Event report: “Dealing with Data – what’s the role for the library?”

Sorry for the lack of posts to this blog, loyal readers. I intend to write something about the work we’re doing on staff profiles here at City soon.

In the meantime, here’s a write-up of an event I attended in Ghent, Belgium (which, by the way, is a lovely little city). Entitled “Dealing with Data – what’s the role for the library?” it was on that perennial hot topic, research data management (RDM). I found two things about the event particularly useful: its focus on how library services (as distinct from all the other parts of universities that have a hand in RDM) can support RDM, and the presence of real live researchers who talked about their experiences with RDM and what library services can do to help. What follows is a summary of a couple of the sessions I found particularly useful, and a lesson I think can be taken from the workshop that I hope to be able to apply here at City.

After a couple of opening sessions that set the scene on RDM in the context of open science and what we mean when we talk about RDM (a slippery subject) came for me what was the best session of the morning, “Meet the Scientists”. Often meetings about RDM (and also open access) limit themselves to librarians and other information professionals talking amongst themselves, without input from those people we are actually supposed to be helping, researchers. Therefore hearing from working scientists was really refreshing, particularly since the scientists came from three disparate research areas (digital humanities, biology and market research) each with their own challenges. A number of issues were identified:

• In the digital humanities, there is a problem that data can be heterogeneous, and this can be a barrier for researchers sharing (“no one else will be interested in my research data”), though this fact can often be used as an excuse not to share.
• In biology, the GenBank repository for genetic datasets is very well established, but researchers do not share experimental data. There are perceived issues with data quality, but also issues with accreditation when sharing (“will people who re-use my data give me suitable acknowledgement?”). The problem of accreditation came up a number of times, and it’s notable that journal publishers are now evolving standards to try to manage citing datasets.
• In market research, data modelling allows for natural experiments to take place, but this can also require versioning of datasets, a potentially formidable task for universities looking at RDM.

The panel suggested a number of ways that libraries can help researchers with RDM:

• Encourage researchers to deposit their data.
• Offer advice to researchers about where they might deposit data, whether this be locally in an institutional repository, in a discipline repository or in a generic repository. Don’t be afraid to offer this advice- researchers will be grateful!
• Think small: running small projects with groups of researchers is a good way to develop services. Don’t bite of more than you can chew, but doing something is always better than doing nothing!
• Think about workflows that created the data in the first place and how these might be preserved.
• Think about providing machine access to data where possible via APIs.

So a lot to be thinking about!

The other session I wanted to highlight was on “Actively Supporting Data Management: Learning From the Approach of Three Research Institutions”. The session covered how three library services had gone about awareness raising about RDM with librarian colleagues. Dr Andrew Cox of the University of Sheffield’s Information School identified barriers to entry for librarians when first thinking about RDM:

1. Librarians are already over-burdened with roles, and RDM is yet another thing to worry about.
2. Librarians don’t often have first-hand experience of the research process and so can feel unfamiliar with some of the issues relating to RDM.
3. Current library skills don’t always translate to an RDM context.
4. The problem of being taken seriously: will researchers be interested in what librarians have to say about RDM?
5. The scale of the task: RDM can involve setting up a service or services from scratch.
6. The lack of infrastructure to assist with RDM.

Therefore Dr Cox recommended the following steps to raise awareness about RDM issues:

• Confidence raising: giving people the confidence to talk about RDM, even if they don’t have all the answers.
• Increasing knowledge about RDM issues.
• Encouraging changes of professional identity to include RDM as one of the things that librarians have expertise in.
• Providing prompts to getting started whether or not policy and infrastructure are in place, and instead of waiting for these to come into place through the actions of others.

In conclusion this was a really good event, and I wish I could have stayed on for the main conference, Elag2013: Inside Out Library. The take-home lesson for me was that it’s better to do something rather than nothing when it comes to RDM, since if we in the library don’t then perhaps no one will.

Filed under: Events, Research Data Management, data, RDM, research data, research data management