General Discussion
Related: Editorials & Other Articles, Issue Forums, Alliance Forums, Region ForumsCNN's Jake Tapper to Clinton why did you loose the Latino vote in NV
This discussion thread was locked as off-topic by etherealtruth (a host of the General Discussion forum).
He's talking about there stupid exit polls showing Sanders winning.the most share of Latino votes over Clinton.I've read she won them by twenty points.Which is it WTF!!!
book_worm
(15,951 posts)while losing both the white and latino vote. And somehow she won in the most populated Latino areas, too, according to real votes cast. This link is helpful:
http://www.vox.com/2016/2/20/11079660/latinos-nevada-sanders
bigdarryl
(13,190 posts)MgtPA
(1,022 posts)Not all Latinos live in Latino precincts, I guess.
book_worm
(15,951 posts)by 10 or 11 points and still wins by 6? Obviously the entrance poll was off.
Predictive statistics based on unproven assumptions that say I'm right must always be valued over descriptive statistics that say I'm wrong.
I predicted last year that I'd have $20k in the bank by now for a new car. I had my assumptions and a baseline. Therefore, when the bank says I have far less than that in the bank the bank must be mistaken and stole my money. (Even if there's no mechanism I can show that I had the money and the bank actually took it. Either the bank's lying, or they stole it.)
And, yes, if I applied that level of innumeracy to literacy skills, eyed bea a rightn reel goodnow, lighk juan've demdair proepheshunl riduhs.
daleo
(21,317 posts)Though "properly done" is the operative phrase here. There is always the possibility for selection bias, on the part of both pollsters and respondents.
gyroscope
(1,443 posts)riversedge
(70,181 posts)bigdarryl
(13,190 posts)riversedge
(70,181 posts)FarPoint
(12,309 posts)It's all good.