VA: SUSA shows a 4-point race...

[Brittain33]

...Obama leads McCain, 50%-46%.

http://www.surveyusa.com/client/PollReport.aspx?g=c3b77d5f-0d1a-4f6e-b195-ae05355e8eb8

All eyes on Rasmussen at 6.

You must be logged in to read this quote.

[Brittain33]

33% (R) - 37% (D) - 23% (I)

29% (C) - 38% (Moderate) - 15% (Liberal)

The most damning: 44% College Graduate. Although both categories here poll equally.

[Verily]

33% (R) - 37% (D) - 23% (I)

29% (C) - 38% (Moderate) - 15% (Liberal)

The most damning: 44% College Graduate. Although both categories here poll equally.

56% college grads is definitely an overestimate, but they're probably also overestimating McCain's performance among them. (The usual paradigm is that the best group for Republicans are those with some college or only a bachelor's degree, but those who have college degrees includes the very Democratic groups with Master's degrees and higher and generally favors the Democrats as a whole.)

All of this is complicated somewhat by race, though, so the paradigm is more easily applied in very white states and less so in states with large black or Hispanic populations, like Virginia.

[ChrisFromNJ]

Nice.

It appears that John McCain is bleeding an abnormal amount of Republican support here, while Barack's Democratic support is unified.

The party ID numbers are a bit questionable, but not nearly as bad as SurveyUSA's previous general election polls.

Here's hoping this is the return of the SurveyUSA that knocked the ball out of the park in the Democratic primary season.

[Alcon]

Party affiliation = no.

[Rowan]

I don't buy it. Party ID is off, and McCain WILL NOT lose 17% of R's. This is just a terrible poll.

[ChrisFromNJ]

Party affiliation = no.

If you adjust the party affliation to 35 D, 35 R, 25 I, you get a dead heat.

Not bad for Obama considering the fact that McCain has a national lead and Virginia is a historically Republican state.

[KeyKeeper]

This is the same polling company that showed McCain leading by 20% in North Carolina, trash it.

[© tweed]

the 2006 Senate race self-identified at 39R-36D-26I.

[Alcon]

It is worth noting one thing about SUSA, though:

You can't just look at 2006 exit polls and dismiss their net partisan affiliation.  Why?  Because SUSA undercounts independents, almost entirely independents with partisan leans -- they tend to be forced into partisan groups.  I think this is because their question is phrased in a way that asks about partisan registration, as opposed to self-identification.  In either case, they get fewer self-identified indies than other organizations.  In the current climate, that can cut two ways:

First, it can inflate Republicans, because it "assigns" Republicans who are reluctant to identify as Republicans.  The effect of this vis a vis other pollsters should decrease over time, as partisan identification becomes more galvanized, as it already has.

Alternatively, it can inflate Democrats, because among independents leaning to either party, Democrats (I believe, as of recently) outnumber Republicans.  Self-identified "pure independents" tend to lean McCain.  This results in a smaller, more Republican, group of independents.  I'm not convinced that this will, or will not, decrease as we get closer to election day.

Effects combined, my general guess is that SUSA assigns more independents as Democrats than Republicans, and McCain-supporting independents are less likely to be assigned as Republicans, than Obama-supporting independents as Democrats.  The net effect:  a higher Democratic partisan affiliation, and a lower Democratic support among independents.

Does that make D+3 reasonable?  Probably not, but it comes closer to explaining it.  I have qualms with SUSA outside of partisan identification, but it shows that SUSA results can't just be given a "dumb weight" to exit poll results.

Another note: Perennial caveat about margin of error on partisan sub-samples.  Here, it's +/-6.0% to 7.6%.

[Vsanto5]

This is the same polling company that showed McCain leading by 20% in North Carolina, trash it.

Of course that poll had been an outlier, but I wouldn't trash this one just yet.

[ChrisFromNJ]

It is worth noting one thing about SUSA, though:

You can't just look at 2006 exit polls and dismiss their net partisan affiliation.  Why?  Because SUSA undercounts independents, almost entirely independents with partisan leans -- they tend to be forced into partisan groups.  I think this is because their question is phrased in a way that asks about partisan registration, as opposed to self-identification.  In either case, they get fewer self-identified indies than other organizations.  In the current climate, that can cut two ways:

First, it can inflate Republicans, because it "assigns" Republicans who are reluctant to identify as Republicans.  The effect of this vis a vis other pollsters should decrease over time, as partisan identification becomes more galvanized, as it already has.

Alternatively, it can inflate Democrats, because among independents leaning to either party, Democrats (I believe, as of recently) outnumber Republicans.  Self-identified "pure independents" tend to lean McCain.  This results in a smaller, more Republican, group of independents.  I'm not convinced that this will, or will not, decrease as we get closer to election day.

Effects combined, my general guess is that SUSA assigns more independents as Democrats than Republicans, and McCain-supporting independents are less likely to be assigned as Republicans, than Obama-supporting independents as Democrats.  The net effect:  a higher Democratic partisan affiliation, and a lower Democratic support among independents.

Does that make D+3 reasonable?  Probably not, but it comes closer to explaining it.  I have qualms with SUSA outside of partisan identification, but it shows that SUSA results can't just be given a "dumb weight" to exit poll results.

Another note: Perennial caveat about margin of error on partisan sub-samples.  Here, it's +/-6.0% to 7.6%.

You must be logged in to read this quote.

Why the conflicting messages?

[Alcon]

They're not inconsistent.  The 2006 Virginia partisan affiliation was R+3.  I doubt 2008 party affiliation will be as Dem-favorable as 2006.  Even adjusting for SUSA's under-independenting, a six-point swing seems excessive.  Is it ridiculous-dismiss-the-poll stuff?  No, but it probably means they got a sample on the Democratic side of things.  Then again, you can tell that from the result they got compared to other pollsters, really.

[KeyKeeper]

This is the same polling company that showed McCain leading by 20% in North Carolina, trash it.

Of course that poll had been an outlier, but I wouldn't trash this one just yet.

I guess you are right.

[Brittain33]

Note that the Warner-Gilmore numbers are exactly where they've been in SUSA and other polls for months, include relative to last week's poll where McPalin was leading Obama narrowly.

However, if one was expecting Gilmore to pick up support in line with general Congressional tightening... a not unreasonable expectation, if less likely in this race than in many others.

[Ronnie]

Even though I doubt this poll is legitimate, 538 will weight it pretty high.

Ugh.

[Lief 🗽]

Virginia will be won by GOTV operations and registration efforts.

[Ronnie]

Virginia will be won by GOTV operations and registration efforts.

I think McCain will win by 3-4 points, simply because of the Demographics of the state.  It isn't trending that much to the Democrats, and I don't think that higher black turnout will attribute to more than an 8 point swing.  Plus, I believe that Obama will lose in the western Appalachian region of the state by a greater margin than Kerry.

VA is likely to be more Democratic by the 2012 election than now.

[KeyKeeper]

Even though I doubt this poll is legitimate, 538 will weight it pretty high.

Ugh.

They will give it the same weight as the Rass. Poll coming out.

[Ronnie]

Even though I doubt this poll is legitimate, 538 will weight it pretty high.

Ugh.

They will give it the same weight as the Rass. Poll coming out.

Yeah, but depending on what Rasmussen says, it will likely still stabilize more in Obama's favor than without the SUSA poll.

[mypalfish]

If true, McCain is a goner in November.

[??????????]

If true, McCain is a goner in November.

You must be logged in to read this quote.

[Franzl]

If true, McCain is a goner in November.

You must be logged in to read this quote.

You didn't contradict his statement with that. Does this mean you agree that McCain is sure to lose if this is the result in Virginia?

[Sbane]

Those southeast VA numbers look a little weird. I can understand that Obama has a huge lead in Richmond but what about the Virginia beach area? Are those military people ditching Mccain?

[Democratic Hawk]

Republican 33% (35%): McCain 80% (85%); Obama 17% (11%)

Democrat 37% (36%): McCain 12% (10%); Obama 87% (89%)

Independent 23% (20%): McCain 48% (55%); Obama 44% (34%)

Conservative 29% (35%): McCain 80% (81%); Obama 19% (15%)

Moderate 38% (39%): McCain 38% (37%); Obama 58% (58%)

Liberal 15% (12%): McCain 23% (11%); Obama 75% (88%)

(denotes SUSA, Sep. 5-9, 2008)