Comments on: Gun fear

By: john b

john b — Wed, 06 Jul 2005 12:20:00 +0000

(also note that the 33,000 is for one dodgy bit of Newcastle – the actual figure now appears to be closer to 10,000, which implies each person reports a crime every 10 years or so, which sounds about right)

By: Chris Lightfoot

Chris Lightfoot — Wed, 06 Jul 2005 12:07:00 +0000

What, from 258/100,000 to 33,000/100,000 over the course of a hundred years? I assume that you’ve verified that this isn’t explained by changes in numbers of police, social attitudes to crime, the creation of new criminal offences, increasing urbanisation of the population, and any number of other enormous changes that have taken place in this country since the beginning of the last century….

By: Squander Two

Squander Two — Wed, 06 Jul 2005 11:43:00 +0000

If you’re talking about a few percentage points up or down from one year to the next, then recorded crime stats are not meaningful. When you’re talking about an increase from 258 per 100,000 to 33,000 per 100,000, putting it all down to changes in recording methodologies just doesn’t cut it.

By: Chris Lightfoot

Chris Lightfoot — Wed, 06 Jul 2005 11:12:00 +0000

Yeah, that's a possibility. But Shipman was quite an extreme case; the interesting question is how many similar cases there are. I suppose the way to do this is to look at inquest results in the case of today's homicide victims and see how many of the findings would have differed if attitudes, medical knowledge or technology had been at their earlier levels. Not sure if that can be done in any reasonably sensible way, though.

"Or, as above, a cherrypicked example like the total number of recorded crimes" -- can we please kill this once and for all. The number of crimes recorded by police conveys very little information about the total number of crimes committed, and changes in the number of crimes reported even less.

By: dave heasman

dave heasman — Wed, 06 Jul 2005 08:45:00 +0000

Chris Lightfoot asks :-
"Does anyone have any insight into how much of it is the result of more comprehensive reporting, i.e. whether there were in the past deaths which were not reported as homicide which were they to occur now would be so reported? "

I don't have any insight, but my instinct suggests that in the 40s & 50s Shipman wouldn't have been nicked, and those (215?) deaths wouldn't have been "murders".
John Bodkin Adams was far more conspicuous and he got off.

By: john b

john b — Wed, 06 Jul 2005 08:05:00 +0000

No, just that your chances of being murdered were sod-all 40 years ago and are still sod-all, whereas the recent fall (and previous rise) in the rate of burglary directly impacts your and my everyday lives.

By: Squander Two

Squander Two — Wed, 06 Jul 2005 08:03:00 +0000

> No, they haven't

I know; that's why I said that they can, not that they have. And I'm sure they can, if they twist the right figures the right way. Lies, damn lies, and all that, is all I meant to imply there.

> cherrypicked examples of very rare crimes like homicide and armed robbery

Or, as above, a cherrypicked example like the total number of recorded crimes.

Anyway, what are you saying here? That, because homicide is rare, changes in its rate of occurrence are meaningless? It may be rare, but it's also (a) a hell of a lot less rare than it used to be and (b) exceedingly important. That the death penalty was rare was not an argument against its abolition.

By: john b

john b — Wed, 06 Jul 2005 05:40:00 +0000

No, they haven’t – that’s part of the problem. They’ve just asserted that ID cards will stop terrorism, without even *trying* to back that up with evidence or analysis.

Overall, it looks like since 1980 crime overall has risen then fallen back to its original rate (peaking in 1995, then falling over the last 10 years). The only examples to the contrary are cherrypicked examples of very rare crimes like homicide and armed robbery. The latter is a particularly dodgy example to use, since rates apparently fell by 1/3 between 1991-94 (here. I’m aware it’s not a great source, but recorded crime stats don’t break out armed robbery so this is the best I can find…)

(actually, another thing I need to do in my Sharpener post on homicide is come up with a convincing breakdown between homicides where the victim was a member of a drug gang, and ones where s/he wasn’t – the former are a substantial proportion of the total, and a rise in this figure is rather less worrying for a member of the general public than a rise in the latter figure…)

By: Squander Two

Squander Two — Wed, 06 Jul 2005 05:04:00 +0000

Home Office stats? Gosh. You mean the same organisation that has promised, year after year, to deliver low crime rates has produced reports showing low crime rates? Astonishing. One can't help but wonder why, when they have so much success, they keep promising to reverse the increasing crime rate. Surely that promise was fulfilled years ago.

Oh, look:

At the beginning of this century (1900-04) the total number of
crimes recorded by the police in the whole of England and Wales ran
at an annual average of just over 84,000. The rate was 258 per 100,000
population. At the beginning of the 1990s the number of crimes
recorded by the police in a twelve-month period in one district of
one city, the West End of Newcastle upon Tyne, was 13,500. The rate
was one in three of the residents — 33,000 per 100,000.

....

If we take the figure for armed robbery, an offence the growth of
which in the statistics could not be significantly accounted for by
changes in reporting (more telephones, for example) and recording
(changes in the law, changes in police procedures), we see that it was
such a small problem that no figures were generally published until
twenty years ago. In 1970 there were 480 armed robberies. By 1990
there were 3,900, and this rose in the following year to 5,300. This
was an eleven-fold increase on 1970, and the increase in the single
year was three times the total in 1970. If we consider the total
number, and not the rate of all cases of robbery, armed or not, the
rise in England and Wales in the twelve months from 1990 to 1991
was two-and-a-half times all cases of robbery recorded in the entire
period between the two world wars.

You are aware that the Home Office can also produce statistics which prove that ID cards will stop terrorism, right?

By: Rob

Rob — Tue, 05 Jul 2005 14:04:00 +0000

Oh, and according to this PDF – http://www.homeoffice.gov.uk/rds/pdfs2/hosb1203.pdf – two years ago, our homicide rate was about a quarter of that in the US, as was London’s in relation to New York’s. A little persistence should find you more detailed stats if you want them.