A poster prepared for the JISC MRD Programme meeting in Nottingham at the start of December (and then also displayed at 7th IDCC in Bristol the following week) is now available via our Hydra repository.
As part of our project plan we provided our thoughts on the anticipated impact of the project’s work. Now sometime into the project we reflect on the degree to which the originally anticipated impact is still the case, based on evidence gathered from our interviews with academics.
- Data management as a support tool for teaching – The University Strategic Plan is strong on the value of research-led teaching and on identifying ways in which this can be embedded. The explicit use of data within teaching was not immediately apparent in our interviews, although there are good examples of how it has been used in specific cases. We would, though, continue to anticipate that by making data better managed, and thereby hopefully more accessible, it will be more available for use in teaching.
- Data management as a stimulus to interdisciplinary research – Another major theme of our Strategic Plan is interdisciplinary research. There is already evidence that datasets have been used across disciplines within the University, and, again, we would hope that making the data more accessible will foster more of this. Interestingly, one of the main areas of past activity, the HMAP project, originated as an interdisciplinary undertaking, combining scientific investigation of marine populations with historical examination of what these were in the past. This highlighted the value that datasets generated for one purpose can have use in very different areas as well.
- Data management as a means of departmental research coordination – Much research in history can be solitary, albeit with notable exceptions (e.g., at Hull, the Slave Voyages work in collaboration with Emory University and others). With different areas of research being undertaken across the Department, it is not always clear what is being generated by whom. Data management planning is in part intended to address this, so that the Department has a clear appreciation of research data outputs, and can thereby stimulate interaction between those who may be in a position to collaborate on future analysis.
- Data management, locally provided – We already manage some datasets within our Hydra digital repository (e.g., the Domesday book, as well as the HMAP data mentioned above). However, we are conscious that the means for depositing such datasets is not as developed as it might be, and ease of how to manage datasets has emerged as a big area of concern, but also interest, when discussing data management with academics. By enhancing our local repository provision, and also roadtesting additional capabilities within our VLE and using linked data, we hope to provide good local technical capability for addressing researcher needs. We do not see this as replacing national or subject-based alternatives, but as a complement to them.
In conversations, other areas of anticipated impact have also emerged. Whilst the focus on the impacts above is largely on departmental and local institutional impact, two other themes have come to light:
- The value of data management in promoting and sharing research. There are some clear differences here between members of staff (some very willing, some very reluctant), but those who are willing will be interesting examples to others of what might be enabled through such an approach.
- The role of a data management plan for other disciplines within the University. Noting the interdisciplinary aim mentioned above, the idea of generating a DMP that is adaptable for other disciplines with which there is no obvious link has taken greater shape. We don’t yet know if this is feasible or practicable, but we shall have it at the back of our minds as we proceed.
Over halfway through our project now, and a chance to reflect on the work we have done so far and how we came to be here. An initial reflection is: why, and how, did we in the Library come to be working with research data in the Department of History? There are a number of reasons why this came about, which can be split into two categories: disciplinary reasons and institutional reasons.
- Disciplinary reasons – History is all about data. It is all about unearthing, gathering, interpreting and presenting data from the past. Hence, there is a need to manage this data so as to be able to effectively draw conclusions and substantiate those conclusions. History also has connections with other disciplines, which can have their own historical aspects. The Department of History at Hull encompasses our courses and research in archaeology. There are also links to Geography and English in some of the research areas being examined. As we have found within the History DMP project, the range of data types is varied, in part because of these interdisciplinary connections, and understanding data management planning within History should thus give us an understanding of how data management can benefit other disciplines as well.
- Institutional reasons – Getting started with data management is one of the questions that has arisen at times within the JISC MRD programme. Where to start? Clearly, data management has been happening for some considerable time already in general, and it was one example of previous experience and good practice within the HMAP project that led us to initially discussing the idea of data management planning with the Department. From this seed we hope many flowers will bloom. The HMAP project is also a key piece of work within the maritime sphere, and International Maritime is now one of the main strategic themes for the University. The theme also carries with it the same interdisciplinary flavour that makes the investigation of data management of interest, as described above.
The Library has been able to pursue data management as part of a key strategic aim of its own – promoting the University’s intellectual capital – but has only been able to do this because of the partnership that has been possible through the willingness of the Department to get involved. How we anticipate the impact of this work together will be the topic of a subsequent blog post.
We are pleased to announce the release of our first formal deliverable, a short report setting out ‘The case for a History Data Management Plan’.
This is a synthesis of the findings from the History DMP Project Team’s interviews with academics in the Department of History. The information gathered has been used to inform this short report on the need for a data management plan and the role of data management in supporting History research.
The report can be downloaded from the University of Hull’s digital repository.
Happy New Year, all.
As we in Hull start the New Year and move into the final phases of our short History DMP project we are turning our attention firmly towards its two intended major outcomes: a document that will allow academics to plan for suitable management of their data, and enhanced technological support for making the data available to others should that be appropriate.
Our Project Proposal and the subsequent Plan indicated that we would base our document on the well-known DCC ‘Checklist for a Data Management Plan’, and we shall. However, it has become clear to us (if we hadn’t realised it already) that in its basic form it is unsuited to use by the average academic. We intend to use the DCC checklist as the basis for a document that is more directly relevant to (in this case) history research and which not only poses questions but also offers some answers – perhaps in the sense of ‘tick boxes’ and the like but also by pointing users to people locally who might offer answers or advice and services. The aim is to make our document obviously relevant and useful rather than an unnecessary chore that has to be dealt with.
Technological support will take two forms. Most obviously we want to ensure that research data stored in the University’s digital repository has appropriate metadata (in the librarian/IT sense, facilitating search and discovery as well as allowing appropriate citation etc) and that it is in at least one format suited to preservation. This last may mean converting (say) an Excel spreadsheet into a csv or tab-delimited file as we ingest it and storing both copies. Naturally, we will also store metadata in the academic sense – documentation about the datasets. The second form of technological support that we want to investigate is linked data. We are going to take one of our existing datasets and see what we can usefully do with it. This will allow us to evaluate both the effort involved in such exposure and the possible benefits of doing so.
It’s going to be a busy few months!
Much of November has been spent interviewing academics in Hull’s history department. The purpose of each interview has been several-fold, including: to find out what datasets they have been/are/will be involved with; to discover the types of data these encompass; to learn how they look after such data in the short and long term; and to ask if they would be prepared to help us create a checklist so that best practice can be shared and used to shape future data management.
On the whole, the response has been extremely positive. A few of our colleagues have been defensive about giving others access to their data and we have had to reassure them that ‘managing’ the data does not necessarily mean sharing it with the public (and thus, importantly, rival researchers!).
Our interviews have uncovered a range of management practices ranging from the carefully thought out and commendable to the, frankly, risky. There has been a general consensus that a checklist reflecting best practice, and targeted at historians, would be useful – especially if it contained pointers to local sources of advice on specific points. Time and again we have heard comments along the lines of “I would have liked advice but didn’t know who to ask.”
We are starting to turn our attention to constructing the first draft of a new checklist taking on board the points raised in these interviews. In parallel we are starting to think about technology solutions, both existing and new, including those integrated with the University’s digital repository, with which to support it – the repository already has somewhat basic support for datasets, we are beginning to understand how that might usefully be enhanced.
We have realised, too, that whilst our primary focus might be on digitised/digital datasets, there is a clear need for the checklist also to cover appropriate management of paper-based data.
The full version of the History DMP Project Plan is now available (16 November 2011) from Hull’s digital repository, Hydra, at https://hydra.hull.ac.uk/resources/hull:4930