On Sat, 15 Dec 2007 19:28:54 -0000, Adam Lea
<
[email protected]>
was popularly supposed to have said:
> Unfortunatly, there will be someone equally nasty and dangerous to take his
> place (assuming he was nasty and dangerous).
All he did was to point out the actual figures for road deaths as
opposed to the hype which the Government had been touting up to that
time.
He also pointed out that no study had been performed to test the degree
to which the regression to the mean phenomenon (that is to say, the
tendency for rare events not to recur in specific areas for a long time)
was affecting the spread of traffic casualties.
To summarise, the Government stance was that if a few accidents occurred
on a stretch of road, then a speed camera installation would cure the
problem.
The problem is, road accidents don't happen very often, and two or three
happening on one stretch of road in six months may be just a
coincidence; pure freak chance, nothing more, and you could then go for
several years without another accident.
So, at that juncture you don't know if any treatment you're applying is
doing anything to help or hinder the problem or if it is doing
absolutely nothing.
The only way you can sort this out is to take candidate sections of road
which meet the criteria for speed cameras, and randomly (flip a coin)
put cameras on some, and no cameras on others. You need to do this a lot
of times, and you need to monitor for a couple of years or so to see if
there's any statistical difference between the two groups of sites.
Paul Smith did this, at one remove, by compelling the Ministery of
Transport to tell him what the accident rates were in different sorts of
motorway roadworks (the Ministry had quietly done the statistics already
by themselves, and said nothing about it).
It turns out that there is a statistically provable difference between
roadworks with Gatso units in, those with SPECS units in, those with no
cameras in and those which were more actively police patrolled.
Of the different types, as compared to no enforcement as a baseline, the
Gatso increases accidents markedly, the SPECS increases accidents
slightly, and the police patrols decrease accidents markedly.
If Paul hadn't asked, we wouldn't know this now.
To date, no statistical comparison of speedcam to no-speedcam sites has
been undertaken, although the two police areas (Durham and North Yorks)
which don't use many fixed cameras are not reporting hugely increased or
decreased road accident rates, which suggests that speed cameras do not
actually do very much at all.
--
Dan Holdsworth PhD
[email protected]
By caffeine alone I set my mind in motion, By the beans of Java
do thoughts acquire speed, hands acquire shaking, the shaking
becomes a warning, By caffeine alone do I set my mind in motion