Gun ownership does not predict firearm homicides. [Or it does in US?]

Amendment Jan 2013: This post looks at numbers available from the UN Office for Drug and Crime on homicide rates per country and rate of gun ownership published in a Guardian article by the blogger . My conclusion from data analysis was that there is no correlation between owning guns and rate of homicides. However, a research article form 2013 indicates the opposite when it comes to domestic US crime data. “Those persons with guns in the home were at greater risk than those without guns in the home of dying from a homicide in the home (adjusted odds ratio = 1.9, 95% confidence interval: 1.1, 3.4)” the authors say. Thank you to Carlos Talavera-López for sharing this paper with me. 
So having guns around increase your risk of dying because of one. The number I show below may not point to the same result simply because my analysis was not accurate or because “big numbers” often fails to show the right thing. Comparing countries with countries may not being as effective as comparing regions of the same State.
– – –
The tragic massive murder in a school of Connecticut has triggered (once more) the conversation in the United State about gun restriction. CNN anchorman Pier Morgan became understandably emotional when he faced a pro-gun officer, questioning “how many kids will have to die before you will says: ‘we want less guns, not more’?”.

In this ‘cloud’ of shouting and finger-points, where mental disorders, media or education have been consequently considered responsible for what happened, The Guardian blogger  had the good idea to let numbers speak. He posted public-available data from UNODC, the United Nation Office for Drug and Crime, that reports on firearm homicides in countries from all over the world.

OBS: These data are not holy gold, considering that developing or in-war countries may have a flaw in data registration (a user comments that Syria is awkwardly down in the list of guns and crimes). Nevertheless, we might draw some interesting observation.

“US has the highest gun ownership rate in the world – an average of 88 per 100 people” the post says. “Thought US counts only 5% of the world population, it houses 30-35% of the world’s total civilian-owned guns“.

I’m not particularly impressed by these conclusions. Factors like welfare, unemployment and gun-driven crimes would make these numbers much more interesting.

The brilliant point comes now: US does not have the worst firearm murder rate. It’s the 28th in world, with 2.97 death of homicide due to guns every 100,000 people [per year]. This means that if you hit the street in US, the absolute risk to be killed by a gun shot is about 0.000008% every day. It’s like rolling a six-faces dice 10 times every morning, after brushing your teeth: if you manage to get all 1s, that is you unlucky day.

This is math playing. Your chance to die from a gun shot differs whether you live in big towns, of you have a low income.
There is more. Simon Rogers encouraged people to download the data and do somehting with it. I did my part.

chart#1

I thought it was a good idea to see if there is a good correlation between civilians that owned guns and the gun-caused homicides. So easy that even Excel can make a linear regression. Apparently, owning a gun do not predict well whether a country will have a high-rate of firearm homicides (fig. above)

chart#2

Very similar story with proportion of homicides caused by guns and number of guns owned by the population. Other factor may play a mush more important role, such as the one I mention above.

These were all the interesting things I could see with the data shared on the post. It doesn’t sound rewarding for who wants to restrict guns in US. Can I do something more meaningful?

I browsed the UNODC webpage and found data on the total number of homicides in US, released by the National Police. In 2011, there have been 14.748 homicides. Crossing numbers, you see that 9.146 were committed with firearms. That’s it. If you ever gonna be killed, you are 60% more likely to be put down with a lethal shot, then with other weapons. If you suddenly remove guns away from owners, 62% of the murders will be (potentially) avoided.

For a weapon-free world.

PS: Please, massively question and review my stat conclusions, for the sake of truth! 😉

Posted in Politics and tagged , , , . Bookmark the permalink. RSS feed for this post. Leave a trackback.

One Response to Gun ownership does not predict firearm homicides. [Or it does in US?]

  1. Pingback: What do we die of? | SciencePlug

Leave a Reply

Swedish Greys - a WordPress theme from Nordic Themepark.

SciencePlug is Stephen Fry proof thanks to caching by WP Super Cache

%d bloggers like this: