On Sat, 22 May 2004, Doug Henwood wrote:
> >Did the AAPOR list discussion make clear which of these tacks the
> >"nonvoters are the same" proponents take?
>
> Most of the arguments were based not on projections from demographics,
> but actual surveys of nonvoters.
If that's all they've got, I'm afraid they've got nothing. If their results contradict all their basic theories and assumptions, then we have to assume there's something wrong with their methods -- unless they can explain what's wrong with the theories and assumptions. The burden is on them to explain the discrepency.
Tossing out their their entire framework on the basis of divergent results would be mindless empiricism. But *not* tossing out their entire framework, and yet still insisting the results that don't fit it are valid, is even worse -- it's incoherent.
Michael