Religious culture and rural car ownership

https://doi.org/10.1016/j.jtrangeo.2021.103035Get rights and content

Abstract

Transportation policy generally restricts the predictors of car ownership to socio-economic variables without consideration of cultural factors. However, culture – the ideas, norms, and objects that a society shares – affects consumption decisions. Excluding this cultural context may limit consideration of potentially successful policy interventions. The research presented here statistically tests the impact of religious culture on rural car ownership in the United States by incorporating religious adherence rates into statistical models of countywide motorization rates. This research explores whether religious affiliation is a statistically significant predictor of car ownership and whether the addition of those religious variables results in a statistically significant improvement in model fit. The findings suggest that religious affiliation, expressed as countywide adherence rates, is a statistically significant predictor of car ownership and that its inclusion improves model fit; however, while statistically significant, that improvement in model fit is quite small. This research concludes that considering culture is a valuable direction for targeting car consumption policies aimed at curbing climate change. This research also demonstrates the need for additional exploration using disaggregated cultural data to better understand the import of this approach to policy development.

Introduction

Motorization rates continue to rise globally despite the known linkage between car ownership and climate change. This paper seeks to expand the purview of transportation research to consider the impact of cultural factors, specifically religious affiliation, on the decision to own private vehicles. Typical predictive models of car ownership emphasize only socio-economic variables. While this approach has many benefits (predictor variables can be readily drawn from census instruments; messy cultural constructs that are politically charged, such as race and religion, never enter policy discussions; statistical relationships derived from data collected in one location are applied elsewhere; etc.), it excludes the cultural context within which people make decisions. This paper claims that car ownership, like many consumption behaviors, is influenced by one's peers in a cultural milieu.

Acknowledging rather than suppressing culture's influence on motorization paves a new avenue for policy intervention. Many fields, from product design to advertising to comedy writing, profitably embrace the cultural context of decision-making. Other policy-oriented fields, notably public health, employ cultural information to better target social programs. This research suggests that transportation planning might productively emulate this strategy to advance policy goals.

Culture is typically defined broadly as “the shared products of society” (Flora et al., 1992) decomposed into three related dimensions: ideas, norms, and objects (Bierstedt, 1963; Giddens et al., 2011). Ideas encompass shared values and knowledge. Norms represent accepted practices or ways of doing things. Objects include all the material products that a society uses for daily life or enjoys for symbolic purposes, such as art. Individuals enter a culture by accepting these three components.

A challenge of incorporating culture into statistical research is that culture is difficult to quantify (Hristova et al., 2018) and attempts to externally tag culture veers perilously close to stereotyping (Flora et al., 2004). One approach to coding culture is through religious affiliation, as congregations and their adherents can be counted (Grammich et al., 2012). Religion represents a subcategory of culture “which is based on the idea of the sacred, and which unites believers into a socio-religious community” (Scott, 2014). This research explores the impact of religious culture on the norm of owning private vehicles.

At first glance this linkage may seem far-fetched or applicable in only relatively exotic cases, such as the Amish rejection of motorized vehicles in the United States or the political-religious decisions to first ban and then allow women to drive automobiles in Saudi Arabia (Construction Week, 2019; PwC, 2018). Religion is rarely associated with differing rates of car ownership in research or planning practice in the Global North, but religious institutions are among the most powerful transmitters or "agents of culture" (Flora et al., 1992). Religious beliefs represent the ideas that are translated into norms of behavior which structure object purchases.

These impacts can be seen in two non-transportation but quite specific home improvement/design interventions. Orthodox Jews observe dietary restrictions that fully segregate meat and milk products. This observance requires two entire sets of pots, pans, dishes, and utensils – and often the purchase of a second dishwasher as a convenience (Smith, 1996). While having two dishwashers would represent non-conforming material culture for most social groups, it is well within the range of normal for adherents of Orthodox Judaism. Latter Day Saints (LDS), known colloquially as Mormons, value self-reliance and, consequently, are enjoined to store a year's supply of food and water (Pond, 2007; Valora, 2012). This goal of home food storage has made room-sized pantries a common design feature in predominantly Mormon communities along Utah's Wasatch Front region. These examples show how religious institutions can socialize members to conform to acceptable consumption behaviors ancillary to specific spiritual beliefs.

Conversely, socialization aspects beyond the spiritual can attract individuals to specific religious groups. This feature is starkly evident in the emergence of Cowboy Churches, which, while Evangelical Protestant, are particularly distinguished by their express embrace of western heritage (i.e. “cowboy”) culture. Cowboy churches meet in remote barns, revival tents, and rodeo arenas surrounded by hay bales, dispense with hymnals, prayer-books, bibles, organs, and choirs in favor of singing along with a guitar or band, encourage a come-as-you bring-your-dog jeans-and-boots aesthetic, and, most famously, baptize adherents in horse troughs. The movement expressly targets working cowboys but is composed of a substantially larger set of rural residents who identify with that culture (Dallam, 2018; Williams, 2011). Religious affiliation, therefore, can serve as shorthand for a larger cultural construct.

Those constructs might themselves influence consumption decisions, including decisions tied to transportation. In introducing the Cowboy Church experience, Dallam (2018, pp. 2–3. Italics added) writes that after the service, “people are up, headed for the door, hopping into their pickups and creating clouds of dust as they pull onto the road. This is cowboy church.” The description implies a linkage between religious culture and vehicle choice. The current research goes further to suggest that linkage may extend to the decision of how many vehicles to own.

This research explores this relationship in rural counties in the continental United States. The selection of rural counties serves two important purposes. First, it controls for the reported differences in urban and rural car ownership patterns (Newmark and Haas, 2015) to explore car ownership in areas with few alternatives to driving. Second, it leverages the stronger role of religious institutions in rural areas (Lapping et al., 1989) to increase the likelihood that the religious affiliation rate used in this analysis is a valid measure of cultural identity. The selection of the continental United States eliminates the unique attributes of the island states and territories (e.g. Hawaii, Puerto Rico, Guam, etc.) and the sui generis landscape of Alaska. Given rural demographics, this selection does limit the religious diversity of the sample almost entirely to denominations of Christianity.

This particular research question arose anecdotally in earlier work by the authors to identify successful policies for reducing rural car ownership (Rearick and Newmark, 2018). That project used socio-economic variables to statistically identify outlier rural counties with lower than expected car ownership growth. While this sorting was designed to reveal effective land use or transportation practices, in most of the cases there were no planning interventions. Instead there was a growing presence of Anabaptist adherents (specifically, Amish in Indiana, Old Order Mennonites in New York, and Hutterites in Montana), whose religious practices either proscribe or reduce the ownership of motor vehicles.

The current work statistically tests the impact of religious culture on countywide car ownership by incorporating religious affiliation rates into statistical models of motorization in rural counties in the continental United States. This research explores whether religious affiliation is a statistically significant predictor of car ownership and whether the addition of those religious variables results in a statistically significant improvement in model fit. The findings suggest that religious affiliation is both a statistically significant predictor and that its inclusion improves model fit; however, while statistically significant, that improvement in model fit is quite small. This research concludes that considering culture is a valuable direction for targeting policies aimed at curbing climate change. This research also demonstrates the need for additional exploration using disaggregated cultural data to better understand the import of this approach.

Section snippets

Literature review

The research literature on the full panoply of travel behaviors typically omits consideration of culture. For example, a sampler of recent studies of car ownership, the focus of this current research, excludes explicitly cultural elements (Dargay and Hanly, 2007; Dargay, 2002; Flamm, 2009; Haas et al., 2016; Oakil et al., 2013; Pyddoke and Creutzer, 2014; Schimek, 1996) although some work on attitudes regarding car purchases (Flamm, 2009; Hafner et al., 2017) might be seen as implicitly

Methodology

This quantitative research combines data from the U.S. Census Bureau and the Association of Statisticians of American Religious Bodies. Multiple regression analysis is used to estimate the motorization rates of rural counties in the continental United States with a reduced model based on traditional socio-economic variables. Additional models build off the reduced model to incorporate rates of affiliation to different religions. Finally, analysis of variance between the full and reduced models

Data collection

This research identified counties within the contiguous 48 states as rural if their U.S. Department of Agriculture (USDA) Economic Research Service (ERS) 2013 Rural-Urban Continuum Codes are above three, signifying that these are nonmetro areas. (For reference, metro areas in the U.S. are defined as having “at least one urbanized area of 50,000 or more population, plus adjacent territory that has a high degree of social and economic integration with the core as measured by commuting ties” (

Variable selection

The availability of the religious affiliation rate data constrains the unit of analysis to the county level. Consequently, the statistical models estimated are aggregate models designed to explore relationships between religious affiliation and car ownership on a countywide rather than individual or household basis.

Model estimation

Four multiple regression models were estimated to predict county motorization rates. The first model used only the socio-economic variables as predictors and serves as the base model. The second model used only the main religious culture groupings as predictors to see if these variables alone are statistically significant. The third model combined the base and main group variables to see if the religious culture information is significant while controlling for the socio-economic variables. The

Discussion

The purpose of this research is to explore whether culture adds explanatory power to socio-economic models of motorization. The specific cultural attribute chosen for this exploration is religious affiliation, which is both a tractable means to quantify culture and a powerful influencer of individual and household decision-making. Exploring religious questions in the United States, however, introduces known analytical challenges. The United States maintains a constitutional commitment to a

Conclusions

This paper finds that religious affiliation is a statistically significant predictor of car ownership rates for rural counties but does not substantially improve prediction relying solely on socio-economic variables. These findings suggest both the importance of cultural variables for structuring transportation decisions and the need to better understand the causal path of these relationships, particularly at the household level, to inform planning and policy. Future research on the nexus of

Acknowledgement

The authors thank an anonymous reviewer from the Transportation Research Board Standing Committee on Social and Economic Factors of Transportation for encouraging this analytical direction as well as Perla Reyes, Michael Higgins, and David Slusky for providing statistical guidance. Initial research efforts were funded by a Dwight David Eisenhower Graduate Transportation Fellowship from the U.S. Department of Transportation and a Leslie E. and Stella M. Ellithorpe Student Research Award from

References (54)

  • N. Collins-Kreiner

    Pilgrimage Tourism in the Holy Land: the behavioral characteristics of Christian pilgrims

    GeoJournal

    (2000)
  • Construction Week

    New women drivers to support 10 million vehicles on Saudi roads

  • M.W. Dallam

    Cowboy Christians, 1st

    (2018)
  • N. Delener

    The effects of religious factors on perceived risk in durable goods purchase decisions

    J. Consum. Mark.

    (1990)
  • W. Elias et al.

    Gender and travel behavior in two Arab communities in Israel

    Transp. Res. Rec.

    (2008)
  • C.B. Flora et al.

    Rural Communities: Legacy and Change

    (1992)
  • C.B. Flora et al.

    Rural Communities: Legacy and Change

    (2004)
  • J.H. Foard

    The boundaries of compassion: Buddhism and National Tradition in Japanese pilgrimage

    J. Asian Stud.

    (1982)
  • A. Giddens et al.

    Introduction to Sociology

    (2011)
  • C. Grammich et al.

    2010 U.S. Religion Census: Religious Congregations & Membership Study

    (2012)
  • P.M. Haas et al.

    Untangling housing cost and transportion interactions: the location affordability index model—version 2 (LAIM2)

    Hous. Policy Debate

    (2016)
  • L. He et al.

    Religion and perceived travel risks

    J. Travel Tour. Mark.

    (2013)
  • D. Hristova et al.

    The new urban success: how culture pays

    Front. Phys.

    (2018)
  • M.B. Lapping et al.

    Rural Planning and Development in the United States

    (1989)
  • K.R. Lord et al.

    Religious influence on consumer behavior: classification and measurement

    ACR North Am. Adv.

    (2005)
  • A.S. Mattila et al.

    The impact of gender and religion on college students’ spring break behavior

    J. Travel Res.

    (2001)
  • G.L. Newmark et al.

    Income, Location Efficiency, and VMT: Affordable Housing as a Climate Strategy

    (2015)
  • View full text