New Connecticut police traffic stop data released

Print More

A new batch of traffic stop data and new tools to explore it were released Tuesday by the Connecticut Data Collaborative and the Institute for Municipal Regional Policy.

There’s increased questioning of whether police officers target drivers based on their race, ethnicity, or gender. In context, Connecticut is one of the few states in which law enforcement agencies are required to submit regular updates on all traffic stops. It’s important that such specific data be analyzed and understood by the public, policy makers and civil rights groups.

The Connecticut Data Collaborative now offers several entry points for those interested in the traffic stop data:

  • Their data portal, where visitors can query specifics, such as time, type, age and town.
  • A data story explaining more intricate details of the statistical analysis conducted by the Institute at Central Connecticut State University using data visualizations such as maps, charts, and infographics.
  • Visualizations comparing last year’s data with the new batch of data.

Here are a couple of examples they created:

From October 2013 to October 2014, Hispanics received a ticket 62% of the time compared to 55% for Black Non-Hispanic drivers and 52% for White Non-Hispanic drivers.

Hispanics experienced a decrease with 58% of stops resulting in a ticket compared to 51% for Black Non-Hispanic and 47% for White Non-Hispanic drivers.

Next, the Collaborative said they plan to create more visualizations to compare post-stop data, an interface that allows exploration of descriptive measures, and a drilldown on individual officers.

Want to work on it in person? The Connecticut Mirror and TrendCT are hosting an October hackathon diving into the previous data as well as this recently updated set. Join us and sign up.

What we’ve done

Here at TrendCT, we’ve generated several stories using the previous version of the dataset, but there’s room for much more analysis and creativity:

What do you think?

  • Joseph Brzezinski

    I wonder if the data being collected is sufficient for the kind of discrimination analysis desired. Demographics of towns are apparently totally ignored on the assumption drivers may not reside in the town incidents occurred. That may not be valid for some purposes, say, if a particular ethnic group being stopped is in an area where the vast majority are another. Likewise, data indicated in the data dictionary has nothing about the vehicle such as make and model or age or operating condition – characteristics related to driver income and ethnicity. Maybe the hackathon will surface some such data.