Official Title: Evaluation of the City of Houston Digital Automated Red Light Camera Program

Authors: Robert A. Dahnke, Benjamin C. Stevenson, Robert M. Stein, Timothy Lomax

This is a 2008 study by the Rice University Center for Civic Engagement on the effects of red-light cameras in the city of Houston, Texas. This study compares the before and after crash statistics of 70 RLC approaches at 50 intersections in that city. The study uses 24 months of pre-RLC crash data compared to at least 12 months of post-RLC data. This study uses a comparison of monitored approaches, not intersections, for it's crash data and analysis. The statistics used in this analysis are not from a defined source and the report did not define what they used as their Distance-From-Intersection inclusion zone for crashes, although the supporting correspondence suggests that 100 feet was used. There are no data or conclusions in this report that suggest a reduction or increase in fatalities due to RLCs. This study does not cover injury crashes. The study concludes that: "The comparison of data between monitored and non-monitored approaches supports the conclusion that red light cameras are mitigating a general, more severe increase in collisions." Another conclusion is: "the absolute number of collisions at camera-monitored approaches is not decreasing"

Download this study below (369k)

This is a brief 16 page report, the data and statistical analysis are not standard or thoroughly explained. The selection of 50 RLC intersections is large number of intersections for this kind of study. The use of "monitored approaches" instead of intersections as the analysis metric is not the standard for this kind of study. In fact, only one other related study(San Diego, CA 2002) uses this method, and it is not explained why they chose this non-standard measurement method. The lack of a defined Distance-From-Intersection inclusion zone for crashes somewhat diminishes the conclusions. The before and after RLC date periods are sequential but only 12 months of after-RLC data is less than optimal for this type of analysis. There is also evidence to conclude that this study was possibly written with bias in mind(external link, external link 2, internal link). The crash data for this same time period in Houston used by this report is also used as a significant portion of the data in the 2008 TxDOT/Texas A&M study which directly contradicts a large sample of the referenced data. Also, data from the Houston Police Department shows that crashes city-wide were decreasing during the time of this study, while this study suggest that crashes have significantly increased. Not all of the conclusions of this study are supported by the data supplied. 

Notes:
A few days before the November 2, 2010 election in Houston, Bob Stein and the Rice University Center for Civic Engagement released the results of a poll showing 55% of likely voters in favor of keeping the city's RLCs and only 36% in favor of removing them(the poll had a stated 4.5% margin of error). When the votes were counted on the night of November 2nd, a very different set of results came out: 47% in favor of keeping RLCs and 53% in favor of removing them, a 17% difference on the RLC removal side, which seems to again call into question the methods of Robert Stein and the Center for Civic Engagement if they can be that far off. Because of this vote, the RLCs in Houston will be deactivated. As of the end of April 2011(five months since the red light cameras were turned off) crashes at former red light camera intersections in Houston have decreased by 16.4%(external link).