Are Americans for or against war? It all depends on how you ask the question

By Eric Quiñones

Princeton NJ -- Less than a week into the U.S.- Iraq conflict, three of four Americans responding to a University of Maryland poll said they supported President Bush's decision to go to war. Yet two-thirds of respondents in the same poll opposed the idea of future American military action without U.N. backing.

Adam Berinsky
Such results underscore the challenge for news outlets scrambling to gauge how Americans feel about the war in Iraq. Adam Berinsky, an assistant professor of politics, is watching closely as part of his research on public opinion. Berinsky is the author of "Silent Voices: Public Opinion and Political Representation in America," which examines biases in polling processes and is slated for publication by Princeton University Press early next year.

Berinsky, who joined the Princeton faculty in 1999, spoke with the Princeton Weekly Bulletin about sentiment on the Iraq war and about how public opinion is collected and reported.

What are your impressions of public opinion reporting in the media since the start of the war?

Going into the conflict, I expected that as soon as troops went into Iraq, there would be a large bump in support for the president and for the military intervention. We saw this kind of bump in the first Gulf War and throughout the last 50 years. And indeed it looks like that's what happened.

I would be skeptical of how long that's going to last. Already there's some discussion of war weariness. Looking at the longer-term dynamics, the key question is how long that increase in support lasts -- does that bump dissipate quickly or does it fade more slowly? There's going to be a lot of initial rallying around the president, but I think that could dissipate pretty quickly as people look toward the future.

What types of biases have you found in researching public opinion polling and reporting?

If there's one thing that we've learned in 60 years of public opinion research, it's that the way that you ask the questions very much affects the kinds of answers that people give. For example, with the Iraq war, if you mention the possibility of severe American casualties or if you mention inflicting damage on Iraqi civilians, support for the war will drop. But if you talk about Bush mentioning links between Iraq and al Qaeda, support is going to rise. The one lesson is that people really need to be skeptical of any claim that a certain percentage of people support a war or oppose a war, because questions always are asked in context. Politicians or those who want to make a case for their side are going to pull that figure out of the context in which it was asked to provide support for their position.

I'm interested in how well public opinion polls broadly represent the voice of the people. In my book, I note that political participation does a good job of representing intense interests, but not a great job of representing everyone equally. That's where public opinion polls come in. We know that public opinion polls might be lacking, but many scholars argue that the polls do a really good job of ensuring that everyone is heard, balancing the biases of political participation.

On the other hand, I look at people who say they don't know where they stand on the issues of the day, and I've noticed in a lot of the polls these last few months leading up to war that there haven't been a lot of those responses. Often professional survey organizations, especially the ones working for the media, won't offer "don't know" as an option. So I think you see mixed signals in a lot of these polls -- people aren't comfortable with the war, but they say now that we've started on that path, we might as well go there.

Are a lot of voices missing in the public opinion polls we see in the media?

There is the question of the people who would answer "don't know" and also the people who don't get interviewed at all. Lately I've seen these flash polls that only show one day of results, saying: "This is what America thinks." They called people one night, and anyone who wasn't home or didn't answer the phone wasn't included in the sample. Generally, polling organizations should call back over several nights because the kinds of people who might not be at home on a given night may be systematically different from those who are. It tends to be older people who are home at night and are more likely to answer the call, so you get kind of a skewed result there. You really need time to get your proper sample.

Another issue is these call-in or online polls. It's similar to political participation because you're getting people with more extreme views who are taking the time to call in or log onto Web sites to have their opinions heard. Fox News or CNN or MSNBC will show results from an online poll, which is basically garbage, because all it's showing you is the opinions of the people who write in.

Does reporting of these poll results, in turn, influence public opinion itself?

There's a line of research that says that people want to jump on the bandwagon with the majority, which I don't believe. Public opinion polls do give people a way to contextualize their personal experiences. For example, let's say you've lost your job. If you see that a public opinion poll says many people think the economy's going in the wrong direction, that gives you a way to say, "It's not just me, it's this broader experience out there." But I don't think it has a direct effect on how people think.

A month before the war started, President Bush generated some controversy when he responded to anti-war protests by saying he would not "decide policy based upon a focus group." Public opinion research, however, clearly is important to politicians -- how do they make use of it?

There's this notion you get from Bush's comment that politicians commission polls or read the newspaper to see what Americans want, and then just go whichever way public opinion tells them to go. That isn't quite true. What politicians often try to do is pitch their message in a way that resonates with the kinds of underlying opinions people have.

Let's say I represent George Bush's 2000 campaign and I'm polling about Al Gore. I might say, "If you knew Al Gore claimed to invent the Internet, does this make you less trusting of him? More trusting?" I'm going to try to poke and prod and see what the soft spot of my opponent is. And I'll do it to myself, too, to see where my opponent might attack me. It's not so much trying to follow public opinion as to see how can I best pitch my policy, given where I stand and what I believe.

It's very interesting that Bush made that comment. He likes to say he doesn't look at polls, but actually he's funding a polling and focus group operation that's not as big as Clinton's, but still is significant in size.


  [an error occurred while processing this directive]