US-Obama Approval: 44% Approve, 51% Disapprove (PPP/Daily Kos/SEIU 10/20-23)

PPP (D) / Daily Kos / SEIU
10/20-23/11; 1,000 registered voters, 3.1% margin of error
Mode: Automated phone
PPP release

National

Favorable / Unfavorable
Barack Obama: 47 / 49 (chart)

Obama Job Approval
44% Approve, 51% Disapprove (chart)

State of the Country
20% Right Direction, 71% Wrong Track (chart)

Comments

[Dave WI]

09:00 AM on 10/26/2011

Mitt Romney needs a SPINE SPINE SPINE plan.

*ba dum cha*

[Dave WI]

12:00 PM on 10/26/2011

Heh. Round and round we go:
http://twitter.com/#!/mmurraypolitics/statuses/129212957472206849

[Dave WI]

06:46 PM on 10/25/2011

More strong leadership from Romney:
http://www.weeklystandard.com/blogs/romney-declines-take-position-ohio-collective-bargaining-law_603987.html

Man, it's going to be a lonely trip to the polls for me to pull the lever for Huntsman next spring.

[Symanski]

08:10 PM on 10/25/2011

Heh, Huntsman is the one guy that I would vote for among the group against Obama-probably. Though, my CA vote is essentially meaningless.

[stillow1]

09:15 PM on 10/25/2011

Dave - Its just politics right now. Its pretty smart to try and avoid taking positions on controversial issues. IMO he continues to run a near flawless primary campaign.

[Dave WI]

09:23 PM on 10/25/2011

That's the thing though, if he was trying to avoid taking a position on it, great.

But back in the summer, he was telling people that he stood with his friend John Kasich. Which is it?

[Dave WI]

09:33 PM on 10/25/2011

To elaborate, if he wants to stay out of state issues, great. If he wants to get involved, great.

But don't get in and then jump out when the going gets tough.

[stillow1]

04:17 PM on 10/25/2011

WhatHave- Notice how the topline resutl changes when a more legit sample is used? Last week PPP used a D+8 and Obama was -3....now they use a pretty decent sample of D+2 and suddenly the topline resutl changes to -7. So much for that uptick you said you were starting to see. I told you there is more to polls than just the top line.

[Field Marshall]

05:14 PM on 10/25/2011

I guess PPP has to do a real poll with a real sample when someone is paying for it.

[Bukama]

05:16 PM on 10/25/2011

Especially when the customer is KOS, since he's been burned before.

[dpearl]

05:23 PM on 10/25/2011

Last week was also a PPP/Kos/SEIU poll so I am not sure what you are getting at FM.
Also, both polls gave almost identical results. We are all free to reweight the results anyway we like since they provide all of the individual subject data.

As Stillow points out - if you apply last week's party identification distribution to this week's approval responses you get back to the 46/49 results they had last week and similarly if you like this week's party identification distribution then go ahead and apply them to last week's data. PPP does not manipulate the party id data in any ways - they simply report what people tell them using a consistent methodology.

[Xenobion]

05:48 PM on 10/25/2011

You're pretty much arguing the MOE on both these polls. If you have such problems with 4% varying difference based off of differing party ID models I'd say I hope you don't lose any sleep over it. What we've come to see with PPP is that they constantly change their party ID model. People want to say that this represents the shifting change of attitudes when simply put polling reflects a separate reality of expectations. This is the explanation of the trend of polling 13 months out. Pollsters are scatter shotting their party ID models and never keeping them consistent if you haven't figured it out yet.

[dpearl]

06:01 PM on 10/25/2011

No - they are not changing any party ID model whatsoever. They ask the self-reported party ID as one of their questions and report their findings. That is all. They get random fluctuations in party ID just as you would expect random fluctuations in any of their results.
That being said...

If you want to make a comparison across different PPP polls then Stillow has a point in that applying a constant party identification distribution will give you a more consistent comparison.

It would be neat if a web site like this one had the software built in to allow us to do that with any pollster that reports the relevant crosstab.

[stillow1]

09:12 PM on 10/25/2011

Thanks dpearl for helping to explain what i was saying.

[Alois SaintMartin]

03:42 PM on 10/25/2011

Something to Pack in your Sack as you head for Cheviot Hills Mr. President !