Matthew Yglesias @mattyglesias
@GarettJones As I said, it's a perfectly reasonable assumption. But I do think the strength of the correlation matters for what inferences we can make using WORDSUM data and it would be nice for someone to take a closer look at what the right answer is. https://t.co/OHrKu1P8jj — PolitiTweet.org
Matthew Yglesias @mattyglesias
tl;dr while the assumption that performance on a brief vocabulary test has some correlation to performance on longe… https://t.co/iNTaRJGl81