Cell Phones Affecting Accuracy of Political Polls?

This is a partial transcript of "Special Report With Brit Hume," Oct. 21 2004, that has been edited for clarity.

Watch "Special Report With Brit Hume" weeknights at 6 p.m. ET!


BOB BECKEL, DEMOCRATIC STRATEGIST: The government and the FCC are going to have to let pollsters get through to cell phone users, because you can't — you simply can't, when tens of thousands of people every month are opting out of land lines, you can't leave that segment out of polling and expect to have anything accurate.


BRIT HUME, HOST: My old friend, Bob Beckel (search) is expressing something that is now part of the buzz in Washington this year, that you really can't trust this year's polls because millions of young people now use only cell phones. And pollsters simply can't reach them, thereby, potentially skewing the results of the polling. But is it true?

Well, who better to ask than John McLaughlin (search), a pollster who numbers among his clients The Wireless Association, a trade group representing the cell phone industry.

John, thanks for join us from our studios up there in New York. What about this challenge to pollsters? It makes all kinds of sense that you've got a lot of young people who only have cell phones, no land lines, they number apparently in the millions. You can tell me whether that's right or not.


HUME: And you can't reach them. How much of a problem?

MCLAUGHLIN: It's a problem because, for those of us in the industry, it makes the poll more expensive, which our clients don't like. But if you're going it right, you have to take the time to make sure that the calls are done right. And you have to make sure that you're getting enough people. Because what you'll have now is you'll have these younger voters one in seven; one in eight don't have a hard line at all. But they tend to be more upscale and they tend to — they tend to be harder to find anyway. But now in this election they're going to vote.

We also do polling for MTV's "Choose or Lose" program, and you see now there's a greater interest in younger voters.

HUME: Well, I know. But how do you compensate for the fact that there's a class of voter, you note that they're upscale and mostly young, that you can't reach because they're using ago cell phone? Can you reach enough other people who have the same profile that your samples are valid? How do you deal with that?

MCLAUGHLIN: Right. And you can — you have to. And you have to at the same time make sure it's random enough that you're capturing the right audience within that. But what you'll normally do is you can set a model based on likely voter turnout. And you can set that model based on past history, plus projected trends.

HUME: What does that mean? Set a model based on voter turnout. I don't understand. I'm still puzzled about how you compensate for the fact you've got these people out there who are not going to answer any calls, because they don't have phones like the ones you can get through to.

MCLAUGHLIN: You have to make enough calls to other people that have landlines. And also from time to time we buy — you can buy samples of voters who have put their numbers on the list. And these voters will put down their cell phone. And then you call them and you saying, ask them how they vote. And they say how did you get the number? Well, they put it on their voter record. So that's in a rare case. A lot don't have the number listed.

And you have to make enough calls to capture people that are similar types of voters. So that when you look at the survey, the quality controls know that the survey is right: that you have enough young voters in it, that you have minorities in it, that you have enough voters who definitely came through as a likely voter to reflect the November turnout.

HUME: OK. Now, so I take it then there's not any evidence that suggests that these cell phones exclusively — exclusive cell phone users don't have other phones are going to vote in any particularly different pattern, than other people who fit the same profile, but do have landlines? Is that the idea?

MCLAUGHLIN: Right. But I think, as you've shown in the clip before, Bob Beckel is right. In time this is going to be a greater segment. It's going to grow in population; it's going to make it harder to poll. And eventually we're going to have to poll over the Internet probably there, as it becomes part of your e-mail people regularly get.

HUME: All right. People are accustomed now to seeing night after night, surveys of what are called "likely voters." And these are in contrast with surveys of people who are merely registered voters. How does a pollster know when it's a registered — merely a registered voter on the phone? Which I guess, in order to be a likely voter, you've got to be. In other words, what do you ask to determine if someone is a likely voter?

MCLAUGHLIN: Well, our company has a fairly loose screen. We just ask them — we ask them how likely they are to vote in this November election for president, very likely or somewhat likely? And if they say they're not likely at all, we don't continue that interview, unless we're trying to get unlikely voters to come out and vote, which we've done in other elections.

HUME: I understand. But in other words, you simply ask the question, are you a likely voter?


HUME: Your company does that. And if the person says yes, you go ahead and include them in the likely voter category?

MCLAUGHLIN: Right. So We have a rather broad screen. But on the other...

HUME: Yes. That's pretty easy. That's a pretty easy test to pass. All they got to do is tell you.

MCLAUGHLIN: Right. And...

HUME: What do other pollsters do?

MCLAUGHLIN: Others, they have degrees of waiting where they'll ask six, seven questions.

HUME: Like what?

MCLAUGHLIN: A scale of one to 100, how likely are you to vote? And they may take 70 percent likely to vote. They may set a different arbitrary scale that they have a history of that they measure success in their firm.

Now, what's interesting is the ones who tend to screen tougher usually are more favorable to Republicans. Because in a smaller turnout those audiences — those voter segments tend to skew Republican. And that's why we did well in the midterm elections. But this election, it looks like you're going to have — 2000 we had 105 million people vote. This election looks like it's going to be higher, much higher; maybe 110, 115 million voters come out.

HUME: So one needs to be careful about being too picky with your likely voter screens, because a lot of ordinary registered voters this time are going to turn out to be actual voters. And therefore, belong in the likely voter category?

MCLAUGHLIN: Right. And the greatest impetus for them to vote is to think that their vote counts. And the more you have all these polls within the margin of error, saying it's too close to call, the more likely you'll have a higher turnout; just like you did on the last weekend of the 2000 election when it was so close.

HUME: Yes. I noticed that Florida, of course, was the place where it was closest of all. And the likely voter — the polls of likely voters were wrong. They showed Bush with a comfortable lead. He didn't have a comfortable lead, because it looked like to me all of the registered voters showed up.

MCLAUGHLIN: Well, there's also the news cycle factor. That night in Florida on Thursday when the DWI story hit, Bush went from a five-point lead in Florida in one night, to a 40-40 — 44 to 44, in our polls dead heat and it stayed that way through the weekend. So that increased voter turnout. So...

HUME: Got you. John McLaughlin, very informative. Thank you very much. Always good to have you.

MCLAUGHLIN: Thank you.

Content and Programming Copyright 2004 Fox News Network, L.L.C. ALL RIGHTS RESERVED. Transcription Copyright 2004 eMediaMillWorks, Inc. (f/k/a Federal Document Clearing House, Inc.), which takes sole responsibility for the accuracy of the transcription. ALL RIGHTS RESERVED. No license is granted to the user of this material except for the user's personal or internal use and, in such case, only one copy may be printed, nor shall user use any material for commercial purposes or in any fashion that may infringe upon Fox News Network, L.L.C.'s and eMediaMillWorks, Inc.'s copyrights or other proprietary rights or interests in the material. This is not a legal transcript for purposes of litigation.