Conceptualizing Data Curation Activities Within Two Academic Libraries

Abstract

A growing focus on sharing research data that meet certain standards, such as the FAIR guiding principles, has resulted in libraries increasingly developing and scaling up support for research data. As libraries consider what new data curation services they would like to provide as part of their repository programs, there are various questions that arise surrounding scalability, resource allocation, requisite expertise, and how to communicate these services to the research community. Data curation can involve a variety of tasks and activities. Some of these activities can be managed by systems, some require human intervention, and some require highly specialized domain or data type expertise. At the 2017 Triangle Research Libraries Network Institute, staff from the University of North Carolina at Chapel Hill and Duke University used the 47 data curation activities identified by the Data Curation Network project to create conceptual groupings of data curation activities. The results of this “thought-exercise” are discussed in this white paper. The purpose of this exercise was to provide more specificity around data curation within our individual contexts as a method to consistently discuss our current service models, identify gaps we would like to fill, and determine what is currently out of scope. We hope to foster an open and productive discussion throughout the larger academic library community about how we prioritize data curation activities as we face growing demand and limited resources.

Maintaining Repositories, Databases, and Digital Collections in Memory Institutions: An Integrative Review

“…In multiple papers, authors wrote that before migration could occur, dataset‐ or object‐level curation had to be performed, including metadata cleanup or remediation (Darcovich et al, 2018; Georgieva and Flinchbaugh, 2021; Gentry et al, 2021; Waldron, 2017; Walsh, 2010), data normalization (Georgieva and Flinchbaugh, 2021), file format migration (Allen, 2017; Dunn et al, 2016; Jeng et al, 2017), file format risk assessment (Lafferty‐Hess et al, 2020), file segmentation (Matusiak et al, 2017), and checking for file corruption (Doty et al, 2015). Two papers included explicit recommendations for how to plan for this work: Build quality control into multiple steps of the procedure .…”

“…Six focused on database maintenance in preparation for migration (Darcovich et al, 2018; Dressler, 2014; Georgieva and Flinchbaugh, 2021; Walsh, 2010; Wu et al, 2016; Wu et al, 2019). 39 papers were about maintenance or sustainability of a system separate from migration (Bond, 2006; Dal Pra et al, 2019; Downs & Chen, 2010; Gordon et al, 2015; Greenberg et al, 2009; Greig & Nixon, 2007; Gries et al, 2018; Han, 2011; Herrera, 2007; Imker, 2020; Jeng et al, 2017; Jiang et al, 2015; Jones et al, 2000; Kertesz et al, 2020; Lafferty‐Hess et al, 2020; Le et al, 2015; Lee et al, 2018; Lutz & Meadow, 2006; Madsen & Oleen, 2013; Manninen, 2018; Marcial & Hemminger, 2010; Nagendra et al, 2001; Nie et al, 2021; Owen & Michalak, 2015; Park et al, 2018; Philipson, 2020; Post et al, 2019; Powell & Moseley, 2021; Reilly, 2009; Rossi & Ahmed, 2016; Rousidis et al, 2014; Schuler & Kesselman, 2018; Stolte et al, 2003; Trisovic et al, 2020; Tuan et al, 2011; Waldron, 2017; Wilkinson et al, 2017; Xu et al, 2016; Zohner et al, 2019). Two papers took a broad look at both migration and maintenance (Stigler & Steiner, 2018; Thomer & Twidale, 2014).…”

“…This included mapping and crosswalking metadata and data schemas (Georgieva and Flinchbaugh, 2021;Gordon, 2015;Kansa, 2005;Lutz and Meadow, 2006;Matusiak, 2017;Owen and Michalak, 2015), unexpected data cleaning (Dunn, 2016;Knight-Davis et al, 2015), and integrating legacy data (Kansa, 2005;Knight-Davis et al, 2015). Several noted that there simply isn't a one size fits all solution to migrations, and that it's necessary to plan for building custom workflows (Lafferty Hess et al, 2020;Mannien, 2018;Post, 2019).…”