Saturday, 7 April 2012

What to make of the decline in Capital crime?

You can take it as a given with any release of fresh crime statistics that newspapers will always find the bad news figures and make them the headline figure.

A slight fall in crime overall, or sudden spike in robberies? Its a bit of a no brainer really. Bad news is apparently more interesting.

Such time honoured traditions perhaps have something to do with research which consistently shows that regardless of what crime stats actually suggest, about to thirds of people think its going up in the country as a whole.

But today, a welcome change in narrative The Times has dug out some figures showing London has its lowest murder rate since the 1960s.

However, a couple of things don't feel quite right about the story.

Firstly, is it even possible to compare murder rates over that long. as Chris Grayling knows only too well, long term trends in crime can only really be established using the British Crime Survey measurement, ie victims fill in surveys listing the crimes against them.

Of course murder victims are at an obvious disadvantage should their input be sought for such surveys, police recorded crime figures (that is, crimes the police actually know about) are all we have to go on.

The potential problem with such figures is the change in definitions and measurements over time so comparing over as many decades as above may not be a fair comparison. (again as Chris Grayling should know by now).

This may not be quite the problem it was for the now Employment Minister, since there can only so many different ways you measure one person killing another. Although  as this piece explains, there have been a few changes in the way things are done.

Not to sniff at a newspaper actually reporting a positive story on crime, but given the way the figures were 'seen by the Times' rather than an official release,  I might wait a bit until I leave my stab vest at home.

Which brings me on to the second potential point of iffiness with the numbers: where did they come from?

The Times simply says they have seen them, for the financial year 2012. The Met's crime stats page only has homicide rates going up to February 2012, so it could well be that they have leaked out ahead of full publication.

Not only is this potentially a breach of the Code of Practice for Officially Statistics (depending on in what respect the Met is covered)  but also dubious given that we are just a few weeks away from a Mayoral election.

Could the selective release of ahead of schedule of certain crime stats by of any benefit to any candidate in particular?  Or should we just welcome a the news that murders are, quite possible, at a historically pretty low level?

No comments: