Penn poll wrap-up (SUSA, PPP) Reverse Cling
by Jerome Armstrong, Mon Apr 21, 2008 at 05:54:25 PM EDT
IA has their poll out in PA, showing a 49-39 point lead for Clinton. They will also have Monday results out, as will Zogby, so we are going to be poll reading tomorrow as well.
A couple things to note about the vote counting tomorrow. Traditionally, I've noticed that Philadelphia suburbs tend to report earliest in statewide elections. Because of the GOP gerrymandering in 2000, there are many CD's that stretch across county lines, so it will be very difficult to get beyond a rough estimate of the delegate breakdown based on the results tomorrow night.
Jay Cost has a post on the likeness of Ohio and Penn in the closing days of the campaign poll numbers. You'll notice from the graph he has on his blog, that basically the candidates close out the contests with Clinton upticking and Obama staying even. He did this a few days ago, when it was about 47.5 to 43, in favor of Clinton. For Ohio, that upticked to, the day before the election, a 49 to 43 poll of polls result. Right now, with the ten latest poll results in PA, its upticked for Clinton to a 49.5 to 43 result, nearly exact the final margin for PA (50-43). We'll see what the final day brings tomorrow.
I also looked into the two polls that are showing divergence from all the others, SUSA and PPP. The numbers for PPP, 49-46 in favor of Obama, over a weekend poll, are wildly different, even from SUSA, as for their reasoning. PPP shows a single-digit lead for Clinton among women and older voters. That alone skews everything else. There's nothing in external events to explain such a movement, and none of the other polls pick it up. Since not even SUSA shows this movement, I hope for PPP's sake that they are doing some final polling today to save some of their credibility. Without PPP, the final poll of polls result would be 49.9 to 42.2, a 7.7 lead by Clinton with 7.7 remaining.
SUSA's movement seems more plausible, because it resonates with the issues that have been brought up recently. I would dub it the 'reverse cling' voter reaction. It's nearly all happened in the southeast Philadelphia region. They are white and black men, predominantly 18-34, but also 35 to 45. Liberal, they are not church goers. However, they are not college graduates either. I wouldn't be surprised if they were mostly single too. Now, this group is what comprises the movement away from Clinton and toward Obama in the poll to poll shift of a 54-40 lead by Clinton to a 50-44 lead. I suspect this group is also not usually a participant in the primary system, so counting on their vote might be dubious (also note that SUSA has a gender turnout gap that shows a M-F 45-55 split instead of the norm 40-60 that has been happening). It's a very interesting shift, if correct, that SUSA points toward having happened. Not surprisingly, Clinton winds up her campaign in Philadelphia for a last minute appeal. We'll see how crazy it is at the poll booths tomorrow for an indication of turnout.
Update [2008-4-21 22:8:21 by Jerome Armstrong]: Terence Samuel sees this Philly electorate as the key that will swing PA to Obama's column, Why Obama Will Win Pennsylvania, predicting Obama by a point and a half.
Update [2008-4-22 6:21:19 by Jerome Armstrong]: Duh. This demographic above is the Jay-Z' Obama vote!