How do voters choose their candidates? How do they process all the political information that they are bombarded with so they can make intelligent choices during elections like next week's primaries? No one knows everything about how voters think and act, but Richard R. Lau, a politics professor at Rutgers, has at least tried to find out. Based on research from experiments with about 700 people, Lau and David Redlawsk wrote "How Voters Decide: Information Processing in Election Campaigns," a 2006 book definitely not aimed at casual readers. I recently spoke with Professor Lau by telephone from the Rutgers campus as he was grading final exams:
Q: Are voters usually rational when they choose a candidate?
A: Well, that depends on what you mean by "rational." Can voters give you a reason for why they did what they did? Yes, absolutely. A more formal economic definition of rationality is ... to very actively and conscientiously consider the consequences of the different alternatives for your own well-being, however you want to define that, and, in this case, vote for the candidate that maximizes your self-interest, however you want to define that. No. Not very many people do that.
Q: What kinds of information or forms of persuasion are voters most likely to be influenced by?
A: They're most likely to be influenced by two things. To the extent that they have strong prior political beliefs -- whether they're Democrats or Republicans or liberals or conservatives -- they really are going to see things in light of their own backgrounds, which is the nice way to say it, or their own biases, which is the less nice way to say it. If you’re a Republican and Dick Cheney tells you something, you are a lot more likely to believe it than if you are a Democrat -- and then you’ll listen to Hillary Clinton. That’s one very big factor. The other factor -- which those of us in political science tend to overlook a lot because we often don’t have the evidence -- is people that you talk to: your friends, your family, your neighbors. If somebody you trust says, "That Mitt Romney is a real jerk," then it is going to be hard for you -- particularly if you don't know enough about (Romney) to counter that information or argue against it -- to reject that person's statement.
Q: Is there anything that is least likely to sway a voter?
A: I don't think very many people are persuaded by detailed policy analysis or arguments or things like that because most people don't have the interest to pay attention to those things and/or the contextual background to make sense of the arguments that a policy wonk would make.
Q: When you did your research, did you look at the effect of TV political ads, fliers, print ads, newspaper editorials? Was that part of your research?
A: We mostly were looking at information that people were selecting themselves, as if they were scanning the headlines of a newspaper, and there’s just 15,000 more things than you could possibly look at, so you’ve got to just choose for yourself what you’re going to learn about these different candidates.
Q: What's the most important thing about voters that you discovered while researching your book?
A: Please do not pose this as though I'm the first one to have found this ... but one of the clear things from our research is that the people who look at the most information, who gather the most information, are not necessarily the people who are going to best be able to determine which candidate is best for them. Really people often do better with little information than with a lot of information. We had two different types of voters who we talked about. One type was what we called "fast and frugal voters." Another we called "cognitive misers" or "heuristic-based voters," who'd look for specific cues and go for them.
A: Heuristic is like a shortcut. It's a term that is used very much in cognitive psychology. For example, a party is probably the most important heuristic that we have in politics. If you learn that someone is a Republican and if you have a well-developed party schema or party stereotype, you don't have to go out and learn this person's positions on abortion or defense spending or gay marriage because you can assume – not perfectly; heuristics are not perfect; they’re shortcuts; sometimes they lead you astray --that a Democrat has one set of opinions and a Republican has a different set of opinions.
Q: You're looking at this as voters looking out for their own self-interest, right?
A: This is the big innovation in our work: We do actually try to define which candidate is best for each voter. We call this "correct voting." What we mean by "correct voting" is choosing the candidate under conditions of incomplete information that you would have chose had you had complete information about all of the candidates. So of course nobody in a real election has anywhere close to complete information.
Q: Is there an easy way to characterize the typical American voter and how he or she makes voting decisions?
A: No. I would not do that. We came up with these four different models or types, if you will, of voters or decision-making strategies. They are all evident in our experiments, and I'm sure there is evidence in real elections. One of those types are the "rational voters" -- the people who seek out a whole lot of information and try to learn as much as they can about all candidates that they have that are available to them. A second type of voter is what we called "confirmatory voters" or "confirmatory decision strategies." These are the Democrats or Republicans who know that they are going to like the Democrat or Republican the best and kind of guide their information gathering and perceptions of the information they see in a partisan manner.
Q: The "crank-pullers."
A: Yeah. And then the other two types are the "fast and frugal" folks who might say, "The only thing I care about is somebody's policy to combat terrorism or something, and so I'm going to learn how all the candidates think about that and then vote for the one that I think is best." The fourth type are these more heuristic-based voters who will look for some cues. Maybe party. Maybe ideology. Maybe endorsements from known groups -- like the National Organization for Women.
A: There were pictures of these people in our experiments, but they were all kind of your standard 50- or 60-year-old people.
Q: No Arnold Schwarzenegger or Jack Kennedys?
A: Intentionally we avoided any familiar figures. We didn’t have any really pretty people or handsome people or ugly people. We didn’t do that. We could. We intentionally did not do this in our experiments, but this could be another big heuristic cue -- familiarity. It’s like, “I know President Bush. I'm going to vote for him because he's doing an all-right job or against him because he's doing a lousy job."
Q: Is there anything that you now know about voters and their behavior that most of us do not know or will never know but probably should?
A: Oh boy, I doubt that. The one thing I would reiterate -- because in a way it is counter to our intuitions -- is the idea that many people don't want and most people can't deal very well with a lot of information. We have evidence that people make better choices when they have less information.
Again, this is kind of an Econ 1 or something -- that you want to provide consumers with as much information as possible and then they’ll make the best choices that they can make. I think a lot of policy in Washington is sort of based on that idea. We think we’re going to help the health care system by providing consumers with 18 million different bits of information about different plans, and doctors and availability and quality. But people are not going to look at this stuff. It’s not helpful to people.
Q: Do most politicians know what you know?
A: Ahhhhhh. Yeah, I would say intuitively they do. They probably wouldn’t lay it out the way I do. But I don’t think most politicians think most people are consciously really seriously paying attention to what they are saying and what their opponents are saying.
Q: It’s almost like you’re saying “less information is more.”
A: “Less information is better." That is often the case.