After writing this post about how homicides affects the poor most, with only one figure, I added another one showing absolute levels for the incidence of homicides across all social strata at the request of some readers. I had made a mistake, so I reposted the corrected figure. But sine not everyone caught that figure and because the fuiure is so important I thought I would show it again.
This is the absolute rate of homicides derived from the INE report per 100,000 people in each of the social strata, where 1 is the top and 5 is the bottom:
This data has been derived from the INE report, which is a poll, not precise crime statistics. INE derives the data from the poll and reaches the conclusion that in Venezuela there were 21,132 homicides in 2009, which somehow gets adjusted down to 19,113 homicides.
In order to calculate what is the rate of homicides per 100,000 people, we need to know how many people there are in Venezuela. Since I can’t find that number in the INE report, I look elsewhere in the INE website and find that for 2009 the projection was 28.8 million inhabitants, which gives 66 homicides per 100,000 inhabitants in 2009.
The chart above was obtained from a Table on page 70, where INE calculates the number of homicides per social strata based on its poll. However, the number of people is based on the sample size of the poll which is 1.8 million, not 28.8 million, so I had to correct for this factor to obtain the chart above.
The numbers are clearly horrific for the poor, the bottom strata of the population, as defined by INE, is the victim of 239 homicides per 100,000 people, almost four times larger than the national average. Those in the next lowest level have 110 homicides per 100,000 inhabitants, almost twice the national average. Curiously, the rich have more homicides per 100,000 inhabitants than the national average at 77. It seems you want to live in the middle class and lower middle class areas where in strata 2 and 3, the rates of homicides are a developed-country-like 9 and 19 homicides per 100,000 inhabitants respectively.
Even more surprising, according to the INE report (page 68) not all homicides get reported, despite the legal requirement that a person dying a violent death has to be autopsied. Clearly, people know the implications of their relatives having to go to the morgue and simply avoid it, likely by asking or bribing the medical doctor that certifies the death, not to specify that it was a violent one. For homicides, INE concludes, 15.% of homicides don’t get reported, which goes up to 36% for sexual abuse, 68% for robbery and 38% for kidnapping. Which shows the lack of trust and confidence Venezuelans have for the police.
Not a pretty picture…
The Devil's Excrement 
October 11, 2010 at 11:42 am
[…] it is getting closer, it had to happen, it is not only the increased crime but it’s profile, as I showed on August 29th. , it is either the poor or the the well to do that have the highest probability of being […]
August 30, 2010 at 8:04 pm
MO
The numbers in an absolute sense are appalling and the news is spreading. The week magazine had a section calling venezuela the murder capital of the world.
One question: if you do the same analysis by social segment would you not also see that the poor in the USA also have the highest murder rate? It does not take away from the gravity of the situation but I believe the same distribution would apply?
Keep up the great work – I despair reading the tepid normal news outlets who have self censored themselves. Chapeau to El nacional
August 30, 2010 at 11:13 am
Some off the cuff reactions follow.
1. That the INE didn’t get the information on homicides from the police, or justice department or attorney general is a rather sad state on statistical validity in Venezuela. [So what else is new?]
2. My initial reaction about using surveys to determine murders is that there is the danger of double – or triple or more- counting. If Juan Fulano says he knows of a murder, and Juan Smith says he also knows of a murder, they may be talking about the same person- unless the survey also took down some information about name, age, and address about homicide victims to cross compare. Did the survey also check with public health to see if that person was listed as dead, and with police to see if the person was listed as being murdered? With the proviso that the survey said that 15% of murders are not reported to the police, of course.
3. If the statistical sample was 1.8 million, that is an enormous amount. Most opinion polls survey only several thousand, and assume that the Law of Large Numbers will insure that the survey results are similar to the entire population. If care was taken to insure that there was not multiple counting, the very size of the survey would increase its validity.
4. I repeat my previous observation that if these murder rates are accurate, a VERY HIGH proportion of deaths in Strata 5/E are from homicide.
I lack the patience to go through several hundred pages of a PDF on a computer screen. I appreciate Miguel Octavio’s having perused the document for us. Perhaps I need to buy a Kindle, to make such perusing more user-friendly.
August 30, 2010 at 5:40 am
Thank for take your time to show the reality of the country, I expect that opposition take note of this, and transmit this to the people.