Authors: Rachael Weiss Riley (Two Sigma), Chris Mulligan (Two Sigma), Vinod Valsalam (Two Sigma), Soumyo Chakraborty (Two Sigma) Jacob Kang-Brown, Christian Henrichson
Presented at: Bloomberg Data for Good Exchange 2017
Abstract: As the rate of incarceration in the United States continues to grow, a large body of research has been primarily focused on understanding the determinants and drivers of federal and state prison growth. However, local jail systems, with 11 million admissions each year, have generated less research attention even though they have a far broader impact on communities. Preliminary time trend analysis conducted by the Vera Institute of Justice (Vera) uncovered disparities in county jail incarceration rates by geography. Contrary to assumptions that incarceration is an urban phenomenon, Vera discovered that during the past few decades, pretrial jail rates have declined in many urban areas whereas rates have grown or remained flat in rural counties. In an effort to uncover the factors contributing to continued jail growth in rural areas, Vera joined forces with Two Sigma’s Data Clinic, a volunteer-based program that leverages employees’ data science expertise. Using county jail data from 2000 – 2013 and county-specific demographic, political, socioeconomic, jail and prison population variables, a generalized estimating equations (GEE) model was specified to account for correlations within counties over time. The results revealed that county-level poverty, police expenditures, and spillover effects from other county and state authorities are all significant predictors of local jail rates. In addition, geographic investigation of model residuals revealed clusters of counties where observed rates were much higher (and much lower) than expected conditioned upon county variables.