Should I Cast an Ill-Informed Ballot? Examining the Contours of the Normative Obligation to Vote

• Date: 01 September 2019
• DOI: 10.1177/1532673X18821359
• Published date: 01 September 2019
• Subject Matter: Articles

https://doi.org/10.1177/1532673X18821359

American Politics Research

2019, Vol. 47(5) 1081 –1099

© The Author(s) 2019

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/1532673X18821359

journals.sagepub.com/home/apr

Article

Should I Cast an

Ill-Informed Ballot?

Examining the Contours

of the Normative

Obligation to Vote

David Doherty1, Conor M. Dowling2,

Alan S. Gerber3, and Gregory A. Huber3

Abstract

Proparticipatory norms play a central role in driving turnout. However, a

broad norm that people are supposed to vote cannot explain why some

people fail to participate or why rates of participation vary sharply across

elections. We argue that the norm of voting extends beyond the mere act of

voting. We present empirical evidence supporting the position that the social

rewards for participating are conditional. The social rewards for casting an ill-

informed vote are far smaller than those associated with casting an informed

ballot. Moreover, some low-information voting strategies are viewed as less

desirable than simply abstaining. Our findings illustrate an important constraint

on the capacity of social norms to foster turnout. The effectiveness of efforts

to translate norms into higher rates of turnout may depend on ensuring that

voters are informed enough to cast a meaningful ballot.

Keywords

turnout, norms, informed voting, survey experiment

1Loyola University Chicago, Chicago, IL, USA

2The University of Mississippi, University, MS, USA

3Yale University, New Haven, CT, USA

Corresponding Author:

Conor M. Dowling, Department of Political Science, The University of Mississippi, 235

Deupree Hall, P.O. Box 1848, University, MS 38677-1848, USA.

Email: cdowling@olemiss.edu

821359APRXXX10.1177/1532673X18821359American Politics ResearchDoherty et al.

research-article2019

1082 American Politics Research 47(5)

Norms about voting, and the social consequences of failure to conform to

those expectations, are powerful explanations for voter participation (e.g.,

Blais, 2000; Dalton, 2008; DellaVigna, List, Malmendier, & Rao, 2017;

Gerber, Green, & Larimer, 2008; Riker & Ordeshook, 1968; Rolfe, 2012;

Sinclair, 2012). In the last two decades, academic researchers and campaign

professionals alike have refocused their attention on the combination of inter-

nalized expectations and social enforcement of those expectations as means

to increase and sustain participation. But if normative expectations about par-

ticipation are so powerful, why do so many eligible voters fail to turn out? We

propose and test the argument that voting norms are more complicated than

simply prescribing an expectation that one cast a ballot.

Democratic theory emphasizes the importance of an engaged citizenry.

Some theorists go further, positing that a—perhaps the—critical feature of

effective democratic governance is participation by an informed electorate

(e.g., Brennan, 2012). For example, “folk theories” of democracy posit

that voters should make choices that reflect their own policy preferences

(e.g., Achen & Bartels, 2016). In light of this perspective, one reason vot-

ing norms might not lead to universal turnout in spite of their apparent

power is that the simple act of voting is insufficient to fulfill one’s norma-

tive obligations. Those who are poorly informed about politics and, thus,

unable to connect their policy preferences to the choices they face on

Election Day, may not be rewarded for the mere act of casting a ballot.

Indeed, those who cast an ill-informed ballot may be viewed less favorably

than those who simply abstain.

This article presents evidence assessing the nature and structure of the

normative obligation to vote. We make two contributions to our understand-

ing of these norms. First, we demonstrate that the social returns to voting

depend substantially on how informed one is about politics. In three survey

experiments, we assess whether the social rewards that flow from voting

depend on whether the prospective voter is informed about the choices they

confront. We find clear evidence that they do. Casting an ill-informed ballot

can lead to greater social disapproval than not voting at all. We also examine

how these effects vary by the level of the election (i.e., national, state, local),

and compare them with common metrics of good citizen behavior.

Second, we assess what people think individuals should do when they are

ill-informed about a particular political choice. In other words, if an indi-

vidual is not informed about a choice they face when casting a ballot, what

do people think they should do? We find that some forms of low information

voting are viewed as unacceptably capricious and unrelated to a candidate’s

likely approach to governing (e.g., voting based on name recognition or bal-

lot order) and that it may be more normatively rewarding for individuals to