Obama's Approval Ratings Down to 49%.

[Big Orange]

OK, it seems like the honeymoon period is over for President Obama, although he is still in fourth place behind President Clinton:

Obama Job Approval Down to 49%
President becomes fourth fastest to slip below the majority approval level

PRINCETON, NJ -- The latest Gallup Daily tracking results show 49% of Americans approving of the job Barack Obama is doing as president, putting him below the majority approval level for the first time in his presidency.

Although the current decline below 50% has symbolic significance, most of the recent decline in support for Obama occurred in July and August. He began July at 60% approval. The ongoing, contentious debate over national healthcare reform has likely served as a drag on his public support, as have continuing economic problems. Americans are also concerned about the Obama administration's reliance on government spending to solve the nation's problems and the growing federal budget deficit. Since September, Obama's approval rating had been holding in the low 50s and, although it has reached 50% numerous times, it had never dropped below 50% until now.

Of the post-World War II presidents, Obama now is the fourth fastest to drop below the majority approval level, doing so in his 10th month on the job. Gerald Ford dropped below 50% approval during his third month in office, and Bill Clinton did so in his fourth month. Ronald Reagan, like Obama, also dropped below 50% in his 10th month in office, though Reagan's drop occurred a few days sooner in that month (Nov. 13-16, 1981) than did Obama's (Nov. 17-19, 2009).

But all presidents except John Kennedy dropped below the majority approval level at some point in their presidencies, and all recovered after the first time below this mark to go back above 50% approval.

Half recovered quickly -- Harry Truman, Dwight Eisenhower, Ronald Reagan, and George W. Bush were back above 50% approval the same month they dipped below that mark, and Lyndon Johnson returned to majority approval in his second month after his initial sub-50% rating. The remainder continued to sink further below 50% and remained there for many months, but eventually recovered, with Ford's effort boosted by the Mayaguez incident.

George H.W. Bush took the longest time to recover, going more than a year before his approval rating was restored to the majority level. Unfortunately for him, the 1992 election at which voters denied him a second term in office occurred during this time, and his rating did not go back above 50% until just before he left office in January 1993.

Notably, most of the presidents who fell below 50% during their first term recovered in plenty of time to be re-elected. This list includes Truman, Nixon, Reagan, Clinton, and George W. Bush. Ford and Jimmy Carter recovered in the short term but could not sustain improvement long enough to convince voters they deserved a second term, and, as noted, the elder Bush recovered after he was voted out of office. Johnson may well have joined that list had he decided to seek a second term. And Eisenhower did not go below majority approval until after he had been re-elected.

Thus, Obama's descent below 50% is an important symbolic milestone in his presidency, but history suggests the odds of his regaining majority approval are high, and he could do so relatively soon, particularly since the individual nightly numbers for him in recent days have been right around the 50% mark. History would suggest his current loss of majority approval bears little relation to his chances of being elected to a second term in 2012.

Gallup

[Crossroads Inc.]

Well the polls can go hang themselves, When Bush was in his mid 20%'s every republican I knew was going around with a stright face saying they support him 100%!

[CarsonPalmer]

I think also that Obama's approval rating is here affected by people to his left; if health care passes, his approval will go back up if only due to that group.

[The Original Nex]

Daily poll trackers and individual polls mean very little. I hate it when news organizations pick out a single poll (on any person or issue) and claim that "so-and-so has an approval of x%" as if that's the end of the discussion.

Polling averages like pollster.com does paint a better picture of the mood of the populace, and indeed Obama's approval has dropped from the highs early in his administration. The big drop can be attributed to several things, he's lost just about all the Republican support he had as they've closed ranks against him, he's lost independent support because a big chunk of independents are actually jaded Republicans who simply left the Party at the end of the Bush administration, as well as Mindless Middle wobblers who hear bad things about Obama in the media and so waver against him. He's also lost points among his own BASE who perceive him as not moving fast enough and not being sufficiently left-wing/assertive/what-have-you.

So yes, Obama is less popular than he was during the Era of Good Feeling that was after his Inauguration. That could only be expected. He's also in the midst of a long, drawn out legislative battle, which has allowed the opposition to slander and fear-monger people into opposition on the one hand, and discourage supporters on the other hand. Don't get hung up on individual polls (especially the Gallup Daily Tracker) or the fluctuations of percentage points that have no statistical value, but supply sensationalist media outlets with a headline.

[Gilthan]

Typically each president starts out with relatively bipartisan approval briefly after inauguration, then polls return to more an usual partisan divide, while influenced particularly by perceptions of the economy:

Deficit reduction has remained number one for voters ever since President Obama listed his four top budget priorities in a speech to Congress in February. Forty-two percent (42%) say cutting the deficit in half by the end of the president's first term is most important, followed by 24% who say health care reform should be the top priority.

Fifteen percent (15%) say the emphasis should be on the development of new energy sources, while 13% say the same about education.

[Coyote]

It has seemed to me that lately the conservatives (from organized GOP straightlaces to teabagger nutbars) have been interpreting the lowering of approval for Obama as some sort of indication that their own ideas are beginning to take hold of the population; ie, that people are leaving Obama's sinking ship and swimming towards their own.

But in order to "leave" Obama you have to have been there with him in the first place. You don't leave the side of someone you never supported; you were never there in the first place. I'd guess that most of the people "leaving Obama" are not fence-sitters or right-wingers, but people who wanted a more progressive agenda but are disappointed they aren't getting it. The people who wanted DADT repealed by Executive Order, the people who wanted us 100% out of Iraq and Afghanistan, the people who are disappointed that Guantanamo Prison isn't gathering cobwebs, the people that wanted Public Health Care.

None of these people are going to go to the Right. In fact, they'll remain Democratic voters, they'll just be "hold your nose" voters; some few will simply drop out of the voting process again (probably where they were before) and some will go back to fringe groups like the Greens Party.

Leaving Obama won't mean a rally to the right wing; it'll just mean a dilution of support for Obama in the future and any future win will be by slimmer margins, making it easier to box him in and stall things. But the GOP could still disintegrate, they actually have (IMO) bigger problems. Their party is losing ground to the teabagger/birther nutbar crowd, and the Libertarians will probably scoop up some of the business-suited types. Some of the hard-core Talibangelicals will either stay to try to usurp the Republican brand name; others may go to the Constitution Party, which wants a national return to the glory days of somewhere around 1780.

[Gilthan]

At the time of the 2008 elections, Democrats held around a 6% edge over Republicans, resulting in the Democratic Party increasing their majority in Congress as well as gaining the Presidency.

For how it has changed in polling since then:

Republican candidates maintain a six-point advantage over Democrats in the latest edition of the Generic Congressional Ballot.

The latest Rasmussen Reports national telephone survey shows that 44% would vote for their district’s Republican congressional candidate while 38% would opt for his or her Democratic opponent.

Support for each party rose one point from last week. Republicans have held the lead for over four months now.

Voters not affiliated with either party continue to heavily favor Republicans, 41% to 24%.

Since late June, support for Republican candidates has ranged from 41% to 44%, while support for Democrats has run from 36% to 40%. Looking back one year ago, the two parties were in a much different place. Throughout the fall of 2008, support for Democratic congressional candidates ranged from 42% to 47%. Republican support ranged from 37% to 41%.

http://www.rasmussenreports.com/public_ ... nal_ballot

[Duckie]

I have no idea why this thread continually cites Rasmussen as if they didn't have a noticeable 5% republican lean in presidential opinion and slightly less in generic ballot compared to every single other polling firm's polls.

Also, the reason why independents heavily favour Republicans is that a significant chunk of republicans have abandoned their party since the bush era, resulting in a 38-38-25 split for DIR, compared to the usual 40-30-30 (which is, despite vast democratic number advantage, about 50-50 since democrats have more problems with party loyalty and independent thinking voters and also have most of their voters in republican states packed into diluted urban districts because of gerrymandering). So most "Independents" now actually is heavily biased with former republicans, moderate and frothing at the mouth "We lost 2008 because we didn't focus on abortion and gays" fundamentalists and teabaggers alike.

[Gilthan]

I have no idea why this thread continually cites Rasmussen as if they didn't have a noticeable 5% republican lean in presidential opinion and slightly less in generic ballot compared to every single other polling firm's polls.

Unadjusted polling of the general populace versus instead polling of registered or likely voters always gives a little different results, as the Democrat-to-Republican ratio is high among demographic groups like young adults who have lower turnout. I watched them during the 2008 elections, and Rasmussen predicted Obama's victory precisely, within a fraction of a percent of the actual results.

See this comparison of poll accuracy, based on predicted versus actual historical results:

http://www.fordham.edu/images/academics ... ection.pdf

[Mr Bean]

I have no idea why this thread continually cites Rasmussen as if they didn't have a noticeable 5% republican lean in presidential opinion and slightly less in generic ballot compared to every single other polling firm's polls.

Unadjusted polling of the general populace versus instead polling of registered or likely voters always gives a little different results, as the Democrat-to-Republican ratio is high among demographic groups like young adults who have lower turnout. I watched them during the 2008 elections, and Rasmussen predicted Obama's victory precisely, within a fraction of a percent of the actual results.

See this comparison of poll accuracy, based on predicted versus actual historical results:

http://www.fordham.edu/images/academics ... ection.pdf

That's a very odd pdf you linked to because it contains no useful information and I'll note it was conducted on November 5th when returns were still coming in. It two almost three weeks for all races to be decided all votes counted.

Poll Accuracy in the 2008 Presidential Election
—Initial Report, November 5, 2008—
Costas Panagopoulos, Ph.D.
Department of Political Science
Fordham University
For inquiries: cpanagopoulo@fordham.edu or (917) 405-9069
For all the derision directed toward pre-election polling, the final poll estimates were not
far off from the actual nationwide voteshares for the two candidates. On average, preelection
polls from 23 public polling organizations projected a Democratic advantage of
7.52 percentage points on Election Day, which is only about 1.37 percentage points away
from the current estimate of a 6.15-point Obama margin in the national popular vote.
Following the procedures proposed by Martin, Traugott and Kennedy (see Public Opinion
Quarterly, Fall 2006, pp. 342-369) to assess poll accuracy, I analyze poll estimates from
these 23 polling organizations. Four of these polls appear to have overestimated McCain
support (indicated with a * below), while most polls (17) overestimated Obama strength.
Pre-election projections for two organizations’ final polls—Rasmussen and Pew—were
perfectly in agreement with the actual election result (**).
The following list ranks the 23 organizations by the accuracy of their final, national preelection
polls (as reported on pollster.com).
1. Rasmussen (11/1-3)**
1. Pew (10/29-11/1)**
2. YouGov/Polimetrix (10/18-11/1)
3. Harris Interactive (10/20-27)
4. GWU (Lake/Tarrance) (11/2-3)*
5. Diageo/Hotline (10/31-11/2)*
5. ARG (10/25-27)*
6. CNN (10/30-11/1)
6. Ipsos/McClatchy (10/30-11/1)
7. DailyKos.com (D)/Research 2000 (11/1-3)
8. AP/Yahoo/KN (10/17-27)
9. Democracy Corps (D) (10/30-11/2)
10. FOX (11/1-2)
11. Economist/YouGov (10/25-27)
12. IBD/TIPP (11/1-3)
13. NBC/WSJ (11/1-2)
14. ABC/Post (10/30-11/2)
15. Marist College (11/3)
16. CBS (10/31-11/2)
17. Gallup (10/31-11/2)
18. Reuters/ C-SPAN/ Zogby (10/31-11/3)
19. CBS/Times (10/25-29)
20. Newsweek (10/22-23)

It does not note how far off they were, what their values were. Something more useful might be Nate Silver of 538 who went back and obtained information on how the average polling places preformed and what their accuracy has been on elections polling VS what actually happened.



What Rasmussen is know for is Push-Polling. Writing terrible misleading questions to get a predicted results like showing Anti-aborting forces as being the majority because the question was written "Should we sell aborted children as meat on the Internationale market" and every "No vote" becomes a Rasmussen Pro Life recorded vote. Or as they got caught recently with an actual question like. "Do you support the takeover of heath care by the government and the rationing of heath care to benfit minorities?"

Well ask a question like that and you will get quite a few "No" votes on the public option. Rasmussen's political counting(IE how many votes a Republican or Democrat will get) is generally speaking spot on. Their issue polling is where they slide off to the righward.

[Duckie]

Well, I'm not claiming Rasmussen are inconsistent and random, to clarify- their bias is very consistant and very accurate. It's probably not purposeful- they probably just oversample older americans possibly due to not sampling cell-phones or the like. So they're a good pollster, if you remember to add about 5% to their presidential approval polls and 'is the conutry on the right track' and the like.

But yeah, Rasmussen has had some pretty terrible question wording. For example, on Health Care:

Suppose that Democrats agreed on a health care reform bill that is opposed by all Republicans in Congress. Should the Democrats pass that bill or should they change the bill to win support from a reasonable number of Republicans?

What's a reasonable number? This question also links 'reasonable' to the Republican position implicitly in people's minds by having the words right next to eachother. It also implies that it's unreasonable to pass a bill without Republicans supporting it.

Another one of the great ones was:

Do you agree or disagree with the following statement... it’s always better to cut taxes than to increase government spending because taxpayers, not bureaucrats, are the best judges of how to spend their money?

Apparantly, 62% of people support tax cuts

Not only does it say "Government spending = bureaucrats taking your money", it also has "Tax Cuts are good because you are the best judge of how to spend your money" right in the question.

So yeah, I don't trust Rasmussen as far as I can throw them. Their polls are fixable if you account for the systematic bias in their sampling (every pollster has systematic bias, but rassmussen are extremely noticeable about this) and then check all their question wordings to throw out the terrible polls like the tax cuts one, but why bother? Just use other people's polls that are far less deviated from the mean.

Actually, looking closer, their approval rating is kinda close recently (it was pretty bad months ago), if still a little biased down from the average and well out of the mean. It's their disapproval rating that's absurdly wrong (either it's wrong or every other pollster and the average of all pollsters is).




The dotted line is rasmussen. Notice how it's at the edge of the aggregate of all polls in a Republican manner. The results are reasonable- it's certainly plausible, especially if they have trouble getting land-line-less young people to talk to. But I don't think they're accurate.

[Gilthan]

Something more useful might be Nate Silver of 538 who went back and obtained information on how the average polling places preformed and what their accuracy has been on elections polling VS what actually happened.

[img.]http://img.photobucket.com/albums/v733/ ... f95a_o.png[/img]

Your chart also shows Rasmussen as unusually accurate, with a little different figures and #3 out of 30-some polls in accuracy instead of #1 out of 20 polls, yet fundamentally again as above-average accuracy.

Rasmussen's political counting(IE how many votes a Republican or Democrat will get) is generally speaking spot on. Their issue polling is where they slide off to the righward.

Here we were talking about political counting of likely voters leaning Republican or Democrat. The separate topic of issue polling does depend greatly upon the wording of a poll.

But yeah, Rasmussen has had some pretty terrible question wording. For example, on Health Care:

Suppose that Democrats agreed on a health care reform bill that is opposed by all Republicans in Congress. Should the Democrats pass that bill or should they change the bill to win support from a reasonable number of Republicans?

What's a reasonable number? This question also links 'reasonable' to the Republican position implicitly in people's minds by having the words right next to eachother. It also implies that it's unreasonable to pass a bill without Republicans supporting it.

Here's the wording of a health care survey question:

Generally speaking, do you strongly favor, somewhat favor, somewhat oppose or strongly oppose the health care reform plan proposed by President Obama and the congressional Democrats?

Such is shown at:
http://www.rasmussenreports.com/public_ ... 21_22_2009

Among likely voters, 21% strongly favor, 17% somewhat favor, 13% somewhat oppose, 43% strongly oppose, and 6% are not sure.

Rasmussen does a variety of poll questions, some of which approach the matter from different angles (as occurs with some other pollsters), yet they're pretty accurate for the results of the questions asked. They even themselves sometimes illustrate how issue polling results can vary depending on how the question is asked, like I recall one case where most Americans supported one program in a generic poll by Rasmussen yet the opposite if told its particular $ billion cost in another Rasmussen poll.

Actually, looking closer, their approval rating is kinda close recently (it was pretty bad months ago), if still a little biased down from the average and well out of the mean. It's their disapproval rating that's absurdly wrong (either it's wrong or every other pollster and the average of all pollsters is).

[img.]http://img81.imageshack.us/img81/2963/firstgraph.png[/img]
[img.]http://img99.imageshack.us/img99/428/secondgraph.png[/img]

Your graph is comparing apples and oranges.

Your chart comes from here:

http://www.pollster.com/polls/us/jobapp ... ded-68228B

Notice where some polls have RV, LV, and some A in the table you didn't show? RV stands for a poll of registered voters, LV for a poll of likely voters, and A for all the populace unadjusted.

With its focus on likely voters, Rasmussen was also above-average accuracy in not only the 2008 but also the 2004 elections, like this snippet of a longer Slate magazine article able to be read here:

Rasmussen cleaned Gallup's clock, missing the spreads by an average of 1.6 points compared to Gallup's 6.2. Rasmussen also whipped Zogby, erring by 1.0 points compared to Zogby's 3.2.

http://www.slate.com/id/2110860/

[Duckie]

Who says their likely voter model is any good? Rasmussen clearly is well outside the mean in those graphs, which use likely-voters as well (for instance, the latest one on Obama's approval was 1500 LV), probably because of its Likely Voter Model which oversamples Republicans, people in rural areas, and undersamples youth with no land-line connection. Further, Rasmussen uses IVR, which is good for avoiding Bradley Effect-equivalents but is terrible for getting youth responses.

Your reply is nonsensical. "They were accurate in the Presidential Election, also, they use a likely voter model."

Well, yes. But accuracy in the Presidential Election does not guarantee they are accurate at approval ratings and at issue polling, which is what they're currently doing. No one cares about the 2008 election or 2004 polling, as Mr Bean states, yet you continue to repeat their accuracy in the presidential elections as if that actually matters for polling about approval ratings- the fact that they can get horse race polling right doesn't mean they're any good at non-horse-race polling. "Do you approve of President Obama's job performance" is not a horse race poll.

I specifically noted that their question wording sucks, which you claim is an "interesting fact" that people's opinions can vary based on how they're asked.

Well no fucking duh, that's why they're a shit pollster- their questions have ranged from innocuous to biased towards republicans. There shouldn't be that range at all, they should all be neutral and well worded.

Further, since their average on these polls is well shifted outside of the mean of every single other poller, that means they're either incorrect and have a systematic bias towards republicans or every single other poller is systematically biased in the opposite direction except Rasmussen. Which do you think is more likely according to Occam's Razor?

[Gilthan]

Who says their likely voter model is any good?

The accuracy of any model is determined by comparing to reality, in this case predictions versus results in the 2004, 2006, and 2008 elections (typically higher accuracy for Rasmussen than other major pollsters, as previously discussed).

But accuracy in the Presidential Election does not guarantee they are accurate at approval ratings and at issue polling, which is what they're currently doing. No one cares about the 2008 election or 2004 polling, as Mr Bean states, yet you continue to repeat their accuracy in the presidential elections as if that actually matters for polling about approval ratings- the fact that they can get horse race polling right doesn't mean they're any good at non-horse-race polling. "Do you approve of President Obama's job performance" is not a horse race poll.

Most of the difference between Rasmussen and some polls on approval ratings is if looking at those who "strongly approve" versus "strongly disapprove." For rather simply total approval, Rasmussen has President Obama at 46% while Gallup has 49%, not a very huge difference when considering that one samples likely voters while the other samples all adults.

Also, the following mentioned before is mostly a "horse race" poll:

The latest Rasmussen Reports national telephone survey shows that 44% would vote for their district’s Republican congressional candidate while 38% would opt for his or her Democratic opponent.

Since late June, support for Republican candidates has ranged from 41% to 44%, while support for Democrats has run from 36% to 40%. Looking back one year ago, the two parties were in a much different place. Throughout the fall of 2008, support for Democratic congressional candidates ranged from 42% to 47%. Republican support ranged from 37% to 41%.

Further, since their average on these polls is well shifted outside of the mean of every single other poller, that means they're either incorrect and have a systematic bias towards republicans or every single other poller is systematically biased in the opposite direction except Rasmussen.

Of the few other major pollsters, here's part of what an investigation at RealClearPolitics concluded, which suggests some of the factors in why Rasmussen may be a better than average predictor of the 2010 elections when the time comes (although of course much can change over the year to go), much like they were in the 2004, 2006, and 2008 elections:

As at least one observer has pointed out, Rasmussen samples likely voters, rather than registered voters or adults.

But there's more going on here than just an A/RV/LV distinction, though.

Good pollsters will generally ask their horse race questions toward the beginning of the poll. This is because "policy preference" questions or questions involving other political figures can ultimately skew the result of a later horse race question, by unintentionally leading readers to view the horse race in a certain frame. ABC/WaPo asks the horse race question as question 18. CBS asks it after some thirty-odd questions. And both pollsters ask the horse race question after a bunch of questions about a very particular policy topic: health care! This is understandable since it's the hot issue of the day, but its also an issue where Democrats typically hold an advantage in the public mind. This in turn probably left respondents somewhat more primed to answer "Democrat" than they would otherwise be. Additionally, the questions about Obama (who is generally regarded somewhat more favorably than "generic Democrat") probably have some priming effects as well.

In other words, if you are asking which pollsters have it right, I'd probably put my money on Gallup-Rasmussen rather than ABC/WaPo-CBS.

http://realclearpolitics.blogs.time.com ... ic_ballot/

Obviously no poll at this time in 2009 means much, yet I'll be watching Rasmussen particularly closely when the 2010 elections approach, as was found to be worthwhile in prior elections.

[D.Turtle]

Right, Rasmussen has no Republican lean at all:

[Gilthan]

The actual results in the 2008 elections were closer to Rasmussen than most there. For instance, that graph depicts ABC/Post as only slightly leaning towards overpolling democrats, while it depicts Rasmussen as greatly overpolling republicans.

Yet here were the predictions versus actual results:

ABC/Post, final prediction:

54% Obama, 41% McCain among RV
Predicted: 13% Obama victory
53% Obama, 44% McCain among LV
Predicted: 9% Obama victory
http://www.washingtonpost.com/wp-srv/po ... 10308.html

Rasmussen, final prediction:

52% Obama, 46% McCain
Predicted: 6% Obama victory
http://www.rasmussenreports.com/public_ ... _did_we_do

Votes counted after the election:

53% Obama, 46% McCain
Actual: 7% Obama victory
http://www.infoplease.com/us/government ... mmary.html

In this example, Rasmussen was closer to accurate than the ABC/Post poll, despite how a golden mean fallacy looking at that graph would lead to assuming the opposite.

This ties back to the other comparisons of Rasmussen versus different polls made earlier, whether the document in which its accuracy was rated as the top #1 out of 20 polls or the one in which its accuracy was rated as #3 out of 30-plus polls.

Of course, arguing over plus or minus a couple percentage points isn't particularly important, but it is noticeable how actually most polls have a bit of a Democrat lean when their predictions are tested against election results.

[D.Turtle]

Frankly, using only the last poll published before the election is disingenious. What matters is how accurate they are in all their polls.

Unsurprisingly, shortly before an election the polls magically converge in a small range, while before that their results are spread around a whole lot more.

For example the last poll results before the November election last year:

Code: Select all

Poll	Date	Sample	MoE	Obama (D)	McCain (R)	Spread
Final Results	--	--	--	52.9	45.6	Obama +7.3
RCP Average	10/29 - 11/03	--	--	52.1	44.5	Obama +7.6
Marist	11/03 - 11/03	804 LV	4.0	52	43	Obama +9
Battleground (Lake)*	11/02 - 11/03	800 LV	3.5	52	47	Obama +5
Battleground (Tarrance)*	11/02 - 11/03	800 LV	3.5	50	48	Obama +2
Rasmussen Reports	11/01 - 11/03	3000 LV	2.0	52	46	Obama +6
Reuters/C-SPAN/Zogby	11/01 - 11/03	1201 LV	2.9	54	43	Obama +11
IBD/TIPP	11/01 - 11/03	981 LV	3.2	52	44	Obama +8
FOX News	11/01 - 11/02	971 LV	3.0	50	43	Obama +7
NBC News/Wall St. Jrnl	11/01 - 11/02	1011 LV	3.1	51	43	Obama +8
Gallup	10/31 - 11/02	2472 LV	2.0	55	44	Obama +11
Diageo/Hotline	10/31 - 11/02	887 LV	3.3	50	45	Obama +5
CBS News	10/31 - 11/02	714 LV	--	51	42	Obama +9
ABC News/Wash Post	10/30 - 11/02	2470 LV	2.5	53	44	Obama +9
Ipsos/McClatchy	10/30 - 11/02	760 LV	3.6	53	46	Obama +7
CNN/Opinion Research	10/30 - 11/01	714 LV	3.5	53	46	Obama +7
Pew Research	10/29 - 11/01	2587 LV	2.0	52	46	Obama +6

Easier to read version in the link.

So, 2 Pollsters hit it completely correct, and ALL of them had the final results in their Margin of Error.

Additionally, you are assuming that Rasmussen is just as accurate now as they were during the election last year - a highly dubious claim, considering that their approval numbers have been consistently and starkly different from the results of other pollsters.

This is a (bad) overlay I just made of the average approval rating results for Obama on pollster.com and the results only from Rasmussen:

The red lines are disapproval, the black lines approval. The higher red line and the lower black line are Rasmussen, the other lines is the average. Rasmussen has between 10-15% higher disapproval numbers and about 5% lower approval numbers.

[Gilthan]

The red lines are disapproval, the black lines approval. The higher red line and the lower black line are Rasmussen, the other lines is the average. Rasmussen has between 10-15% higher disapproval numbers and about 5% lower approval numbers.

On your own chart, the typical difference in approval numbers is closer to 3% than 5%.

Anyway, though, notice that the approval numbers are similar (relatively, considering that multiple percentage points difference should be expected from polling likely voters versus often polling all adults). The main difference is that Rasmussen has more people who don't approve flat-out saying they disapprove, in contrast to some polls who have more "undecided."

Of course people are more frank and more likely to admit disapproval of a prestigious high-ranking figure like the President when talking to Rasmussen's automated polling methodology, more so than the typical alternate pollster using human operators. Such is a little like the difference between how honest a person will be if asked at a public social event versus expressing opinions purely privately. Yet that's a point in favor of Rasmussen's accuracy.

Also, rather than getting results like a Pew poll at the start of this month, where approval ratings were only 2% different from Rasmussen but instead 13% of those polled hesitated to say whether they approved or disapproved, Rasmussen offered more choices. With the options of "Strongly Approve," "Somewhat Approve," "Strongly Disapprove," and "Somewhat Disapprove", the number who didn't express an opinion one way or another was reduced.

Yet another factor, as mentioned before, is polling likely voters versus the total populace. Your graph has some error comparing apples and oranges, because you don't understand that distinction, expecting polls of the whole populace to give the same result as LV polls. Young adults, for instance, lean a few percent more Democrat than the average population, yet they also vote at a lesser frequency, something the most sophisticated and accurate election-predicting models like that of Rasmussen take into account.

Unsurprisingly, shortly before an election the polls magically converge in a small range, while before that their results are spread around a whole lot more.

Rasmussen is known for less fluctuation than many. For instance, there's a Pew research poll on the page of your link which had Obama 15% ahead of McCain at 1.5 weeks to go to the 2008 election, not even remotely close to accurate, likewise a NY Times poll a little earlier in October 2008 showing Obama 14% ahead of McCain. In contrast, Rasmussen consistently during the final month showed Obama 5% to 7% ahead of McCain. Since it is rather unlikely that a large percentage of voters changed their minds in the final few days, that was probably accurate within a percentage point or two. It well predicted the actual 7% victory.

Part of the difference is in their methodology. Here was an article written back in 2008:

Some pollsters, including many academic and media pollsters, argue that partisan identification is fluid and changes frequently. This approach suggests that whatever partisan mix falls out from the results of a random sample is the “right” answer. In the case of the recent L.A. Times poll, this mix was 39% Democrats and 22% Republicans.

Polls that use this approach tend to produce a more volatile set of results (during Election 2004, one national firm reported results days apart that showed more than a ten-point swing in voter preference).

Others, including most political polling firms and Rasmussen Reports, argue that people rarely change their partisan affiliation (how many people do you know who consider themselves a Republican one day and a Democrat the next?). This approach produces more stable results. During Election 2004, Rasmussen Reports polling data never showed either candidate ahead by more than 3 points in our weekly data. Week-to-week changes were never bigger than a point-and-a-half.

It’s inappropriate to simply use the results from the last election. After all, a lot has changed over the past four years (and most of it has not been good for the Republican Party label). If you assumed that the mix of Republicans and Democrats this year will be the same as it was four years ago, you will end up with results far too favorable for the GOP.

At Rasmussen Reports, we address this issue by measuring changes in partisan identification on a monthly basis. We interview 15,000 people each month by telephone to dramatically reduce the level of statistical noise and get a stable result (see our latest partisan trends update and month-by-month numbers). This approach shows gradual shifts over time in keeping with the general flow of the political environment. During Election 2004, the GOP gained ground slightly as the campaign wore on. During Election 2006, the Democrats peaked at just the right time. Democrats struggled a bit (in relative terms) after taking control of Congress in 2007, but experienced a tremendous bounce during the early portions of Election 2008.

http://www.rasmussenreports.com/public_ ... nt_results

By the way, the Realclearpolitics site you referenced is the same that wrote the following article on Rasmussen's relative (high) accuracy:

http://realclearpolitics.blogs.time.com ... ic_ballot/

[Sriad]

I would merely note that Gilthan's initial link/image was baldly deceptive, as it compared only those who "strongly approved" (a plurality of Democrats and some independents) to those who "strongly disapproved" (every fucking Republican in the country and some independents) and thereby posited that there is a ten percent gap between those who approve and those who don't.

In reality, even THEIR numbers (I lol'ed a little when I went to the site and Sarah Palin's beaming plastic visage greeted me in a paid advertisement) only show that among respondents who expressed an opinion 49% were positive and 51% were negative.

I would dig deeper into their bullshit, but 1: other people are doing that for me, 2: I have places to go this evening, and 3: I don't like digging in bullshit.

[Sriad]

http://www.rasmussenreports.com/public_ ... nt_results

By the way, the Realclearpolitics site you referenced is the same that wrote the following article on Rasmussen's relative (high) accuracy:

http://realclearpolitics.blogs.time.com ... ic_ballot/

It is interesting that the link you posted laments the number of leading-policy questions asked by a few other polling houses, but rather obviously has no number for Rasmussen.

Also interesting is that the Realclear link you're responding to is a table of statistics, the article you responded with an editorial by one of their writers, and the polling information on their front page flatly contradicts Rasmussen.

One of the (many) problems with "likely voter" polling is that it disqualifies a high percentage of young voters who wouldn't have voted in a previous election for reasons that should be obvious, or who use cell phones (a prediction bender noted early in the Democratic Party primaries).

[Gilthan]

I would merely note that Gilthan's initial link/image was baldly deceptive, as it compared only those who "strongly approved" (a plurality of Democrats and some independents) to those who "strongly disapproved" (every fucking Republican in the country and some independents)

There's nothing deceptive about a plainly labeled graph like this:



They also show total approval:

and thereby posited that there is a ten percent gap between those who approve strongly approve and those who don't.

I corrected your quote for you.

It is interesting that the link you posted laments the number of leading-policy questions asked by a few other polling houses, but rather obviously has no number for Rasmussen.

As pointed out in the earlier quote, the article was describing how Rasmussen gets higher accuracy than ABC/WaPo-CBS, with a difference being that good pollsters (Rasmussen) ask the horse race questions toward the beginning of the poll, before results are skewed with many other questions (in contrast to CBS doing so after thirty-odd questions, for example).

Also interesting is that the Realclear link you're responding to is a table of statistics, the article you responded with an editorial by one of their writers, and the polling information on their front page flatly contradicts Rasmussen.

I wonder if you even really read a thread before replying or just skim a few sentences to show your "shoot the messenger" denial reaction and brilliance. The average poll estimates 49.9% approval, and Rasmussen estimates 46%, yet much of that difference is to be expected. As previously discussed, Rasmussen considers varying turnout with the likely voter model instead of inaccurately treating all demographic groups as having equal voter turnout.

They don't simply disqualify young voters or treat them as never voting but rather adjust for relative turnout. (In 2004 and 2008, 47% and 49% respectively of 18-24 year-olds voted, versus 70% of 65+ year-olds, for instance).

Meanwhile, the approval figure for the average poll is raised up by such gems of poll-taking as this New York Times - CBS poll, which gave instead a 56% approval last month but did so by counting 165% as many Democrats as Republicans (how they weight the results at the end, 385 versus 234 after their weighting)!:
http://www.cbsnews.com/htdocs/pdf/poll_ ... ontentBody

There's a reason Rasmussen better predicted the 2004, 2006, and 2008 elections than much of its competition -- several reasons actually but most of which have already been discussed before.

[Qwerty 42]

I don't see why failing approval should come as a surprise: the liberals are dissatisfied with progress on items like Guantanamo and UHC, and conservatives think he's the Antichrist.

[Count Chocula]

Honestly, as a man who didn't vote for Obama and would be in the "strongly disapprove" category if asked, I really don't see how President Obama's approval ratings going down will make a difference in policy initiatives. It's not like a low approval rating will get him impeached, and the House and Senate aren't acting like Obama's popularity decline is driving their legislation. I suppose it might matter to White House spin doctors, but unless I see Obama's popularity decline affecting Congress' decisions I fail to see how any of these polls are more than hot air.