Stanford Open Policing Data

A giant dataset of standardized data policing data across different states

logistic regression race police and law


From their site: "the Stanford Open Policing Project is collecting and standardizing data on vehicle and pedestrian stops from law enforcement departments across the country — and we’re making that information freely available. We’ve already gathered over 200 million records from dozens of state and local police departments across the country."

The Stanford Open Policing project is an excellent example of a data source, through the power of standardizing and collecting disparate data sources. But the most exciting part to me, by far, is how far they go in explaining how to use the data and the pitfalls hiding inside. Take a look here, what's the last time you saw a dataset come with a list of best practices and "data notes" for each subset? Compare with the dataset Reuters had to use for their asylum analysis.