John Lott Makes a Mistake. Again. News at 11.

Last week, John Lott released a working paper showing that illegal immigrants in Arizona “are at least 142% more likely to be convicted of a crime than other Arizonans.” I didn’t bother reading it or reporting on it because Lott is spectacularly dishonest and unreliable in his work and it’s not worth the time to pore through his dreck to find out how he tortured the data. Anyway, I figured that someone else would do the work eventually, and today Alex Nowrasteh of the Cato Institute comes through. I’m sorry that Alex had to waste a week of time he’ll never get back on this, but since he did let’s find out what Lott got wrong:

Lott wrote his paper based on a dataset he obtained from the Arizona Department of Corrections (ADC) that lists all admitted prisoners in the state of Arizona from 1985 to 2017….The variable that Lott focused on is “CITIZEN.” That variable is broken down into seven categories. Lott erroneously assumed that the third category, called “non-US citizen and deportable,” only counted illegal immigrants. That is not true….A significant proportion of non-U.S. citizens who are deported every year are legal immigrants who violate the terms of their visas in one way or the other.

I’m sure it was an honest mistake. Nowrasteh uses a more reliable variable to get a better estimate, but even so, it counts both legal and illegal immigrants. So he provides both a high and low estimate for the number of illegal immigrants in Arizona prisons:

In other words, illegal immigrants are a smaller percentage of the prison population than the overall population.

There are all sorts of potential sources of error here, including pretty much all of the estimates of everything. That’s just the way it goes with this stuff. In any case, I imagine that Lott will come roaring back, insisting that he knew all along about the limitations of the CITIZEN variable and then filling the web with statistical gobbledygook to show that he was right all along. Maybe. But given his track record, I’ll believe the Cato estimates until someone I trust comes along to tell me that Lott has a point.