This is one of a two-part blog post that discusses incorporating data management and data sharing plans into researcher workflows. The second blog in the series, by former CLIR/DLF postdoc in data curation Vessela Ensberg, will appear next week and focus on lab-based researcher workflows with non-human biological specimens.
Libraries and library professionals are emerging as major stakeholders in the data management business and continue to position themselves as viable institutional custodians of large collections of research data and research records. In the push to promote open access initiatives and meet recent national funding agency data management planning requirements, academic research libraries are promoting data management education and open access to research outputs, including data. The intent of this post is to encourage library professionals to think about how data management planning and data sharing discussions might be tailored to the research context—in short, to explain how data management planning and sharing protocols can be used to support researcher workflows.
Individual researchers are influenced by particular disciplinary histories and ongoing political, economic and ethical debates that center on access and reuse of research data derived from human participants. They are subject to international treaties, state and local laws, professional ethical codes and institutional policies that define their role and relationship with research participants and research data. That said, academic researchers working with human participants have a great deal of flexibility and control over how they address these common regulatory and ethical issues. At the same time research participants and communities of study also have their own expectations and informal or formal information sharing protocols. Researchers—especially those employing ethnographic, collaborative, applied, and clinical approaches—continually negotiate access to research participants and communities. Digital, linked, public access to information magnifies the implications of unwarranted or unintentional disclosure of sensitive information and transforms debates over intellectual property, data ownership and who has the moral right to exchange knowledge. No one can definitively say who “owns” data or give you a “one-size-fits-all” data sharing protocol or data use agreement that will work in every context. So yes, it’s complicated.
But it’s complicated need not be an end to the discussion. It offers a point for engagement and a way out of what can be a tedious conversation about file formats, standardized metadata, and the latest encryption technology. How might the data management plan help organize the interlocking yet disparate expectations that inform human participant research today? How might it help researchers meet the expectations of their own emergent ethical concerns, collaborative research methodologies, and local community-based research protocols?
A data management plan can summarize countless pages of complex legal agreements and expectations, protocols, funder requirements, ethics applications, interview guides, and survey instruments. It can also help researchers identify competing expectations and gaps between these agreements, policies and protocols. For instance, a local research protocol may call for community ownership of data collected during the course of research. Does the community have the technological and financial ability to maintain the data at standards set by the institutional ethics review board? If not, then how might this planning be incorporated into the research project and funding structure? How will the researcher and community go about protecting individual participant confidentiality and potential reuse of data? Who represents the community? What details might need to be negotiated before or during a project? These issues and proposed solutions can be detailed and updated within the data management plan and inform a wide range of activities within a researcher’s workflow such as the development of data sharing protocols, participant consent forms, operational approvals, grant writing activities, and potential long-term access and preservation planning.
Consent to Share
Academic researchers usually need to inform their participants of the goals of the study and the type of information or materials that will be collected. Participants need to know whether the researcher feels this information might be sensitive, either by itself or if linked with other publically available information, or if it poses a physical or emotional risk to the individual (or community of study). They need to give permission to participate and for their identity to be revealed. Data can sometimes be collected anonymously and at times can be de-identified to the point that it would be extremely unlikely that the participant could be re-identified. Alternately, it may be impossible orunwarranted to take extreme precaution due to the low-risk nature of the study or the desire of participants to be identified.
In some jurisdictions permission is needed if a researcher will use the data to answer another unrelated research question (reuse) whether the participant can be identified or not. But despite lack of unified international regulations to the effect, secondary reuse of human participant data is considered an ethical issue that increasingly needs to be addressed in a research proposal or protocol. Even when this permission is obtained from individual research participants, institutions (such as public health authorities), governments, community elders or representative councils may override this individual permission. When it is necessary to gain operational or community approval to undertake research (or gain a research permit), this permission is paramount.
An underlying principle informing modern research ethics is the right to withdraw from a study, the right to say no, and the right not to answer a question or undergo a procedure. Researchers negotiate with research participants, institutions and communities of study to be able to collect data about private, health-related, sacred or otherwise sensitive information. This is often with the understanding that this information will not be linked (without permission) to other available information about individuals or communities. Researchers and communities are developing a variety of approaches to address these concerns in a digital, online environment. Three projects that I find innovative in their use of visual approaches to managing restriction are: the Digital Dynamics Across Cultures Project which operationalizes Warumungu (Central Australia) protocols such as restricted access to women’s knowledge and images of the dead, Murkutu Tradtional Knowledge (TK) License and Labels developed for sharing indigenous cultural materials, and the DataTags project at Harvard University, which seeks to synthesize legal and policy requirements for the handling and sharing of clinical and health data.
This right of the research participant or community to restrict access to knowledge does not always sit well within the open access, open science, open data movements. One way that this tension might be eased is for the researcher to go public (when possible) about the context of restriction documented in data management plans and sharing protocols. This may mean being more transparent (and thus coming under more scrutiny and control/regulation) about the conditions under which research and data collection take place. In the long run, such transparency may be necessary to counter challenges by overzealous advocates of open data who prioritize the public good (the imagined public majority in publicly funded research) over the research participant or community of study’s well-being or researcher/participant trust. It could also lead to a greater understanding on the part of the researcher, research participant or community of study about the implications of disseminating knowledge.
Data management plans can help support collaborative research governance and document current ethical considerations concerning the access and preservation of research outputs. While they may mark a push to make human subject research more transparent (and thus more highly regulated by institutions, governments, and other financial interests), they can also be used by the researcher to help document and manage the complexities of human participant research today and may offer another space in which to advocate for emergent forms of research governance.
CLIR/DLF Postdoctoral Fellow Kendall Roark is an applied anthropologist investigating clinical and health data sharing practices and governance in Alberta, Canada.