Interesting find in the CBS/NYT internals

[jezebel]

"In 2000, whom did you vote for:

GORE 32 BUSH 42 NADER 2 "
------------------------------------------------------

the other 24% polled didn't vote in 2000.
If I am reading that right, assuming Bush held all who voted for him, and all the Gore voters are voting Kerry, wouldn't that mean that of the 24% who didn't vote last time, but were polled this time, 14% went to Kerry and only 7% went to Bush, since the final tally was Bush 49, Kerry 46?
-------------------------------------------------------------

Or am I just reading to much into it? If not it seems it was a more heavily weighted Bush supporter poll.

[endnote]

1. More of the same...

[endnote]

2. I'm starting to suspect that progressives tend NOT to answer

the phone or something. How else could one explain the consistent sampling bias ??...

[pinkpops]

3. which would mean we can expect 2 out of 3 new voters?

[jezebel]

4. that sure would be nice if it is true NT

[BeyondGeography]

5. Bush's 2000 numbers are always higher

because people want to identify with the "winner" when they are asked that question. This phenomenon also explains the "loser" Gore's numbers, which are always understated in the "who did you vote for last time" query.

[kerry2win]

6. 4% MORE PUBS

someone had a breakdown earlier tonight