>> The original study was supposed to select motorway lengths with and without roadworks and with
>> and without cameras. Without knowing which lengths were used and when again we can't say
>> for certain they were comparing like with like. I'd hope they were as otherwise it
>> negates the whole purpose of the study!
Yes, although it really is tricky to get this right. The only real way, IMHO, is to take a stretch of road that doesn't have cameras, record the accidents, then add cameras, record the accidents and compare.
Presumably there are plenty of roads that have had cameras fitted, so that they could use the accident data from those. It is not perfect but would seem to be the only real like for like test, but wouldn't really work for temporary roadworks.
Trying to control for the differences between roads, by using large amounts of data, probably isn't going to work, because there are numerous factors, not least that roads with cameras could quite likely have been selected to have cameras because they are, or are considered to be, more dangerous.
Last edited by: SteelSpark on Fri 12 Mar 10 at 17:24
|