Q: You suggest there are several stages in the way public opinion forms and solidifies. Where are we now with regard to public opinion on the war on terrorism?

A: The first stage — opinion formation — is when people's consciousness is raised and they develop a sense of urgency about an issue. Consciousness has been raised and a sense of urgency has been stimulated, so we're through that stage.

The second phase — working through — is more complicated. In it, people work through the issue, confront tradeoffs, and wrestle with conflicts and values. We're in the early part of that stage with respect to terrorism. On the one hand, we want to strike back, but we're finding it difficult to strike back. We don't want revenge, but we don't want to be perceived as wimps. We don't want to start a holy war but we want to make sure we can't be pushed around this way. We don't know the full magnitude of the threat. We haven't really wrestled with how much sacrifice we're willing to make.

In the final stage, you have resolution, where people have made up their minds, where people have made judgments with their heads and also their hearts.

Q: Are people's responses even more volatile because the terrorist attack took place in the United States, and because of the outbreak of anthrax?

A: Whoever decided to send anthrax to the media was very shrewd because that was a guarantee that it would be played up incessantly, and it's being played up and people have now raised the question of whether it has stimulated an overreaction. That sense of insecurity and nervousness about anthrax is extremely volatile.

I feel that the American people's response to this threat has been so praiseworthy. I think it's so admirable because there has not been hysteria. People are not bloodthirsty. They aren't running around putting pressure on the administration to nuke 'em or wage unlimited war. People are very thoughtful about the threat that comes from playing into Osama bin Laden's hands in making this a religious issue about Islam. And there's been a wonderful maturity and sense of judgment and discussion that helped to keep the administration from going off the deep end, and the administration helped the public from going off the deep end. It worked both ways.

Q: Are Americans hawks with regard to foreign policy, or do they at least lean in that direction?

A: The tendency on the part of the country is to be hawkish. Dovishness is not our first impulse. If things get painful enough, though, we begin to rethink that position.

Q: Are you surprised by the strong public support for anti-terrorism measures that might impinge on civil liberties?

A: No. I think the single most powerful political emotion is fear of disorder and instability. It transcends everything. These attacks are a threat to order. When you have threat to order and stability, the country will sacrifice. They'll try not to, and they'll try to minimize that sacrifice, but they will do it.

In this case, I think there'll be a lot more discussion about an identity card. My guess is we'll end up adopting an identity card. In the past, the identity card has been perceived as something that gives government too much intrusiveness into privacy. There will be a greater willingness to give up some aspects of privacy in the interest of greater security.

Q: People have been asked in surveys whether they are more suspicious of Arab Americans since the attack on the World Trade Center. Do people answer those questions honestly or are they afraid of how they'll be perceived?

A: People hesitate in a poll to have answers that are not socially acceptable. It's not that they're lying. It's a little self-deception and politeness.

Q: President Bush's approval ratings are at historically high levels. Is it reasonable to expect that he can sustain those high approval ratings?

A: They are at unnatural levels. They are at unsustainable levels, but that doesn't mean that there will be a big fall. If the president went from a 90 percent-type rating down as far as 60 to 65 percent, he'd still be doing extraordinarily well. He isn't necessarily going to fall to a low level unless he does something that undermines his credibility, which is what happened with the elder Bush on the economy.

It's almost inevitable we'll be going back to being partisan, probably in the next couple weeks. The economic stimulus bill in Congress is the first sign of real partisanship. Once you break through the partisan honeymoon, the ratings are going to go down. They're not going to be destroyed, but they're going to start moving to more natural levels.

Q: Trust in government fell after Watergate and Vietnam. But a recent ABC News/Washington Post survey found 64 percent saying they trust the government to do the right thing most of the time or just about always. Do you think this will hold steady, and are there any other signs that government is regaining the public's trust?

A: The high point was reached in the early 1960s where people felt 2-to-1 that you can trust the government to do the right thing most of the time. And then it reached its low point in the early to mid-1990s when it was 2-1 the other way. That's a total reversal over a period of 40 years. The level of trust began to climb back before Sept. 11, going from the 30s to the 40s. The recent jump up to the 60s is artificial and volatile. I think the trend has shifted toward increasing trust in government, but it's not going to go back to the old 2-1, at least not for years and years.

Q: In the 1980s, you pioneered a theory called "the mushiness index," a series of questions asked to determine how firm the public's position might be on a certain issue. Can you discuss the index and offer some ideas as to why it didn't catch on?

A: There's no way you can tell by looking at a poll finding whether it's firm or stable or whether it's unstable or volatile. You could have a very tiny error due to sampling, but you could also have a massive error due to a bad question or people giving a top-of-the-head answer without giving the matter very much thought.

For example, when the Clinton health care plan was first introduced, I examined some 17 different polls that came to an average of 71 percent in favor of the plan. In probing questions we did to determine how soft the support was, we came to the conclusion the 71 percent was really 30 to 35 percent. By the time the Clinton effort was voted down, it was clear to everyone that the public support had eroded. But that had really been clear from the very beginning because when you asked people, "Do you favor a universal health insurance for everybody as proposed by the Clinton health plan," more than 70 percent said, "Sure." But when you asked probing questions — "Supposing you had to accept certain inconveniences or the cost to you went up," or "Supposing there was some compromise in the quality of care" — as soon as people began to think about it that way, that's when the support plunged from majority to minority support. Millions of people, including Clinton, were misled by those sampling issues.

We felt we could ask a few simple questions in conjunction with the poll and then qualify our results or put an asterisk beside the number to indicate people should not take it at face value. It was a little extra time and trouble to ask a few extra questions. But when some of the journalists started using the index, they found it was awkward to explain, and that it added a little to the expense, so they just never bothered. The trend at the time was to go toward quicker and cheaper and simpler polls and soundbite-type polls. This was going in the opposite direction. So it was a wonderful innovation but never used.

The fact it was brushed aside and never used was a great disappointment. It's a symptom of a state of mind. It's a clash of values between media values and polling values. In the clash of values, the media has won, and their values have prevailed.

Q: So where does the "mushiness index" fit in with regard to people's opinions of the war on terrorism and the anthrax outbreak at home?

A: On the terrorism stuff, everybody in the profession knows you don't want an instant question after some national disaster because there's an initial puff of fury, anger or despair, and the puff passes. So if you get a reading in an instant poll, it's "Nuke 'em." A couple days later there's a different response. It's notorious that grieving goes through many stages — anger, despair, sadness — and the country has been going through a shock and grieving response. So on a lot of the terrorism questions, the key is to not take the findings literally until the mood has settled in, which I think could take a matter of months. That said, I think you could surmise that some attitudes are going to be mushier than others.

The issue is: What will happen if there are a lot of body bags? It could be similar to Vietnam. At a certain point, people will raise legitimate questions about the strategy. Is this strategy the best strategy — not the objective, but the strategy? The questions about strategy might also be raised without a lot of American body bags if, say, bombing were to go on through Ramadan, if there were mounting civilian casualties, if there were signs it was destabilizing Pakistan and feeding into Osama bin Laden's purpose of creating solidarity against us throughout the Muslim world. Then there would be questions about things there aren't questions about now.

Q: It seems as if there are a lot more surveys these days, particularly with the proliferation of cable. Is the public well served or less served by all these surveys?

A: Much less well served. The notion that a poll is a poll is a poll, and it doesn't matter how cheap and simple-minded it is, is such a deterioration. It's Gresham's Law of junk driving out quality. They couldn't be more misleading.

Q: What would be your advice to people who are reading about surveys on the war on terrorism?

A: For readers, I'd say the single most useful thing to keep in mind is this notion of mushiness and volatility, that 75 percent may mean 75 percent, but it also may mean 25 percent.