Wednesday, January 27, 2010

The myth of the myth of Hispanic lawlessness

Ron Unz has an analysis of criminal justice data in the American Conservative which finds that Hispanics don't really differ from whites in their levels of crime.


One limitation of the analysis is that it relies on data gathered far down the line from the actual criminal behavior. Between the act and incarceration rates are many steps: reporting the crime, arresting, prosecuting, convicting, and sentencing the defendant as well as deciding when he gets paroled from prison.


The data closest to the act is self-report data. The National Longitudinal Study of Adolescent Health (Add Health) is a longitudinal study of a nationally representative sample of adolescents in grades 7-12.  In the study, 3,911 whites and 719 Hispanics answered if they had committed the crimes listed in the table below. Percentages are displayed as well as the ratio of the Hispanic to white rates.







Of the 15 delinquent acts listed, Hispanics have the higher rate 14 times. Ratios range from 0.95 to 2.33. 


Critics might respond that surveying adolescents captures minor crimes, while data on prisoners, for example, captures serious criminality. But you can see that respondents were asked about very serious crimes as well as about less serious ones. For example, Hispanics are 2.33 times more likely than whites to have stabbed or shot someone (and the difference is statistically significant).

8 comments:

  1. One of the implications of Unz's article was that Hispanic crime may not be worse than that of the more misbehaved whites - those of the south central, for example (Okies, Texas Plains, etc.). Any chance you could look at that data by region?

    ReplyDelete
  2. Something I've been covering for a couple of years now is how the gubmint cooks these particular #'s...

    http://www.nationofcowards.us/?p=4296

    ReplyDelete
  3. Anonymous5:54 AM

    As far as I can tell (I didn't read the article closely - I read it last night before I hit the sack), Unz's argument seems to be that, once you adjust for factors like age and sex, Hispanics have similar crime rates to whites.

    In other words, he's pointing out that crime is a young man's game, and Hispanics are younger and more likely to be male than the median white.

    That's almost certainly true; I just don't see the point of adjusting for age and sex in the first place. Crime rates that are high because of age/sex factors are still high crime rates regardless, even if cohort analysis eliminates the difference.

    ReplyDelete
  4. Anonymous11:25 AM

    They need to ask the adolescents how much they lie. The ones who are more likely to commit crimes are also more likely to lie. So even their higher rates underestimate the differences.

    ReplyDelete
  5. Anonymous4:19 PM

    "Hispanics are 2.33 times more likely than whites to have stabbed or shot someone (and the difference is statistically significant)."


    That pretty much says it all right there. If you are a Mexican-illegal, and you have commited a crime and think you may be a suspect, what do you do? You hightail it back to Mexico, where you will never be arrested for it. Its a wonder that we catch any of them at all.

    ReplyDelete
  6. Have you looked at whether there is a negative correlation between number of children fathered and time spent in prison for men?

    At the risk of sounding un-p.c. (eh, who cares?), it would be nice if the extremely high rates of incarceration among certain groups were having a eugenic effect.

    ReplyDelete
  7. This comment has been removed by the author.

    ReplyDelete
  8. Anonymous8:24 AM

    When I said "almost certainly true", I was referring only to the fact that Hispanics are younger and more likely to be male.

    That Hispanics have higher crime rates is obvious to anyone who lives there. It's remarkable that AmCon would run with this crap.

    ReplyDelete

Are gun owners mentally ill?

  Some anti-gun people think owning a gun is a sign of some kind of mental abnormality. According to General Social Survey data, gun owners ...