to contain details of papers, presentations, books, exhibitions and artworks, but not data files.

But times have moved on.

The research environment now is different from even three years ago.  With many of the major UK funders now mandating open access to data and the REF looming, with its focus on research environment and impact, the possibility of using NECTAR to store research data clearly needs to be considered.  Which begs two questions: how do researchers currently manage their data at Northampton and what might be their data management needs in the future?

So I attended the first KeepIt training day with high expectations and a keen interest in what Sarah Jones and Harry Gibbs had to say about the design and implementation of the Data Asset Framework (DAF).  I had already had a brief exposure to DAF with DCC Digital Creation 101 Lite training a few weeks previously, but Harry’s first hand experience of implementing the tool was new and eagerly anticipated.

As already described by others in this blog, DAF is essentially a framework which enables universities to audit departmental data collections, awareness, policies and practice for data curation and preservation.  The DAF methodology comprises 4 steps:

1. Planning the audit
2. Identifying and classifying assets
3. Assessing management of data assets
4. Reporting and recommendations

where a ‘data asset’ is

numerical data, statistics, output from experimental equipment, survey results, interview transcripts, databases, images or audiovisual files (from the DAF implementation guide)

With the benefit of having been tested in a number of previous pilot studies, these steps appear to offer a straightforward and (importantly) achievable procedure for gathering information about institutional data.

Harry described the aims of the Southampton DAF project as:

• To get an overview of research data holdings
• To find out about data management practices (eg sharing; version control)
• To identify data sets for the institutional repository

Using the School of Social Sciences as an exemplar, Harry and her team first developed and promoted an online questionnaire and then conducted face to face interviews with researchers. The results were reported internally and not shared with the KeepIt course members, but the outcomes were positive:

• An encouraging response from the School which led to an even better library/School relationship
• Raised awareness of data management
• Development of data support web pages on the library website
• Further plans for DAF work.

From this it is clear to see why Debra Morris’s evaluation of the DAF tool focused on the advocacy and engagement.  Indeed, previous DAF studies have emphasised that the value of the tool lies as much in the process as in the final results.  Any mechanism which enables the repository manager (or liaison librarian) to engage meaningfully and usefully with their user community is to be encouraged.

So with this in mind, I took the DAF model to our NECTAR Steering Group.  Both Hilary Johnson (Director of Information Services) and Professor Hugh Matthews (Director of Research and Knowledge Transfer and Dean of the Graduate School) are members of this group, so support from these key people was essential to get a DAF project off the ground.  The challenge was to demonstrate why Northampton needed to undertake a DAF project, so the justification that I offered was as follows:

• Little is known centrally about university researchers’ data storage requirements, or indeed the research workflow that incorporates the creation and management of data.
• No university wide data storage policy or procedure currently exists.
• Research funders are beginning to demand that data as well as published research outputs are made openly available.
• In NECTAR, we have available the infrastructure to store and preserve digital data.
• Previous studies have noted that the process of undertaking DAF has been valuable in itself, even if the resulting inventory of data is only partial.
• There have been a number of previous implementations of DAF, these could be consulted or adapted to meet Northampton’s needs (saving time and ensuring the best possible outcome from the project).

Interestingly, but unsurprisingly, the Director of Research immediately saw the potential for the DAF in contributing to the university’s ‘research environment’.  He recognised the value of appropriately informed data management practices and policies being supported and promulgated throughout the research community.  The Director of Information Services, on the other hand, took a broader view, and stressed the need to consider all data assets – not just those produced by academic researchers.

So it looks as if, resources permitting, we will be using the DAF methodology at Northampton.  At the time of writing this post I am investigating possible sources of research support (Southampton used a research assistant to good effect), and I imagine the scope of the project will encompass both academic and support departments… but the details are for Step 1 of the project to decide.

Watch this space.