February 27, 2007

The Political Cost of Iraq

The numbers from the latest Washington Post poll pretty much speak for themselves:

5. Who do you trust to do a better job handling (ITEM), (Bush) or (the Democrats in Congress)?
The situation in Iraq: Bush 34, Dems in Congress 54
The U.S. campaign against terrorism: Bush 39, Dems in Congress 52
The economy: Bush 36, Dems in Congress 56
The federal budget: Bush 32, Dems in Congress 59
Health care: Bush 25, Dems in Congress 62

Given the decided advantage Republicans have enjoyed on national security issues over the last 20 to 30 years, if someone had told you that five and a half years after September 11 the Republican President who shepherded America through the worst terrorist attack in her history would be running 13 points behind Democrats in Congress on the question of who can better handle the war on terror, you probably would have thought that to be very unlikely, if not a bit nuts. But that's where this President appears to be.

The real question is whether (or perhaps how much) the public's dismal view of President Bush is rubbing off on the Republican party in general, GOP members in Congress, and potentially the party's 2008 Presidential hopefuls as well.

February 26, 2007

WaPo Poll

Initial release: Bush job approval 36%, handling of Iraq at 31%, and Democrats hold a 20-point advantage on who the public trusts to do a better job of handling Iraq. As dismal as those numbers are, they all represent a slight improvement for the President over last month's survey.

More results this evening.

September 22, 2006

The NY Times & Congressional Job Approval

Yesterday, the New York Times offered an analysis of their most recent poll -- asserting that Congress' horrible marks are signs of major trouble for Republicans.

There are a whole host of inferential problems with using congressional job approval as an estimate of seat changes in the next election. At the end of its write-up on the job approval number, which is indeed ridiculously low at 25%, the Times backed off the implication, seemingly aware of the trouble. They have good reason to.

Here is a quick list of reasons why congressional job approval is not the first thing I look to in a poll to get a gauge on November. The 4th reason explains why it is not even the last thing.

1. The implication of the poll is that it reflects an anti-incumbency feeling -- and a very great one at that. However, we can rest assured that 90% of incumbents will be reelected in November. Accordingly, the fact that so many people (a) are voicing disapproval of Congress and (b) will eventually vote for their member anyway is a sign that whatever phenomenon it is picking up is not, in itself, necessarily actionable. At the most, we can only conclude that congressional job approval sets up some kind of context for a vote against the incumbent. This begs the question: are the other factors there, and to what extent are they there? That is what really matters.

2. The ecological fallacy (once again) rears its ugly head. This poll is a poll of the nation. But the House vote is not a national contest. It is a series of 435 individual contests. Accordingly, the distribution of sentiment matters a great deal. And a national poll gives us no indication on how these feelings are distributed -- and so does not, in itself, point to anything whatsoever.

3. Relatedly, there is no necessary reason to accept the idea that this will affect Republicans negatively. These numbers are surely being driven, in part, by voters who are disappointed that Congress has not been sufficiently conservative. They would register disapproval, but this disapproval is not actionable on Election Day. At best, the voter can abstain -- but Republican representatives in competitive districts, those who need their conservative base activated, are very effective at making sure the voters distinguish between the institution and them. That's what the "Homestyle" is all about.

4. The first 3 reasons demonstrate the insufficiency of using congressional job approval as a measure of final vote outcomes. I would actually take things a step further and argue that it is entirely unnecessary. I have written about congressional job approval previously -- and my conclusion was that it is probably a spurious cause of congressional election results. The reason for this is not just that it has a horrible track record in predicting final seat swings -- which itself is a sign that, at best, it is spurious. More importantly, it actually does a better job predicting how many seats the party of the President will gain/lose, regardless of whether the party of the President is actually in control. of Congress. This makes no sense, and supports the idea of spuriousness. My theory is that presidential job approval drives both final elections and congressional job approval.

Let me be clear -- I do not doubt that there is some feeling out there in the nation that this statistic is tapping into. And this feeling is ultimately what will cause Republicans to lose a good number of seats. My objection is that the Times, like many outlets, paints congressional job approval as a cause of what will happen in November. When they point to 1994, that is what they are doing. But the data, as well as the logic, indicates that this is likely a spurious cause. At best, all we can say is that congressional job approval, sometimes but not always, somewhat, but not very much, co-moves with final November results.

Generally, this has been something I have noticed in this season's coverage of the Election. Pundits tend to make a laundry list of causes of GOP trouble. I do not doubt that the GOP is in trouble. What I do doubt is the laundry list. I think that many of the things they point to are in fact all effects of something that is going to cause the result in November. In other words, there is some singular cause that (a) induces congressional job approval to tank, (b) induces right track/wrong track to tank, (c) induces "do you want to reelect your incumbent?" to tank, and (d) induces final election results. (a) through (c) do not cause (d). But, if you are only taking a brief look at the data (for instance...oh I don't know...maybe just looking back to one previous incidence, that happens to be the outlier), and are not careful to distinguish between correlation and causation, you'll conclude that (a) through (c) do indeed cause (d).

This is a symptom of a broader problem I have noticed. Journalists and their editors recognize that they are not able to put together externally valid samples to gauge public sentiment. They know that they need pollsters. However, for some reason, they believe that they can interpret the results. In actuality, interpreting polling is as difficult as creating polling. It requires a thorough grounding in the logic of causal inference -- which is something that journalists obviously lack. They seem to completely lack it, too -- if they knew anything about this logic, they woukd know how difficult and tricky it can be, and they would either outsource the task of interpretation, study up, or express some hestitation. But they don't. They interpret polls just as college freshmen interpret Aristotle -- with a gusto and authority that only comes with complete ignorance of the subject.

A professor friend of mine once told me that the problem with political journalism is that journalists are doing political science with an M.A. in English. I think there is a lot of truth to that -- and the misuse of polling data in journalist circles is one thing I would point to.

September 21, 2006

Getting Seasick With Gallup

Republicans in the last few days have found themselves buoyant -- thanks in large measure, it seems, to what about 2,000 respondents had to say to Fox News and Gallup about which party they prefer in November. Suddenly, in the "generic ballot," the GOP is even with the Democrats.

Unfortunately, some of these buoyant Republicans seem to be the same people who took great heart in my column about the perils of the generic ballot. I am not sure how the two positions are reconcilable -- how you can join me in rejecting the generic in August and now embrace it in September. But I would note that I saw myself quoted by many bloggers as saying, "The generic ballot has a Democratic is just not trustworthy!" The "..." eliminated a very important point, which was that the skew is just the begining of what are very serious problems. So, I am not entirely surprised by the positive reception the recent Gallup generic has enjoyed on the right.

Republicans might be buoyant, but personally I'm getting seasick. In the last month, the Gallup generic has gone back and forth. In late August, it had the Democrats up only 2% among registered voters in the generic ballot. Then the Dems sprang ahead to a 12% lead in the course of about 15 days. And then, another 15 days later, the "likely voters" are driving another tie.

Gallup has done this to us before. As I have mentioned, there is a large and sustained Democratic skew to the generic ballot. But the skew is not omnipresent. A few times it offered up outlying results in the summer such that, when you take the average of their summer polls together, you actually get a result that underestimates Democratic strength. I found this to be the case for a few election years when I analyzed the June/July data. 1986 was a major problem for the model because of its underestimation of Democratic strength. Why? In any estimate where you use one variable (generic ballot) to predict another (final House vote), the error has to be distributed in a bell-shaped fashion. When it is not, often the problem is due to an outlying incidence. Your model is predicting one value that is much, much farther from the actual result than any of the other predictions. This was the case with 1986. In this situation, standard operating procedure is to toss the outlying observation (or to fix the data point if there was a calculation/recording error) after you have determined that there is not some kind of hidden centrality to it (e.g. is the exception actually the rule?). Otherwise, you risk throwing off your ability to draw a prediction, which of course is the whole point.

So, from my perspective -- I have seen Gallup go bouncing back and forth like this before. It was a major pain in my you-know-what this summer as I was trying to make sense of the generic ballot. I'm not jumping on this bandwagon. Note that this should not be taken as good news for Democrats or for Republicans. I would ask that liberal bloggers avoid the "..." that I saw on the right last month. I think the generic ballot is so awash with analytical problems that any kind of direct use of it is trouble.

However, if you are comparing the generic ballot to some other final result, you might not encounter this problem. Emory University's Alan Abramowitz uses the generic ballot as a measure of partisan strength to help predict seat changes, as opposed to vote changes, in an upcoming scholarly article in the journal PS. By using a different dependent variable (seats instead of votes), and by using it in the context of other predictive variables, you can eliminate these problems. But comparing the generic ballot to final popular votes, which is what most everybody else uses it for, is inferential trouble.

September 14, 2006

NBC/WSJ Numbers

Here are the vitals from the latest NBC/WSJ poll out this morning and the updated RCP Averages:

Bush Job Approval
NBC/WSJ: 42%
RCP Avg: 42.5%

Generic Congressional Ballot
NBC/WSJ: Dems +9 (R 39, D 48)
RCP Avg: Dems +9.7 (R 41.3, D 51.0)

Congressional Job Approval
NBC/WSJ: 20%
RCP Avg: 28.3%

Direction of Country
NBC/WSJ: 31% Right Direction, 54% Wrong Direction
RCP Avg: 28.0% Right Direction, 65.3% Wrong Direction

August 09, 2006

Polls in Black and White

Democrats try to figure out how to read polls like Republicans ...

The key insight:

When Republicans segment voters into groups, they typically use the "most important issue" questions to do it, whereas Democrats typically use demographics, speculating, for instance, on how married women or African Americans will vote and then tailoring their campaign strategies to the various groups. While demographics offer insights into who is thinking what, it is much more useful to be able to identify the single issue that will determine someone's vote.

Of course, you also have to have appealing positions on those issues. Maybe the Democrats will figure that one out next time around.

August 07, 2006

A Mixed Bag

New ABC News/Washington Post poll out. Here's the data and the write up by Peter Baker and Claudia Deane which strikes me as a bit overdone with a hint of setting up a "1994 anti-incumbent national wave" meme in advance of a Lamont victory today.

Bottom line: Bush's job approval ticked up slightly this month, as did Congress's. Generic ballot number unchanged. RCP Averages updated for all three indicators: Bush JA | Congress JA | Generic Ballot

July 17, 2006

Zogby Worldwide

Zogby has a new Web site up focused on international polling.

July 12, 2006

Global Preening

The Pew Research Center for the People and the Press has a new survey on global-warming attitudes.

Basically, most people (70 percent) believe global warming exists. Forty-one percent believe it's caused by humans, and 21 percent think it's caused by natural patterns.

Democrats are about 20 percentage points more likely to believe global warming exists; they're more than twice as likely to blame humans.

Most interesting, however, is that while Republicans rate global warming the absolute least important out of 19 issues tested by Pew, Democrats don't consider it all that important an issue either. They rate it 13th out of 19 issues -- health care, education, the economy, Iraq, terrorism, the deficit and the minimum wage all beat out what they claim to believe is a cataclysmic climate shift, caused by man, about to wreak havoc on the earth.

What this shows, I think, is that most global-warming politics is truly about moral preening and liberal guilt over living in a prosperous society. That's not to say one can't legitimately be concerned about climate change. But it does mean that most Democratic voters know they're supposed to believe in global warming and blame SUVs if they want to be considered good, socially conscious people.

When push comes to shove, however, they don't actually expect anyone to do anything about global warming. They just want to know their leaders are properly concerned.

June 30, 2006

American Research Group vs. Rudy

I've had a bee in my bonnet about this for a little while, so I figure I'll get it out, since American Research Group has released yet another poll making it appear as if Sen. John McCain were the undisputed frontrunner in the race for the '08 GOP nomination.

ARG's recent Rhode Island poll shows McCain with the support of 50 percent of likely Republican primary voters in the state. Mitt Romney comes in second with 14 percent. Newt Gingrich rounds out the top three with 4 percent. Giuliani's name isn't even on the list of candidates respondents are asked to choose from.

When Rudy Giuliani is added to a second question on the GOP primary in ARG's poll, it's McCain 43 percent, Giuliani 19 percent.

So, what does this mean? It means that any candidate added in a second question like this is likely to register a far lower level of support than if they'd been included as a top-tier candidate in the first question. ARG lists McCain, Romney, Gingrich, George Pataki (George Pataki!?!?!?), George Allen, Sam Brownback, Bill Frist, Chuck Hagel, and Mike (I heart) Huckabee as first-tier candidates, but thinks Rudy Giuliani shouldn't be on the table until the second question?

Assuming that these two questions must at least be randomized to make these legitimate polls (so that the order wouldn't matter), I shot an email over to ARG to ask what was up. This is the response I got from Dick Bennett:

We added Giuliani after the first ballot and did not randomize the order of the questions.

While he has been more active lately, there are still no signs on the ground that he will run. I continue to hear from activists that Giuliani will only get in the race if McCain does not.

Hope this is helpful.

Yes, it is. It lets me know that as far as the race between McCain and Giuliani going into '08, ARG's polls not only can, but must, be ignored.

So, how does the other polling on the McCain-Giuliani contest pan out? Well, the polling from Strategic Vision, for instance, shows Giuliani well ahead in most of the states the firm has polled.

Here are their respective percentages in a few states, as measured by SV:

[Giuliani / McCain]

PA: 39 / 28
WA: 35 / 28
FL: 39 / 28
GA: 27 / 22
N.J.: 45 / 32
WI: 28 / 25
N.Y.: 53 / 13

One exception in the SV polls is Michigan, where McCain leads 39 / 22.

[except for New York, all those samples are of likely voters -- in New York, the sample is of registered voters]

So, lastly, why does all this matter? Well, because in crucial primary states such as Iowa and South Carolina, ARG polls show McCain with a commanding lead. But it's likely this lead is entirely illusory, based more on a poor survey design than a reflection of reality.

As Giuliani's intentions have become more and more obvious -- and are at least on par with Newt Gingrich's as far as seriousness, to say the absolute least -- ARG should correct this immediately.

Below the fold is a look at two states, IA and MA, where ARG polling seems to conflict rather baldly with other public polls.

Continue reading "American Research Group vs. Rudy" »

Bush at 40, Dems at 47

Three new polls have come out in the last two days (FOX News, LA Times/Bloomberg, and Hotline/Diageo/FD), all showing President Bush's job approval rating at 41%. He's now over 40% in the RCP Average for the first time in months.

Democrats still hold more than a 10-point lead in the RCP Average for the Generic Congressional ballot. CNN, FOX, and Hotline all show the Dems' lead well under 10-points, however, while the LA Times, Gallup, ABC News/Wash Post and Pew all have them leading by 12-16 points.

June 28, 2006

Pew: In 2006, Democrats Enjoy 'Distinct Advantages'

Yesterday, Pew released a poll showing the Democrats with "distinct advantages" going into this year's midterm elections.

The bullet points (summarizing their press release):

* Voters continue to say they favor the Democratic candidate in their district, by a 51% to 39% margin.

* The level of enthusiasm about voting among Democrats is unusually high, and is atypically low among Republicans. In fact, Democrats now hold a voter enthusiasm advantage that is the mirror image of the GOP's edge in voter zeal leading up to the 1994 midterm election. [46% of Democratic voters say they are more enthusiastic about voting than usual, compared with just 30% of Republicans. In October 1994, Republicans held a comparable advantage on this measure (by 45%-30%)]

* Public anger with Congress continues to rise, and anti-incumbent sentiment has reached new highs.

* Increased Democratic intensity is mostly driven by anger toward President Bush and Republican leaders, not by support for the party and its leaders. Fully 64% of Democrats say their party is doing only a fair or poor job in standing up for its traditional positions on such things as protecting the interests of minorities and helping the poor.

Read the whole report here.

June 26, 2006

The West Is Not "the West," Part I

Here's my two-part plea to pollsters: A) When possible, give regional breakdowns of national polls, B) and when you give these breakdowns, don't pretend there's a political entity called the "West."

While the Northeast, the South, and the Midwest are all relatively coherent political beings, which can be discussed as regions despite the obvious fact that there is infinite texture to any measure of public opinion, there is no such thing as the West. There is the Pacific Coast (California, Oregon, and Washington), and there are the eight states of the interior West (Arizona, Colorado, Idaho, Montana, Nevada, New Mexico, Utah, and Wyoming). There is no West.

To take one current example: the recent Pew analysis of President Bush's approval ratings. Now, I love the folks over at Pew, and they're always extraordinarily accommodating about breaking out their data in different ways. But let's just look at Bush's approval rating percentages, broken down by four geographic regions:

Dec. 2004: 39
May 2006: 27
Change: -12

Dec. 2004: 56
May 2006: 36
Change: -20

Dec. 2004: 48
May 2006: 32
Change: -16

Dec. 2004: 46
May 2006: 33
Change: -13

It looks, with a four-region breakdown, as if Bush is strongest in the South (though he's taken the biggest fall there) and weakest in the Northeast.

Now let's look at the West broken down between the coast and the interior (a breakdown provided by the kind folks over at Pew):

Interior West
Dec. 2004: 56
May 2006: 48
Change: -8

Pacific Coast
Dec. 2004: 41
May 2006: 24
Change: -17

Suddenly, we've got a new high and a new low. Bush is doing worse on the Pacific Coast than in the Northeast (24 vs. 27 -- though, granted, that might not be statistically significant) and he's doing far better in the interior West than in the South (a rather stunning 48 vs. 36). What's more, the interior West has seen the least movement in Bush's poll numbers between December of 2004 and May of 2006, a testament to just how solid his support is in the region.

What does it mean? Well, that's a question for another post. For now, my point is simply this: Any regional breakdown labeled "West" is simply an average of the Pacific Coast and the interior West, which are quite separate in their politics. Such a designation is utterly useless for understanding America's political geography.

And given that the interior West is beginning to be seen as a strategically crucial region for both parties (not just in my book, but also in an intriguing new book coming out later this year from Democratic political science professor Tom Schaller), it's time to start looking at it as a distinct entity.

June 13, 2006

Cheating Bush Out of 6.8%

ABC News stretches the boundary of the word "approximately" in this story on Newt Gingrich:

With mid-term elections coming in November and Bush's approval ratings at approximately 30 percent...

Actually, Bush's most recent RCP average job approval rating is 36.8%. I suppose the authors of the ABC News story could be referring to the last ABC News/Washington Post poll - taken nearly a month ago, by the way - which had Bush at 33%. Theoretically, they could also be referring to the new CBS News poll (also 33% job approval) that came out last night.

But every other poll taken in the last month- eight of them, in fact - has Bush at 35% or higher, including a new USA Today/Gallup poll that has the President at 38%. So I suppose, given the new standard offered by ABC News, we could also characterize Bush's approval ratings as being at "approximately" 40%, though I suspect it's highly unlikely you'll find anyone in the MSM rounding up in favor of the President.

Bush Bounce: Statistical Noise or a Modest Uptick? - by Scott Rasmussen

President Bush failed to get an immediate political bounce from the news that al-Zarqawi was killed last week. The reasons why cannot be determined with assurance, but let's first look at the numbers.

Just before al-Zarqawi was killed, 32% of Americans gave the President good or excellent marks for handling the situation in Iraq. That figure actually slipped a point to 31% in our poll conducted in the days following the good news.

At Rasmussen Reports, we also measure consumer and investor confidence on a daily basis. Following the capture of Saddam Hussein, there was an immediate bounce in the nation's economic confidence. By contrast, economic confidence in the U.S. actually fell slightly in the days following the al-Zarqawi news (it's now come back, but no surge is evident).

Same thing on the Bush Job Approval ratings... Initially, there was absolutely no bounce. Today's reading is 42%. That's up a bit from 40% and we will have to watch to see if that's statistical noise or a modest uptick. At the moment, the default assumption is statistical noise (the President's daily Job Approval ratings have been remarkably stable lately--within three points of the 40% level on 58 out of the 60 days).

A couple of other polls have been released suggesting that the President may have enjoyed a bounce from the latest news. But, the news stories assume more than the numbers can justify. Those polls are quite likely a reliable measure of current views, but they are comparing current readings to results from a month or more ago. Many factors (including statistical noise) could account for the slight improvement they have found.

The more intriguing question is why there was no bounce for the President. One possibility is that there have been so many potential "turning points" in Iraq that the public has adopted a wait and see attitude. Rather than celebrating a turning point, Americans may be waiting for proof in the form of decreased violence and reduced U.S. military involvement in Iraq.

Another possibility, suggested by a wealth of polling data, is that Iraq and the War on Terror are no longer the dominant voting issues. For the first time since 9/11, we will have an election decided on issues closer to home. Immigration, the economy, and other domestic topics may ultimately decide the critical election contests this November.

May 11, 2006

Get Your Panic Here

Bush, GOP Congress Losing Core Supporters - Jim VandeHei and Peter Baker, Washington Post

Polls show Bush's base starting to flee - Dick Polman, Philadelphia Inquirer

Bush's low ratings worry Republicans - Linda Feldmann, Christian Science Monitor

Polls show Bush losing conservative support - John Whitesides, Political Correspondent

May 04, 2006

Support For Roe

In today's Wall Street Journal: "U.S. support for Roe v. Wade is at its lowest level in decades, according to a new Harris poll."

April 26, 2006

Quote of the Day

"Nothing kills a political honeymoon faster than the word taxes." - Clay F. Richards, assistant director of the Quinnipiac University Polling Institute, explaining why New Jersey Governor Jon Corzine's negatives shot up to 42% in the lastest Qpoll after registering at only 14% in January.

April 20, 2006

Bush Approval Starts to Near Danger Level

The new FOX News/Opinion Dynamics poll has President Bush's Approval down to 33%. This is 3 points lower than any previous poll FOX News has released, and needless to say it is not good news for the White House. Among Republicans his approval has dropped to 66% and therein lies his major problem and the answer to how he can bounce back.

The President has got to get his Republican base back in the fold. If he can do that his approval ratings will stabilize in the high 30's and low 40's and he will be in a position to mount a comeback this fall. Bush could take sliding approval ratings into the high 30's in the RCP Average, but deterioration below 35% - it's not there yet, but the FOX poll could be a troubling leading indicator for the White House - starts to indicate loss of core support in his Republican base and that can become a very slippery slope which could effectively kill his presidency. He's got to pick a fight with the Democrats on something and get his base reenergized. Compromising with Ted Kennedy on immigration is not the answer to his political problems.

April 07, 2006

GOP: "Scary" New Poll

GOP pollster Tony Fabrizio sums up the results of the new AP-Ipsos survey: "These numbers are scary. We've lost every advantage we've ever had. The good news is Democrats don't have much of a plan. The bad news is they may not need one."

Of particular interest in the AP poll is question number six, which mirrors something we saw a while back from Rasmussen:

6. Who do you trust to do a better job of protecting the country?
Democrats ........................................ 41
Republicans...................................... 41
Both equally (VOL) ........................... 4
Neither (VOL) ................................... 11

Here's the skinny on the rest of the results, along with the updated RCP averages for each data set:

Bush Job Approval: 36 approve, 62 disapprove
RCP Avg: 38 approve, 57 disapprove

Congress Job Approval: 30 approve, 67 disapprove
RCP Avg: 32 approve, 60.3 disapprove

Generic Congressional Vote: Democrats +16 (R 33, D49)
RCP Avg: Democrats +11.7 (R 38, D 49.7)

Direction of Country: Right Track 28, Wrong Track 69
RCP Avg: Right Track 30.5, Wrong Track 64.0

As always, the composition of respondents in the AP-Ipsos (based on self-identification) poll is worth noting: Republicans 26, Democrats 31, Independents 34. Sample: 1,003 adults, 844 registered voters, interviews conducted between April 3-5.

March 23, 2006

Polling Republicans

There's a new Diageo/Hotline poll out looking at the attitudes and preferences of 602 registered Republicans across the nation. The data on 2008 is interesting, though I would echo Jay Cost's recent warning that '08 horserace/preference questions at this point are almost totally useless and represent nothing more than reflexive name recognition on the part of respondents.

That being said, however, we probably can glean a bit of value from questions asking for favorable/unfavorable impressions. With that in mind, take a look at these numbers:

President Bush
Strongly favorable 54%
Somewhat favorable 30%
(Net favorable rating 84%)
Somewhat unfavorable 8%
Strongly unfavorable 6%
(Net unfavorable rating 14%)

Vice President Cheney
Strongly favorable 39%
Somewhat favorable 37%
(Net favorable rating 76%)
Somewhat unfavorable 7%
Strongly unfavorable 9%
(Net unfavorable rating 16%)

Senator John McCain
Strongly favorable 23%
Somewhat favorable 39%
(Net favorable rating 62%)
Somewhat unfavorable 12%
Strongly unfavorable 8%
(Net unfavorable rating 20%)

As you can see, McCain's unfavorable ratings (particularly the strongly unfavorable) aren't significantly different from Bush or Cheney, but his favorable ratings (particularly the strongly favorable) are well below the President and Vice President, suggesting, once again, that McCain's support with the base is positive but still remains on the tepid/shallow side.

March 21, 2006

Approval Ratings For All 50 Governors

SurveyUSA's March rankings are out: North Dakota's John Hoeven tops the list with a plus 59 net rating (76% approve, 17% dispprove), and Ohio's Bob Taft remains at the bottom for the eleventh consecutive month with a minus 60 net rating (17% approve, 77% disapprove).

Here's a run down of Governors up for reelction this November with approval ratings currently below 50% mark:

Ehrlich (R-MD): plus 4 net rating (48 approve - 44 disapprove)
Rendell (D-PA): plus 3 net rating (49-46)
Perry (R-TX): plus 1 net rating (48-47)
Doyle (D-WI): minus 4 net rating (44-48)
Granholm (D-MI): minus 7 net rating (45-52)
Blagojevich (D-IL): minus 8 net rating (44-52)
Minner (D-DE): minus 11 net rating (41-52)
Baldacci (D-ME): minus 16 net rating (39-55)
Kulongoski (D-OR): minus 20 net rating (36-56)
Schwarzenegger (R-CA): minus 25 net rating (36-61)
Murkowski (R-AK): minus 36 net rating (29-65)

March 15, 2006

Bush Approval Sinks Further

Two new polls show Bush 's approval slipping even further. Pew Research has Bush's job rating at 33% this month, a seven-point decline vs. February.

SurveyUSA's new 50-state tracking poll also shows Bush's approval declining. According to SUSA, the President's job approval remains over 50% in only three states (Utah, Wyoming, and Alabama) and his nationwide approval (weighted for population) is down to 36% approve, 60% disapprove. The negative 24-point spread between approval and disapproval is the highest ever recorded by SUSA in eleven months of tracking Bush's approval.

Overall, Bush's approval in the RCP Average dropped to 36.4%. I'll have to go back and check to be absolutely sure, but I believe that's a new low for the President as well.

March 10, 2006

AP Poll: Bush at 37%

New AP-Ipsos poll out showing Bush's job approval at 37% (news story, full results - pdf) That number represents a 3-point drop from last month and puts Bush back to the same level he was in early November after Katrina and the Harriet Miers nomination. Bush's overall job approval in the RCP Average dropped two-tenths of one percent to 39.6%.

Other numbers from the AP poll:

Congression Job Approval: Down 5 points to 31%. (RCP Average now at 32.4%)

Direction of Country: Right track down 5 points to 30% (RCP Average now at 31.4%)

Generic Congressional Vote: Unchanged from last month: Republicans 36%, Democrats 47% (RCP Average is Democrats +11.2%)

March 02, 2006

Two Views of the Battleground Poll

The new bipartisan Battleground Poll is out, and while it was taken too early (Feb 12-15) to give us any new information on the political ramifications of the Dubai Ports World deal, it does contain some interesting findings. Here are the topline numbers: Bush job approval 46%, Congressional Job Approval 37%, Generic Congressional Ballot Dems +5 (46% - 41%).

Here is the analysis from the point of view of Republican pollster Ed Goeas:

The bottom line is that the mood of the electorate is not an anti-incumbent mood, an anti- Democratic or anti-Republican mood, but an anti-Washington mood. When asked if voters wanted a member of Congress to possess “strength of values and convictions,” or “willingness to find practical, workable solutions” to the country’s problems, voters wanted to see practical, workable solutions by a twenty-point margin (58% to 38%). That is not, however, what the American electorate feels they are getting from Washington. When these same voters were asked if they felt lawmakers in Washington “put you first,” or “put partisan politics first,” only four percent (4%) felt that lawmakers put them first, and a whopping ninety-two percent (92%) felt lawmakers put partisan politics first. In other words, Washington is broken and needs to be fixed.

One only has to look at the image ratings of the leaders of both parties to see the effect of this cynical view held by voters. The image of both President George W. Bush (45% favorable/53% unfavorable) and Vice President Dick Cheney (42% favorable/51% unfavorable) are net negative. But equally negative are the image ratings of Democratic leaders like Democratic frontrunner for the 2008 Democratic presidential nomination, Senator Hillary Clinton (45% favorable/51% unfavorable) and Democratic National Chairman Howard Dean (31% favorable/45% unfavorable). Of all the Washington leaders, only Senator John McCain (65% favorable/18% unfavorable) has chiseled out a positive “bi-partisan” image with the American electorate.

Celinda Lake analyzes the number from the Democratic point of view:

As dissatisfaction with the President and Republican-led Congress grows, voters are poised to affect sweeping change. The “sixth year itch”, where voters turn out the party in power in large numbers, is by no means a new phenomenon; only Bill Clinton, in the history of modern campaigns, avoided his party losing seats in the mid-term elections of his second term. Today, Republicans’ failures, especially when they control all the branches of government, suggest the potential for a historical trend will continue in 2006. Importantly, independent and undecided voters are noticeably negative in their assessments of Bush and the direction of the country. [snip]

One of the most important fights in the 2006 elections will be for control of the agenda. Republicans clearly want to centralize this election around security because of their continued advantage on the issue though recent events have tarnished that. In order to make real gains Democrats must at least neutralize the issue. Democrats should put resources into establishing their credentials on security, while emphasizing Republican failures. To this end, the issue over port security should be the start of a frank dialogue on national and homeland security policy, not an isolated news story. This formula – capitalizing on the current political atmosphere, presenting a real alternative to the status quo of costly corruption, and neutralizing the security issue while harnessing their strengths on domestic issues – has the potential to give the Democrats big gains.

Two quick observations. First, the numbers on McCain are really interesting and deserve more discusssion, something I hope to tackle early next week. Second, the operative phrase from Lake's analysis is that Democrats have to "neutralize the issue" of national security. Democrats aren't going to outpoll Republicans on national security (despite that snap Rasmussen poll on the Dubai Ports World deal from last week), but they do need to find a way to close the gap with the GOP on the issue - and the current port controversy offers precisely that kind of opportunity.

February 27, 2006

Bush Approval Slammed By DPW Deal

Rasmussen Reports released a poll last Friday showing Democrats in Congress favored over President Bush on national security 43% - 41%. If that wasn’t enough of a signal that the Dubai Ports World deal was a political loser for the President, the latest polls from CBS News and Cook/RT Strategies end any doubt.

The Cook Political Report/RT Strategies Poll taken Thursday – Sunday shows Bush’s approval dropping 7 points from their poll a month ago, 47% to 40%. CBS News (Wed-Sun) brings similar news reporting an eight point drop from their poll in late January to an all time low of 34%. And then Rasmussen’s daily tracking poll shows a five point drop over the last week to 43%. The RealClearPolitics Average of President Bush’s job approval is at 40.4% , but don’t be surprised to see that slide back to its low in the high 30’s as new post-DPW polls come out in the next few days.

Bottom line, this ports fiasco has been a political fiasco for the White House. The Cheney shooting accident was a trumped-up political story that inflicted no real damage on Bush; the Dubai deal is a completely different story. This seemingly obscure business deal and its impact could be the single biggest political story of 2006, and unlike Abramoff or Katrina or Scooter Libby, Dubai Ports World could be the catalyst the Democrats have been seeking for a big 2006.

January 27, 2006

The Independent Red Flag For Dems

The Los Angeles Times does not have a reputation for producing poll results that tend to favor President Bush. And indeed, their new poll does have plenty of bad news for the president, especially when you look at how self-described independents respond to various questions about his handling of a whole host of issues. Bush has clearly lost a decent amount of support from this group over time, and I don't think that comes as a surprise to anyone.

However, the most interesting results are found when you look at how these same self-described independents respond to questions about national security and about Congress. Here are a few examples:

> When asked who they "trust to do a better job of protecting the nation against terrorism"  Independents favored Bush over Democrats in Congress by 19 points.

> When asked whether they agree with those seeking to reauthorize the Patriot Act, 55% of Independents said they agree with reauthorization, 42% said they disagree.

> When asked whether people "should be willing to give up some of their civil liberties so the government can keep the country safe from terrorism", 50% of Independents responded "yes" while 43% said "no."

> 54% of Independents think hearings should be held to investigate the NSA program, but only 41% think impeachment would be warranted if those investigations concluded the President broke the law (that number is 39% overall).

> Independents give Congress nearly as low of a job approval ratings as self-described Democrats do, 32% vs 30% respectively, but when asked about favorable and unfavorable ratings for the two parties, Independents give Congressional Democrats a only a 31% favorable rating (41% unfavorable) while they give Congressional Republicans a 38% favorable rating (34% unfavorable).

> When asked which party in Congress had "higher ethical standards," Independents gave both parties low marks (Republicans 8%, Democrats 5%) with 79% concluding there is "no difference" between the two. That was 11 points higher than overall.

To summarize, based on the results of this poll (and keep in mind it is only a single poll, though it does comport with other data we've seen recently) Independents aren't thrilled with President Bush and they don't have particularly warm feelings toward Republicans in Congress. However, they seem to have an even lower regard for Congressional Democrats and, even worse, they seem to continue to lack confidence in the Democratic party on matters of national security.

January 19, 2006

The Blurry Face of Beltway Corruption

Republican Ed Rollins on the results from the new Diageo/Hotline poll: “Abramoff is the new face of Beltway corruption – but his mug shot is still blurry to most people.”

Ironically enough, Congress' job approval rating in the new poll - conducted just over a week after news of Abramoff's guilty plea made headlines coast to coast - is up five points to 31% since the last survey in December. Go figure.

Other results from the poll: Bush JA at 46% (down 4 points from December), Generic Congressional vote Dems +7% (down a net of 3 points from December), and a 33/54 right track/wrong track number (unchanged since December).

Updated RCP Indicies:

Bush Job Approval: 43.6%

Congressional Job Approval: 31.6%

Generic Congressional Ballot: Dems +9.4% 

Direction of Country: Right direction 32.3%, Wrong direction 58.8%

January 12, 2006

FOX News Poll

New FOX News/Opinion Dynamics poll out, Bush's job approval is unchanged at 42%. Updated RCP Index for Bush Job Approval = 42.5%.

By the way, here's an oddity from the FOX poll that caught my eye:

11. In general, which party, Democrats or Republicans, do you think is currently doing a better job reaching out and trying to cooperate with the other? (ROTATE)

Democrats Are = 27%
Republicans Are = 27%
Neither = 27%

How's that for an evenly divided country?

January 10, 2006

Poll Break

UPDATE: WaPo/ABC News out now as well. Presidential and Congressional job indicies revised.  

Two new polls out yesterday: CNN/USA Today/Gallup and CBS News.  New RCP indicies are as follows:

Bush Job Approval:  Approve 42.8% 43.0%,  Disapprove 54.0% 54.2% 

Congress Job Approval: Approve 32.6% 34.0%, Disapprove 57.1% 58.3%

Generic Congressional Vote:  Republican 38.2%, Democrat 48.0%

Direction of Country:  32.8% Right Direction, 60.5% Wrong Direction

December 23, 2005

The Robots Hit 50

This morning Bush hit 50% in Scott Rasmussen's three day rolling average for the first time since July. Overall, Bush's average job approval now stands at 46.0% with 51.2% disapproving.

December 21, 2005

Finally, A Ray of Hope For Congress

The improvement in Bush's job approval numbers since he began an aggressive pushback on Veterans' Day has been widely commented on. In fact, a number of other indicators have improved recently as well: consumer confidence, the percentage of people who feel the country is headed in the right direction, etc.

However, one number that hasn't improved at all, despite the uptick for the president and of the mood of the country in general, is the public's impression of the job Congress is doing.  Look at a comparison of job approval rating gains and losses for Bush and Congress in the most recent round of polls:

Bush JA

Chg vs
Last Poll

Congress JA
Chg vs.
Last Poll
CBS News

Congress finally got some good news this week with the release of the latest Washington Post/ABC News poll. The big headline from the poll was that Bush's job approval surged 8 points to 47%, but I'm sure what got the most attention on the Hill is that Congress' job approval surged as well, up 6 points to 43% from 37% in the last poll taken in early November.

Obviously, one poll does not a trend make. But given the steady decline of Congress' job rating throughout 2005 - the folks in Washington started the year with an average approval rating of 42.3% and finished at 29.8% - any piece of good news, no matter how small, has to be welcome.

December 09, 2005

Rasmussen's Robots Backpedaling?

Since peaking at 48% percent on December 5, Bush's job approval has slid back down to 43% in the latest Rasmussen poll. Bush's job approval in the latest RCP Average (11/28-12/8) stands at 41.2% which is an uptick of about three and a half points from where it stood at the middle of November (37.6%, 11/3-11/13).

RELATED: Mark Blumenthal updates his ongoing analaysis of the recent Bush 'bump.'

December 05, 2005

Bush's Curious Numbers

A few interesting highlights from Time's new poll:

> Though President Bush scores his lowest approval rating ever (41%) in the Time poll and gets at or near his worst rankings on nearly every issue, when respondents were asked who they would vote for in a do-over of the 2004 election, Bush loses to Kerry by only a single point (48-47). This probably says more about what a miserable a candidate Kerry was than about the problems of the Democratic party as a whole. Nevertheless it should disturb Democrats that with Bush's numbers so far in the tank there isn't a greater sense of "buyer's remorse" among the public.

> Three out of four (76%) say they are unlikely to change their mind about Bush's job approval in the future, but in a later question 46% say it is likely that Bush can regain higher job approval ratings over the last three years of his presidency.

> On the issue of trustworthiness, 65% say the Bush administration is the same or better than previous administrations, 34% say it is less honest and trustworthy.

> People were asked to rank their impressions of the negative impact of 10 issues on Bush's overall job performance. Response choices included "very negative," "somewhat negative," "negative, but not very or somewhat," "no negative impact," or "no answer/don't know." Here is how the issues rank when viewed by the difference between the percentage of those who feel an issue has had a "very negative" impact (the worst possible rating) and the percentage of those who feel an issue has had "no negative" impact (the best possible rating):

Gas Prices
Social Security

This reconfirms a lot what we already know: high gas prices and the continued struggle in Iraq are the two issues most negatively affecting Bush's overall job rating. The two other issues with a net-negative impact (cronyism and the deficit) are also issues where the Republican base has recently expressed deep dissatisfaction with the president over Harriet Miers and runaway federal spending. Finally, while the Libby indictment has taken its toll on Cheney (his approval rating in the Time poll has fallen 19 points to 32% since October of last year), it hasn't had much of an impact on Bush's overall job approval numbers.

November 23, 2005

Bush Approval at 39.8%

Newly-updated RCP Bush job approval index shows the President edging back toward 40%. 

By the way, we've added the new Cook Political/RT Strategies poll to the mix. It's a collaborative effort by Charlie Cook of National Journal and two veteran pollsters who've opened their own shop: Thomas Riehle, formerly of AP-Ipsos and Lance Tarrance, formerly of Gallup and founder of The Tarrance Group. View the full questionnaire (in pdf format) here. Read Charlie Cook's write up of the results here.

November 18, 2005

Poll Dump

Scott Rasmussen has just dumped a bunch of new poll numbers, including:

Bush Job Approval (11/15-11/17): 42/58

How Is Bush Handling Iraq? (11/16-11/17):  16% excellent, 17% good, 15% fair, 51% poor

Set Firm Deadline For Troop Withdrawal? (11/16-11/17):  45% in favor, 41% opposed

Rasmussen also has recent numbers on 2006 races in Ohio:

Governor:  Ted Strickland (D) 42, Ken Blackwell (R) 36 and Strickland (D) 39, John Kasich (R) 33.

Senate: Paul Hackett (D) 42, Mike DeWine (R) 41 

November 11, 2005

The Vital Signs

It's hard to see how they can get much worse for Bush and the GOP. Here are the latest RCP Indices:

Bush Job Approval: 38.2% approve, 56.2% disapprove

Congress Job Approval:  32.8% approve, 58.3% disapprove

Direction of Country: right track 29.7%, wrong track 63.0% 

Generic Congressional Vote: Democrats +11.7%

November 03, 2005

WaPo/ABC News

New poll from The Washington Post/ABC News (10/30-11/2) has Bush job approval at 39%, which is  unchanged from their last sample taken just days ago (10/28-10/29). We'll have a further update when the full survey results are published on the Washington Post site.

UPDATE: Here are the full results. And here is a look at the sample: Dem 31%, Republican 27%, Independent 38%.

According to the data, that's the same breakdown they used in the survey from early September (9/8-9/11) that resulted in a job approval rating of 42% for Bush.

UPDATE II: Ap-Ipsos (10/31-11/2) is out now with Bush at 37%. That's down two points from their last survey in early October (10/3-10/5). Waiting on full data.


November 02, 2005

CBS News Poll

New CBS News poll out tonight. Highlights:

  • Bush job approval at all time low of 35%
  • Bush favorable rating at 33%
  • Right track 27%, wrong track 68%
  • Congress job approval 34%
  • Congressional Dem favorable rating 41%, Congressional Republican favorable rating 35%

Now for the numbers behind the numbers. Take a look at the composition of the respondents:

Total Respondents (Unweighted) = 936
Republicans =  259 (27.67%)
Democrats = 326 (34.83%)
Independents = 351 (37.5%)

Now look at the weighted sample:

Republicans: 223 (23.80%)
Democrats: 326 (34.79%)
Independents: 388 (41.4%)

The result is a 35% job approval for the president, which is roughly 4-8 points lower than the other polls out right now.

October 31, 2005

Does USA Today Read Its Own Polls?

On the front page of the USA Today web site there is a link to the new CNN/Gallup/USA Today poll that says, "Poll: Presidency failing." Click through to the data and you'll find Bush's job approval for current sample (Friday 10/28- Sunday 10/30) at 41%. That number is unchanged vs. previous pre-Libby indictment sample (Monday 10/24 - Wednesday 10/26).

Other interesting tidbits about the "failing" Bush presidency: Bush up 2 points (versus last survey) on question of whether he's seen as "honest and trustworthy." Bush up 3 points on question of whether he's seen as a "decisive leader." VP Cheney favorable rating down one point and Karl Rove favorable rating up two points.

A forty-one percent job approval rating is nothing to crow about, but it's laughable that USA Today can suggest the Bush presidency is "failing" when the President's approval rating is unchanged from earlier in the week. It would be the equivalent of USA Today running a headline "MARKET CRASHING" with the knowledge that Dow Jones Industrial Average hadn't dropped a point.