Real-Time Data and Analysis to Advance Racial, Social, and Political Justice in Social Work Science: Lessons Learned from a Transdisciplinary Collaboration

Introduction: Social work research has made substantial scientific contributions to nearly all dimensions of human service development and delivery, informing policies and interventions attempting to advance racial, social, and political justice. Foundational to this work have been a variety of qualitative and quantitative methods including in-depth interviews, analyses of social and epidemiological surveys, and designs for testing the efficacy of interventions. While such methods are undoubtably valuable, they are costly, delaying the dissemination of critical findings. Advances in the real-time collection and storage of administrative data, combined with the willingness of researchers to leverage their expertise beyond their own disciplinary expectations, offer unprecedented opportunities for collaborative teams to study social phenomena as they happen, increasing the likelihood that policies and practices are responsive to the immediate needs of impacted populations. This paper describes the novel data sources and analytic methods that are shaping this new frontier, discussed within the context of COVID-19’s effect on housing evictions.

Methods: A transdisciplinary team of scholars with backgrounds in computational social science, computer science, social work science, and housing policy met weekly in a working group that sought to inform timely responses to the post-COVID eviction crisis in New York City. Researchers identified administrative datasets to track the impacts of COVID-19 on housing evictions and utilized various computational approaches to study these associations and predict the impacts of the pandemic on future housing instability and displacement. We chronicle our progress by developing a flow chart of data acquisition and methodological decision points, illustrating the depth of knowledge required to engage in complex research decision making aimed at informing dynamic policy imperatives.

Results: Researchers located current data on eviction filings from the New York State Office of Court Administration (OCA) and NYU’s Furman Center, executed eviction warrants from NYC OpenData, unemployment rates from the NY State Department of Labor, and COVID-19 case rates from NYC Department of Health and Mental Hygiene (DOHMH). Importantly, both OCA and DOHMH data are collected in real-time. These data sources were augmented with community-level variables drawn from the Census Bureau’s American Community Survey (ACS). Techniques borrowed from geographic information systems science were used to harmonize differences in levels of geographic aggregation between data sources and time lags across data sources were overcome by employing forecast modelling. Predictions regarding how these relationships will unfold with and without policy intervention in the future are hypothesized using simulation models. These rich sources of data, combined with substantive and analytic expertise, were used to inform the results provided in the remaining papers of this symposium.

Conclusion: This paper presents the workflow of a transdisciplinary team aimed at impacting housing policy in real-time, illustrating their data source considerations and methodological decision points. In order for policies and the allocation of resources to be responsive to the most pressing issues at hand, social scientists need to embrace these novel data sources and analytic approaches and bring the expertise from their respective fields to help bend the arc of the moral universe toward justice.