Catalyzing Policing Reform with Data: Technical Appendix

Technical Paper

Catalyzing Policing Reform with Data: Technical Appendix

Abstract

In the report Catalyzing Police Reform with Data: Policing Typology for Los Angeles Neighborhoods, we created a typology that elucidates the relationship between resident-initiated and police-initiated activity in the City of Los Angeles, as well as explore how that relationship varies across neighborhoods, by synthesizing data sources on calls for service, stops, arrests, and crime. The open data come from the Los Angeles Open Data Portal, the American Community Survey, and the Department of Housing and Urban Development. This Technical Appendix provides a detailed description of the data sources and methodology used to develop the typology, including processing the raw data, conducting a cluster analysis using Gaussian mixture models (GMM) and performing a stability analysis to identify how individual variables drive clustering. Our code for processing and analyzing the data is on GitHub.

About the Project

The analysis was conducted in collaboration with the Microsoft Criminal Justice Reform team, the Microsoft Data Science and Analytics team, and the University of Southern California’s Sol Price Center for Social Innovation.

This neighborhood-policing typology is part of a larger criminal justice project of the National Neighborhood Indicators Partnership (NNIP), which is a learning network coordinated by the Urban Institute that connects independent partner organizations in 30 cities. NNIP’s mission is to ensure all communities can access data and have the skills to use information to advance equity and well-being across neighborhoods.

Centers

To reuse content from Urban Institute, visit copyright.com, search for the publications, choose from a list of licenses, and complete the transaction.
LATEST IN Crime and Justice
To reuse content from Urban Institute, visit copyright.com, search for the publications, choose from a list of licenses, and complete the transaction.