F you are reading this column, it is clear that you are interested in economics and economic policy. But what about the rest of the population? How do people learn about the economy and economic policy? How much do they know? How does it affect their views?

These are questions that my Princeton colleague Alan S. Blinder and I ask in a study just completed for the Brookings Papers on Economic Activity. Last spring we surveyed a random sample of 1,002 people on their views about economic policy. As part of the survey, we administered a quiz on basic facts, like the size of the federal budget deficit, the level of the minimum wage, the share of income paid in taxes, and the percent of Americans without health insurance.

The good news is that three-quarters of people say it is either extremely important or very important to stay informed.

The bad news is that there is a great deal of confusion about basic facts relevant to policy. Almost half the public, and a quarter of those over age 55, thought Medicare already provided drug benefits for outpatients before legislation providing such coverage was enacted. More than half could not hazard a guess about the size of the budget deficit. The average person thinks 37 percent of Americans lack health insurance, more than twice the actual percentage.

From where do Americans learn about the economy? By far the most common source is television. Those who rely on television the most, however, tend to be among the least informed.

The second most common source is local newspapers, which were cited much more frequently than national or big-city papers.

Friends and relatives came in third, followed by political leaders, radio and economists. The Internet was next, although a sizable contingent listed it as their most important source.

Those who consulted more sources, and consulted them more often, were a bit better informed - but not much. That's a sobering fact for the media.

People who said they voted in the last presidential election were better informed than nonvoters.

Liberals, moderates and conservatives all did about equally well on the test of economic facts. But those who said they hadn't thought much about their ideological leanings - one in three people - were appreciably less knowledgeable.

Politicians pay attention to opinion polls. Policies that poll poorly are rarely pushed. What affects where people stand on the issues?

Those who were more knowledgeable about economic policy facts were less likely to support the last Bush tax cut. Among those who scored in the bottom quarter on the factual knowledge quiz, the ratio of favorable to unfavorable opinions was 2.6 to 1; for those in the top quarter it was 1.1 to 1.

Another study, by Joel Slemrod of the University of Michigan, found that a majority of people mistakenly believed that middle-income families pay a higher percentage of their income in federal taxes than do high-income families. The truth is that the top 1 percent of households paid 24 percent of their income in federal income taxes in 2000, while the middle 20 percent paid 5 percent, according to the Congressional Budget Office. This misperception, Mr. Slemrod concluded, leads many people to favor a flat tax over the current system; they mistakenly think the rich will pay more with a flat tax.

On other issues, however, knowledge and perceptions were only weakly related to positions on policy issues, if at all. For example, the perceived size of the government deficit or general economic knowledge was unrelated to whether people considered the deficit a major problem.

Another surprising result is that political scientists and economists have been unable to find much of a link between individuals' self-interest and their position on a range of public opinion questions, or on their voting behavior.

For example, higher-income people, who stood to benefit much more from the second Bush tax cut than lower-income people, expressed only slightly stronger support for it. And those old enough to receive Medicare were slightly less in favor of expanding it to cover drug insurance than everyone else was.

What does predict public opinion on the economy? In a word, ideology.

On most issues, liberals and conservatives take predictably different positions. And when it comes to the government budget deficit, the views of liberals and conservatives reversed after conservatives enacted record deficits, suggesting that ideological positions are situational.

Ideology may matter because it is costly for individuals to learn about specific issues; taking the party line is easy. Arthur Lupia, a political scientist at the University of Michigan, said, "For voters, 'investing' their attention in ideological or party clues, rather than economic policy details, can be quite rational."

Even so, the view of public opinion that emerges from this research is contrary to the assumption of nonideological, self-interested people that economists typically apply to the political arena. Instead, voters appear to put more weight on what they think is in the country's best interest, using whatever clues they can.

Yet pocketbook issues do affect voting. Many studies have found that economic growth is a good predictor of presidential elections. Incumbents do worse in a weakening economy. Ronald Reagan asked voters whether they were better off than they were four years ago, to great effect.

Self-interest may not affect opinion polls because people lack the interest, information and background to disentangle which policies are to their advantage, or they may feel better by giving the answer that they think would be best for the commonweal. But they can judge whether they are better off than they were before. In the current presidential election, the state of the economy is likely to have much more impact on voters than the specifics of the policies either side proposes.

Still, there are potential costs to systematic misperception and disregard of self-interest. For instance, a large majority can end up on the wrong side of class warfare, as Warren E. Buffett recently warned.

Alan B. Krueger is the Bendheim professor of economics and public affairs at Princeton University. E-mail: akrueger@princeton.edu.