Bush up 47-40 in latest Fox News Opinion Dynamics Poll

[Smash255]

This poll completley contradicts Gallup, completley Contradicts two ARG polls and a Quinnipiac poll, has Bush up 43-42 in blue states.  (I thiink their are some real serious issues with this poll.  Is every other pollster on the national and state level wrong??  I highly dount that, and the fact Bush is doing slighlty better in blue states 43-42 than battlegrounds 43-43 is the real kicker.

[Smash255]

This is 5 different polls.  The main one is where the blue states come from.  AN error in there woudl not effect the 4 state polls, unless there was a systemic problem.

The polls do also support a number of recent polls too, Survey USA in FL, the Harris and Tipp nationals, and carry the trend for MI, PA and OH.  

It is likely the method is overpolling Bush a bit, but not by a whole lot.

Harris poll was over a week ago.  Since then ALL the other polls have showed it close the ABC Poll had its issues the other direction.

7 of the 8 previous PA polls showed Kerry ahead exception Rasmussen by 1 9 of the last 10 MI polls including 5 in a row have showed Kerry ahead in MI.  Ohio polls have gone back & forth the entire way.  ALL FL polls have been close, Survey USA poll that showed Bush up 7 was shown to be flawed with an overampling of Reps  (the poll had something like Reps identity +7 in its sample, when the state is actually Dems +2)

[Smash255]

This is 5 different polls.  The main one is where the blue states come from.  AN error in there woudl not effect the 4 state polls, unless there was a systemic problem.

The polls do also support a number of recent polls too, Survey USA in FL, the Harris and Tipp nationals, and carry the trend for MI, PA and OH.  

It is likely the method is overpolling Bush a bit, but not by a whole lot.

Harris poll was over a week ago.  Since then ALL the other polls have showed it close the ABC Poll had its issues the other direction.

7 of the 8 previous PA polls showed Kerry ahead exception Rasmussen by 1 9 of the last 10 MI polls including 5 in a row have showed Kerry ahead in MI.  Ohio polls have gone back & forth the entire way.  ALL FL polls have been close, Survey USA poll that showed Bush up 7 was shown to be flawed with an overampling of Reps  (the poll had something like Reps identity +7 in its sample, when the state is actually Dems +2)

So you are telling me to ignore a poll that is a week old in the national and pay attention to several month old polls on the state level?  See the contradiction?

ALso, since Harris has been a TIPP/IBD poll that had Bush up.

The Survey USA poll used self identification, they pointed this out and said it may be due to the Reagan funeral that they were skewed toward Bush.

http://www.realclearpolitics.com/bush_vs_kerry.html

Look there and tell me where the national trend lines are.

I was going back further for the state polls because you said you didn't trust the newer Ohio polls that showed Kerry ahead.  The bottom line is as far as the Ohio pokks its been back & forth.

The +2 for Dems was also in the 2000 exit polls for FLA, the Reagan so called boost could impact the results to a point, but its not going to have a 9 point swing in identity.

TIPP had Bush up 3 with Nader, tie without Nader.  Gallup has Bush +1 both with & without Nader, Democorocy Corps has Kerry +1 without Nader and even with Nader.  So the TIPP poll, Gallup & Democrocy Corps all show it really close, meanwhile ABC & Fox don't

[Smash255]

A few things.  First, the Republicans have had a better registration effort in FL the past 4 years.

Second, this is self identification.  You can have 100% ID themselves as republican and not be wrong, if that is what they all say.   During the Reagan funeral a huge swing to the Republicans seems about right.  They did the right thing by pointing out that this had occured.

Finally, on all the polls being all over the place, ain't it fun?  

Well it tends to show inaccuracy.  The Republicans have made gains in vopter drives thats true, but Florida has also seen an increase in women's population and Non-Cuban Hispanics.  So the overall what was added was preety equal.

Anyway point is when the Voter ID in the polls is so far off from the overall id of the state that shows the poll has problems.  The LA Times poll which was inaccurate to the left at a Voter ID of +14 to the dems (+10 over the Nat'l average of +4 dems) this poll had voter ID of +7 to the Republicans which is +9 in favor of the Reps over the state ID of +2 Dems.  So both polls are off about the same.  I'm not saying the Survey USA poll was biased, what I'm saying is its accuracy is a problem because the sample was a much heavier representation of Republicans than the state actually has