What do you enjoy reading the most here on my blog?

Search My Blog

Sunday, March 02, 2008

Immigration and Crime - A Poor Study

If there's one thing I hate even more than a poorly-done analysis claiming to be research that assumes that correlation means causation - it's a poorly-done analysis claiming to be research that assumes that correlation means causation AND misgeneralized the interpretation in the first place AS WELL AS assumes that the analysis has something to do with the issue being researched, which it DOESN'T.

The poorly-done analysis/research in question today is a study that "shows California's foreign-born population -- including illegal immigrants -- makes up only a sliver of the state's population of inmates".

Fine, you found a slightly-negative correlation between California inmates and their foreign birthplaces. Unfortunately, that information is useless. Especially when you try and generalize and apply causation relating foreign-born people to illegal immigrants, relating incarceration to crimes committed, and then also neglect the effects of illegal immigration other than just the initial immigration (like family members) and neglect certain non-jailable crimes (like shoplifting and vandalism).

Are you starting to see why I'm pissed off at this study and the news article that makes the claim that a lower number of foreign-born inmates indicates that stereotypes about illegal immigrants and crime rates are now disproven?

First of all, let's start with the fact that this study can never prove anything about illegal immigrants because all the study inquired about was the nation of birth and who was foreign-born. They make the argument that illegal immigrants are INCLUDED in the foreign-born population. I would make the argument that could also be illegal immigrants included in the USA-born population, assuming that they emigrated first and applied for citizenship elsewhere, THEN illegally re-entered the USA. It's possible, you can't deny that. Would it make a difference? Not at all. I'm just saying that their logic sucks, and equally-sucky logic proves it.

Secondly, this study is about INMATES, not criminals or crime at all. These are people who supposedly have committed a crime and been convicted of it and sentenced to jail. This study cannot possibly fathom the number of crimes being committed that have NOT led to an arrest. This study also ignores the number of crimes being committed that do not have a PENALTY that would lead to an arrest. The biggest problem about trying to make ANY connection of this study and illegal immigration is the fact that an illegal immigrant, once arrested and convicted of a crime, does NOT go to jail! Illegal immigrant criminals who are convicted are supposed to get DEPORTED. Hence, the number of foreign-born jail inmates means VERY LITTLE in relation to illegal immigrants.

Thirdly, any claim about crime rates is bogus. The number of criminals in jail had little to do with crime rates in general. One would HOPE for a correlation between crime rate and criminals being busted, but there are so many other factors in play that any generalization wouldn't be strong enough to beat up a five-year-old girl. Not even a five-year-old girl with brittle-bone disease!

Lastly, illegal immigration is illegal. The crime rate / percentage of criminals in the illegal immigration population is 100%, you dumbasses! At least near the END of the article, there is one quote that actually admits to this glaring error in the statistics: "While acknowledging that staying in the country illegally is a crime, Butcher and co-author Anne Morrison Piehl focused on activities that both foreign-born and U.S.-born adults can commit."

So they basically decided to do a study about crime and foreign-born people, but completely ignore that all of the foreign-born people who entered the country illegally are already criminals.

Any way you slice it, this is a worthless study and an especially-worthless article about the study.


(I can't make up something this atrocious - here's the article)

No comments: