Taking an Equity Lens to Our Data Practice
The COVID-19 pandemic and the killings of Breonna Taylor, George Floyd, and other Black Americans at the hands of police have laid bare the life-and-death racial inequities that have long characterized life in America. The Urban Institute, like many organizations, is taking stock of how our practices cause harm and what it would take to advance racial equity.
Urban’s reputation, built on objective, evidence-based research, carries significant responsibility. As researchers and data experts, we have the power to shape conversations and actions in ways many don’t. As Urban’s president Sarah Rosen Wartell stated recently, “Objectivity requires that we name racism when we see it, that we explain how its roots run through the American experience, understanding its lasting influence in ourselves and our own institution.”
We also need to acknowledge the past and current harms done with data by people in government, the private sector, academia, and other institutions. Advocates, researchers, and others have reported people using data against people of color and other marginalized communities, from historical redlining practices to modern biases in hiring algorithms. Other harms are less visible but just as real, such as when the misuse of data leads to stigmatizing people and places by comparing statistics about white people and people of color on income, educational attainment, or health that readers attribute to individual behaviors or stereotypes and not to the historical and modern context of racist policies and practices that drive these disparities. And there’s the repeated stigmatization that occurs when people use data to label neighborhoods as high poverty, violent, and having poor educational opportunities. Irresponsible data use contributes to racist impressions of people and communities and affect value judgments, policy responses, and investments.
Making data a true community asset requires examining our practices and principles and considering how we can shift power to people who have been systematically denied power and opportunity. The 1979 Belmont Report set out to equalize the balance of power between researchers and people being researched, partly in response to the horrific abuses perpetrated by researchers in the Tuskegee Syphilis study. It set new expectations for researchers and established ethical principles to guide research practice:
· Justice: that people who bear the risks see the benefits
· Respect for persons: that people who could be coerced or deceived have the information and freedom to make choices in their own interests
· Beneficence: that people with power avoid causing harm
These principles have been institutionalized in the practice of institutional review boards (IRBs), such as the Urban’s own IRB committee. But in our experience, IRBs are focused primarily on research that involves direct interaction between people and researchers and on potential harm to individuals, not the collective. Research conducted with secondary data, especially if aggregated, is typically exempt from review. If data are to be used as a tool to advance equity, we need to apply these principles to all types of data and at all stages of the data life cycle, from conception of the questions to dissemination of data and findings. These general principles are needed because rules can’t be written to cover every situation, especially as data and technology are constantly evolving. We’ve written a brief suggesting principle-aligned practices for each stage of the data life cycle and have collected resources to help guide more equitable data practice.
At Urban, there is a great deal more we must do to advance equitable data practice, but we share a range of examples where we’ve been more intentional.
· Our data tool “Closing Equity Gaps in DC’s Wards and Neighborhoods” focuses on the people who would benefit by reorienting how gaps are quantified and providing a scale for the solutions. The tool provides narrative context to situate inequities among DC neighborhoods and provides examples of solutions that focus on structural causes rather than individual behaviors. The tool also enables users to define their own goals for what equity would look like in their communities for each indicator.
· Urban’s “Tracking COVID-19’s Effects by Race and Ethnicity” tool avoids directly comparing racial or ethnic groups by presenting the trends for each group separately. It also presents population averages (nation, state, or metropolitan area) as the comparison, rather than the racial group with the fewest challenges. Displaying data this way encourages the viewer to independently consider each group’s needs.
· Our colleagues developed a data walk methodology (PDF) as a means of returning data to the communities they are collected from and engaging the community to provide their own interpretation of the data and thoughts on how policy and practice should change. This methodology has been used in a number of projects, including the Housing Opportunities and Services Together Initiative in Action and the Promoting Adolescent Sexual Health and Safety Demonstration.
· In a participatory research project titled “Community Voices” (PDF), conducted in collaboration with the Austin Justice Coalition, residents were placed at the center of the project to study resident perceptions of policing and participated in designing and fielding the survey, interpreting the results, and developing recommendations to change policing policy and practice.
We know that a few example projects are not enough. Researchers across Urban are actively examining how our research questions and methods advance or hinder equity and what practices need to change. Even if we are not engaged in directly collecting data and information from people, we have a duty to use and manage data responsibly and ethically. With dedicated effort, our research can benefit the people and communities that we engage with and who are most affected by the policy issues we study.
Leah Hendey is a principal research associate at the Urban Institute. Marcus Gaddy is a former Urban research associate and currently works at Montgomery County’s Department of Health and Human Services.