It appears Donald Trump’s people have discovered Richman et al (2014), or at least the Monkey Cage post about it, to dress up his “rigged” routine. I wrote about brief post about this last year, citing some good skeptical commentary by Ahlquist and Gelbach. It may not surprise you to learn that that skepticism appears to have been warranted. (I can’t tell if that link is going to show up as gated or not, because I’m at work and have institutional access, but the gist of it is this: the Richman finding is premised on an implausibly low level of measurement error for the data they’re working with, which is probably what produces their result. Once more plausible assumptions about measurement error are applied, it appears more likely to suggest that “the rate of non-citizen voting in the United States is likely 0.”) I love the Monkey Cage, and I’m glad it exists, but this is of course a danger of that particular model of overlapping journalism and scholarship–counterintuitive and politically juicy findings get the most attention and no one notices when they’re later debunked.