Getting Ourselves Together: Data-centered participatory design research & epistemic burden

Abstract

Data-centered participatory design research projects-wherein researchers collaborate with community members for the purpose of gathering, generating, or communicating data about the community or their causes-can place epistemic burdens on minoritized or racialized groups, even in projects focused on social justice outcomes. Analysis of epistemic burden encourages researchers to rethink the purpose and value of data in community organizing and activism more generally. This paper describes three varieties of epistemic burden drawn from two case studies based on the authors' previous work with anti-police brutality community organizations. The authors conclude with a discussion of ways to alleviate and avoid these issues through a series of questions about participatory research design. Ultimately, we call for a reorientation of knowledge production away from putative design solutions to community problems and toward a more robust interrogation of the power dynamics of research itself.

Improving Human-AI Partnerships in Child Welfare: Understanding Worker Practices, Challenges, and Desires for Algorithmic Decision Support

“…To ensure that workers who participated in the study would not be identifiable within their workplace, we only report participant demographics at an aggregate level and omit participant IDs for a small number of sensitive quotes. We acknowledge that HCI projects framed as participatory research and design sometimes take up many hours of participants' time, and then yield no tangible follow-through for participants [49]. We intend to continue our collaboration with workers and other relevant stakeholders in Allegheny County to ensure that they can benefit from this research.…”

“…Based on our findings, we provide the following design implications, intended for public-sector agencies deploying or maintaining ADS tools and for researchers exploring ways to design more effective ADS. We note that the following design implications are relevant only when there is good reason to expect benefits of having an ADS in the first place, which outweigh potential harms [6,18,49].…”

Optimizing human-centered AI for healthcare in the Global South

“…utilize the concept of epistemic burden to describe the intensive process of collecting, curating, and disseminating data in participatory design projects. 39 Their work reflects on previous research projects conducted by the authors and incorporates case studies to highlight the extra burden inadvertently placed on community partners, providing insights to evade potentially negative impacts. Other work highlights the power imbalances that continue to exist in participatory design research.…”

Beyond General Purpose Machine Translation: The Need for Context-specific Empirical Research to Design for Appropriate User Trust

“…For example, when Liebling et al tried to understand how migrants in India used MT, female participants were reluctant to be interviewed without their husbands present or rejected the interview without permission from their husbands [25], biasing their participant sample towards male participants. There is also an epistemological burden for conducting research studies with community members, especially to marginalized social groups or people working in high-stakes jobs, as they are often short on time and resources [18,38]. This challenge was reflected in one of the authors' study designs in which we attempted to conduct an ethnographic study with COVID testing volunteers who use MT to help communicate with local immigrants.…”