3 ms·

And Nate Silver: - To follow: some _very_ quick analysis which suggests the claim here of rigged results in Wisconsin is probably BS: http://nym.ag/2gI6YLP htt

by joshwa 10y ago

And Nate Silver: - To follow: some _very_ quick analysis which suggests the claim here of rigged results in Wisconsin is probably BS: http://nym.ag/2gI6YLP http://nym.ag/2gI6YLP - Run a regression on Wisc. counties with >=50K people, and you find that Clinton improved more in counties with only paper ballots. HOWEVER: https://twitter.com/NateSilver538/status/801221546685661184/photo/1 https://twitter.com/NateSilver538/status/801221546685661184/... - ...the effect COMPLETELY DISAPPEARS once you control for race and education levels, the key factors in predicting vote shifts this year. https://twitter.com/NateSilver538/status/801221907609579520/photo/1 https://twitter.com/NateSilver538/status/801221907609579520/... https://twitter.com/NateSilver538/status/801220813890277376 https://twitter.com/NateSilver538/status/801220813890277376

Natsu 10y ago
Nate has a full article here that's likely better than a few tweets for discussing something like this: http://fivethirtyeight.com/features/demographics-not-hacking-explain-the-election-results/ http://fivethirtyeight.com/features/demographics-not-hacking...