HomeLatest ThreadsGreatest ThreadsForums & GroupsMy SubscriptionsMy Posts
DU Home » Latest Threads » Forums & Groups » Retired » Retired Forums » 2016 Postmortem (Forum) » My Critique of Nate Silve... » Reply #9

Response to Time for change (Original post)

Wed May 18, 2016, 12:55 PM

9. I appreciate you taking the time to write it out, but you still misunderstand the purpose of exits

They are not and have not ever been intended to be a harbinger of who will win an election, though they have been used as such in media and by pollsters who have padded the bottom line by touting predictive questions. They are a tool for post-election demographic analysis, which is why they are performed in waves and true-upped to official results after results are realized.

NY is a good example of how this works--the first wave of exits, the ones that purported to show a close race, were overrepresented of Buffalo, NY--Bernie's best metro area in the state (and where I live). The reason for that is practical, Erie Country was one of the very few counties outside of NYC that opened polls in the morning instead of at noon. So it got hit with a lot of exit poll takers, again, for the first, early wave.

There is no exit poll conspiracy, none.

Reply to this post

Back to OP Alert abuse Link to post in-thread

Always highlight: 10 newest replies | Replies posted after I mark a forum
Replies to this discussion thread
Arrow 19 replies Author Time Post
Time for change May 2016 OP
JackRiddler May 2016 #1
Fearful May 2016 #2
reformist2 May 2016 #14
Time for change May 2016 #15
YouDig May 2016 #3
Time for change May 2016 #4
YouDig May 2016 #5
Time for change May 2016 #7
reddread May 2016 #13
2cannan May 2016 #6
Time for change May 2016 #8
zipplewrath May 2016 #17
Time for change May 2016 #19
LineReply I appreciate you taking the time to write it out, but you still misunderstand the purpose of exits
Godhumor May 2016 #9
Time for change May 2016 #10
Godhumor May 2016 #11
Time for change May 2016 #12
Godhumor May 2016 #16
Time for change May 2016 #18
Please login to view edit histories.