PNNL

Abstract

The present study investigates barriers and facilitators to the implementation of Findable, Accessible, Interoperable, and Reusable (FAIR) data processes within the Physical Sciences Division of the Computational Sciences Directorate (PCSD). Employing a dual-method approach consisting of surveys and focus group discussions, the study aims to illuminate the complex interplay between technical and psychosocial factors that influence FAIR data adoption. Key Findings: • Surveys indicated that while staff generally understood the merits of FAIR data, the implementation was hampered chiefly due to concerns of accuracy, trust, and resource constraints. • Focus group discussions further elucidated the nature and extent of these barriers, revealing issues ranging from career risk to administrative burdens. • Despite general apprehensions, there was a common acknowledgment of the positive potential of FAIR data, such as streamlining research processes and fostering a community of shared insights and failures. Recommendations: • Convene a cross-disciplinary working group to facilitate implementation strategies and serve as FAIR data ambassadors. • Implement NEMO, an open-source software for streamlined data handling and robust cost-benefit analyses. • Engage in meta-data identification congruent with community practices. • Foster dialogues with Principal Investigators and Project Managers regarding DataHub costs. • Employ a dedicated “Data Librarian” to manage and curate data repositories. The report underscores the necessity of a nuanced approach that considers both technical and psychosocial variables to accelerate FAIR data integration into the PCSD's research ecosystem. The detailed insights and recommendations aim to provide a roadmap for cultivating a data culture that is both rigorous and collaborative, thereby potentially expediting scientific discovery.