Perhaps it's the "Medicine":

A report from the Times of London:

A UNITED Nations report has labelled Scotland the most violent country in the developed world, with people three times more likely to be assaulted than in America.
England and Wales recorded the second highest number of violent assaults while Northern Ireland recorded the fewest.
Got that? Scotland and England are both far more dangerous than Northern Ireland.

Well, it is a UN report.
It found that people living in Scotland were almost three times more likely to be victims of violent assault than people living in the States and suggests that more than 2,000 Scots are attacked every week, almost 10 times the official police figures.
I've done some work with American crime statistics, and so I know that the manipulation of these things by police departments is quite usual. I don't know how things work in the UK, but in the US the central crime statistics are compiled by the FBI in what they call the Uniform Crime Reports. UCRs are based on stats compiled by local police, and transmitted to the FBI.

There are two serious openings for manipulation in the UCR methodology. The first is the fact that the FBI only tracks certain named crimes. Because local police are themselves compiling the stats, all they need to do is reclassify a "forcible rape" (a UCR tracked crime) as a "sexual assault" (not tracked by the UCR) and the rape disappears off the crime statistics entirely and forever. As far as the statistics are concerned, it never happened, and your city had one fewer rape last year.

Alternatively, if you are lobbying for increased funding, you can start reclassifying things as UCR crimes. This brings us to the second great flaw: the FBI doesn't have a standard for how the police count. The police may report to the FBi the number of crimes that were reported; or the far smaller number "cleared by arrest"; or the far, far smaller number prosecuted; or the very much smaller number for which a conviction was actually obtained. One police department will choose to report on reported crimes, and another only on crimes cleared by arrest (reasoning that they don't know that the other crimes really happened, since they never caught anyone who seemed to be guilty of it).

Thus, a police captain who wants to light a fire under people can cause his city's crime statistics to "soar" just by changing to counting-by-report, and having a policy of classifying reported crimes whenever possible as a report of a UCR crime. A sheriff who wants to show "progress" in his tenure can do the opposite, causing the rates to "fall" again. A clever politician in the police department can play with these statistics both early and late, charging his predecessor with "unethical underreporting" to explain why the rates soar shortly after he enters office, and then making changes over the course of his term to bring the "rates" down.

Do Scottish police do the same thing? I don't know: maybe the UK has a rock-solid methodology. It would be a bit surprising to me, though, to learn that was the case.

No comments: