Speed cameras do not save as many lives as claimed because of a statistical flaw in the calculations used to justify their deployment, a study has found.
Linda Mountain, a senior lecturer at Liverpool University, estimates that the 6,000 or so speed cameras in the UK save about half as many lives as claimed, although they are an important source of revenue.
Official statistics suggest that speed cameras save on average about 100 lives a year based on the fall in fatal accidents at the recognised accident hotspots where they tend to be placed. But Dr Mountain found that about 50 per cent of the decline in accidents would still have occurred irrespective of whether a speed camera was deployed, due to a statistical phenomenon known as regression to the mean.
The phenomenon is a direct result of placing a camera at a site with a disproportionately high number of accidents – by chance alone it is likely that there will be fewer accidents after the camera is in place. "Regression to the mean and trend effects do not mean that speed cameras have no effect on accidents, only that care must be taken when estimating the size of this effect," Dr Mountain told the science festival.
"The effect of regression to the mean is that if we install a speed camera, or just stick a photo of a camera on a road map at locations with very large numbers of accidents, the camera – or the photo of a camera – will appear to be successful in reducing accidents," she said.