Charles Cook is editor of the Cook Political Report and a political analyst for the National Journal and CNN. William Saletan is a Slate senior writer. Saletan penned this "Frame Game" arguing that "every campaign poll that asks about an opponent's flaws is a push poll," and that "real polls" can be just as invidious. In response, Cook posted this message  in "The Fray," Slate's reader feedback forum. Slate has asked them to continue their discussion about the merits and perils of campaign polling in this "Dialogue."   

Dear Will,

First, I think it is very, very important that negative phone banks or advocacy phoning be distinguished from polling. That's why I refuse to even use the term "push polling." One is advocacy, persuasion, just the same as a TV or radio advertisement; the other is intended to learn something, to test an argument to see whether it works or not. I feel very strongly that the media should never use the term "push polling," because it fails to distinguish between a very small number of calls intended to learn something from a large number of calls intended to persuade voters. Stu Rothenberg's piece in Roll Call, which you refer to, goes into that point quite well.

In terms of the use of push questions in polls, what the mass media fail to understand is that these questions, quite often, are used to test the vulnerabilities of the candidates that sponsor the polls. Am I more vulnerable to an attack on this point or that one? Which should I be more worried about? In my previous life, as a campaign consultant back in 1980, I worked for a congressional incumbent who represented a very dangerous district, one that should be represented by someone from the other party. My client always polled strongly early and midway through the race; then invariably, the race would tighten and he would just barely win. So in April or May of the election year, far out from the general election, on the next-to-last substantive (non-demographic) question in the survey, we beat up our candidate badly, then built up our opponent, then asked the trial-heat question one final time. The race went from a runaway with the early question to very, very close with the final one. We then were able to look at the political and demographic attributes of those that defected from us to either undecided or to our opponent, so we could see whom we should focus the rest of the campaign on. We knew what kinds of people they were, even their media habits. My client won by one of his biggest margins in a very tough year. This is a classic use of push questions, even though we "pushed" against our own candidate. That is very different from running a phone bank trashing the opponent.

In terms of negative phone banks, perhaps there should be additional legislation requiring clear disclaimers as to who paid for an advocacy phone call, so there would be some accountability for the substance of these calls. That's fine, but it should not affect legitimate survey research efforts.

This whole episode was triggered by the allegations of a South Carolina woman who spoke up at a McCain rally, saying that her 14-year-old son was in tears, that he was told over the phone that his hero, John McCain, was a liar and crook, or something to that effect. She said her son was push-polled. Putting aside the question of where she learned about push polls, this whole story was quite suspicious. I have never heard of a campaign that polled 14-year-olds. The first question in every political survey I ever heard of is "Are you 18 years of age and registered to vote at your current address?" Why would a pollster bother to interview or, for that matter, attempt to persuade a 14-year-old? Sounds to me like either the kid lied and said he was a registered voter or this never occurred, at least as it was stated.

I know well and am very familiar with both of George W. Bush's pollsters and, for that matter, John McCain's as well. I serious doubt either one would have interviewed a 14-year-old for a presidential primary. I kind of doubt that even a negative/advocacy phone bank would have either. That's what smelled. Then the media took the leap and got into the whole push-polling business. Some handled the story very responsibly, for example the Los Angeles Times, others more ham-handedly, like the New York Times, and others just practiced the standard pack journalism, without bothering to look beneath the surface and examine the practice.

Your piece suggested that campaign polls may be sleazy. Unaddressed is the fact that campaign polls—that is, those that are conducted by respected professionals—are very high-quality survey instruments, equal to or, in some cases, of higher quality than those of the national network surveys. Where there is a problem is many local media-sponsored polls, often shoddily done, sometimes by classes of college students who may or may not be making the calls or introducing interviewer bias. Some of the media-sponsored polls we see coming through our office every day are sloppy and/or methodologically unsound, offering less reliable data than those produced by campaign pollster.

There is considerable skepticism regarding campaign-sponsored polls; some think they are inherently tainted. While that is sometimes the case, most likely what happens is that when campaigns get unfavorable numbers, the data never sees the light of day. When they get numbers that are favorable and may help a campaign's credibility, and perhaps fund raising, they are quick to release that data. Or sometimes the data is selectively released—only the good parts of polls. That's what we have to deal with, with some sleazy pollsters and others of the highest integrity.

You ask whether campaign polls are sleazy: My answer would be sometimes, but we find that in every profession. The bottom line is that campaigns are about trying to win. I frankly have no ethical problems with pollsters asking push questions, to test arguments that they may later use on television, radio, or via direct mail. I do have some problems with negative/advocacy phoning, as it isn't as clear who is calling, what is paying for it, and general accountability about the content of the call.

Frankly though, in the grand scheme of things, I find none of this as troubling as where another part of the political-consulting profession has gone, opposition research. I once sat in on a consultant's professional meeting in which someone on a panel spoke of the importance of obtaining the opponent's date of birth, Social Security number, and mother's maiden name, which made it much easier to obtain the opponents medical and credit records. It is now becoming commonplace that private detective agencies are hired to look into opponents' backgrounds, rather than just campaign consultants looking into opponents' voting records. Frankly, I find that far more appalling than whether pollsters are testing arguments with 300 or 400 voters out of hundreds of thousands in a state or congressional district.

Let's try to get some perspective about this whole thing. Thanks.