Enlisting the Public in Facilitating Election Administration: A Field Experiment

• DOI: http://doi.org/10.1111/puar.12833
• Date: 01 November 2018
• Published date: 01 November 2018

Public Administration Review,

Vol. 78, Iss. 6, pp. 892–903. © 2017 by

The American Society for Public Administration.

DOI: 10.1111/puar.12833.

892 Public Administration Review • November | December 2018

Research Article Enlisting the Public in Facilitating Election Administration:

A Field Experiment

Robert M. Stein is Lena Gohlman

Fox Professor of Political Science at Rice

University, faculty director of the Center

for Civic Engagement, fellow in Urban

Politics at the Baker Institute, and former

dean of the School of Social Sciences

(1996–2006). His research focuses on

election sciences, public policy, and public

opinion. His research has been supported

by the National Science Foundation, Pew

Charitable Trusts, U.S. Department of

Homeland Security, and Kinder Foundation.

E-mail: stein@rice.edu

Andrew Menger is a PhD candidate

at Rice University, where he conducts

research on the effects that election laws

and administration have on political

participation and voter behavior. He expects

to complete his dissertation on voting laws

and the costs of voting in April 2018.

E-mail: andrew.m.menger@rice.edu ;

andrewmmenger@gmail.com

Abstract : The proliferation of election reforms poses a challenge for local election officials (LEOs) charged with

conducting elections. To meet this challenge, LEOs attempt to communicate, inform, and persuade voters how to cast

their ballots in a manner that is efficient and effective for both the voter and the administrator. This article examines

the effects of efforts by LEOs to persuade voters to return mailed ballots before Election Day and in person in order to

facilitate the efficient administration of vote-by-mail elections in Colorado. Field experiments testing the efficacy of

alternative messages find that many messages have no effect on the timing or method of ballot return. Messaging that

focuses on LEOs ’ responsiveness to voters ’ demands is most effective at steering voters to return their mailed ballots in

person but results in later ballot returns.

Practitioner Points

• Experiments are a low-cost and effective way of identifying messaging that motivates voters to change the

timing and method of returning their mailed ballots.

• Messages that remind voters of the convenience of ballot return methods and timing have no effect on when

or how voters return their mailed ballots.

• Reporting accomplishments of local election officials that are consistent with the preferences of voters (a

“responsiveness” message) has a positive effect on in-person return of mailed ballots.

• Encouraging in-person ballot return can result in voters returning their ballots later in the election period,

closer to or on Election Day.

T he adoption by many states of election

reforms such vote by mail (VBM) poses a

challenge for local election officials (LEOs)

who are charged with implementing the new voting

systems. Eliminating or consolidating polling places

allows LEOs to save money by cutting down on

equipment, staff, and location rental costs. However,

an obstacle to the efficient implementation of VBM

elections is the late return of paper ballots, which

adds personnel costs to verifying and counting ballots

and undermines the likelihood that all ballots will be

accurately counted and reported on Election Day. In

Colorado, which recently adopted a VBM system,

election officials sought to encourage the timely return

of ballots through messages aimed at increasing early

ballot return and in-person ballot drop-off, which

avoids potential delays by the U.S. Postal Service.

1

In this article, we examine the effects of different

messages from LEOs on their success at persuading

voters to return their mailed ballots before Election

Day and in person.

Previous efforts to steer how voters cast their ballots

have relied primarily on a narrow set of messages

describing the incentives to voters of using new

voting opportunities. This messaging, which

describes to voters the savings and convenience of

using new voting methods, has produced mixed

results (Michelson et al. 2012 ; Monroe and Sylvester

2011 ). Drawing on Schneider and Ingram ’ s ( 1990 )

description of policy tools, we suggest that messages

designed around appeals to authority and symbolism

may be more persuasive to voters than those

describing direct incentives. We examine a variety

of treatments whose messages focus on different

themes, including voter confidence, convenience,

social benefits, and responsiveness to the voters. Using

the framework of policy tools from Schneider and

Ingram, we describe the mechanisms by which these

treatment messages may influence voters ’ behavior.

We find that a message informing voters how their

local election officials responded to voter demand by

increasing locations for in-person ballot return had

a significant effect of increasing the proportion of

ballots returned in person. However, this treatment

and the social benefits treatment, which was also

intended to increase in-person return, had the

Andrew Menger

Robert M. Stein

Rice University