>Yes, I know that those things are nonsense, but respectable outfits
>like the NY Times seem to refer to the "respectable," Gallup-type
>polls as often as they can, apparently attempting to back up their
>impressions on what the public is thinking with "solid data."
They are pretty solid within their limitations (e.g., they reproduce conventional categories of thought and classification, ask only questions that elites think important, disallow ambivalence or ambiguity, are sensitive to wording, etc.). All the technical problems we've been talking about seem not to have reduced their predictive power concerning elections.
Doug