Lies and damn statistics

The regular complaints from business organisations like the CBI and the British Chambers of Commerce about the weakness of the country’s transport infrastructure are always peppered with estimates of the cost. This morning, when I received a missive from the BCC saying that the ‘cost to business’ was ‘a staggering £17.55 bn per year’ I decided to discover how this figure is calculated.
I discovered that it is based on a survey of 2,260 small businesses in which the respondents were asked to estimate the cost to their firm of congestion. Then those numbers are multiplied by the number of small and medium size enterprises (what about big ones?) given by government, to arrive at the global figure.
I am not denying that congestion is a cost to some businesses – indeed it may be a major one to some. But to aggregate up figures from a survey in which business people – who have a vested interest in exaggerating the cost – simply make an estimate of what congestion costs their business is a fatuous exercise at best and a very dishonest one at worst. It has absolutely no basis whatsoever, not even as an approximation.
This technique, though is effective. These figures are regularly reproduced in newspapers, frequently in headlines, without any analysis of how they are arrived at. They then get picked up by policymakers and become part of conventional wisdom.
In my youth, I was active in squatters’ organisations and we used to get rung up by journalists keen on finding out how many squatters there were. We trotted out random numbers such as 40,000 in London, with little basis in fact – there were probably a few thousand at most – but these figures can now be found in learned tomes.
The BCC and other organisations pride themselves on being rather more respectable than the Advisory Service for Squatters. But in their use of statistics, they are not!