This post was originally published on the Jisc Research Data blog and discusses what we heard from the Jisc RDM Toolkit working group, at a meeting held in London on April 8th.
A recent survey by the European University Association showed that almost 40% of universities in their sample lacked research data management (RDM) policies and were not in the process of developing any. Nonetheless, “researchers both old and new are sharing, citing and reusing data more than ever before”, which means that the level of support that RDM professionals based at higher education institutions need to offer is also expected to increase.
The range of challenges experienced by RDM-focused staff have been explored during a productive face to face meeting of the working group members. In particular, we asked attendees what they felt were the key issues they faced in their roles.
A significant and well-known challenge in RDM is to communicate to all researchers what data is. Particularly, it has been historically difficult to reach audiences less familiar with the “data” terminology, e.g. in the humanities and some branches of social science. Various attempts have been made (including a dedicated page on our RDM toolkit website), but RDM practitioners still consider this a big obstacle.
Even when researchers know they hold data and that this should be managed appropriately, other definitional challenges arise.
During the workshop, we discussed what main forms of data have to be understood by researchers and reached the following conclusions:
- Active vs inactive;
- Shareable vs non-shareable;
- Appropriate for open licensing vs commercially sensitive;
- Personally-identifiable vs non-identifiable/anonymised; and
- Suitable for preservation vs not worth preserving in the long term.
Telling apart these forms of research data requires considerable effort from experts, so this is quite understandably an obstacle for researchers who might have never been exposed to the wide range of terms used!
The above list of forms of research data is considered as a priority, as there are clear implications in terms of compliance – not only with funder requirements, but also with copyright law, data protection and privacy (incl. GDPR) and institutional policies. One of the key challenges highlighted was that funders rarely enforce their research data management and sharing policies, which means that there is little incentive to drive behavioural change among researchers. This is not to say that data sharing is uncommon, but that, in some fields, the practice has not taken off quite as quickly as the open science movement might have hoped.
As mentioned at the top of this post, institutional RDM policies are not particularly common. Additionally, anecdotal evidence from my own conversations with researchers suggests that they are often not aware of whether their organisation has a research data policy in the first place. This indicates a communication gap within higher education institutions but is also a symptom of the fact that researchers are too often pressed for time and might prioritise reading funder requirements rather than those from their own institutions.
Some practical challenges in RDM arise from the complexity of the environment and the “people” factor. These are often intertwined and hint at the wide range of practices, stakeholders and systems involved when data is created, managed, used, shared and preserved.
In the figure above, we mapped the issues highlighted by workshop attendees to the Jisc research data lifecycle diagram. The key problems are rather fundamental, as they are reported at early stages; they are in line with the questions around the definition of “data” and the challenge of telling apart different types of data highlighted in the previous section.
The figure also shows overarching problems related to the technology and systems used to support RDM, including the need to adapt to emerging research areas that come with evolving requirements.
The “people” side of RDM tends to apply to the whole research data lifecycle and a significant concern among RDM practitioners seems to be a lack of communication:
- between university departments;
- with and between researchers (including in terms of what services are available to them);
- between funders and researchers/universities in terms of enforcement of their requirements;
- between universities and external partners (including commercial) in terms of systems used and requirements.
Clarifications are also needed in terms of touch points when it comes to RDM: productive discussions on research data are usually underpinned by well-thought out communication plans, where researchers know who to talk to and when. An example of this could be the use of project initiation meetings, where a researcher could meet RDM specialists before starting their work. However, this highlights another possible roadblock: how can researchers get access to the “right” people? This doesn’t seem to be a major challenge in practice, but it might be seen as such by researchers – perhaps a way to address this concern might be to increase the visibility and roles of RDM staff within individual universities.
The wide-ranging discussion at the RDM toolkit working group meeting helped us identify some areas of focus for future development. It further provided rich detail in terms of the challenges faced every day by practitioners. Clearly, the RDM toolkit website can only address some of the above, as it is meant as a signposting and learning resource; however, over the next few weeks, we will be considering the concerns discussed in this post and aim to implement any lessons learned.
Offer your input
The RDM toolkit has been built to be community governed, so that it is as useful for the research data community as possible – please get in touch with any comments or further thoughts for its development! If you wish to comment on the RDM toolkit’s features and current offering, please visit this page; otherwise, email us directly at firstname.lastname@example.org email@example.com.