The Royal Society for the Prevention of Accidents (RoSPA) has jumped to the defence of speed cameras, after the Association of British Drivers (ABD) yesterday advocated a report that said they were ineffective in reducing collisions.

RoSPA has slammed the ABD's suggestion that they actually cost lives and questioned the validity of the research, which was undertaken independently by engineer Dave Finney and based on a study of cameras in the Thames Valley area.

Far from costing lives, says RoSPA, stats show that speed cameras save lives and prevent injuries on a significant scale.

Speaking to Autoblog, RoSPA's Jo Bullock said: "In Thames Valley itself, an evaluation of fixed speed cameras in 2009, comparing three years before and after each camera was installed, found an overall drop of 94 KSI (killed or seriously injured) collisions and 802 personal injury collisions - a reduction of around 38%. At mobile camera sites, KSI collisions reduced by 61%, and all injury collisions by 30%.

"This latest report only examines one year before and immediately after (although some of the data tables say one year before and two years after) rather than the normal three years before and after periods. Even so, the report finds a reduction of 2.5% at fixed camera sites. However, the report's conclusions interpret this as 'cameras did not result in any reduction'."

RoSPA's speed camera fact website says that during a four-year study period leading up to 2010, speed camera sites saw a 42% reduction in the number of people killed or seriously injured - a total of 1,745 fewer people. Collisions, it says, reduced by 22%, equating to 4,200 fewer.

Autoblog today contacted Mr Finney, the author of the Thames Valley report, for a response to RoSPA's rebuttal of it. Naturally, he has defended his research. He said: "If only 5.5% of collisions involve a vehicle exceeding a speed limit, how can speed cameras cause a 38% or 30% (fixed or mobile) drop in collisions? The reductions at the camera sites were not caused by the cameras, and RoSPA ought know this.

"Sites (selected lengths of road) are generally chosen to have cameras following a spike in the number of collisions, in other words, after an 'unusually' high number. Therefore the numbers tend to drop back to 'usual' over the following years. This is known as RTM (regression to the mean) which, although it sounds complicated, just means 'return to normal'.

"The road safety industry has been unable, or unwilling, to accurately measure or exclude RTM in any of their reports (only estimates and comparisons at best) so I therefore realised I had to do it. As my report shows, there was a significant fall that occurred around a year before cameras (when RTM is expected), but absolutely no fall after the cameras for the first two years of speed camera enforcement.

"Therefore the entire reduction in collisions claimed by RoSPA is demonstrated to be due to RTM with the cameras having achieved no benefit at all.

"If the road safety authorities (including RoSPA) don't understand my report, that might not be surprising because my method of plotting collisions on a graph in order to determine what effect speed cameras have (and then accurately reporting the results) has never been done before, so far as I'm aware."