Franklin and Marshall has a new poll out on Pennsylvania. You can read the full document here. They have pegged Specter at an approval rating of 23%, but it's actually much worse than that. Here's their chart:

In addition, F&M looks at the head-to-heads for Specter with Toomey in the general. Their numbers show Specter at 33% to Toomey's 31%, a statistical dead heat, with 30% undecided.

It's likely actually worse than that because of the weighting problems F&M has, for which they were dinged last year over and over, but still haven't fixed. F&M skews their population data to favour the center of the state (where they are located) and grossly undercounts the southeastern part of the state. To wit, F&M cites "Southeastern Pennsylvania" at 19% of the state population, with Philadelphia at 12%. While the Philadelphia number is close enough, "Southeastern" is actually about 28% of the population. They also undercount Southwestern and Northeastern Pennsylvania, but not by as much. They ascribe 28% of the population to Central Pennsylvania. Amazingly, if you look at a map of the 19 CDs in Pennsylvania, the entire center of the state is TWO CDs.

This means they have a tendency to undercount progressive Democrats, and overcount moderate Democrats and all Republicans. It also brings into question their favourable/unfavouable ratings of Joe Sestak, since they claim that 77% percent of state residents don't know who he is, UP from 73% in August. The thing that makes this claim so specious is that Sestak is a second-term Congressman who advertised on Philadelphia TV. That means that 40% of the population of the state was in the TV viewing area where his commercials were shown. It's completely unreasonable that people in the Southeastern part of the state don't know anything about him.

When F&M therefore says that in the primary match-up, it's Specter 30%, Sestak 18%, they have grossly undercounted the CD Sestak represents, as well as the Democrats least likely to vote for Specter in the primary next spring. As an aside, the is the worse spread F&M has seen for Specter since they started polling this match-up in June., with the spread falling from 20 points to 12 points, with no appreciable difference in the undecideds.

DocJess :: New Polls on Specter Even Worse Than They Seem

Follow Democratic Convention Watch on Facebook and Twitter. Iphone/Android apps available.

Tags: Pennsylvania, Senate, (All Tags)

Print Friendly View

Send As Email

New Polls on Specter Even Worse Than They Seem | 4 comments

More numbers (0.00 / 0)

Actually, Specter's favorability rating in this poll is 28% (compare to 32% for Casey, the other Senator). 23% is the percentage who answer "deserves reelection" to this question:

Do you believe that ARLEN SPECTER has done a good enough job as
SENATOR to deserve re-election, or do you believe it is time for a change?

I don't like that question much. They don't ask an equivalent question about anyone else, and the mild mismatch with favorability and head-to-head contests make it a bit of a headscratcher.

Here's an interesting series of numbers in this poll. They are the percentage who think each person is doing a poor job handling their job:

Specter: 28%
Obama: 28%
Rendell: 35%
State Legislature (overall): 30%

By that measure, Specter is not in particularly bad shape.

Overall, the poll shows Pennsylvanians are hurting, and are not terribly pleased with any of their government right now. Almost no one loves Specter, which is not surprising for a party-switcher, but his favorability ratings are taking a nose-dive, which does seem like it could set off alarm bells for Specter and be good news for Sestak.

by: SarahLawrence Scott @ Fri Oct 30, 2009 at 15:10:26 PM UTC

by: you @ soon

To post this comment click here:

Otherwise click cancel.

Comparative Numbers (0.00 / 0)

The thing is, there IS no one else running for re-election who was polled on. Sestak's not running for Congress, Rendell's not running for Governor...etc. -- so there really is no comparison.

You could nitpick and say they could have said "Do you think Joe Sestak has done a good enough job to warrant election to the Senate?" but it wouldn't really be fair, nor easy to poll.
People might think he'd done a good enough job for re-election to the House, but honestly don't know if he's qualified for a different job (except, maybe, Navy Admiral.)

But on its face, it IS a fair question, and he does poorly. And I think it's more a test of his favourability. As in - even if you like the guy, are you REALLY voting for him?

And he IS in bad shape, at least around here. Every weekend, Sestak is around. The picnics, the functions, the meetings -- he's out meeting everyone. Arlen is NOWHERE to be found.

by: DocJess @ Fri Oct 30, 2009 at 15:36:03 PM UTC

[ Parent ]

Jess (0.00 / 0)

Jess
Up above you said:

In addition, F&M looks at the head-to-heads for Specter with Toomey in the general. Their numbers show Specter at 31% to Toomey's 33%, a statistical dead heat, with 30% undecided.

BUT on page 7 of the Franklin and Marshall report that you gave a link to says

In the general election race, Specter has a small lead over Pat Toomey, 33% to 31%, with nearly a third (30%) undecided

Did you get your numbers mixed up?

And from RealClearPolitic.com:

October 28, 2009
Specter vs. Toomey
Franklin & Marshall
Toomey 31, Specter 33 ... Specter +2

October 28, 2009
Franklin & Marshall
Specter 30, Sestak 18 ... Specter +12

October 15, 2009
Democratic Primary
Rasmussen Reports
Specter 46, Sestak 42 ... Specter +4

October 14, 2009
Susquehanna
Specter 44, Sestak 16 ... Specter +28

So in conclusion, I still have not seen anything that shows that Specter is doing badly in the head to head polls. I understand that you are not backing Specter, but he is still ahead.

by: Leah @ Fri Oct 30, 2009 at 20:43:14 PM UTC

by: you @ soon

To post this comment click here:

Otherwise click cancel.

Leah (0.00 / 0)

Yup, I accidentally reversed the numbers, and have now corrected them. It's an honest mistake...though it actually is within the margin of error and therefore not statistically significant.

The reasons I believe Specter is toast have nothing to do with whether or not I support him personally. It has to do with having lived in Pennsylvania for more than 20 years, and knowing how things work here.

First, the number of people who know EVERYTHING about Specter is relatively huge. He's been the Senator since many people can remember. They do NOT remember who was Senator before him.

On the other hand, Pat Toomey and Joe Sestak are relatively unknown outside their respective Congressional areas.

Arlen can't change people's minds in terms of what they think about him: those images are formed. There is, however, a ton of time for information to be presented on the other candidates. This will be a HUGE election next year.

If people liked Specter, didn't care about his party shift, and were willing to vote for him next year, that would show in the numbers, the undecideds would fall, and he'd have an approval rating of how he handles his job at least double what it is now.

That would mean he'd have the support of enough people for his run to be credible.

Here's the issue: one of the reasons that Clinton won the PA primary last year was because she had the Rendell administration/party force, behind her. To understand Pennsylvania politics, you have to understand the impact of that endorsement, and the money and organization that comes with it. The party is controlled by the incumbent Democratic governor, or the last Democratic governor, and if you (as a mayor, county executive, or county chair) don't play ball - BAD THINGS HAPPEN.

There is an exception for the city of Philadelphia, because the wards are still under direct local control, but there are still issues. Last year, there was a schism in who was supported by the ward leaders and the Council of 70 in the primaries, and the City was "punished" with outcomes related to the schools (partially run by the state under court order) and the casino licenses.

Rendell is supporting Arlen. Again. And this time, the party, the towns, and the counties are NOT falling in line. This hasn't happened before. Ever.

You can see something different in the gubernatorial race: Rendell is supporting Onorato, while Hoeffel is preferred by many who supported him in direct opposition Rendell's disdain the last time he ran for Senate. BUT the money is on Onorato, a beyond-the-blue-dogs-homophobic-misogynistic-anti-choice-pro-gun-Democratic-wingnut. Joe's support is soft because it's the trade for going against Fast Eddie in the Senate race.

THAT'S my analysis: "All politics is local" and this is how we've always played here.

by: DocJess @ Sat Oct 31, 2009 at 08:37:08 AM UTC

[ Parent ]