Comparing International Crime Statistics

Comparing International Crime Statistics

Phil Lee, 1/4/2009

There are significant differences in crime categorization, counting rules and official statistical methodologies used by governments. Using reported data without attempting to compensate for these differences will not provide meaningful comparisons country to country, although trends within a country might be determined if their government does not change the counting rules over a period of comparison. One good discussion of crime counting differences is found here:
http://en.wikipedia.org/wiki/Crime_rate
New Zealand publishes an overview of international comparisons of crime here:
http://www.courts.govt.nz/pubs/reports/2002/intl-comparisons-crime/index.html
Different crime reporting methods are described here:
http://www.courts.govt.nz/pubs/reports/2002/intl-comparisons-crime/section-2.html
and more discussion of counting rules and the issues arising from different rules may be found here:
http://www.courts.govt.nz/pubs/reports/2002/intl-comparisons-crime/section-7.html

A comparison, accounting for the differences between the US and UK crime counting and reporting methods, was published by the US Bureau of Justice Statistics and may be viewed at:
http://bjs.ojp.usdoj.gov/content/pub/pdf/cjusew96.pdf

(Crime and Justice in the United States and in England and Wales, 1981-96)

Except for murder, the UK violent crime was far higher than US violent crime.

The US Bureau of Justice Statistics has a web page of links for International Justice Statistics sources for various countries: http://www.ojp.usdoj.gov/bjs/ijs.htm

For the UK, "‘homicide’ covers the offences of murder, manslaughter and infanticide." (see: http://www.homeoffice.gov.uk/rds/pdfs08/hosb0308.pdf on p. 12 or http://www.crimestatistics.org.uk/output/Page40.asp)

Homicide does not include “Causing death by dangerous driving” which
"includes causing death by careless driving whilst under the influence
of drink or drugs and causing death by aggravated vehicle taking." (see
http://www.crimestatistics.org.uk/output/page41.asp) So, the UK does not include a significant number of careless driving homicides that are included in US homicide numbers.

In the area of murder, analyses typically use FBI homicide reports, but the FBI reports initial police determinations for homicides which tend to undercount civilian defensive shootings (excusable homicides) and over-count felony homicides by about 12%. In UK reporting, homicide classifications and counts are continually revised over time with initial police reports being revised as courts or prosecutors make evaluations. Just this difference alone could alter a typical 6 to 1 ratio of US murders to UK murders to a 5.3 to one ratio for comparisons.

The Brits reclassify homicides from results of investigations. On p. 12 of the “hosb0308” document we find the comment "Where the police initially record an offence as homicide it remains classified unless the police or courts decide later that no offence or homicide took place. Of the 757 offences first recorded in 2006/07, 23 were no longer recorded as homicides by 12 November 2007." About 15% of homicides ultimately become re-defined each year under the UK system, see p18: "Table 1.01 Offences¹ initially recorded by the police as homicide by current classification²: England and Wales, 1954 to 2006/07". This would mean that a defensive killing that a court in England held was justified would no longer be counted in England, but the same would not be true in the US because the FBI homicide classification policy does not allow revisions.

In the US, the FBI classifies homicide according to first police determinations and they are not reclassified by investigations. The FBI’s UCR Handbook clearly establishes a criteria for justified homicide that doesn't match the usual notions of self-defense -- in fact, it obviously promotes a policy of police determining whether a homicide is justified notwithstanding findings by a "coroner, prosecutor, grand jury, or court" (see NOTE on p. 17-18).

No one publishes crime data using demographically similar populations (accounting for differences in racial populations, age distributions and migrant populations – especially considering illegals) to make a comparison between the US and the UK or Canada. However, it is possible to examine crime in selected US states having a large percentage of the population with ethnic origins from Europe. As of 2007, selected FBI murder rates per 100,000 reported are: New Hampshire (1.1), Iowa (1.2), Montana (1.5), Maine (1.6), and Rhode Island (1.79). These states all have lower murder rates (per 100,000) than the Canadian national average (1.8). And these states’ rates are not corrected like those of the UK. These also states have similar murder rates to the UK (1.49 in 2005) but much lower levels of other violent crimes and relatively little gun control when compared to the UK or Canada. So, if the question is whether the UK's or Canada’s gun control creates greater public safety, the evidence is not to be found in murder rates.

Updated by Phil Lee on 9/30/10. Contact maryland_alert at yahoo dot com (sorry for being obscure, but web mail address scavenge programs make this practice necessary).