Neighborhood Ecological Models of Alcohol Outlet Density and Male–on–Female Domestic Violence: Accounting for Adjacent Place and Neighborhood Characteristics

Abstract

Alcohol availability is a consistent predictor of domestic violence, including intimate partner violence. Less is known about the effects of alcohol availability in neighboring units of analysis on domestic violence. This study examined whether alcohol outlet density in Milwaukee, Wisconsin is associated with Male-on-Female domestic violence (MFDV). Using block groups as the unit of analysis (N = 571), spatial regression models were estimated to model the relationship between the density of total, on-premise, and off-premise alcohol outlets on MFDV, while accounting for the spatial spillover effect (i.e., alcohol availability and neighborhood characteristics in focal and surrounding block groups). At the focal level, off-premise alcohol outlets are associated with MFDV, net of concentrated disadvantage, lack of health insurance, MFDV lag and total population. Additionally, off-premise alcohol outlet density in surrounding units of analysis is a significant predictor of MFDV. However, total and on-premise densities did not demonstrate a relationship with MFDV.

Keywords

alcohol availability concentrated disadvantage domestic violence off-premise alcohol outlets spatial regression models

Introduction

In the United States, domestic violence presents a significant, widespread social and public health problem. Domestic violence is defined as violence against a current or former spouse, against an adult with whom the offender resides or has formerly resided, or against an adult with whom the offender has a child (Wisconsin State Legislature, n.d.). As such, it is often manifested as intimate partner violence (IPV), which includes physical and sexual violence, stalking, and psychological aggression by a current or former intimate partner (Breiding, Basile, Smith, Black, & Mahendra, 2015). In the United States alone, roughly 35% of women experience rape, physical violence, or stalking by an intimate partner in their lifetime (Kearn, Reidy, & Valle, 2015). Psychological abuse within an intimate relationship was found to be even more prevalent than sexual or physical violence, with about one-half of both men and women reporting lifetime experiences with this form of violence (Murray, Bunch, & Hunt, 2016).

While men and women perpetrate IPV at similar rates, violence against women is more severe (Cunradi, 2007). In particular, rates of intimate partner homicide starkly diverge according to the gender of the perpetrator. Previous estimates of worldwide intimate partner homicide indicated that while 6% of murdered men were killed by an intimate partner, up to 38% of all murdered women were killed by an intimate partner (WHO, 2013). Recent estimates demonstrate a slight increase, placing the prevalence of women murdered by intimate partners in the range of 38%–40% (WHO, 2018). Male-to-female partner violence (MFPV) is more likely to be repeated and incur serious injuries and death than violence perpetrated by women (Caetano, Nelson, & Cunradi, 2001; Laskey, Bates, & Taylor, 2019). Women who experience IPV are more likely to suffer from a variety of negative health outcomes, such as an increased risk of physical injuries and adaptation of unhealthy coping behaviors (Voith & Brondino, 2017). Compared to women who have not been exposed to IPV, women who experience IPV are more likely to report chronic health conditions such as asthma, irritable bowel syndrome, diabetes, frequent headaches, difficulty sleeping, and chronic pain (Waller et al., 2013).

As the literature review that follows will describe, a consistent and statistically significant predictor of domestic violence is alcohol and its various forms, such as alcohol consumption and alcohol availability (i.e., density). Alcohol consumption is associated with domestic violence at individual and population levels, alcohol outlet density is linked to domestic violence levels in communities, and the effects of alcohol availability matter not only for the local rate of domestic violence but also the rate of domestic violence in adjacent areas. However, we know less about whether greater availability of alcohol outlets in nearby areas, along with nearby levels of concentrated disadvantage in combination with high levels of alcohol availability in the local areas, produces increased levels of domestic violence. Therefore, in this study, we advance the literature by estimating spatially informed regression models to fill the gap and produce new insights about the role of alcohol availability in nearby areas to produce domestic violence in local areas.

Literature Review

Alcohol Consumption and Domestic Violence

A substantial body of prior research indicates alcohol consumption is related to domestic violence and IPV, as one manifestation of domestic violence, at the individual level. Reporting of IPV by both genders was associated with partner’s alcohol consumption and with problematic alcohol use (Cafferky, Mendez, Anderson, & Stith, 2018) and, independent of levels of female substance use, alcohol consumption by males was significantly associated with several forms of IPV (Coker et al., 2002). Also, individuals in committed relationships who engage in binge drinking four or more times per month reported higher rate of alcohol-related problems, including alcohol dependence and social problems, such as problems with police, relationship problems, arguments or fights, and financial problems (McKinney, Chartier, Caetano, & Harris, 2012). Furthermore, measures of heavy drinking have repeatedly been associated with greater odds of IPV. Cunradi (2007) found that men who are past 30-day heavy drinkers were at a more than 6-fold increased risk for mutual IPV, while women in the same drinking category were at a roughly 6-fold risk for mutual IPV. Men in less hazardous drinking categories were at a two to three-fold increased risk of mutual IPV (Cunradi, 2007). More recent studies demonstrate that men who consume alcohol with greater frequency and in greater amounts perpetrate more psychological and physical IPV (Renzetti, Lynch, & DeWall 2018). These patterns are echoed by other prior studies that disaggregated victimization data based on the sex of the perpetrator of violence. Male past-year frequency of drinking and alcohol volume were linked to both Male-to-Female partner violence (MFPV) and Female-to-Male partner violence (FMPV) (Caetano, Ramisetty-Mikler, & Harris, 2010), suggesting that when men consume alcohol frequently and in the context of heavy drinking episodes, MFPV and FMPV is more likely to occur relative to when women drink in similar patterns. Rates of MFPV were much higher among men reporting drinking five or more drinks per occasion at least once per week and two to four times higher among men with alcohol problems (Caetano, Schafer, & Cunradi, 2017).

Alcohol Outlet Density and Domestic Violence

The studies reviewed above describe the connection between alcohol and violence at the individual level. At the community-level, alcohol availability is associated with general violence rates across communities, with between 25% and 50% of domestic violence incidents involving alcohol (Livingston, 2011). Alcohol outlet density has demonstrated a positive, nonlinear association with binge drinking (Ahern, Margerison-Zilko, Hubbard, & Galea, 2013), further complicating the association between alcohol consumption, availability, and violence. Furthermore, the characteristics of places that operate in communities have been linked with violence by the likelihood of violence occurring in an area due to ecological features (Brantingham & Brantingham, 1999). For example, characteristics of a place such as an alcohol outlet can concentrate activity in areas where social control is weakened and hostile interactions are more likely, leading to increased rates of violence in communities (Brantingham & Brantingham, 1999; Gruenewald, Freisthler, Remer, LaScala, & Treno, 2006). At the same time, assaultive violence has been shown to systematically cluster around groupings of alcohol outlets (Grubesic & Pridemore, 2011).

Moreover, some alcohol outlets attract clientele (e.g., young men) that is more likely to be involved in violent acts, experience an increase in aggression associated with alcohol intoxication (Haines & Graham, 2005), and exacerbate violence that is already occurring in communities, increasing both its frequency and severity (Yates, 2019). In addition, total alcohol outlet density is linked to IPV (Cunradi, 2010; Waller et al., 2013) but individuals residing in neighborhoods with a high density of alcohol outlets have significantly increased odds of perpetrating IPV relative to individuals in neighborhoods with low alcohol outlet density (Waller et al., 2013). This can create a cascading effect, wherein neighborhoods already experiencing high levels of violence and greater densities of alcohol outlets are more likely to worsen with the addition of new alcohol outlets, unlike neighborhoods with a sparse distribution of alcohol outlets (Wei, Grubesic, & Kang, 2021). Greater alcohol outlet density within a neighborhood may also indirectly impact the likelihood of domestic violence in general, and IPV in particular, in three ways: by indicating loosened normative constraints against violence, by promoting problematic alcohol consumption for couples at risk of domestic violence, and by providing environments where IPV-related attitudes, norms, and problem behaviors are reinforced (Cunradi, Mair, Ponicki, & Remer, 2011). The presence of alcohol outlets, often surrounded by signs of physical disorder, signals to residents that informal social control is lacking in the community, which in turn may lead residents to not get involved when witnessing IPV or to be less concerned about the social consequences of committing IPV. Greater alcohol outlet density within a neighborhood may provide more opportunities for drinking to individuals whose barriers to aggression are lowered while consuming alcohol or influence couples’ patterns of venue use to include venues associated with heavy drinking. Drinking in bars with physical or social characteristics that make violence more likely or in the company of other intoxicated men in a public setting increases the chance that men will return in a disinhibited, aggressive state which can quickly escalate to IPV (Cunradi, 2010). In addition, community rates of binge drinking have been positively associated with family violence (Stansfield & Doherty, 2019), suggesting that alcohol drinking patterns also play a role in the rate of domestic violence in communities. The findings of recent studies (Lira et al., 2019; Naimi et al., 2017) examined the effect of alcohol policies on intimate partner homicide and suggest that stricter policies, such as reducing alcohol outlet density, are protective for alcohol-related intimate partner homicides. In 2011, Wisconsin had the least restrictive alcohol policies of any U.S. state (Lira et al., 2019), emphasizing the need for further research on the relationship between alcohol outlet density and domestic violence.

Alcohol Outlet Types and Domestic Violence

Although alcohol outlet density has been linked to IPV (Cunradi, 2010; Waller et al., 2013), the effects of alcohol outlet density on domestic violence are not uniform across outlet types, highlighting the need for disaggregation of alcohol availability in communities into two broad categories of alcohol outlets, such as off-premise and on-premise. Off-premise alcohol outlets (e.g., liquor or convenience stores) have consistently shown a stronger association with violence relative to on-premise alcohol outlets (Grubesic, Pridemore, Williams, & Philip-Tabb, 2013; Snowden, 2016; Snowden, 2019), as well as domestic violence (Livingston, 2011; Roman & Reid, 2012). Cunradi et al. (2011) found that each additional off-premise alcohol outlet is associated with increases in rates of IPV-related police calls and IPV-related crime reports, with increases of 4% and 3%, respectively. There is also a relationship between more severe forms of violence and off-premise alcohol outlets. In certain cities, the homicide rate in a census tract with two off-premise alcohol outlets was nearly a quarter greater than census tracts containing only one (Scribner, Cohen, Kaplan, & Allen, 1999), highlighting the important role that alcohol density, rather than alcohol outlet count, has in community violence rates. Thus, we would expect that areas with high levels of off-premise outlets will have high levels of domestic violence, net of other neighborhood characteristics.

Although it is mediated by collective efficacy, concentrated disadvantage predicts violence, violent victimization, and homicide in communities (Sampson, Raudenbush, & Earls, 1997) and has repeatedly been linked with multiple manifestations of domestic violence. Stansfield and Doherty (2019) found that a higher level of concentrated disadvantage was significantly associated with higher rates of family violence. A longitudinal study of neighborhood disadvantage and intimate violence directed towards women found that a one unit increase in neighborhood disadvantage elevated the odds of violence by 50% (Benson, Fox, DeMaris, & Van Wyk, 2003), and a more recent study found that women are more likely to report IPV victimization in neighborhoods with greater levels of concentrated disadvantage (Voith & Brondino, 2017). Other forms of disadvantage, such as lack of health insurance, also produce deleterious effects in communities as a consequence of delayed health care among victims of domestic violence due to a lack of health insurance (Stansfield & Doherty, 2019). Alcohol availability and concentrated disadvantage have direct effects on violence as illustrated by these studies, but the strength of the alcohol/violence association is greater in socially disorganized communities, with more deprived areas experiencing a stronger impact of on-premise outlets on violence (Lightowlers, Pina-Sánchez, & McLaughlin, 2021; Pridemore & Grubesic, 2012).

Spillover Neighborhood Effects of Outlet Density

This review of the literature suggests that alcohol availability (especially availability of alcohol via off-premise alcohol outlets), concentrated disadvantage (and its attendant features, such as lack of health insurance) are consistent predictors of domestic violence, including IPV. What is less clear from prior research is how alcohol availability and concentrated disadvantage in neighboring units of analysis predict domestic violence in the focal units of analysis.

It is important to understand the effect of the characteristics of adjacent areas, such as alcohol outlet density in adjacent neighborhoods, on local rates of domestic violence for theoretical and practical reasons. First, artificial boundaries of units of analysis such as census block groups or census tracts do not align well with the movement of individuals across time and space, including being a victim or a perpetrator of domestic violence. This means that in the course of daily activities, we cross several boundaries of the communities in which we live, including where we purchase alcohol beverages, where we consume them, and where we otherwise might be at risk for victimization. When we only estimate the relationship between alcohol availability levels in the focal census tract or block group and how it produces certain levels of domestic violence in that census tract or block group, we fail to account for Tobler's (1970) first law of geography, which proposes that things that are closer in space are more similar than things that are further apart. If we accept this proposition as valid, we acknowledge that areas that have high alcohol availability are likely bordered by areas that have similarly high alcohol availability. This rich alcohol environment means that alcohol availability will have multiple pathways of influence on alcohol consumption and attendant consequences in the form of domestic violence.

Second, events occurring in the neighboring units of analysis may have an effect on events occurring in the local units of analysis. To illustrate, when communities impose restrictions on issuing new alcohol licenses or when the communities reach their alcohol license quota, an opening of a new alcohol outlet in the adjacent areas will also influence what happens in the focal community because store patrons can travel to the adjacent communities to purchase alcohol. From a practical standpoint, this means that any effort aimed at reducing or limiting alcohol availability must be made in collaboration with nearby communities, because alcohol availability levels in nearby communities may impact domestic violence in focal communities. As such, alcohol licensing committees would consider not only alcohol availability in areas at hand, but also alcohol availability levels in areas that surround them. In spite of all of this, the spatial spillover effect has not been adequately addressed in the literature on alcohol availability and domestic violence.

Our approach builds off the work of Gruenewald et al. (2006), who identified an association between assaultive violence and population characteristics, such as population density, in adjacent neighborhoods. But because some of the prior studies of alcohol availability in adjacent areas found a positive association (Gorman et al., 2018; Zhu, Gorman, & Horel, 2004), while in others no significant spillover effect of alcohol availability on violence was determined (Gorman, Speer, Gruenewald, & Labouvie, 2001), and because few prior studies of domestic violence systematically examined the role of adjacent levels of alcohol availability in addition to neighboring levels of concentrated disadvantage on local domestic violence rate, this study fills this important theoretical and practical gap. Therefore, although prior research unveiled a positive relationship between characteristics of surrounding neighborhoods and violence in general, the findings of the studies that examined the effect of alcohol outlet density in adjacent areas on violence offer mixed conclusions, while less is known about the role of the spillover effect on domestic violence.

Our study advances the literature on domestic violence, often manifested as IPV, and alcohol availability by considering the spatial nature of associations between alcohol outlet density and domestic violence, as well as the characteristics of neighboring units of analysis. Because of the important role that alcohol availability plays in MFDV, our study aims to expand the literature and advance the study on alcohol outlets and Male-on-Female domestic violence when accounting for the spatial spillover effect of alcohol availability, and controlling for common predictors of domestic violence, such as concentrated disadvantage, health insurance coverage, and residential population.

Methods

Research Site and Unit of Analysis

Milwaukee, Wisconsin is a metropolitan city that covers a geographic area of roughly 96 square miles and has a population of 577,222 individuals (U.S. Census Bureau, 2020). The units of analysis for this study are the 571 census block groups that lie within the boundaries of Milwaukee. Block groups were chosen as the units of analysis as they are the smallest and most ecologically meaningful units to serve as a proxy for neighborhoods. Previous research by Sampson and Raudenbush (2004) and Tillyer et al. (2020) demonstrated that block groups appropriately reflect the layout of neighborhoods and the social patterns that occur therein. Using block groups decreases the likelihood of aggregation bias that might occur with larger units of analysis such as census tracts or zip codes (see Andresen & Malleson, 2013; Hipp, 2007; Parker & Wolz, 1979; Wooldredge, 2002). The residential population of these block groups in 2013 ranged between 0 and 3091 individuals, with a mean of 1037.

Data

Licensed Outlets

Data on all active alcohol licenses for 2013 were obtained from the City of Milwaukee. The data included license address, which was geocoded using ArcGIS Pro with a 99.6% successful match rate, and license type¹ (e.g., on-premise, such as bars, pubs, taverns and restaurants, or off-premise, such as package store, liquor store, convenience store). These data were used to calculate on-premise outlet density, which was measured as the number of on-premise outlets per square mile for each block group, and off-premise outlet density, captured as the number of off-premise outlets per square mile for each block group.

Male-to-Female Domestic Violence (MFDV)

Data on violent victimization were obtained from the Milwaukee Police Department and based on incidents investigated by the police between January 1, 2013 and December 31, 2013. The data included the IBRS code, the date of the incident, the offense sequence number, person involvement (e.g., victim, arrestee, or unknown), sex, whether it was domestic violence, and the incident address. These data were geocoded using ArcGIS Pro with a 99.7% successful match rate. These data were used to calculate Male-on-Female domestic violence density per square mile for each block group.

Control Variables

Based on prior studies, we controlled for the lack of health insurance coverage (population without healthcare coverage), the total population for block groups, and socioeconomic disadvantage. Census data were drawn from the 2013 American Community Survey (ASC) 5-year estimates. Based on the prior research in this area (e.g., Benson et al., 2003; Browning, 2002; Stansfield & Doherty, 2019; Voith & Brondino, 2017) and the interactions between deprivation, alcohol availability, and violent crime described by Lightowlers et al. (2021), a scale of concentrated disadvantage was created by conducting principal components analysis of six variables: proportion of total householders with no wife or husband present and with children under 18, proportion of total population that are Black or African American alone, proportion of total civilian labor force who are unemployed, proportion of total households receiving public assistance income, proportion of total population without a high school degree, and proportion of total households with income in the past 12 months below the poverty line. We use the first indicator to measure proportion of single-headed households, which is a commonly used measure of family structure in literature on social disorganization, to capture the dynamic of family structure on guardianship and supervision in communities. The second indicator, the proportion of total population that are Black or African American alone, is included to account for the racial and economic segregation in Milwaukee created through historical redlining policies used to limit the options of potential homeowners, which primarily targeted Black or African American people. Using principal components analyses, we created the concentrated disadvantage factor. The created factor explained 52.3% of the total variance with a Cronbach’s alpha of .73.

Statistical Analyses

Census data were cleaned using R, whereas alcohol outlet data, and MFDV data were pre-processed, managed, and geocoded using ArcGIS Pro. Next, geocoded data were spatially joined to census block groups (N = 571) using ArcGIS Pro. To account for spatial effects, a queen-contiguity weights matrix was created using GeoDa (Anselin, 2005). The spatial nature of the data allowed us to estimate maximum likelihood estimation (MLE) models with the spatial lag function to account for spatial autocorrelation of MFDV and the characteristics of neighboring block groups. Because levels of violence in one area can be predicted from the levels of violence in neighboring areas (Tobler, 1970), all models controlled for the spatial lag of MFDV.

We conducted six sets of analyses. The first model examined whether MFDV is associated with total (i.e., on- and off-premise) alcohol outlet density, as one broad category of alcohol availability in communities. The second model examined whether MFDV is associated with on-premise density, while the third model examined whether MFDV is associated with off-premise density.

The next set of models investigated whether MFDV is associated with alcohol outlet densities in adjacent units of analysis. Therefore, the fourth model examined whether MFDV is associated with total alcohol outlet density in adjacent block groups, the fifth model examined whether MFDV is associated with on-premise alcohol outlet density in surrounding areas, and the last model examined such associations for adjacent off-premise alcohol outlet density. These models controlled for neighborhood characteristics (concentrated disadvantage, health insurance coverage, and population) in the focal and adjacent units of analysis.

Results

Descriptive statistics for the variables used in the spatial regression models are presented in Table 1. On average, Milwaukee block groups contained 93 incidents of Male-on-Female domestic violence per square mile. Block groups contained an average of six off-premise and 15 on-premise alcohol outlets per square mile.

Table 1.

Descriptive Statistics of Milwaukee Block Groups (N = 571).

	Minimum	Maximum	Mean	SD
MFDV density	0	710.44	93.11	99.17
MFDV density lag	0	391.06	89.89	71.01
Off-premise density	0	80.05	5.69	11.07
Off-premise density lag	0	32.22	5.41	5.62
On-premise density	0	340.12	15.25	33.22
On-premise density lag	0	180.39	14.81	23.51
Total outlet density	0	340.12	20.94	36.65
Total outlet density lag	0	192.34	20.22	25.28
Concentrated disadvantage	−2.13	2.84	0	1
Concentrated disadvantage lag	−1.73	1.96	.02	.84
No health coverage	0	712	157.81	123.90
No health coverage lag	16.75	553.67	158.68	90.89
Total population	0	3091	1037.32	453.15
Total population lag	343	2513	1051.9	275.03

Table 2 shows the correlation matrix. As hypothesized, the density of MFDV was positively and significantly correlated with off-premise alcohol outlets and off-premise alcohol outlets in adjacent block groups. However, the density of MFDV was not significantly correlated with on-premise alcohol outlets.

Table 2.

Bivariate Correlations for Milwaukee Block Groups (N = 571).

	1	2	3	4	5	6	7	8	9	10	11	12	13	14
1. MFDV density	1
2. On-premise density lag	−.055	1
3. Off-premise density lag	.480**	.209**	1
4. Total outlet density lag	.056	.976**	.416**	1
5. On-premise density	−.063	.654**	.123**	.636**	1
6. Off-premise density	.344**	.09*	.254**	.14**	.159**	1
7. Total outlet density	.047	.621**	.188**	.619**	.955**	.446**	1
8. No health coverage lag	.239**	.203**	.42**	.282**	.147**	.207**	−.196**	1
9. Concentrated disadvantage lag	.646**	−.27**	.476**	−.145**	−.2**	.262**	−.102*	.202**	1
10. Total population lag	−.235**	.134**	−.136**	.094*	.125**	−.064	.094*	.454**	−.258**	1
11. No health coverage	.343**	.110**	.363**	.183**	.107*	.266**	.178**	.551**	.3**	.013	1
12. Concentrated disadvantage	.634**	−.223**	.426**	−.112**	−.21**	.249**	−.115**	.162**	.82**	−.258**	.284**	1
13. Total population	−.130**	.087	.071	.065	.087*	−.076	.056	.151**	−.174**	.308**	.28	−.101*	1
14. MFDV density lag	.726**	−.081	.631**	.065	−.053	.345**	.056	.347**	.818**	−.271**	.390**	.729**	−.179**	1

Note: Correlations are significant at .001 and .05: P**, P*.

Six models were estimated. All six models were spatially lagged regression models which used a first order queen contiguity matrix. Table 3 shows the results of the three baseline models that investigate the associations between alcohol availability and MFDV when controlling for neighborhood characteristics in focal units of analysis. Model 1 examined the association between total outlet density and MFDV density net of control variables. The model explained 49% of the variance in MFDV density. In Model 1, total alcohol outlet density was not significantly associated with MFDV density. Models 2 and 3 were estimated to provide separate analyses of the relationship between on- and off-premise alcohol outlet density with MFDV. As shown in Model 2, the association between MFDV density and on-premise alcohol outlet density was not significant. The results of Model 3 indicate that off-premise alcohol outlet density was positively and significantly associated with MFDV density (B = 1.09, P = .00). Here, we see that off-premise density is a statistically significant predictor of MFDV, net of other neighborhood characteristics in the focal unit of analysis, suggesting that the effect of off-premises on MFDV exists independent of the neighborhood characteristics. Model 3 explained 50% of the variance in MFDV density.

Table 3.

Male-on-Female Domestic Violence Density Regressed on Alcohol Outlet Density and Control Variables for Milwaukee Block Groups (N = 571).

	Model 1Total alcohol outlet density		Model 2On-premise alcohol outlet density		Model 3Off-premise alcohol outlet density
	B	se	B	se	B	se
Total alcohol outlet density	.14	.08	.06	.09	—	—
On-premise density	—	—	.06	.09
Off-premise density	—	—	—	—	1.09***	.27
Concentrated disadvantage	35.04***	3.63	34.28***	3.66	32.60***	3.58
No health coverage	.09**	.03	.10***	.03	.08**	.03
Total population	−.02**	.01	−.02**	.01	−.02*	.01
MFDV lag	.49***	.04	.50***	.04	.48***	.04
Spatial Pseudo R-squared	.49		.48		.50

Note: Correlations are significant at .001, .01, and .05: p***, p**, p*.

Table 4 displays the results of the three models which estimated spatially lagged measures of alcohol outlets and control variables. Model 4 examined the association of total alcohol outlets in adjacent areas with MFDV net of control variables. The model explained 49% of the variance in MFDV. Total outlet density in adjacent areas was not a significant predictor of MFDV density. This pattern seems to hold true also for Model 5, which examined the role of adjacent levels of on-premise alcohol outlet density on MFDV. Like total outlet density, the density of on-premise alcohol outlets in adjacent areas was not a statistically significant predictor of MFDV, a finding that reflects a similar pattern as what we unveiled in the baseline Model 2 estimated earlier. Last, in Model 6 we see that the density of off-premise alcohol outlets in neighboring areas had a statistically significant and positive association with MFDV density (B = 1.47, P = .00). This model explained 51% of variance in MFDV density.

Table 4.

Spatial Lag Regression Results for Male-on-Female Domestic Violence Density Regressed on Alcohol Outlet Density and Control Variables for Milwaukee Block Groups (N = 571).

	Model 4Total alcohol outlet density		Model 5On-premise alcohol outlet density		Model 6Off-premise alcohol outlet density
	B	se	B	se	B	se
Total alcohol outlet density lag	.22	.15	—	—	—	—
On-premise density lag	—	—	.32	.17	—	—
Off-premise density lag	—	—	—	—	1.47*	.65
Concentrated disadvantage lag	20.64**	7.08	20.92**	7.12	16.17*	6.83
No health coverage lag	.03	.05	.04	.05	.01	.05
Total population lag	−.03	.01	−.03	.01	−.02	.01
Total alcohol outlet density	.08	.10	—	—	—	—
On-premise density	—	—	.05	.11	—	—
Off-premise density	—	—	—	—	1.04***	.27
Concentrated disadvantage	24.56***	4.98	23.95***	5.001	22.92***	4.92
No health coverage	.08*	.03	.08*	.03	.07*	.03
Total population	−.01	.01	−.01	.01	−.01	.01
MFDV lag	.40***	.06	.41***	.05	.37***	.06
Spatial Pseudo R-squared	.49		.49		.51

Note: Correlations are significant at .001, .01, and .05: P***, P**, P*.

Discussion

Using spatially-informed regression models, this study examined the association of alcohol availability in adjacent neighborhoods on Male-on-Female domestic violence in Milwaukee, Wisconsin while accounting for the local and adjacent neighborhood characteristics. The results of these models support the Brantingham and Brantingham (1999) theory that rates of violence are influenced by place characteristics (i.e., alcohol outlet density) and population characteristics (i.e., neighborhood disadvantage, health access barriers) at the neighborhood level and extend this theory to incorporate the characteristics of surrounding neighborhoods as well as places that operate there.

Not only is availability of off-premise alcohol outlets in local, as well as in adjacent communities, a significant contributor to domestic violence against females at the hands of males in Milwaukee, but the effect of availability in adjacent areas is roughly 40% greater than that in local neighborhoods. These findings remain consistent even after controlling for neighborhood characteristics (i.e., concentrated disadvantage) in focal and adjacent areas. Neighborhoods with high levels of concentrated disadvantage and which are in close proximity to neighborhoods with similarly high levels of disadvantage have high levels of MFDV. Although prior studies that we have reviewed previously have controlled for neighborhood characteristics at the focal units of analysis, less is known about alcohol availability and neighborhood characteristics in adjacent areas in predicting domestic violence committed by males against females.

With its focus on spillover effect of alcohol availability, this study advances previous work (Gruenewald et al., 2006; Snowden, 2019; Snowden, Hockin, & Pridemore, 2020; Zhu et al., 2004) that highlighted the positive relationship between densities of off-premise alcohol outlets and greater rates of assault, noting that these types of alcohol outlets directly affect community violence rates. Grubesic and Pridemore (2011), for example, studied how violence concentrates within certain distance thresholds from alcohol outlet agglomerations and found that clusters of excessive risk of violence occur within 575–700 ft, or about 1.5 city blocks away, from alcohol agglomerations (an average Milwaukee street block is about 500 feet in length), and again at the distance of 1500 ft until 1613 ft (or about 3 Milwaukee city blocks away from outlet agglomerations). More severe acts of violence follow similar patterns for the first distance threshold described here, but the second distance threshold has greater spatial distance, clustering at the distance of 1197 ft until 4238 ft (or slightly more than 2 blocks and less than 8 blocks away from alcohol agglomerations).

Grubesic and Pridemore (2011) study of alcohol outlet agglomerations and violence provides several potential mechanisms through which off-premise outlets might explain increased rates of domestic violence. First, purchasing alcohol at these locations allows for consumption in an environment which is absent of capable guardians (Cohen & Felson, 1979). Unlike on-premise outlets which have patrons and staff who may intervene when acts of domestic violence occur, off-premise alcohol outlets provide not only less expensive alcohol beverage (per unit) relative to on-premises, but the consumption of alcohol also purchased at off-premise alcohol outlets typically takes place in private or less supervised drinking settings (e.g., one’s home, someone else’s home, nearby alley, or park) and the interaction between the patron and off-premise clerk is short (Trangenstein et al., 2018). Grubesic and Pridemore (2011) find that alcohol outlets exert influence over clusters of violence well beyond walking distance from the outlets, wherein people buy alcohol, take it home for consumption and proceed to commit a violent act. This aligns with routine activities theory (Cohen & Felson, 1979), which calls for a suitable target, a motivated offender, and a lack of capable guardians. Off-premise outlets provide less expensive alcohol and patrons can purchase greater quantities, and take it to places that have no place managers (Trangenstein et al., 2018). Because domestic violence occurs “behind closed doors,” there is immediately a lack of capable guardians. Heavy intoxication can provide the motive for an offender, and the nearest suitable target will likely be their partner, family member or co-habitant. Moreover, in private drinking settings there is no mechanism to moderate alcohol consumption or limit binge drinking. Previous research on binge drinking and family violence (e.g., Stansfield & Doherty, 2019) and heavy drinking and IPV (e.g., Waller et al., 2013) indicate a positive association between greater quantities of alcohol consumed and violence. In an on-premise environment, a patron can be refused service if they are already intoxicated or aggressive, leading us to consider that greater opportunities to drink heavily without being monitored are present when purchasing alcohol from off-premise outlets.

This study answered previous calls for examination of the effects of alcohol outlets in surrounding neighborhoods (Waller et al., 2012). As hypothesized, the density of off-premise alcohol outlets in adjacent block groups is associated with greater rates of MFDV. Both off-premise alcohol outlets in surrounding neighborhoods and neighborhood characteristics of surrounding areas are positively related to domestic violence carried out by males against females. Moreover, the spatially lagged variables produce differential effects relative to their focal block group counterparts. The spatial lag effect of concentrated disadvantage generated a weaker positive relationship with MFDV than concentrated disadvantage within the neighborhood. Notably, the spillover effect of off-premise alcohol outlets shows a stronger positive association with MFDV than off-premise alcohol outlets within block groups.

Taken together, these findings suggest that alcohol control policies must carefully assess alcohol availability in adjacent communities as well as in local communities. Considering the positive interaction effect between alcohol availability and disadvantage described by Lightowlers et al. (2021), as well as what we find here, alcohol availability must be particularly scrutinized when communities that have high levels of MFDV suffer not only from high levels of alcohol availability via off-premise outlets but also high levels of concentrated disadvantage. The current study highlights the importance of examining this relationship when neighborhoods are adjacent to communities in which off-premise availability and concentrated disadvantage levels are equally high. Addressing the concentration of alcohol outlets is a task that may be more successfully achieved through regulation and policy relative to structural neighborhood characteristics associated with concentrated socioeconomic disadvantage. Examples of this include revoking licenses from nuisance outlets and refusing to issue permits when existing outlets go out of business (Grubesic & Pridemore, 2011). Limiting the number of off-premise outlets by spatial density rather than total number of outlets is an approach which takes into consideration the spillover effect, protecting not only vulnerable neighborhoods but also the adjacent ones.

Limitations

There are several limitations to this study. To begin with, the cross-sectional nature of the data limits conclusions regarding the directionality of relationships and causal effects. It is not possible to determine from the data whether alcohol use and domestic violence co-occurred. Because the dependent variable of MFDV is based on calls for service to the Milwaukee Police Department, unreported instances of MFDV violence are excluded from the analysis, and therefore the data we analyze here provide a conservative estimate of domestic violence. Police data only represent events that are known to the police, where police were called or became involved, which does not cover the full scope of domestic violence incidents. It might be that people residing in neighborhoods that are overly policed and face disproportionate incarceration rates might be hesitant to call the police when hearing or experiencing a domestic dispute. However, the data were recorded and provided by one police department, and reporting and recording procedures for violent crimes tend to be similar across units of analysis (Baumer, 2002), although less is known whether this is the case for intimate partner violence, in particular. Just as we are unable to determine the co-occurrence of alcohol use and violence, the true extent of domestic violence is obscured due to it being a violent crime that occurs “behind closed doors.” Our study was not granular enough to distinguish between simple assault and more extreme forms of violence such as choking or forced sex. This presents an opportunity for future research wherein alcohol outlet density in surrounding areas is considered for other forms of domestic violence. In addition, the nature of our victimization data did not allow for further disaggregation of Male-on-Female domestic violence incidents to be examined as Male-on-Female intimate partner violence. Instead, our study examines a broader indicator of violence (i.e., domestic violence committed by males against female victims), rather than a specific type of domestic violence (e.g., domestic violence committed by males against their female intimate partners). Future studies should be expanded to examine Female-on-Male violence, as well as violence between same-sex partners. Finally, the size of alcohol outlets and their operating hours are important contributors to alcohol availability but were not incorporated into the analysis. These limitations should not detract from the study findings, which furthered the understanding of how alcohol outlets and domestic violence are intertwined. This study was designed to examine this relationship through spatially informed models, controlling for neighborhood characteristics in focal and adjacent areas, and disaggregating total alcohol outlets into on- and off-premise outlets.

Conclusion

This study has important policy implications regarding harm reduction in communities. Based on the findings of this study, in Milwaukee, it may be possible to reduce and prevent domestic violence by instituting changes in alcohol availability, especially when licensing alcohol retailers for sale of alcohol for off-premise consumption. The first step in reducing domestic violence would be to also reduce the number of off-premise alcohol outlets operating within adjacent neighborhoods and prevent new off-premise alcohol outlets from opening in neighborhoods which already have a large concentration of off-premise alcohol outlets, and which border neighborhoods with similarly high concentrations of such outlets. Considering the findings of this study, the unique relationship between outlets where alcohol can be purchased for consumption elsewhere suggests restrictive measures imposed on such outlets could reduce domestic violence committed by males against females.

Future research should examine the causal association between alcohol outlets and domestic violence. Further studies should incorporate neighborhood characteristics (e.g., social disorganization and concentrated disadvantage) which may affect the relationship between alcohol outlet density and domestic violence by examining whether social disorganization moderates the effect of off-premise outlets on MFDV to better understand whether the effect of off-premise outlets on MFDV is greater in communities in which concentrated disadvantage is high. This is important because empirical evidence and theoretical reasoning suggest that the effects of alcohol outlet density on violence are modified by neighborhood social disorganization and concentrated disadvantage (Cunradi, 2007; Mckinney et al., 2012; Pridemore & Grubesic, 2012). In fact, violence is more likely to occur in socially disorganized neighborhoods and neighborhoods with high levels of concentrated disadvantaged (Benson et al., 2003; Voith & Brondino, 2017). Future studies should explore on- and off-premise outlets at a more granular level as there is a difference in settings within these categories. On-premise outlets such as restaurants are dissimilar from pubs and bars, while off-premise outlets such as grocery stores are quite different from packaged liquor stores. Their respective role in domestic violence warrants further investigation.

Finally, further research should be conducted on the intersection of alcohol outlet density and the rise in domestic violence during the COVID-19 pandemic. Restriction of access to on-premise outlets in the early days of the pandemic, in conjunction with extremely limited guardianship, reduced access to support services, and increased stress likely lead to an increase in the role off-premise outlets play in contributing to domestic violence. Clearly, off-premise alcohol outlets matter for violence and more work needs to be done in this area.

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 iD

Aleksandra J. Snowden

Note

Author Biographies

Thomas H. Johnson is a Data Scientist in the Office of Research at University of Wisconsin-Milwaukee. His research interests include violence, incarceration, and algorithmic justice.

Aleksandra J. Snowden is an Associate Professor in the Department of Social and Cultural Sciences at Marquette University. Her research interests include crime mapping and crime analysis, neighborhood characteristics, alcohol availability, violence, and spatial analytical methods.

References

Ahern

Margerison-Zilko

Hubbard

Galea

(2013). Alcohol outlets and binge drinking in urban neighborhoods: The implications of nonlinearity for intervention and policy. American Journal of Public Health, 103(4), e81–e87. https://doi.org/10.2105/AJPH.2012.301203

Andresen

M. A.

Malleson

(2013). Spatial heterogeneity in crime analysis. In Crime modeling and mapping using geospatial technologies (pp. 3–23). Berlin, Germany: Springer.

Anselin

(2005). Exploring spatial data with GeoDa. Chicago, IL: Center for Spatially Integrated Social Science, University of Illinois.

Baumer

(2002). Neighborhood disadvantage and police notification by victims of violence. Criminology, 40, 579–616.

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

Brantingham

P. L.

Brantingham

P. J.

(1999). A theoretical model of crime hot spot generation. Studies on Crime & Crime Prevention, 8(1), 7–26.

Breiding

M. J.

Basile

K. C.

Smith

S. G.

Black

M. C.

Mahendra

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

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. https://doi.org/10.1111/j.1741-3737.2002.00833.x

Caetano

Nelson

Cunradi

(2001). Intimate partner violence, dependence symptoms and social consequences from drinking among white, black and Hispanic couples in the United States. The American Journal on Addictions, 10(s1), s60–s69. https://doi.org/10.1080/10550490150504146

10.

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. https://doi.org/10.1177/0886260509354497

11.

Caetano

Schafer

Cunradi

C. B.

(2017). Alcohol-related intimate partner violence among white, black, and Hispanic couples in the United States. In Domestic violence (pp. 153–160). England, UK: Routledge.

12.

Cafferky

B. M.

Mendez

Anderson

J. R.

Stith

S. M.

(2018). Substance use and intimate partner violence: A meta-analytic review. Psychology of Violence, 8(1), 110–131. https://doi.org/10.1037/vio0000074

13.

Cohen

L. E.

Felson

(1979). Social change and crime rate trends: A routine activity approach. American Sociological Review, 44(4), 588–608. https://doi.org/10.2307/2094589

14.

Coker

A. L.

Davis

K. E.

Arias

Desai

Sanderson

Brandt

H. M.

Smith

P. H.

(2002). Physical and mental health effects of intimate partner violence for men and women. American Journal of Preventive Medicine, 23(4), 260–268. https://doi.org/10.1016/s0749-3797(02)00514-7

15.

Cunradi

C. B.

(2007). Drinking level, neighborhood social disorder, and mutual intimate partner violence. Alcoholism, Clinical and Experimental Research, 31(6), 1012–1019. https://doi.org/10.1111/j.1530-0277.2007.00382.x

16.

Cunradi

C. B.

(2010). Neighborhoods, alcohol outlets and intimate partner violence: Addressing research gaps in explanatory mechanisms. International Journal of Environmental Research and Public Health, 7(3), 799–813. https://doi.org/10.3390/ijerph7030799

17.

Cunradi

C. B.

Mair

Ponicki

Remer

(2011). Alcohol outlets, neighborhood characteristics, and intimate partner violence: Ecological analysis of a California city. Journal of Urban Health: Bulletin of the New York Academy of Medicine, 88(2), 191–200. https://doi.org/10.1007/s11524-011-9549-6

18.

Gorman

D. M.

Ponicki

W. R.

Zheng

Han

Gruenewald

P. J.

Gaidus

A. J.

(2018). Violent crime redistribution in a city following a substantial increase in the number of off‐sale alcohol outlets: A bayesian analysis. Drug and Alcohol Review, 37(3), 348–355. https://doi.org/10.1111/dar.12636

19.

Gorman

D. M.

Speer

P. W.

Gruenewald

P. J.

Labouvie

E. W.

(2001). Spatial dynamics of alcohol availability, neighborhood structure and violent crime. Journal of Studies on Alcohol, 62(5), 628–636. https://doi.org/10.15288/jsa.2001.62.628

20.

Grubesic

T. H.

Pridemore

W. A.

(2011). Alcohol outlets and clusters of violence. International Journal of Health Geographics, 10(30), 30. https://doi.org/10.1186/1476-072X-10-30

21.

Grubesic

T. H.

Pridemore

W. A.

Williams

D. A.

Philip-Tabb

(2013). Alcohol outlet density and violence: The role of risky retailers and alcohol-related expenditures. Alcohol and Alcoholism, 48(5), 613–619. https://doi.org/10.1093/alcalc/agt055

22.

Gruenewald

P. J.

Freisthler

Remer

LaScala

E. A.

Treno

(2006). Ecological models of alcohol outlets and violent assaults: Crime potentials and geospatial analysis. Addiction, 101(5), 666–677. https://doi.org/10.1111/j.1360-0443.2006.01405.x

23.

Haines

Graham

(2005). Violence prevention in licensed premises. In Stockwell

Gruenewald

P. J.

Toumbourou

Loxley

(Eds.), Preventing harmful substance use: The evidence base for policy and practice (pp. 163–176). New York, NY: John Wiley.

24.

Hipp

J. R.

(2007). Block, tract, and levels of aggregation: Neighborhood structure and crime and disorder as a case in point. American Sociological Review, 72(5), 659–680. https://doi.org/10.1177/000312240707200501

25.

Kearns

M. C.

Reidy

D. E.

Valle

L. A.

(2015). The role of alcohol policies in preventing intimate partner violence: A review of the literature. Journal of Studies on Alcohol and Drugs, 76(1), 21–30. https://doi.org/10.15288/jsad.2015.76.21

26.

Laskey

Bates

E. A.

Taylor

J. C.

(2019). A systematic literature review of intimate partner violence victimisation: An inclusive review across gender and sexuality. Aggression and Violent Behavior, 47, 1–11. https://doi.org/10.1016/j.avb.2019.02.014

27.

Lightowlers

Pina-Sánchez

McLaughlin

(2021). The role of deprivation and alcohol availability in shaping trends in violent crime. European Journal of Criminology, 147737082110360. https://doi.org/10.1177/14773708211036081

28.

Lira

M. C.

Xuan

Coleman

S. M.

Swahn

M. H.

Heeren

T. C.

Naimi

T. S.

(2019). Alcohol policies and alcohol involvement in intimate partner homicide in the U.S. American Journal of Preventive Medicine, 57(2), 172–179. https://doi.org/10.1016/j.amepre.2019.02.027

29.

Livingston

(2011). A longitudinal analysis of alcohol outlet density and domestic violence. Addiction, 106(5), 919–925. https://doi.org/10.1111/j.1360-0443.2010.03333.x

30.

McKinney

C. M.

Chartier

K. G.

Caetano

Harris

T. R.

(2012). Alcohol availability and neighborhood poverty and their relationship to binge drinking and related problems among drinkers in committed relationships. Journal of Interpersonal Violence, 27(13), 2703–2727. https://doi.org/10.1177/0886260512436396

31.

Murray

Bunch

Hunt

E. D.

(2016). Strengthening community-level understanding of and responses to intimate partner violence using geographic information systems (GIS). Journal of Aggression, Conflict and Peace Research, 8(3), 197–211. https://doi.org/10.1108/jacpr-01-2016-0209

32.

Naimi

T. S.

Xuan

Coleman

S. M.

Lira

M. C.

Hadland

S. E.

Cooper

S. E.

Heeren

T. C.

Swahn

M. H.

Swahn

M. H.

(2017). Alcohol policies and alcohol-involved homicide victimization in the United States. Journal of Studies on Alcohol and Drugs, 78(5), 781–788. https://doi.org/10.15288/jsad.2017.78.781

33.

Parker

D. A.

Wolz

M. W.

(1979). Alcohol problems and the availability of alcohol. Alcoholism, Clinical and Experimental Research, 3(4), 309–312. https://doi.org/10.1111/j.1530-0277.1979.tb05328.x

34.

Pridemore

W. A.

Grubesic

T. H.

(2012). Community organization moderates the effect of alcohol outlet density on violence. The British Journal of Sociology, 63(4), 680–703. https://doi.org/10.1111/j.1468-4446.2012.01432.x

35.

Renzetti

C. M.

Lynch

K. R.

DeWall

C. N.

(2018). Ambivalent sexism, alcohol use, and intimate partner violence perpetration. Journal of Interpersonal Violence, 33(2), 183–210. https://doi.org/10.1177/0886260515604412

36.

Roman

C. G.

Reid

S. E.

(2012). Assessing the relationship between alcohol outlets and domestic violence: Routine activities and the neighborhood environment. Violence and Victims, 27(5), 811–828. https://doi.org/10.1891/0886-6708.27.5.811

37.

Sampson

R. J.

Raudenbush

S. W.

(2004). Seeing disorder: Neighborhood stigma and the social construction of “Broken Windows”. Social Psychology Quarterly, 67(4), 319–342. https://doi.org/10.1177/019027250406700401

38.

Sampson

R. J.

Raudenbush

S. W.

Earls

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

39.

Scribner

Cohen

Kaplan

Allen

S. H.

(1999). Alcohol availability and homicide in new Orleans: Conceptual considerations for small area analysis of the effect of alcohol outlet density. Journal of Studies on Alcohol, 60(3), 310–316. https://doi.org/10.15288/jsa.1999.60.310

40.

Snowden

A. J.

(2016). Alcohol outlet density and intimate partner violence in a non-metropolitan college town: Accounting for neighborhood characteristics and alcohol outlet types. Violence and Victims, 31(1), 111–123. https://doi.org/10.1891/0886-6708.VV-D-13-00120

41.

Snowden

A. J.

(2019). Exploring violence: The role of neighborhood characteristics, alcohol outlets, and other micro-places. Social Science Research, 82, 181–194. https://doi.org/10.1016/j.ssresearch.2019.04.018

42.

Snowden

A. J.

Hockin

Pridemore

W. A.

(2020). The neighborhood-level association between alcohol outlet density and female criminal victimization rates. Journal of Interpersonal Violence, 35(15–16), 2639–2662. https://doi.org/10.1177/0886260517719080

43.

Stansfield

Doherty

(2019). Neighborhood health, social structure and family violence. Social Science Research, 81, 12–22. https://doi.org/10.1016/j.ssresearch.2019.02.007

44.

Tillyer

M. S.

Wilcox

Walter

R. J.

(2020). Crime generators in context: Examining ‘place in neighborhood’ propositions. Journal of Quantitative Criminology, 37(2), 517–546. https://doi.org/10.1007/s10940-019-09446-5

45.

Tobler

W. R.

(1970). A computer movie simulating urban growth in the Detroit region. Economic geography, 46(sup1), 234–240.

46.

Trangenstein

P. J.

Curriero

F. C.

Webster

Jennings

J. M.

Latkin

Eck

Jernigan

D. H.

(2018). Outlet type, access to alcohol, and violent crime. Alcoholism, Clinical and Experimental Research, 42(11), 2234–2245. https://doi.org/10.1111/acer.13880

47.

U.S. Census Bureau . (2020). QuickFacts: Milwaukee city, Wisconsin. Retrieved June 18, 2022, from https://www.census.gov/quickfacts/fact/table/milwaukeecitywisconsin/LND110210

48.

Voith

L. A.

Brondino

M. J.

(2017). Neighborhood predictors of intimate partner violence: A theory‐informed analysis using hierarchical linear modeling. American Journal of Community Psychology, 60(1–2), 187–198. https://doi.org/10.1002/ajcp.12163

49.

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. https://doi.org/10.1177/0886260511431435

50.

Waller

M. W.

Iritani

B. J.

Christ

S. L.

Tucker Halpern

Moracco

K. E.

Flewelling

R. L.

(2013). Perpetration of intimate partner violence by young adult males: The association with alcohol outlet density and drinking behavior. Health & Place, 21, 10–19. https://doi.org/10.1016/j.healthplace.2013.01.002

51.

Wei

Grubesic

T. H.

Kang

(2021). Spatiotemporal patterns of alcohol outlets and violence: A spatially heterogeneous Markov chain analysis. Environment and Planning B: Urban Analytics and City Science, 48(8), 2151–2166. https://doi.org/10.1177/2399808320965569

52.

WHO . (2013). Global and regional estimates of violence against women: Prevalence and health effects of intimate partner violence and non-partner sexual violence. Geneva, Switzerland: WHO.

53.

WHO . (2018). Global and regional estimates of violence against women: Prevalence and health effects of intimate partner violence and non-partner sexual violence. Geneva, Switzerland: WHO.

54.

Wisconsin State Legislature (n.d.). Domestic abuse incidents; arrests and prosecution (statute 968-075). Retrieved January 10, 2022 from https://docs.legis.wisconsin.gov/statutes/statutes/968/075?view=section.

55.

Wooldredge

(2002). Examining the (ir)relevance of aggregation bias for multilevel studies of neighborhoods and crime with an example comparing census tracts to official neighborhoods in Cincinnati. Criminology, 40(3), 681–710. https://doi.org/10.1111/j.1745-9125.2002.tb00970.x

56.

Yates

(2019). An exercise in careful diplomacy”: Talking about alcohol, drugs, and family violence. Policy Design and Practice, 2(3), 258–274. https://doi.org/10.1080/25741292.2019.1638697

57.

Zhu

Gorman

D. M.

Horel

(2004). Alcohol outlet density and violence: A geospatial analysis. Alcohol and Alcoholism, 39(4), 369–375. https://doi.org/10.1093/alcalc/agh062