The Official Obama Approval Ratings Thread

[Mr. Morden]

[Mr. Morden]

You seem to be forgetting that Obama won NV by 13 points. Also, this is favorables, not approval which would be lower. And plus, it's a Kos poll.

^^^
Correct.  "Favorability" is not "job approval".  One shouldn't treat them interchangeably.  And yes, Obama won NV by ~5 points more than he won by nationally, so the GOP could certainly win without it.  Heck, post-2010 reapportionment, the GOP can win by getting all the states that Bush won twice minus Nevada.  That currently adds up to 269 EV, but would be ~277 EV in 2012.

[Mr. Morden]

Obama went down in today's Gallup:

Favorable: 53% (-2)
Unfavorable: 40% (+2)

That's his job approval, not his favorability.  Should read:

Approve: 53% (-2)
Disapprove: 40% (+2)

[Mr. Morden]

Is it really 2 months now that we had an OH (!) poll ?

PS: pbrower2a, plz colour IN and NE-02 in orange, because the polls are really old and IN was part of an Evan Bayh-sponsored poll.

Have patience. More polls will be out soon. I can't predict which ones, let alone which way they will go. Connecticut appeared tonight. Many agencies will have good cause to show new polls.

For the state of California you will see the number "79" where you expect to see electoral votes. A code for months will be "7"+ X, with X representing the numerical value of the month from January to September or "80" for October, "81" for November, or "82" for December as the opportunities arise.

I am going to give asterisks for the states smallest in territory instead of a code for the month -- states that will all go to the GOP nominee in 2012 only in an Obama loss.   

Cool, but I would just drop the "7" code, and just have the month by itself, leaving out all electoral vote #s.  For states with no month #, just leave them blank.  The 2008 electoral votes aren't going to be used in the 2012 election anyway.

[Mr. Morden]

Again, you are missing the difference between favorability and approval.

^^
Indeed.  Is this a map of favorability #s or approval #s?  Or is it a mix of both?

[Mr. Morden]

"Likely voters in a midterm election (2010)" means something very different from likely voters in a Presidential contest. That is why I cannot accept this polling result.

You mean 2009.  The NJ gov race is this year.

You include other polls of likely voters for 2010 midterms.  These polls are headlined "Colorado Senate" and "New Hampshire Senate":

link

link

so they're for the 2010 midterms, and they poll "likely voters".  They are polling likely voters for next year's midterms, so should you exclude them as well?

[Mr. Morden]

Massachusetts update (no surprise there):

Once again, I would like to point out that you're adding a poll that measures Obama's *favorability* rating, not his *job approval* rating.

[Mr. Morden]

... I promise: unless Indiana or NE-02 is polled by September 31, it goes "orange". Six months is clearly outdated.

I would make the change on Sept. 30th.  Usually, nothing much of anything happens in the world on Sept. 31st. 

[Mr. Morden]

Kentucky? Momentous!

You mean like New Jersey was momentous in the other direction? 

[Mr. Morden]

lol

You must be logged in to read this quote.

Huckabee and McCain might as well drop out.

[Mr. Morden]

lol

You must be logged in to read this quote.

Huckabee and McCain might as well drop out.

Hillary's probably gonna wipe the floor with Obama come '08 anyway.

I don't know.  I could imagine Obama winning Iowa, and creating a wave of momentum going into NH, only to see Clinton pull out a surprise victory in NH, and then a victory in NV, followed by an Obama win in SC.  Then maybe a Super Tuesday that ends in what's virtually a tie, followed by a long slog through to June, with Obama ultimately winning.  Obama might run into trouble though, say if the media manages to find any damaging videos from Rev. Wright sermons.

Actually, scratch all that.  Who am I kidding?  HRC's got it in the bag.

Also, latest Obama approval #s from Zogby in South Dakota (11/13-15):

approve 39%
disapprove 52%

[Mr. Morden]

Considering that it is an internal poll and the fact that Indiana has about 100 House districts and about 50 Senate districts, we shouldn't include this poll into the map (only 20 districts polled).

Plus it's favorability rather than job approval.  Though pbrower seems to be oblivious to that distinction....

[Mr. Morden]

Plus it's favorability rather than job approval.  Though pbrower seems to be oblivious to that distinction....

Indiana checks in, and even if Obama is above 50% disapproval, it's not by much there.

Lol.

[Mr. Morden]

Ask yourself this: why would the Democrats want anything other than an objective poll of support for Obama in Indiana?

Lol.

You must be logged in to read this quote.

Rasmussen doesn't poll for Fox.  Opinion Dynamics does.

[Mr. Morden]

I haven't been rejecting Rasmussen polls because they go through the network that calls itself "Fair and Balanced".

Rasmussen doesn't poll for Fox.  Opinion Dynamics does.

That`s not really true:

http://www.rasmussenreports.com/public_content/politics/elections2/election_20082/2008_presidential_election/missouri/election_2008_missouri_presidential_election

http://www.uselectionatlas.org/FORUM/index.php?topic=87502.0

Ah, OK.  Wasn't aware of that.  Still, most Rasmussen polls are not specifically for Fox.  And Fox more commonly teams with Opinion Dynamics.

[Mr. Morden]

Yet you include (without any similar asterisk on the map) a leaked Democratic internal poll in Indiana, which was a poll of favorability rather than job approval, with the explanation that there aren't many polls of Indiana and "beggars can't be choosers"?

[Mr. Morden]

Approval and favorability differ by little -- 1-2%, which is less than the usual differences between pollsters and certain criteria of selecting "voters" -- adults, registered voters, or "likely voters". A 1-2% difference is within the usual margin of error. The difference between approval and favorability is slight enough for me.

Pollster.com trendline on Obama favorability = 53.7%:

http://www.pollster.com/polls/us/fav-obama.php

Pollster.com trendline on Obama job approval rating = 48.3%:

http://www.pollster.com/polls/us/jobapproval-obama.php

Granted, it's a slightly different mix of polls, with more recent job approval polls included, which could aggravate the gap, but if you look at individual poll-by-poll comparisons, you find that, within the same poll, Obama's current favorability does indeed run something like 4-5% above his job approval rating.  Heck, for the most recent Marist poll, the gap is as large as 9 points.

Personal favorability ratings and opinion of how Obama is doing as president are totally different things.  It's ridiculous to average them together or put them up on the same map side by side.  Especially when both numbers are so close to 50%, so that the choice of which one to use has a significant impact on which direction each state swings.

[Mr. Morden]

OK. Let's suppose that we get favorability polls from North Dakota and Mississippi. Do you want those to be  rejected? If I reject those we have nothing.

So what?  There will inevitably be some states without any polling.  I'm not sure why that's a problem.  We can easily do the mental extrapolation to guess at where things stand in MS and ND, based on what results are showing up in other states.  That would be preferable to making a map that's a mish-mash of polling on two completely different questions.  Combining data from polling on two different questions (approval of the job Obama is doing as president vs. what you think of Obama as a person) makes the map harder to interpret, not easier.

[Mr. Morden]

Posted on Pollster.com, Obama Approval in Wisconsin 60/37 (done by UW)

http://www.pollster.com/blogs/wi_obama_doyle_uwisc_10291120.php

According to that poll, he has 54% approval among likely voters.

94% among Dems, 54% of independents, and 25% of Republicans approve. Probably too high, I think his approval in Wisconsin is around 50\50.

The poll is old, but it is clearly above 50%, even with "likely voters". That marginally breaks a tie. College-based polling takes time because it isn't professional. I'll go with "likely voters".

So now a poll of likely voters for the 2010 midterms is OK?  About 50 pages back in this thread, you were rejecting polls of "likely voters" because "likely voters" in an off-year election aren't the same as likely voters in a presidential year.

Of course, half the polls on your map are probably of likely voters for 2009 or 2010, but you rarely ever pay attention to such things.

[Mr. Morden]

pbrower2a's previous rationale for excluding "likely voter" polls:

The New Jersey poll relates to a gubernatorial race in 2010.

Even when you erroneously thought the NJ gubernatorial race was in 2010, you wanted to exclude the poll.

[Mr. Morden]

Polls sometimes have separate sets of numbers for RVs and LVs though, so you could institute some consistent policy for picking one over the other if a poll has both.

[Mr. Morden]

1. I have no idea of which is more reliable: favorability or approval. I know that there is a difference, and when I see a poll that has both favorability and approval I average them.

And this thread is called "The Official Obama Approval Ratings Thread". I think you have an answer.

Approval and favorability differ by little -- 1-2%, which is less than the usual differences between pollsters and certain criteria of selecting "voters" -- adults, registered voters, or "likely voters". A 1-2% difference is within the usual margin of error. The difference between approval and favorability is slight enough for me.

1. Unless you miss the title of this thread, it is approval and not favorability -- you miss the point.

[Mr. Morden]

CNN/Opinion Research:

http://i2.cdn.turner.com/cnn/2010/images/02/16/rel4a.pdf

approve: 49%
disapprove: 50%

Does Obama deserve to be reelected?

yes 44%
no 52%

[Mr. Morden]

Aren't the Fox News polls simply Rasmussen polls that have been ordered by Fox?  It's the polling organization that conducts the poll that matters, not the news organization that buys them.  Rejecting all "media polls" is kind of silly.  Half the polls out there are "media polls", when you add up everything done for CNN, Fox News, NBC, ABC, CBS, NYT, USA Today, AP, Reuters, McClatchy, etc.  Should you reject a Gallup poll because they've partnered with USA Today?  USA Today didn't conduct the poll.  They're just buying it.

EDIT: OK, apparently, those Fox polls are by Pulse Opinion Research, which is simply a spinoff of Rasmussen Reports:

http://politicalwire.com/archives/2010/09/14/about_those_fox_news_polls.html

You must be logged in to read this quote.

I don't see why you would reject these polls if you're accepting Rasmussen polls.

[Mr. Morden]

What is this "personal approval" thing you guys are talking about?  I think the phrase you're after is "favorability rating".