Deltoid

Dean Payne said:

Centerwall made his comparisons with and without the major (pop. > 1M) metropolitan areas. With these areas, I get the same numbers you list. Without, I get 3.1 for Canadian provinces, and 3.7 for the US states.

I get the same numbers. Here are the homicide rates, inside and outside major metropolitan areas.

             homicide    handgun            % with
             rate        homicide rate      handgun
Canada       2.8         0.3                11
   <1M          3.1         0.2                6
   >1M          2.2         0.4               18
US           8.5         3.5                41
   <1M          3.7         1.1               30
   >1M         14.1         6.4               45

Comparing the regions outside the major metropolitan centres we see that the 0.6 higher US rate (3.7 vs 3.1) is associated with a 0.9 higher handgun homicide rate (1.1 vs 0.2). In the major metropolitan centres, the difference in homicide rates far exceeds the difference in handgun homicide rates, so there are obviously other differences here.

I believe the treatment of these cities is Centerwall's weakest point. His numbers show that each of the Canadian metro areas has a lower homicide rate than the surrounding province, but each US metro area has a higher rate than the surrounding state.

Table 14 of recent FBI Uniform Crime Reports, crime rates by population groups, shows this strong US pattern of homicide (and violent crime) vs. population density even more dramatically.

Australia seems to follow the Canadian pattern. In the period 68-81 the homicide rate in Sydney was 1.9, in the rest of the state it was 2.1. The firearm homicide rate was 0.6 (Sydney) vs 1.0 (rest). Gun ownership was 20% of households (Sydney), 40% (rest).

As I have noted above, it seems unlikely that handguns are the sole cause of this difference, but since handgun availability seems to be higher in urban areas (from the data above, in Canada and the US the % of homicides committed with handguns was higher in urban areas), it may well contribute.

When comparing the many individual states and provinces, Canada's far lower handgun ownership rate did not produce any consistent pattern of lower overall homicide rates.

True, but this just means that handgun ownership is not the sole determinate of homicide rates. It hardly proves Centerwall's conclusion that "Canadians fully compensate for the relative dearth of handguns in Canada by effectively using other means for killing each other". Consider: If this was true then the "homicide by other means" rate would be higher in Canada (since the handgun homicide rate is consistently lower). By the same (invalid) reasoning we can disprove his conclusion by noting that the "other means" homicide rates are not higher in Canada.

Yukon 16.9 Alaska 11.6

I ought to note that the high rate in the Yukon is not particularly meaningful, because the province's population is less than 25,000.

This is what I find most disturbing about Centerwall's paper. As you note, the population of Alaska is twenty times that of Yukon. Furthermore, gun availability may well be HIGHER in the Yukon. (Centerwall does not have any data on gun ownership in Alaska and Yukon.)

Centerwall must have been aware of the gross difference in population between Alaska and Yukon, but he did not report them, and used Alaska and Yukon to claim that his "result" generalized to regions with high homicide rates. This sort of thing does not fill me with confidence in an author.

What I find remarkable is that Sloan and Centerwall managed to come to opposite conclusions from very similar data. For both sets of data, the Canadian region had a lower homicide rate, and the difference was close to the difference in the handgun homicide rates. In both cases when the regions were broken into groups by race (Sloan) or state/province (Centerwall) the Canadian groups did not have a uniformly lower homicide rate.

Sloan concludes that Canadian style handgun controls may reduce US homicide rates, Centerwall concludes that they would not. Hmmm.