United States Traffic Stops - Exploratory Data Analysis

Each year in the United States, there are more than 20 million traffic stops across the country. Until recently, these interactions had not been tracked or analyzed in a systematic way.

This began to change in 2015 with the Stanford Open Policing Project, a unique partnership between the Stanford Computational Journalism Lab and the Stanford Computational Policy Lab. Since its inception, the project has collected data for over 200 million traffic stops and put in a lot of work to standardize and clean up these records from the various states, and has made it available to the public on their website at https://openpolicing.stanford.edu/ and on Kaggle.

Scope

I have chosen to focus on Rhode Island, due to the relatively clean and complete state of the dataset. There are about 509,000 records, which will be plenty to answer some interesting questions and gain insight into how and when police interactions take place, whether there is bias in which people are stopped and who is searched, and any trends in terms of stop outcomes.

Download the Data

The first step is to get the data into the Jupyter notebook. We'll use the opendatasets library from Jovian. Let's install and import it, and use the download method.