Tribune methodology for studying suburban red-light cameras
To determine and confirm where red-light cameras existed that were under IDOT oversight, the Tribune
sought lists from IDOT, poured through thousands of permit records at IDOT’s Schaumburg office and
drove throughout metro Chicago. Reporters determined there were 184 intersections with at least one
approach section covered by a camera, and there were 287 approaches with cameras. To determine
when permits were issued, the Tribune reviewed paper and electronic records both from IDOT and
individual suburbs which were granted permits.
To determine IDOT’s danger score of each intersection, the Tribune obtained from IDOT data detailing
its periodic reviews of intersection crash levels. The reviews are used by IDOT to determine, based on
the number and severity of crashes, intersections that could use the most attention. The most recent
reviews offer a detailed numeric score and put intersections into five danger categories. Previous
reviews offered numeric scores but did not offer detailed category breakdowns. The Tribune, using the
same methodology IDOT used in more recent studies, was able to calculate for previous studies which
category and intersection would have been placed based on its score. To determine the danger level at
the time IDOT issued the permit, the Tribune studied the danger levels of the two studies that would
have been the most recent at the time of the permit. Two studies were chosen because, sometimes, the
permit process can take more than a year before a permit is issued, and IDOT could have based a
decision, in part, on a previous study’s designation. To give the agency the benefit of the doubt, the
Tribune took the highest danger ranking of the two studies for its analysis.
The Tribune separately obtained data that listed specific metrics for every crash reported to IDOT from
2004 through 2015, along with the location of each crash. The Tribune focused on the subset of crashes
deemed related to an intersection, similar to how IDOT analyzes the data for camera permits. Using
mapping software, the Tribune determined every such crash that was within 165 feet of an intersection
with a camera. Because some crashes didn’t list coordinates, the Tribune used other metrics to
determine if those crashes occurred near a specific intersection being studied. Reporters matched more
than 40,000 crashes to the 184 intersections.
For each crash, the Tribune attempted to determine which approach leg had been traveled before the
crash by the vehicle blamed for causing the crash. Intersections with non-traditional directions of travel
were analyzed, with approach directions adjusted to the closest corresponding traditional direction of
travel: North, South, East or West. Crashes at those intersections were also adjusted to conform to
traditional directions of travel. Crashes at those and other intersections that still listed non-traditional
directions of travel were adjusted based on other codes in the data, such as whether police said a
vehicle was turning a certain direction before a crash. (An example: A vehicle at a traditional
north/south/ east/ west intersection that was listed as heading northeast would be coded as originally
heading north if it was also coded as turning right at the time of the crash, or east if it was coded as
turning left.)
IDOT said it focuses its crash analysis on whether an approach leg had at least an average of three
“correctable” crashes per year. It first looks for crashes in which vehicles hit at an angle or one vehicle
was turning, regardless of whether a red light was involved. It reviews paper reports and can adjust
down the tally based on extraneous circumstances, such as whether a driver ran a red light but was part
of a funeral session. To determine whether IDOT approved permits at intersections that met the
threshold of three a year, the Tribune counted any angle or turning crash at an approach in the three
calendar years prior to a permit being issued – regardless of what paper reports may have indicated. By

 doing so, the Tribune analysis gave the benefit of the doubt to IDOT that an approach leg met the
threshold.
IDOT acknowledged that its crash analysis may include crashes that did not involve someone running a
red light. To determine how many crashes were specifically attributed to someone running a red light,
the Tribune focused on the only two indicators in the data that are specific to the scenario: a driver
disregarding a traffic control device or turning right on red.
Those crashes were separately tabulated and analyzed. IDOT told the Tribune it did not do that detailed
of an analysis because it felt it was too difficult.
To view the results of the analysis, please view the graphics attached to this article.