2016 Postmortem
Related: About this forumUPDATED: CNN is a live poll; Gravis and KBUR/Monmouth polling uses the controversial robo-call
Last edited Thu Jan 21, 2016, 06:30 PM - Edit history (1)
polling methodology that has a huge in-house pro-Trump and pro-Clinton effect (I have not seen a convincing explanation for this, but the effect is well documented).
This explains why you see CNN polling that shows Sanders leads in Iowa and New Hampshire with contemporaneous polls from Gravis and Monmouth that show Clinton ahead in Iowa and a tighter race in New Hampshire.
If you do nothing other than exclude robo-call polls from the Pollster aggregator, Sanders is ahead in Iowa and Sanders is comfortably up by double-digits in New Hampshire:
It does Clinton no favors to set her expectations in Iowa based on robo-call polls because, historically, falling short of expectations is almost worse than losing in Iowa.
JonLeibowitz
(6,282 posts)Why robocalls are controversial and unreliable?
Thanks, I actually don't know the answer to this question
Attorney in Texas
(3,373 posts)JonLeibowitz
(6,282 posts)Attorney in Texas
(3,373 posts)Attorney in Texas
(3,373 posts)different turn-out model to reach a different conclusion than what the pollster reached (for example, if some poll had Pres. Obama winning the race, and that poll had a 52% women and 48% men, but Romney thought the actual turnout would be 45% women and 55% men, he would re-calculate the polling results to deflate the women's answers to 45% of the total responses and inflate the men's answers to reach 55% and then he would get a different result from what the pollster projected).
This is not re-analyzing or "unskewing" any polls. It is just
(1) noting the indisputable trend that Clinton and Trump do better -- for whatever reason -- in robo-call polls and
(2) noting that the robo-call polls and the live cell/land line phone polls cannot both be right because they consistently project different results where the contemporaneous polls achieve results that are outside of the margin of error for the other method poll, and
(3) noting how the poll aggregation looks quite different if you exclude the robo-call polls from the mix.
Fawke Em
(11,366 posts)It's against the law, so pollsters who use robo-calls to poll get older voters with landlines.
And older voters tend to like Clinton.
I don't know enough about Trump's demos to say if that's true of him, but, given that he's on TV 24/7 and older people watch more mainstream media than younger voters, his support may also lean to the older voter.
Attorney in Texas
(3,373 posts)so that does not seem to be the entire explanation.
winter is coming
(11,785 posts)names. If you look at the actual questionnaires for live polls, there's usually an instruction to alternate the order of names. The scripts I've seen for robo-call polls have a "Press '1' for so-and-so" format, with no indication that names are being rotated.
msongs
(67,405 posts)Attorney in Texas
(3,373 posts)methodologies being mixed in with polls using proven methodologies.