The BBC is keenly promoting a story about speed cameras.
The cameras have been switched off in Oxfordshire: they still measure the speed of passing motorists, but no longer take pictures. Motorists can therefore speed with impunity and remain point-less.
The shocking news is that, after this change, the cameras have recorded more speeding motorists than before.
In Cowley some 62 people were clocked speeding, representing a rise of 88%.
And in Woodstock 110 drivers were over the 30mph limit, which is 18% more than the average for 2010.
Why daft? Because it tells us nothing useful.
We’d expect more people to speed passing the speed cameras. Previously, everyone drove at whatever speed they drove at everywhere else, then slowed down for the speed cameras.
It seems likely that people are maintaining the same speed they always drove at, but not slowing down any more. Since speed cameras aren’t that common, the average change in driving speeds across the road network may be virtually nil.
It could be, though, that turning off the speed cameras results in more accidents (especially as those cameras are meant to be placed only where there’s been such a problem).
That’s not something it’s possible to tell purely from measuring speed. The evidence for speed cameras reducing accidents (as opposed to reducing the speed of vehicles as they pass the cameras) is mixed – especially when regression to the mean is taken into account.
I don’t know whether turning off the speed cameras will result in more accidents, fewer accidents or make no difference at all. I’ll be very interested to find out – but this evidence tells us absolutely nothing one way or the other.
Share:
22 Comments
Speed cameras reduce accidents
Liverpool Uni showed this in a well-known study: actually a decline of 19% allowing for regression-to-the-mean compared with a fall of 50% if the regression phenomenon had not been allowed for.
so there you have it
similarly in the USA
for example
In 2006, Scottsdale became the first U.S. city to demonstrate the effectiveness of fixed speed cameras on a major highway. Before the cameras were installed, 15 percent of drivers were driving faster than 75 mph (the posted speed limit it is 65 mph). The new IIHS study showed that, with the cameras in place, the number of violators plunged to 1 to 2 percent. The Scottsdale 101 Program Evaluation estimated the total number of target crashes in non-peak periods was reduced by about 54 percent.
The idea that going faster has no effect on accidents, or actually reduces them, will be seen in a similar light to those people who used to say they drove better after a pint.
sorry to labour a point but…
Wiltshire and Swindon Safety Camera Partnership showed:
as at the 31 November 2009, for the preceding three years, at core sites where our cameras are located throughout the county, we have seen a reduction of 68.82% of killed and seriously injured (KSIs) persons. Again, these figures are in comparison to the baseline figures. The cameras really do help to reduce casualties!
Groan, statistics abuse. Let’s see the figures for the entire county, not just the places in front of the cameras. Otherwise all you’re showing is that you’ve moved it to places where cameras aren’t.
Motorways are very different to urban streets.
The evidence is firmly against you on this one. The Germans have studied this carefully and conclusively proved that the accident rate on their unlimited-speed highways is no different to that of their speed-limited ones. The link between speed and accidents only applies to roads with crossings, pedestrians, or other “surprises”. It doesn’t seem to make any difference on motorways.
That study was observing only accidents in the places where cameras were placed on urban roads, gave a result of 8% to 30% decline due to speed change (at 95%), and noted that no consideration was given to whether there was an overall reduction or whether accidents just moved to places where cameras weren’t. It does not prove your claim, and was not attempting to do so; they are quite clear on this. They were investigating a new model for calculating accident rates that accounted for more confounding factors (it worked well and is now used).
Looking at the Swindon switch off, the Telegraph Reports (http://www.telegraph.co.uk/motoring/news/7931842/Speed-camera-switch-off-sees-fewer-accidents.html)
“At the four speed cameras sites themselves, there was one fatal, one serious and 13 slight accidents in the same nine months in 2008-09. In the same period after the cameras were switched off, there were no fatalities, two serious and 12 slight accidents.”
Polite reminder to commenters: please keep on-topic in the thread.
To quote our comments policy:
Labour away – proper studies are good, random information about the speed of cars passing a couple of cameras isn’t.
Speed cameras only cut death by 19%! That’s rubbish; let’s bin them all.
It seems to me that your article implies that you don’t think speeding is dangerous, and it is up to campaigners to prove that it causes accidents. Really they shouldn’t need to do that, not on the Liberal Democrat Voice anyway. Leave the junk science to Christopher Booker and Melanie Philips and their friends on the Daily Mail and the Daily Telegraph.
I struggle to see why anyone would think that people will drive in a more accident-prone way than they otherwise would have done in other areas. You might think the opposite – that the existence of speed cameras would remind people that speed can kill, that people with many points on their licences would be particularly careful, that people not sure where the cameras are would be careful, and that people with cruise control would just set it for 50 in a 50 section, knowing that there is a camera somewhere.
It was blindingly obvious that people would come along and rubbish the statistics. The wilful misrepresentation of the Liverpool Uni study by Andrew Suffield is funny. I would urge interested readers to investigate the study (and others) and make their own minds up.
Using a study on autobahns to try and convince people that speed is not a factor in accident reduction at UK speed camera sites is a good one.
I also liked this: “That study was observing only accidents in the places where cameras were placed on urban roads, gave a result of 8% to 30% decline due to speed change (at 95%), and noted that no consideration was given to whether there was an overall reduction or whether accidents just moved to places where cameras weren’t. ” Just think about it…
Mr Suffield, maybe we should up the speed limit in urban areas to reduce the number of accidents, injuries and fatalities? Make it especially high around sensitive sites like schools perhaps?
Jane – the question isn’t whether speeding is dangerous, it’s simply whether speed cameras make our roads safer. As I hope I made clear, I don’t happen to know the answer to that question, but I’m confident the story being plugged by the BBC today tells us nothing useful either way.
More like daft ‘Liberal Democrat Voice’ story of the day.
Iain R
the information is there – and the best information available says that they do make our roads safer, not by as much as the government said but still saving something like 50 lives per year. It’s not perfect science, there are flaws in the methodology and analysis, but the overall point remains true.
And let’s face it, we all know it’s blindingly bloody obvious, some of us just don’t want to admit it.
@Mark – thanks for the inciteful comment.
@norfolk – quite possibly: I don’t think I ever suggested otherwise. (Although I’d disagree about the blindlingly obvious bit – plenty of things that are blindlingly obvious to some turn out not to be true).
Norman Baker, the Transport Minister, says it’s got nothing to do with saving money. It’s simply a way of allowing communities to decide for themselves how to deal with speeding traffic. Unrelated but another example of
new coalition logic is Chris Huhne claiming that he was never against nuclear power provided it didn’t involve
public money. New boys they are but the spin is familiar.
You’re the one who started talking about high-speed US highways.
(Rest of the posts from the pseudonymous troll had nothing of substance to respond to, it’s now just going “I’m right I’m right I’m right”)
In the past, the argument for getting rid of speed cameras was that they should be replaced by more police traffic patrols, who could watch for all types of bad driving, not just speeding. However, with cuts in police budgets, it seems likely that we will lose speed cameras AND police traffic patrols, making our roads a great deal more dangerous.
Road accidents have been on a downward trend in recent years which seems to suggest that road safety policy (including the use of cameras) has been working.
As for the argument that cameras are unpopular with the public, most surveys show public support for cameras and every year councils get many requests for new cameras to be installed in places where speeding is a serious problem. Cameras are very unpopular with parts of the tabloid press and this decision seems a clear case of pandering to the Daily Mail’s agenda.
If the coalition believes so strongly in localism, why not allow councils to retain the revenues raised from motorists caught speeding and use that funding to maintain the cameras?
No no, you’re missing the point. It’s trivial to show that speeding on urban roads is dangerous. What has very definitely not been proved is that speed cameras do anything useful in reducing speeding on urban roads. All the evidence suggests that they just make speeders drive down different roads where there aren’t cameras. Cameras are annoying enough for people to want to avoid them and easy enough for people to be able to avoid them, so they’re not a good bet. We don’t have hard numbers on exactly how well they compare to other alternatives, but there are certainly plenty of other options that look more promising.
@Andrew Suffield
“All the evidence suggests that they just make speeders drive down different roads where there aren’t cameras.”
Can you provide some references to definitive studies to support such a sweeping statement?
Why are we so determined to have all or nothing in this country. Applying that almost extinct commodity called common sense it is blindingly obvious that there are some places, the Stocksbridge by-pass comes to mind, where almost saturation camera coverage of what was a highly dangerous road has had a beneficial effect. There are other examples, but as one drives around the ‘good’ camera sites are overwhelmed by those where the siting is purely for the purpose of raising revenue. Also, If the statistics are to be meaningful at all, we need to know how many fatalities/serious injuries are attributed to exceeding the speed limit by 20% or less. As far as I can see serious accidents attributed to ‘speed’ reflect not 36mph in a 30 mile limit but 60+mph & whether this is picked up on a camera is irrelevant. The counter productive impression given by the ‘anti speed’ propaganda is that drivers get into their car with the sole ambition of driving like lunatics & seeing how many people they can frighten to death. What utter rubbish but we should remember the ‘give a dog a name’ syndrome & if one has been flashed doing 36mph when the safe speed for the road is clearly 40 or more one is not likely to say what a naughty boy to oneself. More probable that having been irrationally penalised one will put the toe down in anger & frustration. After all there will not be another camera for a bit.
So Mr Suffield
can you explain the influence you think speed has on accidents?