Poll: Bush Approval Numbers Up

Richard Morin
Washington Post Polling Editor
Tuesday, December 20, 2005; 1:00 PM

Washington Post polling editor Richard Morin was online Tuesday, Dec. 20, at 1 p.m. ET to discuss the new Post/ABC poll, which shows an eight point increase in President Bush's approval rating and the impact of his recent explanations about Iraq may have had on the shift.

Read more: President's Approval Rating Rebounds.

Poll Results.

Survey Sample.

The transcript follows.


Richard Morin: Welcome! And happy holidays. We have a lots of good questions (and a couple that made me mad). So let's begin.


Philadelphia, Pa.: Your article makes it clear that this "rebound" is due almost exclusively to more support for Bush from his "base." Unfortunately, the survey information the Post is providing provides no such breakdown, nor is there any "demographic" data.

Could you provide some details on what happened with Bush's approval ratings among "independents?" Also, what percentage of respondents were considered "Bush's base" in this poll, and how does that compare to the percentage in the previous poll?

Finally, why do you describe this as a "bounce" that can be attributed to very recent events (i.e. "surges in the wake of the Iraqi elections.")? The previous poll was taken six weeks prior to this one -- and none of the data shows that people changed their minds in the last week. How do you know that this is, in fact, a "bounce", rather than a far more "linear" change of opinion?

Richard Morin: Thank you for your question. As we noted in the story, Bush's overall job approval rating increased by 9 points among Republicans since our early November poll and by 12 points among self-described conservatives. Clearly, these two groups constitute the president's natural constituency and the fact that so many Republicans and conservatives have returned to the fold must be heartening to the White House.

Here are the shifts among other key groups--but I warn you, the changes are too small and sampling error to high to say with confidence that there has been real change since our November survey. So please don't make too much of them.

Democrats +4 since November

independents +5

moderates no change

Similar patterns were apparent when I looked at shifts over the past five weeks in support for the war and related issues.


Philadelphia, Pa.: Unlike past Post polls, this one does not show the demography of the respondents for factors like Party ID, ideology, income, sex, registered voter, etc.

I'm curious as to why this poll omits that very important data.

Richard Morin: Be curious no longer. My valued colleague and assistant is on a well-deserved vacation this week, leaving me with all this work! Between the story, graphic and assorted other pieces of business, I finally got the demographic breakdowns to our Web site late last evening and they were posted on the site a couple of hours ago.

Here's the link. Enjoy!

Survey Sample.


New York, N.Y.: How is it that the headline reads like a Bush victory when the majority of Americans disapprove of Bush's performance?

Richard Morin: After six months of continuously bad news, the fact that Bush approval was up eight points was the news. But as the story carefully noted, fewer than half of the public approve of the job Bush is doing overall, and fewer than half think he's doing a good job in Iraq or with the economy (despite healthy gains in each area).


New York, N.Y.: Hi-

Can you provide a link for the article and poll that were done when Bush's approval ratings were at the lowest? I wanted to compare the poll results. Thanks.

Richard Morin: Check out the polling corner of the Web site. Go to POLITICS, then click on POLLS. You will find the trend document for this poll, which has all of our approval trend data going back to 2001. You will also find all of our poll stories


Fairfax, Va.: Had Bush's recent PR blitz NOT moved his poll numbers up, would The Post have printed a major front page headline to that effect?

Richard Morin: Yes, though the fact that all of the good coverage following the successful election in Iraq didn't move the needle would be most interesting to me.


Washington, D.C.: Mr. Morin: As I'm sure you know, your most recent poll continues a trend in which the ABC/Post poll has typically found a higher job approval for George W. Bush than those found by other polls conducted more or less simultaneously. Can you explain why this happens?

Richard Morin: I'm sorry, I don't know that. This survey differs from Gallup, as I discussed earlier. But in early November, we had Bush's job approval at 39 percent and Gallup had it at 40. In October, we had Bush approval at 39 and Gallup pegged it at 41. Heck, we weren't even the organization with the highest Bush approval out in the past week. That honor belongs to Diageo/Hotline, which had him at 50 among registered voters.


Fairfax, Va.: I don't recall The Post printing top of the front page stories on the successive declines in Bush's poll numbers. What is it about this new poll that warrants such a focus when Bush's negatives didn't? Given that poll numbers are powerful political ammunition for either party, shouldn't they be reported with equal emphasis by an unbiased paper like The Post?

Richard Morin: You are mistaken. My November survey story that recorded Bush at new lows for his presidency on virtually every key measure was played more prominently on page 1 than this article. Other stories noting Bush's post-spring decline ran prominently on page 1. Your critique is comforting, however--the complaint for months from Bush supporters was that we go out of our way to highlight Bush's bad numbers.


Naperville, Ill.: Why haven't you polled on public support for the impeachment of George W. Bush?

Richard Morin: This question makes me mad...


Seattle, Wash.: How come ABC News/Post poll has not yet polled on impeachment?

Richard Morin: Getting madder...


Haymarket, Va.: With all the recent scandals and illegal/unconstitutional actions of the President, why hasn't ABC News/Washington Post polled whether the President should be impeached?

Richard Morin: Madder still...


Dublin, Ireland: In a statement on Sunday, John Dean, former White House counsel during Watergate, stated that President Bush is "the first President to admit to an impeachable offense." Will The Washington Post be polling about impeachment of the President in the near future, now that this topic has taken on national significance?

Richard Morin: An impeachment demand from Ireland? Oh my gawd. Now I'm furious.

Let me explain.

For the past eight months or so, the major media pollsters have been the target of a campaign organized by a Democratic Web site demanding that we ask a question about impeaching Bush in our polls.

The Web site lists the e-mail addresses of every media pollster, reporters as well as others. The Post's ombudsman is even on their hit list.

The Web site helpfully provides draft language that can be cut-and-pasted into a blanket e-mail.

The net result is that every few months, when this Web site fires up the faithful with another call for e-mails, my mailbox is filled with dozens and dozens of messages that all read exactly the same (often from the same people, again and again). Most recently, a psychology professor from Arizona State University sent me the copy-and-paste e-mail, not a word or comma was changed. I only hope his scholarship is more original.

We first laughed about it. Now, four waves into this campaign,we are annoyed. Really, really annoyed.

Some free advice: You do your cause no service by organizing or participating in such a campaign. It is viewed by me and others with the same scorn reserved for junk mail. Perhaps a bit more.

That said. we do not ask about impeachment because it is not a serious option or a topic of considered discussion--witness the fact that no member of congressional Democratic leadership or any of the serious Democratic presidential candidates in '08 are calling for Bush's impeachment. When it is or they are, we will ask about it in our polls.

Enough, already.


Washington, D.C.: Is it harder to conduct a public opinion poll the week before Christmas than at other times during the year? More specifically, did you find that your contact and response rates were lower on this last survey than what you typically get?

Richard Morin: Great question! I cannot tell you what the cooperation rate for our latest poll because I have not yet received the data. I can't wait to see what it was. My guess is that we found it harder to reach people (lots of no answers) and that cooperation was down a bit. Call it the "Shop-Till-I-Dropped" effect. We did pause before going over this weekend, which is when America--including at least one pollster--did much of its holiday shopping. But the value of going into the field after the Iraq election outweighed these concerns.


Rochester, N.Y.: Recent polling number for Bush's approval rating have been all over the map (in polling statistical terms). Gallop/CNN just released a poll that had his approval rating at 41%. Even taking into account the statistical errors, this seems like a big discrepancy. Is this an artifact of the sampling population/strategy? Also, there appears to be some traditional bias in polls from conservative organizations (not The Post)to rate Bush's performance higher than polling averages (e.g., RealPolitics summaries), so how are readers expected to trust polls of this type in general?

Richard Morin: Ah, trust. Hard to earn, easy to lose. Let me offer up a few points that might ease (but perhaps not eliminate)your suspicions. The Gallup result is initially a bit startling when compared to ours, but less so when you walk through the numbers.

As most of you may know, Gallup reported Bush's overall job approval rating this morning at 41 percent based on a poll conducted Friday, Saturday and Sunday. We reported in this morning's paper Bush's job rating at 47 percent based on our poll conducted Thursday through Sunday. Since we usually track closely with Gallup, your question is a good one.

The first likely suspect is party id: perhaps our poll contained too many Republicans and Gallup's contained too many Democrats. Nice theory, but it is wrong this time. Our sample was slightly more Democratic than Republican (33 percent D versus 32 percent R). But Gallup also had Democrats up by 1 percent.

One other potential reason is that the field periods were slightly different. When I, just for fun, dropped our Thursday interviews and just looked at the results of our Friday through Sunday subsample, I found that Bush's approval rating was 44 percent. That's still different but far less startling than Gallup's 41. In our poll, Bush had good nights on Thursday and Sunday; his worst nights were Friday and Saturday. A weekend-before-Christmas effect? Perhaps, but no one can say for sure.

There are so-called house effects in polls, which is why presidential approval ratings taken by different nonpartisan organizations will differ substantially. Some organizations base their results on voters, others on the general population. Some probe respondents to give an answer, others choose not to.

The thing I look for is the direction of individual surveys. Regardless of the individual results, to the polls agree in the direction of public opinion. Do they say Bush is gaining support or losing support, regardless of what the point estimate is?

When this test is applied, the polls look reasonably well behaved. We had Bush's job approval rating up 8 points from early November. Gallup has Bush up 4 since early November. Our friends at the New York Times and CBS had him up by 5, and so did AP-Ipsos. In my view, there has been movement toward Bush that began even before the Iraqi parliamentary elections and continued after the balloting ended.

Which brings me to this point. Gallup does terrific work. But I found it odd that they had Bush at 41 immediately after the Iraq elections--down insignificantly from 42 in their pre-election poll. Bush didn't benefit--even a little bit--from the success of those elections and the favorable media coverage that followed? Certainly possible, but I find it hard to believe.


Burlington, Vt.: For at least the last several months there has been a wide divergence in Bush's approval numbers among Independents and Republicans of between 40 and 50 percentage points. I suspect that a divergence of this magnitude between Independents and members of a President's party is completely unprecedented. Is this true?

Richard Morin: Interesting, I do not know the answer to that question historically, but I will look into it. Do you have any data on this? For the record, 38 percent of all independents approved of the job Bush was doing as president compared to 87 percent of all Republicans, a 49 point gap. In our November poll, the R/I gap was 45 points.


Arlington, Va.: How do you explain the fact that the major movement in this poll is a shift from "strongly disapprove" (down seven points from prior poll) to "strongly approve" (up eight points in this poll)? That seems like either a shocking shift of opinion for one segment of people or a uniform shift across all categories.

Richard Morin: I find those movements consistent with the overall finding of this survey, namely that Bush has rebounded in recent weeks. Some weak approvers have been heartened by the new, rosy economic numbers and the Iraqi election so they become strong approvers. Likewise, some percentage of strong disapprovers softened. It doesn't mean strong disapprovers became strong approvers, rather some strong disapprovers became weak disapprovers and on the other side of the equation some weak approvers became strong. Perhaps I am missing something, which is entirely possible--you try thinking and typing and trying to figure out how to make paragraphs in this online sustem, and you'll miss stuff, too :)


Albany, N.Y.: Do you believe that the polls done by other news outlets are just as newsworthy as your own poll, and if so, why do they not receive the same degree of prominence in The Post?

Richard Morin: No, frankly, I don't, and for the same reason that we do not publish stories about the same press conference done by USA Today, Wall Street Journal, The Los Angeles Times, that other Times, etc. We run our own. That does not mean we do not cite other poll findings, or that we won't give strong play to a poll on a topic we haven't surveyed on. It's just that we have more confidence and more control over our own polling.


New York, N.Y.: When a newspaper like The Post commissions a poll, it gives the result prominent play, usually on the front page. But when a different organization conducts a separate poll, that poll's results are given much less prominent play, and often not mentioned at all. The implicit assumption is, "Our poll is better than theirs." Is this sound journalism?

Richard Morin: See the last answer. It would be unsound journalism to ignore other survey results, particularly if they offer insights your own may lack. But to give them as prominent play? No, and I think it is unreasonable to expect us to.


Long Beach, Calif.: What do you expect to happen with the next poll, taken AFTER the revelations that Bush has circumvented the courts in obtaining surveillance warrants?

Richard Morin: Interesting question. Let's see how the politics of this spins out after the holidays. One theory is that Americans will be outraged by this invasion of their privacy and further diminution of their rights. Coupled with revelations about secret prisons abroad and torture, this could change the political dynamic. Another theory is that Americans will tolerate this if they can be convinced that it is helping to capture terrorists.

I have to admit I was astonished after 9/11 how my polls and others showed that Americans were more than willing to surrender basic rights in the name of fighting terrorism. I do not have those numbers handy, but I'll dredge them up for the next chat. It really was frightening.

You can see Bush starting to make the case directly that extraordinary surveillance measures were necessary and done in only extraordinary circumstances and under close scrutiny. Also, we're learning more about what Democratic leaders in Congress knew and when they knew it, so we'll see how that affects the mix.

Time will tell. (You don't get insights like that with Tom Edsall!)


Washington, D.C.: About your explanation of "weak approvers," etc, that would make sense if the numbers had gone up in the middle categories, but they didn't. For example, if "weak disapprovers" softened up and moved into the "weak approvers" category, why is there no significant movement in that category? If "strong disapprovers" moved into the "weak disapprovers" category, why hasn't that category jumped up.

Richard Morin: ...because some small percentage of weak disapprovers (eight, to be exact) became weak approvers while some weak approvers became strong? In other words, everybody shifts to the left on the approve/disapprove scale. Again, America, help me--what am I missing here?


Kiawah Island, S.C.: Mr. Morin,

Thanks for taking questions today. With the growing number of households that only use cell phones and have no land lines, do you plan to make any adjustments to your polling techniques? Do you feel that the integrity of the polling data, to this point in time, has been compromised due to this trend?

Richard Morin: Yikes, I have nightmares about this. There are a number of steps that we are taking as a polling organization and as part of the broader polling community to address this issue. My expectation is that we are moving to the day when more polling will be done over the Internet, a la the Knowledge Networks panel. KN randomly selects national samples and hooks them up to the Internet if they aren't. This gets around the socio-economic skews in computer coverage. Also, I expect that we are close to being able to do Random Digit Dial samples calling cell phones (and figuring out a way to do it without making the respondent pay for the call). But here's my wild prediction of the day: More face-to-face interviewing, doing surveys the way that George Gallup, Elmo Roper and the others did them fifty years ago.


Silver Spring, Md.: What is your qualification regarding statistics/poling?

Richard Morin: Here's my official bio:

Richard Morin is director of polling, staff writer and columnist for The Washington Post. Morin came to The Post in 1987 from The Miami Herald, where he had been survey and research editor for five years. Before that, he was an investigative reporter, urban affairs reporter and an editor at the Herald for five years. He worked as night police reporter, city hall reporter and an investigative reporter at The Arizona Republic in Phoenix, Arizona for four years before joining The Herald. His work as a reporter and editor has been honored by the American Bar Association, the association of Investigative Reporters and Editors (IRE), Sigma Delta Chi-Society of Professional Journalists and others. In 1980, he was a finalist for the Pulitzer Prize in Special (Investigative) Reporting. In addition to his duties as a staff writer and director of polling, Morin writes writes a bi-weekly general interest column on new research in the social sciences called "Unconventional Wisdom" for the Sunday Outlook section. He currently serves on the board of directors of the Roper Center for Public Opinion Research at the University of Connecticut. In 1989, he went to Taiwan under the sponsorship of the Asia Foundation to advise Taiwanese journalists how to use public opinion polls to cover political campaigns. In 1993, he conducted workshops in Mexico City on polling and the press, and in 1999 he conducted similar workshops for journalists in South Africa as a consultant to the Independent Newspaper Group, the country's largest newspaper chain. Born and raised in Los Angeles, California, he received a BA degree in 1971 from the University of Redlands in California, and an MA in journalism from the University of Missouri in 1973. He studied survey methodology, computer programming and applied statistics at the Institute for Social Research at the University of Michigan in 1983 and 1984. In the Fall of 1999, he was a Research Fellow at the Joan Shorenstein Center for the Press, Politics and Public Policy at Harvard University's Kennedy School of Government.


Richard Morin: Okay, there are a LOT of angry comments directed at me over my impeachment poll response(s). We're out of time, but I do want to share them with you.

Here they come...Welcome to my world! And see you next time.


Tucson, Ariz.: Will your next poll ask for reactions to the White House contention that when the Bill of Rights interferes with their freedom of action it can be ignored?

Will there be any questions about how this disregard for constitutional limits affects opinions on impeachment?

Richard Morin: still more...


Arlington, Va.: Senate Democratic Deputy Whip has begun the impeachment discussion by publicly asking 4 Presidential scholars about whether this is an impeachable offense. It is too early in this matter for a leader to "call for" impeachment, but it is becoming a serious discussion topic.

Howard Dean, in an email to supporters, just wrote, "Richard Nixon once said in an interview that, "if the president does it, it can't be illegal."

"He found out that wasn't true. This administration may need a reminder."

I don't see why someone has to "call for" impeachment before it becomes a pollable topic. Shouldn't the fact that it's being discussed as a possibility be enough?


Richard Morin: more...


Lansing, N.Y.: Now that President Bush has openly admitted that he does not feel he has to obey laws he finds constraining (the FISA) and has in fact broken this law numerous times and plans to continue doing so, will you include a question in your polls asking if respondents think he should be impeached?

Richard Morin: more...


Washington, D.C.: In her November 13 column, Post Ombudsman Deborah Howell addressed reader requests for The Post to polls to measure public support for impeachment.

Howell wrote: "First, there was a swarm to me and to Post Polling Editor Richard Morin asking that The Post do a poll on whether President Bush should be impeached ... The question many demanded that The Post ask is biased and would produce a misleading result, Morin said; he added that the campaign was started by"

Please explain WHY a question asking if President Bush should be impeached if he lied to the country about war is "biased".

Please also explain how this is consistent with polls the Post ran -- under your direction, I might add -- in 1998 asking whether then-President Clinton should be impeached if he had an affair with Monica Lewinsky. Do you now believe those questions you asked -- and reported on -- throughout 1998 were "biased"? If so, do you believe you and The Post owe Clinton an apology?

Why does The Post think it is appropriate to raise the spectre of impeachment when there is a Democratic president, but not when there is a Republican in office?

For more information, see this post: Media continues to ignore impeachment polling

Richard Morin: still another...


Alexandria, Va.: So what if the questions on impeachment are drummed up by a Web site? Seriously, so what? It would be most interesting to see what percentage of the population actually believes impeachment should be on the table, to see whether it really is just a Democratic canard or the start of a groundswell. After all, the Constitution says impeachment is for "high crimes and misdemeanors", and breaking the law certainly qualifies.

Richard Morin: and another...


Whitewater, Wis.: Do you think the upbeat poll simply indicates a holiday rush of benevolence? Do you think there is a possibility that impeachment proceedings will begin in 2006?

Richard Morin: and another...


Falls Church, Va.: "An impeachment demand from Ireland? Oh my gawd. Now I'm furious," you say. Are there no American citizens living in Ireland??? And though I agree that these mass campaigns to get impeachment questions in your polls are poorly organized and downright obnoxious, now that there are members of Congress suggesting that impeachment would be appropriate if, in fact, the spying with no warrant is illegal (e.g. Boxer, Lewis) will you consider asking the question? Or are members of Congress only important when they are running for president?

Richard Morin: and another...


What About Boxer?: Excuse me - but I don't know the site you refer to. I DO know my representatives' positions.

You just posted this:

"That said. we do not ask about impeachment because it is not a serious option or a topic of considered discussion--witness the fact that no member of congressional Democratic leadership or any of the serious Democratic presidential candidates in '08 are calling for Bush's impeachment. When it is or they are, we will ask about it in our polls."


Not to mention the Nations foremost expert on

presidential hubris, Mr. John Dean.


Enough, already.

Richard Morin: Here's one


Re- Impeachment: Three words for those of you demanding The President be impeached.

President Dick Cheney.

Richard Morin: and let this be the last word...


Richard Morin: Happy Holidays!


Editor's Note: moderators retain editorial control over Live Online discussions and choose the most relevant questions for guests and hosts; guests and hosts can decline to answer questions.

© 2005 The Washington Post Company