Great post. What evidence do you have that the non-voting population is "much more liberal" than the voting population? I was struck by how much turnout increased in 2020 from 2016 and yet the national popular vote went from about +2 Dem to +4.5 Dem.
One thing people get wrong in looking at 2016 -> 2020 and saying “turnout doesn’t help democrats” is (a) controlling for everything else, higher residual increases in turnout were correlated with Democratic gains in vote share last year, (b) Biden made gains v Clinton overall, and (c) we’re still only observing a world where 65% of people are voting. The other 35% are much more Democratic, especially in blue states.
I didn’t know about the turnout being correlated Dem gains. I’ve just heard some say that turnout increases often inspire the other side to turnout as well. That MRP story you shared was quite interesting and intuitively makes sense. I have always wondered about Mississippi!
Polls of the public's attitudes and values are useful and important. However, we have a group of 50 people who have captured the Senate and don't give a flip what the public or their own constituents want or think.
Appendix A: I am also reminded that Pew got the exact percentage correct ahead of an Irish referendum on expanding abortion rights. 66% of Irish adults told Pew they favored expanded rights, and then 66% of Irish voters ended up voting that way: https://www.pewresearch.org/fact-tank/2018/05/29/ireland-abortion-vote-reflects-western-europe-support/
Great post. What evidence do you have that the non-voting population is "much more liberal" than the voting population? I was struck by how much turnout increased in 2020 from 2016 and yet the national popular vote went from about +2 Dem to +4.5 Dem.
Hey Aaron, I explain a lot of the math in this linked piece, but basically the types of people who don’t vote are a lot more Democratic-looking than the types that do. See here: https://medium.economist.com/would-donald-trump-be-president-if-all-americans-actually-voted-95c4f960798
One thing people get wrong in looking at 2016 -> 2020 and saying “turnout doesn’t help democrats” is (a) controlling for everything else, higher residual increases in turnout were correlated with Democratic gains in vote share last year, (b) Biden made gains v Clinton overall, and (c) we’re still only observing a world where 65% of people are voting. The other 35% are much more Democratic, especially in blue states.
I didn’t know about the turnout being correlated Dem gains. I’ve just heard some say that turnout increases often inspire the other side to turnout as well. That MRP story you shared was quite interesting and intuitively makes sense. I have always wondered about Mississippi!
Polls of the public's attitudes and values are useful and important. However, we have a group of 50 people who have captured the Senate and don't give a flip what the public or their own constituents want or think.
All the more reason why polls are valuable, for advocates and the public alike!