We’ve got polls!

[Heffalump_and_Roo]

@dalej42

I can’t bump this thread because of the DMR poll. The release has been canceled.

Des Moines Register, CNN cancel release of pre-caucus Iowa Poll

This election couldn’t get any weirder.

Just adding a quote as to why the release of the poll was cancelled.

Nothing is more important to the Register and its polling partners than the integrity of the Iowa Poll. Today, a respondent raised an issue with the way the survey was administered, which could have compromised the results of the poll. It appears a candidate’s name was omitted in at least one interview in which the respondent was asked to name their preferred candidate.

While this appears to be isolated to one surveyor, we cannot confirm that with certainty. Therefore, the partners made the difficult decision to not to move forward with releasing the Iowa Poll.

According to a NYT article that is paywalled so I can’t see it, the people from the Pete Buttigieg campaign complained when his name was left off the survey by at least one interviewer.

Here’s a quote that’s supposedly from the NYT article taken from a Yang sub thread:

CNN abruptly canceled a special Saturday night broadcast of the results of a new Des Moines Register poll of Iowa caucusgoers, according to a network official, after complaints from Pete Buttigieg’s campaign that his name was not included in at least one of their telephone calls for its poll this week.

[dalej42]

All sorts of rumors about why the poll was tossed out. My WAG and it’s just a WAG is deliberate sabotage by one of the questioners. Remember, it’s not Ann herself actually doing the polling. Now, that doesn’t mean that someone necessarily tried to fix the poll in favor of their preferred candidate, it can also mean someone that just didn’t give a damn and fudged their answers.

Ann Selzer is the gold name in polling and she isn’t going to risk her name over a flawed poll. Every poll is going to have a questioner flub something on a question and that wouldn’t be the reason to throw out an entire poll.

[dalej42]

CBS/YouGov poll for Iowa, basically the same news as always

Bernie and Biden tied at 25%
Pete at 21%
Warren 16%
Klobuchar 5%

Weather for Iowa is unusually warm today, 52 degrees F in Des Moines, typical winter tomorrow with a high of 34 F. So, no weather events to affect turnout or affect last minute canvassing.

CBS News Battleground Tracker: What could happen in Iowa on Monday?

[RTFirefly]

@dalej42

All sorts of rumors about why the poll was tossed out. My WAG and it’s just a WAG is deliberate sabotage by one of the questioners. Remember, it’s not Ann herself actually doing the polling. Now, that doesn’t mean that someone necessarily tried to fix the poll in favor of their preferred candidate, it can also mean someone that just didn’t give a damn and fudged their answers.

Ann Selzer is the gold name in polling and she isn’t going to risk her name over a flawed poll. Every poll is going to have a questioner flub something on a question and that wouldn’t be the reason to throw out an entire poll.

Yeah, I’m wondering about this too.

The possibility that they’d have just left his name off the electronic questionnaire seems improbable. You’d have multiple sets of eyeballs review the questions, and you’d probably test it out in the office just to make sure the software didn’t have any glitches.

Also hard to believe one interviewer would choose to leave his name off. First, they’ve got the unweighted results of the poll, and I’ll bet there’s an interviewer code in the paradata or metadata or whatever the hell it’s called. If one interviewer showed a big zero for one of the leading candidates, that would stick out like a sore thumb.

I don’t know how many interviewers they have, but if each interviewer yields only a handful of responses (which would be my expectation: that it takes a shitload of calls per completed interview), then unless the demographics of that interviewer’s responders are wildly unrepresentative, you just throw his/her interviews out, and take care of it in the weighting.

So I’ve got a certain amount of professional curiosity about this, that goes beyond any interest in what the poll’s results might have been. I’m hoping there’s a presentation on this at the Joint Statistical Meetings this summer; there’s certainly the material for one, no matter what went wrong.