Jessica Simes: The Application of Data Analytics to Mass Incarceration

CDS Affiliated Faculty Jessica Simes is an Assistant Professor of Sociology in the College of Arts and Sciences. Broadly speaking, she examines the consequences of mass incarceration for communities and neighborhoods in the U.S. and uses data science to answer questions of justice and equity. She co-leads a Learning Community on Safety, Justice, and Health with Katharine Lusk and Jonathan Jay and also has a new book out, titled Punishing Places: The Geography of Mass Imprisonment.

We had the chance to sat down with her to dive deeper into the way she builds interdisciplinary data science projects and to find out just how she creates bridges between social science and CDS.

What motivated you to pursue this field and what drew you to academia?

My research focuses on the harms and inequalities that emanate from policing and mass incarceration, with a particular focus on community wellbeing. I was drawn to this field because I have always been driven to understand and contribute to an end to injustice and inequity. I was inspired by professors early in my education who demonstrated to me how research could be an important tool to spark policy reform and social change.

Can you elaborate on your current research?

My current research uses administrative data from police departments and prison systems to study patterns of racial inequality and health disparities. I’m interested in processes of criminalization (e.g., police surveillance, stops, arrests) and punishment (e.g., sentencing, incarceration). My recent work has explored how public investments in healthcare reduce police arrests (PLoS One), how COVID-19 significantly reduced overall rates of arrest but did not impact racial disparities in the policing of urban neighborhoods (Journal of Urban Health), and racial disparity in the population prevalence of solitary confinement (Science Advances). For each of these projects—as well as my recent book Punishing Places: The Geography of Mass Imprisonment (University of California Press), I take records on arrests or prison experiences and create analyzable datasets to explore urgent questions about racial inequality and justice.

In what ways does data science impact, affect and assist in studies of racial inequality and mass incarceration?

Data science – and for my work in particular, spatial analysis, visualization and exploration, data engineering and modeling – provides a crucial set of tools for uncovering often hidden patterns of inequality and injustice. Through data science, we can connect the seemingly unconnected to discover how structural inequality and structural racism persist in the present day. This also means using data that may not have been built as a dataset originally—such as prison or arrest records, and learning from the data.

Describe the real-world impact of the data being used in these studies.

For all of these projects, I think about impact in three key ways: impact for the academic community, for policy reform, and for the public. When possible (as some data are private), I share data and code with academic researchers interested in studying similar patterns or replicating my analyses, and aim to publish in open-access journals to increase the availability of my findings to journalists, other researchers, and the general public. For impact on policy and for the broader public, my collaborators and I are working on several strategies to work directly with practitioners and policymakers, and share our findings with people directly impacted by mass incarceration. This is an important next step and where I’m putting most of my energy in my career: how to have meaningful real-world impact.

What made you decide to publish a book grounded in data and what do you hope to achieve from this publication?

I decided to write a book about the spatial pattern of incarceration because I wanted to explore several aspects of mass incarceration—and it’s just hard to do in a short article format. The book describes the extant theoretical landscape explaining the spatial and contextual patterns of incarceration, and then I model the urban to rural spatial trends, explore racial disparity, and examine novel measures for estimating the community-level impacts of mass incarceration. I hope this book sparks a public conversation about re-framing mass incarceration as a community-level harm, and encourage discussions not just about how to create change, but how to respond to longstanding historic injustices related to crime policy in the United States.

What have been some challenges you faced between your domain and data science?

The biggest challenge I’ve faced is accessing data. Spatial data on criminal justice encounters – from arrest to conviction to incarceration to release from prisons or jails – are hard to access in a timely manner. Data get old very quickly, and often these data are shared through intensive data agreements between researchers and agencies. Part of my work to overcome this is to think of creative strategies and processes for accessing information—from data scraping to public information requests, and eventually leveraging the talented students at BU and especially in CDS to collaborate on these open science projects.