Post by Sophia Lafferty-Hess, DCN Curator at Duke University.
In a recent Journal of Librarianship and Scholarly Communications article, curation and repository staff from Duke and the University of North Carolina at Chapel Hill shared the outputs of a “thought exercise” to conceptualize data curation activities within our individual institutional contexts. This exercise was part of a 2017 Triangle Research Libraries Network (TRLN) Institute, which focused on “Supporting New Directions and Projects in Scholarly Communications.” The inputs for our exercise were the excellent Data Curation Activities defined by the DCN and used in the 2017 ARL Spec Kit on Data Curation. We began by mulling over all these activities as a group and considering what we were currently doing and what we might do. We came to an understanding that while both institutions were “doing data curation” in some form – curation is multi-faceted, curation is complex, and curation can mean different things to different groups and stakeholders.
Therefore, we set out to group the 47 data curation activities into three levels (primarily to aid in our own conversation to conceptualize curation). Level one involves curation activities performed by our systems with potential light mediation from humans for such things as metadata enhancement. Level two focuses on a more complete review of the data package, particularly focusing on reviewing documentation, file formats, and risk management for PII/PHI. Level three involves a more enhanced quality review at the data level often requiring in-depth disciplinary/data type expertise or specialized activities, such as code review or de-identification.
Various groups have been considering and communicating levels of curation (for instance the CoreTrustSeal and the DCN). I would argue that levels are important for a variety of reasons. First, they help us frame various service portfolios and consider what is achievable when anything seems possible but resources are limited. As more and more repositories are launched, more and more institutions develop curation programs, and more stakeholders consider “desirable characteristics of repositories”, transparency around what we do in the realm of the curation universe lends legitimacy to our programs and aids in the assessment of options.
In the days of “FAIR data”, curation is also quickly becoming a buzzword. How do we as curation professionals ensure we are clearly communicating what we mean by curation so the word does not lose its impact? Operationalizing curation within an institutional context is no simple task. Significant resources are needed to implement a program that aims to curate data at the level outlined within the CURATED Model. Making the often-invisible work of curation visible, will lead to more recognition of the effort needed to share well-curated data by stakeholders and thereby affect future support for current or future services.
The Data Curation Network as a community provides us a space to learn from each other – to discuss how we talk about curation both within an institutional setting and with our colleagues in the curation, archival, and information science communities. What has resonated the most when talking to people about curation? How do we make the abstract more concrete? How do we ensure our work is valued and recognized by stakeholders? How do we prioritize new services? These questions and more lay the groundwork for continuing conversations for conceptualizing curation. As a member of the DCN and broader curation community, I look forward to engaging in these conversations with DCN members and beyond!
Article Citation: Lafferty-Hess, S., Rudder, J., Downey, M., Ivey, S., Darragh, J. and Kati, R., 2020. Conceptualizing Data Curation Activities Within Two Academic Libraries. Journal of Librarianship and Scholarly Communication, 8(1), p.eP2347. DOI: http://doi.org/10.7710/2162-3309.2347