People are stupid
True, but they’re not nearly as dumb as landline only, robocall polls in a state with large young and minority populations. It looks like all of the Emerson polls have leaned pretty heavily toward Republicans and have been out of line with other available polling.
I don’t get it. Didn’t Josh just yesterday have a long piece on how Emerson’s polling is landline only, and therefore is less likely to survey younger people? I’m not in a position to evaluate Josh’s analysis, but it seems reasonable to suspect that Emerson isn’t entirely accurate with its polling this year (538 rates them a B, not the worst, but not the best either).
If that’s the case, though, why the breathless OMG take in the article? Everyone has NC going blue this year.
Relax. The Emerson polls always skew right. Nothing to see here.
Landline only and his lead is just two? Hillary must be ahead by quite a bit.
Emerson’s finding disagree with all the other polls, therefore its findings are suspect. Wait for more data to come in.
Emerson ergo pfft.
Norman Ornstein
‏@NormOrnstein
What many poll-watchers know: Quinnipiac polls are about publicity, not accuracy:
https://www.washingtonpost.com/local/education/how-did-marist-monmouth-suffolk-and-quinnipiac-get-known-for-political-polling/2016/08/02/15429962-532b-11e6-bbf5-957ad17b4385_story.htmlOkay, Ralph, reading through some of the responses here, I agree with your point from the other thread a lot more than I did an hour ago. LOL
I was fooled!
Nate Silver has Emerson rated as B, not sure if it is good for polls.
The fact that she’s not winning by a blowout pisses me off. Get it together people!
Josh reported that to offset the likelihood of older/whiter bias in a landline only poll, Emerson has stated that it automatically adjusts any poll by adding 3% to the D column. That seems pretty darned unscientific unless they have conducted some control group experiments comparing landline and landline plus cell results. But I am with Josh and most reporters generally, it doesn’t pay to unskew the polls.
Be forewarned, but as Abramowitz says: “absurd”
Absurd Reuters/Ipsos poll shows Clinton lead shrinking from 12 pts. on Aug. 22 to less than 1 pt. on Aug. 27, in five days!
12:56 PM - 30 Aug 2016
https://mobile.twitter.com/AlanIAbramowitz/status/770711752761892864It’s probably because lots of people think Hillary is unhealthy, and so better to vote for Trump since he is healthy. We need a healthy president, because of ISIS. That’s the most important thing.
Yeah but what happens when you add in his support from African Americans?
No one is unskewing polls. It’s perfectly o.k. to be skeptical of polling results and question whether or not the statistics are robust.(1)
(1)Robust statistics are statistics with good performance for data drawn from a wide range of probability distributions, especially for distributions that are not normal.
For statistics, a test is claimed as robust if it still provides insight to a problem despite having its assumptions altered or violated.
The running EV polling in the Prime Hive…
~OGD~
Exactly and Sam’s 341 EV confirm pretty well to Larry Sabato’s 348 predicted EV for Clinton. I.e., robust.