Since Alexis de Tocqueville described democracy in America in the 1830s, admirers of our political system have been struck by a particular combination of behavior. Our most active citizens both attend public meetings on community issues and work with other people directly on public problems. Tocqueville observed, "the most democratic country on the face of the earth is that in which men have, in our time, carried to the highest perfection the art of pursuing in common the object of their common desires and have applied this new science to the greatest number of purposes." 3 The "art and science of association," as he called it, required discussion to define issues, inform citizens, help them to understand one another's interests and values, and knit together communities. However, talking was not enough. By actually working on public problems, Americans made their talk consequential, and they learned lessons from experience that they could bring back into their conversations.

According to our Harris survey, fifteen percent of Americans (age 15 and older) say they do both of these things. This is certainly a minority, but it includes roughly 36 million people, which (for comparison) is equivalent to the population of the entire state of California and is a greater than the populations of Canada or the combined populations of 21 U.S. states.

Following the terminology of Cynthia Gibson's white paper for the Case Foundation, we say that this 15 percent of Americans does "citizen-centered" work.4 Citizens may sometimes choose to influence or collaborate with the government, but when they decide for themselves how to address issues, they are appropriately at the center of politics and civic life. Another term for citizen-centered engagement is "public work." 5

A more stringent definition would require that people discuss issues, work directly on issues, and talk with other people who hold views different from their own. Eleven percent of the population—or 26 million Americans—would meet that more stringent definition. In this section, we use the looser standard (not requiring that the discussions be diverse), but we turn to diversity of perspectives below.

A highly engaged group

Our survey shows that the 36 million people who engage in citizen-centered work are remarkably engaged and attentive:

They are leaders. Sixty-three percent of them are officers of clubs, compared to only 28 percent of the rest of the population.

They know more about politics than other people do. We asked the whole sample three factual questions about politics that have been pre- tested to measure broad knowledge.6 The people involved in "citizen- centered" work answered two of the three questions correctly, on average, and one third answered them all correctly. Their score was significantly better than the average for the whole population, who got between one or two correct answers (an average of 1.6) out of three.

They are also engaged in electoral politics. Even though we define the "citizen-centered" group as people who collaborate with peers on local or community issues, more than 70 percent of them say that they "always" vote in local and national elections, and 63 percent say that they participate in campaigns and elections in several different ways (for example, attending rallies or persuading other people to vote). Less than half of the rest of the population claims to vote regularly.

Citizen-centered people have different beliefs about government than other citizens do. They are more eager to be partners with government agencies. Thirty-four percent of them believe that it is "extremely" important for government agencies to give citizens like themselves constructive roles. Only 17 percent of other people agree (half the rate).

The survey asked whether respondents favored seven different strategies for addressing issues and being involved in their communities. Those involved in citizen-centered work were more likely than average Americans to support every one of the seven strategies. However, the gap was biggest when we asked about "attending community meetings sponsored by citizens' groups." About 40 percent of all Americans like to engage in this way, but it attracts 69 percent of the people who do citizen-centered work.

More than other people, those who engage in citizen-centered work believe that they have a duty to protest when something in society needs changing. Yet they are somewhat more trusting of the government, suggesting that they believe their voices can effect changes in policy. Finally, they are more likely than other people to say that it is important to serve their country through military service or community service.

How representative is citizen-centered engagement?
So far, we have shown that the roughly 36 million people involved in "citizen-centered" politics are dedicated, impressively informed and active, and influential. It is therefore important that they reflect all Americans so that the views of diverse groups have a voice.

African Americans and Whites appear to participate at roughly equal rates in various forms of citizen-centered engagement. Unfortunately, our samples of Latinos, Asian Americans, and Native Americans are small enough, and demographically different enough from the national populations of those groups, that we are unable to estimate their rates of "citizen-centered" participation with precision. As noted below, educational attainment correlates strongly with deliberation and public work. Since Latinos have lower average rates of college attendance than Whites, we would expect Latinos to be somewhat less represented in meetings and public work.

African Americans are the most philosophically committed to citizen-centered work. They are more likely than Whites to say that they would participate more if they had more opportunities and the most likely to say that a lack of venues for discussion is a serious problem. When offered a list of ways to address issues and be involved in their communities, they are the most likely to choose participating in community meetings, attending community meetings sponsored by churches (there is a 16-point gap on this question compared to Whites), and to gather with other citizens to identify problems and solutions.

Although liberals have a somewhat higher rate of participation in "citizen-centered" work than moderates and conservatives do, there is ideological diversity among the 36 million Americans. About 43 percent of them are liberals or lean to the liberal side, 35 percent are conservatives or lean to the conservative side, and 17 percent are moderates.

The greatest gap in participation is between the most and least educated. Education can also be seen as a measure of social class. We return to this problem below in the section on "Gaps in Participation." However, in keeping with previous research on political and civic participation in America, we find that religious congregations and labor unions reduce disparities by education. Those who attend church regularly are 8 points more likely to participate in "citizen-centered" work; union members are 9 points more likely to participate. After all, religious congregations and unions provide venues for discussing issues, offer opportunities for working together, and teach relevant skills. Unfortunately, both union membership and attendance at religious congregations have fallen over time.