Abstract
Recent developments in both theory and research on neighborhood crime have focused attention on the role of organizations. The current research builds on the existing literature to examine the relationship of churches to neighborhood “street” and domestic violence. The findings suggest that churches are fairly stable neighborhood organizations and not importantly heterogeneous, and neither factor appears to be related to neighborhood crime. Alternatively, the number of churches in a neighborhood is positively related to both street crimes and domestic assaults, and denomination does play an important role especially as it relates to domestic assaults. Theoretical and empirical implications are discussed.
Notes
1Most theorists and researchers in this tradition currently use the word “institution” rather than organization. Rosenfeld (Citation2010), however, argues that is important to maintain the distinction between institutions, such as religion, which he defines as “broad value complexes and accompanying social positions and roles that together constitute the pillars of society” (1) and their embodiment in local organizations, such as churches. It is an important conceptual distinction that we wish to follow.
2We use the word church to refer to what Chaves (2002) defines as congregants. This includes synagogues, mosques, and temples.
3We also assessed stability in two other ways. First we examined the correlation between the number of churches in a given block group from one year to another. Using this method, we find that the number of churches across the block groups in Norfolk is remarkably stable within the nine-year period measured. The highest correlation was between the years 2008 and 2007 (r = 0.954) and the lowest was between the years 2000 and 2008 (r = 0.700). The average correlation between the number of churches across all the years is r = 0.831. A second way of thinking about stability is to ask whether, and how often, churches move from one location to another. Of the 555 churches that existed at least one year during the time period, only 20 moved from one location to another within the city. An examination of the addresses shows that when churches did move, they often moved only a few blocks away suggesting that they were interested in maintaining local neighborhood congregants. The degree of stability provided by these first two measures precludes the use of either of them in multivariate models.
*p < .05, **p < .01.
4Although criminologists in this area have yet to explore whether the effects of churches on neighborhood crime may vary by the level of socioeconomic disadvantage of the neighborhood, the literature on religion does suggest the possibility. Thus, we did test for interactions between the number of churches and levels of socioeconomic disadvantage but found no significant results for either street or domestic crimes.
*p < .05; **p < .01.
*p < .05; **p < .01.
*p < .05; **p < .01.
*p < .05; **p < .01.
*p < .05; **p < .01.