a few numbers behind the claims on demographic change

[podfish]

I'm sure everyone's been hearing for years about the threat of changing demographics to Republicans. Somehow it hasn't seemed to impact anything. But I just saw this, which was much more dramatic than I'd expected (even if it's not wholly true.. I'm not sure I trust this as rigorously statistically valid - it seem cherry-pickish, and HuffPo isn't exactly a-partisan, but still!)

On the Democratic side, Sen. Sanders’ success among the youth vote has been a consistent part of the primary narrative. But a similar analysis of youth in GOP primaries isn’t even possible. That’s because Republican turnout among voters ages 17-24 has been too small to be measured in 20 of 25 state exit polls. Only the Texas and Ohio GOP state primaries had enough participation by voters age 25-29 to measure in the exits.

Similarly, non-white voters have cast so few Republican ballots that they couldn’t be analyzed in 19 states, while non-white voters are represented in every Democratic primary exit poll. Geographically, fully half of this year’s GOP primary vote has come in counties Mitt Romney won by at least 10 points (https://fivethirtyeight.com/features/the-four-corners-of-the-gop-donald-trump-owns-three-of-them/) in 2012 (very unrepresentative regions of the national election he lost by 5 million votes).


https://www.huffingtonpost.com/greg-speed/16-gop-primaries-a-previe_b_9808372.html

[podfish]

and along similar lines:
https://www.usnews.com/opinion/articles/2016-05-05/donald-trumps-mojo-with-white-working-class-voters-is-greatly-exaggerated


Let's begin with exit polling. In 18 states (https://www.cnn.com/election/), so few Republican voters reported earning less than $30,000 per year that exit polls show "n/a" in that income column. There wasn't enough data. In six of those states no data exists on Republican voters earning less than $50,000 a year.

In other words, virtually everyone who voted for a Republican in Illinois (https://www.cnn.com/election/primaries/polls/il/Rep), Connecticut (https://www.cnn.com/election/primaries/polls/ct/Rep), Maryland (https://www.cnn.com/election/primaries/polls/md/Rep), Massachusetts (https://www.cnn.com/election/primaries/polls/ma/Rep), Vermont (https://www.cnn.com/election/primaries/polls/vt/Rep) and Virginia (https://www.cnn.com/election/primaries/polls/va/Rep) reported earning more than $50,000 a year. These are Trump voters. Even if they never went to college, they earn more than the average wage, which was $44,569.20 in 2014, according to the Social Security Administration (https://www.ssa.gov/cgi-bin/netcomp.cgi?year=2014).