Fight Clubs: Media Coverage of Party (Dis)unity and Citizens’ Selective Exposure to It

DOI10.1177/1065912919827106
Date01 June 2020
Published date01 June 2020
Subject MatterArticles
https://doi.org/10.1177/1065912919827106
Political Research Quarterly
2020, Vol. 73(2) 276 –292
© 2019 University of Utah
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/1065912919827106
journals.sagepub.com/home/prq
Article
Members of the American public perceive the Democratic
and Republican parties to be more than mere vehicles for
enacting specific policies—the parties are viewed as two
sets of different kinds of people. A growing body of
research contends that citizens harbor consistent beliefs
about how the two major parties in the United States
relate to particular social groups (Campbell, Green, and
Layman 2011; Rothschild et al. 2018) and that such sym-
bolic, “group-centric” considerations are consequential
for party identification (Green, Palmquist, and Schickler
2002; Miller, Wlezien, and Hildreth 1991), policy sup-
port (Nelson and Kinder 1996), political attitudes, and
polarization (Ahler and Sood 2018; Mason and Wronski
2018). Mason (2015), for example, finds that citizens
who more strongly identify with the groups typically
associated with each party are more politically active and
affectively polarized.
Less understood, however, is how members of the
public come to be aware of such stereotypes and, in par-
ticular, the role that mass media may play in transmitting
linkages between parties and various politically relevant
groups (i.e., each party’s various political, social, and
economic groups in both government and the mass
public). The present study, therefore, first reports results
from an exploratory content analysis of over 1,300 news-
paper articles published during the Obama presidency. A
primary aim of this analysis was to assess the degree to
which news media discussed both inter- and intra-party
dissension between President Obama (leader of the
Democratic Party) and each party’s coalition—that is, not
only relations between Obama and elected officials but
also the various class-based, racial, religious, and social
groups that typically align with one of the two major par-
ties. As many journalists have noted, Obama demon-
strated a willingness to both please and disappoint
elements of the Democratic Party’s base, whether in the
political stances he adopted or in not acting aggressively
enough on behalf of these groups (e.g., Baker 2013;
Chapman 2010; Halloran 2013; Wolfgang 2013). By
exploring patterns in how President Obama was reported
827106PRQXXX10.1177/1065912919827106Political Research QuarterlyKane
research-article2019
1New York University, New York City, USA
Corresponding Author:
John V. Kane, Center for Global Affairs, New York University, 15
Barclay Street, New York City, NY 10007, USA.
Email: John.Kane@nyu.edu
Fight Clubs: Media Coverage of Party
(Dis)unity and Citizens’ Selective
Exposure to It
John V. Kane1
Abstract
News media play a key role in communicating information about political parties to the American public. However,
our understanding of how media depict relations between elites and the broader party coalitions remains limited.
Moreover, while research suggests that forced exposure to such information can affect political attitudes, it remains
unclear whether citizens are willing to selectively expose themselves to such communications. To address these two
interrelated questions, this study first employs a content analysis to explore patterns in news coverage of inter- and
intra-party relations throughout the Obama presidency. Next, two survey experiments investigate the degree to
which such relations affect citizens’ self-exposure to such information. Taken together, the analyses uncover two
important asymmetries. First, throughout Obama’s presidency, mass media depicted a Republican coalition virtually
always against the president, yet substantial discord within the Democratic Party. Second, though partisans show no
propensity to consume news depicting inparty unity (vs. disunity), both Republicans and Democrats exhibit a strong
tendency to consume news stories depicting disunity in the outparty. Insofar as partisans’ self-exposure to such
information is a necessary precondition for attitudinal and behavioral change, these findings have notable implications
for how mass media stand to shape partisanship in the United States.
Keywords
media, self-exposure, partisanship, party unity, cognitive dissonance, schadenfreude
Kane 277
to relate to both the Republican Party coalition and his
own, such an analysis stands to provide a more holistic
understanding of how national media communicate infor-
mation about the parties and, specifically, (dis)unity
within and between party coalitions.1
Importantly, extant research contends that awareness
of such relations within and/or between the parties can
influence how citizens evaluate presidents and elected
officials, as well as how citizens reason about the merits
of complex policy proposals (Jones 2017; Rottinghaus
and Tedin 2012). For example, Groeling (2010, 140–46)
finds that intra-party disunity can lower approval of pres-
idents among both Independents and inparty members,
while Kane (2016a, 2016b) finds that intra-party disunity
garners greater political support from outparty members
compared with intra-party unity. However, if such down-
stream effects are to be observed in the real world, a key
precondition is that citizens—and (perhaps especially)
partisan citizens—are willing to selectively expose them-
selves to this kind of political information. It should not
be taken for granted that partisans will do so, particularly
in light of mounting research documenting partisans’ ten-
dency to consume news stories that are unlikely to chal-
lenge their existing political views, evaluations of elites,
and so on (e.g., Levendusky 2013). Therefore, following
the content analysis results, I report findings from two
survey experiments, one involving the Obama presidency
and one involving the Trump presidency, designed to
determine whether news stories involving intra-party
unity or disunity can affect partisans’ likelihood of self-
exposure to such information.
Overall, the results of these two complementary analy-
ses make novel contributions to previous studies of politi-
cal news content and partisans’ selective exposure to it.
If, for example, beliefs about the social composition of
each party are consequential for mass partisanship and
political behavior (Campbell, Green, and Layman 2011;
Green, Palmquist, and Schickler 2002), then it is crucial
that scholars develop a better understanding of how mass
media communicate relations between elected elites and
politically relevant social groups. On this point, I find that
nearly 30 percent of the 1,360 content-analyzed articles
discussed the Obama presidency in relation to particular
partisan groups, suggesting that the practice of using
political groups to contextualize the president’s actions is
quite prevalent in mass media. More broadly, the content
analysis reveals an important asymmetry: throughout the
Obama presidency, mass media communicated that
Democratic groups were opposed to the president approx-
imately as often as they were aligned with him, while
Republican groups were, by comparison, nearly always
opposed to President Obama. This finding suggests that,
at least with respect to the president, mass media do not
simply present a political landscape in which one party
coalition is uniformly aligned with the president’s agenda
while the other is uniformly opposed—the reality is
somewhat more complex.
The survey experiments reveal a second asymmetry.
Specifically, the experimental results indicate that, while
partisans were nearly equally likely to select news arti-
cles discussing inparty unity (vs. news articles discussing
inparty disunity), both Republicans and Democrats were
far more likely to select a news story depicting outparty
disunity (vs. outparty unity or the control condition)—
that is, the story in which the president was reported to
have alienated a key group in his party’s base. The find-
ing concerning inparty members’ behavior deviates
somewhat from the “cognitive dissonance” argument,
which would suggest that partisans will avoid self-expo-
sure to information that challenges their existing beliefs
and loyalties (Festinger 1962). At the same time, the find-
ings regarding outparty members’ behavior offer addi-
tional support for the notion that partisans seek out
information that is likely to result in “political schaden-
freude” (Combs et al. 2009; Hareli and Weiner 2002).
Again, in light of research which argues that beliefs
about the parties’ relations with groups are consequential
for citizens’ partisan orientations (Green, Palmquist, and
Schickler 2002), as well as recent research on how infor-
mation about partisan polarization and partisan infighting
can affect citizens’ policy attitudes and willingness to
compromise (Druckman, Peterson, and Slothuus 2013;
Groeling 2010; Kane 2016b), the findings of the present
study come with important implications for the manner
by which mass media can shape partisanship.
Party Coalitions and News Media
Research on the social-group bases of partisanship argues
that citizens’ perceptions of, and feelings toward, the
kinds of people in each of the party coalitions structure
these citizens’ own partisan orientations (Ahler and Sood
2018; Green, Palmquist, and Schickler 2002; Mason
2016; Miller, Wlezien, and Hildreth 1991). In other
words, the composition of, and politics within, the two
parties’ political coalitions may function as cues, strongly
pulling citizens toward—or pushing them away from—
one of the two major political parties. Corroborating this
argument, recent scholarship finds that negative senti-
ments toward outparty members have intensified concur-
rently with the systematic sorting of particular social
groups into one of the two parties (Abramowitz and
Webster 2018; Achen and Bartels 2016; Levendusky
2009; Mason 2015).
Such research necessarily implies that citizens have, at
some point, learned about how the parties relate to vari-
ous politically relevant groups. Assuming that citizens
learn this information about the parties (at least in part)

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