Who Represents Me? Race, Gender, Partisan Congruence, and Representational Alternatives in a Polarized America

• Author: Kathryn Pearson,Dara Z. Strolovitch,Ashley English
• DOI: 10.1177/1065912918806048
• Published date: 01 December 2019
• Date: 01 December 2019
• Subject Matter: Articles

https://doi.org/10.1177/1065912918806048

Political Research Quarterly

2019, Vol. 72(4) 785 –804

© 2018 University of Utah

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/1065912918806048

journals.sagepub.com/home/prq

Article

Few concepts are as fundamental to American democ-

racy—and few issues more essential to democratic legiti-

macy—than Americans’ beliefs that their views and

interests are represented in political institutions. But

while there are rich bodies of research examining issues

such as Americans’ trust in government and whether their

preferences are congruent with the behavior of their

members of Congress (MCs) and policy outcomes, schol-

ars have paid less attention to what happens when people

do not feel that the MC sent to Washington by voters in

their district actually represents them and their interests.

Do they, for example, seek out other options when their

own MC is a member of the opposing party? Likewise,

although scholars have demonstrated that descriptive rep-

resentation has positive effects on a range of political out-

comes for members of historically marginalized groups

such as women and people of color, we know little about

who they seek out when they are represented by MCs

who are not “like” them when it comes to salient identi-

ties and formations such as race, ethnicity, and gender.

Recognizing that Americans are simultaneously for-

mally represented by multiple elected officials and

informally represented by other political actors and entities

such as advocacy organizations, we use original survey

data to answer critical questions about their beliefs about

who represents them in national politics and to show how,

in a pluralist system, they decide which of many possible

actors represent them the most (Saward 2010, 38). In so

doing, our study provides a novel approach to thinking

about the connection between symbolic representation—

defined by Hanna Pitkin (1967) as representation that

evokes “emotional, affective” beliefs that can lead people

to feel represented by political actors that they identify

with—and descriptive representation, or representation

that mirrors “the people, the state of public consciousness,

or the movement of social and economic forces in the

806048PRQXXX10.1177/1065912918806048Political Research QuarterlyEnglish et al.

research-article2018

1University of North Texas, Denton, USA

2University of Minnesota, Minneapolis, USA

3Princeton University, NJ, USA

Corresponding Author:

Ashley English, Department of Political Science, University of North

Texas, 1155 Union Circle #305340, Denton, TX 76203, USA.

Email: Ashley.English2@unt.edu

Who Represents Me? Race,

Gender, Partisan Congruence, and

Representational Alternatives in a

Polarized America

Ashley English1, Kathryn Pearson2, and Dara Z. Strolovitch3

Abstract

The belief among citizens that their views are represented is essential to the legitimacy of American democracy, but few

studies have explicitly examined which political actors Americans feel best represent them. Using data from the 2006

Cooperative Congressional Election Study, we ask new questions about whether respondents who share a partisan,

racial, or gender identification with their members of Congress (MCs) feel those members best represent them.

Although the framers designed the House so that individuals’ own MCs would be their closest and most responsive

representatives, a majority of respondents turn to other actors for representation. Partisanship is a key reason for

this attenuated connection, as respondents who do not share a partisan identification with their MCs are more likely

than those who do to rely on their party’s congressional leaders or advocacy organizations for representation instead.

Sharing a racial identification with one’s own MC can strengthen representational connections as respondents who

share a racial identity with their MCs are significantly more likely than respondents who do not to indicate that their

MC represents them “the most.” These results shed light on enduring questions about the significance of symbolic

representation and its link to partisanship and descriptive representation.

Keywords

symbolic representation, descriptive representation, partisanship, race, gender

786 Political Research Quarterly 72(4)

nation” (p. 61). We do so by asking people who they feel

represents them the most effectively and by examining

how those attitudes are connected to the partisan, racial,

and gender identifications of the respondents and their rep-

resentatives. That is, we take up a provocative empirical

question—“What makes men [sic] feel represented?”—

that Pitkin left unanswered in her pathbreaking and con-

ceptually focused examination of the circumstances under

which people should believe in their representatives or

should feel represented (Pitkin 1967, 9, 111). Our approach

is, therefore, in conversation with work examining the rela-

tionship between symbolic and descriptive representation,

on the one hand, and work about people’s beliefs in the

legitimacy of their political institutions or the fairness of

the decision-making process, on the other. At the same

time, we depart from those studies by emphasizing peo-

ple’s beliefs about their own individual House members

and by providing them with the opportunity to indicate that

they would prefer to seek out another political actor for

representation instead (Abney and Hutcheson 1981;

Atkeson 2003; Atkeson and Carrillo 2007; Banducci,

Donovan, and Karp 2004; Barreto, Segura, and Woods

2004; Gay 2001, 2002; Hansen 1997; Hayes and Hibbing

2017; High-Pippert and Comer 1998; Howell and Fagan

1988; Jones 2014; Pantoja and Segura 2003; Reingold and

Harrell 2010; Sanbonmatsu 2003; Sapiro and Conover

1997; Verba, Burns, and Schlozman 1997).

Three central themes emerge from our analyses. First,

although the framers of the U.S. Constitution designed

the House of Representatives so that individuals’ own

MCs would be their closest and most responsive repre-

sentatives, the majority of Americans do not necessarily

believe that their MCs represent them better than other

political actors. Second, this attenuated connection is

conditioned by partisanship: People who share a partisan

affiliation with their MC are significantly more likely

than those who do not to look to them for representation,

and they are also significantly less likely to report that

other actors such as advocacy groups and congressional

party leaders represent them. Third, people are more

likely to “prefer” their MCs when they share a racial

identification with them, suggesting that Americans from

both dominant and marginalized groups recognize and

value the symbolic nature of descriptive representation.

Assessing the Framers’

Representative Claim

Americans are formally represented by four federal elected

officials: their MC, two senators, and the president. When

the framers designed American political institutions, how -

ever, they made what Michael Saward (2010) calls a

“representative claim” that the House of Representatives

would be the “people’s House.” They codified that aspira-

tion by creating small districts, holding elections every two

years, and making House members the only directly

elected federal officials (at the time). They hoped that these

provisions would ensure that House members would be the

most proximate to the American people. Proximity to one’s

MC is not the same as being represented by them, however.

Cindy Simon Rosenthal (1995), for example, suggests that

it is important for constituents to believe in their represen-

tatives. Likewise, Andrew Rehfeld (2006) argues that rep-

resentation can only occur when people recognize political

actors as their legitimate representatives, and Iris Young

(1997) explains that this recognition is more likely to occur

when people share a social perspective with political

actors, believe that political actors look after their interests,

and share political actors’ opinions, principles, or values.

People who share social perspectives with their MCs

should, therefore, be more likely to feel both emotionally

connected to them and to feel symbolically represented by

them.” They should also, in turn, be more likely to trust

them and to believe that their groups’ interests are being

represented (Gay 2002; Young 1997). A growing body of

research indicates, for example, that constituents who

share a gender or racial identification with their MCs are

more likely to trust and contact their representatives and to

participate in politics than those who do not (Abney and

Hutcheson 1981; Howell and Fagan 1988; Pantoja and

Segura 2003; Tate 2003).

Building on these suggestive insights about the condi-

tions under which people are likely to “feel” symboli-

cally represented, we argue there are several reasons to

believe that some Americans reject the framers’ notion

that their House members represent them the best. For

example, in a society in which race and gender shape

people’s understandings of their political identities and

interests (see, inter alia, Beltrán 2010; Celis et al. 2014;

Cohen 1999; Cramer 2016; Crenshaw 1989; Dawson

1994; Mansbridge 1999; Sapiro 1981; Strolovitch 2007;

Strolovitch, Wong, and Proctor 2017; Young 1994,

2000), constituents and political actors may be more

likely to share a perspective when they also share social

positions in terms of gender, race, ethnicity, and other

salient identities. In an increasingly polarized environ-

ment, people may likewise believe that only a represen-

tative from “their” political party will share their interests

and values (Mason 2018). People whose race, gender, or

partisan identifications are not congruent with those of

their MCs may consequently not “feel represented,” par-

ticularly in the contemporary political environment in

which voters increasingly disapprove of the rival party,

and the House is more competitive and polarized along

partisan lines than it has been in nearly a century.1 In

such an environment, it may also be difficult for MCs to

overcome partisan disagreements with their constituents

through their nonpartisan “home styles” and constituent

case work (Fenno 1978; Mayhew 1974; Young 1997). In

theory, these discontented partisans could vote their MCs