PA-PPP: Clinton +3 in 4-way, +4 in Head to Head

[Maxwell]

I'm still actually quite chocked by this poll. Hilary's up 7-9 points nationally but only up 3 in PA. Where is she over performing, that's what I'm wondering.

Comparing it with the other PPP is probably more accurate, and she leads 50-45 in nationally and 49-45 in Pennsylvania. Negligible.

[Ljube]

This is an indication that in a close election, PA should vote R.

[Absolution9]

Age demographics skew old even for PA, one of the oldest states in the country. In 2012, 17% of the PA electorate was 65 or older. PPP has 23% of the electorate as senior citizens.

PPP also projects that 80% of the PA electorate will be white when only 78% of PA voters were white in 2012 and most of the voter population loss was older white voters while a disproportionate number of new voters in the state are non-white (especially Latino, even there)

You are going by the exit polls when you say PA electorate was 78% white in the last election.  There is a lot of accurate analysis (Nate Cohn's linked below) showing that the electorate was whiter and older nationally than the exit polls indicated in 2012 and beyond.  Simply put Obama did better with blue collar whites and whites in general than exit polls presented.  Romney probably won only 56-57% instead of 59% of whites.

Can't include a link due to under 20 posts but google Nate Cohn's article "There Are More White Voters Than People Think."  Its a pretty open and shut case, local voter file data and the CPS survey show that the electorate was probably 74-75% white nationally in 2012 and will be 72-73% white this year.

[Virginiá]

Can't include a link due to under 20 posts but google Nate Cohn's article "There Are More White Voters Than People Think."  Its a pretty open and shut case, local voter file data and the CPS survey show that the electorate was probably 74-75% white nationally in 2012 and will be 72-73% white this year.

http://www.nytimes.com/2016/06/10/upshot/there-are-more-white-voters-than-people-think-thats-good-news-for-trump.html?_r=0

You must be logged in to read this quote.

Which is sort of why I personally have just stuck to the exit polls. They provided a compelling argument but gave 3 sets of numbers and suggested all have their own issues. I suppose the CPS numbers are probably the best bet, but I'm hoping after this election we can get a clearer picture of the electorate with this issue factored in.

[Absolution9]

Not disagreeing with you, I was just responding to the claim that the poll got the electorate wrong at 80% white due to the exit polls.  They probably used the CPS survey to estimate the voting population.  I have seen a lot of variety in estimates for both national and state level polls.  Nationally I have seen between 70-73% white so both methods are being used I guess.

I'm not sure why there might be a discrepancy, maybe they over sample in larger urban areas and underestimate the numbers for small towns/rural areas/second tier cities, etc, but I always thought it was fishy that the exit polls produced a 54/46 ratio for non-college/college educated when the population over 25 is currently closer to 71/29.  I get that the college educated group would have a higher proportion voting but not that huge a spread!

[pbrower2a]

A likely-voter  model implies a low turnout in which Republicans are more heavily co0ncentrated in the voting. Democrats and Democratic-leaning interests try to change reality to an extended vote involving not-so-likely voters. Low turnout helps Republicans; high turnout helps Democrats. 

[Erich Maria Remarque]

A likely-voter  model implies a low turnout in which Republicans are more heavily co0ncentrated in the voting. Democrats and Democratic-leaning interests try to change reality to an extended vote involving not-so-likely voters. Low turnout helps Republicans; high turnout helps Democrats. 

Not necessary. The only group, Trump is doing great with is non-college-educated Whites, had a really low turnout 57% (in 2012) and it is still 32-35% of RV. If he manage to energise them more than nonwhites, he'll benefit of a higher turnout.