Obama poll numbers in Fresno, CA and Louisville, KY

[Poundingtherock]

Obama's favorable/unfavorable rating in Louisville is 42/41.  Sarah Palin's favorable/unfavorable rating in Louisville is 36/35.

Obama won 56% of the vote in the county where Louisville, Kentucky is located.

http://www.surveyusa.com/client/PollReport.aspx?g=fb3d4e10-81bc-485d-9be1-c5e60623d362

Obama's favorable/unfavorable rating in Fresno, California is 41/39.  Sarah Palin's favorable/unfavorable rating in Fresno, California is 35/28.

Obama won 50% of the vote in the county where Fresno, California is located.

http://www.surveyusa.com/client/PollReport.aspx?g=cbdc3ca0-7948-4c09-b3c3-0008af50d939

Surprised that Palin would have a higher or the same net favorable rating as Obama in two cities located in counties that he won last November?

[pbrower2a]

Obama's favorable/unfavorable rating in Louisville is 42/41.  Sarah Palin's favorable/unfavorable rating in Louisville is 36/35.

Obama won 56% of the vote in the county where Louisville, Kentucky is located.

http://www.surveyusa.com/client/PollReport.aspx?g=fb3d4e10-81bc-485d-9be1-c5e60623d362

Obama's favorable/unfavorable rating in Fresno, California is 41/39.  Sarah Palin's favorable/unfavorable rating in Fresno, California is 35/28.

Obama won 50% of the vote in the county where Fresno, California is located.

http://www.surveyusa.com/client/PollReport.aspx?g=cbdc3ca0-7948-4c09-b3c3-0008af50d939

Surprised that Palin would have a higher or the same net favorable rating as Obama in two cities located in counties that he won last November?

SurveyUSA has been fielding some strange, inexplicable poll results -- so strange that they suggest fabrication or distortiion. One set of polls suggests that Obama approval is now as low in Missouri and Virginia as it is in Kentucky, Kansas, and Alabama. That is beyond explanation.

[Poundingtherock]

That's quite a serious charge that you are making.

Here's SurveyUSA's track record from the last election:

http://www.publicpolicypolling.com/pdf/images/Wall%20Street%20Chart.jpg

SurveyUSA also pretty much nailed CA-9, VA-gov, and NJ-Gov (unlike Democracy Corps(D) and the NY Times/CBS news, which embarrassed themselves with their NJ polling):

http://www.surveyusa.com/index.php/2009/11/04/1641/

http://www.surveyusa.com/index.php/2009/11/04/no-pollster-more-accurate-than-surveyusa-in-nj-governor-contest/

http://www.surveyusa.com/index.php/2009/11/04/no-pollster-more-accurate-than-surveyusa-in-va-governor-contest/

[Sbane]

So Obama's approval rating is higher than his disapproval rating in a county he won by a hair's breadth and this is supposed to be bad for him? And I do believe these results are for the metro area and the Louisville metro area consists of more than just Jefferson county.

[Poundingtherock]

Sbane,

If the data is just for the Louisville metro area, then his overall numbers in the county are very likely much worse.  Obama would obviously poll stronger in the metro area than he would in the non-metro portion of the county.

I'd say his Louisville numbers are bad but you are right, his numbers in metro Fresno (though note my point above about his countywide numbers being worse than his numbers in the metro area) are not that bad.

But what to make of Palin's numbers in these two metro areas?

[Sbane]

Sbane,

If the data is just for the Louisville metro area, then his overall numbers in the county are very likely much worse.  Obama would obviously poll stronger in the metro area than he would in the non-metro portion of the county.

I'd say his Louisville numbers are bad but you are right, his numbers in metro Fresno (though note my point above about his countywide numbers being worse than his numbers in the metro area) are not that bad.

But what to make of Palin's numbers in these two metro areas?

Please read this first. Click on Kentucky in the link.

As for Palin's numbers, they don't surprise me much. Places like Fresno and Louisville are her base so if she doesn't put up respectable numbers here she won't do better elsewhere. Some on this forum think Palin is widely unpopular and that she has no chance of winning the nomination but I do believe they are wrong. Though she will be a worse candidate in the general than either Romney or Huckabee.

[Poundingtherock]

Even if we add all those counties together, they probably come far short of the numbers coming out of Jefferson county.  But your point is well-taken.  Thanks for the link.

However, that wouldn't explain the numbers in Fresno (my point about the metro area would seem to hold here, though I'll concede the point as to Kentucky, see California)

http://www.surveyusa.com/SUSA_Regional_Definitions_As_Of_081029.htm

I wouldn't call any "metro" area Palin's base.  For any Republican, the "base" areas are rural areas where he or she will have to run up huge numbers.  However, the Republican candidate will need to do well in "metro" areas like Fresno and Louisville to a lesser extent because in swing states, you cannot just count on the rural vote and will need to lessen the margin in cities such as Fresno and Louisville to win the state.

[Sbane]

Actually the republican base is concentrated in smaller cities and towns (in addition to suburbs of larger metro areas), not rural areas since they don't have that many people to begin with. They may give the largest margins but that is not where the "base" is. Fresno and Louisville are the smaller cities where you would expect Palin to do well in. She won't be doing as well in traditionally republican areas like Orange county, NJ suburbs, Dupage county, Oakland county etc. I would think her approval rating is lower in those areas than you would expect for a republican. Plus the most damning statistic in both polls is the high numbers of people who don't want her to run in 2012. They may "approve" of her but it doesn't necessarily mean they want her to be president. Still as I said before, she is pretty underrated. She will do better than most expect.

[Zot]

This is pretty good news.  Jefferson County (Louisville) gave 0bama his second best margin of Kentucky's 120 counties.  Maybe we can get rid of Yarmuth next year and keep the senate seat in good hands.

[Alcon]

The Louisville DMA is much more than Jefferson County:



McCain won the DMA 52-47.

[Zot]

So are they including the Hoosier counties as well?  I didn't look at the survey and assumed the Louisville results were a subset of a statewide Kentucky poll.

[Alcon]

So are they including the Hoosier counties as well?  I didn't look at the survey and assumed the Louisville results were a subset of a statewide Kentucky poll.

I don't know, but if their survey area is the "Louisville DMA" (anywhere that picks up Louisville television stations, basically) it would have to include parts of Indiana.

If they're only including the portions of the Louisville DMA falling within Kentucky, the result would have been 50.7%-48.0% McCain instead.

If they're oversampling the "Louisville area" sub-section of their state poll, that area would have voted McCain narrowly, 49.7%-48.9%.  I doubt they're doing this; their "Louisville area" sample does not contain four counties -- Carroll, Henry, Oldham and Trimble -- that the DMA does, and "DMA" is a pretty technical term.

No matter what, their sample area voted McCain, and the 52%-47% area is probably what they polled.

[Poundingtherock]

Louisville DMA in SurveyUSA is the following:

http://www.surveyusa.com/SUSA_Regional_Definitions_As_Of_081029.htm (click on Kentucky)

Breckenridge
Bullitt
Grayson
Green
Hardin
Jefferson
Larue
Marion
Meade
Nelson
Shelby
Spencer
Taylor
Washington

[anvi]

I'm as much of a fan of polling numbers as anybody on this forum...but, I'm sorry, what is this debate about?  There are no ordinary circumstances within which Obama would win Kentucky in 2012, and there are no ordinary circumstances under which Palin would win California in 2012.  Shouldn't we be looking and favorables/unfavorables in states that are likely to be...you know...competitive?

[Poundingtherock]

They are relevant because the Louisville area and the Fresno area are relevant as swing counties.  They are similar to counties in states that could be competitive.

The Republican will probably need to lose the Louisville area slightly while winning Fresno county by a somewhat significant margin to win the election.

[anvi]

But counties in swing states are demographically unique: they have different ethnic ratios, age distributions, gender ratios, registration and voter trends, ect. ect.  How do these two districts in Kansas and California help us predict county results in eastern Ohio, southern Virginia and southeastern Florida?   

[pbrower2a]

That's quite a serious charge that you are making.

Here's SurveyUSA's track record from the last election:

http://www.publicpolicypolling.com/pdf/images/Wall%20Street%20Chart.jpg

SurveyUSA also pretty much nailed CA-9, VA-gov, and NJ-Gov (unlike Democracy Corps(D) and the NY Times/CBS news, which embarrassed themselves with their NJ polling):

http://www.surveyusa.com/index.php/2009/11/04/1641/

http://www.surveyusa.com/index.php/2009/11/04/no-pollster-more-accurate-than-surveyusa-in-nj-governor-contest/

http://www.surveyusa.com/index.php/2009/11/04/no-pollster-more-accurate-than-surveyusa-in-va-governor-contest/

Yes, it is a serious charge. SurveyUSA may have been accurate enough in 2008, but it seems way off now. I was referring more to some statewide polls not in that batch -- polls that seem to have contradictory patterns such as contradictory correlations between age and support and the unlikely similarity between approval levels in states with very dissimilar political tendencies. How could Obama support be within 1% in Alabama, Kansas, Kentucky, Missouri, and Virginia?

Another pollster, Strategic Vision, was caught plagiarizing other pollsters, and I can just imagine what happened with some of its employees. I hope that they got jobs elsewhere -- as in the commercial food-service business, segment specializing in low-cost sandwiches.


  

[Poundingtherock]

Of course, you didn't note SurveyUSA being the #1 pollster for CA-9, Va-Gov, and NJ-Gov.

The explanation is one that I've made before.  More of his supporters in Virginia and Missouri disapprove of him than in other states.  That doesn't mean they wouldn't vote for him against a Republican because they would.

He doesn't have much support in Kansas, Kentucky, and Alabama to begin with.

For any approval/disapproval poll, it matters what group is the one doing the disapproving.  If it's liberals/young people/African-Americans/left-of-center folks, they are irrelevant because they will vote for Obama against the Republican even if they disapprove of Obama.

So Obama would probably outpoll his approval rating in Missouri and Virginia by 8-10 points.  I don't recall a poll showing Corzine with over 40% approval and yet, he won 45% of the vote because people who disapproved of his performance were liberals/African-Americans who would never vote for the Republican.

[Alcon]

Louisville DMA in SurveyUSA is the following:

http://www.surveyusa.com/SUSA_Regional_Definitions_As_Of_081029.htm (click on Kentucky)

Breckenridge
Bullitt
Grayson
Green
Hardin
Jefferson
Larue
Marion
Meade
Nelson
Shelby
Spencer
Taylor
Washington

Like I said, a "DMA" has a very specific technical meaning.  SUSA's regions don't line up with DMAs; they just line up with common regional terms (...generally.)  If SUSA is polling the "Louisville area" subsample of their KY polls, calling it the "Louisville DMA" would make no sense.  Why would they even think to use the term DMA unless they were actually polling based on media market boundaries?