Religious congregations and crime incidents: opportunity and bias

ABSTRACT

Religious congregations’ experiences with crime victimization are an under-researched and poorly understood phenomenon. Religious congregations represent a street crime target as a potential source of money, electronics, furniture and other goods, or as places where potential victims come and go. Furthermore, religious congregations are potential targets of bias crimes, especially certain minority religious traditions. This study examines survey data from a national sample of 1,385 religious congregations to investigate the role of religious tradition, criminal opportunity/routine activities factors, and surrounding structural disadvantage on criminal events experienced by congregations. We examine overall, property, and violent crime. We find that opportunity factors and structural disadvantage foster overall reported crime, as well as property and violent crime separately. By contrast, religious tradition is only associated with violent crime, specifically threats of violence. Jewish and Muslim congregations are more likely to report experiencing violent incidents, and report experiencing them more often.

KEYWORDS:

• Religion
• victimization
• religious congregations

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. Our sample did not include Roman Catholic congregations, for two reasons: 1) Roman Catholic congregations only represent 6% of the congregational population, and more importantly, 2) we expected that the risks faced by Catholic congregations would parallel those found in some other types of congregations (e.g., large, urban, predominantly Hispanic, etc.) Thus, we chose to focus our limited survey resources on generating as large a sample of diverse types of Protestant, contested, and religious minority congregations as possible.

2. Sixty-three percent of respondents indicated that they were the congregation’s ‘head religious leader,’ 9% stated that they were a ‘supporting religious leader,’ and 10% stated they were a manager or secretary. The rest indicated that they were a leader or member of the congregation’s board, an executive director, or some combination of other categories.

3. In this paper, we report a differing response rate than [[authorship identifying citation]], which reported the response rate as 15.4%. This is because in reviewing the data further, we determined that there was a higher number of congregations than originally thought that had closed, had bad addresses, invalid contact information that could not be looked up online, or other problems. This makes the response rate a very slightly better 17.5%.

4. One hundred and twenty-two congregations had missing data on at least one of the independent variables. In order to conduct our analyses with the full dataset, we used the dummy variable adjustment method for handling missing cases (Allison 2002). When congregations had missing data for the religious tradition, money received last fiscal year, or the percent membership live within five-minute variables, we created a dichotomous measure for missing data. For the weekly worship attendance variable, we created a dichotomous measure that was equal to 1 for missing data and 0 otherwise. The dummy variable adjustment method is appealing and intuitive as it uses all the information available, but it can sometimes produce biased estimates (Allison 2002). We ran our models with and without the dummy variable adjustments, and each method produced substantively the same results. We also ran the models with congregations with missing data deleted listwise, and with mean substitution, and results were substantively the same. Thus, we are confident that our results are robust to different strategies for accounting for missing data.

5. We estimated a number of alternative model specifications involving the congregation surrounding context variables. First, in models with only the congregation characteristics and none of the contextual variables included, the effects of the congregation characteristics are substantively the same. Second, we estimated models that omitted the non-significant census tract population size variable, and this made no substantive difference to our findings. Third, we included different specifications of the concentrated disadvantage scale which did not include census percent non-Hispanic black, and instead included this as a separate predictor. In these models, the disadvantage variable was positive but non-significant, and percent non-Hispanic black had a significant positive effect on overall crime and property crime. Thus, it appears that including percent black in the concentrated disadvantage scale contributes much of the disadvantage scale’s predictive power for overall and property crime.

Additional information

Funding

This work was supported by the National Science Foundation [SES-1349728];

Notes on contributors

Jeffery T. Ulmer

Jeffery T. Ulmer is Professor of Sociology and Criminology at Penn State University. His work spans topics such as courts and sentencing, criminological theory and symbolic interactionism, religion and crime, and violent crime rates. He has been awarded funding for his research from the National Science Foundation, National Institute of Justice, the Pennsylvania Interbranch Commission on Gender, Racial, and Ethnic Fairness, and others. He received the 2001 Distinguished New Scholar Award and the 2012 Distinguished Scholar Award from the American Society of Criminology’s Division on Corrections and Sentencing. He and coauthors won the American Society of Criminology’s 2012 Outstanding Article Award. He and Darrell Steffensmeier were also awarded the ASC’s 2006 Hindelang Award for Confessions of a Dying Thief: Understanding Criminal Careers and Illegal Enterprise (2005, Transaction).

Christopher P. Scheitle

Christopher P. Scheitle received his Ph.D. in sociology from Penn State University in 2008. His research examines the social structure and dynamics of religion in the United States, with a focus on three specific areas. The first explores the relationship between religion and science. The second area looks are religious discrimination and victimization against congregations and individuals. A third area examines innovations in how religion is organized in the United States, especially regarding the growth of so-called parachurch organizations. He has published three books and over fifty-five scholarly articles. His most recent book is Religion vs. Science: What Religious People Really Think (2017, Oxford, with Elaine Howard Ecklund).