A Review of COVID-19 Intersectional Data Decision-Making: A Call for Black Feminist Data Analytics, Part I

Credit: Mary Long/Shutterstock Images

This piece is the first in a series of reflections on data analytics and COVID-19.

Now that the United States is a little over half-a-year into the coronavirus pandemic, data has revealed that Black Americans and other communities of color have disproportionately contracted and died from COVID-19. The reasons for this have been well-discussed by physicians, public health experts, and scholars who study race and racism in the United States. Rhea Boyd, a pediatrician and child health advocate, and her colleagues have challenged the medical community and others to place systemic racism at the center of the reasons for the racial health disparities that are highlighted in the pandemic’s impact. Boyd et al. write:

“In short, racism kills. Whether through force, deprivation, or discrimination, it is a fundamental cause of disease and the strange but familiar root of racial health inequities.”

Structural racism, as a fundamental source of elevated sickness and death rates in Black communities impacted by the pandemic, is a critical factor in understanding COVID-19. Less consideration, however, is devoted to the relationship between racism and critical data analysis. What is its significance for developing interventions that could mitigate the impact of the coronavirus on Black people and other marginalized racial groups?

One approach to COVID-19 racial data that might serve as a model but stops short is featured in the article, “A Data-driven Approach to Addressing Racial Disparities in Health Care Outcomes.” The piece recounts the work of health data analysts, physicians, and researchers at Brigham Health in Boston. They worked with hospital leaders in diversity, equity, and inclusion in clinical and hospital care to analyze subgroup data that primarily included: race, ethnicity, language, sex, insurance status, geographic location, and health-care worker status.

The Brigham Health team adopted legal scholar and critical race theorist, Kimberlé Williams Crenshaw’s intersectionality framework, which she defines as “a lens through which you can see where power comes and collides, where it interlocks and intersects” and created filters which allowed them to better understand how COVID-19 impacted some groups more than others. Their approach revealed silences in the data that would have remained so had they not applied an intersectional method to their data analysis. For example, Brigham Health found that “Hispanic non-English speaking patients were dying at higher rates than Hispanic English speaking patients.”

Acknowledging people’s multiple identities, which shaped their vulnerability and experience with the coronavirus inspired the Brigham Health team to make data-driven decisions that were also intersectional. For example, the Brigham Health team reports that by identifying the geographic location of individuals as an intersection of their identity, they were able to set up free COVID-19 testing in hotpot neighborhoods. This created opportunities to conduct additional health screenings and highlighted other issues that made people of color vulnerable to the coronavirus. Clearly, using an intersectional approach to data analysis is quite effective.

Even as the Brigham Health team’s work highlights the value of an intersectional data analysis, they applied a narrow understanding of intersectionality that tends to be limited to intersecting identities. This concept of intersectionality, as it was in the Brigham Health Study, is often disconnected from the larger Black Feminist framework from which it is derived.

Yet, the Brigham Health Team’s application of intersectionality is not an isolated incident. Other data scientists have also settled on an incomplete definition of intersectionality to analyze data and develop strategies for ameliorating the impact of COVID-19 on Black and Brown people. Several pieces touting intersectional approaches to analyzing COVID-19 data have emerged over the past spring and summer. Many of them are connected by their pared-down application of intersectionality as a determinant for how “power and inequality are structured differently for groups.” Even as these commentaries serve as important and necessary calls for the application of an “intersectional lens” to data in the studies of COVID-19, they remain inadequate in the effort to fully “interrogate racism as a critical driver of racial health inequities.”

When we return to Crenshaw’s original conceptualization of intersectionality, it is clear that she was just as concerned with the systematic design of power as she was with how individual intersecting identities experienced them. Crenshaw writes in her analysis of discrimination, “the paradigm of sex discrimination tends to be based on the experiences of white women; the model of race discrimination tends to be based on the experiences of the most privileged Blacks.” She goes on to argue that race and sex discrimination are constructed to address a narrow set of circumstances that deny Black women’s lived experiences. To this end, a fuller accounting of intersectionality requires a deep and thorough analysis of interlocking identities as well as an interrogation and overhaul of the powerful and oppressive systems that negate them.

Catherine D’Ignazio and Lauren F. Klein’s book, Data Feminism offers an intersectional approach to data, otherwise known as data feminism, that provides a more substantive grounding in Black feminist thought that grapples with oppressive systems. D’Ignazio and Klein, heavily drawing on Black feminist frameworks, argue that data feminism concentrates on “intersecting forces of privilege and oppression” that demonstrate that power is not shared equally in society. However, their analysis insightfully transcends (as does Sasha Costanza-Chock’s work) the way that minoritized identities intersect. Instead, they also focus on software and data analytic design flaws that are predicated on what Patricia Hill Collins calls a “matrix of domination,” or race, class, and gender as interlocking systems of oppression. Like Crenshaw, Collins argues for a much deeper conceptualization of intersectionality that is designed to dismantle systems of oppression and power.

Let’s return to Brigham Health’s team intersectional approach to COVID-19 data. By applying a limited understanding of intersectionality that solely focused on overlapping identities, the research team neglected to consider the ways that the algorithms used to filter the data may have reflected a matrix of domination. Even more, the team failed to disclose how they defined intersectional categories such as race, sex, ethnicity, geography, etc. and whether these definitions reflected communities of color. Nor does this failure account for how these categories are socially constructed within systems of power. Costanza-Chock writes, “Universalist design principles and practices erase certain groups of people, specifically those who are intersectionally disadvantaged or multiply-burdened under capitalism, white supremacy, heteropatriarchy, and settler colonialism.” This results in algorithms of oppression, which Safiya Umoja Noble discusses in her work. In the same vein, Ruha Benjamin calls our attention to the New Jim Code, which is “a specific manifestation of discriminatory design in which racist values and assumptions are built into our technical systems.”

“Design injustice”(design processes that intentionally and unintentionally engender individual and systematic inequity) becomes more difficult to identify without applying the mutually constitutive Black feminist frameworks, intersectionality, and matrix of domination, to COVID racial data analysis.

The Brigham Health Team’s lack of discussion about the possible “black box” issues raised by the lack of transparency about the algorithms and definitions they used does not invalidate their COVID-19 data analysis nor the decisions they made with regard to health interventions for racial communities. Nonetheless, it does beg the question about whether the team’s data transformation practices and algorithms failed to capture important information or indeed misinterpreted it. Inaccurate data analysis leads to decisions that could negatively impact people’s health and vastly change understandings about Black and Brown people’s susceptibility to the coronavirus.

One might argue that the need for a more substantive data analysis of COVID-19 racial data could easily be resolved by D’Ignazio, Klein, and Costanza-Chock’s insightful approaches to software design and data analysis. They all, in part, build on Black feminist thought that includes intersectionality and a matrix of domination.

So, why does it matter whether data scientists and health researchers focusing on COVID-19 racial data use intersectional frameworks that fully converge in Black feminist theory?

The answer is quite simple. As noted above, COVID-19 has had a devastating impact on Black communities relative to white people. Racial data analyses that are situated outside or alongside a Black perspective, even ones that purport to be inclusive of diversity, equity, and inclusion, are short-sighted. Rather what is required to understand and use COVID-19 racial data in transformative ways is a centering of Blackness.

The Insight Center for Community Economic Development recently called for the centering of Black perspectives:

“… in the creation of new policies, systems, and institutions — in the pursuit of economic liberation for all — we can and must reject the ideology grounded in white supremacy and anti-blackness, shift narratives to reinvigorate our shared imagination, and disrupt the imbalance of power in our society.”

COVID Black joins them in this call. Yet, the expansiveness and breadth of COVID-19 call for methods that are immersed in Black women’s “subjugated knowledge,” and “culture of resistance,” in other words, Black feminist data analytics. To this end, our work is in conversation with the Black Women Best Framework where centering a Black woman standpoint means that:

“…deeply entrenched racist, sexist, queerphobic, abelist, and xenophobic policies that harm the overall economy for everyone are ulimately unraveled”

As more health researchers, policymakers, and data analysts use intersectional frameworks to analyze COVID-19 racial data and other health information, they must pay close attention to the significance of interlocking identities and the way power is embedded in algorithmic applications and data analytics. A Black feminist data analytical framework offers one of the more expansive and liberatory means for doing this.

Look for the next essay in this series to learn what Black feminist data analytics are and the value they hold for analyzing COVID-19 racial data.


Kim Gallon is the Director and Co-Founder of COVID Black. Learn more about COVID Black by clicking here.



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store


Redefining Health Data…Connecting Black Communities to their Data.