The Battle of the Polls: A Look Behind the Influential Polling Organizations that Shape the Presidential Race

StoryOctober 27, 2004
Watch iconWatch Full Show
Media Options

A look at the methods behind the influential polling organizations that could have a major impact on November 2nd and who is behind them. We speak with Ruy Teixeira, who tracks the daily presidential polls and publishes a weekly column on the polls. [includes rush transcript]

Much of the focus of the election campaign right now is on whether next Tuesday’s vote will be fair. As we have been reporting consistently on the program, there are already widespread concerns that certain voters will again be disenfranchised or prevented from voting. The Democrats say they are deploying some 10,000 lawyers across the country to challenge any scandals that emerge. The Republicans are deploying thousands of operatives in places like Ohio, they say to ensure that everyone that votes is a legitimate voter. Some charge that the Republican effort is nothing short of a voter intimidation or suppression campaign. Meanwhile, thousands of people are volunteering to serve as Voter Protection workers, where they will spread out to districts of concern and monitor the fairness of the vote. These efforts are being organized by groups like People for the American Way, the NAACP and other civil rights organizations.

As the various factions spread their people across the country and prepare for a showdown, another battle is being fought in the media. And that is the battle of the polls. Both campaigns point to polls that show them leading the race, embracing the old addage that if you look like a winner, you’ll win. A new CNN/Gallup poll released earlier this week showed President Bush with a 5 percent lead over John Kerry. The Republicans have seized on this going into the final stretch of the campaign, with pundits citing it over and over on TV and in the papers. But who controls the polls?

Last month, took out a full page ad titled “Gallup-ing to the Right” charging that the poll was biased toward Bush.

  • Ruy Teixeira, Joint Fellow at the Center for American Progress and The Century Foundation. He edits the website Donkey Rising which tracks the daily presidential polls and publishes a weekly column on the polls titled Public Opinion Watch. He is the author of five books, including The Emerging Democratic Majority, which he co-wrote with John Judis.

Related Story

Video squareStoryMay 20, 2015Sgt. James Brown, 26, Survived Two Tours in Iraq Only to Die Begging for His Life in Texas Jail
This is a rush transcript. Copy may not be in its final form.

AMY GOODMAN: We are joined by Ruy Teixeira, a joint fellow at the Center for American Progress and the Century Foundation. He edits the website donkeyrising, which tracks the presidential polls and publishes a weekly column on the polls entitled “Public Opinion Watch.” He’s the author of five books, including The Emerging Democratic Majority, which he co-wrote with John Judis. Welcome to Democracy Now!, Ruy Teixeira.

RUY TEIXEIRA: Delighted to be here.

AMY GOODMAN: Tell us about these polls.

RUY TEIXEIRA: Well, the one you were referring to, the Gallup poll, which showed a five point lead and — for Bush — and they actually had some 13 point leads in September, these leads in particular are among what they call likely voters. And you know, you hear the phrase “likely voters” and that sounds pretty good, you know. Gee, who wants the unlikely voters, but the thing is that the way they create these samples they call likely voters is all driven by a very elaborate set of questions that tends to screen out, at least in this campaign, as tending to screen out a lot of democratic-leaning voters, that you can easily wind up with fewer minorities and young people than you would normally get in a sample of real voters on election day. So, to make a long and technical story short, the likely voters, in fact, are not particularly likely voters. The way that Gallup uses its methodology, it tends to produce these samples that under-represent constituencies in the electorate that are more likely to favor democrats than republicans. So we just — I and a lot of other people feel that the way they do their sampling, they wind up with these likely voter samples that just are not right, and then they present that to the public as representing the way voters really feel, and then they draw another likely voter sample a week later, which may change quite a bit, and then they say, “Oh, look what the voters think now.” The fact of the matter is, what they’re measuring is just their construct of likely voters and how it’s changing over time. They’re not necessarily capturing the way voters’ sentiments are really changing. So, we feel that’s quite misleading, and as you pointed out in your intro, these things really do have an effect, because they do enter the echo chamber of the media, particularly something like a Gallup poll, which still has a certain amount of prestige, still has a certain profile with a lot of people. So, when people hear Gallup poll they think, “Oh, that must be pretty reliable.” I guess the contention of myself and a lot of others is they’re not so reliable, at least anymore.

AMY GOODMAN: In the Gallup ad that MoveOn did, the full page ad, they pointed out that George Gallup, Jr., is the son of the Gallup Organization’s founder and long-time executive of the firm, a devout evangelical Christian. What is that supposed to mean, what does that have to do with anything?

RUY TEIXEIRA: Well, I think what they were trying to suggest is that, at least what I would take from this, is not so much that George Gallup and the people who work for him aren’t consciously manipulating the data, but rather that, let’s just hypothesize that their methodology works in such a way as, at least in this campaign cycle, there may be some underlying structural reasons, as well, it tends to favor republican candidates and more conservative causes. Now, if you are running the show at Gallup, and you’re probably not too interested, then, in sort of updating your methodology and changing it to deal with changing circumstances. You could be kind of happy with the way things are. And if you are a professional working for George Gallup, you’re perhaps less likely to push for change than you might otherwise be, given what you know about the predilections of your boss. I cannot prove any of that, but to some extent it’s just a given that when an organization is controlled by people with a certain agenda, things are less likely to change that perhaps need to be changed if those things actually suit that agenda. So that’s what I take from it, not conscious manipulation.

AMY GOODMAN: Ruy Teixeira, what about this whole issue of who gets polled, and the whole idea of the likely voter being a voter who voted before, and the level of voter registration, new voters who are being registered right now, where they get counted?

RUY TEIXEIRA: The nature of most likely voter models is such that it’s unlikely to pick up a lot of the new voters. I mean, a lot of the questions that people ask is about your past voting behavior. People ask about things like, do you know where your polling place is, people ask — and, of course, there’s all kinds of questions about interest level and commitment to voting, which may, in fact, be a little less among people who are newly registered. So for a variety of reasons, these questions tend to screen out the new voters, they tend to screen out young voters, they tend to some extent screen out minority voters who typically tune in very late in the process. For a variety of reasons, you know, without any necessarily conscious intent, the nature of these questions is such that you’re going to wind up screening out these kinds of voters. To the extent these voters have leanings toward a certain candidate or certain kind of political leanings, that’s consequential for how the poll turns out.

AMY GOODMAN: Do we know who are these newly registered voters? I mean, is it overwhelmingly democrat, is it equal republican and democrat?

RUY TEIXEIRA: Well, there’s some debate about that. My sense is that the registration wars have been won by the democrats, by and large. You can tell that both from looking at the states where they keep figures by registration and even where they don’t, where have the registrations been concentrated, the democratic leaning areas or republican leaning areas? For example, it’s clear that in Ohio, the democrats just completely cleaned the republicans clock in terms of new registrations. That’s one reason why the republicans are out there busily trying to challenge every newly registered voter they can find. That’s a whole other story. But I think the democrats have net out-competed the republicans on this. And then if you look at the political leanings of newly registered voters or people who will be new voters, who say they will be new voters, by and large, though there are some exceptions, most polls tend to suggest that they’re leaning fairly heavily toward the democrats this year. So this the kind of thing where these voters are not being picked up by the polls, by and large, as we said, and then, you know, the problem and then — and I think you were also alluding to this — could sort of pre-justify in a way the — Paul Krugman wrote about this in his column — if , in fact, there is a lot of disputes about newly registered voters and in fact the election is skewed by that towards the republicans, right, they can sort of point to the polls that have been taken before the election and say, “Well, look, there’s no problem with this outcome because the polls before the election said this is how it would come out.” But if the polls before the election are to some extent driven by who has been excluded from the polls before the election, then it becomes a self-justifying cycle, which I think is a very bad thing for, you know, obviously for the democrats, but for democracy in general.

AMY GOODMAN: What about who answers the polls, and specifically the issue of cell phones?

RUY TEIXEIRA: Cell phones are yet another thing that pollsters are scrambling to try to figure out how to deal with. The thing that mitigates the cell phone problem is that most people who have cell phones also have landlines. The number of pure cell phone users is relatively small, though it is growing fast. However, even if you confine your intention to that group, there is some evidence that by excluding the cell phone-only users, it is a group with a fairly distinct demographic profile which leads to a certain kind of politics. They tend to be poor, they tend to be renters. There is some evidence that excluding them from polls does skew the polls slightly.

AMY GOODMAN: And young as well?

RUY TEIXEIRA: Toward the republicans.

AMY GOODMAN: Are we talking young, as well?

RUY TEIXEIRA: Yes. These voters tend to be young, as well. Another very interesting study was done just recently by the Consumer Electronics Association, which — I’m not ready to take it to the bank yet, it needs more study, but — it’s suggested that not only are cell phones a problem, but the kind of people who are willing to answer polls who don’t, you know, who don’t screen their calls, who are likely to answer the phone and do these things actually tend to skew toward the republicans. In other words, democrats, you know, sort of even controlling for the cell phone issue, are just less likely to answer the phone and do a poll, because they’re screening their calls or for other reasons. All of that adds up to a series of response biases that, you know, in today’s world may be making the polls less reliable and skewing them at least somewhat toward the right. So, pollsters, their first instinct was denial on all of this stuff, right? But now they’re moving toward the “Well, it’s a problem, but it’s not a big one.” I think eventually, we’ll drag them kicking and screaming toward the idea it’s not only a problem, it’s actually big enough that it needs to be solved.

AMY GOODMAN: What about the position of the master republican strategist, Karl Rove, around polls?

RUY TEIXEIRA: Well, Rove is a notorious, as you mentioned in your intro, if you look like you’re winning, you’re going to win. So he loves to spin those polls. The guy who’s actually best at it is not Karl Rove, it’s Matthew Dowd, who is the chief political strategist for the campaign and a pollster. He’s just relentless in spinning those polls, in fact, a lot better than the democrats are, I think, in some ways. They very much believe in the momentum theory. If you can convince people that you’re going to win, that enhances your likelihood of winning, because it pumps up your side and it demoralizes the other side and it makes them less likely to show up at the polls. So they love these skewed polls. They just love them, love them, love them.

AMY GOODMAN: Finally, Ruy Teixeira, for listeners and viewers right now who where we’re being fed polls constantly throughout the day and night, both television, newspapers, what do you think is the most important question to ask? What kind of critical eye should we have? How do we dissect them? How does a layperson do that?

RUY TEIXEIRA: Well, gosh, in a way that’s such a big subject, I hardly know where to start. One thing is just to look at as wide a range of polls as possible, and make sure you’re not looking at outliars. So when Gallup comes in with a big Bush lead, for example, and everybody else has it close to a dead heat, well, that’s some sort of indicator. And one way you can look at as wide a range polls as possible is you plug my website, I’ll plug somebody else’s website. A guy named Bob Poulson has a site called That’s probably the best site for tracking all the latest polls, all the national polls, all the state polls. He does a great job. So my advice is to look across as wide a range of data as possible, and then for sort of extra credit, you can look at the internals of these polls, see if they have got any demographics, look at their distribution of party identification or, you know, does it seem like there’s too many republicans, and so on, so by all means, a critical eye is very much in order, and the more data that you have the better.

AMY GOODMAN: Ruy Teixeira, I want to thank you for being with us, joint fellow with the Center for American Progress. His poll is — his website is donkeyrising, which tracks the daily presidential polls.

The original content of this program is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 United States License. Please attribute legal copies of this work to Some of the work(s) that this program incorporates, however, may be separately licensed. For further information or additional permissions, contact us.

Up Next

Sgt. James Brown, 26, Survived Two Tours in Iraq Only to Die Begging for His Life in Texas Jail

Non-commercial news needs your support

We rely on contributions from our viewers and listeners to do our work.
Please do your part today.
Make a donation
Up arrowTop