Washington Post Senior Political Reporter Dan Balz Yesterday and Today on the Latest Polling
HH: Right now, however, I’m joined by Dan Balz, senior political reporter for the Washington Post. Dan, welcome back, it’s good to have you.
DB: Thank you. Good to be with you again.
HH: Your reaction to Rick Santorum’s announcement today?
DB: Well, you know, I mean in a sense, the suspense in this Republican nomination battle had been drained out, I think, really with last week’s contests in Maryland and the District of Columbia, and particularly Wisconsin. Rick Santorum had a very, very difficult path forward. He was hoping he could win his home state of Pennsylvania, and maybe make a run in a bunch of primaries in May in some of the southern states where you know, the electorate would set up much better for him than some of the northern states had done. But the reality was he still wasn’t going to be able to become the nominee, I think. And you know, a lot of factors entered into it. A loss in Pennsylvania on the 24th of this month would be a real setback for him, politically, and would cloud his political future. And he’s made a remarkable run in this campaign, and one that you know, almost nobody other than perhaps him and his wife, and a few loyalists, thought was even possible. And so you have to give him a lot of credit for what he accomplished. And I think he realized that there’s a right time and a wrong time to get out, and you don’t want to overstay in a contest like this, and I think he decided that this was the right moment to get out.
HH: Clearly, Mitt Romney is happy, because now he can devote all of his primary dollars that he can continue to raise to building the network in those 14 states. And that brings me to your story today, along with Jon Cohen in the Washington Post about this poll. And I want to begin with a question, Dan. How much attention would you pay to a poll of adults as opposed to registered voters or likely voters as indicative of November’s results?
DB: Well, we generally don’t do likely voters at this stage of the campaign. We wait until much later for that. We do both, and this poll had it, we do both adults and registered voters, and the head to head match-up between the President and Governor Romney was on the basis of registered voters. The other thing we find, that when we look at these specific numbers, there’s not a significant difference in a lot of the findings between a poll of all adults and a poll of registered voters. So we feel pretty confident that, you know, that this is a good representation of where things stand at this point.
HH: Can people get into the crosstabs, because I couldn’t find them. And I know that in the registered voters match-up, it’s 51-44, Obama-Romney, which I also, I just don’t think it’s indicative of much, because it’s not the 14 states, and it doesn’t do likely voters. But the likely impact of the headline that Obama is more likable than Romney may have some secondary impact that I think may not be warranted, given that that might be one of those things where adults are really very different from likely voters, Dan Balz. Would you agree with that?
DB: I’m not sure that I would agree with that, Hugh. And the second thing I would say is that the margin on this question of sort of the likeability is huge. I mean, it was, I don’t have the thing in front of me, but it’s like 64-29 or something. I mean, you know, you could draw the sample any way you wanted, and that is a huge gulf. I mean, maybe, you know, I’m not saying it would, but even if it moved a couple of points, it’s a huge difference. And to the question of who’s the more inspiring candidate, you know, it was a 25 or so point margin on that. So you know, I think those are problems that Governor Romney has to address, or else has to figure out how to make them not part of the campaign.
HH: You know, I’m just not buying it, Dan, because I A) I think a gap could be much more significant than a couple of points between likely voters and adults. But I also wonder whether, if you have anything to compare it to 1980 numbers, because I’ll bet you at the same time in 1980, Ronald Reagan was viewed with a great deal more dislike than Jimmy Carter, and it turned out to not be an issue at all as, of course, the conventions and the general election campaign brought them both into focus for adults, registered voters and likely voters. Is there anything to compare the data set to?
DB: I’ve never looked back, and I mean, that’s a good question. I’ve never looked back at the 1980 numbers. My guess is that, you know, at different points in that campaign, Jimmy Carter was pretty unpopular, too. I mean, his approval rating, I mean, once you got through that first, that kind of rally around the president after the Iranian hostage taking, his numbers started to go down. So my guess is that on those kinds of questions, he was not in particularly good shape through much of that 1980 campaign. But you know, let me go back to the point I just made. You can look at these personal attributes and say the President has a significant lead over Governor Romney, and that those may or may not be significant. You can look at other elements of this, and you know, I think the story, and the lead of the story stress this, that while the President has advantages on a number of these personal attributes and on some issues, that there is one big looming vulnerability, and that is that people don’t think this economy is in a genuine recovery, that he gets negative reviews, and negative approval ratings, on his handling of the economy, and that on the questions of who’s better at creating jobs, or who would handle the economy better, the President and Governor Romney are very, very close to one another. So I mean part of this is a question of what becomes really important to voters in this choice in November, and I think that’s a function both of how voters view the candidates, and what the campaigns are able to do in trying to frame the race. So you know, going forward, you know, you have to say this is likely to be a pretty competitive race. We’ve thought that from the beginning, and I don’t think anything’s happened that changes that. Governor Romney does begin the general election with some deficits that he’s got to make up. I mean, I think that the numbers on his support among women are a potential problem. And he was just speaking in Delaware just before I came on the air, and you know, went right at this question of the Democrats’ accusation that there’s been a Republican war on women, and said the real war on women has been the economic hardship that this president has imposed on them, because his policies have not brought the economy around.
HH: Now Dan…
DB: That’s going to be part of the debate going forward.
HH: I want to phrase this, I’m going to call it my Andrew Breitbart memorial question, but I want the audience to understand that I say it with the greatest amount of respect for your reporting. It’s not about Dan Balz. But when I read this, this morning, I thought what Andrew would say to me is this is the mainstream media blocking and tackling for President Obama. We have gas prices going out of the sky, the term malaise is in the headline. And you guys are running likeability and women polls. And I frankly, I just don’t trust your pollsters. I don’t. I don’t believe a word of this.
DB: Well Hugh, a month ago, you know, we had this race looking a lot different, and we got criticism from Democrats that we had somehow, you know, we were putting up a poll that didn’t reflect reality. I mean, one of the problems today is that people on left or right read into the kinds of things that we do as some effort on our part to carry water for the other side. You know, I mean, you know how I operate.
HH: You don’t do it, but I don’t know Peyton Craighill or Scott Clement. And why aren’t the crosstabs available?
DB: The crosstabs should be available. I mean, if you email Scott or Peyton or somebody, the crosstabs are generally available.
HH: Okay, they’re not linked at the story. That’s why I didn’t think to write the pollster and say can I…but I would think that in this day and age, and I wanted to give you the last 45 seconds here, in this day and age, given that environment, and the Andrew Breitbart question, why should anyone believe unnamed, not unnamed, but unknown quantities when it comes to the general election campaign?
DB: Well, I mean, we’re not asking people to do that. You know, the Post/ABC poll, I think, has been one of the most consistent polls, and you know, I’ve been dealing with it for 25 or 30 years, has been a consistent and a high quality poll. You know, we take enormous pride in the attention to detail that we put into it. The people who run the poll are first rate. And I mean, their goal is not to skew numbers by any means. It is to try to get at…
HH: The truth. All right, Dan Balz, thank you. It is always a pleasure to talk to you. I look forward to checking in frequently in 2012.
End of interview.
– – – –
HH: I begin this hour today and next segment with Dan Balz, who is back to complete the conversation we began yesterday. I couldn’t keep him over yesterday, because Del Wilber was hard on his heels. But Dan Balz, thank you so much for coming back, and I apologize for the hard out yesterday, because that was abrupt, and we had a lot to talk about.
DB: We did, and those things happen in radio and TV, so no problem. I’m glad I could make it back.
HH: Well, where we were yesterday when we led off was the ABC News/Washington Post poll that showed President Obama with some significant statistical leads. And I had raised the question about my unease with the poll. Since then, Dan, two other conservative by admission commentators, Ed Morrissey at Hotair.com and Jay Cost at the Weekly Standard, have gotten into the crosstabs of this most important poll, and said look, there was an eleven point Democrat advantage over Republicans, which is really a very significant advantage. Does that add to your disquiet with the poll? Or is that something that you don’t concern yourself with?
DB: Well, I mean, these are questions that are better answered by Jon Cohen, our polling director, in whom I trust implicitly on all matters of methodology. I did talk to him about that. This is an issue, this is an issue in polling. Let’s just start with that. I mean, this is something that pollsters discuss and debate, and actually have different views on. Our feeling is this. Party identification is not a demographic. You can’t go to the Census and say the country is X% Democratic and X% Republican, and X% independent. It’s something that varies over time. And this is something that we don’t say argue a registered Democrat. No pollster, you can do that in a poll, but you also ask people do you identify yourself as a Republican or a Democrat, that number, you know, poll to poll, news organization to news organization, you know, conservative pollster, liberal pollster, that number can vary from time to time. And in this poll, compared to the last one, the Democratic number moved up a little, which, but not out of the ordinary for the normal fluctuation to see. The Republican number moved down a little bit, but again, not out of the ordinary. In those circumstances, we kind of let things float. And our feeling is that you know, over time, these things iron themselves out, and that the poll still has validity. So I mean, I went back and looked at, for example, party identification in exit polls in presidential and mid-term years. And in almost, in most cases, Democrats outnumbered Republicans on election day, except for a few important differences. One was 2004 when George W. Bush won reelection. It was even – 37% Democrats, 37% Republicans. That was a year in which the Bush campaign recognized that one of the most important things they had to do was try to be at parity on election day, because normally, Republicans are at a disadvantage on election day in terms of party identification. It was also at parity in 2010, when Republicans had a big victory. Democrats had a big advantage on party ID on election day in 2008. So it is something that moves. I know that I got several emails on this same point, but I don’t think that is a valid criticism of this or any other poll, unless there’s a dramatic move, and then there’s, for which there’s no adjustment. But this was not dramatic enough. And I mean, I guess the other thing I would say is the findings in this poll are not out of line with what we’re seeing in most every other poll right now. I mean, there’ve been, I did a calculation today, there have been 29 polls that are on the Real Clear Politics list that have been taken on Romney V. Obama since the beginning of February. Romney has been ahead in two of those, including one of ours. He’s been tied in three others, and Obama’s been ahead in 24 of those. And in the most recent polling, including ours, Obama’s margin has been plus two, plus eleven, plus four, plus eight, plus seven. That was ours, and tie, which I think was in the Rasmussen track, although our poll obviously is not a tracking poll. So this is a debate. It has become actually a more sort of partisan or polarized debate over some of the mechanics of polling. We think we’re on pretty solid ground, and we think that over time, our polls have held up. Every polling organization occasionally has a bad one. I don’t think this one is, frankly.
HH: I’m talking with Dan Balz, senior political correspondent for the Washington Post, and I think your admonition to me is a very good one, that I really ought to talk to your pollster about this. But I want the audience to hear what Jay Cost, who’s a very significant, talented, even-keeled superstar, sort of the new Michael Barone, says in his Weekly Standard piece. He writes that the poll has an inexplicably large Democratic advantage. The party breakdown in the poll is 34% Democrat, 23% Republican, and 34% independent. As a point of historical comparison, the party spread in four of the last five elections since 2002 has been an even split between the two sides. In 2008, a perfect storm of bad news for the GOP, the party ID advantage was only plus seven. So a Democratic advantage of plus eleven in the Washington Post/ABC poll is an unjustifiable number, his emphasis, at least in terms of what the electorate is thinking. Now I know that that’s a technical issue for pollsters to debate. Maybe Jay and Jon, I can get them on and do that. What I want to go to, Dan Balz, with you is the political importance of this.
DB: Can I raise one other thing?
HH: You bet. Go ahead.
DB: …which goes to one of the questions we talked about yesterday, which was you said to me I really don’t like a poll of all Americans, or all adults at this point. I’d prefer registered voter. So I got our polling folks to give me some registered voter numbers on the attributes and issue stuff that we talked about yesterday.
HH: Oh, terrific.
DB: For example, on that one we talked, we didn’t talk about this, seems more friendly and likable. When you do all adults, it was 64% said Obama, 26% said Romney. When you do registered voters, it’s 64% Obama, and 28% Romney.
HH: So you were right. There’s no statistical difference there at all.
DB: In most of these, there is, you know, there’s a marginal difference. It’s a couple of points. Interestingly, one where there was a bigger difference, but it went against Romney, was on the question of addressing women’s issues. Obama was plus ten on that among all adults. He was actually plus seventeen among all registered voters. Conversely, on handling the federal budget deficit, Romney was plus thirteen among all adults. He was plus seventeen among registered. But again, these things, you know, they can have a marginal difference. But I think at this point in a campaign, to be talking about a two point difference, let’s say, on the horse race, you know, is it plus seven or is it plus five, you know, which if you had a slightly different party identification split, you might bring it to that, these are, I’m just guessing on these numbers. These are not real. But you know, a slightly less Democratic or more Republican sample in this poll is still going to leave Obama ahead of Romney. And I think a lot of the issues that were raised by the poll are still there for Romney. I mean, one of the things that’s been interesting, you know, you know, our poll had him at a 19 point disadvantage among women, among registered voters in the ballot test with the President. You know, you could quibble and say okay, maybe it’s 15. But everything that the Romney campaign has done over the last 24 hours says to anybody they recognize that they have a problem that they’ve got to do better with women voters, and they are addressing it. So again, I think the polling is born out by the way the two campaigns are acting right now.
HH: And that data certainly blows up my assumption about adults versus registered voters, perhaps not my assumption about likely versus registered, but that is a different question. After the break, I want to explore with you, and I’ll set it up right now. Jay Cost’s bigger issue, he runs through the Washington Post data fairly extensively, and says look, if you look at the numbers about right track/wrong track, if you look at the numbers on the economy still being in a recession, most Americans still believe it’s in a recession by a fairly astonishing number of 76%-21%, if you look at all those numbers, he said there’s incredibly bad news for President Obama in this poll. And here’s his summary, and we’ll come back from break and have the debate with this with Dan Balz. So here’s my bottom line, Jay Cost writes. ABC News/WaPo has again offered up a pro-Democratic sample that helps Team Obama spin the day’s news. Yet dig a little deeper, and there is bad news here for the President, even if his friends in the media do not want to talk about it. We’ve got 30 seconds to the break, so hold off and we’ll come back, and I’ll set it up again, and I’ll give you the floor, Dan Balz, for the entire, as much time as you want to respond to that.
– – – –
HH: Dan Balz, when we went out to break, I read you Jay Cost’s conclusion from the Weekly Standard, and I want to give you the floor, because he kind of articulated better than I did not a suspicion of your reporting of the poll, I think you reported it exactly as it is, but the suspicion among center-right people generally that pollsters for mainstream media tweak these numbers in such a way as to set up stories that help Democrats.
DB: Well, you know, I mean, I will just speak as candidly and as honestly as I can, that we don’t do that. I mean, I don’t do that, and I know Jon Cohen doesn’t do that. Let me read you the lead that Jon and I wrote on this poll. It said with the general election campaign beginning to take shape, President Obama holds clear advantages over Mitt Romney on personal attributes and a number of key issues, but remains vulnerable to discontent with the pace of the economic recovery according to a new Washington Post/ABC News poll. That was the way we laid it out. It said that the President has some advantages, particularly on the attributes and on some key issues, and we just went through a few of those, but remains vulnerable to discontent with the pace of the economic recovery. And all of the things that Jay cost cited are cited in our story as being the reasons why the President is clearly vulnerable. I mean, my shorthand on this was that this general election campaign starts with him having some important advantages, but with one large looming vulnerability, and it’s the one that’s always been out there, and it’s why I’ve thought, and obviously a lot of people think, this is likely to be a very close general election campaign. And so you know, I mean, people read into the way we write things, or read into the findings, or read into one phrase or another something to suggest that we’re trying to tip it. I know we were very careful, and we thought about this, and we talk about it before we actually sit down and try to write the story. How do we want to present this? How do you present it in the most balanced way possible? Everybody’s got a different prism through which they read things, and we can’t account for that. We’ve got to do it the way we think is the fairest way possible.
HH: Now Dan Balz, is there any conversation within the Post, which is obviously a key opinion center as well as reporter here, about running two samples – one that models up at 11 point Democratic registration advantage, and one that models at the actual turnout of ’08, and one that models, you know, three or four, and then reporting the different results? Or is there something that weds you, other than dollars, to one sample at one time, because it would seem to me that you referenced what I think is the great innovation in the last ten years, the Real Clear Politics average.
HH: But of course, we all use that, except you guys don’t, because you’ve got a proprietary poll, and you want to report on your poll. Is there something to be said for all pollsters running a few different models and disclosing all their crosstabs and all their data before setting into analysis?
DB: Well, you cited one reason – cost.
DB: I mean, this is not the period of news organizations being as flush as we might wish. So I mean, we’re in an era where we’re tight on budgets. Polling costs a lot of money. It is a huge investment, and we all spend a lot of money on it. But we don’t have unlimited money to do it. The second, a second reason is one of the values of polling is that we have trends. We talk about a poll being a snapshot in time, which it certainly is. But one of the important values of having regular polling, and we do a national poll almost every month, is that you collect data over time, and you can watch movements. If you are changing the sample, readjusting the sample, you lose the value of that. Which one do you end up with? And how do you determine, should the party balance be, you know, 11 or 7 or 3, or you know, I mean, you can drive yourself crazy with that. Now when you get closer into the general election, and everybody begins to move toward a likely voter screen, you know, that’s another whole art and science. Everybody’s likely voter screen is a little bit different. And you can screen very tightly for likely voters, you can screen a little bit looser. And news organizations sometimes do that, and we look at those to try to get a sense of it. But we don’t do that at this stage of the campaign. I mean, we’re so far out, Hugh, that you know, this election is going to change fifteen times between now and November. You can’t put that kind of precision on an electorate that is just beginning to pay serious attention, if they are, to a general election match-up. So you know, we’ve got to do it in the way that we have done it consistently. We think that provides the fairest look at the way things are or aren’t moving. One month may move back. Another month, the Real Clear Politics average, particularly when they’re averaging, you know, all the really good polls, gives you a sense of it. The Real Clear average today, or yesterday, was, I think, plus five for Obama. We were at plus seven. So there’s no statistical difference in those two. So there are a lot of reasons we don’t do what you suggest.
DB: I don’t think it would actually help. It would be great for everybody, you know, for all of us to talk about. But I don’t think it would tell us anything more than we know from the way we’re doing it now.
HH: Okay, a last question, and I want to give you two, three minutes to answer it. We know there’s some, at least I know, I don’t want to put words in your mouth, I know there are some traditionally really rotten polls – the Minnesota poll comes to mind out of the Minneapolis Star-Tribune, and the L.A. Times poll has often been an extraordinarily obvious fashion skewed left. But let’s assume for the moment that you’ve got good pollsters working really hard to do the right thing. We still have a 14 state election, and I think you’d agree with me, and we can run through them from Florida to Nevada. And we know what the swing states are. And they’re all, none of them are reflected in the national polls. They’re all different. By running national polls, one way or the other, is big media not accurately reflecting to the country where the race is, and how it will be decided, because we all know it’s those 14 states, and we all know that big polls don’t tell us anything about those individual states, so it’s all kind of a charade, Dan Balz?
DB: No, I don’t think so, because national polls, I mean, the only way you can get presidential approval over time, which is an important thing to have, is to do a national poll. You can’t simply say well, we’re just going to look at the president’s approval in swing states as we get close to an election. The second point I would make is that as you get closer to an election, news organizations do do swing state only polling. I mean, there is a way to do it, and we all sometimes do it. There’s also good individual state polling. But you know, the national horserace number, as you get into the fall, and you have good pollsters and polling, that national number doesn’t deviate a lot from the national number on election day. And that national number gives you an idea of how those swing states are probably going to go, or how the overall election is going to go. It is a rare case, as you know, in which somebody wins the popular vote and doesn’t win the Electoral College. You can win the popular vote narrowly and win the Electoral College by a significant margin. That’s certainly true, and that’s a function of the way some of those swing states break. But national polling still has a value. I mean, I think we don’t want to become so balkanized, so focused only on the idea that this is a campaign that only plays out in X counties in Y number of states. This is a national election, and national polling still has a value and an importance. That doesn’t take away anything from really good state polls. We all pay attention to what’s going on in Ohio or Florida or North Carolina or Virginia or Colorado at this point, and all the way through the year. But when you aggregate it up, national polling still has something to tell you, and particularly at this point in the election cycle.
HH: Dan Balz of the Washington Post, thank you so much.
End of interview.