[Terp]

This original story was not posted under this headline so you would never have found it. The discussion in the threads last night pointed out Newsweek's poll sampling pre and post debate is faulty.

I followed the other thread before FR went off-line for a upgrade. The first NewsWeak poll showing GW with the big lead was faulty because they sampled to many Republicans and this latest poll is faulty becasue they sampled to many Democrats.

Here's the scoup from another site.

SAMPLE SIZE/MARGIN OF ERROR FOR REGISTERED VOTERS SUBGROUPS: 1,013 Total Registered voters (plus or minus 4)

481 Men (plus or minus 5)

532 Women (plus or minus 5)

345 Republicans (plus or minus 6)

364 Democrats (plus or minus 6)

278 Independents (plus or minus 7)

As you can see they oversampled Dems and women.

[hobson]

I can't find it now but the pre and post polls had the EXACTLY the same sample for men and women. I don't believe the poll AT ALL.