Are Perceptions of Local Conditions Rooted in Reality? Evidence From Two Large-Scale Local Surveys

• DOI: 10.1177/1532673X19885863
• Published date: 01 July 2020
• Date: 01 July 2020
• Author: Aaron C. Weinschenk,Thomas M. Holbrook
• Subject Matter: Articles

https://doi.org/10.1177/1532673X19885863

American Politics Research

2020, Vol. 48(4) 467 –474

© The Author(s) 2019

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/1532673X19885863

journals.sagepub.com/home/apr

Article

Although the literature on retrospective voting has largely

focused on national- and state-level politics (Atkeson &

Partin, 1995; Cohen & King, 2006; Fiorina, 1981; Healy &

Malhotra, 2013; Lewis-Beck & Stegmaier, 2000; Stein,

1990), the question of whether local electorates hold local

political officials accountable for local conditions has started

to attract serious attention, producing a body of research

indicating that local conditions, such as the state of the local

economy, taxes, schools, and crime, are connected to local

election outcomes (Arnold & Carnes, 2012; Berry & Howell,

2007; Burnett & Kogan, 2016; Holbrook & Weinschenk,

2014b; Hopkins & Pettingill, 2018; Howell & Perry, 2004;

Kaufmann, 2004; Lay & Tyburski, 2017; Oliver & Ha, 2007;

Oliver, Ha, & Callen, 2012). Implicit in these studies of local

elections is the important assumption that voters have some

understanding of the state of local conditions; that local elec-

torates can distinguish between good times and bad times.

Although Howell and Perry (2004) note that “there are many

areas in which city residents can reasonably evaluate local

performance,” we actually know very little about whether

perceptions of local conditions are connected to reality

(p. 38).

The question of whether people make accurate assessments

about political, social, and economic conditions has long been

of interest to scholars (Duch, Palmer, & Anderson, 2000;

Holbrook & Garand, 1996; Niemi, Bremer, & Heel, 1999),

although there is disagreement about the extent to which

perceptions match reality. Recently, Achen and Bartels (2016)

have argued that voters are quite irrational and generally do

not make well-informed retrospective assessments about

political and economic conditions. Instead, according to

Achen and Bartels (2016), most people make political evalua-

tions on the basis of political and social identities rather than a

sincere assessment of reality. Other scholars, too, have found

evidence that the public’s perceptions of conditions are not

always fully rooted in reality (Bartels, 2002; Duch et al., 2000;

Gramlich, 2016; Shao & Goidel, 2016). To be fair, some stud-

ies are more optimistic about the capability of the public to

connect their assessments of conditions to reality. For exam-

ple, Niemi et al. (1999) show that “state economic perceptions

are clearly grounded in economic reality, that is, in the actual

conditions of the state” (p. 188). Both Franko (2017) and Xu

and Garand (2010) find that perceptions of income inequality

are connected to objective measures. In addition, Erikson and

Wlezien (2012) and Lewis-Beck, Martini, and Kiewiet (2013)

find strong relationships between objective economic

885863APRXXX10.1177/1532673X19885863American Politics ResearchHolbrook and Weinschenk

research-article2019

1University of Wisconsin–Milwaukee, USA

2University of Wisconsin–Green Bay, USA

Corresponding Author:

Aaron C. Weinschenk, Department of Political Science, University of

Wisconsin–Green Bay, P.O. Box 413, Green Bay, WI 54311, USA.

Email: weinscha@uwgb.edu

Are Perceptions of Local Conditions

Rooted in Reality? Evidence From

Two Large-Scale Local Surveys

Thomas M. Holbrook1 and Aaron C. Weinschenk2

Abstract

In this research note, we test an assumption that is often made in the literature on local retrospective voting—that peoples’

perceptions of local conditions are well-grounded in reality. To do so, we examine the relationship between objective

measures of local conditions and aggregated survey measures of perceptions of those conditions. We focus on three different

conditions that have been shown to influence vote choice and approval at the local level—the state of the local economy,

the quality of local public schools, and levels of local crime—and find strong evidence that perceptions of these conditions

reflect actual local conditions. This important and previously unreported finding helps bolster the connections some scholars

have found between objective indicators and election outcomes at the local level, as those indicators are tied to mass

perceptions of related local conditions, which are connected to evaluations of incumbents. Overall, our results indicate that

local electorates are well-positioned to hold local officials responsible. Given the general conception of the local electorate

as disengaged, the strength and consistency of our findings are somewhat unexpected.

Keywords

local politics, elections, retrospective voting