PA PrimD: Survey USA: Clinton by 6

[Verily]

Clinton: 50 (-4)
Obama: 44 (+4)

http://www.nbc10.com/politics/15943088/detail.html

[Verily]

And I'm ready to make my call. Clinton 53, Obama 46.

[Sam Spade]

It's as good of a comment here as in the other thread, so...

I don't really care about the quality of the poll, fwiw, if it has any.

The point I will continue to make is this:  There have always been in my mind two results:
1) The "race is over" result, where Clinton wins by 6-8 points, where the working-class Catholics in the Philly area decide that the race is over and split.
2) The "race continues as it did before" result, where Clinton wins by 12%-15%, and the working-class Catholics in Philly do their thing.

Now, with SUSA having her lead down to 6 points, my tendency would be leaning towards the former result, rather than the latter, whereas last night, I would be leaning more towards the latter, rather than the former.

The key point is the SE PA subsample - before SUSA had that as tied.  Now they match the other polls which show Obama +15 there.  That's your gap.

Of course, this doesn't mean we won't split the difference and see an 8%-12% margin.  Quite possible, actually.  And SUSA could be wrong here, who knows...

Of course, now Drudge says that Clinton "internals" show her up 11 points.  Caveat emptor.

[Verily]

Crosstabs!

http://www.surveyusa.com/client/PollReport.aspx?g=b92f9f10-4d6b-4f93-b747-e06a37fce20f

[Verily]

White: 81%
Black: 14%

Seems like an underestimate of the black vote. Voting splits look right, though.

18-34: 21%
65+: 21%

Might be an overestimate of the young voters, but I'm not certain on that. Again, the splits look right.

[Alcon]

Obama performance versus last poll:

Men: 53% (+8)
Women: 37% (+4)

Young voters: 57% (+13)
Seniors: 32% (+3)

Whites: 36% (+4)
Blacks: 87% (+13)

College graduates: 49% (?)
Proles: 40% (?)

Southeast: 55% (+12)
Southwest: 36% (+1)

I like how 1% of those with their "minds made" are undecideds.  Commitment to noncommitment!

[Alcon]

White: 81%
Black: 14%

Seems like an underestimate of the black vote. Voting splits look right, though.

18-34: 21%
65+: 21%

Might be an overestimate of the young voters, but I'm not certain on that. Again, the splits look right.

In 2004, Kerry voters identifying themselves by one of the three main races split thusly:

White 74%
Black 22%
Latino 4%

The age split was:

Under 30: 25%
Over 65: 21%

Even accepting that most of the Democrats for Bush were whites, it's still a clear undersampling of blacks.

[Verily]

So, being safe, 18% black would be a good estimate.

Regions are worth looking at, too. This poll has the race unusually close in South Central PA, closer than I would have expected (although maybe majority-black Harrisburg is driving that). I also don't know what the regional % breakdown should look like.

[I spent the winter writing songs about getting better]

So, being safe, 18% black would be a good estimate.

Regions are worth looking at, too. This poll has the race unusually close in South Central PA, closer than I would have expected (although maybe majority-black Harrisburg is driving that). I also don't know what the regional % breakdown should look like.

Sample size in South Central PA of approx. 71.

[Verily]

Bah, I'm overthinking it.

[© tweed]

I have never seen this much primary polling for one state in my life

[Fmr. Pres. Duke]

Clinton wins 55-44. Obama's ceiling is probably 44% in Pennsylvania.

[Eraserhead]

Clinton wins 55-44. Obama's ceiling is probably 44% in Pennsylvania.

O RLY?