New Mexico: Report Documents Benefit to Longer Yellows

This was a very interesting study prepared by the University of New Mexico (Institute for Social Research)
Albuquerque Yellow Light Timing Change Effectiveness Study (University of New Mexico, 9/30/2012)
http://thenewspaper.com/rlc/docs/2012/nm-longeryellow.pdf

While it is quite long, I would hope that you will read the report to get an idea of what it takes to do statistical analysis of systems that have multiple variables involved.

The report reviews and cites a lot of research that lengthening yellow timing results in decreased crashes. The report data generally supports that hypothesis. While this report should hearten opponents of red light camera, it should be noted that the shortest yellow timing pre-study was 4.0 seconds and the increases were modest (.2 to .5 seconds). In my opinion, making sure that yellow intervals are appropriate should be a first step before cameras are considered.

While the data set is not large enough to be statistically significant (involving only two intersections), I was quite amazed to read that increasing the All Red Length (ARL) actually increased crashes over the 11 month comparison data. I would have thought that instituting a lengthening of ARL would have decreased crashes.

This report was something thenewspaper.com could have reported without any need for being misleading - but that was not to be. The sub-title of their article reporting this study said "University of New Mexico study finds accidents go down when red light cameras are removed and yellows are lengthened."

thenewspaper.com is planting the impression that the New Mexico study tied the removal of red lights with accidents going down. If I read the following paragraph from the report correctly, the authors of the study actually note a "beneficial effect of the RLC system found in the RLC study".

"It is important to note the RLC system was in operation during calendar year 2010 at most of the study intersections and so the RLC system was in effect during most of the pre-study time frame at most of the study intersections that are part of this study. This is important given that prepost-studies assumes the study group and comparison (or control) group are similar and that changes detected in the study group, but not the control group, were actually a result of the intervention. We are also not able to disentangle the possible safety effects of the RLC system in the post-time period. There is the possibility that the beneficial effect of the RLC system found in the RLC study extended beyond the end of the program. The study intersections and comparison intersections differ in this important way. For this reason it is difficult to attribute changes in crashes and crash rates solely to the increases in the yellow light time intervals and the all-red light clearance intervals. It is important to separate the effects of different improvements because studies have shown that treatments vary in their effectiveness. There is very little research that has studied the relative contribution of yellow light timing intervals and RLC systems on red light running and traffic crashes (Retting, et. al, 2008)."

On the Albq. news last night they said that the company who ran the red light cameras was going to take everyone with outstanding tickets to civil court and the City is backing them up.

I wonder who will pay the process server.