Greeny News

media reports this week that some cameras have failed to reduce accidents - and in some places have even led to an increase - ignore the power of random variation

"The radars could not reduce accidents on the roads and many, in fact led to an increase in the number of victims on certain roads," said the Telegraph.

Is this true? He has speed cameras "no" to reduce accidents? Do you have "led" to an increase in accidents? A close examination reveals little of these two bold assertions may be false.

When considering a large number of radars and guidance, we expect by chance alone, there will be sites where the number of accidents increased after installing a radar . This will happen even if the chances of an accident, actually declined.

Any analysis must take into account the randomness. The figures provided by local councils must also allow people to get the full picture, which means giving the public information on how the data are variable. For example, the board does Humberside data on average three years and collisions at low and no indication of the variability around these values.

or not a particular accident happens mathematically is a random event - you can not predict. The number of accidents that occur every hour or every day or every year is a random number that depends on a variety of factors, such as the speed limit on the stretch of road on which a camera is placed in particular, how busy the road is and the time. We can not predict exactly how many accidents at some point, but we can estimate with any degree of certainty.

With this in mind, we can consider applications of radar in the Telegraph. These claims are an example of the falsity of the tax, which may arise when multiple comparisons, as in the years between the tips