Divisions in the Big Tent: Group Sentiments and Candidate Preferences within the Democratic Party

• Published date: 01 July 2022
• Date: 01 July 2022
• DOI: 10.1177/1532673X221081481
• Subject Matter: Articles

Article

American Politics Research

2022, Vol. 50(4) 488–502

© The Author(s) 2022

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/1532673X221081481

journals.sagepub.com/home/apr

Divisions in the Big Tent: Group Sentiments

and Candidate Preferences within the

Democratic Party

Jonah B. King

1



Abstract

Work in political psychology has found that the two major political parties are sorte d based on individual’s group attachments.

This sorting has resulted in the Democratic Party’s electorate and elites being made up of members from various ra cial,

religious, and social groups. Given this heterogeneity, how do social group identities impact voter preferences in the

Democratic primary? Based on social identity theory, I hypothesize that an individua l’s feelings toward racial or gender groups

should be associated with their candidate preference in the primary election, but only when a candidate’s group identities are

known. Using the 2018 American National Elections Studies (ANES) pilot and a nationally diverse survey of 2020 Democratic

primary voters, I f‌ind that the strength of sentiments toward racial or gender identiti es is associated with candidate preference

and this association is conditioned by the ability to link the candidates to racial or gender groups. This paper sho ws that the

effects of group identity extend beyond simply sorting individuals into one of the two major parties and have the potential to

create group-based divisions within the “big tent”of the Democratic Party.

Keywords

Social identity, voting behavior, social groups, group identity, Democratic Party

The base of support for the Democratic Party has been de-

scribed as a “big tent”that includes individuals of various

races, genders, sexual orientations, and religions (Pew

Research Center, 2018). A simple look at the candidates in

the Democratic2020 Presidential primaries or thecomposition

of Members of Congress (Pew Research Center, 2018)shows

that this diversity extends to the elite level. Black individuals,

Latinos, women, members of the LGBTQ community, and

non-Christians comprise the Democratic f‌ield. The base of

support as well as elites of the Democratic Party stand in stark

contrast to the largelyWhite and Christian base of support and

elites of the Republican Party (Abramowitz, 2018;Mason,

2018). Research has shown identity-based sorting of these

groups into the Republican and DemocraticParties over recent

years (Achen & Bartels, 2016). In this instance, sorting means

that as individuals belong to and feel closer to groups aligned

with their party, the strength of their partisan identif‌ication

increases (Mason & Wronski, 2018). These past studies ex-

amine the cumulative effects of party aligned group identities.

The amount of diversity and the potentialfor cross pressures in

the Democratic Party, however, permits an examination of

whether group sentiments within the Democratic Party in-

f‌luence intraparty candidate preference.

The existing work on social identity in politics is extensive

and offers a great deal of insight into how our group

attachments inf‌luence our political behavior. This literature

has largely focused on partisanship, ideological identif‌ica-

tion, and participation, but is the impact of social group

sorting limited to these factors? It is reasonable to think that

the effects of sorting will be more muted in the Republican

Party due to its homogeneity, but the heterogeneity of the

Democratic Party may lead to additional group divisions

beyond partisanship. This means that Republicans will tend

to f‌ind that the Republicans around them look like them, hold

similar religious beliefs, and sexual orientations. On the other

hand, if an individual Democrat were to take a look around

their “big tent,”they may see members of other racial groups,

people of varying gender identities, individuals with different

sexual orientations than themselves, and people who do not

share their religious beliefs. I aim to determine if the potential

for varied group sentiments within the Democratic Party will

result in voting within the party based on social group

sentiment. The crowded 2020 Democratic presidential

1

Political Science, University of Mississippi, MS, USA

Corresponding Author:

Jonah B. King, Political Science, University of Mississippi, University, MS

38677, USA.

Email: jbking@go.olemiss.edu