Analysis of Local Police Data
This repository includes methodologies, data, and code supporting the following articles, published by The Trace and BuzzFeed News:
- "Shoot Someone In A Major US City, And Odds Are You’ll Get Away With It" (January 24, 2019) — The Trace / BuzzFeed News
- "5 Things To Know About Cities’ Failure To Arrest Shooters" (January 24, 2019) — The Trace / BuzzFeed News
The Trace and BuzzFeed News analyzed internal data on homicides, aggravated assaults, and non-fatal shootings from 22 municipal police departments in the United States. You can find a description of the standardization process and the data fields in this methodology, and you can download a copy of the standardized offense dataset.
notebooks/analyze-local-police-data.ipynb notebook, written in Python, takes the standardized data, and does the following:
- Identifies the main outcome metric (arrest vs. closure) for each agency/offense combination
- Calculates arrest/closure rates, using the main outcomes identified above
- Analyzes Houston's closure rate trends
- Calculates median arrest disparities by offense and victim race
- Calculates agggregate arrest disparities by offense and victim race
Note: To reproduce the findings, you'll need to unzip the
inputs/offenses-standardized.csv.zip file before running the notebook.
Questions / Comments?
Please contact Jeremy Singer-Vine at firstname.lastname@example.org.