Ballot Secrecy Concerns and Voter Mobilization

• Author: Gregory A. Huber,Alan S. Gerber,Daniel R. Biggers,David J. Hendry
• Published date: 01 September 2014
• Date: 01 September 2014
• DOI: http://doi.org/10.1177/1532673X14524269
• Subject Matter: Articles

American Politics Research

2014, Vol. 42(5) 896 –923

© The Author(s) 2014

Reprints and permissions:

sagepub.com/journalsPermissions.nav

DOI: 10.1177/1532673X14524269

apr.sagepub.com

Article

Ballot Secrecy Concerns

and Voter Mobilization:

New Experimental

Evidence About Message

Source, Context, and the

Duration of Mobilization

Effects

Alan S. Gerber1, Gregory A. Huber1,

Daniel R. Biggers1, and David J. Hendry1

Abstract

Recent research finds that doubts about the integrity of the secret ballot as

an institution persist among the American public. We build on this finding

by providing novel field experimental evidence about how information about

ballot secrecy protections can increase turnout among registered voters

who had not previously voted. First, we show that a private group’s mailing

designed to address secrecy concerns modestly increased turnout in the

highly contested 2012 Wisconsin gubernatorial recall election. Second, we

exploit this and an earlier field experiment conducted in Connecticut during

the 2010 congressional midterm election season to identify the persistent

effects of such messages from both governmental and non-governmental

sources. Together, these results provide new evidence about how message

source and campaign context affect efforts to mobilize previous non-voters

by addressing secrecy concerns, as well as show that attempting to address

these beliefs increases long-term participation.

1Yale University, New Haven, CT, USA

Corresponding Author:

Alan Gerber, Yale University, New Haven, CT 06520, USA.

Email: alan.gerber@yale.edu

524269APRXXX10.1177/1532673X14524269American Politics ResearchGerber et al.

research-article2014

Gerber et al. 897

Keywords

secret ballot, voter mobilization, field experiment, message source,

participation

How do individuals’ beliefs about the voting process affect political partici-

pation? Can communication during a campaign change those beliefs? If

changes in beliefs affect participation, do those effects endure over time?

Recent research suggests that one impediment to greater participation is

doubt about ballot secrecy among potential voters. In particular, individuals

who have not previously voted are more likely to believe that their vote

choices will be revealed and that election officials and others at the polls may

seek to intimidate them (Gerber, Huber, Doherty, & Dowling, 2013). Building

on this finding, Gerber, Huber, Doherty, Dowling, and Hill (2013) present

field experimental evidence that communication from a government source

about ballot secrecy protections appears to ameliorate these concerns. In that

experiment, treated registrants who had not previously voted participated at

higher rates in the 2010 Connecticut midterm election than those who were

not sent this information.

This article addresses the robustness and significance of that study in two

ways. First, we report results from a new field experiment conducted in

Wisconsin prior to the 2012 recall election, in which Governor Scott Walker

retained office. A private nonpartisan and nonprofit voter mobilization orga-

nization implemented a randomized mailing campaign that provided assur-

ances about ballot secrecy protections to selected registrants who had not

previously voted. We examine whether that intervention increased turnout in

the June 2012 recall election and the November 2012 presidential election.

Second, we assess whether the initial increase in 2010 turnout associated

with the intervention reported by Gerber, Huber, Doherty, Dowling, and Hill

(2013) persists over time by examining turnout in the 2012 primary and gen-

eral elections in Connecticut. In so doing, we both replicate prior research

and assess three important questions about the efficacy of voter mobilization

efforts.

The first question concerns whether campaign mobilization efforts that

attempt to change beliefs can be successful (i.e., persuasive) when under-

taken by private (i.e., non-governmental) actors. Whereas the earlier

Connecticut study measured the effects of reassurance from a government

office, the Wisconsin study explores whether non-governmental actors can

successfully reach and persuade individuals despite their lack of inherent

credibility as overseers of the election process. Alternatively, do members of

the targeted population ignore those private communication efforts or regard