Community, faith, and public violence: A county-level examination of religious institutions and mass public shootings in the United States

ABSTRACT

Americans confront public forms of violence on a strikingly regular basis. Shootings within schools, religious institutions or simply crowded spaces have attracted significant media attention. The question remains, then, as to what role, if any, can existing institutions play in mitigating these factors – and, if so, what institutions? I argue that the presence of religious institutions will decrease the likelihood of a mass public shooting occurring. Religious institutions are particularly effective at building social capital. Mass public shootings in particular speak to widespread alienation from parochial environments. I test this theory using county-level data on religious congregations and mass public shootings for the lower 48 states. I find that that over the range of religious congregation presence, the likelihood of a mass public shooting declines from 1 in 40 to effectively zero. This result is robust across multiple definitions of ‘mass public shooting’ but decreases in effectivity as the definition is broadened to other types of mass shootings. This study fills a well-documented void of county-level structural analysis in the existing literature.

KEYWORDS:

• Mass public shootings
• religion
• social capital

Acknowledgments

The author would like to thank Professor Gregory A. Robinson of Binghamton University, Professor Nathan Monroe of University of California, Merced, and the reviewers for their input and support

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. Throughout this study, I will use the phrase ‘religious institution’ to refer to the theoretical concept, and the phrase ‘religious congregation’ to refer to its empirical manifestation; I do so to draw both a distinction and logical connection between the two. I discuss the theoretical definition in the section titled ‘Religion and Social Capital’, and the empirical definition in the research design.

2. Many of the scholars included in the following section use the phrase ‘social disorganization’ to describe communities lacking in social capital; these phrases are functionally equivalent in describing a community’s capacity for collective action.

3. As this study focuses on mass public shootings, it is not necessary to discuss the definition of rampage school shootings; they are defined by Newman et al (50).

4. The discussion of Christian nationalism and gun violence is somewhat tangential here, although I have sought to account for it empirically below. Preliminary tests of my own doing indicate that the presence of religious congregations is in fact negatively associated with the presence of Christian Identity groups. Such research, however, falls outside the purview of this particular and warrants further examination in its own right.

5. These same studies also avoid using region fixed effects; I included particular regions of interest in the factor analysis, as discussed below.

6. For a discussion on the definitions of ‘mass public shootings’, ‘active shootings’ and ‘failed mass shootings’, see Freilich et al 2020.

7. As per the NRCRD, these institutions include bowling leagues, recreation facilities, golf and sports facilities, and civic, labor, political, professional and business associations. The NRCRD features data for 2009 and 2014. In order to adjust for the 2010 census, I used rounded, imputed values.

8. I also repeated these tests using census region and division fixed effects in place of southern and midwestern states in the factor analysis, as well as rural population percentage in place of the rural-urban continuum. Such tests did not have an impact on any of the main text findings and little-to-no impact on appendix findings. For the sake of simplicity, I have used the tests following the factor analysis guidelines as laid out in Land, McCall, and Cohen (1990).