These days, US media have been focused on the impact that gun ownership has in homicides, due to the tragic massive-murder that occurred last month in Newtown. I previously covered the overall inconsistency of data that correlate gun ownership with increased murder rates. That analysis was done with the limitation of comparing different countries with each-other, and left a number of question unsolved. On top: what leads people to commit violent crimes?

Challenged by a friend, I decided to look at more numbers to know the real impact of murders to society, the proportion of death caused by guns and if it’s true that wealthier countries experience less violent crimes.

I hope I haven’t cherry-picked data, if so, as usual, leave a comment and let’s debate on that.

First: What do we die of? 

57.000.000 (million) people dies every year in the world, WHO says. That’s the entire Italian population. Half of all these deaths have only 10 causes:

10-top causes of death
10-top causes of death *WHO

People die primarily of heart diseases and vascular-bran diseases (like strokes). By themselves, such cases call for 24% of all deaths. That’s quite a number: if you are on the end of your PhD studies, look for a post-doc in these fields!

Watch out: looking at the world as a whole tells us little on the real situation. Indeed, the picture is very different form rich to poor populations. In high-income countries, more then 3/4 of the deaths interest the old population (>70 years old). Which make perfectly sense. However, in the low-income countries 40% of death affects children (0-15 years old). Another 40% hits on “young population” (15-70 years old). Such estimation were done 4 years ago, reason to believe that the picture today haven’t changed dramatically. Though, things are getting better.

Now, how does the gun killing enter this frame?

First, let’s look at all homicides. The number of people murdered every year is about 0.77 million. That’s account for less then 0.02% of death in the world. In prospective, if your whole body was a graveyard, only your right hand will be occupied by murdered corps. Sure this is not a meaningless amount, and I am more and more convinced that we should effectively prevent and punish such bad habit.

Splitting the world into regions, the rate of all homiocides gives the following picture:

Homicide rates 2008 divided by world regions - data for UNDOC
Homicide rates 2008 divided by world regions – data for UNDOC

Souther and middle Africa and Central America pop out with average rate of homicides above 50 (per 100.000 ppl). The country with the worse situation is Honduras with 91.8 murders rate (at 2012). The countries where everyone loves everyone (of those we have data from, at least) are Iceland, Monaco, Palau (“island country located in the western Pacific Ocean” Google says) and Slovenia (analysis for 2008).

A very important things to keep in mind: average doesn’t tell the whole story. Never. Funny enough, an average is a number that correspond to the reality for the minority of the population observed. If I say that the mean of the Swedish population height is 1.75m, this only tells you that 39.9% of all Swedes are this tall (yes STAT nerds, assuming numbers are Gaussian, ok?).

That’s it. Without knowing something about the distribution of the data, we cannot make any other assumption.

Normal distribution

How much gunpowder has been used to commit these crimes? In average, the proportion of murders committed with firearms is 30% in the world. Only in US, it rises to 66%. Interestingly enough, such proportion is fairly stable since 2003. So in the latest 10 years, 3 out of 5 times people have used guns to kill in US, ranking weapons as the first cause of homicides in the US. Good habits never change, ehm?

Proportion of homicides caused by fireamrs in US over the years
Proportion of homicides caused by fireamrs in US over the years

Finally, does poverty leads to homicides? Yes and No. A short study conducted by Blake Taylor in 2006 pulls numbers from 322 metropolitan areas in the States. Holding all the other variable constant (unemployment, education, racism/hate, minorities, pop. density) the increase of 1% in population poverty leads to an increase in violent crime of 2.6%, the author concludes. These founding are consistent with older studies.

However, the picture may still mislead. The other variables (unemployment, ecc…) also positively contributes to violent crime, the author says. Is then poverty the first leading force of homicides? May not be, but analysis is missing. “For a more accurate picture of the relationship between poverty and crime, variables such as police force, family structure, climate, divorce rate, religion, and educational attainment could be added” the author conclude.

If you are a number nerd, please help me getting the whole picture straight: I am still looking for the main leading cause of homicides (or violent crimes) in developing countries. If you get a reference mail that to me or leave it in a comment.

– – – –

PS: The numbers comes form: the UNODC “Intentional homicide, count and rate per 100,000 population (1995 – 2011)” data-sheet, updated every year. Unfortunately  not all country provides reliable count and rate of homicides every year. For some reasons, year 2008 was the reaches column in terms of numbers, so the following data comes only from 2008 registrations.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.