Urban-Rural Differences in Non-Voting Political Behaviors

• Published date: 01 June 2023
• DOI: http://doi.org/10.1177/10659129221119195
• Author: Jennifer Lin,Kristin Lunz Trujillo
• Date: 01 June 2023
• Subject Matter: Articles

Article

Political Research Quarterly

2023, Vol. 76(2) 851–868

© The Author(s) 2022

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/10659129221119195

journals.sagepub.com/home/prq

Urban-Rural Differences in Non-Voting

Political Behaviors

Jennifer Lin

1

and Kristin Lunz Trujillo

2,3



Abstract

Existing studies on the contemporary U.S. urban-rural divide have neglected its potential role in non-voting political

participation. Theoretically, there are mixed expectations: for example, higher social capital in rural areas, alongside a

generally older population, suggest rural areas should have greater political participation. Conve rsely, lower socio-

economic indicators and more physical distance barriers suggest the opposite. Using nationally stratif‌ied survey data from

the 2018 CCES (N= 61,000) and 2020 CES (N= 60,000), we f‌ind that specif‌ic participation behaviors do not consistently

vary across the urban-rural spectrum, controlling for demographic variables. The few consistent ly signif‌icant differences

relate to the nature of location-based access. For instance, using 2020 and 2021 ACLED data, we f‌ind that an activity

where non-rural residents participate more—protesting—occurs less often in rural areas, thus stymieing participation

opportunities for rural and small-town residents. Alternatively, rural and small-town residents are consist ently more

likely to put up a sign, which may ref‌lect a greater incidence of living in houses with yards compared to urban residents.

Social media political participation behaviors do not yield urban-rural differences, further suggesting that once geographic

access-related barriers are removed, participation rates are essentially similar across the urban-rural spectrum.

Keywords

political participation, urban, rural, political behavior, geography, access

The vast majority of recent work on American urban-rural

political differences focuses on voting patterns, parti-

sanship, and political attitudes (Cramer Walsh 2012;

Gimpel et al. 2020;Johnson and Scala 2020;Lunz Trujillo

2022;Lyons and Utych 2021;McKee 2008;Mummolo

and Nall 2017;Munis 2020;Nelsen and Petsko 2021;

Parker et al. 2018;Scala and Johnson 2017). This work

points to rural areas being more right-wing or anti-

establishment in nature, compared to non-rural areas.

There is, however, a gap in understanding whether urban-

rural differences in various forms of political participation

exist, particularly among non-voting political behaviors.

If there are indeed urban-rural variations in participation,

this could impact to what extent urban-rural divisions in

preferences are implemented. Such differences would also

inform how disparate groups participate or not in politics,

which could create group-based inequalities in represen-

tation. For instance, disproportionately high urban partici-

pation could edge out rural interests, or vice versa. How,

then, doespolitical participation in the contemporary United

States differ ac ross the urban -rural spectr um, if at all?

The theoretical expectations of urban-rural political par-

ticipation suggest opposing effects. First, demographic

differences (such as variation in age or education level across

population density) may be driving urban-rural participation

splits, though in different and sometimes countervailing ways.

Second, rural areas tend to be higher in social capital—which

is associated with increased political participation—compared

to non-rural areas (Mazumdar et al. 2018;Putnam 2000,205).

Third, rural identity and rural resentment may impact rural

participation, though it is unclear in which direction. Finally,

there are differences in access to avenues of participation in

rural versus non-rural areas based on location. For example, if

protests tend to occur in urban centers, they would be much

more accessible for a suburban or urban resident than for a

rural resident, on average.

1

Northwestern University, Evanston, IL, USA

2

Harvard University, Cambridge, MA, USA

3

Northeastern University, Boston, MA, USA

Corresponding Author:

Kristin Lunz Trujillo, Network Science Institute, Northeastern

University, 144 Huntington Avenue, 10th Floor, Boston, MA 02116,

USA.

Email: k.lunztrujillo@northeastern.edu

Due to these countervailing pressures on political par-

ticipation, then, it is unclear whether political participation

should be more prevalent among rural or urban residents. Or,

it is also reasonable to expect a lack of urban-rural differ-

ences in political behaviors because these differing expec-

tations may cancel out one another. In addition, many

political behaviors can be accomplished regardless of lo-

cation due to online, phone, and mail-based modes of

participation, thus promoting similar access across geo-

graphic space, especially as the increased “nationalization”

of politics in the United States minimizes sub-national op-

portunities for political participation (Hopkins 2018).

Using two nationally representative surveys of

American adults, the 2018 Cooperative Congressional

Elections Study (CCES; N= 61,000) and the 2020

Cooperative Election Study (CES, N= 60,000), we f‌ind

that there are often no consistent differences in how

rural/small-town versus suburban and urban residents

participate in specif‌ic behaviors, particularly after

controlling for individual-level demographic variables.

Measuring rurality in terms of self-categorization versus a

Census-def‌ined place of residence does not result in sys-

tematic differences in descriptive self-identif‌ied participa-

tion rates.

Further, the two areas where there are consistent urban-

rural differences are those where physical location comes

into play: protesting and putting up a political yard sign.

Rural areas have higher levels of residence in houses

versus apartments, compared to non-rural areas (Mazur

2017); since rural residents are more likely to have a yard,

there are more opportunities to put up a sign. Similarly,

using data from the Armed Conf‌lict Location and Event

Data (ACLED) Project, we f‌ind that protests are mod-

erately correlated with urbanicity in 2020 and 2021, even

controlling for left-leaning respondents.

1

This provides

additional evidence to our overall argument that the un-

derlying mechanism for the few urban-rural non-voting

behavioral differences relate to geographic location itself.

In other words, we f‌ind access to be key. Since many

political behaviors can be done using the internet, phones,

or mail, inherent differences in population density-related

access are removed in nearly all political behaviors ex-

amined. The few differences that are robust to year (and

controlling for other factors) relate to participation be-

haviors that still vary in access determined by physical

location. In line with these results, we also f‌ind that there

is no consistent difference in urban-rural political be-

haviors on social media once other factors are controlled

for, at least among the mostly online samples used. In

other words, o nce urban-rural difference s in internet access

and demographics areaccounted for, political participation

on social media does not vary across the urban-rural

spectrum suggesting that there is no inherent difference

once access to the internet is available.

Urban-Rural Differences in Voting

Behavior and Partisanship

The most researched and discussed urban-rural political

differences in the U.S. are vote choice and political af-

f‌iliation. The U.S. urban-rural divide in vote choice has

existed for quite some time—in the 1960s, there was

approximately a 15-point difference in presidential vote

share between the most and least population dense areas

(Rodden 2019,6)—though this division has especially

widened since the mid-2000s (Gimpel et al. 2020;McKee

2008;Scala and Johnson 2017). In other words, there has

been a longstanding and increasing trend of less pop-

ulation dense areas being more Republican and sup-

porting right-wing candidates, while more population

dense areas tend to support left-wing candidates and

identify as Democrats (Gimpel et al. 2020).

Why is there an urban-rural difference in vote choice and

partisanship? Typical explanations either posit demographic

sorting, economic differences across the urban-rural spectrum,

or deep-seated sociocultural differences and values. The f‌irst

of these—demographic sorting—attributes the urban-rural

divide to certain demographic groups being more prevalent

in population sparse areas, while other groups tend to live in

more population dense areas. Younger and more educated

individuals are increasingly f‌inding jobs in cities, and whites

(especially older individuals) have been moving out of city

centers or have remained in rural America (Kaufman 2019b;

Rodden 2019). In other words, demographic groups dis-

proportionate to rural areas also align neatly onto the Re-

publican base, which tends to be, among others, whiter, male,

lower in education level, more evangelical Christian, and so

on. According to this line of thinking, then, the urban-rural

divide results from geographic demographic sorting, with

right-wing sympathizers increasingly moving to (or remaining

in) rural areas, and left-wing supporters concentrating in urban

areas (Bishop and Cushing 2009;Brown and Enos 2021).

Though urban-rural political divisions to some extent are

accounted for by differences in demographic makeup across

population density, it is also worth noting that the prevalence

of geographic sorting has been challenged empirically by

different researchers (such as: Abrams and Fiorina 2012;

Darmofal and Strickler 2016;Gimpel and Karnes 2006;

Mummolo and Nall 2017).

The urban-rural divide may also be based on economic

differences. An out-migration of educated workers alongside a

poor recovery since the Great Recession in terms of county

job, wage, and GDP growth (Pender et al. 2019) has caused

some rural areas to experience f‌lagging economic conditions

(Kaufman 2019b;Wuthnow 2018) relative to non-rural areas.

Rural-specif‌ic sectors, such as agriculture, manufacturing, and

mining have been hit particularly hard after the Recession due

to a commodity price drop, as well as long-term sector decline

(Pender et al. 2019;Rodden 2019). Skills training and higher

852 Political Research Quarterly 76(2)