Improving data management in academic research: Assessment results for a pilot lab

Abstract

Common practices for data collection, storage, organization, documentation, sharing, re-use, and preservation are often suboptimal. Issues often arising from common data practices include data loss, corruption, poor data integrity, and an inability to demonstrate the provenance (i.e., the origin) of the data. Ineffective data management can result in data that are unusable for re-use and re-analysis. However, effective data management practices exist to support data integrity, interoperability, and re-use. These practices maximize the value and potential impact of any particular dataset. In light of the gap between common practice and known effective strategies, we developed an intensive lab curriculum to train students and research support staff in implementing these strategies. This lab addresses the lack of formal data management training available on our campus and targets key processes in the data life cycle, promoting strategies that facilitate generation of quality data appropriate for re-use.