Reported individual costs and benefits of sharing open data among Canadian academic faculty in ecology and evolution

Abstract

Open data facilitate reproducibility and accelerate scientific discovery but are hindered by perceptions that researchers bear costs and gain few benefits from publicly sharing their data, with limited empirical evidence to the contrary. We surveyed 140 faculty members working in ecology and evolution across Canada’s top 20-ranked universities and found that more researchers report benefits (47.9%) and neutral outcomes (43.6%) than costs (21.4%) from sharing data. Benefits were independent of career stage and gender, but men and early career researchers were more likely to report costs. We outline proposed mechanisms to reduce individual costs of data sharing and increase benefits.

Share the code, not just the data: A case study of the reproducibility of articles published in the Journal of Memory and Language under the open data policy

“…Several surveys of data sharing practices show that less experienced researchers are hesitant to share data and analysis code out of fear of public shaming, loss of reputation, etc., if any errors are found (Meteyard & Davies, 2020;Soeharjono & Roche, 2021). It is possible that errors may be found, but this does not necessarily lead to an adverse effect on one's career (Ebersole et al, 2016).…”

Slow improvement to the archiving quality of open datasets shared by researchers in ecology and evolution

“…We found no evidence to suggest that PI gender influences the quality of open data (Figs 2 , S2 A,B). This result is encouraging given that men in E&E selfidentify as experiencing more costs than women as a result of sharing open data 2 . Women accounted for almost half of the PIs assessed in our study, yet far fewer than 50% of PIs in biology departments at Canadian institutions identify as women 27 .…”

“…Many journals, funding agencies, and policymakers agree that the societal benefits of publicly sharing (non-sensitive) research data outweigh any perceived or reported costs to individual researchers 2-6 . Indeed, while some researchers remain reluctant to share the data underlying their published results 3,7 , most view open data positively 2,8,9 . Since 2010, when a handful of journals began requiring open data in ecology and evolution (E&E) 10 , policies encouraging this practice have grown rapidly.…”

“…The fact that corresponding author status had no bearing on data quality is worrying and suggests that PIs do not understand the responsibilities associated with this role, do not have the tools or training to ensure that open data are compliant with journal policies, or simply do not care. Education, capacity-building, and incentives targeted at individuals are needed to address these issues 2,26 . We had no a priori hypothesis for why PI gender might influence data quality.…”

“…This consistency within research groups suggests both a robust lab culture promoting good research data management or, alternatively, a PI's lack of competence or reluctance to engage in data sharing and student training in this regard. PIs who oppose open data initiatives 2,7 are unlikely to respond positively to incentives or training opportunities to improve data quality and FAIRness. However, our results suggest that only a minority of PIs potentially fall within this category (approximately 10%; Fig.…”