Campaigning with Class: The Impact of Candidate Class Background and Current Socio-Economic Achievement on Voters’ Evaluations

Download data and study materials from OSF

Principal investigator:

Meredith Sadin

Princeton University

Email: msadin@princeton.edu

Homepage: https://sites.google.com/site/meredithsadin/


Sample size: 1721

Field period: 7/27/2010-10/29/2010

Abstract

When deciding who to vote for, citizens often use crude information shortcuts that are based on a candidate's demographic characteristics. Despite the fact that one particular characteristic – a candidate's social class – is often made salient during a campaign, we know little about whether and how voters process this information. This experiment tests a theory of the way in which candidate social class operates as a heuristic in a campaign by providing information to voters about a hypothetical candidate's class origins, current class status and social class mobility. I find that citizens are more likely to vote for a candidate with working class origins and perceive a candidate with current working class status as less competent and more liberal than his upper class counterpart. I also show that a candidate's social class affects respondents' vote intentions largely by altering their perceptions of the candidate's ability to understand the concerns of people like them. This study is the first to treat social class as a heuristic and has implications for understanding how voters make inferences about class and incorporate these inferences into their vote calculus.

Hypotheses

Hyp. 1. Voters will prefer candidates with working class origins

Hyp. 2. Voters will prefer candidates with current upper class status.

Hyp. 3. Voters will prefer candidates that have worked their way up as opposed to those that were born privileged and remained so.

Hyp. 4. Voters' partisan identification will moderate the aforementioned hypotheses, with Democrats being more easily primed by social class information.
Hyp. 5. Voters' own current and past social class will moderate the aforementioned hypotheses.

Hyp. 6. The effect of candidate social class on citizens' vote likelihood will be mediated by the perception that a candidate is more likely to understand the concerns of voters like them.

Experimental Manipulations

The social class background and current social class of a hypothetical congressional candidate, each with three levels (control, working class, and upper class). These manipulations yield a 3x3 experimental design.

Outcomes

Respondent's vote intention, feeling thermometer, candidate ideology, four personality traits, three economic policy positions

Summary of Results

I find that citizens are more likely to vote for a candidate with working class origins and find him to be more understanding of their concerns, and more liberal on a variety of economic policy measures. I also find that citizens perceive a candidate with current working class status as less competent, less intelligent, less hardworking and more liberal than his upper class counterpart. Additionally, I show that individuals prefer a candidate that has worked his way up as opposed to a candidate with a privileged upbringing who retained it. I find evidence that a voter's partisan identification moderates this relationship, with Democrats being more susceptible to being primed by class information than Republicans. I also show that a candidate's social class affects respondents' vote intentions largely by altering their perceptions of the candidate's ability to understand the concerns of people like them. This study is the first to treat social class as a heuristic and has implications for understanding how voters make inferences about class and incorporate these inferences into their vote calculus.