Eviction Data: process and practice
Data are often viewed as value-neutral – a way of measuring a social issue. Yet what and how is collected, as well as how data are disseminated reflect existing framing of issues and, by extension, people and places. By extension this also means a reflection of existing power. For example, as a new article by Meghan Hatch, Elora Raymond, Benjamin Teresa and NCSG director Kathryn Howell discusses, the ways eviction data are presented, collected and, indeed, sold, imperil the ability for tenants to find and maintain housing as well as their ability to collectively organize. The new article A data feminist approach to urban data practice: Tenant power through eviction data in the Journal of Urban Affairs argues for an approach to data practice that is clear about the goals and potential harm or benefit to that data.
What is Data Feminism?
Part of the larger movement for data justice, data feminism offers a framework for examining power, harm and goals in data collection, use and dissemination. Far from being tools for measuring an issue, data feminism exposes a truth of all research: the framing of an issue and the questions you ask about it can determine what you get out of the data and ultimately, how you interpret it. Data feminism centers seven principles for collecting, analyzing and deploying data: examine power, challenge power, elevate emotion and embodiment, rethink binaries and hierarchies, embrace pluralism, consider context, and make labor visible. These principles fall into three broad categories, data as process, data as power and data as participation. In other words, Hatch et al write, “data feminism does not separate the research process from the wider social environment in which researchers carry out data collection, analysis, and interpretation of findings” (p. 3).
Why Eviction Data?
The authors examined eviction as a critical space for applying data feminism because “at its root, eviction is the continuation of long processes of dispossession and displacement traced from early segregation and exclusion through urban renewal, redlining, and foreclosure” (p.4). Indeed, research has consistently found that Black renters – particularly Black women – are at highest risk of eviction (Desmond et al., 2013; Hepburn et al., 2020). Further, evictions are spatially concentrated in racialized submarkets (Teresa & Howell 2021, Raymond et al., 2018, Immergluck et al., 2020), and disproportionately instigated by large investor-owner landlords (Gomory, 2022; Raymond et al., 2016, 2018). Yet, data are typically used to assess the risk of renting to particular tenants, ultimately weaponizing rental history that grew out of racialized roots exclusion to homeownership, employment and political power (Diamond & Wilson, 2023).
Data Feminism for Evaluation
The paper critiques two processes for building eviction data tools undertaken in Richmond, Virginia and Atlanta, GA in which the authors participated. In both cases, the coalitions building data tools faced challenges related to data structures from the courts, competing interests and long term funding. Through the paper, the authors take on the tensions of power and goals for the data as each stakeholder had different, and sometimes conflicting, goals. For example, organizers need data to understand ownership, common behaviors, and risk for tenants, while elected officials want data that can help them monitor change over a crisis. This was exacerbated by the limited access to the data at the government level that made it difficult to use data to challenge patterns of behavior. At the same time, the paper wrestles with questions of pluralism and hierarchy in the data. Though both cases engaged a wide group of partners to understand data needs, the data were ultimately held centrally, meaning, as the paper argues, it becomes critically important to have ongoing dialog with partners and continue to critique the implications of concentrated data for power and representation.
As the paper argues, “these cases do not illustrate the perfect application of data feminism. Yet, they offer space to understand both the barriers to that approach in practice and the opportunities that are created by reimagining ways data can be used to change not just outcomes, but indeed, ongoing community processes” (p 14). Both projects are ongoing and evolving and continue to be challenged with bringing the resources to a data process that will support the goals of examining and challenging power, embedding in context and engaging in authentic relationships.
These cases illustrate the critical need to understand eviction as more than a moment in time. Instead, eviction is a process that is part of the history of racialized dispossession, which is reproduced through data tools and representations. This process can be better illuminated by applying the principles of data feminism to analysis of eviction data.