ERDMAS: An exemplar-driven institutional research data management and analysis strategy

Devising fit-for-purpose research data management strategies within a university is challenging. This is because the five ‘Vs’ for generated research data; its Volume, Variety, Velocity, Veracity and its Value must be constantly considered. Invariably, a combination of data V’s for any given research endeavour determine how best to manage it appropriately addressing archiving, compliance, security, privacy, sharing, reuse and so forth. As such, institutions are faced with defining, shaping and refining strategies and practicies to ensure there are consistent and adequate research data management polices and guidelines in place for their researchers. FAIR data principles are very important for embracing open data opportunities, but more broadly, research data management practices need to be established in a comprehensive way. Additionally, new ICT options have rapidly become available where institutions can make considered choices on whether to continue to use ‘on prem’, private Cloud or public Cloud infrastructure. If a hybrid approach is adopted, then the potential impact on existing institutional research data management strategies must be continually assessed and revised accordingly. Getting the balance right between developing a relevant institutional policy on the one hand yet also dynamically catering for the eclectic research data management and analytics needs of researchers and their evolving interactions with external collaborators on the other, must be continually navigated. In this manuscript, an exemplar-driven research data management and analytics conceptual framework is introduced. A key feature of this framework is that it is couched in two dimensions. On one axis is the ‘standard’ linear approach of developing the research data management policy, guidelines, procedures, audit and risk assessment and an options matrix. Importantly, a second axis comprising a researcher-driven focus is introduced where exemplar research activities are used to define ‘classes’ of research data management and analysis requirements. This exemplar-driven dimension enables an ongoing system-wide comparative review to occur in parallel that can continually inform policy and guidelines refinement.