Book Review – Dividing Lines: The Politics of Immigration Control in America

Dividing Lines would be a dry read if immigration control weren’t a hot issue now.  And hey, maybe it’s still dry to you, but I found it interesting.  It’s a blow-by-blow history of immigration policy in the United States since the founding, with special attention to the political coalitions that supported several very different policy regimes.
Continue reading

Machine learning and racial bias

Warning: Second-order contrarianism.
You may remember this Propublica article from about a year ago, arguing that the COMPAS scores, a machine-learning algorithm that predicts risk of criminal recidivism, is racially biased.  Their methodology was a bit strange and they made their data available openly, so I had been intending to reanalyze using more straightforward methods.  Fortunately, several people have already done this, sparing me the effort; several of them found that according to commonly accepted standards, the COMPAS algorithm is not racially biased.  The Washington Post also published an article, saying the question is complicated. Continue reading

Obama/Trump voters in the National Election Studies data, Part 4


I conducted three analyses and found that:

  1. There were a surprisingly large number of Obama/Trump voters, and with some exceptions they matched the media portrait – white, working class, low income, older, and Midwestern.
  2. These voters probably chose Trump based on racially loaded policy issues; in particular, “law and order” issues related to fear of criminal violence.
  3.  These voters held economically populist views on a number of issues, even if those views did not drive their votes in the 2016 election.  However, their economic views on most issues were at least slightly to the right of the views of Obama/Clinton voters.

Continue reading

Obama/Trump voters in the American National Election Studies data, Part 3

Many Democrats have wondered whether economic populism might help win the support of white working class Trump voters.  We saw in Part 2 that economic populism didn’t seem to play much of a role for Obama/Trump voters in the 2016 election; however, it’s still possible that in a future election, focused on different issues, that approach might work.  My third analysis explores that question.  Note that as per previous analyses, I restricted this sample to white voters.

Continue reading

Obama/Trump voters in the American National Election Studies data, Part 2

Multiple regression analysis is a technique that measures correlations between variables, but adjusts things so that each variable’s influence is shown “all other things being equal.” Let’s imagine that Trump voters tend to be against both abortion and immigration, and most voters who are against abortion are also against immigration. But the few people who are fine with abortion and against immigration still tended to vote for Trump, and those who are against abortion and fine with immigration tended to vote for Clinton. In that case, multiple regression analysis will show a strong association between views on immigration and Trump voting, but only a weak relationship between views on abortion and Trump voting; one might reasonably guess in this case that immigration mattered more to voters than abortion did.

Continue reading

Obama/Trump voters in the American National Election Studies data, Part 1

We’ve all had several months to make up our minds about whether Trump voters are racist or not, and one might think that there’s nothing more to be said on the topic.  But the American National Election Studies (ANES) group recently released its 2016 data file – the gold standard of election surveys in the United States, with data on thousands of subjects who answered hundreds of questions about political topics, including voting behavior.

Continue reading