More about crime stats: the importance of the methodology
I just received an email from Nicholas Marsh, researcher at the Norwegian Initiative on Small Arms Transfers. Mr. Marsh is reacting to the video I posted here earlier, titled “Choose Your Own Crime Stats” (see below).
Mr. Marsh authorized me (thanks!) to quote him at length, and that’s all I’ll do, hoping this would trigger more discussion on the topic.
a) Yes, there has been a steady decline in violent crime in the USA over the past two decades. This has been remarked upon at length in books, newspapers and magazines. So he is wrong to suggest that no one has talked about it. Though its my impression that the decline hasn’t featured prominently over the last few weeks. In general, I agree that as far as statistics are concerned, the recent debate has featured an absence of stats, or people cherry picking the stats that support their case and ignoring the ones that don’t.
b) Yes, more violent crime tends to occur in urban areas - I’m guessing that’s the same in Belgium.
c) There is though a big problem with his comparison of violent crime in the UK and USA. The UK headline numbers (available here http://www.homeoffice.gov.uk/publications/science-research-statistics/research-statistics/crime-research/hosb0812/hosb0812?view=Binary) are for ‘violence against the person’. If you look at the methodology (see here http://www.homeoffice.gov.uk/publications/science-research-statistics/research-statistics/crime-research/counting-rules/) they include predictable things like manslaughter, murder or assault, but they also include things such as: conspiracy to murder, causing death by dangerous driving, harassment, illicit abortion, illicit possession of firearms, and assault without injury. The ‘violence against the person’ figures do not include sexual offenses (eg rape), robbery or burglary which are presented separately in the report.
The FBI statistics (see here http://www.fbi.gov/about-us/cjis/ucr/crime-in-the-u.s/2011/crime-in-the-u.s.-2011/tables/table-1) cover ‘violent crime’ which is defined as: murder and nonnegligent manslaughter, forcible rape, robbery, and aggravated assault (see http://www.fbi.gov/about-us/cjis/ucr/crime-in-the-u.s/2011/crime-in-the-u.s.-2011/violent-crime/violent-crime)
So the UK and US violent crime statistics presented in the video are actually covering completely different things. They can’t meaningfully be compared.
The differences in definition are a general problem when comparing crime across countries. In addition, you have big differences in the level of reporting of crimes.
Homicide is one offence which tends to have similar definitions (though even then there are lots of differences, eg how manslaughter is treated) and a generally high level of reporting at least in developed countries. So its more meaningful to compare homicide than to other violent crimes.
Mr. Marsh adds, a bit later, in a second email:
It should be possible to compare violent crime in the USA and UK if someone could get data on specific offenses (eg armed robbery).