Our view is that strides in effective RDM will only be possible if researchers and their institutions proactively create DMPs; we view formal data management as a critical gap essential to address to drive improvement. A DMP is a formal document that outlines how to handle data during and after a research project. It helps ensure compliance with data management best practices and facilitates responsible data sharing, emphasizing data integrity, transparency, and reuse. If one of the goals of the Tri-Agency RDM policy is to increase the availability of data for downstream use—ie., reproducibility checks, big data analytics, and innovation—we need formal planning to achieve this. Despite institutional commitments to data management, in the absence of formalization of data management practices, we are concerned that many data deposits will not be fit-for-purpose.
The Bigger Picture: Culture Change Needed
Right now, researchers aren’t always rewarded for good RDM practices. In many scholarly disciplines, the currency for success is publishing academic papers and particularly doing so in high impact factor journals. If publishing papers alone is what counts for career advancement, and sound RDM practices such as depositing clean, well-documented datasets are not “counted,” we anticipate limited cultural change in RDM practices. We are nonetheless encouraged by grassroots initiatives, including “data champions” programs to promote best practices. Some institutions are also setting up or reinvigorating governance committees to oversee RDM strategies long-term, and we anticipate discussions of incentives and rewards for RDM to be widely discussed.
Final Thoughts: What’s Next?
Variability in resources and capacity identified across institutional strategies highlights the need for resource sharing and novel funding mechanisms. As Canada moves towards full enforcement of the Tri-Agency RDM policy, we see need for greater coordination. Possible actions could include to:
- Develop a community of practice to share resources and address similar concerns in a unified way to reduce spending and duplication of effort.
- Make RDM training mandatory and more accessible to varied disciplinary contexts. These training resources and practical tools are best co-built with researchers who use them. Take as an example the suite of online modules addressing various aspects of RDM produced by the Ottawa Data Champions Team to address the diverse needs of the biomedical research community.
- Develop better support systems for researchers handling sensitive data. This includes addressing best practices for deidentification.
- Offer real incentives for good data management and consequences for failing to manage data robustly.
Ultimately, Canadians fostering a robust RDM culture benefits all actors in the research ecosystem both domestically and internationally.
