An API for Officer Involved Shootings
The anger after the officer involved shooting (OIS) incidents in Ferguson and many other cities in the United States is well-justified. A hard and personal question is why these shootings occur. A much easier (or at least a more manageable) question is how many officer involved shootings occur in the US. Is the recent attention due to an increase in shootings, or just more public awareness? I couldn’t find basic statistics on police shootings. This seems wrong.
I was only able to find official statistics from the Dallas and Philadelphia police departments. Here is a visualization of of the Dallas data. This department should be commended for their efforts toward public accountability, even though the statistics are troubling. Why are there so many officer involved shootings? Zoom and click on the points for details on the shooting.
The CartoDB map comes with a built-in API, based on SQL queries. You can access the data in a structured way with the following query:
This data is being incorporated into a more comprehensive database, which also includes media reports of OIS incidents. This service rides on elasticsearch, so the database is searchable and scalable.
Another field will ultimately be added for report type, either official or unofficial. The unofficial reports are powerful for two reasons. First, they apply pressure to police departments nationwide to release official statistics. The information will be publicly available, whether the police departments like it or not. With more exposure, police departments should release their own data to ensure that the information true, rather than filtered through a media machine. Second, the reports are striking. It is incredibly troubling to read through the aggregated reports. This sort of transparency is steeped in history. Ida Wells simply reported the disgusting frequency of lynchings in the South in the late nineteenth century. Just seeing the numbers on lynchings created a movement to stop them. Hopefully people will contribute to the open-source repository to continuously scrape the various sources of this information and serve them back out in a structured, reliable way.