Gender, Race, Ethnicity, and the Racialization of Attitudes Toward Descriptive Representation

Date01 September 2021
Published date01 September 2021
Subject MatterArticles
American Politics Research
2021, Vol. 49(5) 517 –533
© The Author(s) 2021
Article reuse guidelines:
DOI: 10.1177/1532673X211022620
To better understand the political underrepresentation of
marginalized groups, research increasingly points to not only
the supply of potential candidates, but also the demand for
diverse representatives by both party elites and voters
(Crowder-Meyer, 2013; Karpowitz et al., 2017). Who is most
likely to support diverse candidates, and what are some of
the factors that influence these views? And, will attitudes
about one marginalized group predict support for increased
representation by members of other underrepresented
groups? For instance, are racial attitudes related to opinions
about greater representation by women? Is feminism con-
nected to views about Hispanic representation? We explore
these questions using data from the 2016 American National
Election Study (ANES), which asks respondents their atti-
tudes about increased representation by women and
Hispanics. Specifically, we focus on the extent to which
racialized attitudes and a sense of group connection shape
public opinion about increasingly diverse representation.1
Although conventional wisdom suggests that partisanship
drives attitudes about descriptive representation, in recent
years, polarized partisan fault lines have increasingly aligned
with racial ones, even spilling over into issues not histori-
cally associated with racial divides (Enders & Scott, 2019;
Hajnal, 2020; Tesler, 2012, 2016; Tesler & Sears, 2010).
Much research on this topic focuses on racial resentment, a
common measure of racial animus that seeks to capture
“modern” or less overt racism by asking respondents a series
of questions about their views of Black Americans (Kinder &
Sanders, 1996). Racial resentment is thought to be predictive
of a growing range of political behaviors and attitudes,
including voting (Enders & Scott, 2019; Sanbonmatsu,
2020), as well as the evaluation of policies (Tesler, 2015,
2016) and candidates (Kinder & Dale-Riddle, 2012; Tien,
2017; Yadon & Piston, 2018). We examine whether racial
resentment toward members of one group influences atti-
tudes about representation by members of other marginal-
ized groups, including by gender.
Scholarship on descriptive representation additionally
argues for the primacy of identity congruence in shaping vot-
ers’ preferences for representation, with women preferring
more women representatives, and individuals of all racial/
ethnic groups generally preferring representatives who share
their race/ethnicity (English et al., 2019; Rosenthal, 1995;
Sanbonmatsu, 2003). However, it may not be just group
membership, but the strength of connection to one’s group
that creates demand for descriptive representation. A deeper
attachment to a group has been linked not only to demand for
1022620APRXXX10.1177/1532673X211022620American Politics ResearchLucas and Mohamed
1Saint Anselm College, Manchester, NH, USA
2Clark University, Worcester, MA, USA
Corresponding Author:
Heather Silber Mohamed, Department of Political Science, Clark
University, 950 Main Street, Worcester, MA 01610, USA.
Gender, Race, Ethnicity, and the
Racialization of Attitudes Toward
Descriptive Representation
Jennifer C. Lucas1 and Heather Silber Mohamed2
Who is most likely to consider diverse representation desirable? Previous literature typically emphasizes the importance
of partisanship and group identity congruence (i.e., women representing women) in influencing attitudes about descriptive
representation. Alternatively, we test whether the racialized politics that emerged in 2016 might now shape views about
representation by members of underrepresented groups. Using data from the 2016 ANES, we examine attitudes toward
increasing the number of women and Hispanic representatives. Rather than partisanship or identity congruence, our results
point to the primacy of racial attitudes, linked fate, and feminism in shaping views about diverse representation. Indeed,
even when examining attitudes about increased representation by women, we find strong evidence that opinions are now
racialized. We argue for a more expansive understanding of support for descriptive representation, which may reflect an
individual’s opinions of marginalized groups and structural inequities more broadly.
descriptive representation, racial resentment, Latino elected officials, women elected officials
518 American Politics Research 49(5)
in-group descriptive representation, but in some cases, in
support for co-ethnic representation, such as increased sup-
port for Black representation among Latino respondents
(McConnaughy et al., 2010). But, the question remains to
what extent a sense of connection with any marginalized
group might equate to greater support for increased represen-
tation by other marginalized groups, particularly across gen-
der/racial/ethnic groups.
We use two distinct measures to analyze the extent to
which group connection influences attitudes about diverse
representation, both by members of one’s own group and by
members of other underrepresented groups. First, we exam-
ine linked fate, or the extent to which members of a racial/
ethnic group see their own fate as tied to that of other group
members (Dawson, 1994). While measures of linked fate are
far less common in gender and politics research (but see Gay
et al., 2016), surveys frequently include questions about atti-
tudes toward feminism, which captures a combination of
gender policy attitudes and group identity (Bittner &
Goodyear-Grant, 2017; Harbin & Margolis, 2020; Zucker,
2004). Thus, our second measure uses attitudes about femi-
nism as an indicator of shared interest or group connection
among women. We explore the extent to which identifying as
a feminist is associated with support for more diverse repre-
sentation, including increased Hispanic representation.
Notably, measures of “modern” racist and sexist attitudes
emerged in tandem, grounded in an underlying attempt to
capture a more subtle prejudice combined with a sense of
individualism (Benokraitis & Feagin, 1986; Swim et al.,
1995). In contemporary research, the connection between
racist and sexist attitudes has typically been decoupled, or
remains underexplored (Winter, 2008). However, there is
still much to be learned about how gender and race intersect;
as Cassese and colleagues (2015) note, “there is a critical
need to better understand the process by which gender issues
become racialized” (see also García Bedolla, 2020). Our
paper heeds this call by exploring the extent to which views
about one marginalized group are connected to attitudes
about representation by another, returning to a tradition in
which racial and gendered attitudes are understood as dis-
tinct but similar concepts (Dovidio et al., 1989; Sears &
Kinder, 1971).
We seek to understand whether there are commonalities
that drive attitudes about diverse representation more gener-
ally. Instead of partisanship or shared identity, we find that
support for more diverse officeholders depends on individu-
als’ racial and gendered attitudes. For instance, we find that
racial resentment is a significant predictor of attitudes about
descriptive representation, even when respondents are asked
specifically about representation by women. We also find that
a sense of group connection, rather than a shared identity, is
particularly influential in attitudes about descriptive repre-
sentation, with Black and Latino linked fate, and feminist
attitudes, all being strong predictors of attitudes about diverse
representation, even in the absence of identity congruence. In
contrast to conventional wisdom, in multivariate models, we
find no statistically significant differences based on partisan-
ship, and in some cases, variables for identity congruence are
significant in the opposite of the expected direction once
measures for racial resentment, linked fate, and feminism are
included. Our research suggests that candidates of diverse
backgrounds may find support among members of both
political parties, with racial and gendered attitudes being a
key factor in shaping demand for increasing representation
by members of currently underrepresented groups.
Attitudes Toward Descriptive
A representation gap exists between women and men, and
among racial/ethnic minority groups, in comparison to non-
Hispanic white elected officials. In the 115th Congress,
which was seated soon after completion of the 2016 ANES,
there was nearly a nine-point difference between the percent
of Latinos in the general population and their overall share of
Congressional seats. The gender gap in representation was
even more striking. While women make up roughly half the
population, the gap was closer to 30 percentage points for
Congress; despite being nearly 51% of the U.S. population,
women still constitute less than a quarter of congressional
members. While Latinos comprise approximately 16% of the
population, they represented just 4% of Senators and 9.4% of
House members in 2017 (Zweigenhaft, 2018).
Does the American public want this inequality remedied?
A 2018 Pew Research study found that 59% of all Americans
agreed there are too few women in high political offices
today (Horowitz et al., 2018; see also Dolan & Sanbonmatsu,
2009; Gallup News Service, 2001), suggesting that while a
majority prefer more diverse representatives, a sizeable
minority does not. Given such gaps, it is important to under-
stand who supports more diverse representation, and whether
the same factors drive attitudes about representation by dif-
ferent marginalized groups.
Much of the research around preferences for diversity in
representation is rooted in scholarship on descriptive repre-
sentation. Pitkin (1967) distinguishes between descriptive
(“standing for”) and substantive (“acting for”) representa-
tion. While descriptive representation in itself may be insuf-
ficient to represent members of marginalized groups,
descriptive characteristics are often thought to be a proxy for
substantive representation, with women representatives pri-
oritizing women’s issues (Barnes et al., 2021; Reingold,
2009; Swers, 2013; Taylor-Robinson & Heath, 2008).
Likewise, across racial/ethnic groups, co-ethnic representa-
tives are thought to be more aligned with group members’
policy preferences, with Black representatives more likely to
vote for legislation supported by the Black community (Tate,
2003; Whitby, 1997), and Latino elected officials more likely
to represent “Latino” interests (Wilson, 2009, 2017), albeit
with variation by context and party (Casellas, 2010;

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT