National Opinion Dynamics Poll - Tied with Nader Kerry+2 Head to Head

[The Vorlon]

http://www.foxnews.com/story/0,2933,122378,00.html

[classical liberal]

net fav change from last time is Bush gained 3 and Kerry gained 8.  undecideds broke 2-1 for Kerry in that time.

[millwx]

http://www.foxnews.com/story/0,2933,122378,00.html

This latest round of polling is interesting.  The only one showing Bush with a lead (TIPP) has absolutely atrocious internals for Bush (half unsupportive of war in Iraq, 40% saying we're losing the war on terror, over half saying they're dissatisfied with the direction of the country, a favorability rating of 42%, and a job approval rating of 43%).  I know the LA Times poll has serious problems, and the Gallup and Rasmussen polls may have some modest problems.  And, frankly, as you know, I'm not even as fond of Fox/OD as most people are.  BUT... in almost every category in every one of these polls (including the LA Times one!) Bush's internal's are better.  Yet he only leads in the TIPP poll?!?!?  Odd.

[Reds4]

It is very odd, I suspected Bush would be up 2 or so in the Opinion Dynamics poll, and yes I expected a better overall round of polling for Bush than what we have seen. Isn't it true that Vorlon and many others see the summer as a time where Democrats generally poll slightly better? maybe 3 points or so? Summer is here now so maybe that could be it, but other than that I dont know. I must say it seems somehow Kerry's favorability has increased some over the past few weeks as well. Possibly tv ads, who knows?

[Fmr. Gov. NickG]

Can someone please give me a brief explanation of why Democrats would poll better in the summer if the party ID weighting is consistant?

[© tweed]

Can someone please give me a brief explanation of why Democrats would poll better in the summer if the party ID weighting is consistant?

The GOP Base 'tunes out'.  Don;t know why though.  Maybe because they are big family people and go campign in Montana for the whole summer.

[The Vorlon]

Can someone please give me a brief explanation of why Democrats would poll better in the summer if the party ID weighting is consistant?

Not all polls do a hard weight by party ID:

Polls that DO weight by Party ID:

Zogby 37/35/28 Dem/Rep/Indy
Rasmussen 38/35/27 Dem/Rep/Indy
Battleground 37/36/27 Dem/Rep/Indy (Vorlon Estimate)
CBS - Varies -- Last poll was +7 to the Dem side
Survey USA (Dependant upon the State being polled)

Polls that DO NOT weight by party ID

Gallup
TIPP
ABC (With some qualifiers)
Fox (With some qualifiers)
ARG
Quinnipiac
PEW
Harris
Teeter/Hart (The do, and they don't - too complicated to get into actually)
Democracy Corps
Mason Dixon
LA Times - (Between 5-18% more democrats in evey poll since since 2001 however
Newsweek

The Weighted group will NOT show a "summer effect"
The unweighted group will, to varying degrees, show a summer effect.

[millwx]

Can someone please give me a brief explanation of why Democrats would poll better in the summer if the party ID weighting is consistant?

Good question.  I like to pretend I know a bit about polls  :-)  but... you're right, this makes no sense.  Folks continually argue that Reps tend to go on vacation more, so they get polled less during the summer.  Fine.  But if you force your sample back to about a 35/33/32 (D/R/I) sample, it should make a darned bit of different, even if your raw sample is heinous... like 45/25/30... it gets weighted back.

Frankly, I'd like to see the stats showing Dem bias... because I'm not arguing that it doesn't exist.  I just don't know why.  For example, the low income conservatives may be the non-traveling Republicans more likely to vote Dem.  So, the Rep sample in the summer might be too centrist of a sample.  But that's a shot in the dark.

[The Vorlon]

Can someone please give me a brief explanation of why Democrats would poll better in the summer if the party ID weighting is consistant?

Good question.  I like to pretend I know a bit about polls  :-)  but... you're right, this makes no sense.  Folks continually argue that Reps tend to go on vacation more, so they get polled less during the summer.  Fine.  But if you force your sample back to about a 35/33/32 (D/R/I) sample, it should make a darned bit of different, even if your raw sample is heinous... like 45/25/30... it gets weighted back.

Frankly, I'd like to see the stats showing Dem bias... because I'm not arguing that it doesn't exist.  I just don't know why.  For example, the low income conservatives may be the non-traveling Republicans more likely to vote Dem.  So, the Rep sample in the summer might be too centrist of a sample.  But that's a shot in the dark.

See the post DIRECTLY above your last post - the answer is there

Zogby - Rasmussen - Both hard weight - no change last few weeks (Rasmussen always tied, Zogby stable to mild move to Bush)

Others NOT a hard weight by ID, drift towards Kerry.  The degree of the "summer fade" varied from poll to poll, depending of the integrity of the polls basic methodology.

[millwx]

See the post DIRECTLY above your last post - the answer is there

Yup.  Saw it.  I was writing mine when you posted, so I didn't see yours first.  All thanks to The Vorlon!  :-)