How to curate research data: An 8 step guide with incentives to collaborate

Abstract

As reproducibility and data sharing emerge as key issues for academic researchers, the data management services offered by the library must continue to scale. Building from a 2013 pilot (http://hdl.handle.net/11299/162338) the data curation workflows used at the University of Minnesota Libraries have grown into a robust service involving multiple data curation specialists that curate data deposited into our Data Repository for the University of Minnesota (DRUM) and appropriate subject data repositories. Data curation steps, including quality assurance, file integrity checks, documentation review, metadata creation for discoverability, and file transformations into archival formats, are value-add services that enhance digital data for long-term preservation and reuse.

This talk will explore the data curation workflows in place not only at my institution but from 20+ disciplinary- and institutional-based data repositories such as Dryad, ICPSR, Yale, and the U of New Mexico. These experiences, collected in a new ACRL book due out in September 2016 titled "Curating Research Data: Practical Strategies for Your Digital Repository," span the sequential actions that you might take to curate a dataset from receiving the data (Step 1) to eventual reuse (Step 8). And individual institutions putting these key data curation workflows into action is just the first step. We will also discuss a new Sloan-funded project called the Data Curation Network (https://sites.google.com/site/datacurationnetwork/) that aims to create a shared staffing model for providing data curation services across academic institutions thus allowing our services to scale beyond what a single institution might offer alone.