Welcome to another in my occasional series on useful, interesting or controversial findings from academic studies. Today’s it is the question of speed cameras, for which a study of international evidence has recently been conducted:
To evaluate the effectiveness of speed cameras, the authors examined all eligible studies, that is, studies that met pre-set standard criteria. We analysed the effect of speed cameras on speeding, road traffic crashes, injuries and deaths by comparing what was happening in road areas before the introduction of speed cameras and after their introduction, and also by analysing what was happening in comparable road areas where no speed cameras were introduced during the study period.
The authors accepted a total of 35 studies for review which met the pre-set criteria. All studies reporting speed outcomes reported a reduction in average speeds post intervention with speed cameras. Speed was also reported as either reductions in the percentage of speeding vehicles (drivers), as percentage speeding reductions over various speed limits, or as reductions in percentages of top end speeders. A reduction in the proportion of speeding vehicles (drivers) over the accepted posted speed limit, ranged from 8% to 70% with most countries reporting reductions in the 10 to 35% range.
Twenty eight studies measured the effect on crashes. All 28 studies found a lower number of crashes in the speed camera areas after implementation of the program. In the vicinity of camera sites, the reductions ranged from 8% to 49% for all crashes, with reductions for most studies in the 14% to 25% range. For injury crashes the decrease ranged between 8% to 50% and for crashes resulting in fatalities or serious injuries the reductions were in the range of 11% to 44%. Effects over wider areas showed reductions for all crashes ranging from 9% to 35%, with most studies reporting reductions in the 11% to to 27% range. For crashes resulting in death or serious injury reductions ranged from 17% to 58%, with most studies reporting this result in the 30% to 40% reduction range. The studies of longer duration showed that these positive trends were either maintained or improved with time.
The quality of the included studies in this review was judged as being of overall moderate quality at best, however, the consistency of reported positive reductions in speed and crash results across all studies show that speed cameras are a worthwhile intervention for reducing the number of road traffic injuries and deaths.
To affirm this finding, higher quality studies, using well designed controlled trials where possible, and studies conducted over adequate length of time (including lengthy follow-up periods) with sufficient data collection points, both before and after the implementation of speed cameras, are needed.
12 Comments
Report can be downloaded at http://onlinelibrary.wiley.com/o/cochrane/clsysrev/articles/CD004607/pdf_fs.html without going via s–dd–g b—-y facebook
So cutting speed cameras is a false economy.
What you save in not setting the system up, you lose in terms of the costs for the emergency services. The misery is being made to adhere to the law is nothing like as bad as the misery of being mangled up in a traffic accident.
This research is the key test as to whether the policy should be implemented or not. Don’t be surprised is libertarians find fault with the research, just as they do with global warming or the findings of the Spirit Level (http://www.amazon.co.uk/dp/0141032367/?tag=libdemvoice-21)
There’s a download button at the top of the embedded report – that gives a pdf download, so I don’t quite follow your reference to Facebook?
Very poor analysis of the effects of Regression To The Mean.
Put a hidden flag up and the rate of accidents will be lower in future months from whatever peak accident the flag started from.
For a sequel you could give us the Cochrane review on ‘Area wide traffic calming for preventing traffic related injuries’ by Bunn et al., It was one of the first things I read when I became a councillor given my residents enthusiasm for traffic calming 🙂
Please add possessive apostrophe to residents’
“There’s a download button at the top of the embedded report – that gives a pdf download”
Tried it – It takes one to Facebook!!!!!!!!!
It’s working here ok giving a pdf download, so do you want to email me further details of exactly what you’re doing and what browser you are using? [email protected]
One of the key issue in studies of the effect of speed cameras on accident rates is “regression to the mean” (RTM). See this article for a detailed discussion. The argument is basically that if you put speed cameras at places where there have been more crashes due to random fluctuation, the law of averages says that the number of crashes is going to decrease anyway.
So it’s disappointing to see that the report says (page 34), “Only ten studies out of a probable twenty four studies, where RTM may have been a sizeable factor, either described and/or controlled for its effects.” Furthermore, it adds “Likewise, only a small number of studies controlled for other long term trends in crash rates and changes in traffic volumes.” Bizarrely, in its conclusion the report says, “…speed cameras do reduce road traffic crashes, as well
as those resulting in road injuries and deaths. However, in this study, an overall magnitude or significance of this effect could not be deduced” So speed cameras do reduce crashes, but we can’t say how much by or even whether it’s significant, despite looking at so many previous studies.
This seems a rather slender basis for drawing the conclusion that speed cameras are wonderful and anyone who disagrees is a raving libertarian bent on ignoring the evidence.
HF: as Phil highlights, the authors looked at regression to the mean, concluding that it wasn’t a sizeable factor in 11 of the studies and was controlled for in some of the others.
So what’s “very poor” about their analysis of the issue? Isn’t checking to see whether studies allowed for it and whether it could have had a sizable impact good analysis?
Of course ignoring it would have been very poor, but that’s not what they did?
Geoffrey Payne. I realise this is off topic, but I would not describe myself as a libertarian, except in the broad JS Mill sense, and I’m pro speed cameras and very concerned about global warming, but even I have a problem with the findings of the Spirit Level.
The flaw in every analysis of the effects of speed limit enforcement on accidents, especially the severity of accidents, is the unquestioned assumption that any given speed limit is sensible & rational. But it clearly is not. Any driver knows of roads where 40mph is perfectly safe on grounds of visibility, road width, distance of houses from the road, likelihood of child running unseen into the road and so on & so on yet there is a 30mph limit based on no science or logic & usually carrying a camera because it makes money & conversely roads where 30mph is downright dangerous yet legal & usually do not have a camera. Also the statistics do not usually distinguish between a collision where vehicles are doing less 20% above this arbitrary limit ( 36mph in a 30 limit) and those that are going seriously fast. If speed limits were sensible, reflected real road conditions & the technical capability of modern cars they would be obeyed because that’s how the vast majority of drivers drive. When the speed limit of 30mph was introduced in 1934 most vehicles could only manage 65mph max. & braking distances were at least twice what they are now.