Wednesday, July 21, 2010

Loads and Loads of Analysis

Anyone who loves digging into drug war data should check out this blog from Diego Valle-Jones, and this post in particular. It's a really top-notch resource. The most important conclusion I've come to so far is that the typical UN and Mexican National System of Public Security murder rates are significant underestimates, and that we should all favor the Inegi numbers when looking for an official source.


Richard said...

Of course, the data begs the question of how one comes up with a precise number of "unreported" crimes. The analysis says, "In Mexico about 85% percent of crimes go unreported, so it's difficult to obtain exact estimates other than general trends."

Is that 85% an even inexact estimate, or pulled out of thin air?

The murder rate chart looks much more alarming than it is, the y axis only running from 9 to 18 per 100,000... a huge number, but not out of line with "normal" Latin American homicide rates. And, it assumes "present trends continue" which is always a big "IF"

pc said...

Someone actually asked him about the 85 percent figure in comments, and he had referenced a survey from 2008 or something like that. Of course, such a survey would seem to still be ripe for misreporting, so the figure is far from airtight.

Don Gringo said...

Beautiful statistics. So Mexico is finally getting back to normal!

pc said...

Yeah it's a veritable flood of information.