The Norm Chronicles (52 page)

*
Kelvin’s son.

*
See, for example, the murder of the 94-year-old Emma Winnall.
¹

†
A common complaint is that the crime statistics lie. Perhaps people have given up reporting crime because they don’t think anything will be done. Perhaps the police fiddle the data. But almost all figures here are based not on what people tell the police or on what the police record. They are separate surveys of what people say has happened to them personally. They are approximate, as surveys always are, but they are reasonable. These surveys used not to include the under-16s, now there are also surveys for 11- to 16-year-olds. One offence not included in victim surveys is homicide, for the good reason that homicide victims can’t answer surveys. The homicide data are from what’s known as the Homicide Index, maintained by the Home Office.

‡
The mistake of a paediatrician for a paedophile really happened, but not the mob attack sometimes reported.
³

*
Using 2006–7 as a rough guide, we find that stabbing – or, more correctly, killing by sharp instrument – is the most frequent method of homicide, accounting for 41 per cent of incidents in London, with shooting next, at 17 per cent. We therefore estimate that four stabbings has a probability of 0.41 × 4 = 0.028, assuming that the causes of death are independent for multiple murders on the same day, and that the observed rates can be considered as estimates of the risks for each murder, and given that there are four murders in a day.

†
In fact, the Poisson distribution, which has nothing to do with fish but is named after a Professeur Poisson.

*
See, for example, his lecture ‘Longer, Healthier, Happier? Human Needs, Human Values and Science. Sense About Science’, 2007.
¹

*
In one episode of
Scrubs
the ‘hero’ is shown on his rounds with a ghost in tow – the ghost of a patient whose death he has caused by medical error.

*
The standard way of measuring income here is by household. Since different households have different numbers of people in them, the figures are adjusted to take account of this. So it is assumed that a single person living alone needs 67 per cent of the income of a couple to achieve the same standard of living. Similarly, people with children need more than a couple. By this means, all households can be compared on the same scale, approximately.

*
We concede that at the sub-atomic level there may well be irreducible, unavoidable and inevitable chance, which means that ‘determined probabilities’, to use Stephen Hawking’s expression, could be said to exist. But this does not seem particularly relevant to our judgement about who will win the Grand National.

*
The Bank of England uses a similar metaphor when communicating uncertainty about its forecasts with fan charts, in which the line describing predicted economic growth or inflation blurs out into a huge fan of graded possibilities, expressing what might be expected ‘if economic circumstances identical to today’s were to prevail on 100 occasions’. You can ask a computer to play out these futures multiple times, a technique known as a ‘Monte Carlo simulation’, which started with the US project to build a hydrogen bomb. Similarly, there are ‘ensembles’ for weather forecasting in which a number of different predictions are based on slightly perturbed assumptions about what is happening now, and chaos means that these small differences may result in wildly different predictions after a few days. Unfortunately there is still a reluctance to talk in public about the chances of different weather patterns, although in the US ‘possible paths’ of hurricanes are routinely shown on public news broadcasts.