Who Else Is Running? Reference Dependence in Candidate Evaluations

• Author: Eric Loepp
• DOI: 10.1177/1532673X18803881
• Published date: 01 March 2020
• Date: 01 March 2020
• Subject Matter: Articles

https://doi.org/10.1177/1532673X18803881

American Politics Research

2020, Vol. 48(2) 238 –251

© The Author(s) 2018

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/1532673X18803881

journals.sagepub.com/home/apr

Article

Models of candidate evaluation reasonably assume individu-

als review a candidate’s political attributes and express pref-

erences based on those characteristics. Scholars have had

success identifying causal effects associated with features

such as race (e.g., Mcdermott, 1998; Sigelman, Sigelman,

Walkosz, & Nitz, 1995; Terkildsen, 1993), gender (e.g.,

Huddy & Terkildsen, 1993a, 1993b; Kahn, 1994; King &

Matland, 2003), and partisanship (e.g., Rahn, 1993;

Schaffner, Streb, & Wright, 2001). Such studies, however,

are limited in that voters rarely, if ever, evaluate a single can-

didate in isolation. Instead, save for contests featuring an

unchallenged candidate, they compare multiple individuals

simultaneously. An individual’s perception of a particular

candidate may be partially a function of the larger candidate

choice set from which individual voters must make electoral

judgments. Drawing on behavioral economics and the con-

cept of reference dependence, I argue that the same candidate

may be perceived in systematically different ways based on

the electoral context in which they are evaluated.

This project contributes to our understanding of how vot-

ers form impressions of singular candidates in a dynamic

political environment. Specifically, I test the claim that the

same political candidate will enjoy higher or lower levels of

support among voters based on (a) their electoral competi-

tion and (b) ideological characteristics of individuals evalu-

ating them. To do this, I first discuss the concept of reference

dependence and apply its logic to American elections. I then

present the design of an experiment in which voters are asked

to evaluate candidates either (a) in isolation or (b) contrasted

with a challenger, thereby facilitating analysis of whether or

not the presence of political competition informs evaluations

of individual candidates running for office. Results demon-

strate that voters in both political parties are susceptible to

challenger effects, although not in the same way. I close with

a discussion of the work’s implications both for scholars ana-

lyzing elections as well as elites deciding whether or not to

participate in them.

Reference Dependence in an Electoral

Context

Behavioral economists and psychologists have long noted

that individuals have a tendency to perceive particular phe-

nomena based on reference points (Emery, 1970; Monroe,

1973; Tversky & Kahneman, 1991). Tversky and Kahneman

(1974), for instance, find in a simple experiment that student

estimates of the quantity 8 × 7 × 6 × 5 × 4 × 3 × 2 × 1 are

roughly 4 times as large as estimates of the equivalent 1 × 2

× 3 × 4 × 5 × 6 × 7 × 8. Consumer research uncovers a

similar pattern: some reference point (or “anchor”) affects

individuals’ perceptions of concepts such as value and attrac-

tiveness of goods (e.g., Bagwell & Riordan, 1991; Biswas &

Blair, 1991; Briesch, Krishnamurthi, Mazumdar, & Raj,

1997; Hardie, Johnson, & Fader, 1993). The best example of

803881APRXXX10.1177/1532673X18803881American Politics ResearchLoepp

research-article2018

1University of Wisconsin–Whitewater, WI, USA

Corresponding Author:

Eric Loepp, University of Wisconsin–Whitewater, 5100 Laurentide Hall,

800 Main Street, Whitewater, WI 53190, USA.

Email: eric.loepp@gmail.com

Who Else Is Running? Reference

Dependence in Candidate Evaluations

Eric Loepp1

Abstract

Research assessing how voters evaluate political candidates often focuses on the effects of particular attributes (e.g., race,

gender, partisanship). I submit that voters’ perceptions of candidates may depend not only on candidates’ own traits and

features, but those of other candidates running against them. Drawing on literature on reference dependence, I argue

that the same candidate may be perceived in significantly different ways depending on whether or not voters evaluate the

candidate as a single entity or as one option in a multicandidate field. An original survey experiment reveals that under certain

circumstances, Republicans and Democrats both adjust their evaluations of party candidates as a function of the presence

of other candidates. I conclude with a discussion of this project’s implications for a larger body of work looking at reference

dependence in American elections.

Keywords

reference dependence, elections, candidates, ideology, primary elections