Parsing out the “Hispanic Effect” in Disaggregated Homicide Trends at the Intersection of Race,Ethnicity,and Gender from 1990 to 2016

Abstract

This study contributes to homicide research by parsing out the Hispanic Effect and applying an intersectional approach to examining U.S. homicide victimization trends by race, ethnicity, and gender, jointly. Drawing on mortality data, we document and describe total, firearm, and non-firearm homicide victimization rates from 1990 to 2016 for six subgroups: Black women, Black men, Hispanic women, Hispanic men, White women, and White men. The analysis of within- and between-group homicide trends reveals important subgroup-specific patterns that prior studies using aggregate or confounded data have masked. The findings have important research, theory, and policy implications and advocate for an intersectional approach to studying homicide.

Keywords

homicide intersectionality gender and crime race and crime firearms crime trends violence

Introduction

Homicide, often considered the most severe and reliably measured crime, represents a serious concern for public health, safety, and policy. The United States’ high rate of homicide is one of the key characteristics setting it apart from other industrialized nations (Messner & Rosenfeld, 2013; Zimring & Hawkins, 1997). Homicide is the leading cause of death for some segments of the population (Heron, 2017, p. 32), having devastating effects on families, communities, and the larger society (DeLisi et al., 2010; Mastrocinque et al., 2015; Redelings et al., 2010). For these reasons, social scientists have devoted extensive empirical attention to homicide, assessing longitudinal homicide trends (Blumstein et al., 2000; Blumstein & Rosenfeld, 1998; Blumstein & Wallman, 2005; Parker et al., 2016; Rosenfeld & Oliver, 2008), analyzing micro- and macro-level correlates of homicide (Land et al., 1990; Messner & Golden, 1992; Ousey, 1999; Ousey & Lee, 2004; Parker, 2001, 2008; Parker & Hefner, 2015; Peterson & Krivo, 1993), and underscoring the unequal distribution of homicide across the population (Krivo & Peterson, 2000; Parker, 2008; Sampson & Wilson, 1995). The literature indicates that men and persons of color, for example, have substantially greater homicide involvement—as both victims and perpetrators—than their counterparts (Fox & Fridel, 2017; Fox & Zawitz, 2004; Parker, 2008). While these gender and racial/ethnic differences in homicide risks are well-documented, scholars have paid less attention to the intersectionality of these characteristics and how they coalesce to affect homicide victimization risks over time.

Intersectional research acknowledges the fact that individuals occupy multiple, rather than single, social identities. Intersecting social characteristics—such as race, ethnicity, gender, age, sexuality, and class—simultaneously create different structural arrangements and shape exposure to risks (Burgess-Proctor, 2006; Parker & Hefner, 2015; Potter, 2006, 2015). For example, a number of homicide studies show that homicide involvement is higher for persons occupying multiple marginalized social locations, such as for those who are young, Black, and men (Blumstein, 1995), reside in predominately Black, economically distressed neighborhoods (Sampson & Wilson, 1995), and are Black and women (Haynie & Armstrong, 2006; Parker, 2008; Parker & Hefner, 2015). Moreover, the literature on homicide trends is dominated by analyses of single social characteristics, such as by race/ethnicity or gender. Such studies may mask important patterns within and between subpopulations over time, creating a key knowledge gap that has important implications for research, theory, and policy. Policy prescriptions and prominent explanations of homicide changes might hold in the aggregate or by a single demographic characteristic but not for subgroups according to their intersecting identities. These limitations are likely exaggerated when considering the “Hispanic Effect”—a longstanding methodological limitation that results in over/underestimating race-specific crime and victimization rates due to the disregard of Hispanic origin in race categories (Steffensmeier et al., 2011).

In the current study, we aim to redress this knowledge gap by parsing out the Hispanic Effect and examining the intersection of race, ethnicity, and gender in U.S. homicide trends between 1990 and 2016. Specifically, we document and describe homicide victimization rates for six race/ethnic-gender-specific subgroups: Black women, Black men, Hispanic women, Hispanic men, White women, and White men. Moreover, given the prominence of firearms in homicide, we document and describe firearm and non-firearm homicide trends for the six subgroups. We draw on national mortality data from the Centers for Disease Control and Prevention’s National Vital Statistics System to examine four overarching research foci. First, we examine and describe each subgroup’s homicide trend relative to the aggregate homicide trend throughout the 27-year period. Second, we examine the role of firearms by analyzing trends in firearm versus non-firearm homicides for the subgroups. Third, we focus on between-subgroup homicide trends by comparing homicide disparities between the six subgroups. Lastly, we estimate and illustrate the impact of the Hispanic Effect on homicide trends for Black women, Black men, White women, and White men. We conclude by discussing our study’s implications for homicide research and policy.

Background

Intersectionality Perspective

The concept “intersectionality” was introduced by Kimberlé Crenshaw in her seminal 1989 paper titled “Demarginalizing the Intersection of Race and Sex: A Black Feminist Critique of Antidiscrimination Doctrine, Feminist Theory and Antiracist Politics.” In this article Crenshaw (1989) provided examples of court cases that denied Black Women protection from discrimination, suggesting that 1) Black women were not protected as a class and could be discriminated only as Black or as women, separately and 2) Black women could not represent Black people or women at large. The examples illustrated the ways that oppression can be distinct for Black women sitting at the intersection of racism and sexism. Intersectionality specifically acknowledges that the combination of multiple social characteristics allows for distinct forms of oppression. The combination, or intersection, of social characteristics influences individuals’ interactions with others and institutions. As criminologist Hillary Potter contended, “identities and power are relevant throughout all social aspects of human life, so they must also be considered within the contexts of criminality, victimization, and informal and formal responses to crime” (Potter, 2013, p. 305).

Importantly, a study is not intersectional for disaggregating by multiple social characteristics. Intersectionality suggests that there is meaning in the social spaces created by interdependent and mutually constructive intersecting characteristics (Bowleg, 2008; Ransford, 1980). This understanding rejects, as Crenshaw (1989) did, the additive approach which suggests that the adversity faced by people with a single marginalized identity (i.e., Black or woman but not both) can be added to represent the adversity faced by those at the intersections. Rather, those at the intersections experience something reminiscent of and distinct from its parts. Thus, when Bowleg (2008) discussed the obstacles in intersectional research, she specifically addressed the inherent additive approach of extant analytical techniques. She argued that intersectional researchers have a responsibility to provide interpretations that incorporate “the sociohistorical realities of historically oppressed groups” (p. 318). That is to say that intersectional research requires an analysis of the legacy of oppression and power and its present impact on the groups and phenomena in question. As such, the current study is framed through an intersectional lens suggesting that the multiplicity of systems of power related to race, ethnicity, and gender are related to homicide victimization and lays the framework for future studies to continue from an intersectional lens in building the homicide literature.

The following section overviews the homicide literature that has considered the importance of race, ethnicity, and gender as separate identities as well as the smaller number of studies that have examined multiple identities intersectionally.

Race, Ethnicity, and Homicide

Research on race and homicide overwhelmingly focuses on two groups: Black and White people. For instance, it is widely accepted that Black Americans are disproportionately involved in homicide offending and victimization, a pattern that has persisted for decades (Fox & Zawitz, 2004; LaFree, 1995; LaFree et al., 2010; Parker, 2008; Velez et al., 2003). The Black-White difference in homicide involvement is generally attributable to each group’s differential exposure to criminogenic risk factors including: structural disadvantage, racial segregation, family disruption, unemployment, neighborhood concentrated disadvantage, and racial discrimination (Kaufman, 2005; Krivo & Peterson, 2000; McNulty & Bellair, 2003; Sampson et al., 2005; Sampson & Wilson, 1995; Unnever & Gabbidon, 2011).

Far less attention has been paid to other racial/ethnic groups, such as Hispanic Americans whom are the largest group of color and fastest growing group in the U.S. This uneven focus is largely due to longstanding limitations in crime data. For example, the FBI (e.g., UCR, SHR)—the leading source of crime data in the U.S.—does not consistently categorize Hispanic Americans separately from Black and White populations. The ethnicity data in the SHR are sparsely collected, and extant datasets often code Hispanic Americans as White. Known as the “Hispanic Effect,” this measurement limitation imposes the consequences of inflating non-Hispanic White homicide rates, underestimating disparities between non-Hispanic Whites and other racial/ethnic groups, and omitting Hispanic-specific analyses on crime and criminal justice (Steffensmeier et al., 2011).

The limited empirical knowledge on Hispanic Americans’ homicide involvement largely comes from cross-sectional studies using data from smaller aggregation units, such as neighborhoods and cities (Chauhan et al., 2011; Kaufman, 2005; Light & Ulmer, 2016; Martínez, 2014; Phillips, 2002). These studies collectively show that Hispanic Americans’ homicide involvement is intermediate to that of Black and White Americans and is generally explained by structural disadvantage. In a national-level study, Langley and Sugarmann (2018) used a combination homicide data from the CDC and FBI to describe Hispanic American’s homicide victimization. In 2015, Hispanic Americans were victimized by homicide at a rate that doubled the White homicide rate but was only one-fourth the Black American rate (Langley & Sugarmann, 2018, p. 4). This report also noted that homicide was the leading cause of death for Black youth aged 15 to 24 years old and the second leading cause of death for Hispanic youth (Langley & Sugarmann, 2018, p. 5). Of these homicides, 93% and 82%, respectively, involved a firearm (Langley & Sugarmann, 2018, p. 5).

The handful of longitudinal race/ethnic-specific homicide studies suggests that race and ethnic disparities in homicide have markedly declined in contemporary decades (Cook & Laub, 2002; Fox & Zawitz, 2004; LaFree, 1995; LaFree et al., 2010; LaFree et al., 2006; Parker, 2008; Rosenfeld & Fox, 2019; Tonry & Melewski, 2008). However, this literature is generally confounded by the Hispanic Effect. To address this limitation, Steffensmeier et al. (2011) estimated “clean” trends for Black and White homicide arrests using New York and California data that did not confound race and ethnicity. They found that interpretations of the racial gap were highly dependent upon the time period under study and that overall, there had been little change in disparities over time. Specifically, the Black-White gap decreased from the 1980s to 1990s but increased from the beginning of the 1990s to the end. However, this study did not directly measure Hispanic American’s homicide involvement due to data limitations associated with relying on official data from the FBI. In another study also accounting the Hispanic Effect but finding stronger evidence of convergence during a later time period, Light and Ulmer (2016) drew on data from the CDC’s Underlying Causes of Death for 131 metropolitan areas and examined race/ethnic-specific homicide trends from 1990 to 2010. In 1990, the Black-White homicide rate disparity ratio was nearly 30 but dropped to approximately 17 by 2010. The Black-Hispanic gap was also lower in 2010 than in 1990 (ratio of 13.9 vs. 21.5); however, they found evidence of a growing disparity from 2000 to 2010 (ratio of 11.5 vs. 13.9). The ratio gap between Hispanic Americans and White Americans decreased from 7.6 in 1990 to 3.5 in 2010.

Considering Hispanic origin is necessary for the advancement of intersectional theory and research related to violence. Steffensmeier et al., 2011 explained the “Latino Paradox.” Specifically, many Hispanic cultures experience structural and systemic disadvantages in the U.S. that are in some ways similar to those faced by Black Americans. Yet, Hispanic Americans have lower levels of homicide involvement than Black Americans. Rather than suggesting that this difference is inherently cultural, Steffensmeier et al. (2011) acknowledged important differences in these experiences as well, including the legacy of enslaving Africans and group differences in the degree of disadvantage (see also Light & Ulmer, 2016). In the same way that Bledsoe (2019) suggested that anti-Blackness is distinct from and surpasses class discrimination, the adversity that Black and Hispanic women and men face is different and appears to differentially influence their homicide risks. A prime example is racial/ethnic residential patterns for which the most extreme residential segregation, social isolation, and concentrated disadvantage—robust predictors of violence—are reserved for Black Americans, although this literature does not address racial/ethnic disadvantages at the intersection of gender (Charles, 2003; Iceland et al., 2002; Sampson & Wilson, 1995). In a U.S. Census Bureau report, Semega et al. (2019) noted that U.S. poverty rates have declined for non-Hispanic White, non-Hispanic Black, and Hispanic women, yet more women continue to live in poverty than men. The literature on socioeconomic conditions suggests that disadvantage in the U.S. is strongly influenced by racialized, nationalized, and gendered systems. The culmination of these systems could be influencing homicide risks (as suggested in Parker & Hefner, 2015; Reckdenwald & Parker, 2008).

The homicide literature on race and ethnicity is important yet limited by excluding a consideration of gendered systems of inequality and oppression, an ommission also found in the seminal studies by Steffensmeier et al. (2011) and Light and Ulmer (2016). Indeed, the literature is sparse in homicide studies that examine race and ethnicity and women’s experiences of violence. This is true for studies of violence more generally.

Gender and Homicide

In most homicide incidents, perpetrators and victims share similar sociodemographic characteristics, such as race, socioeconomic status, and age (Fox & Fridel, 2017). This pattern, however, does not always hold when gender is examined. Homicide is largely a male event wherein men commit most of the nation’s homicides, including those involving men and women victims (Fox & Fridel, 2017). To this end, homicide research tends to be male-centric given men’s higher and women’s lower homicide risks, leaving women-specific patterns lost in the aggregate. However, just as the literature on race/ethnicity and homicide lacks gender consideration, much of the gender and homicide research fails to account for race and ethnicity.

Theories related to macro-level women’s homicide victimization often suggest that women’s homicide is associated with their socioeconomic status, relative or absolute (Bailey, 1999; Martin et al., 2006; Peterson & Bailey, 1992; Titterington, 2006; Vieraitis et al., 2007; Whaley & Messner, 2002; Xi, et al., 2012). Whaley and Messner (2002) reviewed the state of literature regarding these relationships and found that the direction of the hypotheses varies across studies. Some hypotheses suggest that increases in women’s absolute status or women’s status relative to men is associated with lower levels of victimization while others suggest the opposite (Bailey, 1999; Martin et al., 2006; Peterson & Bailey, 1992; Titterington, 2006; Vieraitis et al., 2007; Whaley & Messner, 2002; Whaley et al., 2013; Xie et al., 2012). It appears that the general consensus presently is that the hypotheses may be corollary rather than contradictory. Nevertheless, the findings have been mixed and are generally considered inconclusive.

Within the literature on women’s macro-level violent victimization, the bulk of the studies investigates intimate partner homicide specifically or non-fatal forms of violence, most being cross-sectional designs (Bailey, 1999; Peterson & Bailey, 1992; Whaley & Messner, 2002; Whaley et al., 2013; Vieraitis et al., 2015; Xie et al., 2012). In addition, the literature rarely attends to the potential role of race or ethnicity. For instance, Whaley and Messner (2002) used FBI data (SHR) for 191 large cities to investigate the relationship between gender equality and male-on-female, female-on-male, male-on-male, and female-on-female homicide in a 1990 to 1994 panel study. They found that gender equality was associated with higher rates of male-on-female and male-on-male homicide in the south. Outside of the south, they found that gender equality was associated with lower rates of male-on-male violence. Whaley and Messner (2002) theorized that the functional form of the relationship might be related to race; however, they were unable to study this relationship directly.

These studies collectively highlight and attempt to reduce the lack of attention paid to women’s homicide involvement at the macro-level. However, these studies generally examine gender without considering race and ethnicity.

The Intersection of Race, Ethnicity, Gender, and Homicide

An abundance of research has examined the influence of race/ethnicity and gender on homicide separately, but few studies have applied an intersectional approach to jointly studying them. Exceptions include the emphasis often placed on young, Black men in homicide research, specifically (Blumstein, 1995; Fingerhut et al., 1992; Fingerhut & Kleinman, 1990), and in violence research, more generally (Lauritsen & Heimer, 2010). Empirical attention to young, Black men, whose homicide involvement is exorbitantly higher than other groups, has far outpaced that for other race/ethnic-gender subgroups (e.g., Black/Hispanic/White women, Hispanic/White men). Another exception was a 2012 report to the National Academy of Sciences by criminologist Janet Lauritsen. The report depicted trends of rape and sexual assault by race/ethnicity-gender using the National Crime Victimization Survey from 1993 to 2009. Lauritsen illustrated that for rape and sexual assault, Latina women experienced rates of violence much lower than those for non-Latina Black women while rivaling non-Latina White women rates, sometimes being higher and sometimes being lower (Lauritsen, 2012). This study challenged our understanding of the Latino paradox by illustrating the similarity between Latina women’s and White women’s levels of sexual victimization.

Prior studies that come closest to reflecting intersectional homicide research tend to examine race and gender jointly without considering ethnicity. In a Bureau of Justice Statistics report, Fox and Zawitz (2007) used FBI data (SHR) to depict homicide offending and victimization trends from 1975 to 2005 for Black women, White women, Black men, and White men across age groups. The report emphasized that homicide involvement was highest among young adult Black men and had declined among all subgroups since the 1990s. Using the same data source, Parker (2008) conducted a city-level analysis of the intersectionality of race and gender in homicide from 1980 to 2003 in her book, Unequal Crime Decline. Results from her intersectional analysis challenged the universality of the well-known great crime drop. For instance, Black men and White men homicide rates rose during the late 1980s—in varying magnitudes—while Black women and White women homicides were declining. Moreover, the 1990s homicide drop was greater among Black women than among the other three subgroups (Parker, 2008, p. 27). Haynie and Armstrong (2006) found some support for a relationship between gender equality and homicide offending disaggregated by race, gender, and victim-offender relationship. Based on city-level FBI (SHR) data, the results were consistent with those from a study by Vieraitis and Williams (2002), showing that gender equality was more strongly related to White women’s homicide involvement than Black women’s homicide involvement.

In these race-gender-specific homicide studies, however, the researchers relied on FBI data, which were confounded by the Hispanic Effect and precluded their ability to study Hispanic women and Hispanic men homicide. As such, these results are likely based on overestimated rates for White men and underestimated rates for Black men. The potential impact of the Hispanic Effect on women’s homicide trends is unclear. Moreover, due to the time periods under study, neither analysis was able to examine homicide after 2005, including the widely-speculated homicide rises in 2015 and 2016. Finally, these studies excluded a consideration of firearms in homicide at the intersection of race, ethnicity, and gender. Firearms have been shown to play an increasingly critical role in homicide trends, especially for men and people of color (Blumstein, 1995; Blumstein & Wallman, 2005; Cook, 2018; Fox & Fridel, 2017; Rosenfeld & Fox, 2019), leaving more to be learned about firearm violence among women of color. There is a possibility that Black women’s and Hispanic women’s firearm victimization risks are more elevated than currently known and should be centered in anti-violence public discourse, policy, and practice. The current study builds upon extant literature through a longitudinal investigation of race/ethnic-gender-specific homicide trends from an explicitly intersectional lens while considering the role of firearms.

Current Study

As reviewed above, virtually all prior homicide research is either cross-sectional, confounds race and ethnicity, or fails to apply an intersectional approach to assess the import of race/ethnicity-gender in homicide risk over time. Studies that only investigate race, ethnicity, gender, or neither may be masking distinctive risks and patterns by race/ethnicity-gender for which theories have not accounted or policy prescriptions might be futile. To this end, the current study contributes to the homicide literature by applying an intersectional lens to document and describe national homicide trends over a 27-year period from 1990 to 2016 for six racial/ethnic-gender subgroups: Black women, Black men, Hispanic women, Hispanic men, White women, and White men. We rely on data that disaggregate Hispanic origin from racial designations. In addition to total homicide, we consider the increasingly important role of firearms by examining firearm and non-firearm homicide for the subgroups (Blumstein et al., 2000; Kegler et al., 2018). We draw on national mortality data uniquely suited for the disaggregation required for our research foci and aim to advance the literature in several ways.

First, we move beyond race-specific, gender-specific, and race-gender-specific analyses by applying an intersectional approach that appreciates the interactive effect of race, ethnicity, and gender, jointly. Second, unlike prior race and homicide studies that almost exclusively study the Black-White racial gap, we also consider homicide trends for the Hispanic population, the largest group of color in the U.S. that has unfortunately received limited empirical attention. Third, our ability to disaggregate Hispanic origin from White and Black racial designations allows us to assess the consequence of the Hispanic Effect that has plagued most of the race and homicide literature. Fourth, we examine homicide trends over 27 years. This expansive focus allows us to account for several significant changes in homicide trends, including the recent homicide increases in 2015 and 2016 that have garnered widespread public attention (Gaston et al., 2019; Rosenfeld et al., 2017). Fifth, departing from most prior studies, we rely on an underused national homicide data source suited for intersectional research that also averts longstanding concerns about the reliability official crime data (e.g., FBI data), especially by race/ethnicity.

The previously-described homicide literature underscores at least four patterns, most of which have become widely-accepted conventional knowledge: 1) men have substantially greater homicide involvement—as both victims and perpetrators—than women, 2) persons of color are at greater risk of homicide than White Americans, 3) Hispanic Americans’ homicide involvement is intermediate to that of Black and White Americans, and 4) the Black-White homicide gap has considerably declined over time. Because intersectionality is seldom of account, these patterns might not hold by race, ethnicity, and gender, jointly. For example, while we expect Black men’s homicide to exceed those of other men and women of any race or ethnicity, it is unclear whether Hispanic and White men’s homicide risks, too, will exceed those of Black women, or whether gender trumps race/ethnicity or race/ethnicity trumps gender in this regard. Considering the Latino Paradox, we expect the homicide trend for Hispanic men to be lower than Black men’s and higher than White men’s. However, the literature provides little guidance about whether the Latino Paradox holds for women—that is, whether Hispanic women’s homicide risk is intermediate to that of Black and White women. Moreover, we hypothesize that parsing out the Hispanic Effect will reveal exaggerated homicide risks for White men and White women and slightly underestimated homicide risks for Black men and Black women. By accounting for the Hispanic Effect, we expect minimal homicide convergence between Black and White women and men. Finally, considering that Blumstein and colleagues (2000) found that the increases in aggregate homicide victimization in the 1980s and 1990s were reserved to firearm homicide, we expect greater variation in firearm homicide than non-firearm homicide over time.

Data

We investigate race/ethnic-gender-specific homicide using data from the National Vital Statistics System (NVSS), a monitoring effort by the National Center for Health Statistics (NCHS), a branch of the Centers for Disease Control and Prevention (CDC). Data on official events (e.g., births, deaths) are recorded by health departments located in the 50 states and District of Columbia and collated by the NCHS. These data are available through the Fatal Injury Reports housed in the Web-Based Injury Statistics Query and Reporting System (WISQARS). From these data, rates can be computed by victim’s race, gender, and mode of death, but Hispanic origin was not included until 1990.¹ For this reason, our analysis begins in 1990. We created total, firearm, and non-firearm homicide victimization trends for Black women, Black men, Hispanic women, Hispanic men, White women, and White men from 1990 to 2016. The Black and White trends exclude victims of Hispanic origin, overcoming a significant, longstanding data limitation, as described earlier.

These data are collected and reported by individual locales. Therefore, it is possible that classifications and data collection processes have varied over time. While the legal definition of homicide has not changed considerably since 1950, advances in technology may have made determining the cause of death more precise over time. Each of these caveats should be considered when interpreting results from the current study. Nevertheless, the practice of recording homicides has a long history and is generally viewed as the most consistently, reliably documented crime over time. In addition, relying on CDC homicide data rather than FBI SHR data likely yields more accurate results (Loftin et al., 2008).

Analysis

We examine aggregate and subgroup-specific homicide trends from 1990 to 2016, the period allowed by the data. First, we compute and analyze race/ethnic-gender-specific total, firearm, and non-firearm homicide rates and rate changes for each subgroup relative to those in the aggregate. This analysis is organized according to discrete subperiods defined in the extant literature on aggregate total homicide trends. Based on the homicide literature, the 1990 to 2016 period can be deconstructed and characterized into four subperiods: (1) homicide peaks from 1990 to 1993, (2) homicide declines from 1993 to 2000, (3) homicide lows and stabilization from 2000 to 2014, and (4) homicide growth from 2014 to2016. Hence, the analysis will determine the degree to which these aggregate patterns hold for each subgroup’s total, firearm, and non-firearm homicide victimization.

Second, we employ a between-group analysis by assessing race/ethnic-gender-specific disparities in homicide across subgroups. We compute disparity ratios that compare each subgroup’s total, firearm, and non-firearm homicide rates with those of White men.² Ratios are computed by dividing a subgroup’s homicide rate by the White men homicide rate. Then we examine and graph disparity levels and disparity changes over the 27-year period.

Third, we measure the degree to which the Hispanic Effect confounds race/ethnic-gender-specific homicide rates and trends during the study period. To do this, we compute and contrast “correct homicide rates” against “confounded homicide rates.” Correct homicide rates properly designate race and ethnicity and are calculated by excluding people of Hispanic origin from Black and White women and men homicide rates, also known as non-Hispanic Black and non-Hispanic White women and men rates. Confounded homicide rates refer to race- and gender-specific calculations that include people of Hispanic origin in Black and White women and men rates, therefore confounding race and ethnicity. Given that most prior race and homicide studies rely on the latter, this comparative analysis illuminates the impact of this longstanding measurement limitation on homicide research.

Results

Figure 1 depicts age-adjusted total homicide victimization rates while Figure 2 displays age-adjusted firearm and non-firearm homicide victimization rates. Both figures show graphed trends for the aggregate population and for Black women, Black men, Hispanic women, Hispanic men, White women, and White men between 1990 and 2016. The percent-change in age-adjusted homicide rates (i.e., total, firearm, and non-firearm) over the aforementioned four subperiods, in the aggregate and by race/ethnic-gender subgroupings, can be found in Appendix A. We now turn to describing subgroup trends relative to aggregate trends for each of the four subperiods.

Figure 1.

Age-adjusted total homicide vicitmization trends by race/ethnicity and gender subgroup.

Figure 2.

Age-adjusted firearm and non-firearm homicide victimization trends by race/ethnic-gender subgroup.

Aggregate and Subgroup Homicide Trends

Homicide Peaks, 1990 to 1993

The subperiod from 1990 to 1993 comes at the tail end of substantial homicide growth in the 1980s for the aggregate population. U.S. homicide rates peaked at 9.9 per 100,000 in 1991, the highest rate in the 27-year series and one that almost rivals the homicide peak of 10.3 in 1980. Homicide rates also peaked for each subgroup in 1991, with a rate of 66.6 for Black men, 27.3 for Hispanic men, 13.4 for Black Women, 5.9 for White men, 4.5 for Hispanic women, and 2.6 for White women, each per 100,000. While homicide rates were at high levels during this subperiod, the aggregate homicide rate increased by only 2% from 1990 to 1993. Diverging from aggregate trends, White men and Hispanic men experienced 2% and 8% decreases, respectively, while the Black men homicide rate increased by 6%. Homicide also increased among each of the women subgroups to similar degrees, as Black women, Hispanic women, and White women homicide rates rose by 6%, 7%, and 8%, respectively.

Aggregate and subgroup-specific homicide increases from 1990 to 1993 were largely driven by homicides involving firearms. Firearm homicide increased in the aggregate by 11% and among each subgroup to varying degrees. Black men and Black women experienced the largest firearm homicide increases of 19% and 16%, respectively, followed by White women (13%) and Hispanic women (12%). White men and Hispanic men firearm homicide rates increased modestly by 3% and 2%, respectively. On the other hand, non-firearm homicide decreased in the aggregate by 13% and across most subgroups except White women whose homicide victimization rate increased by 3% and Hispanic women whose rate was unchanged.

Homicide Declines, 1993 to 2000

The second subperiod from 1993 to 2000 is commonly referred to as the “Great Crime Decline” (Zimring, 2007). As shown in Figure 1, aggregate total homicide victimization rates sharply declined during this period, resulting in an overall 39% decrease, the largest change across the four subperiods. Homicide victimization also declined across all subgroups, although to varying degrees. Hispanic men, Black women, and Black men experienced the steepest declines, by between 43% and 52% during this subperiod. The homicide victimization rate for Hispanic women, White men, and White women decreased by 38%, 32%, and 26%, respectively.

Firearm homicide victimization demonstrated a similarly stark decrease of 44% in the aggregate and of varying magnitudes across the six subgroups. White women experienced the smallest decrease of 32% while Hispanic men had the greatest decrease of 57%. Non-firearm homicide rates also decreased but less dramatically in the aggregate (25%) and for every subgroup. Hispanic men, Black men, and Black women experienced the greatest decreases in non-firearm homicide of 37%, 36%, and 32%, respectively, while the smallest decreases were observed among White men (13%), Hispanic women (18%), and White women (18%). Thus, during the 1993 to 2000 subperiod, the three subgroups that tended to experience the highest homicide levels—Black men, Hispanic men, and Black women—experienced the steepest declines in total, firearm, and non-firearm homicide.

Homicide Lows and Stabilization, 2000 to 2014

During the third subperiod between 2000 and 2014, aggregate total homicide victimization continued to decline, but to a lesser degree, showing a bit of stabilization with the exception of 2001, the year of the September 11th fatal terrorist attacks. Over this subperiod, homicide declined by 14% in the aggregate and with wide variation of 8% to 39% across subgroups. The decline was the greatest for Hispanic men, Hispanic women, and Black women (39%, 39%, and 32% respectively). Moreover, the 2000 to 2014 period illustrated a great deal of convergence among some subgroups. As Figure 1 depicts, by the end of the subperiod, Hispanic women and White women homicide trends completely converged with one another. Homicide trends for Hispanic men also converged with Hispanic women’s and White women’s, although their homicide levels remained the second highest among all subgroups.

Both firearm and non-firearm homicide also dropped in the aggregate and among all subgroups, with substantial variation in the declines. For the aggregate homicide trend, the decline was greater in non-firearm homicide (29%) than firearm homicide (5%). This pattern generally held across subgroups as well. Black women firearm homicide decreased by 15% and non-firearm homicide decreased by 47%, the largest non-firearm homicide decline among subgroups. Similarly, Hispanic women firearm homicide dropped by 31% while their non-firearm homicide rate decreased by 46%. Hispanic men and Hispanic women experienced the largest decline in firearm homicide of 40% and 31%, respectively, while declines in firearm homicide among the other subgroups ranged from 2% for White men to 15% for Black women. Thus, unlike other subperiods, non-firearm homicide rates changed more dramatically than firearm homicide and appeared to have had a stronger impact on total trends.

Homicide Growth, 2014 to 2016

In the fourth and most recent subperiod from 2014 to 2016, total homicide victimization grew in the aggregate by 22% and among each subgroup. The increase was lowest for Hispanic women whose total homicide victimization rate grew by only 6%. The other subgroups experienced an increase between 18% and 25%. In the aggregate, the 28% increase in firearm homicide and unchanged non-firearm homicide rates suggest that firearms drove nearly all of the total homicide increase. The impact of firearms on homicide during this subperiod coincides with research pointing to the increasing significance of firearms in homicide (Kegler et al., 2018; Lauritsen & Lentz, 2019; Rosenfeld & Fox, 2019). Each subgroup experienced an increase in firearm victimization that ranged from 11% for Hispanic women to 40% for Black women. To the contrary, non-firearm homicide for Black men and Hispanic men decreased by 7% and 2%, respectively, and increased by 10% for White women, and by between only 1% and 3% for the remaining subgroups.

Racial/Ethnic and Gender Disparities in Homicide Trends, 1990 to 2016

In addition to subgroup-specific homicide rates and rate changes, disparities between subgroups are of notable concern. Prior studies suggest convergence, or decreased disparities, between subgroups, specifically between Black and White people, over time (LaFree et al., 2006, 2010; Light & Ulmer, 2016; Tonry & Melewski, 2008). We investigate and build upon this finding by examining disparities by race, ethnicity, and gender, jointly, and properly disaggregating Hispanic victims from Black and White trends for total, firearm, and non-firearm homicide. Disparity ratios comparing each subgroup’s homicide rates with those of White men are graphed in Figure 3 (total homicide) and Figure 4 (firearm and non-firearm homicide) for the 1990 to 2016 period and are tabulated by subperiod in Appendix B. Together, these findings reveal mixed evidence of convergence by race, ethnicity, and gender over the series.

Figure 3.

Disparity ratios between race/ethnic-gender subgroups and White men for age-adjusted total homicide victimization trends.

Figure 4.

Disparity ratios between race/ethnic-gender subgroups and White men for firearm and non-firearm age-adjusted homicide.

During the entire 27-year series, the largest disparity was between Black men and White men, a finding consistent with our earlier outlined expectations. This gap is not only the largest, but is multifold the gaps between the other subgroups and White men. In 1990, for every White man killed, 11.3 Black men died by homicide, as shown in Appendix B. This disparity ratio increased to 12.1 in 1993 then decreased to 10.4 by 2016. Thus, while the disparity between Black men and White men has not grown, its convergence is slight and not as large as prior studies suggest (LaFree et al., 2006, 2010; Light & Ulmer, 2016; Tonry & Melewski, 2008). We also do not see the shift from convergence to divergence illustrated in Light and Ulmer (2016). The second largest disparity was between Hispanic men and White men, a gap that peaked at 5.0 in 1990 and decreased for the remainder of the series to a ratio of 2.2 in 2016, showing convergence. Therefore, this pattern confirms the Latino Paradox (Steffensmeier et al., 2011) for Hispanic men whose homicide risk is intermediate to that of Black men and White men throughout the series. Among women victims, Black women were the only subgroup whose homicide rates exceeded those of White men. Thus, contrary to conventional knowledge that men are more likely than women to become homicide victims, results show that throughout the 27-year period, Black women were more likely than White men to be murdered. In 1990, 2.2 Black women were killed for every White male victim, a ratio that increased to 2.4 in 1993 and decreased to 1.5 by the series’ end, showing some convergence. Disparity ratios for Hispanic women and White women were similar in magnitude and consistently below 1.0, showing that Hispanic women and White women similarly had lower total, firearm, and non-firearm homicide rates than White men. Moreover, the increasing similarity between Hispanic women and White women homicide complicates the Latino Paradox thesis. As Figure 1 showed earlier, during the study period, Hispanic women’s homicide rates were intermediate to Black and White women’s until near the end of the series when their rates were nearly identical to White women’s.

The Black men-White men and Hispanic men-White men gaps were the only disparities that appeared to be consistently driven by firearm homicide, as illustrated in Figure 4. Compared to their ratios for non-firearm homicide, the subgroups’ firearm homicide rate ratios were greater and fluctuated less throughout the series. For example, between 1990 and 2016, the disparity ratio comparing firearm-involved homicide between Black men and White men changed slightly from 13.2 to 13.7. For the same period examining non-firearm homicide, this ratio changed more dramatically, decreasing from 8.2 to 3.9. These ratios further underscore the importance of distinguishing firearm versus non-firearm homicide and firearm versus total homicide, especially for men of color. Thus, contrary to prior conclusions that Black-White disparities in violence and homicide have substantially converged (Cook & Laub, 2002; Fox & Zawitz, 2004; LaFree, 1995; LaFree et al., 2010, 2006; Light & Ulmer, 2016; Parker, 2008; Rosenfeld & Fox, 2019; Tonry & Melewski, 2008), our data suggest that this claim is overstated at the intersection of race, ethnicity, and gender and when accounting the Hispanic Effect.

Measuring the Hispanic Effect on Subgroup Homicide Trends, 1990–2016

As previously mentioned, most prior race-specific homicide studies were unable to disaggregate Hispanic people from data on Black people and White people, creating a significant and longstanding methodological limitation and knowledge gap in homicide research. As Steffensmeier et al., 2011 discussed, this limitation has the effects of 1) neglecting the study of Hispanic-specific trends, 2) confounding Black, but especially White homicide, and 3) masking the true extent of racial disparity between Black people and White people. Our previous analyses addressed these concerns using data that disaggregated Hispanic victims from non-Hispanic Black and White victims and by analyzing Hispanic women and Hispanic men homicide trends juxtaposed to those for other race/ethnic-gender subgroups. Going a step further, we take advantage of our data’s ability to demonstrate the consequence of discounting the Hispanic Effect on subgroup-specific homicide trends. In this vein, for Black and White women and men, we compute “correct” total homicide trends, which exclude Hispanic origin, and “confounded” total homicide trends, which include Hispanic origin. These correct and confounded subgroup-specific trends are displayed in Figure 5.

Figure 5.

Confounded and correct age-adjusted total homicide victimization rates by race/ethnic-gender subgroups.

Figure 5 shows that throughout most of the 1990 to 2016 period, relying on data that misclassify Hispanic victims as Black or White has generally misestimated homicide rates by race, ethnicity, and gender, although to varying degrees. The Hispanic Effect is greatest for White men, whose homicide rates are inflated and misestimated more than any other subgroup. This is largely due to the tendency to overwhelmingly classify Hispanic people as White versus Black and subgroup differences in homicide involvement. When averaging over the 27-year period, failing to appropriately classify Hispanic people has inflated White men homicide rates by a rate difference of 1.8 or 31%.³ This discrepancy was greatest during the early 1990s when Hispanic men homicide rates were highest, and it was lowest near the end of the series when Hispanic men rates were declining. As a result, conclusions drawn from most prior studies that examine confounded White homicide in general, and confounded White men homicide specifically, are premised on inflated rates. White women homicide rates are also affected by the Hispanic Effect, but to a lesser degree. Over the series, White women homicide rates were inflated by an average rate difference of 0.2 or 7%. The correct and confounded White women rates were mostly similar to one another, although to a lesser degree during the early 1990s when the inflation was a rate difference of 0.3 or 11% in 1990.

In comparison to White men and White women trends, the Hispanic Effect is less consequential for Black women and Black men. However, as expected, Black men and Black women homicide rates were underestimated due to the Hispanic Effect. Black men rates were underestimated nearly every year during the series, deflating rates by an average rate difference of 1.4 or 3.0%. An exception was 1990 when their rates were inflated by a rate difference of 2.5 or 4%. Similar to their male counterparts, the Hispanic Effect deflated Black women homicide for nearly the entire series, underestimating their homicide by an average rate difference of 0.2 or 2.6%. An exception was 1990 when Black women homicide was inflated rather than deflated due to the Hispanic Effect by a rate difference of 0.4 or 3.3%. Therefore, homicide victimization among Black men and Black women has been slightly greater than previously documented.

In addition to misestimating subgroup trends, relying on data that fail to disaggregate Hispanic origin from racial categories has underestimated the true extent of racial/ethnic-gender disparities between subgroups. As both Figure 3 and Figure 5 for total homicide show, the greatest disparity was between Black men and White men. However, relying on confounded rates substantially underestimates this disparity. As illustrated in Figure 5, when averaging over the 27-year period, the confounded Black men-White men disparity ratio was 6.9 but the correct ratio was 10.2. Stated differently, the Hispanic Effect has underestimated the gap in homicide risk between Black men and White men by 48%. The disparity between Black women and White men was similarly misestimated. The average Black women-White men ratio was 1.2 using confounded rates but 1.8 using correct rates, a 50% underestimation. Notably, using confounded data gives the incorrect impression that Black women and White men homicide rates converged and were almost the same near the series end, especially between 2008 and 2016. In contrast, correct rates reveal less similarity, a larger gap, and less convergence, therefore challenging claims of large Black-White convergence in prior research.

In addition to the Hispanic Effect underestimating disparities between Black men and White men and Black women and White men, the gap between White women and White men has been underestimated. When averaged over the 27-year period, the White women-White men confounded disparity ratio was 0.4 whereas the correct disparity ratio was 0.5. This discrepancy translates to an average disparity underestimation of 25%, giving the incorrect impression that the homicide risk between White men and White women is more similar than it is.

Finally, because the Hispanic Effect is less pronounced for Black women, Black men, and White women, misestimated disparity ratios were relatively small for these subgroup comparisons. For example, the Black men-White women disparity was underestimated by 11% (confounded ratio = 18.9, correct ratio = 21.0), and the Black women-White women disparity was underestimated by 10% (cofounded ratio = 3.4, correct ratio = 3.8). The Hispanic Effect underestimated the Black men-Black women disparity only negligibly by less than 1% (cofounded ratio = 5.67, correct ratio = 5.70).

Conclusion and Implications

The current study’s goal was to advance the homicide literature by attending to the Hispanic Effect and applying an intersectional approach to studying homicide trends. Averting methodological concerns that arise when using official data, we departed from convention by instead drawing on mortality data from the CDC’s WISQARS. These data allowed us to compute and analyze age-adjusted homicide trends at the intersection of race, ethnicity, and gender from 1990 to 2016 while considering the role of firearms. Results from the analyses reveal patterns that interrogate and add complexity to conventional knowledge deriving from homicide research.

Criminologists generally understand that people of color and men, separately, experience the highest homicide victimization and that White people and women experience the lowest. Our results complicate these widely-held simplifications. We find that Black men, followed by Hispanic men, consistently had the highest homicide risks among all six subgroups. Black women were at third-highest risk of being killed throughout the 27-year series. Therefore, Black women’s homicide trends, especially relative to White men’s, challenge the notion that women are at lowest and men are at highest risk of homicide. These findings suggest that race is a more powerful predictor of homicide risk than gender and that Black women sitting at the intersection of racism and sexism face unique homicide risks that have been understudied.

Another widely-held convention is that Hispanic Americans’ homicide involvement is intermediate to that of Black and White Americans, referred to a “Latino Paradox.” Our analysis revealed mixed support for this longstanding understanding. We found evidence of a Latino Paradox among men across race/ethnicity but not always for women. Specifically, in recent years, Hispanic women homicide rates have been similar to—instead of higher than—White women’s. Thus, the Latino Paradox does not always hold across gender or time, neither does the notion that Hispanic women are at greater risk for being killed than White women. This finding illuminates the distinct experiences and risks between Hispanic versus Black women.

The literature also indicates that the Black-White homicide gap has considerably declined over time. Our results suggest this claim is an overstatement when accounting the Hispanic Effect at the intersection of race, ethnicity, and gender. While the Black women-White men and Black women-White women homicide gaps have converged some, the gap between Black men and White men—the largest disparity—has remained relatively stable. Therefore, race/ethnic-gender disparities in homicide are largely enduring and have not improved to the degree scholars have contended.

Moreover, we found support for our hypothesis that parsing out the Hispanic Effect would reveal exaggerated homicide risks for White men and White women and slightly underestimated homicide risks for Black men and Black women. Our results suggest that the Hispanic Effect is consequential to the study of homicide. It has made White men, and White women to a lesser degree, appear to be of greater risk of being killed than they are while deflating and masking the reality of Black women’s and Black men’s elevated risks.

These findings are further nuanced by the additional consideration of firearm and non- homicide. Attending to firearm versus non-firearm homicide does not discount conclusions about total homicide but rather adds a layer of information for consideration. For example, disaggregating the homicide trend by firearm usage reveals that guns heavily influence homicide trends during some subperiods but less so during others. The homicide increase in the early 1990s was primarily driven by Black men and Black women firearm homicide victimization. Firearms drove almost all of the recent homicide rises from 2014 to 2016, in the aggregate and for most race/ethnic-gender subgroups. Moreover, firearms contributed to race/ethnic-gender disparities in homicide, such as those between Black men and White men over time. These findings, coupled with the fact that guns are used in most homicides and might be of increasing lethality, underscore the import of attending to firearms in homicide research and policy, especially at the intersection of race/ethnicity-gender.

Together these findings point to several implications for research, theory, and policy. First, a fruitful endeavor is for agencies to drastically improve data collection efforts by accurately classifying and coding race and ethnicity alongside conducting analyses at the intersection of race, ethnicity, and gender. The inability and difficulty associated with disaggregating by race, ethnicity, and gender are present in most datasets, including the most authoritative ones generally used to inform national policy decisions (e.g., FBI). Knowledge generated from studying homicide in the aggregate or by a single identifier does not always hold by race/ethnicity-gender, and the Hispanic Effect has resulted in misleading conclusions about subgroup-specific homicide risks and intergroup disparities. This also raises concerns about theories of homicide, whether they are invariant across race/ethnicity-gender, and their need to account for intersectionality in order to account for race/ethnic-gender differences in homicide levels, changes, and patterns. For example, social disorganization perspective—a leading explanation of violence—has attended to racial and ethnic homicide differences, pointing to racial/ethnic differences in concentrated structural disadvantage, family disruption, racial segregation, employment, and racial discrimination. Scholars should work to expand social disorganization theory and develop new theories that simultaneously account for both racialized and gendered oppressions and disadvantages. These endeavors should consider the Black Feminist literature that stresses the role of power structures and disparate disadvantages that multiply-marginalized groups, such as Black Women, experience. This means centering Black women and their experiences in homicide research and accounting for predictors salient to them, such as racism-sexism and domestic violence. Similarly, evaluative research should also critically consider subgroup differences. Just as it is possible that homicide victimization is explained by different factors for different subgroups, it is probable that prevention and intervention effectiveness varies across the population. Initiatives that decrease victimization for one group, for example, might elevate it for another or have a null or lesser effect.

While this study is novel in its longitudinal consideration of the intersection of three identities, additional identities would be useful to consider in future studies and data designs, including class, sexuality, gender expression, and physical and mental ability, amongst others. Lacking these variables does not necessarily disqualify the knowledge generated from this or prior studies. However, their inclusion could contribute substantially to our understanding and prevention of homicide. There is reason to believe that other identifiers are related to violence, such as gender identity and sexuality. Numerous news outlets have been reporting the elevated risk of violent victimization that transwomen (especially of color) face. Our major crime data sources do not allow researchers to track and study the trend of transwomen’s homicide victimization. Future data collection efforts should attempt to capture these trends to broaden our collective understanding of violence and produce effective prevention and intervention efforts. Another avenue for future research is to examine intersectional homicide risks at regional, state, and local levels over time. It is possible that certain geographic units have stronger impacts on national trends than others. Our national-level analyses, though informative, might mask important geographical differences.

In conclusion, the current study represents more than a task of disaggregation but is also framed through an intersectional lens, suggesting that the multiplicity of systems of power related to race, ethnicity, and gender are related to homicide victimization. Therefore, we lay the framework for future studies to continue from an intersectional lens in building up this important body of literature. With this foundation, future studies should be able to consider explanatory factors and investigate whether some relationships are contextual or specific to certain subgroups while also building and refining theories. Ideally, future studies will incorporate variables that reflect the mechanisms central to intersectional theory, such as disparity in social status and structural inequality. Such research endeavors might be better suited to informing policy prescriptions and responses to violence than traditional approaches.

Footnotes

Appendix

Appendix B.

Ratios of Total, Firearm, and Non-Firearm Age-Adjusted Homicide Victimization Rates between Race/Ethnic-Gender Subgroups.

	1990	1993	2000	2014	2016
Total
Black females	2.2	2.4	2.0	1.5	1.5
Black males	11.3	12.1	10.1	9.8	10.4
White females	0.4	0.5	0.5	0.5	0.5
Hispanic females	0.8	0.9	0.8	0.5	0.5
Hispanic males	5.0	4.7	3.3	2.2	2.2
Firearm
Black females	1.7	2.0	1.6	1.4	1.5
Black males	13.2	14.7	13.8	13.5	13.7
White females	0.4	0.4	0.4	0.4	0.4
Hispanic females	0.7	0.8	0.6	0.5	0.4
Hispanic males	5.5	5.4	4.0	2.4	2.5
Non-Firearm
Black females	3.1	3.2	2.5	1.6	1.6
Black males	8.2	7.3	5.4	4.2	3.9
White females	0.6	0.7	0.6	0.6	0.7
Hispanic females	0.9	1.0	1.0	0.6	0.6
Hispanic males	4.2	3.3	2.4	1.8	1.7

Note. White men are the reference group and included in the denominator of ratio calculations.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Shytierra Gaston

Notes

Author Biographies

Shytierra Gaston is an assistant professor in the Department of Criminal Justice and Criminology at Georgia State University. She uses quantitative and qualitative methodologies to investigate issues related to race, crime, and criminal (in)justice as well as corrections, prisoner reentry, and collateral consequences of criminal justice contact. Dr. Gaston’s scholarship can be found in Criminology, Crime & Delinquency, Criminal Justice & Behavior, Urban Affairs Review, and Homicide Studies.

CheyOnna Sewell is the assistant director of Social Science Education with the Miramontes Arts and Sciences Program at the University of Colorado Boulder. She applies an intersectional-lens to the study of gendered violence, responses to violence, and abolition.

References

Bailey

(1999). The socioeconomic status of women and patterns of forcible rape for major U.S. cities. Sociological Focus 32(1), 43–63.

Bledsoe

(2019). The primacy of anti-blackness. Area, 1–8. https://doiorg.colorado.idm.oclc.org/10.1111/area.12599

Blumstein

(1995). Youth violence, guns, and the illicit-drug industry. The Journal of Criminal Law and Criminology, 86(1), 10–36. https://doi.org/10.2307/1143998

Blumstein

Rivara

F. P.

Rosenfeld

(2000). The rise and decline of homicide—And why. Annual Review of Public Health, Palo Alto, 21, 505–541.

Blumstein

Rosenfeld

(1998). Explaining recent trends in U.S. homicide rates. The Journal of Criminal Law and Criminology, 88(4), 1175–1216. https://doi.org/10.2307/1144254

Blumstein

Wallman

(Eds.). (2005). The crime drop in America (2nd ed.). Cambridge University Press.

Bowleg

(2008). When Black + Lesbian + Woman ≠ Black lesbian woman: The methodological challenges of qualitative and quantitative intersectionality research. Sex Roles, 59(5), 312–325. https://doi.org/10.1007/s11199-008-9400-z

Burgess-Proctor

(2006). Intersections of race, class, gender, and crime: Future directions for feminist criminology. Feminist Criminology, 1(1), 27–47. https://doi.org/10.1177/1557085105282899

Charles

C. Z.

(2003). The dynamics of racial residential segregation. Annual Review of Sociology, 29(1), 167–207. https://doi.org/10.1146/annurev.soc.29.010202.100002

10.

Chauhan

Cerdá

Messner

S. F.

Tracy

Tardiff

Galea

(2011). Race/ethnic-specific homicide rates in New York City evaluating the impact of broken windows policing and crack cocaine markets. Homicide Studies, 15(3), 268–290. https://doi.org/10.1177/1088767911416917

11.

Cook

P. J.

(2018). Gun markets. Annual Review of Criminology, 1(1), 359–377. https://doi.org/10.1146/annurev-criminol-032317-092149

12.

Cook

P. J.

Laub

J. H.

(2002). After the epidemic: Recent trends in youth violence in the United States. Crime and Justice, 29, 1–37. https://doi.org/10.1086/652218

13.

Crenshaw

(1989). Demarginalizing the intersection of race and sex: A Black feminist critique of antidiscrimination doctrine, feminist theory and antiracist politics. University of Chicago Legal Forum, 1989, 139.

14.

DeLisi

Kosloski

Sween

Hachmeister

Moore

Drury

(2010). Murder by numbers: Monetary costs imposed by a sample of homicide offenders. The Journal of Forensic Psychiatry & Psychology, 21(4), 501–513. https://doi.org/10.1080/14789940903564388

15.

Fingerhut

L. A.

Ingram

D. D.

Feldman

J. J.

(1992). Firearm homicide among black teenage males in metropolitan counties: Comparison of death rates in two periods, 1983 through 1985 and 1987 through 1989. JAMA, 267(22), 3054–3058. https://doi.org/10.1001/jama.1992.03480220072030

16.

Fingerhut

L. A.

Kleinman

J. C.

(1990). International and interstate comparisons of homicide among young males. JAMA, 263(24), 3292–3295.

17.

Fox

J. A.

Fridel

E. E.

(2017). Gender differences in patterns and trends in U.S. homicide, 1976–2015. Violence and Gender, 4(2), 37–43. https://doi.org/10.1089/vio.2017.0016

18.

Fox

J. A.

Zawitz

M. W.

(2007). Homicide trends in the United States. Bureau of Justice Statistics. https://www.bjs.gov/content/pub/pdf/htius.pdf

19.

Fox

J. A.

Zawitz

M. W.

(2004). Homicide trends in the United States: 2000 update (No. NCJ 204885). Bureau of Justice Statistics. https://www.bjs.gov/content/pub/pdf/htus00.pdf

20.

Gaston

Cunningham

J. P.

Gillezeau

(2019). A ferguson effect, the drug epidemic, both, or neither? Explaining the 2015 and 2016 U.S. homicide rises by race and ethnicity. Homicide Studies, 23(3), 285–313. https://doi.org/10.1177/1088767919849642

21.

Haynie

D. L.

Armstrong

D. P.

(2006). Race and gender-disaggregated homicide offending rates: Differences and similarities by victim-offender relations across cities. Homicide Studies, 10(1), 3–32. https://doi.org/10.1177/1088767905281518

22.

Heron

(2017). Deaths: Leading causes for 2015. National Vital Statistics Reports, 66(5). https://www.cdc.gov/nchs/data/nvsr/nvsr66/nvsr66_05.pdf

23.

Iceland

Weinberg

D. H.

Steinmetz

(2002). Racial and ethnic residential segregation in the United States: 1980–2000 (p. 151) [Census 2000 Special Reports]. U.S. Census Bureau.

24.

Kaufman

J. M.

(2005). Explaining the race/ethnicity–violence relationship: Neighborhood context and social psychological processes. Justice Quarterly, 22(2), 224–251. https://doi.org/10.1080/07418820500088986

25.

Kegler

S. R.

Dahlberg

L. L.

Mercy

J. A.

(2018). Firearm homicides and suicides in major metropolitan areas—United States, 2012–2013 and 2015–2016. Morbidity and Mortality Weekly Report, 67(44), 1233–1237. https://doi.org/10.15585/mmwr.mm6744a3

26.

Krivo

L. J.

Peterson

R. D.

(2000). The structural context of homicide: Accounting for racial differences in process. American Sociological Review, 65(4), 547–559.

27.

LaFree

(1995). Race and crime trends in the United States, 1946–1990. In Hawkins

D. F.

(Ed.), Ethnicity, race, and crime (pp. 169–193) State University of New York press.

28.

LaFree

Baumer

E. P.

O’Brien

(2010). Still separate and unequal? A city-level analysis of the Black-White gap in homicide arrests since 1960. American Sociological Review, 75(1), 75–100.

29.

LaFree

O’Brien

R. M.

Baumer

(2006). Is the gap between Black and White arrest rrates nnarrowing? National trends for personal contact crimes, 1960–2003. In Peterson

R. D.

Krivo

L. J.

Hagan

(Eds.), The many colors of crime: Inequalities of race, ethnicity, and crime in America. NYU Press.

30.

Land

K. C.

McCall

P. L.

Cohen

L. E.

(1990). Structural covariates of homicide rates: Are there any invariances across time and social space? American Journal of Sociology, 95(4), 922–963. https://doi.org/10.1086/229381

31.

Langley

Sugarmann

(2018). Hispanic victims of lethal firearms violence in the United States. Violence Policy Center. http://www.vpc.org/studies/hispanic18.pdf

32.

Lauritsen

J. L.

Heimer

(2010). Violent victimization among males and economic conditions. Criminology & Public Policy, 9(4), 665–692. https://doi.org/10.1111/j.1745-9133.2010.00660.x

33.

Lauritsen

J. L.

(2012). Subpopulations at high risk for rape and sexual assault: What does the NCVS tell us? Report prepared for the Panel on Measuring Rape and Sexual Assault in Bureau of Justice Statistics Surveys, Committee on National Statistics. Retrieved June 5, 2012, from https://sites.nationalacademies.org/cs/groups/dbassesite/documents/webpage/dbasse_080065.pdf

34.

Lauritsen

J. L.

Lentz

T. S.

(2019). National and local trends in serious violence, firearm victimization, and homicide. Homicide Studies, 23(3), 243–261. https://doi.org/10.1177/1088767919848665

35.

Light

M. T.

Ulmer

J. T.

(2016). Explaining the gaps in White, Black, and hispanic violence since 1990: Accounting for immigration, incarceration, and inequality. American Sociological Review, 81(2), 290–315. https://doi.org/10.1177/0003122416635667

36.

Loftin

McDowall

Fetzer

M. D.

(2008). A Comparison of SHR and vital statistics homicide estimates for U.S. cities. Journal of Contemporary Criminal Justice, 24(1), 4–17. https://doi.org/10.1177/1043986207312585

37.

Martínez

(2014). Latino homicide: Immigration, violence, and community. Routledge. https://doi.org/10.4324/9781315776064

38.

Martin

Vieraitis

L. M.

Britto

(2006). Gender equality and women’s absolute status: A test of the feminist models of rape. Violence Against Women, 12, 321–339.

39.

Mastrocinque

J. M.

Metzger

J. W.

Madeira

Lang

Pruss

Navratil

P. K.

Sandys

Cerulli

(2015). I’m still left here with the pain: Exploring the health consequences of homicide on families and friends. Homicide Studies, 19(4), 326–349. https://doi.org/10.1177/1088767914537494

40.

McNulty

T. L.

Bellair

P. E.

(2003). Explaining racial and ethnic differences in serious adolescent violent behavior. Criminology, 41(3), 709–747. https://doi.org/10.1111/j.1745-9125.2003.tb01002.x

41.

Messner

S. F.

Golden

R. M.

(1992). Racial inequality and racially disaggregated homicide rates: An assessment of alternative theoretical explanations. Criminology, 30(3): 421–442.

42.

Messner

S. F.

Rosenfeld

(2013). Crime and the American Dream (5th ed.). Wadsworth.

43.

Ousey

G. C.

(1999). Homicide, structural factors, and the racial invariance assumption. Criminology, 37(2), 405–426. https://doi.org/10.1111/j.1745-9125.1999.tb00491.x

44.

Ousey

G. C.

Lee

M. R.

(2004). Investigating the connections between race, illicit drug markets, and lethal violence, 1984–1997. Journal of Research in Crime and Delinquency, 41(4), 352–383. https://doi.org/10.1177/0022427803262059

45.

Parker

K. F.

(2001). A move toward specificity: Examining urban disadvantage and race-and relationship-specific homicide rates. Journal of Quantitative Criminology, 17(1), 89–110. https://doi.org/10.1023/A:1007578516197

46.

Parker

K. F.

(2008). Unequal Crime Decline. New York University Press.

47.

Parker

K. F.

Hefner

M. K.

(2015). Intersections of race, gender, disadvantage, and violence: Applying intersectionality to the macro-level study of female homicide. Justice Quarterly, 32(2), 223–254. https://doi.org/10.1080/07418825.2012.761719

48.

Parker

K. F.

Mancik

Stansfield

(2016). American crime drops: Investigating the breaks, dips and drops in temporal homicide. Social Science Research, 64, 154–170. https://doi.org/10.1016/j.ssresearch.2016.09.029

49.

Peterson

R. D.

Bailey

W. C.

(1992). Rape and dimensions of gender socioeconomic inequality in U.S. metropolitan areas. Journal of Research in Crime and Delinquency, 29, 162–177.

50.

Peterson

R. D.

Krivo

L. J.

(1993). Racial segregation and Black urban homicide. Social Forces, 71(4), 1001–1026. https://doi.org/10.2307/2580128

51.

Phillips

J. A.

(2002). White, Black, and latino homicide rates: Why the difference? Social Problems, 49(3), 349–373. https://doi.org/10.1525/sp.2002.49.3.349

52.

Potter

(2006). An Argument for Black feminist criminology understanding African American women’s experiences with intimate partner abuse using an integrated approach. Feminist Criminology, 1(2), 106–124. https://doi.org/10.1177/1557085106286547

53.

Potter

(2013). Intersectional criminology: Interrogating identity and power in criminological research and theory. Critical Criminology, 21(3), 305–318. https://doi.org/10.1007/s10612-013-9203-6

54.

Potter

(2015). Intersectionality and Criminology: Disrupting and revolutionizing studies of crime (1st ed.). Routledge.

55.

Ransford

H. E.

(1980). The prediction of social behavior and attitudes. In Jeffries

Ransford

H. E.

(Eds.), Social stratification: A multiple hierarchy approach (pp. 265–295). Allyn and Bacon.

56.

Reckdenwald

Parker

K. F.

(2008). The Influence of gender inequality and marginalization on types of female offending. Homicide Studies, 12(2), 208–226. https://doi.org/10.1177/1088767908314270

57.

Redelings

Lieb

Sorvillo

(2010). Years off your life? The effects of homicide on life expectancy by neighborhood and race/ethnicity in Los Angeles county. Journal of Urban Health, 87(4), 670–676. https://doi.org/10.1007/s11524-010-9470-4

58.

Rosenfeld

Fox

J. A.

(2019). Anatomy of the homicide rise. Homicide Studies, 23(3), 202–224. https://doi.org/10.1177/1088767919848821

59.

Rosenfeld

Gaston

Spivak

Irazola

(2017). Assessing and responding to the recent homicide rise in the United States (No. NCJ 251067). National Institute of Justice. https://www.ncjrs.gov/pdffiles1/nij/251067.pdf

60.

Rosenfeld

Oliver

B. E.

(2008). Evaluating recent changes in homicide and robbery rates. Justice Research and Policy, 10(2), 49–65. https://doi.org/10.3818/JRP.10.2.2008.49

61.

Sampson

R. J.

Morenoff

J. D.

Raudenbush

(2005). Social anatomy of racial and ethnic disparities in violence. American Journal of Public Health, 95(2), 224–232. https://doi.org/10.2105/AJPH.2004.037705

62.

Sampson

R. J.

Wilson

W. J.

(1995). Toward a theory of race, crime, and urban inequality. In Hagan

Peterson

R. D.

(Eds.), Crime and inequality (pp. 37–56). Standford Univerity Press.

63.

Selvin

Sacks

S. T.

Merrill

D. W.

(1980). Standardization of age-adjusted mortality rates (No. LBL-10323). https://doi.org/10.2172/5277554

64.

Semega

Kollar

Creamer

Mohanty

(2019). Income and Poverty in the United States: 2018 (No. P60-266). U.S. Census Bureau. https://www.census.gov/library/publications/2019/demo/p60-266.html

65.

Steffensmeier

Feldmeyer

Harris

C. T.

Ulmer

J. T.

(2011). Reassessing trends in Black violent crime, 1980–2008: Sorting out the “Hispanic Effect” in uniform crime reports arrests, national crime victimization survey offender estimates, and U.S. prisoner counts. Criminology, 49(1), 197–251. https://doi.org/10.1111/j.1745-9125.2010.00222.x

66.

Titterington

V. B.

(2006). A retrospective investigation of gender inequality and female homicide victimization. Sociological Spectrum, 26, 205–236.

67.

Tonry

Melewski

(2008). The malign effects of drug and crime control policies on Black Americans. Crime and Justice, 37(1), 1–44. https://doi.org/10.1086/588492

68.

Unnever

J. D.

Gabbidon

S. L.

(2011). A theory of African American offending race, racism, and crime. Routledge.

69.

Velez

M. B.

Krivo

L. J.

Peterson

R. D.

(2003). Structural inequality and homicide: An assessment of the black-white gap in killings. Criminology, 41(3), 645–672. https://doi.org/10.1111/j.1745-9125.2003.tb01000.x

70.

Vieraitis

L. M.

Britto

Kovandzic

T. V.

(2007). The impact of women’s status and gender inequality on female homicide victimization rates: Evidence from U.S. counties. Feminist Criminology, 2, 57–73.

71.

Vieraitis

L. M.

Williams

M. R.

(2002). Assessing the impact of gender inequality on female homicide victimization across U.S. cities: A racially disaggregated analysis. Violence Against Women, 8(1), 35–63. https://doi.org/10.1177/10778010222182937

72.

Whaley

R. B.

Messner

S. F.

(2002). Gender equality and gendered homicides. Homicide Studies, 6(3), 188–210. https://doi.org/10.1177/108876790200600302

73.

Xie

Heimer

Lauritsen

J. L.

(2012). Violence against women in U.S. metropolitan areas: Changes in women’s status and risk, 1980–2004. Criminology, 50(1), 105–143. doi: 10.1111/j.1745-9125.2011.00245.x

74.

Zimring

F. E.

(2007). The great American crime decline (1st ed.). Oxford University Press.

75.

Zimring

F. E.

Hawkins

(1997). Crime is not the problem: Lethal violence in America (Univ PR Pbk ed.). Oxford University Press.