Why the Tribune’s Red Light Camera Story Is Garbage Journalism

The Chicago Tribune’s reporting ignored the human toll of car crashes at intersections with red light cameras in failing to consider the severity of injuries in right-angle crashes. Image: Wikipedia

In a huge front-page story Friday, the Chicago Tribune published yet another installment in its long-running vendetta against the city’s photographic traffic enforcement program. Because the Trib chose to obscure key information about the severity of crashes, the story is worthless as an evaluation of the city’s red light camera program.

The article lavishes attention on a $14,000 study “commissioned by the Tribune [that] concluded the cameras do not reduce injury-related crashes overall.” But if you manage to get 2,000 words into the article, authors David Kidwell and Alex Richards acknowledge that nine years ago the Federal Highway Administration also commissioned a study on red light cameras. And if you take a close look at the FHWA study, it debunks the entire premise behind the Tribune’s analysis.

The FHWA employed a methodology that closely resembles the Tribune’s, with one all-important difference: The feds incorporated the severity of crashes into their calculations. Both studies found that red light cameras tend to prevent right-angle crashes, while rear-end crashes increase. But since FHWA also acknowledged that right-angle crashes are more severe and impose higher costs on society than rear-end crashes, it found that even with increases in one crash type, the benefits of red light cameras outweigh the costs.

On its website explaining the FHWA study, the University of North Carolina’s Highway Safety Research Center clearly states: “Since the angle and rear-end crashes are of different severities, you must combine both the change in frequency with differences in severity in the analysis. This is why looking… just at changes in total crash numbers is not correct.” While the Trib interviewed a UNC researcher about camera site selection, it failed to note this basic conclusion that upends the paper’s own methodology.

Put simply, the Tribune’s methodology ignores the most important factor – the number of people killed and the severity of injuries sustained at intersections with red light cameras.

The FHWA didn’t make that mistake. Factoring in “the lesser severities and generally lower unit costs for rear end injury crashes” the agency concluded that red light cameras achieved $14-18 million in savings to motor vehicle occupants in urban and rural intersections in seven municipalities. Only thousands of words into the piece, when mentioning the FHWA study, does the Tribune admit that there is a difference in severity of different crash types.

The Tribune also drew unwarranted conclusions from its own study, which found that the change in total crash numbers was not statistically significant. The study itself says “the increase in crashes may not necessarily be because of [adding red light cameras], but may just have happened by chance.”

But the data the Tribune collected is sufficiently robust to bolster the conclusion that red light cameras reduce right-angle crashes.

“[The] Chicago Tribune’s study confirms that these cameras help to reduce dangerous angle crashes that are more likely than rear-end crashes to cause serious injury or death,” Chicago Department of Transportation Commissioner Rebekah Scheinfeld said in a statement today. They’re less severe because they’re slower – drivers are decelerating, and going the same direction, during rear-ends – and cars have crumple zones in the front and rear, but not the sides.

CDOT, responding to the Tribune, said it is proceeding with a review to make sure the cameras are placed where they’re most effective, specifically where there’s a demonstrated problem with angle crashes. Additionally, previous research [PDF] by one of the researchers behind the Tribune’s study suggests that red light cameras work best at intersections with four or more red light-running crashes per year.

Based on their review of crashes at a subset of all red light cameras the Tribune’s study authors said this group comprises about 40 percent of cameras. There’s validity in this recommendation and Scheinfeld has recognized it. She told aldermen gathered at an October hearing that she has continued cleaning up the red light camera program the Emanuel administration inherited, and has even shut down several cameras this year at some low-crash intersections.

Trib’s Shoddy Journalism Provides Fodder for Mayoral Candidates

Today, 2nd Ward alderman and mayoral candidate Bob Fioretti announced that he will be introducing an ordinance to shut down red light cameras by the spring.

He wasn’t the only mayoral candidate to latch onto the flawed premise of the study. Mayoral candidate Jesus “Chuy” Garcia, a Cook County commissioner, said that Chicago’s red light cameras are “pickpocketing” people. People who broke a traffic law and put other motorists, pedestrians, and bicyclists in danger. (Garcia, as it happens, does support cameras when it suits him: He voted for a red light camera in suburban River Forest and also accepted a donation from Safespeed, the would-be camera operator.)

Even in the Tribune’s reporting, none of the experts recommended putting an end to red light cameras in Chicago. The closest anyone came to that was Joseph Hummer, a Wayne State University engineering professor, who said in a follow up article that CDOT “[needs] to reimagine the entire program and do it well,” he said. “Do it right this time.”

Both the FHWA and the National Highway Traffic Safety Administration, which sets standards for vehicle safety features and conducts traffic safety research, endorse using red light cameras to reduce injuries to people. But the Tribune continues to obscure the safety benefits of automated enforcement and produce stories that lead political opportunists to take aim at the program. This is putting Chicagoans at risk: The data is out there for the Tribune to use, if it actually cares about preventing injuries and deaths, that red light cameras improve safety on the streets.

  • Excellent rebuttal.

  • Aaron Berlin

    Maybe we should add tailgating cameras to cut down on the number of rear-end collisions? I’m sure that’s what the Tribune/Fioretti etc. is actually worried about.

  • 1976boy

    There is an entitlement mentality to driving that always amazes me; it makes people who are otherwise rational to ignore common sense and insist on unlimited freedom for drivers, despite the costs in damage and deaths.

  • Anne A

    In the worst cases, where drivers are acting out in very immature, selfish ways, I would consider these situations “vehicular tantrums.”

  • Anne A

    Doesn’t matter whether it’s on local streets or highways. Too many drivers have either forgotten the “2 second rule” or choose to disregard it. I see way too many close calls out there due to tailgating.

  • The 2 second rule should really be the 2-10 second rule as the distance needed to decelerate prior to rear-ending the car in front of you increases the faster you go.

  • That’s an interesting technology challenge.
    We could also reduce speeding.

  • Let us say a prayer for those noble souls who sacrificed life and limb to spare some crumpled bumpers.

  • JacobEPeters

    the 2 second rule does increase the distance needed to stop as speed increases, but I agree at faster speeds I lean towards 4 second rule…then someone cuts between me and the car ahead of me, causing me to slam on the brakes.

  • Fred

    Collision avoidance systems that automatically apply the brakes are already starting to appear in cars. They will be standard equipment in the next decade.

  • JacobEPeters

    Thank you, I was trying to explain this to a coworker yesterday, and this article doesn’t even get into the effect that ticketing traffic violations has on decreasing more egregious motor vehicle crimes. The responses from the mayoral candidates perfectly exemplifies why I can’t vote for either. Which is sad because we need a real debate about issues not reactionary point scoring based on assumptions rather than looking at the totality of a complex issue.

  • Oppelganger

    Thanks Mayoral hacks! This program was marred fraud from the outset and is a disservice to the people of Chicago- but glad you want it to stay! You must love having robots replace our police. Glad you’re counting the beans over severity of injuries rather than total number of injury-causing accidents that the Trib study looked at- cuz making sure our injurious accidents are cheaper is more important than making sure they’re less-frequent, right?

    By the way, where was the FHWA analysis conducted?? Pretty sure it wasn’t in Chicago (like the Trib study was) or on the basis of CHICAGO date (like the Trib study was). You also forgot to mention that the red light cameras fail to meet the minimum 3-second standard required by the same federal transportation regulations.

    Not serious journalism…

  • The Tribune’s commissioned study had the same conclusions as the FHWA’s study and nearly every other red light camera study that used the same or similar methodology. This is evidence that the results of a red light camera program will be the same anywhere: rear-end crashes go up while angle (“T-bone”) crashes go down.

    The severity of injuries is more important than the raw number of crashes, or even the percent change in crashes after an RLC is turned on at an intersection, because the cost to the vehicle occupants who are injured, and to their families, are higher (both in real dollars expended for recovery and personal/mental health issues) in angle crashes than in rear-end and other crash types.

    The City of Chicago *never* changed the 3-second standard for yellow lights at traffic signals. It only changed the threshold for which it accepted a violation from 3.0 to 2.9 in February and back, from 2.9 to 3.0 in September. This was legal and the Inspector General found that nearly all of those who received this ticket (about 37% of all RLC tickets issued in that time period by estimate) were indeed running a red light. We’ve covered this previously.

  • Print it out. Or email it. Thanks for sharing!

  • tooter turtle

    As we witnessed this morning, when a closure of NB LSD (due to a fatality collision) led some drivers to cross park land and drive on the LFP. A person has to feel pretty damned entitled to damage public property and endanger those on foot or on bikes. But we already know that drivers will take crazy risks with their lives (or more typically other’s) to ‘get ahead’.

  • drivebetter

    Wow, did this really happen? Totally unacceptable and typical, if so. Got a link/pics?

  • tooter turtle
  • Chris Chaten

    The Trib study did isolate crashes to injury crashes; it doesn’t count all crashes. There’s not enough data in the published study to indicate the national study is applicable locally.

    The 3 second yellow standard is empirically too short based on standards, but that’s largely a separate issue, aside from making the optics of the program dubious (revenue vs safety).

  • Jmk

    Steve – if you want to see garbage, you should look in the mirror. I don’t know why, when presented with facts, you continue to support cameras.
    You support the illegal issuing of tickets (short yellow lights), and you state it’s ok for the number of accidents to increase.

    You have no integrity and nothing but a hack. And I would say this to your face.

  • Chris Chaten

    I was taught 1 car length per 10 mph.

  • Shlabotnik

    JMK, I am all for serious debate about the red light cameras but name calling is in no way a serious debate. Make your argument about the program and data and skeptics are more likely to appreciate what you have to say. Name calling gets you nowhere and it never will.

  • Matt Florell

    A few words on the 2005 FHWA red light camera study that you hold up here:

    – The 2005 FHWA study uses data collected in differing ways using differing specifications by different jurisdictional entities at various state, county and city levels. As such, there are several mentions of inconsistencies in the data collected

    – All of the data used in the FHWA study at this point is at least 11 years old, including the economic impact figures that are based on 2001 numbers which have undoubtedly changed in the last 13 years.

    – In spite of the fact that the FHWA had stated that this study would be updated it has not been updated in the last 9 years.

    – The lack of any meaningful crash data per jurisdiction being shown, as well as the fact that the municipalities were guaranteed anonymity in the limited reporting of crash data that is shown means that the data is not verifiable

    – Since fatalities were excluded from the crash costs in the FHWA study, and the fatalities increased during the study period, if those fatalities are included in the calculations the crash costs are shown to increase after red light cameras were installed

    The FHWA study has more problems than this new Chicago Tribune study has. The Tribune study is verifiable and repeatable, it uses consistent data sources, it is from a very recent time period and it does not exclude some of it’s statistics from portions of it’s analysis like the 2005 FHWA study does.

  • Matt Florell

    You conveniently ignore that in the 2005 FHWA study fatalities actually increased after red light cameras were installed, and that they excluded fatalities from their economic impact calculations. If they had been added into the economic impact, the crash costs after cameras would have gone up, not down.

    As for the 2.9 second yellows, the FHWA says yellows lights should be a minimum of 3.0 seconds, and in a separate document they state that that electrical components can be off by as much as 100 milliseconds. That is not the same as saying you can ticket people for electrical anomalies. It is a gray area and not in the spirit of the law to ticket people for electrical anomalies. Whether it was technically legal or not, it was wrong for them to do it.

  • Guest

    As was I

  • kastigar

    Plus one extra car length added.

  • ohsweetnothing

    Unreal.

  • hello

    Now who will I vote for for Mayor? I want, not-Rahm, but this pandering to voters by going anti-traffic cam has really thrown me for a loop.

  • I believe that fatalities were included in the crash costs of the FHWA study. Look at Table 3 where it says there, and in the preceding paragraph, that they made four groups:

    1. No injury, angle crashes.
    2. No injury, rear-end crashes.
    3. Killed + A, B, C injuries, angle crashes
    4. Killed + A, B, C injuries, rear-end crashes

  • Juvenal66

    Mr Vance nice attempt to cherry pick favorable arguments. Curious you do not mention Erke’s meta-analysis of all available literature — even that financed by the red cam companies and their fellow travelers at the IIHS — which also concludes there is no safety benefit to redcams: http://www.ncbi.nlm.nih.gov/pubmed/19664425

  • Matt Florell

    They may show that label in the table(which is a standard table used in other similar studies), but they didn’t actually include fatalities in their totals. When researchers at the University of South Florida in 2011 tried to recreate the 2005 FHWA study, the numbers didn’t add up, so they asked the FHWA why fatalities weren’t included in their study:

    “They excluded the cost of fatal crashes in their economic analysis because the cost of a single fatal crash “could significantly bias the results” due to the limited number of fatal and serious crashes in their study. In other words, the authors spotlight the statistical difficulties of including the cost of fatalities, while ignoring the practical implications of such events.”

    “Using their data, the actual estimated cost of an angle injury crash was $82,816 before RLCs and $100,176 after RLCs were implemented”

    http://health.usf.edu/NR/rdonlyres/2511FA2D-6BC2-4091-9FD5-DBF711F420AA/0/2011pp00109FPHROrbanetal.pdf

  • Guest

    And from the abstract on the website that you link:

    “In Red Light for Red-Light Cameras?, Alena Erke concludes that “… on the whole, redlight cameras do not seem to be a successful safety measure.” Although Erke’s survey of the literature appears to have been comprehensive, her review of the studies was not critical. She appears to have accepted the authors’ descriptions of their analyses rather than providing readers with her own considered opinion of how valid those analyses were and what their true implications might be. For the meta-analysis leading to her final conclusion, Erke combines data from two questionable studies with three other “well-controlled” studies. Non-peer-reviewed studies received substantial statistical weight in the meta-analysis. These problems likely produce misleading results. If the highway safety field is to succeed in identifying for policymakers those strategies that are most likely to reduce the human tragedy of motor vehicle crashes, we need first to focus on conducting valid research and analysis. Adding precision to the estimated benefits of those strategies through meta-analysis is important, but secondary, and cannot replace the function of a systematic and critical review.”

    Lies, damned lies, and statistics can be employed on both sides of an argument.

  • Guest

    I can’t speak to some of their other criticisms of red light cameras in general, but I consider it a red flag that they take FHWA’s fatality numbers, which were excluded specifically because of their small sample size, and try to re-spin the numbers by including them.

    Reporting that fatal angle crashes were “10% higher” after introduction of cameras and that overall fatalities increased by 33.6% seems a tad irresponsible with so few data points – and especially since the base from which they are calculating the increase is 4.8 fatal angle crashes and 1.28 fatal crashes per 100 injury crashes respectively.

  • David Altenburg

    And yet, compare the quality of the discussion here with the comments section of any website of “serious journalism”.

  • Guest

    A few published responses to the article (and responses to the responses) also raise some interesting points (including a detailed rebuttal from the former Director of USF’s CUTR):
    http://hsc.usf.edu/NR/rdonlyres/E07FD238-8640-4EB9-8CFE-85B1100F1792/0/2008pp047052FPHRLetterstotheEditorMarch2008_2_.pdf

    The takeaway? Smart, informed people have come to different conclusions. I would be very careful giving more weight to the Tribune article than others. Especially since there’s a significant profit motive to the Tribune’s research (controversy = $)

  • Matt Florell

    I consider it a red flag that the FHWA included a label for fatal crashes in their data table(K = Killed), but purposely excluded the fatality data. They also did not mention this omission anywhere in their published summaries, you have to read the full 98-page report to find that out.

  • Matt Florell

    I would argue that the profit motive of the City of Chicago is many times greater than the Tribune on the subject of red light cameras. And given the games that the City of Chicago has been playing with homicide statistics in the last few years, they have a well documented recent history of data manipulation.

  • Hugh

    The Tribune-Texas A&M study is vastly superior to the City of Chicago’s own evaluations of its program in place since 2002. Oh, wait, the City has never evaluated the program. Why are you criticizing the Tribune for doing something? I will look forward to your follow-up post, “Why the City of Chicago’s Negligence in Evaluating Red Light Camera Safety is Garbage Governance.”

  • Hugh

    “…the Tribune’s methodology ignores the most important factor – the number of people killed and the severity of injuries…” Great idea: consider severity. The City of Chicago’s Commissioner of Transportation did not include severity in her testimony in defense of the red light camera program in a public hearing before the City Council in October, 2014. The City’s report on crash statistics, linked from its website in defense of the red light camera program, makes no distinction between the severity of injuries. Why are you calling a local newspaper garbage?

    http://www.cityofchicago.org/city/en/depts/cdot/supp_info/red-light_cameraenforcement.html

  • Hugh

    “The data is out there for the Tribune to use, if it actually cares about
    preventing injuries and deaths, that red light cameras improve safety
    on the streets.” An impressive contribution to the public dialog, because in my humble experience most Chicagoans have questions about the City’s motives behind the red light camera program.

  • Hugh

    Starts out as a promising methodological comparison between two studies, but you did yourself a disservice with the turn to the mayoral challengers in the 2nd half. I’m sure the shot of Fioretti will help depoliticize this public safety issue. Nice touch.

  • tooch

    Is the current system perfect? Nope. But, I don’t want police in Chicago chasing people who go through red lights. Their time is better spent focusing on more serious crimes. Automating tickets for red-light offenders makes a ton of sense from a resource management prospective.

  • Guest

    I won’t dispute the profit motive from the city and I expect and welcome the scrutiny into the program. I believe there are lots of shady things involved in the program, including the vendor corruption fiasco, questionable yellow light timing, etc. I wouldn’t ask anyone to believe the city any more than the Tribune.

    My point is simply that there are many disinterested parties with credible evidence to support both sides of the argument. I wouldn’t lump the Tribune study in with those disinterested parties.

    I also take issue with the conflation (seemingly purposeful in some quarters) of red light camera programs in general with Chicago’s specific implementation of the cameras. No one on this website or elsewhere would defend the corruption at the heart of Chicago’s program. But that doesn’t mean RLCs, properly implemented, are inherently corrupt or somehow unfair to motorists.

  • We’ve previously covered CDOT’s gross mismanagement of the red light camera program and how CDOT has implemented new checks and is re-evaluating camera replacement.

    1. http://chi.streetsblog.org/2014/10/17/inspector-general-issues-a-reality-check-on-tribs-red-light-cam-spin/

    3. http://chi.streetsblog.org/2014/10/30/cdot-tells-council-improved-red-light-cameras-might-mean-more-tickets/

  • Hugh

    I’m re-reading the head post above looking for anything vaguely critical of the City. I’m not finding anything and I don’t understand why. I find: “…the Tribune … is putting Chicagoans at risk:” Wow. An astonishing alternate reality.

  • LIM

    If u think those cameras r there for safety.. U r in Rahms pocket… The lights give tickets for yellows that r not 3 seconds longs and turning on right on red with no signage that says u can’t… Total garbage … I’m very scared of anyone who wants to live in a city that videotapes everything u do then ur nothing but a sheep

  • Norm

    When the city collects Thousands of Dollars in tickets at a given intersection and does absolutely nothing to improve safety at that intersection – beyond cashing ticket payment checks – it’s all about the revenue – pure and simple! Plus, since there’s no penalty beyond fines, the wealthy can run all the lights they can afford to run while the poor pay the price. Revenue Rahm says he pays all the tickets his motorcade continually gets for running traffic lights – no other consequence – nothing about public safety – Police say they have no intention to do anything about it!

  • The mayor’s motivations for installing red light cams don’t really matter, as long as the cams improve safety. As detailed above, they do. If we can reduce serious injuries and fatalities, punish scofflaws, and raise revenue for our broke city with the cams, that’s a win-win-win. If so, who cares if the mayor is mostly interested in revenue?

  • Charlotte Frei

    Does anyone know if the Tribune study did some “yellow light dilemma” analysis? That is, are there intersections in Chicago where 3 seconds is not even long enough to stop if a driver is going the speed limit?

ALSO ON STREETSBLOG

Trib Launches War on Speed Cams, CDOT Releases Data Showing They Work

|
The Chicago Tribune’s David Kidwell and his colleagues have written extensively about the city’s red light camera program. Some of that reporting has been constructive, including revelations about the red light cam bribery scandal, unexplained spikes in ticketing, and cameras that were installed in low-crash locations during the Richard M. Daley administration. Other aspects of the […]