Tuesday, December 18, 2007
[See Correction appended]Pollster.com's Mark Blumenthal notes a big reason why those gratuitous network Iowa caucus "entrance" polls might be wrong. As he puts it in an email, "There is zero check against younger interviewer bias"--meaning that when the networks employ young interviewers older caucusers tend to avoid talking to them.
Keep in mind that the 2004 [exit poll] debacle was partly the result of younger, presumably Dem interviewers having greater trouble approaching or interviewing older Republican voters. This despite the age / gender "adjustment" that [CBS pollster] Frankovic talks about.
Why this matters: As Frankovic notes, Obama's support is much higher among voters under 45. So never mind the deliberation, post threshold reallocation, etc. The entrance poll will likely show Obama doing better than he'll really do even among those entering.
The networks could leave the Iowa caucuses to their own perverse, undemocratic and historically misguided devices without making them more perverse and undemocratic. But then how would network polling divisions justify their existence? ...
See earlier post. ...
Correction: It turns out the networks will attempt to correct for the tendency of older caucusers to avoid young entrance poll interviews. But the fix requires the already-harried entrance polltakers to keep an accurate tally of the voters who don't talk to them (are they old, male, female etc.). It's not clear that this can be pulled off in the crush "as voters stream in for the 6:30 p.m. caucus start," Blumenthal notes. But he has posted a correction. ....11:48 P.M. link
What to expect when you're expecting: Drudge teases the National Enquirer ... Update: The Enquirer posts the gist ...Update: The full Enquirer story is now up. ... One initial point: There's no reason to conclude this story was planted by one campaign or another. I'm familiar with how the initial Rielle Hunter/Edwards rumors, true or not, got to at least one news outlet--and no campaigns, Dem or GOP, were involved. It was a story going around--I'd been hearing it for months. Not all rumors are plants. And some are true. Even in the Enquirer. .. P.S.: Here's an earlier analysis of the potential effect of this scandal on Edwards--and Hillary. It doesn't seem all that complicated. Until recently, Edwards not very subtly put his wife's illness. and his loyalty to her, near the center of his campaign. In the process, he said:
In so many ways, you're the guardians of what kind of human being, we're going to have as president. ... And you get to judge us.
and, on 60 Minutes: