Survey reveals a majority of Americans believe their Members of Congress do not represent them the best

James Madison wrote in Federalist 52 that the House of Representatives “should have an immediate dependence on, and an intimate sympathy with, the people.” This is why the framers of the U.S. Constitution structured House elections so that they would occur in small districts every two years.

That House members actually do closely represent the American people has been an operating assumption. Two hundred years after Federalist 52, congressional scholar David Mayhew reaffirmed this belief in his foundational work on the electoral connection, which noted that Members of Congress’ (MCs’) single-minded quest for re-election established an important accountability relationship with the electorate.

In spite of such longstanding assumptions, public approval of Congress as an institution is consistently low (currently at 19 percent). Moreover, recent research that we conducted calls into question these foundational assumptions about the elections and congressional representation. Are small districts with regular elections enough to ensure that the American people feel they are represented in Congress? If not, who do the American people turn to for representation in Washington, DC instead, and why?

To answer these questions, we crafted an original question for the 2006 Cooperative Congressional Election Study (CCES) to determine which political actors people believe represent them the most effectively at the national level. Respondents were presented with a list of representatives that included: “the House member from your district;” President George W. Bush; Republican congressional leaders such as Dennis Hastert, Democratic congressional leaders such as Nancy Pelosi, the Congressional Black Caucus, the Congressional Hispanic Caucus, the Congressional Asian Pacific American Caucus, the Congressional Caucus for Women’s Issues, and “organizations that speak for different groups or about different issues.” Then, we asked them to tell us which of those representatives they “look to MOST to represent” them “in Washington.” Our list of possible representatives includes a wide array of political actors beyond MCs because we assumed increasing partisan polarization and the underrepresentation of women and people of color in the House mean that many people do not believe they are able to elect House members who legitimately represent their interests.

The answers they gave reveal that the connection between MCs and their constituents is far more tenuous than the framers intended.

First, we found that the majority Americans did not believe that their MC was the political actor that represented them the best. Only 27% of respondents indicated their MC represented them “the most” and even when included their second choices, only 48% of respondents chose their own House members.

Second, we found that partisanship drives this attenuated electoral connection. Using multivariate statistical analyses, we found that sharing a partisan affiliation with one’s MCs increased the chances that respondents reported their own MCs represented them “the most” by 38% for Democrats and 28% for Republicans.

Third, we found that people who do not share a partisan identification with their MCs often look to party leaders and advocacy organizations instead of the House member from their district. Democrats represented by Republican MCs preferred Democratic congressional leaders and advocacy organizations to their own MCs, while Republicans represented by Democratic MCs turned to President George W. Bush and Republican congressional leaders instead.

Fourth, we found that people were more likely to look to their House members when they shared a racial identification with them. Our multivariate analyses indicated that racial congruence (African Americans represented by African American MCs, whites represented by white MCs, Latinos represented by Latino MCs, and Native Americans represented by Native American MCs) increased the chances that respondents reported that their own MC represented them the best by 8%.

Together, these findings provide important reminders that frequent House elections and geographically-based districts are not enough to ensure the American people feel represented in Washington, particularly in our current polarized era. Shared identity and experiences are an important component of “feeling” represented for Americans from both dominant and marginalized racial groups. For MCs, our findings suggest that it is difficult to bridge divisions between themselves and their constituents from the opposing party.

For more information, please see our full study in Political Research Quarterly at:

Ashley English is an Assistant Professor of political science specializing in gender and politics, interest groups and women’s organizations, and representation in the policymaking process at the University of North Texas. Kathryn Pearson is an Associate Professor specializing in American politics; her research focuses on the United States Congress, congressional elections, political parties, and women and politics. Dara Strolovitch is Professor at Princeton University, where she holds appointments in Gender and Sexuality Studies and the Department of Politics.

Filed Under:
Topics: Representation & Leadership