Fatal and Non-Fatal Intimate Partner Violence

Abstract

Comparing male and female victims across lethal and non-lethal intimate partner violence (IPV) can provide a better understanding of these incidents and assist policy makers in developing more tailored victim services and prevention programs. To date, little research has examined this issue. This study compares the characteristics that predict female and male IPV victimization. Given the role that law enforcement can play in promoting victim assistance programs, police data are used to explore this issue. This study finds that victim sex differences do exist within and across lethal and non-lethal IPV. These findings are discussed as well as their implications for future policy and research.

Keywords

intimate partner victim/offender relationship victimization gender NIBRS methodology public policy

The vast majority of research examining intimate partner violence (IPV) has concentrated on women. With regard to studying victims, this focus is not too surprising since women are more likely than men to be victims of IPV (e.g., Catalano, 2007; Rennison, 2003; Smith & Farole, 2009; Warner, 2010). For IPV offenders, recent attention also has been dominated by research centered on women mainly due to concerns about the unintended consequences of mandatory arrest policies (Hirschel & Buzawa, 2002; Melton & Belknap, 2003; for a review, see Malloy, McCloskey, Grigsby, & Gardner, 2003). Largely missing from this literature are male victims. Although some men are assaulted by their partners as a result of self-defense measures (Bouffard, Wright, Muftic, & Bouffard, 2008; Kruttschnitt & Carbone-Lopez, 2006; Miller, 2005; Swatt & He, 2006), not all male victims of IPV are the original aggressors (Felson & Lane, 2010). Understanding how circumstances compare between male and female victims can provide a more complete picture of IPV. This information can assist policy makers in developing more programs targeted to the appropriate population and clients (Johnson, 1995). Within this comparison, we are particularly interested in examining fatal and non-fatal IPV. If similar predictors are identified for fatal and non-fatal victimizations, such a pattern suggests a continuum of IPV. Such a finding would lend support to different conclusions, explanations, and policies than if patterns of fatal and non-fatal IPV suggest each is a distinct form of violence.

In addition to the limited attention on male victims, few IPV studies rely on police records. These data have limitations that are well documented, especially for studying IPV (Buzawa & Hotaling, 2006; Felson, Messner, Hoskin, & Deane, 2002; Felson & Pare, 2005; Fleury, Sullivan, Bybee, & Davidson, 1998; Thompson, Saltzman, & Bibel, 1999; Vazquez, Stohr, & Purkiss, 2005). Despite these limitations, understanding the characteristics of IPV that come to the attention of police is important from a policy perspective to help educate police on how best to handle IPV and to identify effective prevention and intervention strategies. As gatekeepers for the criminal justice system, the police hold an important key in the process of identifying those individuals in need of assistance (Buzawa & Buzawa, 2008). Victim services are particularly important in this area since a larger percentage of IPV victims seek out these services as compared to non-IPV victims (Langton, 2011). In examining all non-fatal crime victims, however, more women than men receive victim services (Langton, 2011). In this study, we sought to contribute to this literature and the understanding of these incidents by examining IPV cases that come to the attention of police for both male and female victims.

Background and Current Literature

Theoretical Framework

A long-standing debate in the study of IPV concerns whether this violence is gendered or not.¹ Initial research by feminist criminologists supported the perspective that IPV is gendered or asymmetrical violence and primarily a male-on-female phenomenon (Dobash & Dobash, 1998; Melton & Belknap, 2003; Tjaden & Thoennes, 2000). Under this view, male victims are generally characterized as the original abusers and their female offenders are the previous victims who fight back (Bouffard et al., 2008; Swatt & He, 2006). The opposing perspective is that IPV is non-gendered, family violence where victims are equally male and female (Archer, 2000; Straus & Gelles, 1986). Contributing to this continued debate is the fact that both theoretical frameworks have received empirical support (Chan, 2011; Melton & Sillito, 2011).

In an attempt to resolve this dispute, Johnson (1995) provided a third approach that synthesizes these two disparate perspectives. He hypothesized that IPV incidents have different processes, which can be categorized into typologies. Initially, Johnson identified two typologies—common couple and patriarchal terrorism—that correspond to the IPV debate. The “common couple” violence typology encompasses the non-gendered IPV findings. This type of violence is characterized by the occasional couple arguments that, according to Johnson, “get out of hand” (p. 287). These incidents are less serious in nature, result in minor injuries, and do not escalate over time. The common couple typology involves both male and female victims equally. The “patriarchal terrorism” typology corresponds to gendered IPV findings and is characterized by the male’s continued need to dominate and control his female partner. These incidents concern male-on-female violence that involve more serious violence and injury and typically escalate over time.

Subsequently, Johnson and Ferraro (2000) extended this approach to include two additional IPV categories: violent resistance and mutually coercive violence. “Violent resistance” is characterized by self-defense, typically involves female offenders, and results in homicide. The “mutually coercive violence” typology is characterized by domination and control by both partners in the relationship and is the least common of the four typologies. According to Johnson and Ferraro, the distinctions among the four typologies are determined through assessing the patterns across many interactions for a particular couple and assessing the motivation of the violence committed.

Empirical research has provided some support for the typologies identified by Johnson (1995) and Johnson and Ferraro (2000). The gender-neutral common couple typology has most frequently occurred in studies relying on samples drawn from the general population (e.g., Archer, 2000; Capaldi, Kim, & Shortt, 2007; Capaldi et al., 2009; Straus, 2011). The male-on-female typologies characterized by patriarchal terrorism and violent resistance tend to be identified in cases obtained from official police records (e.g., Bair-Merritt et al., 2010; Melton & Sillito, 2011; Warner, 2010). These typologies also are frequently used by IPV scholars as a way to frame their research (Melton & Sillito, 2011; Swatt & He, 2006). Given the empirical support and trend in the literature, we also used these typologies in our present study. Although we were limited in our ability to identify specific typologies (as we discuss below), this framework provided a useful rubric through which to guide our study.

Current IPV Research

IPV has received a great deal of research attention from criminologists as well as scholars in other disciplines. This work has not been distributed evenly, however, and a few areas have received rather limited study. Three of particular relevance for our study are the comparison between male and female IPV victims, the comparison of lethal and non-lethal IPV, and the use of police-based data.

Despite the gender-focused theoretical debate in the IPV literature, little attention has been given to specifically comparing male and female victims. The work in this area has considered mutual partner violence (Caetano, Ramisetty-Mikler, & Field, 2005; Caetano, Vaeth, & Ramisetty-Mikler, 2008; Field & Caetano, 2005), perpetrator arrest to prevent re-victimization (Cho & Wilke, 2010), and predictors of offender sex (Swatt & He, 2006). Other studies have compared male and female victims with a focus on particular characteristics such as alcohol use during IPV incidents (Thompson & Kingree, 2006), social economic status (Cunradi, Caetano, & Schafer, 2002), and psychological effects (Halpern, Oslak, Young, Martin, & Kupper, 2001). Although researchers have described sex differences for IPV victims, no study to date has specifically addressed what characteristics might predict this difference in victim sex for IPV.

Comparing lethal and non-lethal incidents is another area largely missing in the IPV literature overall, but particularly in discussions of victim sex. Previous researchers have suggested the need to study intimate partner murders with related offenses such as assault to gain a more complete picture of IPV (Block & Christakos, 1995), particularly “the complexity of female-perpetrated homicide incidents” (Swatt & He, 2006, p. 289). In particular, examining fatal and non-fatal IPV can help provide a better context for fatal incidents. Two alternate ways in which intimate partner murder can be viewed are (a) as part of a continuum of violence and an extension of aggravated assault and (b) as an act of violence distinct from non-lethal IPV assaults. Previous studies have described sex differences in murders involving intimate partners (Belknap, Larson, Abrams, Garcia, & Anderson-Block, 2012; Block & Block, 2012; Catalano, Smith, Snyder, & Rand, 2009; Rennison, 2001), but little attention has been devoted to extending this assessment across fatal and non-fatal incidents (Durose et al., 2005; Rennison, 2003). One reason for this omission may be a result of how researchers measure IPV. A common approach is to categorize all violent crimes together, which blurs murders with non-fatal crimes such as aggravated assault, sexual assault, robbery, and simple assault. The result is an inability to parse out lethal and non-lethal incidents as well as an assessment that is dominated by the most common and least serious crimes in the category (typically, simple assault).

IPV research also has largely discounted official data collected by police (for exceptions, see Kernsmith & Craun, 2008; Melton & Sillito, 2011; Thompson et al., 1999; Vazquez et al., 2005; Warner, 2010). The limitations of police data for studying victimization in general and IPV in particular are well known (Barnett-Ryan, 2007; Buzawa & Hotaling, 2006; Melton & Belknap, 2003). Although official data do have certain problems, they are important sources of information for IPV that comes to the attention of police. For researchers interested in these issues, two limitations that likely contribute to the underutilization of police data for IPV are concerns of underreporting and lack of readily accessible data. Underreporting to the police appears to be more common among female—rather than male—victims of IPV. Findings from the 2008 National Crime Victimization Survey indicated that 72% of cases of IPV against men were reported to police as compared to 49% of cases of IPV against women (Catalano et al., 2009). Another reason for the limited research relying on police records is a lack of readily accessible data. The only national source of police data is collected by the Federal Bureau of Investigation (FBI) in its Uniform Crime Reporting Program (UCR), which historically has collected only aggregate-level crime data.² As a result, researchers interested in analyzing incident characteristics from police records needed to collect their own data (e.g., Kernsmith & Craun, 2008; Melton & Sillito, 2011).

The UCR currently is converting its summary data system to an incident-based one known as the National Incident-Based Reporting System (NIBRS). NIBRS data provide a unique opportunity to explore officially recorded IPV. Despite the availability of these data, researchers only recently have started to use the NIBRS to study IPV. Roberts (2009) and Addington (2009) both reviewed studies that used NIBRS data to study IPV as well as victimization overall. Illustrative examples of this work include state-specific studies exploring patterns of IPV (Thompson et al., 1999; Vazquez et al., 2005) as well as national-level studies comparing heterosexual and homosexual IPV (Pattavina, Hirschel, Buzawa, Faggiani, & Bentley, 2007) and examining IPV in the context of family violence (Warner, 2010). Researchers also have capitalized on the strength of police data in providing reliable arrest information to study factors related to patterns of arrests (Dichter, Marcus, Morabito, & Rhodes, 2011), including mandatory arrest policies (Eitle, 2005) and the effect of IPV victim and offender use of alcohol on arrests (Hirschel & Hutchinson, 2011). Despite this growing interest in—and acceptance of—using the NIBRS to explore IPV, no published study has used these data to predict victim sex differences across fatal and non-fatal IPV incidents.

Research Questions and Hypotheses

As the previous discussion indicated, IPV research has not compared the predictors of male and female victims across fatal and non-fatal IPV, and little attention has been given to studying IPV using police data. Our study addressed these issues so that we may begin to gain a better understanding of possible variation across victim sex³ to fill this gap in the literature. In particular, we examined two primary research questions:

Research Question 1: Do victim, offender, and incident attributes for IPV incidents reported to police differentially predict victim sex?

Research Question 2: Do these predictors of victim sex vary across fatal and non-fatal IPV incidents?

Based on the typology framework presented by Johnson (1995) as well as the current literature, we hypothesized that different characteristics will predict the victim’s sex and that these characteristics will vary based on the lethality of the incident outcome.

Method

Data

Our study used victim-level data from the 2008 NIBRS, which is one of the most recent years of publicly available data. We relied on the NIBRS Extract Files provided by the National Archive of Criminal Justice Data (2010). NIBRS data are essential for this study as they provide victim and offender demographics, victim-offender relationship, incident details, and clearance status. One caveat in analyzing NIBRS data is its limited coverage. The NIBRS is a substantial departure in crime data collection for law enforcement agencies and requires a lengthy certification process (see Addington, 2004, for details). As a result, the conversion to the NIBRS has been gradual and has yet to achieve 100% coverage nationally. In 2008, 31 states were NIBRS certified.⁴ Within these 31 states, 9 states fully reported in the NIBRS and an additional 4 states had more than 94% NIBRS participation (Justice Research and Statistics Association [JRSA], n.d.-b). Overall, NIBRS agencies covered approximately 25% of the U.S. population in 2008 (JRSA, n.d.-b). Law enforcement agencies that participate in the NIBRS tend to represent smaller population areas. No agency covering a population of more than 1 million participated in the NIBRS (JRSA, n.d.-a). Because participation is voluntary, NIBRS states and law enforcement agencies do not constitute a representative sample of U.S. law enforcement agencies or states. This nonrepresentativeness suggests exercising caution when interpreting the results and generalizing beyond the NIBRS-participating agencies included in this study (but see Addington, 2008).

Case Selection

To study intimate partner violence, we concentrated our research on the most serious incidents that involve murder and non-negligent homicide (collectively referred to as murder) and aggravated assault. Intimate partner relationships are defined as dating (defined as boyfriend or girlfriend victim-offender relationships) and spousal (defined as spouse, common-law spouse, or ex-spouse victim-offender relationships). This distinction is relevant based on previous studies that suggest a slightly higher percentage of IPV reported to police involves spouses rather than dating partners (Durose et al., 2005). Because adolescent IPV may be conceptually different from adult IPV (see Durose et al., 2005; Halpern et al., 2001), we included only adult victims and offenders (defined as age 18 or older). In addition, our study focused on heterosexual relationships due to the very small number of homosexual relationships appearing in the data as well as previous studies indicating that homosexual IPV may involve different predictors from heterosexual IPV (Pattavina et al., 2007; Swatt & He, 2006). Finally, our study included only incidents involving one victim and one offender based on research that suggests distinct patterns for IPV involving multiple victims or offenders (Thompson et al., 1999) as well as to ensure statistical independence (Regoeczi, Jarvis, & Riedel, 2008). These case selection criteria resulted in a sample size of 267 intimate partner murders and 28,534 intimate partner aggravated assaults for 2008.

Measures

To explore our research questions, we used variables found to be predictive of IPV by previous studies. These variables include characteristics about the victim, offender, and incident and are described more particularly below. Frequencies for all variables are presented in Table 1.

Table 1.

Frequency of Aggravated Assaults and Murders [% (N) or M (SD)].

	Aggravated Assault
Characteristic	Population (N = 28,534)	Sample (n = 2,794)	Murder (N = 267)
Victim age (in years)	33.6 (11.1)	33.4 (10.9)	42.3 (15.4)
Victim sex
Female	75.2 (21,466)	75.9 (2,122)	81.3 (217)
Male	24.7 (7,056)	24.0 (671)	18.7 (50)
Missing	0.0 (12)	0.0 (1)	0 (0)
Victim race
Non-White	38.2 (10,901)	36.0 (1,007)	22.7 (74)
White	60.4 (17,228)	62.6 (1,749)	72.3 (193)
Missing	1.4 (405)	1.4 (38)	0 (0)
Offender age (in years)	34.5 (10.7)	34.3 (10.8)	43.7 (15.5)
Offender race
Non-White	43.5 (12,402)	41.3 (1,155)	31.1 (83)
White	55.3 (15,791)	57.6 (1,608)	68.5 (183)
Missing	1.2 (341)	1.1 (31)	0.4 (1)
Offender used alcohol/drugs
Not applicable	78.5 (22,392)	77.5 (2,165)	86.5 (231)
Yes	21.5 (6,142)	22.5 (629)	13.5 (36)
Offender weapon use
Firearm	9.3 (2,650)	10.5 (294)	49.4 (132)
Knife	22.5 (6,408)	22.6 (631)	22.1 (59)
Blunt object	13.9 (3,959)	12.8 (358)	3.7 (10)
Personal contact	34.5 (9,846)	34.0 (949)	10.1 (27)
Other	18.5 (5,287)	18.8 (524)	10.5 (28)
Missing	1.3 (384)	1.4 (38)	4.1 (11)
Location of assault
Outside home	19.7 (5,632)	19.5 (544)	18.7 (50)
Inside home	80.3 (22,902)	80.5 (2,550)	81.3 (217)
Cleared
No	34.4 (9,814)	32.8 (916)	24.3 (65)
Yes	65.6 (18,720)	67.2 (1,878)	75.7 (202)
Intimate partner relationship
Dating	62.9 (17,953)	63.8 (1,782)	42.3 (113)
Spouse	37.1 (10,581)	36.2 (1,012)	57.7 (154)
Injury
None	30.5 (8,689)	30.1 (841)	n/a
Minor	45.9 (13,095)	46.0 (1,285)	n/a
Serious	23.7 (6,750)	23.9 (668)	n/a

Victim characteristics

Victim characteristics include demographics (sex, age, and race), victim-offender relationship, and injury (for aggravated assaults). Victim sex was coded as female and male. Victim age is a continuous variable and was included based on research suggesting that female intimate partner murder victims are younger than male victims (Catalano et al., 2009). Victim race was included as White and non-White due to the small number of minority victims who are not African American. Previous studies indicated that IPV, especially incidents reported to the police, is more likely to involve White rather than non-White victims (Durose et al., 2005; Thompson et al., 1999; for an exception, see Catalano et al., 2009, with respect to Black female homicide victims). As described above, victim-offender relationship indicates whether the relationship involved dating partners or spouses.

The aggravated assault models also included the type of injury sustained by the victim.⁵ Since murders are by definition lethal crimes, no injury information is collected by the NIBRS. Previous IPV research suggested that incidents involving serious injuries are more likely to be reported by male victims of intimate partner violence than female victims (Kernsmith & Craun, 2008; Warner, 2010). Type of injury was coded as major, minor, or no injury. Major injuries include apparent broken bones, possible internal injuries, loss of teeth, severe lacerations, unconsciousness, and other major injuries. Minor injuries include “apparent minor injuries” as designated by NIBRS coding. For the aggravated assault multivariate models, “no injury” was used as the comparison group.

Offender characteristics

Offender characteristics include offender age and race, but not sex, since the cases are restricted to heterosexual relationships. Offender age was coded as a continuous variable. Research indicates that male arrestees are typically older than their victims whereas female arrestees tend to be younger than their victims (Henning & Feder, 2004). As with victim race, offender race was coded as White and non-White due to the small number of minority offenders who are not African American. Offender race is of interest since previous studies found that the majority of both male and female arrestees are African American (Henning & Feder, 2004). Offender used drugs/alcohol measured whether the offender used drugs or alcohol during the incident. Previous IPV research has suggested that incidents with male victims that end in an arrest are more likely to involve offenders under the influence of alcohol as compared to those with female victims (Henning & Feder, 2004). Overall, IPV tends to be characterized by the use of alcohol and drugs by the offender (Thompson & Kingree, 2006; Thompson et al., 1999).⁶

Incident characteristics

One limitation with the NIBRS data is that we were unable to identify specific typologies comparable to those suggested by Johnson (1995) and Johnson and Ferraro (2000). We can, however, use certain incident characteristics that are available to compare IPV across sex. With regard to incident characteristics, we included type of weapon, incident location, and clearance. Previous research highlighted the use of weapons in IPV incidents, particularly the use of firearms in lethal IPV (Durose et al., 2005; Thompson et al., 1999). For aggravated assaults, type of weapon was coded as firearm (which includes handguns and long guns), knife, blunt object, personal weapon (such as hands and feet), and other weapon. “Other weapon” includes other specified weapons such as fire, poison, motor vehicles, a generic category of other weapon, and no weapon. The category “personal weapons” was the comparison group for the aggravated assault model. For murders, a low occurrence of particular types of weapons necessitated coding the type of weapon more broadly into firearm, knife, and other weapon. Here, “other weapon” includes blunt object, personal weapons, and all of the categories used in other weapon for aggravated assault with the exception of “none.” Unlike aggravated assaults, murders do not include the “none” weapon category since “no weapon” is not a valid coding choice for NIBRS murders. For the murder multivariate models, knife was the comparison group. Location where the incident occurred was coded as home or not home location. Home location was included given that previous studies found that these incidents in general tend to occur in the victim’s home, regardless of victim sex (Rennison & Welchans, 2000; Vazquez et al., 2005). Clearance measured whether the police cleared the IPV case or not. Male IPV offenders are more likely to be arrested than female offenders (Pattavina et al., 2007). Clearance includes both cases that resulted in an arrest and cases cleared by exceptional means (FBI, 2004).⁷

Analytic Strategy

With our research questions, we sought to examine predictors of victim sex across both fatal and non-fatal IPV. To address these questions, we modeled victim sex as our dependent variable. By so doing, we were not seeking to predict victim sex per se but, rather, were predicting which characteristics are associated with incidents where the victim is male and those where the victim is female. Although scientifically regression may examine a causal relationship between an independent and dependent variable, a technical approach to regression analysis has “no implication that the latter are actually caused by the former” (Cohen, Cohen, West, & Aiken, 2003, p. 33). We modeled victim sex using regression in this technical sense. This approach allowed us to directly compare (through bivariate and multivariate analyses) as well as to quantify the magnitude of (through multivariate analyses) each predictor’s effect on differentiating between male and female victims.⁸ Considering sex as a dependent variable has support in the IPV literature. For example, Swatt and He (2006) modeled offender sex as a dependent variable in their study of factors predicting intimate partner murder.

The relative rarity of intimate partner murders in relation to aggravated assaults required estimating separate models for each form of IPV to compare the predictors of victim sex across the lethality of the incident. Bivariate relationships among the victim’s sex and particular victim, offender, and incident characteristics were initially assessed with either chi-square or independent t tests. Multivariate binary logistic regressions were modeled to examine the predictors of victim’s sex for aggravated assaults and murders. Multicollinearity and model fit were assessed, and no violations were evident for murder and the sample of aggravated assaults. Given the small amount of missing data (see Table 1), we used complete case analysis (Allison, 2002).

One issue in analyzing NIBRS data is the extremely large sample size. For our study, this concern mainly arose for the aggravated assaults (n = 28,534). Since the probability of rejecting the null hypothesis increases with large sample sizes (Lehmann, 2004), it was possible that significance would be found merely due to samples involving thousands of cases. Although statisticians have devoted a great deal of attention to addressing the issue of very small samples (fewer than 30), little attention has been devoted to the best practice for handling very large samples. Three possible options involved the following: (a) using a smaller p value to discern statistical significance (such as p < .001), (b) focusing on substantive or clinical significance, and (c) drawing a sample of the data for analysis to ascertain the consistency of the findings.⁹ Our third approach was more novel than the other two. Here, we drew a random sample of 10% of the aggravated assaults (n = 2,794) to obtain a smaller number of cases to analyze. Because no set standard exists for handling large sample sizes, we discuss all three of these approaches in interpreting our findings. As Table 1 indicates, similar frequencies between the sample and the population were obtained. For the aggravated assault logistic regression, the stability of the coefficients and statistical significance were assessed for the sample by conducting an additional nine randomly obtained samples from the aggravated assault population. All the results presented concern analyses that used the full population and the initial 10% sample.

Findings

Aggravated Assaults Between Intimate Partners

In looking at the frequencies for all victims of aggravated assault (Table 1), the majority were female (75.2%), White (60.4%), and approximately 34 years of age (SD = 11.1 years). For the offenders overall, most were White (55.3%), 34.5 years of age (SD = 10.7 years), and did not use alcohol or drugs (78.5%). The most common weapon type was personal contact (34.5%). The majority of all the assaults occurred inside the home (80.3%), were cleared (65.6%), involved a dating relationship (62.9%), and resulted in minor injuries (45.9%).

Different characteristics were observed when the victim’s sex was taken into account (see Table 2).¹⁰ Here, the majority of female victims were younger (women = 32.5 years, men = 37 years; t = −30; p < .001) and White (women = 64.4%, men = 51.6%; X² = 367.9; p < .001) when compared to male victims. Female victims tend to be assaulted by offenders who are White (women = 57.3%, men = 52.2%; X² = 54.4; p < .001) and have been using alcohol or drugs (women = 22.6%, men = 18.3%; X² = 59.7; p < .001). Most of these male offenders used a firearm (women = 10.9%, men = 5.0%) or personal contact weapons (women = 42.8%, = 11.3%; X² = 3535.6; p < .001). In contrast, male victims of aggravated assaults typically are attacked with a knife (men = 43.1%, women = 16.1%), blunt object (men = 18.4%, women = 12.6%), or other weapon (men = 22.2%, women = 17.7%; X² = 3535.6; p < .001).

Table 2.

Characteristics of Aggravated Assaults and Murders by Victim Sex [% (N) or M (SE)].

	Aggravated Assault
	Population		Sample		Murder
Characteristic	Female Victim (N = 21,466)	Male Victim (N = 7,056)	Female Victim (n = 2,122)	Male Victim (n = 671)	Female Victim (N = 217)	Male Victim (N = 50)
Victim age	32.5*** (0.0)	37.0 (0.1)	32.3*** (0.2)	36.7 (0.5)	41.9 (1.1)	44.0 (2.0)
Victim race
Non-White	35.6*** (7,515)	48.4 (3,384)	33.2*** (693)	46.8 (313)	24.0** (52)	44.0 (22)
White	64.4 (13,624)	51.6 (3,601)	66.8 (1,393)	53.2 (356)	76.0 (165)	56.0 (28)
Offender age	34.7 (0.1)	33.9 (0.1)	34.5 (0.2)	33.8 (0.4)	44.1 (1.1)	42.1 (2.0)
Offender race
Non-White	42.7*** (9,057)	47.8 (3,337)	40.2** (843)	46.8 (311)	27.8* (60)	46.0 (23)
White	57.3 (12,140)	52.2 (3,647)	59.8 (1,254)	53.2 (354)	72.2 (156)	54.0 (27)
Offender used alcohol/drugs
Not applicable	77.4*** (16,612)	81.7 (5,768)	76.5* (1,623)	80.6 (541)	88.5 (192)	78.0 (39)
Yes	22.6 (4,854)	18.3 (1,288)	23.5 (499)	19.4 (130)	11.5 (25)	22.0 (11)
Offender weapon use
Firearm	10.9*** (2,300)	5.0 (347)	12.3*** (256)	5.7 (38)	53.1*** (110)	44.9 (22)
Knife	16.1 (3,402)	43.1 (3,003)	16.0 (335)	44.4 (296)	17.9 (37)	44.9 (22)
Blunt object	12.6 (2,673)	18.4 (1,283)	11.3 (237)	18.0 (120)	4.3 (9)	2.0 (1)
Personal contact	42.8 (9,053)	11.3 (791)	42.1 (880)	10.4 (69)	13.0 (27)	0 (0)
Other	17.7 (3,736)	22.2 (1,550)	18.2 (381)	21.5 (143)	11.6 (24)	8.2 (4)
Location of assault
Outside home	19.8 (4,255)	19.5 (1,375)	19.1 (406)	20.6 (138)	19.4 (42)	16.0 (8)
Inside home	80.2 (17,211)	80.5 (5,681)	80.9 (1,716)	79.4 (533)	80.6 (175)	84.0 (42)
Cleared
No	35.1*** (7,525)	32.4 (2,287)	33.7 (716)	29.8 (200)	25.8 (56)	18.0 (9)
Yes	64.9 (13,941)	67.6 (4,769)	66.3 (1,406)	70.2 (471)	74.2 (161)	82.0 (41)
Intimate partner relationship
Dating	63.6*** (13,650)	60.9 (4,294)	64.1 (1,360)	62.7 (421)	41.0 (89)	48.0 (24)
Spouse	36.4 (7,816)	39.1 (2,762)	35.9 (762)	37.3 (250)	59.0 (128)	52.0 (26)
Injury
None	30.7 (6,591)	29.7 (2,095)	30.7 (652)	28.2 (189)	n/a	n/a
Minor	45.6 (9,797)	46.7 (3,293)	46.0 (976)	45.9 (308)	n/a	n/a
Serious	23.7 (5,078)	23.6 (1,668)	23.3 (494)	25.9 (174)	n/a	n/a

p < .05. **p < .01. ***p < .001.

With regard to the multivariate model for aggravated assault, most of our findings are consistent regardless of which of the three methodologies we used to interpret significance for this large sample. First, if we use a smaller p value (p < .001), we find that all but two of the predictors remain statistically significant. These two are home location and clearance. Next, we examined the substantive or clinical significance. The standard to use here is not as clear as selecting a smaller p value. If we choose to use the predictors closest to 1.0 (which suggests less substantive significance), we find that those values are associated with age (both victim and offender) and clearance. Finally, if we look at the sample data,¹¹ all of the predictors that are statistically significant in the population model are also significant in the sample model, with the exception of dating relationship (see Table 3). All of these findings taken together suggest that the majority of the relationships between the predictors and victim sex are not driven by the large sample size. The relationships for home location, clearance, and dating relationship are the ones that should be interpreted with caution. Given these consistent findings, our discussion uses the full “population” model.

Table 3.

Logistic Regression Model Predicting Male Victims for Aggravated Assaults and Murders.

	Aggravated Assault
	Population			Sample			Murder
Characteristic	beta	SE	Exp(β)	beta	SE	Exp(β)	beta	SE	Exp(β)
Victim age	0.11	0.003	1.11***	0.10	0.01	1.11***	0.08	0.03	1.09**
White victim	−1.00	0.06	0.37***	−1.01	0.19	0.36***	−0.95	0.63	0.39
Offender age	−0.10	0.003	0.91***	−0.09	0.01	0.91***	−0.07	0.03	0.93**
White offender	0.71	0.06	2.04***	0.62	0.19	1.86**	0.44	0.63	1.55
Offender used alcohol/drugs	−0.32	0.04	0.73***	0.27	0.13	0.77*	1.14	0.47	3.13*
Offender used firearm^a,b	0.61	0.07	1.84***	0.73	0.23	2.09**	−1.05	0.41	0.35*
Offender used knife^a	2.31	0.05	10.06***	2.41	0.16	11.18***	n/a (ref. group)	n/a (ref. group)	n/a (ref. group)
Offender used blunt object^a	1.69	0.05	5.42***	1.86	0.18	6.41***	n/a (included w/ other)	n/a (included w/ other)	n/a (included w/ other)
Offender used other weapon^a,b	1.59	0.05	4.90***	1.55	0.17	4.73***	−2.25	0.59	0.11***
Home location	−0.12	0.04	0.89**	−0.29	0.13	0.75*	0.14	0.49	1.15
Cleared	0.11	0.03	1.12**	0.27	0.16	1.30*	0.37	0.46	1.45
Dating relationship	−0.19	0.03	0.83***	−0.06	0.11	0.95	−0.27	0.40	0.77
Minor injury^c	0.25	0.05	1.28***	0.31	0.12	1.36*	n/a	n/a	n/a
Serious injury^c	0.32	0.09	1.38***	0.63	0.15	1.87***	n/a	n/a	n/a
−2 log likelihood	24199.99			2325.74			209.6
Model χ²	6422.01***			638.02***			39.53***
Hosmer and Lemeshow	28.94***			3.76			4.97
Nagelkerke R²	0.31			0.32			0.23
N	27,304			2,674			254

Compared to personal weapons for aggravated assaults.

Compared to knife for murders.

Compared to no injury.

p < .05. **p < .01. ***p < .001.

In reviewing the population model presented in Table 3, the model predicts victim sex, which was coded 1 for men and 0 for women. As discussed above, this approach permits us to directly compare the predictors for victim sex for non-lethal and lethal IPV. Because we have restricted the sample to heterosexual relationships, the offender is of the opposite sex. Thus, the victim is more likely to be male (and the offender female) when the adjusted odds ratio (AOR) is greater than 1. Conversely, female victims (and male offenders) are more likely when the AOR is less than 1. With regard to demographics of the victims and offenders, differences are observed. Male victims are less likely to be younger and White than female victims of IPV (AOR = 1.11 and .37, respectively). Male victims are more likely to be assaulted by female offenders who were White and older (AOR = 2.04 and .91, respectively).

Certain incident characteristics also predict male rather than female victims. These characteristics include clearance, victim injury, and use of a weapon. Clearance is more likely to occur for male rather than female victims (AOR = 1.12). Injury (either minor or serious) as compared to no injury is more likely to occur for male rather than female victims (AOR = 1.38 and 1.28, respectively). A finding that may be related to injury is that weapons are more likely to be used against male victims than female victims. For instance, female-on-male aggravated assaults are twice as likely to involve a firearm, approximately 5 times as likely to involve an “other” weapon or blunt object, and more than 11 times as likely to involve a knife when compared to male-on-female aggravated assaults (AOR = 1.84, 5.42, and 10.06, respectively). Other incident characteristics predict female victimization. Assaults in the home and by an offender who used alcohol or drugs are more likely to occur for female rather than male victims (AOR = .89 and .73, respectively).

Murders Between Intimate Partners

In looking at the frequencies for all murder victims (Table 1), the majority were female (81.3%), White (72.3%), and approximately 42 years of age on average (SD = 15.4 years). Most offenders were White (72.3%), did not use alcohol or drugs (86.5%), and used a firearm (49.4%). Murders typically occurred inside the home (81.3%), were cleared (75.7%), and involved a spouse (57.7%).

Different characteristics can be observed when the victim’s sex is taken into account (Table 2). Female murder victims are more likely to be White as compared to male victims (76.0% and 56.0%, respectively; X² = 8.14; p = .004). More female victims were killed by male offenders who were White (72.2% and 54.0%, respectively; X² = 6.28; p = .012) and used firearms (53.1% and 44.9%, respectively), blunt objects (4.3% and 2.0%, respectively), personal contact (13.0% and 0%, respectively), and other weapon (11.6% and 8.2%, respectively; X² = 20.43; p < .001). With respect to male victims, female offenders typically used a knife (44.9% and 17.9%, respectively; X² = 20.43; p < .001).

When controlling for other variables, the statistically significant predictors of the victim’s sex included age (both victim and offender), offender’s use of alcohol or drugs, offender’s use of a firearm, and offender’s use of other weapons (Table 3). Male victims were more likely to be older than were female victims of murder (AOR = 1.09), and male victims were more likely to be killed by older offenders than were female victims (AOR = .93). Firearms (AOR = .35) and other weapons (AOR = .11) rather than knives were more likely to be used in murders of female rather than male victims. The use of alcohol or drugs by the offender was almost three times (AOR = 3.13) more likely to be associated with male rather than female victims.

Discussion

Our study hypothesized that we would observe different characteristics for male and female victims of aggravated assault and murder involving their intimate partners. With regard to our specific research questions and hypotheses, our first question concerned whether victim, offender, and incident characteristics differentially predict the sex of the victim. We hypothesized that we would observe variations by victim sex based on previous research and the IPV typology framework described by Johnson and his colleagues. To explore this question and hypothesis, we focused on the main effects identified in the multivariate analyses (Table 3) for sex across both crimes, specifically those predictors that are statistically significant in both aggravated assault and murder models and in the same direction. Here, we found that victim and offender age are significant predictors of victim sex for both aggravated assault and murder. The presence of more age-mature IPV couples might be suggestive of “common couple” violence in which the couple is more prone to report to the police due to their age (Felson et al., 2002) or “patriarchal terrorism” in which an abusive relationship develops over time before coming to the attention of the police. In addition, although both female and male victims and offenders of IPV are middle-age, male victims and offenders are older than their female counterparts. This pattern may reflect a tendency for these offenses to be dominated by more traditional relationships in which the male is older than the female partner. Outside of the victim and offender age, no other statistically significant main effects were observed for the other victim and offender demographics or for the incident characteristics.

Our second research question builds upon these findings by examining the predictors of victim sex within the context of crime lethality. Here, we also hypothesized that sex differences would be observed based on Johnson and colleagues’ typological framework. To address this question, we again focused on the multivariate analyses. Here we were interested in examining the interaction effects between sex and crime. These effects were reflected in those predictors that are statistically significant in both the aggravated assault and murder models but in the opposite direction. Our multivariate model identified sex differences across lethality (Table 3) with regard to both offender and incident characteristics. No statistically significant differences were observed across lethality for victim or offender demographics.

For offender characteristics, we found that the offender’s use of alcohol or drugs was significantly more likely to predict female victimization in aggravated assaults but male victimization in murders. This divergent finding has support in the existing literature and indicates different underlying typologies from Johnson and his colleague’s work (Johnson, 1995; Johnson & Ferraro, 2000). The finding that alcohol or drug use by male offenders is more predictive of female aggravated assault victimization is similar to previous studies (Field & Caetano, 2005) and may reflect a precipitory condition to the “patriarchal terrorism” IPV typology. The use of drugs or alcohol may reduce an individual’s ability to control his behavior. With regard to our finding for murders, female offender alcohol or drug use predicts male victimization. This pattern appears to be consistent with previous research indicating that alcohol use in female-on-male violence is associated with women who have “a significant amount of stress through their lifetime” and who “experience a broad spectrum of difficulties” (Caetano et al., 2005, p. 402). An extension of this explanation would suggest that these murders may reflect drug or alcohol use as a precipitory condition in the “mutual violent control” typology or a coping mechanism in the “violent resistance” typology. Two caveats should be noted in interpreting these findings with regard to drug and alcohol use. One is the large sample size for the aggravated assault models (both the overall and subsample). The second caveat is the measurement issues previously discussed concerning the variable measuring offender’s use of drugs and alcohol.

As we indicated above, although we were unable to discern specific typologies from our data, we can use this framework to place our findings concerning certain incident characteristics into a preliminary context. As Table 3 shows, type of weapon is an important incident characteristic that varied by sex and across lethality. For non-lethal IPV, male victims are significantly more likely than females to be assaulted with weapons involving various instruments such as firearms, knives, blunt objects, and other weapons as compared to personal weapons. For example, male victims as compared to female victims are almost 11 times more likely to be attacked with a knife rather than a personal weapon. For lethal IPV, a slightly different picture emerges. Here, female victims as opposed to males are significantly more likely to be killed with either a firearm or other weapon when compared to a knife. The range of weapon use against male victims might be suggestive of a female offender trying to locate a weapon of convenience for self-defense in a “violent resistance” situation. In contrast, a male offender’s intent for lethal violence may be evident in his use of a firearm and this intent might be more reflective of a general tendency toward violence as suggested by the “patriarchal terrorism” typology.

The logical extension of weapon use is the amount of injury the victim sustains. This relationship is reflected in our findings and also can be explored using Johnson and Ferraro’s typologies. For aggravated assaults, we found that male victims were more likely to sustain both minor and serious injuries compared to female victims with serious injury having a slightly stronger effect (AOR = 1.36 and 1.87, respectively). For murder, we found that the majority of victims were female (81.3%). This finding may suggest that the minor injuries of male victims are reflective of common couple violence whereas the serious injuries and relatively few murders of male victims are reflective of violent resistance by the female offender to protect herself. The larger number of female murders might reflect the progression of patriarchal terrorism.

These findings—particularly those for weapon use—provide mixed support for our hypothesis that we would observe sex differences when looking across lethal and non-lethal IPV. Variation by type of weapon appears to be the most consistent finding based on both our bivariate and multivariate analyses. Male victims appear to be attacked mainly by knives for both fatal and non-fatal incidents. Female victims, conversely, are attacked with distinctly different weapons for aggravated assaults (personal weapons) as compared to murders (firearms). These patterns are worth some additional discussion of possible explanations.

One explanation for the different weapon patterns is that different underlying processes may be at work for male and female IPV victims. Since the same weapon tends to be used for both lethal and non-lethal IPV involving male victims, this pattern suggests a similar underlying process and a continuum of violence, such as suggested by Johnson and Ferraro’s violent resistance. The different result may occur more due to chance (such as proximity to medical care) than the female offender’s intent for self-protection. In contrast, the difference in weapon type for female aggravated assault and murder victims suggests distinct underlying processes for these crimes such as evident by Johnson’s common couple and patriarchal terrorism violence. That is, aggravated assaults might be capturing more common couple violence in which arguments simply “get out of hand,” whereas murders are more likely the result of patriarchal terrorism in which male domination escalates over time, resulting in death. If this preliminary observation is accurate, different explanations for IPV based on victim sex are needed as well as more tailored prevention and assistance programs. For example, if the etiology of female non-fatal IPV victims differs from those of fatal IPV and if this violence escalates over time, service providers have an opportunity to intervene. Police can play an important role in identifying at-risk women based on calls for service. Policies for male victims also could involve such interventions, but given the continuum of violence within the incident, prevention may play a greater role in a holistic approach to male IPV.

A second explanation for the weapon patterns we observed focuses on reporting behaviors. Here, a male victim might report to police only if he feels that an incident is “serious” enough for him to be a “credible” victim. As such, a male victim might come forward only if he has been attacked by a weapon or injured in some way, and he may be reluctant to report if he is assaulted by a female using her hands or other personal weapons.¹² Women may be more willing to report IPV in seemingly less serious forms, but even for female victims, more incidents are reported when the offender is intoxicated (on alcohol or drugs) and thus may be less “responsible” for the IPV. Socially accepted stereotypes of sex and IPV might affect reporting practices for both men and women. If this underlying explanation is accurate, it suggests more outreach by police departments and community-based IPV services to promote IPV reporting for all victims. Primary prevention programs aimed at educating adolescents and the larger community about IPV might assist in reducing the stereotypes of sex and IPV. Further research is needed to explore the veracity of these explanations.

Although our study provides several important insights with regard to differences across IPV victim sex, it is not without limitations. Whereas police data have many limitations (e.g., Barnett-Ryan, 2007; Buzawa & Hotaling, 2006; Melton & Belknap, 2003), of relevance for our study are two issues. One is the limited contextual information concerning motive, frequency of IPV within the relationship, and related factors that would allow us to identify whether male and female victims experience IPV arising from different typologies as described by Johnson and Ferraro (2000). A second issue is the fact that not all crimes are reported to the police, and this concern is greater with IPV, especially for female victims (Catalano et al., 2009). As a result, caution should be used in extending conclusions from our study to all IPV victims. Although this limitation is important to keep in mind, one interesting pattern in our study is the fact that many of our findings comport with previous studies relying on victimization data.

Three additional limitations stem from our use of NIBRS data in particular. One is that they do not yet have national coverage, which suggests exercising caution in drawing conclusions to non-NIBRS reporting jurisdictions. Another limitation with NIBRS data is that we cannot identify whether repeat victims appear in the file. It is possible that the police repeatedly respond to IPV involving the same couple. This situation could introduce multiple observations of the same couple for which we cannot control. A third limitation with the NIBRS is that we cannot identify whether the victim recorded in the NIBRS records was the original aggressor or not. One concern in examining male victims of IPV is that they may have been the original aggressor and are injured as a result of the initial female victim defending herself. Although these limitations suggest interpreting our findings with some caution, overall our study provides important insights with regard to officially reported serious IPV that warrant further study and exploration.

Conclusion

Much attention has been given to studying IPV, but relatively little has been devoted to understanding sex difference for IPV victims, particularly victims of aggravated assault and murder. In our study, the majority of sex differences are observed when comparing across aggravated assault and murder. This finding emphasizes the importance of accounting for the type of crime in future IPV research, especially lethal and non-lethal incidents. For aggravated assaults, male victims tend to encounter intimate partners who are armed with a weapon, especially a knife, and sustain minor to serious injuries. For murder, female victims rather than male victims are more likely to encounter an offender with a firearm. Determining the roles of victim and offender is notoriously blurred in IPV incidents as the current offender may have been the previous victim in an ongoing violent relationship (see Miller, 2005). Although this issue is important to keep in mind, previous research suggests that not all of male IPV is retaliation (Felson & Lane, 2010; Kruttschnitt & Carbone-Lopez, 2006). Our study highlights the need for further exploration of sex differences, especially for IPV that comes to the attention of police.

Given these findings, our study suggests the importance of acknowledging, and accounting for, sex differences in IPV prevention measures. Our findings provide initial support for tailored policies that address sex differences in the reporting and recording of IPV as well as the availability of victim services. As Johnson (1995) has noted, “advice based on a mistaken assumption [about the underlying dynamics of IPV] may do some women a great disservice” (p. 292). We would suggest that this observation could be extended to include assumptions about victim sex as well. The variations we observed, particularly those concerning seriousness of the incident (in terms of injury and weapon use), indicate initial support for exploring particular policies. For example, men should be encouraged to report minor IPV incidents instead of waiting until their injuries are more serious. IPV services need to tailor their services to the different sexes and recognize that the barriers to service may also differ by sex. Our study also indicates that many IPV characteristics are similar between police and non-police data. Given increased attention to IPV, reporting patterns may have changed over time. Further research should examine whether the undercounting of IPV incidents is consistent across characteristics such that police data obtain comparable cases, just fewer of them.

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.

Notes

Author Biographies

Lynn A. Addington is an associate professor in the Department of Justice, Law & Society at American University in Washington, DC. Her research focuses on violent victimization, the measurement of crime, and utilization of national crime statistics. Her publications have appeared in a range of outlets including the Journal of Quantitative Criminology and Justice Quarterly. She also is the co-editor (with James P. Lynch) of Understanding Crime Statistics: Revisiting the Divergence of the NCVS and UCR (2007, Cambridge University Press).

Suzanne E. Perumean-Chaney is an assistant professor in the Justice Sciences Department at the University of Alabama Birmingham. Her research interests include youth violence, school violence and related policies, homicide, intimate partner violence, media, statistical methods and spatial analysis. She has published in Journal of Statistical Computation and Simulation, Journal of School Health, Security Journal, American Journal of Criminal Justice, and other outlets.

References

Addington

L. A.

(2004). The effect of NIBRS reporting on item missing data in murder cases. Homicide Studies, 8, 193–213.

Addington

L. A.

(2008). Assessing the extent of nonresponse bias on NIBRS estimates of violent crime. Journal of Contemporary Criminal Justice, 24, 32–49.

Addington

L. A.

(2009). Studying the crime problem with NIBRS data: Current uses and future trends. In Krohn

M. D.

Lizotte

A. J.

Hall

G. P.

(Eds.), Handbook on crime and deviance (pp. 23–42). New York: Springer Science+Business Media.

Allison

(2002). Missing data. Thousand Oaks, CA: SAGE.

Archer

(2000). Sex differences in aggression between heterosexual partners: A meta-analytic review. Psychological Bulletin, 126, 651–680.

Bair-Merritt

M. H.

Crowne

S. S.

Thompson

D. A.

Sibinga

Trent

Campbell

(2010). Why do women use intimate partner violence? A systematic review of women’s motivations. Trauma Violence Abuse, 11, 178–189.

Barnett-Ryan

(2007). Introduction to the Uniform Crime Reporting Program. In Lynch

J. P.

Addington

L. A.

(Eds.), Understanding crime statistics: Revisiting the divergence of the NCVS and UCR (pp. 55–92). New York: Cambridge University Press.

Belknap

Larson

Abrams

M. L.

Garcia

Anderson-Block

(2012). Types of intimate partner homicides committed by women: Self-defense, proxy/retaliation, and sexual proprietariness. Homicide Studies, 16, 359–379.

Block

C. R.

Block

(2012). Margo Wilson’s contribution to the Chicago Homicide Dataset: Sexual rivalry and sexual jealousy. Homicide Studies, 16, 404–427.

10.

Block

C. R.

Christakos

(1995). Chicago homicide from the sixties to the nineties: Major trends in lethal violence. In Block

Block

(Eds.), Trends, risks, and interventions in lethal violence: Proceedings of the third annual spring symposium of the homicide research working group (pp. 17–50). Washington, DC: National Institute of Justice.

11.

Bouffard

L. A.

Wright

K. A.

Muftic

L. R.

Bouffard

J. A.

(2008). Gender differences in specialization in intimate partner violence: Comparing the gender symmetry and violent resistance perspectives. Justice Quarterly, 25, 570–594.

12.

Bureau of Justice Statistics. (1997). Implementing the National Incident-Based Reporting System: A project status report. Retrieved October 10, 2011, from http://www.ojp.usdoj.gov/bjs/pub/pdf/inibrs.pdf

13.

Buzawa

E. S.

Buzawa

A. D.

(2008). Courting domestic violence victims: A tale of two cities. Criminology & Public Policy, 7, 671–685.

14.

Buzawa

E. S.

Hotaling

G. T.

(2006). The impact of relationship status, gender, and minor status in the police response to domestic assaults. Victims and Offenders, 1, 323–360.

15.

Caetano

Ramisetty-Mikler

Field

C. A.

(2005). Unidirectional and bidirectional intimate partner violence among White, Black, and Hispanic couples in the United States. Violence and Victims, 20, 393–406.

16.

Caetano

Vaeth

P.A.C.

Ramisetty-Mikler

(2008). Intimate partner violence victim and perpetrator characteristics among couples in the United States. Journal of Family Violence, 23, 507–518.

17.

Capaldi

D. M.

Kim

H. K.

Shortt

J. W.

(2007). Observed initiation and reciprocity of physical aggression in young, at-risk couples. Journal of Family Violence, 22, 101–111.

18.

Capaldi

D. M.

Shortt

J. W.

Kim

H. K.

Wilson

Crosby

Tucci

(2009). Official incidents of domestic violence: Types, injury, and associations with nonofficial couple aggression. Violence and Victims, 24, 502–519.

19.

Catalano

(2007). Intimate partner violence in the United States. Washington, DC: Bureau of Justice Statistics, U.S. Department of Justice and Office of Justice Programs.

20.

Catalano

Smith

Snyder

Rand

(2009). Female victims of violence. Washington, DC: Bureau of Justice Statistics, U.S. Department of Justice and Office of Justice Programs.

21.

Chan

K. L.

(2011). Gender differences in self-reports of intimate partner violence: A review. Aggression and Violent Behavior, 16, 167–175.

22.

Cho

Wilke

D. J.

(2010). Gender differences in the nature of the intimate partner violence and effects of perpetrator arrest on revictimization. Journal of Family Violence, 25, 393–400.

23.

Cohen

West

S. G.

Aiken

L. S.

(2003). Applied multiple regression/correlation analysis for the behavioral sciences (3rd ed.). Mahwah, NJ: Lawrence Erlbaum Associates.

24.

Cunradi

C. B.

Caetano

Schafer

(2002). Socioeconomic predictors of intimate partner violence among White, Black, and Hispanic couples in the United States. Journal of Family Violence, 17, 377–389.

25.

Dichter

M. E.

Marcus

S. C.

Morabito

M. S.

Rhodes

K. V.

(2011). Explaining the IPV arrest decision: Incident, agency, and community factors. Criminal Justice Review, 36, 22–39.

26.

Dobash

R. E.

Dobash

R. P.

(1998). Rethinking violence against women. Thousand Oaks, CA: SAGE.

27.

Durose

M. R.

Harlow

C. W.

Langan

P. A.

Motivans

Rantala

R. R.

Smith

E. L.

Constantin

(2005). Family violence statistics: Including statistics on strangers and acquaintances. Washington, DC: U.S. Department of Justice.

28.

Eitle

(2005). The influence of mandatory arrest polices, police organizational characteristics, and situational variables on the probability of arrest in domestic violence cases. Crime & Delinquency, 51, 573–597.

29.

Federal Bureau of Investigation. (2004). Uniform Crime Reporting handbook. Retrieved from http://www.fbi.gov/ucr/handbook/ucrhandbook04.pdf

30.

Felson

R. B.

Lane

K. J.

(2010). Does violence involving women and intimate partners have a special etiology? Criminology, 48, 321–338.

31.

Felson

R. B.

Messner

S. F.

Hoskin

A. W.

Deane

(2002). Reasons for reporting and not reporting domestic violence to the police. Criminology, 40, 617–647.

32.

Felson

R. B.

Pare

(2005). The reporting of domestic violence and sexual assault by nonstrangers to the police. Journal of Marriage and the Family, 67, 597–610.

33.

Field

C. A.

Caetano

(2005). Longitudinal model predicting mutual partner violence among White, Black, and Hispanic couples in the United States general population. Violence and Victims, 20, 499–511.

34.

Fleury

R. E.

Sullivan

C. M.

Bybee

D. I.

Davidson

W. S.

II. (1998). Why don’t they just call the cops? Reasons for differential police contact among women with abusive partners. Violence and Victims, 13, 333–346.

35.

Halpern

C. T.

Oslak

S. G.

Young

M. L.

Martin

S. L.

Kupper

L. L.

(2001). Partner violence among adolescents in opposite-sex romantic relationships: Findings from the National Longitudinal Study of Adolescent Health. American Journal of Public Health, 91, 1679–1685.

36.

Henning

Feder

(2004). A comparison of men and women arrested for domestic violence: Who presents the greater threat? Journal of Family Violence, 19, 69–80.

37.

Hirschel

Buzawa

(2002). Understanding the context of dual arrest with directions for future research. Violence Against Women, 8, 1449–1473.

38.

Hirschel

Hutchinson

(2011). Unraveling the relative contributions of his, her, and their drinking to the likelihood of arrest in intimate partner violence cases. Journal of Interpersonal Violence, 26, 3050–3079.

39.

Johnson

M. P.

(1995). Patriarchal terrorism and common couple violence: Two forms of violence against women. Journal of Marriage and the Family, 57, 283–294.

40.

Johnson

M. P.

Ferraro

K. J.

(2000). Research on domestic violence in the 1990s: Making distinctions. Journal of Marriage and the Family, 62, 948–963.

41.

Justice Research and Statistics Association. (n.d.-a). 50 largest NIBRS agencies. Retrieved July 4, 2011, from http://www.jrsa.org/ibrrc/index.html

42.

Justice Research and Statistics Association. (n.d.-b). Status of NIBRS in the states. Retrieved July 4, 2011, from http://www.jrsa.org/ibrrc/background-status/nibrs_states.shtml

43.

Kernsmith

Craun

S. W.

(2008). Predictors of weapon use in domestic violence incidents reported to law enforcement. Journal of Family Violence, 23, 589–596.

44.

Kruttschnitt

Carbone-Lopez

(2006). Moving beyond the stereotypes: Women’s subjective accounts of their violent crime. Criminology, 44, 321–351.

45.

Langton

(2011). Use of victim services agencies by victims of serious violent crime, 1993–2009. Washington, DC: Bureau of Justice Statistics.

46.

Lehmann

E. L.

(2004). Elements of large-sample theory. New York, NY: Springer.

47.

Malloy

K. A.

McCloskey

K. A.

Grigsby

Gardner

(2003). Women’s use of violence within intimate relationships. Journal of Aggression, Maltreatment, & Trauma, 6, 37–59.

48.

Maltz

M. D.

(1999). Bridging gaps in police crime data: A discussion paper from the BJS Fellows Program. Washington, DC: Bureau of Justice Statistics.

49.

Melton

Belknap

(2003). He hits, she hits: Assessing gender differences and similarities in officially reported intimate partner violence. Criminal Justice and Behavior, 30, 328–348.

50.

Melton

H. C.

Sillito

C. L.

(2011). The role of gender in officially reported intimate partner abuse. Journal of Interpersonal Violence, 27, 1090–1111.

51.

Miethe

T. D.

Regoeczi

W. C.

(2004). Rethinking homicide: Exploring the structure and process underlying deadly situations. Cambridge, UK: Cambridge University Press.

52.

Miller

S. L.

(2005). Victims as offenders: The paradox of women’s violence in relationships. New Brunswick, NJ: Rutgers University Press.

53.

National Archive of Criminal Justice Data. (2010). National Incident-Based Reporting System, 2008: Extract files [Computer file]. Ann Arbor, MI: Inter-university Consortium for Political and Social Research. doi:10.3886/ICPSR27741

54.

Pattavina

Hirschel

Buzawa

Faggiani

Bentley

(2007). A comparison of the police response to heterosexual versus same-sex intimate partner violence. Violence Against Women, 13, 374–395.

55.

Regoeczi

W. C.

Jarvis

J. P.

Riedel

(2008). Clearing murders: Is it about time? Journal of Research in Crime and Delinquency, 45, 142–162.

56.

Rennison

C. M.

(2001). Intimate partner violence and age of victim, 1993–99. Washington, DC: Bureau of Justice Statistics.

57.

Rennison

C. M.

(2003). Intimate partner violence 1993–2001. Washington, DC: Bureau of Justice Statistics.

58.

Rennison

C. M.

Welchans

(2000). Intimate partner violence. Washington, DC: Bureau of Justice Statistics.

59.

Roberts

(2009). Contributions of the National Incident-Based Reporting System to substantive knowledge in criminology: A review of research. Sociology Compass, 3, 433–458.

60.

Smith

E. L.

Farole

D. J.

(2009). Profile of intimate partner violence cases in large urban counties. Washington, DC: Bureau of Justice Statistics.

61.

Straus

M. A.

(2011). Gender symmetry and mutuality in perpetration of clinical-level partner violence: Empirical evidence and implications for prevention and treatment. Aggression and Violent Behavior, 16, 279–288.

62.

Straus

M. A.

Gelles

R. J.

(1986). Societal change and change in family violence from 1975 to 1985 as revealed by two national surveys. Journal of Marriage and the Family, 48, 465–479.

63.

Swatt

M. L.

(2006). Exploring the difference between male and female intimate partner homicides: Revisiting the concept of situated transactions. Homicide Studies, 10, 279–292.

64.

Thompson

M. P.

Kingree

J. B.

(2006). The roles of victim and perpetrator alcohol use in intimate partner violence outcomes. Journal of Interpersonal Violence, 21, 163–177.

65.

Thompson

M. P.

Saltzman

L. E.

Bibel

(1999). Applying NIBRS data to the study of intimate partner violence: Massachusetts as a case study. Journal of Quantitative Criminology, 15, 163–180.

66.

Tjaden

Thoennes

(2000). Extent, nature, and consequences of intimate partner violence. Washington, DC: National Institute for Justice.

67.

Vazquez

S. P.

Stohr

M. K.

Purkiss

(2005). Intimate partner violence incidence and characteristics: Idaho NIBRS 1995 to 2001 data. Criminal Justice Policy Review, 16, 99–114.

68.

Warner

T. D.

(2010). Violent acts and injurious consequences: An examination of competing hypotheses about intimate partner violence using agency-based data. Journal of Family Violence, 25, 183–193.