What's the Real Story behind Obama's Poll Numbers? They're all Relative.

It's amazing how many spurious headlines come out about the President's poll numbers. With all the negativity around the job approval ratings one might assume that the President is among the worst politicians around these days.

Media figure heads and writers rarely put these numbers in any meaningful context other than a trend, which could be a month ago, a year ago, since he took office, or whatever works for the story. Most of the time the changes in the president's job approval numbers are not outside of the margin of error. Here's a quick tutorial, if the margin of error is +-4% then showing a 3% decline suggests no change. Also, remember the error is plus or minus both the new number and the old one.

More curiously, many of the attention grabbing headlines are negatively framed. That is, we see headlines when there is a 2-3% drop, but not when there are similar marginal gains.

So, I thought I'd take a moment to present an alternative analysis of the polls. Reading the data is both an art and a science, and shallow headlines can give polling a bad reputation.

Polls are not an inherently bad political phenomenon. When done properly they can be useful tools to enhance democracy. But when pollsters and media producers misinterpret the patterns and meanings within the data it raises serious questions about both their skill sets and their desired narratives. To use a bad example often expressed by gun rights advocates, "polls don't kill politics, the inexperienced users of polls kill politics."

There are three important questions that one can and should ask about the polls we've seen over the past 21 months since the president took office. Each question helps give some comparative context to nature of the political climate for the President.

First, who beats the President in head-to-head match ups for his leadership post? Second, among the nation's leaders--ignore those who have no elective governing responsibilities--who has higher job approval and favorability ratings than the president right now? Third, are there any positive trends that suggest things aren't horribly bad?

First, let's look at early viability ("can he still win?") measures. One indication that the President still has political legs can be seen in this year's head-to-head match-ups with other candidates. Pollingreport.com tracks horse race questions for potential 2012 candidates.

See Pollingreport.com's numbers: http://www.pollingreport.com/2012.htm

Among the various topline reports, one finds that President Obama consistently wins match-ups versus Sarah Palin, Mitt Romney, Newt Gingrich, Mike Huckabee, Jeb Bush, and even a hypothetical "Tea Party Candidate." While some of the margins between Obama and potential candidates have decreased since he took office, there are no horse races where the President loses. In addition, no other Democrat is within 50% points of Obama.

This all suggests that while the public has issues with Obama as leader, they are not convinced there is someone who would do better.

Moving to job approval and favorability, it helps to use more consistent data. Since the AP-Gfk poll does monthly tracking (1,000+ U.S. adults) on the president and other public attitudes let's take a look at some of their numbers.

See the AP-Gfk poll results here: http://www.ap-gfkpoll.com/pdf/AP-GfK%20Poll%20September%20Topline%2009.15.10%20final%20FULL.pdf

According to AP-Gfk's poll results, President Obama's approval rating for September of this year is 49%. The approval rating for "Congress" is 26%; for "Democrats in Congress" it's 38%; and for "Republicans in Congress" it's 31%. It's true the President's numbers have decreased more than either political party's; however, after contentious battles on health care, financial reform, economic legislation, Supreme Court nominees, immigration bills, military strategy, taxes, and the Gulf Spill, the President still wins the job approval comparison.

Some think Bill Clinton was the teflon president, but over less than two years Obama has weathered far more storms, during a much more difficult period in the country's history and still sits pretty well in the eyes of the public.

Turning to favorability, the President's September rating is 57%. Compare this with a 45% favorability rating for Republicans. Sarah Palin's favorability is at 38%, 6% points lower than former President George W. Bush (44%). Democrats hold a 48% favorability rating, and House Speaker Nancy Pelosi's rating is at 33%. And, according to pollingreport.com the favorability trends for Republican figures Romney, Gingrich, and Huckabee also fall short.

Romney's Ratings: http://www.pollingreport.com/r.htm#Romney

Gingrich's Ratings: http://www.pollingreport.com/g.htm#Gingrich

Huckebee's Ratings: http://www.pollingreport.com/h-j.htm#Huckabee

AP-Gfk does not asses Republican Minority Leader John Boehner; but, according to an April 2009 FOX News Poll of 900 registered voters, Boehner is largely unknown (55% never heard of him). Still, the FOX News Polls reports his favorability at 12%. Oddly, a Gallup Poll of 1,033 U.S. adults, conducted a month earler, reported Boehner having a 29% favorability rating, and only 26% had never heard of him.

Overall, the President registers higher job approval and favorability ratings than other national leaders. I'm actually still looking for any prominent national leader--elected by the public--who rates higher than Obama in the same poll. I'm sure it's out there, I just haven't seen it. (hint hint...for media commentators...)

That said, the AP-Gfk poll shows Michelle Obama (68%) and Hillary Clinton (62%) both have higher favorability ratings than the President, suggesting he does not reign as most favorable among the Washington politicos. Then again, neither Michelle Obama nor Hillary Clinton has elective responsibilities.

Keeping with the theme of presidential performance, we can also compare some numbers about the economy. On handling the economy, 42% approve of the President's performance, while Democrat's and Republican's have ratings of 39% and 34% respectively. These are all low numbers but it's all relative.

In terms of blame for the recession, 26% say President Obama shares "a lot" or "quite a bit" of blame, compared George W. Bush's 51%, the Democrat's 40%, and the Republican's 38%. Accurately, the public perceives financial institutions, the federal government's regulators, and people's bad decisions are mostly to blame: "the banks and lenders that made risky loans" (82%), "the people who borrowed money that they could not afford to repay" (67%), and "the federal government for failing to regulate banks closely enough" (67%).

Finally, are there any noteworthy poll results that are positive and might provide some additional context on the state of the union?

Well, since July 2009, the percent of people saying they are either "somewhat happy" or "very happy" has not gone below 75%. And, the percent of people saying they are "very happy" (39%) is now at its highest level since the trend started in the summer of 2009. In fact, it's up 9% points since August of this year. That's meaningful change, right?

In October of 2008, the AP-Gfk poll showed that 17% of the public felt the country was headed in the right direction. By September 2009, the number was at 37%. This month (September 2010), that number is at 40%, 5% points up from the previous month. In fact, there is a significant linear trend upward in the percentage saying "right direction" and a corresponding downward trend in the percentage saying "wrong direction." This too is meaningful change, right?

In summary, a headline could read, "President's Poll Numbers hit all-time Low," or it could read, "President's Poll Numbers Continue to Stay above Other National Leaders." Hearing that a team scored only one touchdown is bad, until you find out that no one else scored at all in any other games. Right now, almost everyone is doing poorly in the polls; let's not highlight only the President.

I know many public poll readers might find it mutinous to report positive data from polls, but the fact that there is a consensus that Americans are more happy than unhappy (i.e., subjective well being is good), and that the direction of the contry is improving are worth noting. In other words, the media can play a role in improving things in this country too, just look for the silver linings in the data....too.

I'm not acting as general counsel for the President, I'm just trying to help the public understand that interpreting poll numbers is not always as straightforward as some would have us think.

Everything is relative.

UPDATE: Since CNN's headlines about President Obama's approval ratings being at an all-time low of 42%, the following numbers are have been reported (see below). I was waiting on the headlines reporting a bounce, but alas, no luck. Turns out the 42% might have been more random error than previously thought, OR the President is experiencing a BOUNCE.

Pollster Dates N/Pop Approve Disapprove
Rasmussen 9/25-27/10 1500 LV 50% 49%
Gallup 9/25-27/10 1500 A 45% 49%
Zogby (Internet) 9/24-27/10 2068 LV 45% 54%
Rasmussen 9/22-24/10 1500 LV 45% 54%
Gallup 9/22-24/10 1500 A 44% 48%
CNN 9/21-23/10 1010 A 42% 54%

 

Follow David C. Wilson on Twitter: www.twitter.com/dcwilsonphd

Comments

[Wolf LeMay]

10:55 PM on 10/02/2010

It is amazing how some talk out of one side of there mouth and then start talking out the other. 1st I would say yes President Obama is low in the polls along with everyone else but lets look at that.
At this time during the President Bush's rain his numbers were highers even though the talking heads were saying he is doing everything wrong. But now the talking heads are saying President Obama is doing every thing right but the polls are low along with everyone else. Could it be that people are just tried of government the way it is? The Congressmen and Senators that have are and will be in office are still following the same old path so the pollsters are asking the same old things. How about we do this lets have a poll on how tried people are of the some old Government. Have every Poll get the same question to ask across the board no twist no promoting one side or the other and see what happens. But you know that is not going to happen cause Pollster, Rasmussen, Gallup, Zogby, CNN, FOX, and on and on are going to put their own spin to make us believe what they want. All I can say is this David C. Wilson is right in a lot of ways but more of the point is being missed. that's why the polls go lower and lower.

[Sam Arnold]

01:15 PM on 09/29/2010

this article needs way more exposure. Good for Huffington Post for finally posting some sober political analysis. There's a big question that's been missing in the Obama reporting on this site, and most other media: "how could any president do any better under the same constraints?"

Obama's has been simultaneously the most crippled, picked-apart, and astoundingly successful presidencies since Truman.

The problem is that the biggest constraint is the "party of no" and their recipe for failure. And the mainstream reporting seems custom-designed to help them succeed.

[David Wilson phd]

12:55 PM on 09/30/2010

Agreed. It's true the president is "in charge," but clearly "CHANGE" cannot happen if the rules don't allow for it. It's one thing if people don't like Obama, it's another to criticize him because of his policies, most of which have yet to have an impact because they operate on a fiscal calendar..

[suzc]

10:50 AM on 09/28/2010

Very useful explanation.
Thank you.

[Chris1962]

09:49 AM on 09/28/2010

>>>I'm not acting as general counsel for the President, I'm just trying to help the public understand that interpreting poll numbers is not always as straightforward as some would have us think.>>>

Yes, poll numbers are easy to spin, like this entire article demonstrates. For instance, comparing a seated president's numbers with people who neither hold office nor have declared candidacy nor have anywhere near the public focus upon them as Obama's, means absolutely nothing. Likewise, comparing a president with a congressional body is apples and elephants.

You also compare favorabilities by pointing to the "favorable" number alone. This is inherently deceptive, since many times, even if a "favorable" appears low, it can be because the "unsure" number is high.
Example: Romney. While citing Obama’s September personal  favorability as 57%, you provide a link to Romney’s July favorability, which is 36%. Romney’s “unfavorable” rating, however, is only 28%, while 36% are either unsure or have never heard of him. (This trick is often used when citing Palin’s popularity.)
 
Another entirely misleading point you make is that “AP-Gfk does not assess Republican Minority Leader John Boehner; but, according to an April 2009 [sic] FOX News Poll of 900 registered voters, Boehner is largely unknown (55% never heard of him).” You fail (deliberately?) to take into consideration that Boehner was not Obama’s target of blame and suntan-ridicule back in April. Obama was, instead, still using Bush to blame everything on.

Then you say “Still, the FOX News Polls reports his favorability at 12%,” while neglecting to mention his unfavorability, which was only 18%, with 15% unsure. And while, true, the “AP-Gfk does not assess Republican Minority Leader John Boehner,” other polls do. There’s a Likely Voters poll out from a few days ago, showing Boehner’s favorability at 31%, which is a great deal more than 12%, isn’t it?
 
You compare Hillary’s favorability to Michelle instead of taking into consideration that Hillary may challenge Obama in 2012.  And you also ignore a current Rasmussen Likely Voters poll citing Hillary as more mainstream than Obama (51%-48%) and yet another, from three months ago, that says "Rasmussen Reports national telephone survey shows that 57% of voters feel Clinton is qualified to be president, but 34% disagree and say she is not. As for President Obama, 51% say he is fit for the job. However, 44% say he is not qualified to be president, even though he has now served 17 months in the job.”
 
That “44%” figure is supported by his JOB approvals of late, plus a Likely Voters Battleground poll that came out yesterday, saying only 38% would vote for him in 2012. That’s pretty astounding, yet you ignore it all while neatly comparing potential Republican candidates to his 57% PERSONAL favorability rating instead of his JOB approval, which is in the mid-40's.  

[David Wilson phd]

08:07 PM on 09/28/2010

Not sure what the point is here. Are you making a case for Boehner's 31% approval rating? Are you really trying to say there is a candidate who could beat Obama right now? Are you serious? Citing Obama numbers isn't the important part, it's that it's all relative.

[bytheC]

08:34 AM on 09/28/2010

As someone who does not know their polls, Mr. Wilson's explanation was simple to understand. I have quit paying attention to the media hype long ago as they just aren't reliable, truthful or accurate!

[isis]

11:56 PM on 09/27/2010

This analysis seems to go with the feedback you get just as a normal person talking to others. Obama is OK. The democrats could get their act together and the republicans are kind of scary.

[justalurker]

08:36 PM on 09/27/2010

Please, please, Huffpost, PLEASE put this on the Front page or on the Politics page to counter the pervasive negative spin that Obama has bad ratings. The truth is his are the BEST AROUND!

[TylerDurdenSays]

08:33 PM on 09/27/2010

Public opinion polls are misleading, they become the focus of the news—rather than the issues--, and, polls suggest that what’s popular is what ought to be.

Take this most common example. “Do you approve of the job the President is doing?” What does this actually measure? If the number is low, does that mean a progressive President is being “too progressive”? Absolutely not. But, on the face of it, that’s what it suggests. One could easily disapprove of the President because he’s not, in their estimation, being progressive enough. Yet, we tend to assume that disapproval ratings mean that a Democrat is being “too far left” if his numbers are low.

Likewise, I’ve seen polls that ask, “Do you agree with health care reform?” If I answer “No,” then what am I really saying? The poll does not measure what we might think. I could disagree for any number of reasons, most of which give a false impression. I know dozens of people who would answer “no” to that question because they are in favor of a single payer system. I know other people who want no reform at all. So, oddly, all of these groups would be shown to agree (about something)—even though they are on opposite sides of the political spectrum—if they answered this poll. Yet, the headline in a newspaper or blog would read something like, “58% say they disagree with health care reform.” See the problem with that?

[TylerDurdenSays]

08:36 PM on 09/27/2010

Part II:

Second, the media treat polls as if they are newsworthy. In fact, they report more on polls than on virtually anything else. This is called “horserace journalism.” Rather than debating the merits of any particular idea, every day, we have a different winner. It’s silly, of course, but the media need something to talk about on slow days, right?

Moreso, news anchors and commentators treat polling data as if it is fact. They are opinion, and, as I’ve already shown, they are not even accurate gauges of public opinion. The way the media treat polls as fact is by constantly citing them as “evidence” of what to do or what not to do, or as some kind of proof that something will or will not happen.

Finally, popular support does not mean it’s the right—or the best—thing to do. Using public opinion as a reason to do or not to do something is a weak form of reasoning known as “argumentum ad populum.” It’s the classic scenario where your mother asks you if you’d jump off the bridge if Johnny did it. You know the rest. Our founding fathers knew that might does not make right, which is why there are so many checks and balances to negate potential majority rulings. Merely because—allegedly—a narrow majority of United States citizens “oppose” health care reform does not mean we should do nothing.

[gsmp]

08:40 PM on 09/27/2010

Faved both parts and already a fan, Tyler!

[AaronTX]

01:32 AM on 09/30/2010

The media is reflecting the public's desire to do more and more navel gazing. I'm also annoyed that opinion is so often portrayed as news.

[TylerDurdenSays]

08:21 PM on 09/27/2010

I know this is technically an "opinion" piece, but it's needs to get promoted so more than 10 comments are here!

Thanks for an ANALYSIS of all the polling data. Even though I have little love for polls, I agree with your overall arguments.

[Harbinger08]

07:42 PM on 09/27/2010

I've been saying the same thing. Irritating that the news media keeps pushing the "President Obama" is unpopular meme. Wonder why?

[justalurker]

07:03 PM on 09/27/2010

Thank you! Thank you! THANK YOU!!!

For six months I have been noting (see sample repost below) exactly what your article shows: Obama is still the most popular national politician.

The question is: How do we make the corporate news media tell the WHOLE TRUTH about Obama's ratings and not their negative-spin HALF TRUTH ? Any suggestions, David Wilson?

REPOST FROM JULY:
The REAL story is that Obama is BY FAR the most popular national politician in the news right now. [In July] I compiled the numbers side-by-side below to refute the false "narrative" the corporate "news" media would have you believe.

BOTTOMLINE is that Obama is running 14-33 points higher in favorable ratings than other national leaders and he is running 14-26 points higher than likely 2012 Republican opponents.

THAT IS THE REAL NARRATIVE ON OBAMA'S RATINGS!!!!!!

Here's the data:

FAVORABLE RATINGS -- OBAMA
47.4% Obama -- July 2010 RealClearPolitics
48% Obama -- June 2010 Public Policy Polling

FAVORABLE RATINGS -- OTHER NATIONAL LEADERS
34% McConnell -- June 2010 PollTracker KY (state)
15% McConnell -- April 2010 Public Policy Polling
15% Boehner -- April 2010 Public Policy Polling
27% Pelosi -- June 2010 Pew Poll
33% Reid -- June 2010 Las Vegas Review Journal Poll (state)

FAVORABLE RATINGS -- 2012 CONTENDERS
29% Palin -- July 2010 NBC WSJ
34% Palin -- June 2010 Public Policy Polling
28% Gingrich -- June 2010 Public Policy Polling
22% Ron Paul -- June 2010 Public Policy Polling
31% Romney -- June 2010 Public

[Chris1962]

10:14 AM on 09/28/2010

Try comparing the "unfavorables," instead, which is where the rubber meets the road. And do it with Likely Voters instead of "registered voters," half of whom don't vote. You get a very different, and much more accurate, picture of where Obama stands if the election were today.

And since polls are fluid and offer only a snapshot of the moment, check the latest Likely Voters Battleground snapshot from Politico/George Washington University, conducted by Republican pollster Ed Goeas of The Tarrance Group and Democratic pollster Celinda Lake of Lake Research Partners. It shows only 38% would vote for Obama in 2012, with 44% saying they will vote to oust him, which is consistent with his JOB approval among Likely Voters.

Don't confuse Obama's personal favorable ratings with his job approval. Lots of people personally like Obama as a man, but that doesn't translate into voting for him.

[justalurker]

10:44 AM on 09/28/2010

Yawn. Already spent hours on this discussion in another thread.

I'm not convincing you and you are not convincing me. Best to agree to disagree.

Enjoy your worldview. I'll do likewise.

[David Wilson phd]

08:03 PM on 09/28/2010

Job approval is correlated to party ID more than anything else (this is well known...see Bush 2004). Also, we know from research that favorability ratings are more predictive of votes than job approval...so higher favorability/likeability means that one can always "convince." Voters will have 2 years to see how the parties govern before 2012. Honestly, we all know that none of the horse races today have anything to do with 2012 (just ask Rudy and Hillary). In the end, the only tested candidate for 2012 is Obama; that actually makes his run easier not harder.

[Daveg000]

11:50 AM on 09/27/2010

It's all relative. there are many things I'm not happy about, there are many things that I am.

His problem is that he is always too ready to compromise.

It doesn't work when your competition is completely ready and willing: to lie, ignore the facts, to stick together 100% and obstruct when they are completely wrong, even if it is to the detriment of the country, to engender hate and fear.

I am not completely happy with Obama, but he is not the problem.

He did not create the problems that we have now. They have been in the making for 40 years, as the Republicans (28 years) and Bill Clinton (8 years), (forget Jimmy Carter), sent our jobs out of this country, let foreign subsidized products be dumped on the economy destroying industries, and deregulated in favor of greed instead of stability.

The middle class has been under attack for 40 years, as the country went back to 1928 tax rate levels for the rich, and deregulated all the protections, and outsourced jobs and industries. SEPERATION of wealth is back to the 1928 levels.

The middle class is the backbone of this country.

If you think that Republicans, who are fighting for more deregulation, more tax breaks for the rich, Demonizing WORKERS (the other word for unions, Unions that are the only way for the common worker to fight Greedy CEO's and deregulation) and engendering hate and fear are the way to go - vote for them.

[Daveg000]

12:13 PM on 09/27/2010

The Robert's Court has said that Corporations are people, can Spend all the money they want to Buy elections. All that voted for that were Republican activist Conservatives.

I am a retired working man, with normal children and Grandchildren, I see that the Republican Policies have hurt them more than the Democrats could ever have.

I have heard nothing new from the Republicans since OBAMA took office, fighting against every policy, some that were their own.

NOBODY will fix this in a decade, the Republican stance, protect the rich and lie to, manipulate the middle class(for the short time it will last) will NEVER FIX THIS.

If you show me sombody on the Republican side that supports working people - the middle class, and can drag everybody else in the Republican party into supporting the Middle class, I will change my mind.

What's Good for the middle class is good for the country.

I see what Republican policies have done to the middle class, and I won't forget it in 23 months. I have watched it for 40 years.

I agree with the premise of the article, everything is relative, Obama and the Democrats have passed good legislation for the middle class. Stopped the economic freefall.

But in doing so, they didn't go far enough, tried to placate the Very Republicans that put us in this position.

Many seem to believe the party of no, think that putting more obstructionalists in office, to fiddle while Rome is burning is the

[kenhamlett]

10:38 AM on 09/27/2010

I am a Democrat who did not vote for President Obama. I am also a gay man, and I will never again vote for any candidate (obviously, I did not vote for John McCain, either) who does not believe that I deserve the same rights that they have in this country. That said, virtually all of my Democratic friends did vote for him -- the straight ones and the gay ones and the white ones and the black ones and the Muslim ones and the Christian ones and the Jewish ones. The real problem President Obama faces is that of all the people I know who voted for him, I do not know a single one who does not consider him a disappointment. He has broken so many promises (to gays, about closing Guantanamo, about the wars, about supporting universal health care, about standing up to China, etc.) and performed so poorly or indifferently (the Gulf oil crisis, the economy, etc.) that these people -- some of whom also campaigned for him and contributed money -- all feel a strong sense of disillusionment. The President could actually win my vote, if he decided to show some courage and assert his support for equal rights. But, many of my friends tell me that they "will never vote for him again." It is this loss of support about which the President and Democratic leaders need to be concerned.

[David Wilson phd]

11:22 AM on 09/27/2010

I think the point of my piece above is that disappointment is 1) relative, and 2) deserves to be spread around a bit more by the media (and those observing politics in general). In psychology we study something called the fundamental attribution error; it's when individuals make judgments about people over-emphasizing WHO they are (gay, Jewish, Christian, etc.), and under-emphasizing the situation they're in (e.g., recession, 2 wars, Gulf Spills, election season). The situation in America is undoubtedly [economically] bad, but it's not where it was.... and despite people's desire to have the President "be" change, right now, it just happens slowly.

[TylerDurdenSays]

08:24 PM on 09/27/2010

And do you all plan to vote for, who, Palin instead???

[emma richmond]

02:30 AM on 09/27/2010

The MEDIA have gotten to be nothing but slugs they lie and Pander for the Republicans and Tea Baggers, we know the MEDIA have been lying for a long time, CNN and MSNBC, and NBC and ABC, have made it a Habit of Misinterpret the President Poll, we are a group of people you can't fool and not Ignorance, when a Poller called us last week we let it all poured out about how sorry and Corrupt the MEDIA have gotten, they lie and lazy and don't try to find out the Real facts, half of these Pundits are not qualified as Pundits most of the time they don't even know what they are talking about and these so called Journalist, are so bad, they get these Characters on they show they let the wing nuts they hear one sided NEWS and they don't Challenge, sure they might ask one question then they get a lie they say OK thank you, most of them are Cowards, to the Republicans or the Right Wings, but with the Democrats they such nasty attitudes, that's the time to change channel to our cooking show or movie, it's getting sad when you can't get honest and fair NEWS with out the Hate for the President, they hate the President so much they rather support Ignorance and Stupidity, they pander to the Republicans and Palin and her Daughter who have never done one thing for this Country but fill they Pockets yet they are Praised.

[emma richmond]

02:13 AM on 09/27/2010

David C. Wilson we have been saying this for a long time, that the MEDIA have been Misinterpret the Poll Numbers and Tampering with the President and the Democrats numbers and it's on Purpose, what really get us it's every other day polling, by CNN and MSNBC and NBC and ABC, it's strange they can't pass his skin Color a Black can Come in and get so much accomplished then an President in History, they might can Misinterpret his Poll numbers but the world knows what he have accomplished and this is not the 1950'S where the White Man Steal the Black man accomplishment, just as the Republicans trying to run on his accomplishment and trying to take Credit for saving the American Auto and Millions of Jobs in which the Republicans said let them failed, the Auto Workers was making to much money, when at the sametime they have a foreigning Car Company in they back yard where they give them incentives and Big tax's Break for haveing they company in this country, they would have rather the American Auto MFG., Fail, just last week Corker Republicans of TN tried to take Credit, we're so happy the people didn't for get they told him to get out, what get us they are so use to taking from people of Color for years and think they can still do that, they are so wrong. they voted against this President for everything, But GOD Stepped in. yes they are Misinterpret.