One wonders why these polls are all over the place. The ABCNews and Gallup Polls showed little change in voter sentiment, and the Gallup poll in particular showed Bush more favored than Kerry on certain key issues (e.g., likeability). The NewsWeek poll shows a major movement, and Kerry favored on many of the same things that Gallup said Bush was favored for.
Why might that be? Well, here's their sample:
SAMPLE SIZE/MARGIN OF ERROR FOR REGISTERED VOTERS SUBGROUPS:Newsweek has been oversampling Republicans for a month, since the convention -- you've all heard the complaints. This time, they ran a poll that oversampled Democrats, and (especially in the registered voters section) women, who are far more likely to be Kerry supporters than men (women are evenly split, that is, whereas Bush is heavily favored by men).
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)
SAMPLE SIZE/MARGIN OF ERROR FOR DEBATE VIEWERS SUBGROUPS:
770 Debate viewers (those who say they watched at least some of the
debate) (plus or minus 4.1)
369 Men (plus or minus 6)
401 Women (plus or minus 6)
265 Republicans (plus or minus 7)
274 Democrats (plus or minus 7)
215 Independents (plus or minus 8)
Just chance that the samples shifted right after the debate, which Kerry's allies have been touting as his big chance for movement? Well, we all know that the media isn't biased towards Kerry -- and certainly not NewsWeek.