Tuesday, April 20, 2010

Does IQ explain racial differences in violent crime?

Adolescents in the ADD Health Study were asked if they had intentionally injured someone in the past year. I estimated logistic regression models to see if: 1) non-Asian minorities (NAMs) were more likely to answer yes, and 2) are the racial/ethnic differences explained by lower IQs.

The coefficient for blacks is 0.34 (p = .001) and 0.24 for Hispanics (p = .013). The numbers indicate that members of both groups are more likely to have injured someone. When a measure of IQ is added to the equations (it's a vocab test), the results look like this: 

Logistic regression coefficients

Hispanic model (n = 6,150)
Hispanic 0.16
IQ -0.01*

Black model (n = 6,154)
Black 0.28*
IQ -0.01*

* p < .05, two-tailed test

When IQ is entered into the model the Hispanic effect disappears. In other words, lower IQs explain why Hispanics are more likely to be violent. On the other hand, entering IQ into the model for blacks reduces the relationship between being black and violent, but it does not eliminate it. According to ADD Health data, other factors are necessary to fully explain high levels of violence by blacks.

8 comments:

Jason Malloy said...

"When IQ is entered into the model the Hispanic effect disappears. In other words, lower IQs explain why Hispanics are more likely to be violent."


When these researchers excluded Puerto Ricans (i.e. blacks), Hispanics in the AddHealth dataset were found to be no more violent than whites. This would mean that Hispanics are much less violent than whites of the same IQ level.

Anonymous said...

Hey Malloy, when are you going to start blogging more?

Anonymous said...

Perhaps you are more likely to have injured someone if you socialise with low IQ violent people, and Blacks socialise with Blacks.

Or perhaps, you are more likely to have injured someone if you are physically stronger, and Blacks are physically stronger than others at a young age. Might this also explain Hispanics being less violent?

Anonymous said...

"Or perhaps, you are more likely to have injured someone if you are physically stronger, and Blacks are physically stronger than others at a young age."

Is that really true? I think I read that blacks are comparable to whites on average.

Anonymous said...

"Is that really true?"

I don’t know if it is true. Black babies mature faster than white babies. I don't know if they are effectively more mature by adolescence or even whether this would imply greater strength.

The more I think about it the less I like this hypothesis.

My guess would be that it does come down to (partly) biologically determined personality factors.
I recall reading of studies which show Blacks as having higher self-esteem as well as studies showing such is linked to higher rates of violence. Whether these latter studies controlled for race I don’t know.

Anonymous said...

The hispanic IQ is 89, modestly above that of blacks, and the hispanic crime rate is 3 times that of whites- yet far below that of blacks.

Hispanics seem to often be virtually absent in discussions of the relationship between IQ and crime when it comes to using racial comparisons.

Anonymous said...

Malloy, why do you claim puerto ricans are "black?" They have far higher degrees of black ancestry than groups like mexicans, but they're simply not "black."

Also, if you're to use SAT scores as proxies for IQ, then puerto ricans have one of the lowest average IQ's of any hispanic group.

Still doesn't change how hispanics throw a big monkey wrench into the idea of IQ and crime being huge correlates.

Anonymous said...

The PR's do indeed have low IQ's. They do poorly in school and commit a great deal of crime. Their taste in clothing, music and lifestyle can be classified as barbaric.