Minority threat and police strength from 1980 to 2000: a fixed-effects analysis of nonlinear and interactive effects in large US cities

Citation

(2006), "Minority threat and police strength from 1980 to 2000: a fixed-effects analysis of nonlinear and interactive effects in large US cities", Policing: An International Journal, Vol. 29 No. 1. https://doi.org/10.1108/pijpsm.2006.18129aaf.004

Minority threat and police strength from 1980 to 2000: a fixed-effects analysis of nonlinear and interactive effects in large US cities

Stephanie L. Kent and David JacobsCriminologyVol. 43 No. 32005pp. 731-60

This article takes a thorough look at the existing research on threat theory and its focus on the size of minority populations and police strength. Noting that the research that does exist is based on data that may not reflect the current status or trends of the relationships between minority populations and police strength, the authors set out to “provide superior tests of refinements to the racial threat version of conflict theory” (p. 734).

A number of theoretical explanations exist and may apply, creating some primary hypotheses for the instant research. It is predicted that there is a non-linear relationship between the size of the Hispanic population and size of police departments; it is expected that the size of the Hispanic population must reach a threshold level before it is perceived as a threat necessitating additional police personnel. It is predicted that whites are not threatened by black populations that are residentially segregated from their own; in those communities where residences are more segregated there will be less police strength. When main effects are held constant, it is predicted that with higher segregation and higher percentage of blacks there will be fewer officers. And, factoring in regional perceptual differences, it is predicted that there will be fewer police in Southern cities with highest levels of segregation; it is believed that in this region that whites have less contact with blacks and are therefore less threatened by them. The authors also expect to find previously unmeasured period effects (specifically, an increase in the relationship between racial threat and police strength) and political differences (fewer officers in cities that employ city managers).

U.S. census data is analyzed with a pooled time-series analysis for the years 1980, 1990, and 2000, including cities with populations greater than 100,000 in 1980. In 1980 and 1990 the number of cities was 125 and in 2000 the number was 123, for a total of 373 city-years in all analyses. Measured variables include racial threat, ethnic threat, residential segregation, city manager presence, tax base, violent and property crime rates, and social disorganization.

Statistical analysis tended to support the authors’ hypotheses, primarily that police growth is related to population. Social disorganization theories had the weakest results, showing mixed support with the measures used here. Political science theories were supported as greater median family incomes resulted in the ability to support larger police departments. Although property and violent crime failed to explain variation in police strength, a threshold effect was found with murder rates. Ethnic threat was found to explain police strength in all models, but its effects were weaker than those of racial threat. The strength of the relationship between black population size and police strength has increased in the census years 1990 and 2000; this is due in part to increased strength in the relationship between racial threat and the financial resources of the community to respond with increased police strength. Conversely, police strength is reduced in those areas of more defined residential segregation, including the South.

The authors follow their findings with cautions to the reader. While the current study looked at numbers of officers, it did not look at number of officers actually on the street. It cannot be assumed that because a department increases in size, it will increase in beat officers. Also, because the analysis is done at the city level, the authors caution that intervening explanatory factors may have been missed.

J.W. Carter, Luahna Winningham