Structural Covariates of Homicide Rates

• Date: 01 August 2007
• Author: Patricia L. McCall,Paul Nieuwbeerta
• DOI: 10.1177/1088767907304072
• Published date: 01 August 2007
• Subject Matter: Articles

167

Homicide Studies

Volume 11 Number 3

August 2007 167-188

© 2007 Sage Publications

10.1177/1088767907304072

http://hs.sagepub.com

hosted at

http://online.sagepub.com

Structural Covariates of

Homicide Rates

A European City Cross-National

Comparative Analysis

Patricia L. McCall

North Carolina State University, Raleigh, USA

Paul Nieuwbeerta

Netherlands Institute for the Study of Crime and

Law Enforcement (NSCR), Leiden, The Netherlands &

Utrecht University, Utrecht, The Netherlands

Most previous empirical comparative studies of homicide examine homicide rates

across nations or subnational units within a single country. This study is the first in

which a European cross-national city comparison is made. The article aims to provide

insight into the extent that the homicide rates are related to the social and economic

forces characterizing a sample of European cities. Derived from theories rooted in clas-

sic works, including social disorganization, strain, and urbanism, are six hypothesized

effects of structural forces on homicide rates. Analyses show that cities’deprivation and

population structure indexes are strong predictors of homicide rates. The predicted

effects of unemployment rates, population heterogeneity, and age structure on homi-

cide rates, however, were not consistently corroborated by these results. Comparing

Eastern and Western European countries,the authors also find support for the influence

of the country’s level of development on city-level homicide rates.

Keywords: city homicides; cross-national studies; European homicides; homicide rates

Studies of the spatial distribution of crime, including homicide, have their origins in

the cartographic or statistical school in the 19th century in Europe. Pioneers such as

Guerry (1833), Fletcher (1849), and Quetelet (1847) started empirical studies investi-

gating differences in crime rates across European geographical areas. In the same cen-

tury, the French scholar Durkheim (1895/1982) questioned how and why rapid

socioeconomic changes in European societies caused by industrialization and urban-

ization fragmented people’s social ties, thereby freeing them to deviate from social

norms and engage in deviant behavior including crime and homicide. Durkheim was the

first scholar to address the crime problem at a structural theoretical level.

In the 20th century, the focus of empirical ecological research and the develop-

ment of theories on the geographical distribution of crime moved away from Europe

168 Homicide Studies

toward the United States. Ecological research made enormous progress in the first

decades of the 20th century, especially at the University of Chicago. Studying the

city of Chicago, scholars described the distribution of crime and delinquency,

including homicide, across neighborhoods and identified structural factors responsi-

ble for that clustering (Burgess, 1925; Park, 1925; Park & Burgess, 1925; Shaw &

McKay, 1942; Wirth, 1938). In their study of Chicago neighborhoods in the early

1900s, guided by their social disorganization perspective, they in particular found

high rates of delinquency in socially disorganized areas often characterized by pop-

ulation heterogeneity, poverty, and high population turnover.

A review of current research on ecological research on crime in general and on

homicide in particular shows that the vast majority of today’s research is carried out

in the United States.1A vast body of homicide research exists in the United States

that covers a variety of topics, time periods, and levels of analysis. Because of ready

availability of macro-level social indicators, aggregate-level studies of homicide

characterize much of this literature–typically examining homicide rates across

states, metropolitan areas, and/or cities within the United States. These extant homi-

cide analyses have done particularly well to identify classic covariates of homicide

at these various levels of geographic location in the United States.

In this article we build on the theoretical and empirical work in the United States

but aim to bring the discussion back to its theoretical and empirical origin, that is,

the European continent. Following the lead of most criminologists who have studied

U.S. homicide rates, we also use a structural approach in our research and argue that

the same theories used to study city-level homicide rates in the United States also are

applicable to studies of cross-national city-level European homicide rates. Yet this

remains an empirical question. Therefore, we address the question: Do those struc-

tural forces identified by criminologists as important in explaining homicide rates in

the United States also apply in understanding the extent to which social and eco-

nomic factors affect homicide rates in European cities and countries? We address

this question by analyzing data on homicide rates and structural covariates from

more than 100 cities from 16 European countries.

By analyzing city-level homicide rates from several European countries, our

study is the first that studies aggregate homicide rates in a large sample of European

subnational (city) units cross-nationally. There have been a limited number of

European studies on homicide that examine the extent to which social and economic

factors explain the variation in homicide rates across European cities; however, these

typically analyze cities within a single European country (Gartner & McCarthy,

1991; Kim & Pridemore, 2005; Liu, 2005; Pridemore, 2005; Villarreal, 2004).

Furthermore, there have also been studies analyzing European homicide cross-

nationally. However, almost without exception, the cross-national homicide studies

have been conducted using national-level data. Limitations of data availability and

comparability across countries, to a large extent, have restricted cross-national homi-

cide research to the national level.2Not only have varying definitions of homicide