Arrest History and Intimate Partner Violence Perpetration in a Sample of Men and Women Arrested for Domestic Violence.

Intimate partner violence (IPV) is a serious and prevalent problem throughout the United States. Currently, individuals arrested for domestic violence are often court mandated to batterer intervention programs (BIPs). However, little is known about the arrest histories of these individuals, especially women. The current study examined the arrest histories of men (n = 303) and women (n = 82) arrested for domestic violence and court-referred to BIPs. Results demonstrated that over 30% of the entire sample had been previously arrested for a non-violent offense, and over 25% of the participants had been previously arrested for a violent offense other than domestic violence. Moreover, men were arrested significantly more frequently for violence-related and non-violent offenses than their female counterparts. In addition, men were more likely than women to have consumed binge-levels of alcohol prior to the offense that led to their most recent arrest and court-referral to a BIP. Lastly, arrest history was positively associated with physical and psychological aggression perpetration against an intimate partner for men only, such that more previous arrests were associated with more frequent aggression. These results provide evidence that many men and women arrested for domestic violence have engaged in a number of diverse criminal acts during their lifetimes, suggesting that BIPs may need to address general criminal behavior.

INTRODUCTION

Intimate partner violence (IPV) is a prevalent and serious problem throughout the United States and the world. This form of violence knows no boundaries, as people of all racial, religious, sexual, and minority and majority groups are affected by it each year. Thus, it is crucial to conduct research that aims to better understand the characteristics and life histories of perpetrators of IPV. Individuals who perpetrate the most severe IPV are often court-referred to batterer intervention programs (BIPs), which are designed to reduce the recidivism of IPV. Unfortunately, these programs have questionable utility in reducing IPV (Babcock, Green, and Robie 2004; Stuart, Temple, and Moore 2007). Therefore, continued research is needed to better understand the individuals mandated to these programs, as this could help to inform more effective interventions. The purpose of the present study was to examine the arrest histories of men and women arrested for domestic violence and court-referred to BIPs. Knowledge about whether these individuals commit crimes in addition to IPV may signal a need for BIPs to also focus on reducing general criminal behavior. Moreover, knowing whether men and women differ on arrest histories could help BIP providers to determine the necessity of employing gender-specific interventions.

IPV: Prevalence and Treatment

IPV consists of physical, psychological, sexual, and stalking behaviors directed by one partner toward another (Shorey, Cornelius, and Bell 2008). While IPV can consist of these unique forms of aggressive behavior, the most prevalent forms of aggression are physical and psychological (Archer 2000), and are the types of IPV that were examined in the current study. Physical aggression consists of behaviors such as slapping, kicking, shoving, punching, or using a weapon against a partner (Straus ). In contrast, psychological aggression consists of verbal and behavioral acts directed against a partner that often diminish one’s self-worth and produce fear (Follingstad 2007), and includes acts such as name calling or swearing at one’s partner, threatening aggression, and breaking or destroying a partner’s personal belongings (Straus ). The yearly prevalence of IPV is estimated to be 20–30% for physical aggression and 70–90% for psychological aggression (Lawrence ). In addition, mounting research clearly demonstrates that women perpetrate as much or more acts of IPV compared to their male counterparts (Archer 2000; Leisring 2011).

Despite an abundance of research in the past thirty years on the characteristics of individuals who perpetrate aggression, efforts aimed at reducing IPV have had challenges in terms of reducing recidivism (Stuart ). Men and women who are arrested for domestic violence in the United States are often court-mandated to attend BIPs. While it is the aim of these programs to prevent violence recidivism, research on their effectiveness at achieving this aim is questionable. Two recent meta-analyses on the effectiveness of BIPs have been conducted, with both demonstrating only small improvements in violence recidivism (Babcock ; Feder and Wilson 2005). The meta-analysis that included the 10 most methodologically rigorous studies found that the effect size for BIPs, depending on the source of outcome data, ranged from d = 0.00 – 0.26 (Feder and Wilson 2005). Thus, it is clear that BIPs have substantial room for improvement.

Criminal Behavior among Perpetrators of IPV

There has been some research to date on the criminal behavior of male batterers court-mandated to BIPs. For instance, using a sample of 4,032 male abusers, Maxwell and colleagues (2001) found that 40% of their sample had a prior criminal history; other researchers have found similar rates among male batterers (Baba ; Ventura and Davis 2005). Klein and Tobin (2008) followed a sample of male batterers (n = 342) for nine years, finding that over the nine year period the sample of men had been arrested for non-domestic violence related offenses for a cumulative total of 632 times. Thus, male batterers often have extensive criminal histories and arrest records, indicating that a general delinquent way of interacting with the world may be common for many of these men.

While male batterers are often generally criminally aberrant, the research is less clear for women in BIPs. Babcock, Miller, and Siard (2003) asked a sample of women in BIPs (N = 60) about their previous arrest histories, finding that 62% reported a prior arrest, although they did not specify whether previous arrests were domestic violence or a non-domestic violence arrest(s). Dowd and colleagues (2005), using a sample of 107 domestically violent women who were mandated to anger management, found that over 70% reported being arrested at least once during adulthood, although they did not specify what criminal behavior led to being arrested. They did report, however, that 8.6% of the sample had been previously arrested for driving under the influence (DUI). Research also suggests that a large percentage of women court-mandated to BIPs perpetrate aggression against non-intimates (Shorey ; Stuart ), suggesting that some of these women may be generally aggressive, increasing their risk for arrests for non-domestic violence related offenses. Thus, preliminary research suggests that women referred to BIPs or other treatment programs may also have extensive criminal backgrounds and general aggressive tendencies, although more information regarding the criminal behaviors that lead to arrest (e.g., violent-related, substance-related, non-domestic violence) is needed.

In addition to examining the arrest histories of men and women arrested for domestic violence, and specifying the crimes that led to arrest, research would benefit from examining whether men and women arrested for domestic violence differ in their previous arrest histories. Some researchers have argued that female BIP programs should be specifically tailored for women, as they often differ substantially from their male counterparts on many personal characteristics, such as childhood abuse and trauma histories (Dowd and Leisring 2008). Thus, given potential differences in life histories of men and women in BIPs, it is possible that women court-referred to BIPs are less generally criminally delinquent than their male counterparts. However, some researchers have argued that men and women in BIPs may be more similar than dissimilar (Busch and Rosenberg 2004; Carney, Buttell, and Dutton 2007). Clearly, research is needed to empirically determine whether men and women in BIPs differ on arrest histories, which may help to inform BIPs. That is, if one or both genders are found to have extensive criminal histories, BIPs may want to consider addressing more general criminal behavior in addition to domestic violence.

Current Study

Due to little research on the criminal arrest histories of men and women court-referred to BIPs, and increasing evidence that perpetrators of IPV may have general aggressive and criminal tendencies, the current study examined the prevalence of lifetime arrests for different types of criminal behavior (e.g., substance-related; violence-related) among a sample of men and women arrested for domestic violence and court-referred to BIPs. In addition, we examined whether men and women differed in their criminal arrest histories and whether history of arrests was associated with more frequent psychological and physical aggression perpetration. To our knowledge, this is the first study to examine differences in arrest histories of men and women arrested for domestic violence. We hypothesized that a large percentage of our sample (i.e., > 30%) would have histories of non-domestic violence related arrests. We also hypothesized that those with more previous arrests would report more frequent aggression perpetration. Due to the limited research on the criminal histories of women arrested for domestic violence, we did not have any a priori hypotheses regarding gender differences in arrest histories.

METHOD

Participants

A sample of men (n = 303) and women (n = 82) arrested for domestic violence and court-referred to BIPs in the state of Rhode Island participated in the current study. This sample of men and women represent a subsample of men and women reported on previously (Shorey ; Stuart ; Stuart ). The state of Rhode Island requires mandatory arrests in cases of alleged domestic violence, which can include a wide-range of offenses, such as assault and battery, stalking, harassment, and violation of orders of protection. No information was obtained on the specific reasons why participants were arrested, although it is likely that the participants in the current study were suspected of committing a range of different domestic violence offenses.

Participants reported a mean age of 32.53 years (SD = 10.09), education of 12.0 years (SD = 2.24), which is equivalent to a high school education, and annual income of $31,504 (SD = $22,566). The ethnic composition was as follows: 70.1% white, 12.1% black, 9% Hispanic, 2% American Indian/Alaskan Native, 1.8% Asian/Pacific Islander, and 5.1% other. A few participants did not indicate their race (n = 4). At the time of the study, 24.5% of participants reported being married, 32.4% reported cohabiting and not currently married, 21% were dating, 11.2% were single, 6.1% were separated, 4.6% were divorced, and .2% were widowed. The average length of the participant’s current relationship was 6.13 years (SD = 7.01), length of time living with their current intimate partner was 5.55 years (SD = 6.58), and number of children was 1.86 (SD = 1.83).

Procedure

Participation in the current study was voluntary and all questionnaires were completed during participant’s regularly scheduled BIP sessions. Participants completed the measures of interest in small groups. Groups were open and the mean number of intervention sessions attended prior to participation in the current study was 9.78 (SD = 7.07). The number of intervention sessions attended was unrelated to IPV perpetration (physical and psychological) and history of arrests. None of the information gathered from participants was shared with the intervention facilitators or the criminal justice system. Participants did not receive any compensation for their involvement in the study. Upon providing informed consent, participants completed a packet of questionnaires with a research assistant present to answer any questions. Further information about this study and its procedures can be obtained elsewhere (Stuart ; Stuart ).

Measures

Intimate Partner Violence

The Revised Conflict Tactics Scales (Straus ) was used to assess IPV perpetration in the year prior to coming to the BIP. The CTS2 is the most widely used self-report measure for assessing IPV perpetration. For the current study, only the physical assault (12 items) and psychological aggression (9 items) subscales were examined. Example items for psychological aggression include “Insulted or swore at my partner” and “Threatened to hit or throw something at my partner.” Example items of physical assault include “Slapped my partner” and “Kicked my partner.” Participants indicated on a 7-point scale (0 = “never”; 6 = “more than 20 times”) the number of times they used a particular form of aggression against their intimate partner in the previous year. Scores for each subscale are obtained by taking the midpoint for each item (e.g., “4” for a response of “3 to 5 times”) and then adding the frequency of each of the behaviors for each subscale. Scores for each item could range from 0 to 25 and higher scores are reflective of more frequent aggression perpetration (Straus, Hamby, and Warren 2003). In the present study, internal consistency was .78 for psychological aggression perpetration and .80 for physical aggression perpetration, which are consistent with most research conducted with the CTS2. Both subscales were positively skewed and thus were log-transformed (natural log) prior to statistical analyses.

Arrest History

We created a 9-item questionnaire for the current study that inquired about participants’ prior arrest history. The questions included: (1) Did you have at least one drink of alcohol prior to or during the event that led you to the BIP? (2) Did you have four (for women)/six (for men) or more drinks prior to or during the incident that led you to the BIP? (four drinks for women and six drinks for men correspond to binge levels of alcohol consumption for each gender) (3) Did you feel intoxicated or under the influence of alcohol prior to or during the incident that led you to the BIP? (4) Did you use drugs prior to or during the incident that led you to the BIP? (5) How many times have you been arrested for or charged with an alcohol-related offense? (6) How many times have you been arrested for or charged with a drug-related offense? (7) How many times have you been arrested for or charged with domestic battery, spouse assault, or any other offense against a relationship partner? (8) How many times have you been arrested for or charged with a violence-related offense against someone other than a partner? (9) How many times have you been arrested or charged with any other offense? The first four questions were rated using a yes/no format. Questions 5–9 were rated using a 0–10 or more scale. Examples of the different types of offenses were provided for questions 5–9. The internal consistency for this questionnaire was .75.

RESULTS

All analyses were conducted using SPSS version 18.0. First, we examined the prevalence of previous arrests for men and women, as well as differences between men and women in previous arrests, which are presented in Table 1. Among male participants, 42% indicated that they had at least one drink prior to/during this incident, while 32.5% of men had consumed binge levels of alcohol (defined as 6 or more drinks on one occasion). A similar number of men felt intoxicated prior to/during this incident, while just over 15% had consumed drugs. Women had a similar level of substance use prior to or during the incident that lead them to the BIP, with the exception of binge levels of alcohol. The percentage of women who had consumed binge levels (defined as 4 or more drinks on one occasion) of alcohol (20.2%) was significantly less than their male counterparts (32.5%).

Table 1

Prevalence of Substance Use for Incident that Lead the BIP and Different Types of Prior Arrests

Arrest History	Men (n= 303) %	Women (n = 82) %	χ2 (df), p
1. At least 1 drink prior/during arrest incident	42.0	33.3	2.04 (1), > .05
2. Binge drinking prior/during arrest incident	32.5	20.2	4.70 (1), < .05
3. Felt intoxicated prior/during arrest incident	32.8	25.0	1.86 (1), > .05
4. Used drugs prior/during arrest incident	16.4	13.1	.54 (1), > .05
5. Prior arrest for an alcohol-related offense	36.7	22.6	5.86 (1), < .05
6. Prior arrest for a drug-related offense	38.2	28.6	2.62 (1), > .05
7. Prior arrest for domestic violence against a partner	75.0	67.9	1.72 (1), > .05
8. Prior arrest for a violence-related offense against non-intimate	30.9	15.5	7.86 (1), < .01
9. Prior arrest for any other offense	40.1	26.2	5.48 (1), < .05

Note: Questions 1–4 refer to the arrest that led participants to the BIP.

Among male participants, 36.9% had been previously arrested for an alcohol-related offense and 38.2% for a drug related offense. Further, 75% of men had a prior domestic violence arrest history, with 30.9% with a prior violence-arrest history with someone other than an intimate partner. Finally, 40.1% of men had been previously arrested for any other offense that was not domestic violence or substance use related. As compared to women, men had a greater percentage of prior arrest histories for alcohol related offenses, for domestic violence offenses with someone other than an intimate partner, and for any other offense other than substance use or domestic violence (see Table 1).

We next examined correlations among arrest histories for all previous offenses, arrest histories for non-domestic violence offenses, and the perpetration of physical and psychological aggression for men and women separately. For men, history of prior arrests for any offense was positively associated with psychological aggression (r = .13, p < .05) and physical aggression perpetration (r = .24, p < .001). Further, history of prior arrest for non-domestic violence offenses was positively associated with physical aggression perpetration (r = .19, p < .01), but not psychological aggression perpetration (r = .07, p > .05). For women, history of prior arrests for any offense was not associated with psychological aggression (r = .13, p > .05) or physical aggression perpetration (r = .16, p > .05). Similarly, a history of prior arrest for non-domestic violence offenses was not associated with psychological aggression (r = .13, p > .05) or physical aggression perpetration (r = .16, p > .05).

Finally, we examined whether there were differences in the frequency of psychological and physical aggression perpetration between (1) individuals who consumed substances prior to or during the arrest that led them to the BIP, relative to those with no substance use at the time of the arrest (2) individuals with any prior arrests, relative to those with no arrests, and (3) individuals with any non-domestic violence prior arrests, relative to those with no prior non-domestic violence arrests. We created a dichotomous variable for any substance use prior to or during the arrest that led them to the BIP based on the four questions for substance use (questions 1–4 on the Arrest History questionnaire). That is, if any of the four substance use questions were endorsed, these individuals were placed in the substance use group. The same method was employed for any prior arrests and non-domestic violence arrests. Independent sample t tests were used to examine differences between groups, with men and women examined separately.

Table 2 presents the differences in frequency of violence perpetration for the substance use and non-substance use, and prior arrest and no prior arrest groups. For men, results demonstrated that men with substance use prior to or during the arrest incident that led to the BIP, those with a history of any prior arrests, and those with a history of any prior non-domestic violence arrest reported significantly more frequent psychological and physical perpetration than their respective male counterparts. For women, the only significant difference between groups was for substance use prior to or during the arrest, with women who had consumed substances reporting less frequent physical aggression perpetration than women who had not consumed substances prior to or during the incident that led to the BIP.

Table 2

Differences between Arrest Groups on Frequency of Violence Perpetration

	Men
	Substance Use Group (n= 142)M (SD)	No-Substance Use Group (n= 161)M (SD)	t, p
Psychological Aggression	35.48 (31.32)	25.44 (28.97)	3.58, < .001
Physical Aggression	11.18 (19.50)	5.31 (12.61)	4.25, < .001
	Any Prior Arrest (n= 269)M (SD)	No Prior Arrest (n= 34)M (SD)
Psychological Aggression	32.40 (31.20)	11.76 (14.35)	4.27, < .001
Physical Aggression	8.75 (17.25)	1.97 (2.96)	2.65, < .01
	Non-DV Arrest (n= 218)M (SD)	No Prior Arrest (n= 85)M (SD)
Psychological Aggression	32.15 (31.59)	24.78 (26.86)	2.20, < .05
Physical Aggression	9.52 (18.28)	4.07 (9.18)	2.89, < .01
	Women
	Substance Use Group (n= 32)M (SD)	No-Substance Use Group (n= 50)M (SD)	t, p
Psychological Aggression	42.34 (34.70)	45.56 (41.31)	.17, > .05
Physical Aggression	13.59 (23.15)	24.30 (33.85)	2.02, < .05
	Any Prior Arrest (n= 63)M (SD)	No Prior Arrest (n= 19)M (SD)
Psychological Aggression	47.66 (40.88)	33.15 (28.33)	.83, > .05
Physical Aggression	21.58 (32.72)	15.26 (21.23)	.86, > .05
	Non-DV Arrest (n= 38)M (SD)	No Prior Arrest (n= 44)M (SD)
Psychological Aggression	55.52 (43.28)	34.61 (31.60)	1.74, > .05
Physical Aggression	27.89 (39.05)	13.41 (18.23)	1.73, > .05

Note: Substance use group refers to participants who had used either alcohol or drugs prior to or during the arrest incident that lead them to the BIP; DV = Domestic Violence.

DISCUSSION

Findings from the current study demonstrated that a large percentage of men and women arrested for domestic violence and court-referred to BIPs had previous arrest histories for substance use, domestic violence, and non-violent offenses. These findings add to a growing body of literature on the arrest histories of men court-referred to BIPs (e.g., Baba ; Klein and Tobin 2008; Ventura and Davis 2005), suggesting that a general criminal propensity may be present for many of these men. Further, this study adds to a budding literature on the arrest histories of women court-referred to BIPs (e.g., Babcock ; Dowd, Leisring, and Rosenbaum 2005), being one of the only studies to date to elucidate the specific types of criminal behavior for which these women had been previously arrested.

Our findings also demonstrated that men were significantly more likely than women to have prior arrest histories for a number of offenses, including alcohol-related offenses, domestic violence involving someone other than a partner, and for any offense other than substance use and domestic violence (e.g., disorderly conduct; robbery). One possible explanation for why men had more previous arrests for a number of different offenses could have to do with personality differences between men and women. That is, previous research has demonstrated that men arrested for domestic violence have a high prevalence of antisocial personality traits (e.g., Dutton, Starzomski, and Ryan 1996; Holtzworth-Munroe and Stuart 1994), and antisocial personality is strongly associated with general delinquent and criminal behavior. Although women who have been arrested for domestic violence and court-referred to BIPs also evidence antisocial personality traits (e.g., Shorey ), it is possible that men and women in BIPs differ in the prevalence of antisocial traits. For instance, using a subsample of the men and women from the current study, Stuart and colleagues (2006) found that antisocial personality traits were higher among men than women, and other research has shown antisocial personality traits to be more prevalent among men than women across a number of populations (Cale and Lilienfeld 2002). Future research should therefore examine whether arrest differences between men and women in BIPs may be partly explained by personality differences, namely antisocial personality.

Our findings are fairly consistent with research suggesting that women arrested for domestic violence and court-referred to BIPs may have different life histories and risk factors for violence than their male counterparts (e.g., Dowd and Leisring 2008). For instance, previous research suggests that women in BIPs report greater IPV victimization histories (Stuart ), lower relationship satisfaction (Stuart ), less general violence perpetration (i.e., physical violence against non-intimates) (Stuart ), and may have more extensive trauma histories (Dowd and Leisring 2008) than their male counterparts. Thus, our findings add to a growing body of research suggesting that women in BIPs may be different from their male counterparts in many important ways, including having fewer prior arrests for a number of distinct offenses. This suggests that female-specific BIPs may be needed as opposed to mirroring what is done in BIPs for males (see Dowd and Leisring 2008, for a review of this topic). However, it is also possible that females are less likely to be arrested than their male counterparts despite criminal behavior, which is an empirical question for future research to explore.

Our findings are also consistent with the general theory of crime (Gottfredson and Hirschi 1990), which postulates that a general lack of self-control is responsible for deviant behavior, including aggressive, criminal, and substance use behavior. That is, a lack of self-control is responsible for behavior that satisfies immediate desires at the expense of long-term negative consequences. Previous research has demonstrated that both male and female perpetrators of domestic violence report high levels of impulsivity, low levels of various indicators of self-control (Shorey ; Stuart and Holtzworth-Munroe 2005; Tager, Good, and Brammer 2010), and high levels of problematic substance use (Moore ; Stuart ). Thus, BIPs may be more successful at reducing violence recidivism if they focus on self-control, which may in turn serve to reduce a number of related problematic behaviors (e.g., substance use, general aggression, and criminal behavior). Although it is likely that many programs do, indeed, focus on these associated behaviors, especially substance use, it is possible that these programs may need to focus more heavily on ways to reduce general delinquent behavior in general.

Improving BIPs

The question becomes, then, how treatment providers could enhance BIPs to more effectively reduce domestic violence and other delinquent behaviors (i.e., crime; substance use). Stuart and colleagues (2007) have discussed how BIPs may benefit from incorporating substance use treatment components, since a substantial number of BIP participants meet criteria for a substance use disorder. Indeed, research suggests that substance use treatment results in reductions in IPV perpetration among substance abusers in treatment for addictive behaviors (Stuart, O’Farrell, and Temple 2009). Thus, research is needed to determine whether adding substance use treatment components to BIPs results in reduced IPV and other criminal behaviors.

Another approach could be to implement mindfulness programs in BIPs. Mindfulness is a nonjudgmental, open, and nonreactive awareness of the present moment (Kabat-Zinn 1994). Mindfulness interventions, which include formal meditation practices often rooted in Buddhist traditions, have demonstrated robust improvements across a range of populations, disorders, and problem behaviors (Baer 2003; Keng, Smoski, and Robins 2011). While no known research has examined whether mindfulness interventions reduce IPV, research has demonstrated that mindfulness-based interventions (i.e., Vipassana Meditation) with prison populations increased positive mood, emotional intelligence, less substance use, less trauma symptoms, and less behavioral infractions for prisoners who received the intervention relative to inmates who did not receive the intervention (Perelman ; Simpson ), and less substance use and psychiatric symptoms after release from prison when compared to inmates who did not receive the intervention (Bowen ). Thus, mindfulness interventions may hold promise in reducing IPV and other delinquent behaviors among men and women court-referred to BIPs, although research is needed in this area.

Limitations

The current study has a number of limitations that deserve mention. First, the cross-sectional design prohibits the determination of causality among study variables. Longitudinal research is needed to determine whether involvement in criminal behavior other than domestic violence precedes, co-occurs, or follows the onset of IPV. In addition, our assessment of prior arrests relied on participant self-report, and we cannot rule out that social desirability may have impacted these reports. The use of state and federal arrest records would enhance future research and provide more definitive results concerning prior arrest histories. Our sample of participants was largely non-Hispanic Caucasian in ethnicity, limiting the generalizability to more diverse populations. In addition, we did not have a comparison group of men and women who perpetrated IPV but had never been arrested for domestic violence to compare arrest histories. An interesting question for future research would be to determine whether there are arrest history differences between men and women arrested for domestic violence and men and women who perpetrate IPV but who have not been arrested for it.

In summary, this is the first known study to examine and compare the arrest histories of men and women arrested for domestic violence and court-referred to BIPs. Results suggest that men have more extensive arrest histories than their female counterparts, although a substantial percentage of women had been previously arrested for a variety of offenses. These findings indicate that BIPs may benefit from targeting general delinquent and criminal behavior in their programs, and gender specific programs may be warranted.