Ah, good information. Yes, this does skew the results quite a bit. Just confused about Gallup’s involvement in this particular poll.
Gallup in its poll does NOT adjust for party identification, according to what was written about it last week. As such, this poll is probably a gallup poll result massaged by USA Today’s party base and turnout concepts.
So, it is like the guy hiring a CPA and asking a simple math question to determine whom would be his CFO. The proper answer to the question was “what do you want the answer to be”. So, while these polls have some value, they are as valid as the overlay.
As far as the gender difference I believe that large difference takes place in decreasing numbers as the age brackets go up. I have not seen the degree of interest in the younger voters this time as I did in 08 so my suspicion is that large a difference is probably over sampled.