The Details
The data on suicides, homicides, and race for each state and Washington DC came from the Centers for Disease Control (CDC) Web-based Injury and Statistics Query and Reporting System (WISQARS) (link) for the years 2008 – 2010.
Data on gun ownership was collected by the CDC's Behavioral Risk Factor Surveillance System survey from 2004 (link). The Behavioral Risk Factor Surveillance System (BRFSS) is on ongoing telephone survey of US households that collects data about chronic health conditions, health-related risk behaviors, and utilization of preventative healthcare services. Data is collected from all 50 states, Washington DC, and three US territories. Every year, over 400,000 US adults are surveyed, either by phone or in-person.
Linear regression was used to examine the data for correlations and trends.
The data table, showing rates of homicides and suicides among blacks and whites for all 50 states plus Washington D.C., can be viewed here.
For all 50 states plus Washington DC, the average percentage of gun owners was 36.4%; the average rate of homicide using a firearm of any type was 3.73 homicides per 100,000 population (37.3 / 1,000,000 population); and the average rate of suicide using a firearm of any type was 6.06 suicides per 100,000 population (60.6 / 1,000,000).
For all 50 states plus Washington DC, the average rate of homicides using a gun of any type for whites was 15.88 homicides per 1,000,000 population, while the average rate of homicides using a gun of any type for blacks was 155.51 homicides per 1,000,000 population. On average in the USA, the rate of gun homicides among blacks is nearly ten times higher than the rate of gun homicides among whites.
For all 50 states plus Washington DC, the average rate of suicide using a gun of any type for whites was 83.68 suicides per 1,000,000 population; the average rate of suicide using a gun of any type for blacks was 31.0 suicides per 1,000,000 population. On average in the USA, the rate of suicide using a gun is 2.7 time greater among whites than among blacks.
Using linear regression, the number of whites and blacks dying by gun homicide and gun suicide in each state was regressed on the gun ownership measurement (percentage in each state who own a gun). The percentage of gun ownership was highly positively correlated with the number of gun suicides by both blacks and whites. The percentage of gun ownership was highly positively correlated with the number of gun homicides for whites, but the correlation of gun ownership with number of gun homicides among blacks did not achieve statistical significance (correlation of gun ownership and whites dying by gun homicide r = 0.503 (p < 0.001); correlation of gun ownership and blacks dying by gun homicide r = -0.109 (p = 0.261); correlation of gun ownership and whites dying by gun suicide r = 0.706 (p < 0.001); and correlation of gun ownership and blacks dying by gun suicide r = 0.567 (p = 0.001)).
The relationships between gun ownership and death from gun homicide and gun suicide are illustrated in the scatterplots presented in the supplemental materials (found here). Each state is plotted in the graphs by both the percentage of gun owners and number of blacks or whites dying by either gun homicide or gun suicide. The least-squares regression line was also plotted to show the overall relationship between gun ownership and dying by gun homicide and gun suicide. (View scatterplots)
The results of the linear regression of gun homicides and gun suicides on the number of gun owners were statistically significant in the case of white suicides, black suicides, and white homicides, but were not significant for black homicides (Homicide among whites regressed on gun ownership: B = 0.347 (t = 3.73, p < 0.001). Homicide among blacks regressed on gun ownership: B = -0.521 (t = -0.647, p = 0.522). Suicide among whites regressed on gun ownership: B = 1.83 (t = 6.915, p < 0.001). Suicide among blacks regressed on gun ownership: B = 0.488 (t = 3.51, p = 0.002)). The regression allows us to predict how suicides and homicides numbers will change with changes in the number of gun owners. For the country as a whole, were the percentage of gun owners to increase from a national average of 36.24% to 37.24%, we would expect the rate of suicides among whites to increase from a national average rate of 83.69/million to a rate of 125.44/million, the rate of suicides among blacks to increase from a national average rate of 31.0/million to a rate of 40.98/million, and the rate of homicides among whites to increase from a national average rate of 15.88/million to a rate of 19.90/million. Similarly, were the percentage of gun owners to decrease from a national average of 36.24% to 35.24%, we would expect the national average rate of gun suicides among whites to decrease to 41.92/million, the national average rate of gun suicides among blacks to decrease to 21.02/million, and the national average rate of homicides among whites to decrease to 11.86/million. Because the correlation of homicides among blacks with gun ownership did not achieve statistical significance, we would not expect the rate of homicide among blacks to increase or decrease with changes in the number of gun owners.
This study not only provides further evidence that more guns are associated with greater number of gunshot injuries, but also gives us some idea of how increases and decreases in gun ownership might reflect on the number of gun homicides and gun suicides.
There is evidence from other studies that the population density of a community may effect gun violence. Gun homicides have been shown to be more frequent in densely populated urban communities, while gun suicides have been shown to be more frequent in less populated rural communities (here, and here). The study data presented here does not include information on the type of neighborhood where the shooting occurred, so we can not independently assess how rural vs. urban communities differ in gun violence.
|