The sample in the last poll in Pennsylvania from Public Policy Polling before today: 46 percent Democrat, 44 percent Republican, 9 percent independent/other.
The sample in the new poll in Pennsylvania from Public Policy Polling from today: 48 percent Democrat, 41 percent Republican, 11 percent independent/other.
Headline from Public Policy Polling: “The race for Senate in Pennsylvania has seen dramatic movement over the last two months and Joe Sestak has now taken the slightest lead over Pat Toomey, 46-45. On PPP’s previous survey of the state, in mid-August, Toomey led by a 45-36 margin.”
Why, it’s almost as if a more heavily Democratic sample dramatically alters the results!
PPP says they see an enormous awakening among Democrats, worth 5 percent in the past two months. We’ll see.
UPDATE: For perspective:
The 2004 exit poll in Pennsylvania: 41 percent Democrat, 39 percent Republican, 20 percent independent.
The 2006 exit poll in Pennsylvania: 43 percent Democrat, 38 percent Republican, 19 percent independent.
In 2008, the exit polls in this state split 44 percent Democrat, 37 percent Republican, 18 percent independent.
I don't know about this poll, but Fox News has Boxer safely ahead now by 4 pts and Blunt's lead down to 6 in MO and Buck only ahead by a single point. This sure doesn't look like much of a wave election, but we'll see.
Reply to this commentLinkReport AbuseIs this the polling that Obi Wan predicted we would see showing a Democrat surge?
Reply to this commentLinkReport AbuseToo bad PPP isn't required to justify the 4-point enthusiasm gap between 2008, when Pennsylvania was crazy about Democrats and Obama, and the 2010 mid-term elections. The lop-sided and obviously wrong sample used by PPP proves that certain polling organizations are as politically motivated as the media, which makes them useless for the purpose they are intended to serve.
Reply to this commentLinkReport AbuseThe question is: Which is the most accurate party breakdown? It seems unbelievable that the party breakdown would equal that in the democrat heavy year of 2008. My guess is it wouod be a lot closer to 2004, even though that was a highly competitive presidential year. What was 2002 midterm breakdown?
Reply to this commentLinkReport AbuseYou don't think this election is going to be more Democratic than 2008? Oh, Jim, now you're just being silly. Obama has woken up all his supporters and they're all rushing to the polls. Republicans no longer care and will stay home again.
Reply to this commentLinkReport AbuseI know participation by Independents drops off in midterms, but this sample just seems way off. They're predicting a huge drop off among unaffiliated voters from even the 2006 midterms. It just doesn't seem to make any sense. Even if PPP is seeing an increase in enthusiasm among PA Dems, is there any reason to believe this sample?
Reply to this commentLinkReport AbuseOne thing to remember about political polling is the results are only as reliable as the answers of those who are polled. Not everyone tells the truth, especially if they intend to vote for candidates they normally would not.
Reply to this commentLinkReport AbuseA 5-point change in the sampling doesn't account for an 11-point swing in the race. All other things being equal, this would still indicate a tightening of the race.
Reply to this commentLinkReport AbuseSeems the absentee ballot requests tell a completely different story!
Reply to this commentLinkReport AbuseCrunching the numbers a little more, though, most of that swing appears to be in the sampling selected. It doesn't look like it's tightening up all that much.
Reply to this commentLinkReport AbuseRemember folks, PPP is the polling firm for Daily Kos now. Maybe they don't want to hurt their pay masters.
And the D's are going to get thumped at the congressional level in PA, the Gov races was never even a race, etc. This is a garbage poll akin to a newseek or NY times/cbs poll.
Reply to this commentLinkReport Abuse"Boxer is safely ahead by 4 pts and Blunt's lead is down to 6." That is interesting phrasing, mypalfish.
Seriously, if I see another, similar Pennsylvania poll by a different outfit, I think we should all start to worry, but if the PPP sample is overly light in self-styled independents, it strikes me that they (we) are the ones who are most fired up this year. In fact (for example) I already voted: All GOP, except where there was a Libertarian candidate and the race was not crucial.
Toomey is probably safe, even in his blue state.
Reply to this commentLinkReport AbuseWell the oversampling of Democrats alone does not explain why Toomey's numbers are crashing with women. Toomey is bleeding women, which I think it is because of the ads Sestak is running linking Toomey to Palin. I have never seen a Republican so universally despised by women in this state. She is probably even disliked more than Santorum.
Reply to this commentLinkReport Abuse"Toomey is bleeding women, which I think it is because of the ads Sestak is running linking Toomey to Palin. I have never seen a Republican so universally despised by women in this state. She is probably even disliked more than Santorum."
******************
That's an interesting point. Even more interesting is the question of why Palin has this double-edged sword quality. Unfortunately, that's not a question one can discuss honestly in conservative circles without being attacked. In a way, that is the double-edged sword quality of the tea party movement itself.
Reply to this commentLinkReport AbuseSeriously, PPP gave a 4 percent Democrat improvement in turn out over 2008, and all it did was allow Sestak to lead Toomey by _1_ point?
Is Tom Corbett going to be trailing in the governor race by 1 point too under this turnout model?
Reply to this commentLinkReport AbuseOn second thought, maybe the PPP poll is to provide cover for Sestak skipping out on yet another scheduled debate with Toomey. 'I need to raise money to increase my lead, so that is why Toomey will be debating an empty podium.'
Reply to this commentLinkReport AbuseI don't understand the Sestak "skipping out" on a debate comment. They are debating tomorrow night.
Also, I'm not a statistician, but does the bump up in number of responders who identify themselves as Democrats cancel out the heretofore significant over-sampling of responders who identify as Republicans?
Reply to this commentLinkReport Abuse