uOttawa Response to Draft Tri-Agency Research Data Management Policy

The University of Ottawa provides the following comments on the draft Tri-Agency Research Data Management Policy on behalf of its research community of more than 1200 researchers.  These comments are informed by comments received in a focused researcher survey on this topic during late July and August 2018 and internal discussions across key institutional areas -- Library, Information Technology and the Office of the Vice-President, Research.

Our response:

We support the overall intent and policy objectives outlined in the draft Policy and the three main requirements: institutional strategy, data management plans and data deposit.  However, there is insufficient detail to adequately assess key aspects such as implementation costs and timelines. For this reason, we recommend further clarity on key requirements, followed by an incremental and flexible implementation approach.

Specific considerations:

  1. Throughout the document, there is a lack of clarity on key concepts and specific requirements. For example, for the institutional strategy, institutions need to understand specific requirements in order to develop their policy framework and strategy and identify resource requirements.  It is not clear whether the final policy will have explicit expectations for institutional conformity in key policy areas, such as information classification, data protection, management and ownership; or whether these will be determined within each institution.  We suggest that some degree of conformity is desirable to minimize duplication of efforts and maximize the value of research data.  We envisage that concepts and requirements could be developed through a national dialogue of key institutional stakeholders (library, research, IT, compliance) prior to the finalization of the policy. We note that work done by the Portage Network can be a helpful resource. (Appendix A cites other examples where clarity is needed.)
  2. Many of our researchers voiced concerns about researcher and institutional readiness to support the development of data management plans and data deposit. The key barriers identified for developing data management plans were: training, tools, institutional infrastructure and support services (less than 15% reported expertise, with readiness varying across disciplines). Health researchers reported greater capability, social scientists the lowest level, with natural scientists and engineers in between.  This divergence in readiness can both inform institutional level strategies and may also support a different national rollout for each funding agency.
  3. Many concerns were raised regarding the mandatory deposit of data. The majority, though not all, of the concerns came from researchers using qualitative research methods. These included:
    • potential risks to human participants
    • risks to researchers (for example, research being done in politically unstable areas)
    • predicting the long-term risks vs. the short-term risks for both participants and researchers
    • the protection of intellectual property (e.g., having research ideas stolen and used by others who haven’t done any work; particularly for qualitative and community-based research), proper citation
    • curation challenges (not merely storing data but doing so in a manner that is useful and ensuring long-term access and usability of data)
    • secondary use of data (e.g., ethics requirements, potential risks)
    • interpreting/using data out of context
    • potential impact of security breaches
  4. Resource requirements are difficult to quantify, but it is clear that the costs of complying with the RDM policy will weigh heavily on institutional budgets and individual researchers’ project budgets and will divert valuable resources away from research to RDM.  Additional dedicated resources for improved RDM will be required to prevent a deterioration in research.
  5. Need for realistic implementation approach and timelines. Research data management in Canada is in the embryonic phases, and it is important that we learn to walk before we run.  To this end, we support a staged, multi-year approach, which starts with institutional strategies, rather than firm policies. This will allow institutions to conduct thorough assessments of institutional readiness and researchers’ needs and develop appropriate actions to support research data management practices. It will also allow institutions and researchers to assess the limitations, benefits and potential risks of mandatory data deposits.


Appendix A: Additional Examples of Key Concepts and Requirements to be Clarified

  • Section 1 – Preamble
    • Clearer definitions needed for “research data”, “operational data”, “third party data”, “public sector data”, and “repurposed data”. Depending on the field of study these terms mean different things to different people.
  • Section 3.1 – Institutional Strategy:
    • Reference to TCPS 2 and the RCR framework: it is unclear how the policy requirements will be linked to these two documents and whether any changes will be made in order to evaluate compliance with the RDM policy.
  • Section 3.2 – Data Management Plans
    • “Methodologies that reflect best practices in research data management practices” – need to clarify what these best practices are; consider what applies in different fields of research.
  • Section 3.3 – Data Deposit:
    • “Recognized digital repository” needs to be defined. What is recognized; by whom is it recognized; what are the criteria for recognition?
    • “…encourage researchers to provide access to the data where ethical, legal, and commercial requirements allow ….” Need to clarify whether this is a requirement and how to determine what exceptions would be acceptable. Examples would be helpful.
    • No mention of personal and sensitive data, and specifically personal health information – more guidance is needed.
    • Reference needed for standard of de-identification of data to be used (the terms used in the TCP 2 article, Chapter 5A could be referenced) and responsibility for checking the appropriateness of storing various types of data.
    • Need to clarify issues around consent: how to inform participants and get their informed consent to have their data in a public repository.
  • Indigenous Research data: There is no proper consideration regarding specific requirements for these communities; the policy does not mention whether any discussion has taken place to date, there is merely a brief note in the FAQ stating that the agencies welcome feedback on how the policy could affect “Indigenous research, knowledge and data”.
Back to top