U.S. flag

An official website of the United States government, Department of Justice.

Social Control in the Metropolis: A Community-Level Examination of the Minority Group-Threat Hypothesis

NCJ Number
Justice Quarterly Volume: 20 Issue: 2 Dated: June 2003 Pages: 265-295
Date Published
June 2003
31 pages

This study examined whether there were links between changes in minority representation within a single city (New York City) over time at the community level and changes in the allocation of police resources.


The study focused on 74 New York City police precincts in 1975, 1982, and 1992. All of the precincts were assigned a deployment count for each of the years, for a total of 222 observations. In addition, the NYPD provided annual population counts of police officers who were assigned to city precincts in 1975, 1982, and 1992. Annual averages of the number of sworn personnel were derived by summing the monthly populations and dividing them by 12. Information about community structure was obtained from the U.S. census for 1970, 1980, and 1990. Independent variables measured were crime rates for both property and violent crimes, the percentage of households in poverty, percentage of households receiving public assistance, both adult and male unemployment rates, previous police deployment, and percentage nonwhite racial composition. The dependent variable was change in police deployment over time. The study found that increases in the percentages of Latinos in the precincts have predicted changes in police deployment, and the link between Latino populations and police deployment was nonlinear. Although this finding is consistent with prior studies of the minority group-threat hypothesis that have found curvilinear relationships, the shape of the curve is inconsistent. When previous studies have found nonlinear relationships between minority populations and allocations of police resources, they have generally observed inverted U-shaped curves. The current study found a parabolic curve, which strongly supports Blalock's (1967) power-threat hypothesis. Current results suggest that Latino populations did not become threatening until they represented approximately one quarter of the precinct-level populations, at which point precincts significantly increased their levels of police deployment. 5 tables and 47 references

Date Published: June 1, 2003