Quote:
Originally Posted by well named
You are still confusing two different parts of the methodology. There are three steps
1) Identify a population of illegal immigrant inmates:
2) Collect arrest data for a sample of this population:
Correct. It's probably as good an estimate of the crimes committed by the criminal illegal immigrant population as you can get.
Quote:
Note that I made a mistake earlier. I was on my phone and couldn't check. Their estimates are to arrests for the 249k that had IAFIS data, not the full 500k population as I said a bit earlier. But also note the difference here between the source of data for step (1) and step (2). The time ranges they used to get the population of inmates does not limit the time range of the arrest histories, as they make clear in the parts I've quoted.
Correct, but this is irrelevant, because we are looking forward to a representative sample of crimes committed bu illegal immigrants per 100000 population, and this is a good estimate.
Quote:
3) Estimate the "Number and Percent of Criminal Alien Arrest Offenses by Type of Offense" (Table 2)
Let me walk through this in (even more excruciating) detail. This is the claim I am disputing:
So far we've estimated the proportion of illegal immigrant crimes to total population of illegal immigrants. Now you can compare the crime rate within that population to the crime rate of the total population over a few years which is what they did here. There are some problems with the math as you've pointed out; it would be almost perfect if we had total illegal immigrant crimes for the time period of the total population it is compared to, but that is obviously not possible. Something to note is that the total crimes estimated here are probably less absolutely speaking due to unsolved crimes or unprosecuted ones, so there are multiple problems in both directions. However, when you look at the other statistics coupled with this:
-illegal immigrants make up a significant % of the prison population (way more proportionally than their native counterparts)
-50% of all federal crimes are committed in border counties of California, Arizona, New Mexico
-the vast majority of illegal immigrants in prison are from Mexico
then we can see that there is something there not to be simply discounted.
From a strictly deductive point of view, we would expect that crimes committed in legal immigrant populations to be fairly low from an absolute point of view. These people go through the painful process of immigrating legally. They want to be here and they respect the law.
On the other hand, we should expect crime rates in illegal immigrant populations to be relatively high. These people are not concerned with following immigration laws, so it's more likely they have either committed crimes that prevent them from immigrating, or are willing to commit other crimes as well.