
Nate Silver @NateSilver538
@Nate_Cohn @davidshor @algaraca @ercjhnkrbs @SpecialPuppy1 Our density calcs are at the census tract level so you can do it by e.g. zip code. It definitely adds a lot if say you're doing a big multivariate regression on CCES data. — PolitiTweet.org