Michael Marshall, moderator: Our final speaker today, I originally had him first on the list but he asked to go last as I think he was expecting his profession to be maligned in the course of the panel, is Scott Rasmussen, who is the founder and president of Rasmussen Reports, which is a polling organization. He’s been an independent public opinion pollster since 1994. He specializes in electronic polling and publishing and has been an innovator in the field. Michael Barrone, senior writer for U.S. News & World Report, called him one of America’s most innovative pollsters. His polls are covered extensively by news channels – Fox News, CNN, CNBC, MSNBC, the BBC, and other media outlets – and his company’s impact on the popular culture has been recognized by mentions on the Daily Show with Jon Stewart and a segment on The West Wing, the very popular political drama series.
He was also – he said it was a long time ago – he’s also, together with his father, a co-founder of ESPN, the cable sports network. I said, well, if I had done that, I would have stuck that near the top of my resume. But he’s moved on to other things.
There are a couple of questions which previous speakers have raised, which I’m sure he’ll address, and also tell us more about the whole polling process from a pollster’s viewpoint. Please welcome Scott Rasmussen.
Scott Rasmussen: It’s really nice to have a politician have to sit next to you and listen to your remarks.
I want to begin just with a very simple example of the interaction we’re talking about. Yesterday morning I woke up, had some new polling data come into my laptop and before the day was out there was a headline on the Drudge Report – actually before like three hours were out there was a headline on the Drudge Report saying 22 percent of Americans want Hillary to drop out of the race, 22 percent want Barack to drop out. That was the headline. By the end of the day, Hillary Clinton was saying that polling just out today showed that 62 percent of Americans don’t want either one of us to drop out and I’m going to stay in the race.
This is the kind of dynamic that is created between polling and the 24-7 news media. The candidates are trying to deal with it in new ways, and it is a serious responsibility. Yesterday was frustrating for me because when we put that story out, my focus was on the fact that a majority of Democrats weren’t ready to close down this race yet. But it instantly became in the news cycle a comparison between Clinton and Obama.
You have to have a thick skin when you’re in the polling business in this era. Politicians tend to love you when you say polls that support their position, and ignore you other times, or worse. One U.S. Senator in 2006 early in February found a poll result of ours he didn’t like. He went to the local papers and demanded that they not publish it, and talked about how bad we were and trashed us. In September our polls showed him coming back and he sent out a press release citing our work. This is the nature of the beast that we deal with.
But what happens, it’s not the poll itself that is really the problem. It is the story it becomes. Over the last week or two in the Democratic presidential race – we track it every single day – one day Hillary Clinton’s up by a point or two, the next day Barack Obama is up by a point or two. So I go on a news channel and they say, how come Barack Obama has lost ground in the last three days? And the answer is, he hasn’t. The two candidates have been bouncing around in the margin of error for the last two weeks. Nothing has really changed, but some people are looking at one part of the data and trying to come to a different story line.
It’s just as frustrating for me when we put out a poll and the story goes off on its own way, as it is for the politicians who have to deal with it, because that tends to reflect on our credibility as well.
Now as you’ve heard from a couple of different people, there are limitations and challenges to polling in today’s world and they’re very real. Mr. Wattenberg, whose work I’ve admired for a long time, didn’t begin to enumerate all the challenges and problems with polling. I mean, when you start talking about cell phones and competition for somebody’s mindset, and how do you interrupt their evening for five minutes and get them to think about at topic that they’ve never thought about before. I mean, there’s a whole series of things. We also have the fact that women answer the phone more than men, and whites answer the phone more than non-whites, and there’s older people answer the phone more than younger people. And oh, by the way, try and find a 30-year-old mom at home who will answer the phone in the evening. I mean, it is really difficult to do this in today’s world. I can go on. The list of challenges continues. Polling is a very difficult profession in today’s world and you do have some mistakes.
I think part of our problem is the potential of polling was oversold for a few decades. It is not precise, it is not perfect. We heard about New Hampshire earlier from a couple of people. Black day for all pollsters. Terrible, terrible evening when we all got it wrong. I was happy because I only projected Hillary losing by seven points. That was better than some of my competitors. But the worst part was, when I went back in the data, we showed Hillary losing by 12 points on Saturday, by 10 points on Sunday, by 7 points on Monday. We knew something was happening but the story never focused on the trend. The story focused on the absolute numbers. Since ours was among the closer numbers, it just became oh, Obama is going to win.
And by the way, this was not the same as the Bradley effect. We actually got Barack Obama’s total right. What we underestimated was the support of especially older white women for Hillary Clinton, who turned out on a 60-degree weather day in January in New Hampshire. But there were a lot of things that went on. While I didn’t enjoy living through that particular experience of New Hampshire this year, I think it was a good thing to have early in the campaign because it reminded us of the limits of what polling can do.
Overall through this campaign cycle we have seen a pretty good sense of where the race was most of the time through the polling results that had been projected before the campaigns, before the votes. Not perfect, but fairly good.
Now, when people ask about – and we get this question a lot. Are you guys influencing the race? Don’t you feel guilty that you put this candidate out or something? No. I think polling is a very good thing for the process. I believe when it is properly done and properly interpreted, polling adds a tremendous amount to the political process in today’s world. When we’re at our best as an industry, and we’re not always there, we can give voice to the voiceless, we can highlight the gaps between the political elite and the general public, we can find issues that nobody knew that were bothering people.
Here’s a couple of examples, one silly, one a little more serious. As you recall, there were a bunch of politicians all last year who thought the way to have their state influence the campaign was to move the dates up sooner and sooner so that they would all be voting shortly after Iowa or maybe before Iowa. Who knew. We actually went out and told people around the country and in Iowa, and we asked them really – and we did it in a variety of ways because no one question is perfect – but essentially we’re trying to find out, are the people of Iowa lucky to have all of these presidential candidates living with them for half a year and messing up their holiday seasons, or are they unlucky. And overwhelmingly the American people said I’m glad it’s in Iowa. They didn’t look at this as a positive benefit to influence the presidential race. All they thought about was, can you imagine having those television commercials all through Christmas, through New Years? It was a very strong reaction, and we see that every state we go to.
Right now the people in Pennsylvania are not happy that they get to have this much influence over the political process because what they see is tremendous disruption in lots of things going on in the day-to-day routine.
Now a little more serious example took place when there was a debate in this city over the topic of immigration last year. There was a bipartisan consensus in Washington, D.C. on what had to be done on the immigration issue. There was also a bipartisan consensus on what had to be done throughout the nation. Unfortunately they weren’t the same. What Americans were talking about when they debated the immigration issue was not the proper path to citizenship, or what the right amnesty provisions were. The American people did not trust the federal government to enforce any laws that were passed. It was a lack of trust that blocked the issue to fail in the U.S. Senate, and by the way, it is the same issue that is causing problems for the candidates today when they talk about immigration.
Another issue that always shows at the top of the polls in terms of voter interest in this election cycle is government ethics and corruption, but it’s not being talked about all that much. And by the way, when voters talk about ethics and corruption, they’re not talking about passing new campaign finance laws. It is a much deeper corruption they see. Voters actually believe that having a friendly reporter is more valuable than having a wealthy campaign donor, and they believe that reporters in the media are part of a corrupt process. When we polled about Jack Abramof, and he went to prison, only 15 percent of Americans believed he was doing anything different than what normal lobbyists do. This was their perception.
Now government ethics and corruption is not being talked about much on the campaign trail and neither is immigration because they both reflect an anger at Washington, and the people in Washington are guiding this debate. So part of what a poll can do at its best is bring up some uncomfortable issues. In that regard – I’m sure Michael has felt this in the past – a pollster is a little bit like – I’m old enough that when I go to the doctor I get bad news. He tells me what I should eat, or how I should exercise, or all kinds of things, stuff I don’t want to hear. When a pollster does good work on a political race, it usually tells the candidates something they don’t want to hear.
I know without a doubt that when you were running for the U.S. Senate you liked Survey USA polls better than mine because they showed you very close to winning.
Michael Steele: Actually I’m weird that way. No, actually, I’d like to know the other side of it.
Scott Rasmussen: But there is this sense that you’re providing some information that people don’t want to hear. That’s the second great value of polls. You tell not only candidates but you tell reporters something about the state of the race, and you tell it to them in a way that limits their wiggle room. A reporter has a hard time writing about this story if there are facts that get in the way.
In 2002 there was George Bush campaigned throughout the fall, really tried to build support for his position on Iraq. Shortly after he won, or the Republicans won control of the U.S. Senate because of his campaigning, the U.N. Security Council knew how to read election returns and they unanimously endorsed Bush’s position on Iraq.
One reporter wrote a story and said, because of what’s happened, because of the U.N. Security Council’s action, perhaps public support for the situation in Iraq will improve. Well, the truth was, we had done polling for several days before and several days after that U.N. Security Council resolution. It had absolutely no impact on public opinion about Iraq. The only thing that it did was that it made Americans less pessimistic about the United Nations because the Security Council was going along with where the U.S. position was at that point in time. Once that polling data was out there, it became a little bit difficult to present the story in a different way.
I tend to look at this as a positive good because a candidate has a difficult time and a journalist has a difficult time moving on in something else.
Finally, this whole notion that polling does impact the process is absolutely true. It does have an impact. It’s not as if a pollster sits there and tries to figure out who’s going to win or lose. I don’t know how you do that. Our view is, if our final poll this year shows John McCain winning by three points, I want John McCain to win by three points. If it shows him losing to anybody by three points, I want him to lose by three points because that’s our job is to be as accurate as we can be.
But we do know that there is a strong interaction. We know that in the primary season this year voters will go into the polls being aware of what the polls said and what the strategic impact of their vote would be. When they were going to vote in Florida in the Republican primary, they knew that Mike Huckabee and Rudy Giuliani were fading, so the choice was between two other people and they made their decision accordingly. We saw lower tier candidates losing support in the final days.
We also know that when you begin to cover the race, you affect – when you cover it as a pollster, you affect the way the reporters write about the race. The Democratic presidential primary right now, if there was no polling out there, we’d probably be reading a whole bunch of really boring issue analysis things about the slight differences between Hillary Clinton and Barack Obama that were put together by people, very smart people at the Brookings Institute and other places, and we would be totally ignoring the fact that there are strong demographic alliances that are supporting each of those candidates. That has been the enduring story of this campaign, was the demographic constituencies that are out there. In fact, after Super Tuesday you could look ahead and project just about what was going to happen the rest of the way out because you knew the demographics of the various states.
Now that has changed reporting. It has brought the subject of race and gender and other topics that are maybe uncomfortable to talk about, but it has brought them front and center in the Democratic campaign. That may be – some might say that’s influencing the race inappropriately, but it’s really making it more transparent. It is letting people know what is really going on. And ultimately when polling shines a light on some problems with the political process, it’s not always the poll that’s the problem. We have a political process that is badly broken. Only once in the last generation has a presidential candidate managed to win a decisive majority, and that was Ronald Reagan in 1984. Before that it used to happen all of the time.
I think part of the problem is our leaders have not found a way to resonate – to come up with a message that resonates beyond their own particular narrow interest group.
Again, I’m not going to say that polling is perfect. We’ve made our mistakes, and we’ll make many more before this election year is over. We recognize there is an impact on the process. We try to be faithful to the overall flavor of the story more than of getting hung up on a specific number. And there are lots of things we have to do going forward to change the way we gather information from the public. We’re experimenting with them right now, but bottom line is, I wouldn’t be doing what I’m doing if I didn’t think it was good and healthy for the political process. Thank you.
Michael Marshall: Thank you very much, Scott. Very interesting overview from a pollster’s perspective, and reminding us that it’s not the poll that is the issue as much as the story that it becomes, and a lot of other very interesting insights. Perhaps he’s left us all with a little more sympathy for pollsters and the tough row that they have to hoe.
SPEAKER BIOGRAPHY
Scott W. Rasmussen
Scott Rasmussen is the founder and President of Rasmussen Reports. He has been an independent public opinion pollster since 1994. Rasmussen Reports is an electronic publishing firm specializing in the collection, publication, and distribution of public opinion polling information.
The Company’s flagship publication is RasmussenReports.com. During Election 2004 and again in Election 2006, the site attracted more than twice as many visitors as its nearest direct competitor (Gallup.com) peaking at 1.8 million daily page views.
Mr. Rasmussen has developed a reputation for delivering reliable, actionable public opinion data. His work has been cited as among the most accurate in the nation by many including Slate magazine, RealClearPolitics.com, and noted political commentator and Larry Sabato.
Rasmussen polls are covered extensively Fox News, CNN, CNBC, the BBC, and other media outlets. The Company’s popular culture presence has been highlighted by mentions on The Daily Show with Jon Stewart and a segment on the West Wing.
Scott has pioneered the use of automated technology to conduct public opinion polling and developed new research techniques to not only measure current public opinion but also to gain better understanding of what events or actions actually change public opinion. He has been called “one of America’s most innovative pollsters” by Michael Barone, Senior Writer of US News & World Report.
While Rasmussen Reports is most noted for its political polling, the Company also provides a variety of economic tracking indicators including the Rasmussen Consumer Index, a daily measure of consumer confidence, and other monitors in conjunction with Discover Card, the Hudson-Highland Group, and Country Insurance.
Scott grew up in the broadcast business. Earlier in his career, he and his father founded ESPN, the cable sports network.