Individuals perceive the gay non-partisan candidate as significantly more liberal; & conservative individuals reliably prefer in-group gay candidates to straight candidates from the other party

Partisanship, sexuality, and perceptions of candidates. Eric Loepp & Shane M. Redman. Journal of Elections, Public Opinion and Parties, Jan 12 2020. https://doi.org/10.1080/17457289.2019.1711099

ABSTRACT: Cultural debates over gay rights have prompted a great deal of scholarship assessing the nature and resiliency of citizen attitudes toward homosexuality in American political life. We posit, however, that more attention should be devoted to LGBT individuals themselves as potential office holders. Specifically, we argue that in an era of increased affective polarization and partisan sorting, party identification of candidates, as well as affectual and ideological attitudes of voters, must be integrated into analyses of the effects of sexuality as a voting heuristic. Drawing on social identity and subtyping theories, we contend that candidate sexuality will influence attitudes about some candidates, but not others. We then present the results of an original survey experiment in which candidate partisanship and sexuality are both manipulated. The data reveal substantial support of our theory: while sexuality is a relevant consideration in the candidate evaluation process, both partisanship of candidates and political attributes of voters can severely condition its effects.


Discussion
In a world of increasing affective polarization (e.g. Iyengar and Westwood
2015) and partisan-ideological sorting (e.g. Levendusky 2009), studying the
effects of candidate sexuality requires analyzing its interaction with partisanship. We demonstrate that while gay stereotypes remain a potent force in
shaping impressions of political candidates in a non-partisan setting, they
are largely, though not wholly, neutralized in partisan contexts. Voters in
both parties consistently express more support for co-partisans than out-partisans. Even more conservative individuals reliably prefer gay candidates from
their own party to straight candidates from the other. At the same time, candidate sexuality can be an important determinant of voter attitudes when
evaluating gay Republican candidates. These candidates represent a
subtype of Republicans, possessing cognitively inconsistent attributes that
complicate the evaluation process. This tension manifests in the use of sexuality as a conditioning heuristic that generates differential perceptions of gay
and straight Republicans. The results speak to behavioral patterns in the electorate that have important implications for gay candidates that are beginning
to run for office at higher rates than ever. 2018 saw over 600 non-straight candidates run for public office in the United States, including a four-fold increase
over 2010 in the number of candidates nominated for a seat in Congress.16
The 2020 presidential election cycle witnessed the first openly gay candidate
run for his party’s presidential nomination.17 Our results suggest mixed prospects for these candidates. On one hand, sexuality remains a pervasive
influence in non-partisan contexts. Given that political careers often start at
lower levels of public office—which are often non-partisan—the results
imply that if openly gay candidates seek a non-partisan office, their sexuality
status may—if divulged—be used as a cue by voters, such that the candidate
is perceived to be more ideologically liberal. In heavily Democratic areas, this
may be an asset; in more Republican regions, a liability. However, many elections are not non-partisan, and this study contributes important caveats concerning the applicability of sexuality to partisan races. Consistent with recent
research on the effects of partisan affect and sorting (e.g. Mason 2015), we
observe that shared partisanship compels voters to express (dis)approval of
Democratic candidates principally as a function of their party affiliation.
Voters are no more likely to reject a gay Democratic candidate than a straight
one, or vice versa. Popular wisdom, along with some of the extant literature,
suggests that gay individuals face additional barriers that may discourage
them from running for office. While we do not claim gay Democrats will
never face electoral challenges when running for office that their straight
counterparts do not, this research suggests that when it comes to judging
Democratic politicians, the effects of sexuality are diminished. At the same
time, the data suggest a dilemma for gay Republicans that is not dissimilar
to challenges female or non-white Republicans have historically faced:
voters drawing on global stereotypes about partisanship and sexuality must
process conflicting information about them. This could prove advantageous
at times: for instance, a gay Republican candidate may enjoy more support
among Democratic voters than would straight Republican candidates. Yet
Democrats may still perceive gay Republicans to be “too conservative”
when they are contrasted with Democratic candidates from the voters’ own
party. Simultaneously, we show that Republicans may perceive gay Republicans as “not conservative enough,” and express less support for them than
for straight Republican candidates. This challenge is compounded by the
fact that primary elections in the United States mean gay Republicans
would first have to endure the penalties imposed by their sexuality among
party constituents. Even if they were to secure a party nomination, there is
little evidence Democrats would defect from their party to support a gay
Republican in a general election. Patterns of stereotypes and partisan affect
suggest the prospects for openly gay Republican candidates remain bleak.
This research does not purport to be the final authority on the interaction
of sexuality and partisanship. Indeed, multiple caveats are important to
acknowledge. First and foremost, this project represents an initial foray into
this subject; further research building upon the findings uncovered here is
essential. Different treatment designs may yield additional results of interest.
Relatedly, a number of potentially consequential variables that will further
qualify the impact of sexuality in American elections are now ripe for exploration, such as gender, race, and political status (e.g. incumbency). Indeed, just
as women may still be held to different standards than men even if they are
not overtly discriminated against at the polls (e.g. Teele, Kalla, and Rosenbluth
2018), it may also be the case that candidate sexuality matters in particular
electoral contexts more than others. Future work can also extend this analysis
to other phenomena of interest like vote choice in multi-candidate fields.
More expansive conjoint analyses would also be useful to further identify
the causal effects of candidate attributes on voter behavior. Despite these
limitations, we offer an important extension of early research on sexuality
cues that do not integrate partisanship: candidates’ sexuality does not systematically impact how all voters perceive them; instead, its impact is felt when it
does not align with partisanship, when partisan information is altogether unavailable, and among individuals with particularly strong political attitudes.