This study will examine the relationship
between city-level structural factors and interracial homicide rates.
Specifically, we will use data from the Supplemental Homicide Reports, the U.S.
Census, and the Uniform Crime Reports to test the assumption that racial
antagonism and conflict mediate the relationship between racial inequality and
interracial homicide, and to assess an alternative hypothesis that draws on
macro-structural opportunity theory and the routine activities perspective to
understand the racial inequality-interracial homicide relationship. First, we
will focus on the prevalence, the nature, and the motivation of different
subtypes of interracial killing, and examine the structural predictors of those
subtypes. We suggest that the motivation of the homicide, largely ignored in
previous research, is the best indicator of the mechanism by which structural
characteristics can influence interracial killing (homicides with expressive
motives are more likely to be manifestations of antagonism and conflict than
those with instrumental motives). We will then determine whether robbery rates
mediate the relationship between racial inequality (and other structural
characteristics) and interracial homicide. If this is the case, we argue that it
is premature to suggest that racial antagonism and conflict mediate the racial
inequality-interracial homicide relationship, and instead propose that the
relationship may be the result of a combination of racial interaction and
patterns of economic crime.
Interracial homicide is one of the most
discussed, yet least studied, types of violence in America. While theories
abound, there has been little empirical consensus concerning an explanation for
variation in rates of interracial killing. One argument is that competition for
jobs and political power fuels racial antagonisms which, in turn, engender
interracial violence. Studies that adopt this perspective predict that cities
with greater levels of racial inequality and/or racial competition will have
greater rates of interracial killings, controlling for other factors. While
some empirical support for this argument has been documented (Balkwell 1990;
Jacobs and Wood 1999; Parker and McCall 1999) and these studies are
provocative, they do not provide a decisive test because they fail to directly
measure the social psychological process hypothesized to mediate the racial
inequality-interracial homicide relationship. That is, the hypothesized
relationship looks like this: Racial Inequality/Competition à Racial Hostility à Interracial Homicide. Yet, the intervening mechanism—racial
hostility—has gone largely unmeasured. This is because all of these studies
have used aggregate-level data, which precludes the measurement of racial
hostility, generally considered to be an individual-level measure. While it may
be possible to identify aggregate measures of racial hostility independent of
interracial violence, this has not been demonstrated in the literature. This
being the case, there is a risk of incorrectly identifying the causal processes
driving variation in interracial homicide.
The objective of the study is to empirically
evaluate the causal mechanisms that have been proposed in previous research and
to test an alternative hypothesis that incorporates the homicide motive. We
will do this first by focusing on the prevalence, the nature, and the
motivation of different subtypes of interracial killing, and by
examining the structural predictors of the different subtypes. Despite a
rigorous search, we were unable to locate any study that examines interracial
homicide rates disaggregated by motive. We will then determine whether the
relationship between macro-structural characteristics and interracial homicide
can be explained, in part, by high rates of other types of crime, such as
robbery. We believe this dual-pronged approach can significantly advance our
understanding of the causes of interracial homicide. The goal of this study is
to answer a set of questions we believe will shed light on the processes by
which macro-structural factors can influence rates of interracial homicide:
1)
What
structural characteristics explain variation in rates of interracial homicide
across U.S. central cities? For example, how are racial inequality and
opportunities for interracial contact (both measured in various ways), among
other factors, related to interracial killings?
2)
Does
the influence of these structural characteristics on interracial killing vary
depending on the motive of the homicide? That is, is the effect of racial
inequality (and other factors) larger or smaller if the interracial homicide
occurs during an economically-motivated crime (versus during an argument or
fight)?
3)
How
do rates of other types of criminal behavior (e.g., robbery) influence rates of
interracial homicide? Can variation in rates of non-lethal
economically-motivated criminal behavior explain, in part, the relationship
between macro-structural characteristics and interracial homicide?
We elaborate the logic of these questions and the
theoretical basis for our predictions below.
THEORETICAL BACKGROUND
Explanations for interracial
homicide have emerged from two different perspectives. One approach, informed
by Blau’s (1977) macrostructural opportunity theory and elements of the general
criminal opportunity perspective, focuses on opportunities for
intergroup contact as a central factor influencing interracial violence. This
perspective argues that interracial homicide, like interracial marriage, is a
byproduct of intergroup contact. The likelihood of killing someone, or marrying
someone, of another race is significantly influenced by the amount of
interracial contact in a given area. According to Blau, opportunities for such
contact are shaped, primarily, by the fundamental properties of urban social
structure. For example, one such property is racial inequality. Inequality
refers to the differences among persons with respect to status characteristics
such as education, income, wealth, and prestige. Inequality has a distinctive
effect on associations when it is correlated with group membership. For
example, when race is strongly correlated with socioeconomic status,
opportunities for contact between members of different groups are likely to be
infrequent. Macrostructural opportunity theory therefore proposes that racial
inequality will reduce opportunities for contacts between members of different
races, which, in turn, reduces the probability of interracial associations. As
a result, racial inequality will be negatively related to interracial
homicide. Likewise, other factors that directly affect opportunities for
interracial contact—such as racial residential segregation and racial
heterogeneity—are also hypothesized to influence interracial homicide.
Recent advances extend
Blau’s theoretical framework by incorporating insights from the criminal
opportunity perspective (Messner and South 1992). This approach derives largely
from the routine activities perspective (Cohen and Felson 1979), which argues
that for a successful criminal victimization to occur, a motivated offender
must come into contact with a suitable target in the absence of guardians who
could intervene and prevent victimization. The criminal opportunity perspective
suggests that routine activities (e.g., employment status or living in a
racially heterogeneous area, etc.) that encourage members of different races to
interact more frequently are likely to influence levels of interracial
killings.
Another explanation for
interracial homicide focuses less on opportunities for intergroup contact and
more on the motivation for intergroup homicide. Of importance here is
the extent to which structural conditions operating in urban areas contribute
to competition and conflict between racial groups. Social psychological
processes such as frustration-aggression, racial antagonism, black rage, and
stress-anger-displacement are used to explain the observed relationships
between macro-structural characteristics such as racial inequality and
interracial homicide (Balkwell 1990; Jacobs and Wood 1999; Olzak et al. 1996).
For example, Blau and Blau (1982) posit that racial inequality leads to strong
pressures to commit violence and that this process derives from the inherent
contradiction between ascriptive inequality and democratic values. Key to this argument
is the fundamental principle in democratic societies that rewards should be
distributed in accordance with merit and effort. Thus, persons who receive
meager rewards because of their race are likely to develop feelings of
resentment, frustration, and hostility, which are likely to be expressed
through aggression. This perspective predicts that racial inequality will be positively
related to interracial homicide.
Empirical research that has
examined the relationship between racial inequality and interracial homicide
has been inconsistent. Messner and South (1992) find reasonably strong support
for macrostructural opportunity theory; cities with extreme socioeconomic
differentials between blacks and nonblacks have relatively low levels of
interracial homicide, controlling for other factors. Messner and South argue
that this is so presumably because the consolidation of race with socioeconomic
status inhibits the formation of interracial associations. On the other hand,
Parker and McCall (1999) and Jacobs and Wood (1999) find a positive
relationship between racial inequality (measured in various ways) and
interracial homicide, and conclude that economic inequalities rooted in
ascribed positions likely create alienation, despair, and conflict—all of which
are associated with higher levels of intergroup violence, and most
consistently, with black interracial homicide.
One major concern with the
empirical literature is that the relationship between racial inequality and
rates of homicide in conflict-based explanations often suggests specific social
psychological processes, most of which are not tested in structural analyses of
interracial killings. Given the difficulty with measuring such processes,
particularly in national studies, this is understandable. Still, the implied
micro-macro linkage has not been adequately analyzed which raises questions
about its validity. While the proposed study does not measure the social
psychological assumptions that underpin conflict-based explanations of
violence, it contributes to the debate in two ways. First, we examine a thus
far unexamined aspect of the racial inequality-interracial homicide
relationship: motive of the homicide. Second, we determine whether rates of
robbery mediate the influence of racial inequality on interracial homicide.
Concerning point one,
findings from a number of studies underscore the importance of disaggregating
by motive when determining the structural characteristics of homicide (Kubrin
forthcoming, Kubrin and Wadsworth forthcoming; Parker and McCall 1999; Peterson
and Krivo 1993; Williams and Flewelling 1988). The motive of the homicide is
especially critical for understanding interracial killings because it is our
best indicator of the underlying social psychological processes that fuel lethal
violence. We argue that altercation-based or other expressive interracial
homicides, more so than robbery or other instrumental homicides, better reflect
the causal mechanisms (e.g., racial hostility) that have been proposed in
previous research. If these causal processes are accurate, the relationship
between racial inequality and interracial homicide will be stronger for
expressive altercation-based killings than for instrumental homicides motivated
by economic gain.
Point two assesses another
possible explanation for the variation in rates of interracial homicide. This
approach may be included under the umbrella of macrostructural opportunity
theory, and draws from ideas utilized in the routine activities perspective. We
posit that high rates of interracial killing may be the result of an interaction
between high rates of interracial contact and high rates of interpersonal crime
(e.g., robbery), which intentionally or unintentionally, can lead to homicide.
Put simply, some cities may have more interracial killings because they have
higher rates of robbery and because blacks and whites interact more
frequently in public spaces. Robbery creates a context in which an interaction
between two people (independent of race) is likely to turn violent. Thus the
combination of high rates of interracial interaction with high rates of robbery
creates an increased likelihood, independent of racial conflict or antagonism,
of interracial homicide.
If it is the case that
robbery rates are strong predictors of interracial homicide, and may in fact
mediate the influence of macro-economic factors on interracial killing, we
suggest that attributing high interracial homicide rates to social
psychological processes stemming from antagonistic race relations is premature.
While this explanation cannot be ruled out (or confirmed) without empirically
measuring the processes, a more suitable explanation may be related to the
calculation used to choose a suitable target for a predetermined criminal
offense. In this study, we propose two empirical models (described below) that
will address these issues.
Consistent with past
research, our unit of analysis is the central city. Parker (1989) suggests that
the central city is the most appropriate level of aggregation for two reasons.
First, given the geopolitical process by which SMSAs (a larger level of
aggregation) are designated, these units can be very heterogeneous. Such
heterogeneity can mask meaningful relationships between structural
characteristics and patterns of behavior. Central cities tend to be more
homogenous in patterns of land use and the distribution of human activity.
Second, homicide is, for the most part, a central city phenomenon. Parker’s
(1989:994-95) analysis shows that roughly 80 percent of the homicides took
place within the boundaries of the central city.
In
creating rates of interracial homicide, we limit our analysis to murders and
non-negligent homicides with a single victim and single offender. This approach
is consistent with previous research (Kovandzic et al. 1998; Messner and South
1992; Miethe and Drass 1999; Parker and McCall 1999). SHR data only have
information on the first victim and offender in cases where there are multiple
victims and/or offenders, making it impossible to determine the characteristics
(by which to classify the homicide) for the second and subsequent victims
and/or offenders.
Given
the relative rarity of interracial killings disaggregated by motive, this
research users data from 1985-1995 in order to increase the number of
interracial killings with different motives, a common practice in the
literature (Jacobs and Wood 1999; Kubrin forthcoming; Kubrin and Wadsworth
forthcoming; Morenoff and Sampson 1997; Parker and Pruitt 2000). This is
necessary, as creating rates from fewer years would result in many cities
having few or no recorded interracial homicides, thus skewing the distributions
toward zero.[1]
We
will use the race-adjusted SHR data to compute thirteen city-level outcome
measures representing various rates of inter- and intraracial
homicide. These include five non-motivation-specific measures including the
total rate of interracial homicide (based on blacks killing whites [B-W] and
whites killing blacks [W-B]), the rates of B-W, B-B, W-B, and W-W homicide. The
remaining eight motivation-specific outcome variables include the rates of B-W,
B-B, W-B, and W-W financially-motivated and non-financially-motivated
homicides.
Census Data: We regress the overall and
subtype-specific homicide rates on the structural characteristics and crime
rates of U.S. central cities. Information on the structural characteristics
comes from the 1990 U.S. Census. Measures representing region, overall income
inequality, racial inequality, interracial contact, as well as overall and
racially disaggregated measures of population size, poverty, per capita income,
unemployment levels, and female-headed households will be included in the
analyses. Region will be measured as a series of dummy variables. As in
previous research, overall income inequality is measured using the Gini Index
of Family Income Concentration (Kovandzic et al. 1998, Parker and McCall 1999).
There will be three measures of racial inequality that reflect racial
differences in education, income, and employment opportunities (Messner and
Golden 1992). Racial income inequality is operationalized as the ratio of white
to black median income. The ratio of white to black median years of schooling
for the respective populations aged 25 and over serves as an indicator of
educational inequality. Lastly, the ratio of white to black unemployment rates
is used as a measure of inequality in employment opportunity. Depending on the
correlations between these measures and results from collinearity diagnostics,
it is likely that the three indicators of inequality will be combined using
factor analysis. To measure levels of interracial interaction, we use
indicators of racial heterogeneity (Lieberson 1969:851), residential
segregation (index of dissimilarity), and exposure and isolation indices (p*)
(Massey and Denton 1993). Total and racially disaggregated measures of the
other variables will also be computed from the census data.
The
two sets of models to be tested are both aggregate level models that use the
central city as the unit of analysis. In the first set of models, we
disaggregate homicide by both the race of the offender and victim as well as
the circumstances under which the homicide took place. The circumstance of the
homicide (e.g., robbery, argument) is a good indicator of the motive behind the
killing.[2]
This disaggregation process will be used to create dependent variables
representing the rates of a variety of different types of interracial and
intraracial homicide (e.g., B-W financially motivated, B-W non-financially
motivated, etc.).
If interracial homicide
stems from a process in which macro-structural forces encourage social
psychological processes such as frustration-aggression, black rage, or racial
antagonism, we would expect the factors driving these processes to have a
stronger influence on expressive or non-financially motivated killings than
instrumental or financially-motivated killings. We would also expect these
factors to influence interracial killing more strongly than intraracial
killing. To test whether the structural characteristics discussed above are
more strongly related to expressive than to instrumental killings, and to
interracial vs. intraracial homicide, we will run a series of regressions with
various race and motive specific homicide rates as the dependent variables and
employ a standard test for coefficient differences across equations (Kubrin
forthcoming; McNulty 2001; Paternoster et al. 1998). Figure 1 offers a
graphical representation of some equations that will be tested.
Our second analytic approach
treats the total rate of interracial homicide, the rate of B-W homicide, and
the rate of W-B homicide as the dependent variables. These rates will be
regressed on a number of structural characteristics identified in previous
research (Jacobs and Wood 1999; Messner and Golden 1992; Parker and McCall
1999). After attempting to replicate the analyses of previous studies, we will
add variables measuring central-city level rates of robbery. The degree to
which the effects of macro-structural factors representing racial inequality
are attenuated when robbery rates are included in the models offers insight
into the process by which structural forces influence interracial homicide.
Significant attenuation would suggest that it is the frequency of economically
motivated criminal behavior, driven by structural characteristics, that
influences interracial homicide. Support for such a finding casts doubt upon,
or at least calls into question, explanations that focus on social
psychological processes stemming from racial antagonism. Figure 2 offers a
graphical representation of this model.
The project will span 1 year
from May 1, 2003 - April 30, 2004. The tasks and timeline are as follows: 1st
Phase: Data cleaning and the construction of variables. City-level
structural characteristics will be created from 1990 Census data and racially
and motivationally disaggregated homicide variables will be constructed from
the race-adjusted SHR data. We will also perform a thorough review of the
literature on interracial homicide. These activities are to be carried out in
May and June of 2003. 2nd Phase: Data analysis will be
conducted. We will run descriptive analyses first to determine the distribution
of rates of disaggregated interracial homicide, and determine their correlates.
This disaggregation will be based on victim and offender race and homicide
motive. We will then run series of regression analyses that correspond to the
conceptual models presented earlier. These activities are to be carried out
from July-November of 2003. 3rd Phase: Writing the
final report. We will write a final report detailing the study and the major
findings. We will also submit our data and supporting documentation. These
activities are to be carried out in December 2002 and January 2003. 4th
Phase: We will write two to four papers, depending upon the results,
which will be submitted to scholarly journals such as Criminology, Journal
of Research on Crime and Delinquency, and Homicide Studies for
publication. We also intend to submit an abstract to present the major findings
at the 2005 American Society of Criminology Conference (abstracts are due in
March 2004). Finally, we will present our findings and meet with BJS staff to
discuss follow up research. These activities are to be carried out from
February-April 2004.
FIGURES
REFERENCES
Balkwell,
James W. 1990. “Ethnic Inequality and the Rate of Homicide.” Social Forces
69:53-
70.
Political and Economic Rivalries Explain White
Killings of Blacks or Black Killings of Whites?” American Journal of
Sociology 105:157-190.
Kovandzic,
Tomislav, Lynne M. Vieratis, and Mark R. Yeisley. 2001. “The Structural
Covariates of Urban Homicide: Reassessing the Impact
of Income Inequality and Poverty in the Post-Reagan Era.” Criminology
36:569-597.
for Racial Differences in Process.” American
Sociological Review 65:547-559.
Kubrin,
Charis E. Forthcoming. “Structural Covariates of Homicide Rates: Does Type of
Homicide Matter?” Journal of Research in Crime
and Delinquency.
Kubrin,
Charis E. and Tim Wadsworth. Forthcoming. “Identifying the Structural
Correlates of
African-American Killings: What Can We Learn from
Data Disaggregation?” Homicide Studies.
LaFree,
G., Drass, K.A., & O’Day, P. (1992). Race and Crime in Postwar America:
Determinants of African-American and White Rates,
1957-1988. Criminology 30:157-188.
Macmillan, Ross and
Rosemary Gartner. 1999. “She Brings Home the Bacon: Labor-Force
Participation
and the Risk of Spousal Violence Against Women.” Journal of Marriage and the
Family 61:947-958.
Massey, Douglas S.
and Nancy A. Denton. 1993. American Apartheid: Segregation and the
Making
of the Underclass.
Cambridge: Harvard University Press.
McNulty,
Thomas. 2001. “Assessing the Race-Violence Relationship at the Macro Level: The
Assumption
of Racial Invariance and the Problem of Restricted Distributions.” Criminology
39:467-487.
Messner,
Steven F. and Reid M. Golden. 1992. “Racial Inequality and Racially
Disaggregated
Homicide Rates: An Assessment of Alternative
Theoretical Explanations.” Criminology 30:421-447.
Messner,
Steven F. and Scott J. South. 1992. “Interracial Homicide: A Macrostructural-
Opportunity Perspective.” Sociological Forum
7:517-536.
Expressive Homicides: An Application of Qualitative
Comparative Analysis.” Journal of Quantitative Criminology 15:1-21.
Miles-Doan,
Rebecca. 1998. “Violence Between Spouses and Intimates: Does Neighborhood
Context
Matter?” Social Forces 77:623-645.
Morenoff,
Jeffrey D. and Robert J. Sampson. 1997. “Violent Crime and the Spatial Dynamics
of
Neighborhood Transition: Chicago, 1970-1990.” Social
Forces 76:31-64.
O’Brien,
Robert M. 1987. “The Interracial Nature of Violent Crimes: A Reexamination.”
American Journal of Sociology 4:817-835.
Olzak,
Susan, Suzanne Shanahan, and Elizabeth H. McEneaney. 1996. “Poverty,
Segregation,
and the Race Riots: 1960 to 1993.” American
Sociological Review 61:590-613.
Ousey,
Graham C. 1999. “Homicide, Structural Factors, and the Racial Invariance
Assumption.”
Criminology 37:405-425.
Parker,
Robert Nash. 1989. “Poverty, Subculture of Violence, and Type of Homicide.” Social
Forces 67:983-1007.
Parker,
Karen F. and Patricia L. McCall. 1999. “Structural Conditions and Racial
Homicide
Patterns: A Look at the Multiple Disadvantages in
Urban Areas.” Criminology 37:447-473.
Parker,
Karen F. and Matthew V. Pruitt. 2000. “Poverty, Poverty Concentration, and
Homicide.”
Social Science Quarterly 81:555-570.
Peterson,
Ruth D. and Lauren J. Krivo. 1993. “Racial Segregation and Black Urban
Homicide.”
Social Forces 71:1001-1026.
Shihadeh,
Edward S. and Darrell J. Steffensmeier. 1994. “Economic Inequality, Family
Disruption, and Urban Black Violence: Cities as
Units of Stratification and Social Control.” Social Forces 73:729-751.
Williams, Kirk R.
and Robert L. Flewelling. 1988. “The Social Production of Criminal
Homicide:
A Comparative Study of Disaggregated Rates in American Cities.” American Sociological Review 54:421-431.
APPENDIX A
(more)
APPENDIX B
The race adjustments take
into account two otherwise problematic sources of missing data. First, some
cities fail to submit homicide reports each year, and while counts are usually
available through the UCR, the race and sex of the offender and victim, as well
as other characteristics of the homicide, are unknown. The adjustment procedure
estimates the characteristics of the offenders, the victims, and the crimes
based on all of the known characteristics of the reported events.
This
project will increase our understanding of interracial homicide, and the
structural correlates that contribute to this form of violence. We will build
on recent work by challenging the assumption that racial inequality and other
structural forces are linked to higher rates of interracial killing through
racial hostility and antagonism, and by offering an alternative hypothesis that
incorporates the motive of the homicide—an important yet thus far unexamined
aspect of the racial inequality-interracial homicide relationship. This study
will also determine how rates of other types of criminal behavior such as
robbery influence rates of interracial homicide. In addition to theoretical
development, we believe that our findings can potentially make a substantial
contribution to policy aimed at reducing both inter- and intra-racial violence.
Previous results from
research conducted by the P.I.s have been published in various refereed
academic journals including Journal of Research in Crime and Delinquency,
Social Forces, and Homicide Studies. Results have
also been presented at numerous academic conferences including the American
Society of Criminology and Academy of Criminal Justice Sciences conferences. We
expect that this publication and presentation record will be sustained with
this project.
The research team includes
two P.I.s who bring different areas of expertise and research skills to the
project. Charis E. Kubrin (Assistant Professor of Sociology and Senior Fellow
at the Institute on Crime, Justice and Corrections at George Washington
University) has expertise in aggregate-level analyses of homicide rates,
statistical methods, and has published a number of studies on race and violent
crime. She has received funding from the National Science Foundation, the
National Consortium on Violence Research, and the SOROS Foundation to conduct a
variety of studies all of which deal with race and crime. Her two most recent
publications examine the structural correlates of homicide rates, and an
earlier publication examines how changes in racial composition of neighborhoods
have influenced changes in violent crime rates in the last decade. Tim
Wadsworth (Assistant Professor of Sociology at University of New Mexico) has
expertise in individual and aggregate-level analyses of labor market
participation, industrial composition, and criminal behavior. He has received
funding from the National Institute of Justice to conduct studies focusing on
individual employment, city and county labor market characteristics, racial
segregation, and violent criminal behavior. His most recent publications
examine the influence of race and poverty on violent crime.
APPENDIX E
CURRICULUM
VITAE
[1] One potential problem with
using homicide data that span a 10-year period is that the results of the
analyses could be affected if one or more of the motive subtypes became more or
less prevalent over time. Drastic overtime changes in the proportion of
homicides with different motives can lead to misleading results for the
structural covariates. We examined the subtypes for each year between 1985 and
1995 and found that the percentage of interracial homicides that were
financially motivated ranged from 16.8% to 20.4%. This provides strong evidence
that the mixture of interracial homicide types remained relatively invariant
over the 10-year period.
[2] As motivation stems from the mental or emotional process of the offender (be it conscious or unconscious), researchers are rarely able to confidently identify the precise motive of any crime. However, the circumstances surrounding the crime provide enough information to categorize the offense as financially motivated or not. Thus for our purposes, circumstance acts as a very good, yet imperfect, indicator of motive.