Community Characteristics and the Geographic Distribution of Intimate Partner Violence: A Partial Test of Social Disorganization Theory

Abstract

Shaw and McKay’s social disorganization theory has provided a framework to examine the relationship between community-level structural variables and neighborhood crime. Although empirical support for the theory has been widely demonstrated using property and violent crime data, the body of literature examining the theory’s applicability to intimate partner violence (IPV) is more limited. Further, much of the existing literature in this area applies the theory’s macro-level variables to individual outcomes instead of assessing community effects. Using negative binomial regression to analyze incident data from the Austin (TX) Police Department and demographic information from the United States Census Bureau, this study assesses the relationship between concentrated disadvantage, racial/ethnic heterogeneity, residential instability, and the geographic distribution of IPV incidents in a major U.S. city with no racial or ethnic majority. The independent variables of interest were constructed using principal axis factoring, and a spatial lag variable was included in the model to control for spatial clustering. The analysis showed that concentrated disadvantage was significantly, positively associated with annual counts of IPV incidents in neighborhoods, as was the control variable total crime reports. These results demonstrate that the geographic distribution of IPV is influenced by community factors. They underscore the importance of considering community-wide prevention and intervention efforts in tandem with individual services to those impacted by IPV.

Keywords

intimate partner violence social disorganization neighborhoods geography

Intimate partner violence (IPV), defined by the Centers for Disease Control and Prevention (CDC) as “physical violence, sexual violence, stalking and psychological aggression (including coercive tactics) by a current or former intimate partner (i.e., spouse, boyfriend/girlfriend, dating partner, or ongoing sexual partner),” is a pervasive global public health issue (Breiding et al., 2015, p. 11). The CDC estimates that in the United States, more than 36% of women and more than 33% of men have experienced IPV in their lifetime. IPV has several adverse health effects, including anxiety, depression, chronic pain, injury, and death (Smith et al., 2018). Despite this pervasiveness, the extant IPV literature largely focuses on individual risk factors for both victimization and perpetration such as age, race/ethnicity, education level, relationship type, pregnancy status, substance use, prior victimization, or personal/household income (e.g., Brownridge et al., 2011; Caetano et al., 2010; Capaldi et al., 2012; Han & Choi, 2021; Reingle et al., 2012; Rennison & Rand, 2003; Wood et al., 2021).

The relationship between community characteristics and IPV is less studied, perhaps because the influence of these characteristics is thought to be limited to public crimes, whereas IPV typically occurs in private. This discrepancy persists despite evidence that variables related to socioeconomic status, racial composition of neighborhoods, and residential mobility are often among the strongest macro-level predictors of crime (Pratt & Cullen, 2005). Whether focusing on individual or macro-level outcomes, criminologists have most commonly used social disorganization’s theoretical framework to assess the impact of neighborhood-level factors on the incidence of IPV (Beyer et al., 2015; Pinchevsky & Wright, 2012; Sampson & Groves, 1989; Shaw & McKay, 1942). Some skeptics might contend this is a misapplication given that social disorganization theory’s causal mechanism hinges on how its independent variables of interest affect the ability of a community to leverage informal social control of crime (Bursik, 1988; Kornhauser, 1978; Sampson & Groves, 1989). We believe this skepticism, while helping to foment more meticulous scholarship, is overly conservative. The influence of one’s social and community environment persists behind closed doors, as empirical evidence regarding social context and IPV has shown (Benson et al., 2003; Browning, 2002; Goodson & Bouffard, 2019; Miles-Doan, 1998).

In addition to the relatively few studies that use aggregate units of analysis such as neighborhoods and/or Census tracts (Boggess & Chamberlain, 2021; Miles-Doan, 1998; Miles-Doan & Kelly, 1997), and counties (Blumenstein & Jasinski, 2015; Goodson & Bouffard, 2017, 2019), existing research often centers the micro-level outcome in multilevel analyses (Benson et al., 2004; Browning, 2002), and constrains the outcome to certain offense types such as assault or rape (Goodson & Bouffard, 2017; Miles-Doan, 1998). Given the limited study of social disorganization’s neighborhood-level variables on macro-level IPV outcomes, more research is needed in this area. The current study contributes to this body of research by analyzing the relationship between three social disorganization variables—concentrated disadvantage, racial/ethnic heterogeneity, and residential instability—and annual neighborhood counts of IPV incidents reported to the police in Austin, Texas in 2018 (Sampson & Groves, 1989; Shaw & McKay, 1942). This study also examines a rarely considered type of geographic area in the United States, a city with no racial or ethnic majority, in which a substantial portion of households speak Spanish as a first language (Austin City, 2021; Data USA, 2021).

Literature Review

Social Disorganization Theory

Shaw and McKay (1942) contended that high levels of three neighborhood-level characteristics—low socioeconomic status, ethnic heterogeneity and immigrant concentration, and residential mobility—indirectly affect neighborhood crime rates by inhibiting the ability of residents to embrace shared values and goals in order to exert informal social control and maintain order. Absent this informal social control, crime and delinquency will flourish (Bursik, 1988; Kornhauser, 1978; Shaw & McKay, 1942).

In 1989, Sampson and Groves added family disruption and urbanization to Shaw and McKay’s model. Together, these variables were contended to lead to three endogenous factors: “sparse local friendship networks,” “unsupervised peer groups,” and “low organizational participation,” which partially mediated the effects of the five exogenous variables on crime and delinquency (Sampson & Groves, 1989, p. 783). These seminal works inarguably established the macro-level context of place and crime as a major pillar of criminological theory but did not consider whether that context extended off the street and into homes.

Social Disorganization Theory and IPV

The influence of neighborhood-level covariates on IPV has been relatively understudied compared to individual-level covariates. When neighborhood-level variables have been included, they are most often used to predict an individual’s risk for victimization or offending rather than a community’s risk for IPV occurrence within it (e.g., Benson et al., 2004; Browning, 2002; Chang et al., 2015). Overlooking community-level factors minimizes the importance of societal context in shaping individual behaviors. It also discounts the potential importance of the same crime indicators possessing varying levels of empirical support at different units of analysis (Pratt & Cullen, 2005).

Application of neighborhood-level variables to IPV was first done by O’Campo et al. (1995), who found that low neighborhood per-capita income and higher neighborhood unemployment rates both increased a woman’s risk of IPV victimization during her childbearing year regardless of individual per-capita income. This study introduced a much-needed perspective to IPV scholarship by highlighting the influence of social factors on IPV risk, but nevertheless applied neighborhood variables to individual outcomes. Soon after, Miles-Doan and Kelly (1997) demonstrated that IPV assault incidents clustered in Duval County (FL) neighborhoods with high levels of concentrated poverty. Miles-Doan (1998) further found an association between concentrated poverty and neighborhood intimate partner assaultive violence rates over and above those of concentrated poverty and non-intimate assaultive violence in Duval County.

Subsequent research using aggregate-level covariates has most often tended toward O’Campo et al.’s application to individuals instead of the geographic focus of Miles-Doan and Kelly, with rare exception (e.g., Blumenstein & Jasinski, 2015; Goodson & Bouffard, 2017; 2019; Gracia et al., 2015; Wooldredge & Thistlethwaite, 2003). This application has persisted even when study authors explicitly state that a social disorganization theoretical framework is being used to guide their research, despite social disorganization theory being intended to explain community crime rates, not individual likelihood of victimization or perpetration (Shaw & McKay, 1942).

Regardless of whether the theory has been applied to individual- or aggregate-level outcomes, at least partial support has been found for a link between social disorganization’s variables of interest and IPV (Beyer et al., 2015; Pinchevsky & Wright, 2012; VanderEnde et al., 2012).

Concentrated disadvantage

The most consistent statistically significant predictor of IPV in the social disorganization literature is neighborhood disadvantage, conceptualized and operationalized differently (e.g., socioeconomic status, deprivation, poverty, and concentrated disadvantage) across studies, albeit with several common indicators (Beyer et al., 2015; Pinchevsky & Wright, 2012). These indicators often include median income, poverty rate, use of public assistance, education level, percent of population below age 18, female-headed households (both with and without minor children), percent unemployment, percent non-white, and percent Black (Beyer et al., 2015; Blumenstein & Jasinski, 2015; Pinchevsky & Wright, 2012). Studies have generally found the disadvantage-IPV relationship to be positive (e.g., Benson et al., 2003; Benson et al., 2004; Lee et al., 2013; Miles-Doan, 1998; Van Wyk et al., 2003; Wooldredge & Thistlethwaite, 2003), or nonsignificant (e.g., Lanier & Maume, 2009; Caetano et al., 2010; Li et al., 2010). Discrepancies in these findings could be due to differences in measurement and model specification, use of Census data from 1990 versus 2000, or a combination of the two.

In some instances, the disadvantage-IPV relationship was attenuated by other variables. Browning (2002) found that concentrated disadvantage was positively associated with intimate partner homicide, but was nonsignificant when mediated by collective efficacy. Wright and Benson’s (2010) findings contradicted Browning: although concentrated disadvantage and collective efficacy were related to IPV in the expected directions in main effects models, collective efficacy did not mediate the effect of neighborhood disadvantage, which remained significant in a model that included both variables. Fox and Benson (2006) found that IPV was more prevalent among economically vulnerable couples and, separately, couples who lived in more highly disadvantaged neighborhoods. Although neighborhood disadvantage did not change economically vulnerable couples’ IPV risk, it had a significant and positive effect on IPV risk among economically secure couples living in disadvantaged neighborhoods versus those living in advantaged neighborhoods.

Racial/ethnic heterogeneity and immigrant concentration

The significance and direction of the relationship between racial/ethnic heterogeneity and/or immigrant concentration and IPV has been inconsistent, particularly in more recent research. The indicators most commonly used to construct this variable are percent foreign born, percent Hispanic/Latino, Blau’s index of diversity, percent non-White, and percent Black (Beyer et al., 2015). Percent of limited-English-speaking households has sometimes been used as an indicator (Beyer et al., 2015; Blau, 1977; Pinchevsky & Wright, 2012). Recent research has generally found this variable to be nonsignificant (Browning, 2002; Frye et al., 2008; Frye & Wilt, 2001). However, Goodson and Bouffard (2017) found a significant and positive relationship between ethnic heterogeneity and intimate partner assault across several relationship types in both rural and urban settings. They proposed this result could stem from a difference in level of analysis and/or measurement. Some studies measured race and ethnicity at the individual level instead of or in addition to measuring neighborhood racial composition (Benson et al., 2003; Frye et al., 2008). Other studies included percent Latino as one of the indicators of an area’s heterogeneity (Browning, 2002; Frye et al., 2008; Frye & Wilt, 2001; Wright & Benson, 2010). In contrast, Goodson and Bouffard relied on Blau’s index of diversity and did not include Hispanic/Latino ethnicity in that indicator (Blau, 1977).

Two studies conceptualizing this variable as immigrant concentration found it was negatively related to IPV outcome. Wright and Benson (2010) proposed this was due to the “immigrant paradox”; status as “other” created greater in-group cohesion and a resulting protective effect against violence. Pearlman et al. (2003) found that Spanish-speaking-only Hispanic women living in linguistically isolated areas had lower rates of police-reported IPV.

Similar to prior studies contradicting Shaw and McKay, Blumenstein and Jasinski (2015) found that an increase in a county’s racial/ethnic heterogeneity was associated with a 79% decrease in the rate of intimate partner assaults. However, they also found that an increase in a county’s percent Hispanic residents was significantly related to an increase in the rate of intimate partner assaults. Kimber et al. (2013) found that first-generation immigrant status was a protective factor against physical/sexual IPV in an unadjusted model. In contrast, an interactive effect between sex and immigrant status showed first-generation female immigrants living in areas of high immigrant concentration were significantly more likely to report physical/sexual IPV victimization than third-generation females. Finally, Lee et al. (2013) found neighborhood immigrant concentration to predict arrest for IPV, which they attributed to a conflict theory perspective in which law enforcement maintain order in service to the majority by exerting power over minority populations (Black, 1976).

Residential mobility

Whether conceptualized as stability or instability, nearly every social disorganization-IPV study uses percent renters and percent new residents to operationalize residential mobility (Beyer et al., 2015; Blumenstein & Jasinski, 2015; Pinchevsky & Wright, 2012; VanderEnde et al., 2012). Rapid residential turnover is said to inhibit the development of bonds with neighbors that facilitate effective informal social control. Moreover, residents who do not intend to stay in a particular community will not financially or emotionally invest in its well-being (Bursik, 1988; Kornhauser, 1978; Sampson & Groves, 1989; Shaw & McKay, 1942). Therefore, greater residential instability should predict higher crime. Recent scholarship, however, does not support this hypothesis. Instead, residential instability has been found to have a protective or nonsignificant effect against IPV. Benson et al. (2003), finding that increased residential mobility had a significant negative effect on IPV, suggested that this result reflected a changing residential landscape in which individuals faced sociostructural barriers to leaving disadvantaged areas while more educated individuals moved more frequently (Farley, 1996; Wilson, 1987). Similarly, Li et al. (2010) found that residential stability increased the risk of IPV for low-income pregnant women in Jefferson County, Alabama, independent of individual- and household-level variables. The authors attributed this result to the lack of resources to leave a bad situation. Conversely, Boggess and Chamberlain (2021) found that homeownership, typically correlated with longer residential tenure and greater investment in one’s community, had a mitigating effect on partner violence. Waller et al., (2012) found residential mobility to be nonsignificant.

The Present Study

Consistent with the diversity of IPV incident types, this study includes a full range of reported IPV offenses. This range encompasses not only violent offenses like aggravated assault, sexual assault, and rape, but also includes non-violent offenses such as burglary and vandalism intended to intimidate and harass victims. Lastly, this study does not distinguish gender or gender pairing of victim and perpetrator, which is consistent with its focus on community context, not individual characteristics. Guided by Shaw and McKay’s (1942) social disorganization theory, it is hypothesized that concentrated disadvantage, racial/ethnic heterogeneity, and residential instability will be positively related to annual counts of IPV incident reports per neighborhood.

Method

Data Collection and Sample Selection

This present research was conducted using incident report data from the City of Austin, Texas Police Department for calendar year 2018, downloaded from the city’s open data portal; as well as the U.S. Census Bureau’s GIS mapping resources and its American Community Survey (ACS) Five-Year Estimates for 2014–2018.

Unit of Analysis

The unit of analysis, neighborhood, was operationalized using U.S. Census tracts. Some critics argue that Census tracts impose artificial delineations that do not accurately reflect existing neighborhoods’ socially defined boundaries; however, tracts are nevertheless widely used as a proxy measure in the extant literature (Beyer et al., 2015; Pinchevsky & Wright, 2012). Tracts also facilitate the use of detailed U.S. Census statistics: the variables used in this study were constructed using ACS data measured at the Census tract level. Therefore, it was determined that the Census tract, as a proxy for neighborhood, was the most appropriate unit of analysis for this research. Austin’s city limits were used to isolate the Austin tracts (n = 219) from a U.S. Census TIGER/Line file of all Texas Census tracts. Three tracts were then eliminated from the sample because they did not meet the conceptual definition of a residential neighborhood—one was an assisted living facility, one was the Travis County Correctional Complex, and the last was the Austin-Bergstrom International Airport. A total of 216 Census tracts were included in the final sample.

Dependent Variable

The dependent variable, annual counts of IPV incident reports per neighborhood, was sourced from the Austin Police Department’s (APD) Crime Reports dataset (Austin Police Department, 2020). This dataset, updated weekly, contains a record of every call for service for which APD officers have written a report, dating back to 2003. Each record includes either an exact street address or the nearest intersection and indicates whether each incident is family-violence-related, a classification that includes IPV per department policy (Austin Police Department, 2017).

A public information request was made to APD for the victim-offender linkage data for all family violence incidents that had both occurred and been reported in calendar year 2018 (n = 8010). Incidents were then selected into the sample only if two criteria were met: the linkage data indicated a spouse, ex-spouse, common-law spouse, boyfriend/girlfriend, or ex-boyfriend/girlfriend relationship type; and the incident did not involve minors. Next, the data were uploaded to ArcGIS Pro and geocoded. All incidents in the dataset uploaded to ArcGIS Pro were successfully matched to an address or intersection. Incidents falling outside Austin’s city limits or within one of the three Census tracts that did not qualify as residential neighborhoods were eliminated from the dataset. Because the addresses used for geocoding indicated where the reports were taken, not where the incidents occurred, the incident data where APD is headquartered were specifically examined. This was done to ensure that a disproportionate number of reports taken at headquarters did not distort the data such that it changed the context of “incident location.” Both a visual assessment of the mapped incidents and an examination of the addresses in the data used to geocode these locations indicated a small number of incident reports matching APD headquarters (n = 8). The final count of IPV incident reports was 4500.

A count outcome variable was used instead of a rate outcome for a few reasons. Counts are more appropriate than rates when the value of the dependent variable cannot be negative, which is the case with this research (Long, 1997). Crime, including IPV, is geographically concentrated (Sherman et al., 1989; Kubrin & Weitzer, 2003; Weisburd, 2015). Using counts better reflects this uneven spatial distribution and facilitates the mapping of geocoded incident report data, thereby providing more detailed information about possible clusters of IPV incident reports than a rate.

Independent Variables

The social disorganization independent variables of interest—concentrated disadvantage, racial/ethnic heterogeneity, and residential instability—were derived using the ACS Five-Year Estimates for 2014–2018. This dataset provides the most robust and reliable statistics collected by the ACS at the Census tract-level unit of analysis (U.S. Census Bureau, 2019). These three variables were constructed using confirmatory factor analysis and principal axis factoring. Indices of each measure were constructed using factor scores. Higher index values represent higher levels of the variables; zero values represent the average level of each, not their absence.

The concentrated disadvantage variable was initially constructed using percent of labor force unemployed, percent of adults without a high school diploma or equivalent, percent of female-headed households with children under 18 years of age, percent of households below the poverty line, and percent of households using the Supplemental Nutrition Assistance Program (SNAP). All indicators had salient loadings (>.3) except for percent unemployed, which was dropped from the factor. The remaining four indicators were then assessed for reliability using Cronbach’s alpha (α = .89).

Racial/ethnic heterogeneity was constructed using three indicators: percent Hispanic or Latino, Blau’s index of diversity¹ (k = 3), and percent foreign born. All three indicators were salient and returned an acceptable Cronbach’s alpha (α = .81).

Residential instability was constructed using percent of householders who rent and percent of residents who moved to their current dwelling within the prior 4 years. Both indicators were salient, and the factor had acceptable reliability (α = .88). The latter indicator was used in lieu of prior research’s 5-year timeframe because the U.S. Census Bureau transitioned to using a 1-year timeframe to measure that value. Therefore, the 4-year timeframe was the closest to the literature’s previously used timeframe that could be constructed using available data.

The control variables used for this study were total crime reports, males per 100 females, median age, and population density (measured as population per square mile). Total crime incidents that resulted in a report for 2018 were retrieved from the Austin open data portal. These incidents excluded the IPV incidents, and were geocoded to the 216 tracts using XY data, resulting in 93,771 reports. The males per 100 females measure was employed as an approximation of “maleness,” similar to prior research that considered the effects of community-level gender dynamics (Beyer et al., 2015; Goodson & Bouffard, 2019). The use of median age was based on previous findings showing that higher age served as a protective factor against IPV victimization (Rennison & Rand, 2003; Rennison & Welchans, 2000). Population density was included to account for the possibility of proximal third parties, such as neighbors, calling the police to intervene (Libertun de Duren, 2020).

Analytic Strategy

Characteristics and distributions of the data warranted considering count regression models (Hilbe, 2011; Osgood, 2000). The dependent variable of interest, IPV incident reports, was positively skewed in Census tracts, possessed a non-normal distribution, and had a substantial zero count in tracts. Model fit characteristics and information criterion values were compared among four types of models, and the data most appropriately fit negative binomial regression models. Actual versus apparent dispersion was checked to confirm overdispersion (Hilbe, 2011). Cook’s D values suggested that no influential observations were present, and no multicollinearity issues were detected (VIF < 4).

Due to the nature of the data and unit of analysis, spatial autocorrelation was checked for both incidents and aggregated incidents to tracts. This is a necessary and underutilized step for geographic units of analysis because of how crime concentrates spatially due to many factors, such as various kinds of facilities (Cozens et al., 2019, pp. 8–9). To assess whether this was an issue for the present study, Average Nearest Neighbor (NNR = .31, Z = −87.97, p < .01) and Global Moran’s I (I = .35, Z = 12.23, p < .01) were used to determine whether crime was concentrated (Mletzko et al., 2018). Both statistics indicated that spatial clustering occurred, warranting the use of a spatial lag variable. Spatial lag variables consider the unmeasured factors that influence spatial dependence (Kubrin & Weitzer, 2003); using a spatial lag variable in the model considers the possibility that crime may be influenced by unmeasured independent variables, but also by crime itself. The spatial lag variable was created in R using Anselin and Bera’s (1998) equation and then added to the negative binomial regression model for this study (Mletzko et al., 2018).

Results

The purpose of this study was to assess whether social disorganization variables are related to the geographic distribution of counts of annual IPV incident reports per neighborhood. It was hypothesized that three social disorganization variables—concentrated disadvantage, racial/ethnic heterogeneity, and residential instability—would be positively related to annual counts of IPV incident reports per neighborhood. Results from negative binomial regression analysis partially supported this hypothesis.

Descriptive statistics for the model are shown in Table 1. The mean annual count of IPV incident reports per neighborhood was 20.83, with a median of 15 and a range of 111, indicating that the variable is positively skewed, which is common with crime data (e.g., Sherman et al., 1989). More than 10% (n = 26) of the neighborhoods in the sample did not have any IPV incident reports in 2018. Among tracts with the highest IPV counts, 10% of tracts contained approximately 20% of all reported IPV incidents in the city.

Table 1.

Descriptive Statistics for IPV and Neighborhoods (n = 216).

Variable	Min. (Max.)	Mean (Median)	SD	Range	Variance
IPV incident reports	0 (111)	20.83 (15)	20.94	111	438.37
Crime reports	0 (6325)	454.61 (344)	544.04	6325	295,976
Concentrated disadvantage^a	−.99 (4.22)	.00 (−.44)	.99	5.21	.97
Racial/ethnic heterogeneity^a	−1.38 (3.06)	.00 (−.28)	1.00	4.43	1.00
Residential instability^a	−1.68 (2.39)	.00 (−.10)	.94	4.06	.88
Population density	80.282 (26,850)	4601.05 (4023.88)	3843.52	26,769.72	14,772,623.57
Males per 100 females	67.60 (183.80)	103.93 (101.15)	16.63	116.20	276.47
Median age	19.80 (54.20)	35.14 (34.20)	5.67	34.40	32.10

^aZero represents the mean level of the variable, not its absence.

Descriptive geographic results are provided in Figures 1 and 2. Figure 1 visualizes the 4500 IPV incident reports across the 216 Census tracts in this study. Incidents are unevenly geographically distributed; not only are IPV incident reports more concentrated in certain neighborhoods, they are concentrated in certain parts of certain neighborhoods (Weisburd, 2015). This was also suggested by the significant Average Nearest Neighbor and the Global Moran’s I values. Figure 2 provides a graduated symbol map of IPV incident reports superimposed over a choropleth map of concentrated disadvantage in Austin, Texas. Concentrated disadvantage is symbolized with the standard deviation of scores in tracts. Racial/ethnic heterogeneity and residential instability are similar to concentrated disadvantage in that they are generally more prevalent on the east side of the study city.

Figure 1.

Map of IPV incident reports and Census tracts in Austin, TX.

Figure 2.

Graduated symbols map of IPV incidents and choropleth map of concentrated disadvantage in Austin, TX.

Table 2 provides the multivariate negative binomial regression results. The specified model was statistically significant overall (LR χ² [8] = 166.40, p < .01). It also partially supports this study’s hypothesis and is consistent with Shaw and McKay’s (1942) original contention linking neighborhood social disorganization and crime. Concentrated disadvantage (z = 2.34, p < .05) was statistically significant in the expected positive direction when controlling for all other variables in the model. Racial/ethnic heterogeneity (z = .72, p > .05) and residential instability (z = .88, p > .05) were in the positive direction but not statistically significant. Crime reports were positive and significantly (z = 6.44, p < .01) associated with IPV. However, the other control variables, population density, males per 100 females, median age, and spatial lag, were all non-significant (p > .05) after considering all variables in the model.² Using the rate ratio, for every one-unit increase in concentrated disadvantage, there is a 22% increase in IPV incident reports in neighborhoods.

Table 2.

Negative Binomial Model of Interest Variables on Intimate Partner Violence (IPV).

	IRR	Std. Error	Z
Crime reports*	1.00	.00	6.44
Concentrated disadvantage*	1.22	.10	2.34
Racial/Ethnic heterogeneity	1.05	.08	0.72
Residential instability	1.08	.10	0.88
Population density	1.00	.00	−0.15
Males per 100 females	1.00	.00	−0.80
Median age	0.97	.02	−1.55
Spatial lag	1.00	.00	1.46
Constant*			4.24
Likelihood ratio χ² (7)*			166.40
Pseudo R²			0.10

(*) p < .01; IRR = incidence rate ratio.

Discussion

This study examined the effects of three social disorganization variables, concentrated disadvantage, racial/ethnic heterogeneity, and residential instability, on annual counts of IPV incident reports per neighborhood. These variables were also analyzed in models with empirically relevant control variables, a crime variable that included all non-IPV crime, and a spatial lag variable. The results of this study partially supported the hypotheses that all three social disorganization variables would be significant and positively associated with IPV incident report counts. Although findings diverge from Shaw and McKay’s original propositions, they are consistent with more recent studies finding only partial support for the theory. These results suggest that although social disorganization theory was originally proposed to explain public crime promoted by group behavior, it also sheds light on how social factors impact IPV, which happens “behind closed doors,” and how those impacts manifest at the community level.

The significance of the crime reports control variable on annual neighborhood IPV incident report counts was notable, particularly given that it had the largest effect size out of all independent variables in the model. This finding suggests that the spatial distribution of IPV is not so different from the geography of crime in general. It also calls into question the contention that IPV has an etiology distinct from that of other types of crime, at least at the aggregate level. In other words, where one can expect to find non-IPV crime one can also expect to find IPV, after considering relevant control variables. Perhaps this relationship is due to normativity of law enforcement presence in higher-crime neighborhoods and subsequently reduced stigmatization of calling police to resolve disputes. The significant and positive relationship between total crime reports and IPV incident reports may also be driven by the same influences that underlie the significant and positive relationship between concentrated disadvantage and IPV incident reports found in this study. The police have been found more likely to file reports in low-socioeconomic neighborhoods, and less likely to file reports when victims are female (Smith, 1986). However, police discretion research has several limitations as well (see e.g., Nickels, 2007; Riksheim & Chermak, 1993). As with concentrated disadvantage, areas that have higher crime might also lack informal intervention resources, rendering police response the most viable resource available to intervene in crime (Avakame et al., 1999; Edwards et al., 2014; Goodson & Bouffard, 2017).

As in many prior social disorganization-IPV studies, concentrated disadvantage had a positive and significant effect on the dependent variable. In neighborhoods with high levels of concentrated disadvantage, law enforcement is often the most visible and/or accessible intervention resource available (Avakame et al., 1999). “Urban and poor” residents are more likely to call the police for IPV than those who are more educated (Avakame et al., 1999, p. 782). Avakame et al. attributed this finding to more affluent individuals having greater access to non-police interventions, including family and friends, while those who are more disadvantaged must rely on law enforcement for help (Avakame et al., 1999; Flicker et al., 2011).

Socioeconomically driven reporting bias of victims and/or third-party complainants could also be influencing this outcome, given that official police data were used in the present study. More affluent victims might not report IPV in order to preserve their social standing or economic status (Miles-Doan, 1998). Cattaneo and DeLoveh (2008) found greater incident severity was associated with increased likelihood of calling for police intervention among higher-income victims but had no effect for victims at lower income levels. They proposed that less affluent victims might have a lower threshold for summoning law enforcement.

Police bias along with neighborhood racial composition and socioeconomic profile may also impact these results (Gelman et al., 2007; Smith, 1986). For example, Smith (1986) found that police officers are more likely to file crime reports in socioeconomically distressed neighborhoods. Similarly, Lee et al. (2013) found higher levels of concentrated disadvantage and immigrant concentration were positively correlated with IPV arrest, which they attributed to conflict theory (Black, 1976).

Concentrated disadvantage also appears to have the greatest effect when less severe IPV offenses are included, as they are in this study (Blumenstein & Jasinski, 2015; Pinchevsky & Wright, 2012). Despite this possible reporting bias, we believe including the full range of offense types is important to accurately reflect the complexity of IPV. In addition, the statistical significance of the concentrated disadvantage variable here, in light of prior research, suggests further study is warranted to see if this significance holds when lesser offense types are excluded.

Neighborhoods with high levels of concentrated disadvantage not only have higher rates of poverty, mental and physical illness, and substance abuse, they also have fewer resources to alleviate them compared to more affluent areas (Edwards et al., 2014; Goodson & Bouffard, 2017). Residents of disadvantaged areas have weaker social networks and experience more isolation, compounding health strains, and difficulties caused by lack of access to resources, all of which are demonstrated risk factors for IPV (Miles-Doan, 1998; Sampson & Groves, 1989; Sampson et al., 1997). Individuals living in these areas may be so focused on merely surviving that they cannot unite to foster a social environment visibly disapproving of IPV. As a result, perpetrators of IPV in these neighborhoods may go unsanctioned by their community (Edwards et al., 2014; Pinchevsky & Wright, 2012; Van Wyk et al., 2003). Regardless of whether differential reporting practices of victims and/or third-party complainants between levels of disadvantage, disparate access to non-police resources, or disproportionate stressors underlie the relationship, these findings suggest that greater concentrated disadvantage equals diminished ability to exert informal social control to intervene in and prevent IPV in communities.

Given IPV’s pervasiveness yet uneven geographic distribution, it must be addressed on a systemic level while simultaneously acknowledging its disproportionate effect on certain populations. To do so is not to stigmatize these populations, but to recognize that they often contend with complex factors that make combatting IPV more difficult.

This research indicates that further analysis of the geography of IPV is merited. Future studies could investigate geographical disparity between IPV incident reports and arrests, the distribution of IPV offense harm, geographical distribution of IPV at different units of analysis, and change in that distribution over time.

Strengths and Limitations

To our knowledge, this is the first social disorganization-IPV study taking place in a city without an ethnic majority, as well as the first to be conducted in Austin. In 2018, Austin’s population was 48.8% non-Hispanic White, 30.2% Hispanic, 8.1% Black or African American, and 7.6% Asian (Data USA, 2021). Nearly a third (32%) of Austinites speak a language other than English in the home (Austin City, 2021; U.S. Census Bureau, 2019). Considering our results, this demographic diversity provides additional support for the generalizability of a link between concentrated disadvantage and increased incidence of IPV.

This study also expands the extant social disorganization-IPV literature in three important ways. First, it considers all types of IPV incidents instead of limiting its scope to only a few types. Second, it contributes by including total crime reports as a control variable. Doing so provides valuable insight into the spatial patterning of IPV incident reports versus all crime incident reports and calls into question the contention that IPV’s etiology is distinct from that of other types of offenses. Finally, we included all victim-offender gender pairings in our outcome variable instead of limiting our dataset to incident reports solely concerning male perpetration of IPV toward female victims. This decision acknowledges that male-on-female IPV is not the only way in which this type of violence is perpetrated. Future research in this area might employ a similar model to analyze disaggregated gender pairings of victim and offender, allowing for consideration of differential effects of social disorganization variables on spatial patterning of IPV incident reports for various victim-offender gender pairings.

These results only reflect the prevalence and distribution of reported IPV incidents that also resulted in responding officers filing a report. It has been estimated that as few as 25% of IPV incidents are brought to the attention of the police (Felson & Paré, 2005); therefore, this research likely underrepresents the true number of IPV incidents, as with any research based on official data. Minor IPV incidents are probably disproportionately underrepresented in the data, particularly in more affluent neighborhoods, where evidence suggests victims are less likely to call police in response to IPV they do not perceive as severe (Avakame et al., 1999; Cattaneo & DeLoveh, 2008; Miles-Doan, 1998). Future research could address that issue by including models that distinguish between misdemeanor and felony IPV offenses or by using self-report data that includes information about more minor IPV incidents.

The statistical significance of concentrated disadvantage may be driven by police officer bias against racial and ethnic minorities, leading to a greater likelihood of filing reports for minor IPV incidents in disadvantaged neighborhoods compared to affluent ones. Prior research has shown Black and Latino races/ethnicities to be overrepresented in arrests resulting from “stop-and-frisks” compared to whites (Gelman et al., 2007). The present study’s outcome may be affected by a similar phenomenon if people from racial/ethnic minorities are overrepresented in the incident report data from areas with high concentrated disadvantage.

Another limitation is that as a partial test of social disorganization theory that centered the independent variables of interest in Shaw and McKay’s original model, this analysis did not consider other possible causal factors related to the spatial pattern of IPV incident reports. Most notably excluded are those related to social control, such as community cohesion and friendship networks, which could also be addressed by future research.

Conclusion

While Shaw and McKay’s work concerned street crime and delinquency, their key propositions of social disorganization theory are partially supported by these present findings. This study adds to the body of literature on aggregate-level effects of IPV. It underscores the need to address IPV as a community issue and demonstrates that IPV is a burden disproportionately shouldered by disadvantaged populations. Non-IPV crime is a significant predictor for the location of IPV crime, at least within Census tracts. Further, the figures in this study illustrate that even within neighborhoods, IPV incident reports cluster in certain areas, a finding which merits additional inquiry.

Footnotes

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 iDs

Molly McDowell

Daniel Reinhard

Notes

Author Biographies

Molly McDowell is a Ph.D. candidate at Texas State University. Her primary areas of scholarly interest are structural inequality and crime, context of place and incidence of crime, criminological theory, gender violence, and quantitative methods.

Daniel Reinhard is a recently graduated PhD working as the Chief Data Analyst for the Boulder Colorado Police Department. Some of his recent publications have involved homeless income generation, crime prevention, and the use of spatial and temporal methods to understand various public safety topics.

References

Anselin

Bera

A. K.

(1998). Spatial dependence in linear regression models with an introduction to spatial econometrics. In Ullah

(Ed.), Handbook of applied economic statistics (Chapter 7). CRC Press.

Austin City . (2021). Top ten demographic trends in Austin. https://www.austintexas.gov/page/top-ten-demographic-trends-austin-texas

Austin Police Department . (2017). Austin police department policy manual. https://www.austintexas.gov/sites/default/files/files/Current_APD_Policy_Manual_2017-1.5_issued_7-20-2017.pdf

Austin Police Department . (2020). Crime reports [Data file]. https://doi.org/10.26000/043.000005

Avakame

E. F.

Fyfe

J. J.

McCoy

(1999). “Did you call the police? What did they do?” An empirical assessment of Black's theory of mobilization of law. Justice Quarterly, 16(4), 765–792. http://doi.org/10.1080/07418829900094361

Benson

M. L.

Fox

G. L.

DeMaris

van Wyk

(2003). Neighborhood disadvantage, individual economic distress and violence against women in intimate relationships. Journal of Quantitative Criminology, 19(3), 207–235. https://doi.org/10.1023/A:1024930208331

Benson

M. L.

Wooldredge

Thistlethwaite

A. B.

Fox

G. L.

(2004). The correlation between race and domestic violence is confounded with community context. Social Problems, 51(3), 326–342. http://doi.org/10.1525/sp.2004.51.3.326

Beyer

Wallis

A. B.

Hamberger

L. K.

(2015). Neighborhood environment and intimate partner violence: A systematic review. Trauma, Violence, & Abuse, 16(1), 16–47. http://doi.org/10.1177/1524838013515758

Black

D.J.

(1976). The behavior of law, Academic Press.

10.

Blau

P. M.

(1977). Inequality and heterogeneity: A primitive theory of social structure (Vol. 7). Free Press.

11.

Blumenstein

Jasinski

J. L.

(2015). Intimate partner assault and structural-level correlates of crime: Exploring the relationship between contextual factors and intimate partner violence. Criminal Justice Studies, 28(2), 186–210. http://doi.org/10.1080/1478601x.2014.1000488

12.

Boggess

L. N.

Chamberlain

A. W.

(2021). Investing in neighborhoods: The mitigating role of home mortgage loans on intimate partner violence. Journal of Interpersonal Violence, 36(23–24), 1–24. http://doi.org/10.1177/0886260519900312

13.

Breiding

M.J.

Basile

K.C.

Smith

S.G.

Black

M.C.

Mahendra

R.R.

(2015). Intimate partner violence surveillance: Uniform definitions and recommended data elements, version 2.0. National Center for Injury Prevention and Control, Centers for Disease Control and Prevention.

14.

Browning

C. R.

(2002). The span of collective efficacy: Extending social disorganization theory to partner violence. Journal of Marriage and Family, 64(4), 833–850. http://doi.org/10.1111/j.1741-3737.2002.00833.x

15.

Brownridge

D. A.

Taillieu

T. L.

Tyler

K. A.

Tiwari

Chan

K. L.

Santos

S. C.

(2011). Pregnancy and intimate partner violence: Risk factors, severity, and health effects. Violence Against Women, 17(7), 858–881. http://doi.org/10.1177/1077801211412547

16.

Bursik

R. J.

(1988). Social disorganization and theories of crime and delinquency: Problems and prospects. Criminology, 26(4), 519–552. http://doi.org/10.1111/j.1745-9125.1988.tb00854.x

17.

Caetano

Ramisetty-Mikler

Harris

T. R.

(2010). Neighborhood characteristics as predictors of male to female and female to male partner violence. Journal of Interpersonal Violence, 25(11), 1986–2009. http://doi.org/10.1177/0886260509354497

18.

Capaldi

D. M.

Knoble

N. B.

Shortt

J. W.

Kim

H. K.

(2012). A systematic review of risk factors for intimate partner violence. Partner Abuse, 3(2), 231–280. http://doi.org/10.1037/e621642012-252

19.

Cattaneo

L. B.

DeLoveh

H. L.

(2008). The role of socioeconomic status in helpseeking from hotlines, shelters, and police among a national sample of women experiencing intimate partner violence. American Journal of Orthopsychiatry, 78(4), 413–422. http://doi.org/10.1037/a0014558

20.

Chang

L. Y.

Foshee

V. A.

Reyes

H. L. M.

Ennett

S. T.

Halpern

C. T.

(2015). Direct and indirect effects of neighborhood characteristics on the perpetration of dating violence across adolescence. Journal of Youth and Adolescence, 44(3), 727–744. http://doi.org/10.1007/s10964-014-0190-z

21.

Cozens

Love

Davern

(2019). Geographical juxtaposition: A new direction in CPTED. Social Sciences, 8(9), 1–22. http://doi.org/10.3390/socsci8090252

22.

Data USA . (2021). Census place. https://datausa.io/profile/geo/austin-tx/

23.

Edwards

K. M.

Mattingly

M. J.

Dixon

K. J.

Banyard

V. L.

(2014). Community matters: Intimate partner violence among rural young adults. American Journal of Community Psychology, 53(1–2), 198–207. http://doi.org/10.1007/s10464-014-9633-7

24.

Farley

(1996). The new American reality: Who we are, how we got here, where we are going. Russell Sage Foundation.

25.

Felson

R. B.

Paré

P. P.

(2005). The reporting of domestic violence and sexual assault by nonstrangers to the police. Journal of Marriage and Family, 67(3), 597–610. http://doi.org/10.1037/e535952006-001

26.

Flicker

S. M.

Cerulli

Zhao

Tang

Watts

Xia

Talbot

N. L.

(2011). Concomitant forms of abuse and help-seeking behavior among White, African American, and Latina women who experience intimate partner violence. Violence Against Women, 17(8), 1067–1085. http://doi.org/10.1177/1077801211414846

27.

Fox

G. L.

Benson

M. L.

(2006). Household and neighborhood contexts of intimate partner violence. Public Health Reports, 121(4), 419–427. http://doi.org/10.1177/003335490612100410

28.

Frye

Galea

Tracy

Bucciarelli

Putman

Wilt

(2008). The role of neighborhood environment and risk of intimate partner femicide in a large urban area. American Journal of Public Health, 98(8), 1473–1479. http://doi.org/10.2105/ajph.2007.112813

29.

Frye

Wilt

(2001). Femicide and social disorganization. Violence Against Women, 7(3), 335–351. http://doi.org/10.1177/10778010122182479

30.

Gelman

Fagan

Kiss

(2007). An analysis of the New York City police department’s “stop-and-frisk” policy in the context of claims of racial bias. Journal of the American Statistical Association, 102(479), 813–823. http://doi.org/10.1198/016214506000001040

31.

Goodson

Bouffard

L. A.

(2017). The rural/urban divide: Examining different types of assault through a social disorganization lens. Journal of Interpersonal Violence, 35(17–18), 3530–3553. http://doi.org/10.1177/0886260517711179

32.

Goodson

Bouffard

L. A.

(2019). Social disorganization and gender equality as correlates of family violence and rape. Journal of Crime and Justice, 42(3), 274–287. http://doi.org/10.1080/0735648x.2018.1535995

33.

Gracia

López-Quílez

Marco

Lladoza

Lila

(2015). The spatial epidemiology of intimate partner violence: Do neighborhoods matter? American Journal of Epidemiology, 182(1), 58–66. http://doi.org/10.1093/aje/kwv016

34.

Han

Y. R.

Choi

H. Y.

(2021). Risk factors affecting intimate partner violence occurrence in South Korea: Findings from the 2016 Domestic Violence Survey. PLoS ONE, 16(3), 1–16. http://doi.org/10.1371/journal.pone.0247916

35.

Hilbe

J. M.

(2011). Negative binomial regression. Cambridge University Press.

36.

Kimber

M. S.

Boyle

M. H.

Lipman

E. L.

Colwell

S. R.

Georgiades

Preston

(2013). The associations between sex, immigrant status, immigrant concentration and intimate partner violence: Evidence from the Canadian General Social Survey. Global Public Health, 8(7), 796–821. http://doi.org/10.1080/17441692.2013.814701

37.

Kornhauser

R. R.

(1978). Social sources of delinquency: An appraisal of analytic models. University of Chicago Press.

38.

Kubrin

C. E.

Weitzer

(2003). New directions in social disorganization theory. Journal of Research in Crime and Delinquency, 40(4), 374–402. http://doi.org/10.1177/0022427803256238

39.

Lanier

Maume

M. O.

(2009). Intimate partner violence and social isolation across the rural/urban divide. Violence Against Women, 15(11), 1311–1330. http://doi.org/10.1177/1077801209346711

40.

Lee

Zhang

Hoover

L.T.

(2013). Police response to domestic violence: Multilevel factors of arrest decision. Policing: An International Journal of Police Strategies & Management, 36(1). 157–174. http://doi.org/10.1108/13639511311302524

41.

Kirby

R. S.

Sigler

R. T.

Hwang

S. S.

LaGory

M. E.

Goldenberg

R. L.

(2010). A multilevel analysis of individual, household, and neighborhood correlates of intimate partner violence among low-income pregnant women in Jefferson County, Alabama. American Journal of Public Health, 100(3), 531–539. http://doi.org/10.2105/ajph.2008.151159

42.

Libertun de Duren

N. R.

(2020). Effects of neighborhood upgrading programs on domestic violence in Bolivia. World Development Perspectives, 19(September), 1-8. https://doi.org/10.1016/j.wdp.2020.100231.

43.

Long

J. S.

(1997). Regression models for categorical and limited dependent variables. SAGE Publications.

44.

Miles-Doan

(1998). Violence between spouses and intimates: Does neighborhood context matter? Social Forces, 77(2), 623–645. http://doi.org/10.2307/3005541

45.

Miles-Doan

Kelly

(1997). Geographic concentration of violence between intimate partners. Public Health Reports, 112(2), 135-141. 9071275.

46.

Mletzko

Summers

Arnio

A. N.

(2018). Spatial patterns of urban sex trafficking. Journal of Criminal Justice, 58(3), 87–96. http://doi.org/10.1016/j.jcrimjus.2018.07.008

47.

Nickels

E. L.

(2007). A note on the status of discretion in police research. Journal of Criminal Justice, 35(5), 570–578. http://doi.org/10.1016/j.jcrimjus.2007.07.009

48.

O'Campo

Gielen

A. C.

Faden

R. R.

Xue

Kass

Wang

M. C.

(1995). Violence by male partners against women during the childbearing year: A contextual analysis. American Journal of Public Health, 85(8), 1092–1097. http://doi.org/10.2105/ajph.85.8_pt_1.1092

49.

Osgood

D. W.

(2000). Poisson-based regression analysis of aggregate crime rates. Journal of Quantitative Criminology, 16(1), 21-43. https://doi.org/10.1023/A:1007521427059.

50.

Pearlman

D. N.

Zierler

Gjelsvik

Verhoek-Oftedahl

(2003). Neighborhood environment, racial position, and risk of police-reported domestic violence: A contextual analysis. Public Health Reports, 118(1), 44–58. http://doi.org/10.1016/s0033-3549(04)50216-9

51.

Pinchevsky

G. M.

Wright

E. M.

(2012). The impact of neighborhoods on intimate partner violence and victimization. Trauma, Violence, & Abuse, 13(2), 112–132. http://doi.org/10.1177/1524838012445641

52.

Pratt

T. C.

Cullen

F. T.

(2005). Assessing macro-level predictors and theories of crime: A meta-analysis. Crime and Justice, 32(1), 373-450. http://doi.org/10.1086/655357.

53.

Reingle

J. M.

Staras

S. A.

Jennings

W. G.

Branchini

Maldonado-Molina

M. M.

(2012). The relationship between marijuana use and intimate partner violence in a nationally representative, longitudinal sample. Journal of Interpersonal Violence, 27(8), 1562–1578. http://doi.org/10.1177/0886260511425787

54.

Rennison

C. M.

Rand

M. R.

(2003) Nonlethal intimate partner violence against women: A comparison of three age cohorts. Violence Against Women, 9(12), 1417–1428. http://doi.org/10.1177/1077801203259232

55.

Rennison

C. M.

Welchans

(2000). Intimate partner violence: Bureau of Justice Statistics special report. US Department of Justice.

56.

Riksheim

E. C.

Chermak

S. M.

(1993). Causes of police behavior revisited. Journal of Criminal Justice, 21(4), 353–382. http://doi.org/10.1016/0047-2352(93)90019-j

57.

Sampson

R. J.

Groves

W. B.

(1989). Community structure and crime: Testing social disorganization theory. American Journal of Sociology, 94(4), 774-802. https://doi.org/10.1086/229068.

58.

Sampson

R. J.

Raudenbush

S. W.

Earls

(1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science, 277(5328), 918–924. http://doi.org/10.1126/science.277.5328.918

59.

Shaw

C. R.

McKay

(1942). Juvenile delinquency and urban areas. University of Chicago Press.

60.

Sherman

L. W.

Gartin

P. R.

Buerger

M. E.

(1989). Hot spots of predatory crime: Routine activities and the criminology of place. Criminology, 27(1), 27–56. http://doi.org/10.1111/j.1745-9125.1989.tb00862.x

61.

Smith

D. A.

(1986). The neighborhood context of police behavior. Crime and Justice, 8(1), 313-341. http://doi.org/10.1086/449126.

62.

Smith

S.G.

Zhang

Basile

K.C.

Merrick

M.T.

Wang

Kresnow

Chen

(2018). The National intimate partner and sexual violence Survey (NISVS): 2015 Data Brief – Updated Release. National Center for Injury Prevention and Control, Centers for Disease Control and Prevention.

63.

United States Census Bureau . (2019). When to Use 1-year, 3-year, or 5-year Estimates. https://www.census.gov/programs-surveys/acs/guidance/estimates.html

64.

Unites States Census Bureau (2019). American community Survey 2014-2018 5-year estimates [Data files and code books]. https://data.census.gov/cedsci/

65.

Van Wyk

Benson

M. L.

Fox

G. L.

DeMaris

(2003). Detangling individual-, partner-, and community-level correlates of partner violence. Crime & Delinquency, 49(3), 412–438. http://doi.org/10.1177/0011128703049003004

66.

VanderEnde

K. E.

Yount

K. M.

Dynes

M. M.

Sibley

L. M.

(2012). Community-level correlates of intimate partner violence against women globally: A systematic review. Social Science & Medicine, 75(7), 1143-1155. http://doi.org/10.1016/j.socscimed.2012.05.027.

67.

Waller

M. W.

Iritani

B. J.

Christ

S. L.

Clark

H. K.

Moracco

K. E.

Halpern

C. T.

Flewelling

R. L.

(2012). Relationships among alcohol outlet density, alcohol use, and intimate partner violence victimization among young women in the United States. Journal of Interpersonal Violence, 27(10), 2062–2086. http://doi.org/10.1177/0886260511431435

68.

Weisburd

(2015). The law of crime concentration and the criminology of place. Criminology, 53(2), 133–157. http://doi.org/10.1111/1745-9125.12070

69.

Wilson

W. J.

(1987). The truly disadvantaged. University of Chicago Press.

70.

Wood

Baumler

Rinehart

J. K.

Temple

J. R.

(2021). Risk for dating violence and sexual assault over time: The role of college and prior experiences with violence. Journal of American College Health, 1-7. http://doi.org/10.1080/07448481.2021.1910273.

71.

Wooldredge

Thistlethwaite

(2003). Neighborhood structure and race-specific rates of intimate assault. Criminology, 41(2), 393–422. http://doi.org/10.1111/j.1745-9125.2003.tb00992.x

72.

Wright

E. M.

Benson

M. L.

(2010). Immigration and intimate partner violence: Exploring the immigrant paradox. Social Problems, 57(3), 480–503. http://doi.org/10.1525/sp.2010.57.3.480