Understanding Data Sharing Practices in the School of Biological Sciences (SBS)

Recent quantitative research from the Biological Research Data Management team BioRDM (Deeb et al., 2025) indicated that data sharing in the School of Biological Sciences (SBS) at the University of Edinburgh (UoE) appeared to be on the rise since 2014, and especially after 2020. Designed to investigate such perceptible increases in collaborative, interdisciplinary, and Open working in our university, this qualitative study is based on 30 semi-structured interviews of research group leaders (“PIs”) conducted between August 2025 and February 2026. Findings in this report are presented in two ways: via an ‘Interview Inventory’ and through a thematic analysis. From the inventory of exact interview transcripts, we show that: o Openness was very rarely articulated as a defining attribute of research data, with only one PI making this link, whereas Openness was a defining attribute of code/software for half the PIs; o two-thirds of participating PIs factor both time and money into data sharing, suggesting that data sharing is usually seen as a significant effort. These findings seem aligned: the effort of Openness is additional to generating data, not an attribute that is inherent in generating data (in contrast to code). o eighty percent of PIs did not change their views on data sharing before and after the COVID-19 pandemic, though half did change some of their data sharing practices, suggesting that data sharing was largely accepted before the pandemic. A thematic analysis of the interviews generated findings that augment the existing survey-based literature. These included five broad categories of commentary, Personal Values, Social Dynamics, Research Practice, The Temporal, and Research Infrastructure, with sub-categories within each of these. The most-referenced sub-category was Research Practice - Making Research Visible (205 references). The Research Infrastructure category is most directly relevant for the School and University. Within Research Infrastructure, Journal Requirements were mentioned more than Institutional Support but Institutional Critiques were also frequent (85, 72 and 56 references, respectively). Two issues surfaced for future consideration. 130 references to Morals indicate that the ethos of data sharing is widely recognised among the participating PIs. However, the University policy on research data management was scarcely mentioned, indicative of a knowledge gap and an opportunity for the institution to build awareness and legitimacy. Second, further School and University actions can progressively reduce the effort, time and cost involved in data sharing. More broadly, we suggest that academic societies could also be prime actors in the Open movement due to their place in publishing, public engagement, and links to the academy.