2018 POP Conference
November 5–7, 2018
Providence, Rhode Island

Site Menu ▼
To print this guide, click on your web browser's "Print" icon, or go to the menubar and select "File…Print"

Domestic Violence

Guide No. 45 (2006)

by Rana Sampson

Translation(s): Violência Doméstica (Portuguese) PDF

The Problem of Domestic Violence

This guide begins by describing the problem of domestic violence and reviewing factors that increase its risks. It then identifies a series of questions to help you analyze your local domestic violence problem. Finally, it reviews responses to the problem and what is known about these from evaluative research and police practice.

Domestic disputes are some of the most common calls for police service. Many domestic disputes do not involve violence; this guide discusses those that do, as well as the measures that can be used to reduce them. In the United States, domestic violence accounts for about 20 percent of the nonfatal violent crime women experience and three percent of the nonfatal violent crime men experience.[1]Harm levels vary from simple assault to homicide, with secondary harms to child witnesses. Domestic violence calls can be quite challenging for police as they are likely to observe repetitive abuse against the same victims, who may not be able to or may not want to part from their abusers. Police typically view these calls as dangerous, partly because old research exaggerated the risks to police.

Domestic violence is but one aspect of the larger set of problems related to family violence. Related problems not directly addressed in this guide, each of which requires separate analysis, include:

† Much of the recent research about domestic violence refers to the problem as “intimate partner violence.” Mostly this guide keeps to the term domestic violence, not because it is more accurate, but simply because it is still so widely used by police. Also in this guide, the term domestic violence is intended to include violence perpetrated by current and former intimates or dating partners, including those of the opposite or same sex.

†† Originally researchers failed to separate domestic disputes from other types of “disturbance” calls and raw percentages stretched the findings beyond what they reasonably meant (Fridell and Pate, 1997). Of the 713 officers feloniously slain in the United States between 1983 and 1992, 33 percent (235 officers) were slain while intervening in a crime. Of those, 24 percent (56 officers) were slain during a domestic disturbance. In other words, about five officers a year in the United States over that 10-year period were killed during domestic disturbance calls. The frequency of the call likely makes claims of dangerousness in terms of injury to officers overblown as well.

In addition, police must address a range of disputes among intimates, former intimates, and family members that may or may not involve violence, including

Some of these related problems are covered in other guides in this series, all of which are listed at the end of this guide. For the most up-to-date list of current and future guides, visit www.popcenter.org.

General Description of the Problem

Domestic violence involves a current or former intimate (and in many states, a current or former dating partner). Domestic violence tends to be underreported: women report only one-quarter to one-half of their assaults to police, men perhaps less.[2]The vast majority of physical assaults are not life threatening; rather, they involve pushing, slapping, and hitting.[3]Most women victims of domestic violence do not seek medical treatment, even for injuries deserving of it.[4]

Surveys provide us with estimates of the level of domestic violence in the United States, but there are wide differences among them depending on the definitions of domestic violence used and populations surveyed.[5]Two large surveys provide some insight into the level of domestic violence in the United States. The first, the National Violence Against Women Survey (NVAWS), conducted in 1995 and1996, found that nearly one in four women and nearly one in 13 men surveyed experienced rape and/or physical assault by a current or former spouse/partner/dating partner at some time in their lifetime, with about one and one-half percent of women and about one percent of men having been so victimized in the 12 months before the survey.[6]The National Crime Victimization Survey’s (NCVS) estimates, however, are about one-third lower for women and more than two-thirds lower for men. Differences in survey administration and methodology may account for the large differences in the numbers.

† NVAWS is a telephone survey of a representative sample of 8,000 U.S. men and 8,000 U.S. women.

†† The National Crime Victimization Survey (NCVS) collects data about criminal victimization from an ongoing nationally representative sample of U.S. households. The survey is administered every six months to about 100,000 individuals in approximately 50,000 households. Interviewers ask questions about crime victimization of all household members age 12 and older. The survey attempts to capture two types of crime, victimization that was reported to the police and victimization that was not reported to the police. (Rennison and Welchans, 2000).

††† The NCVS, administered by census workers as part of a crime survey, does not conduct all of its interviews in private because all members of the household are interviewed for different portions of the survey; also in contrast, the NVAWS survey uses more questions to screen for intimate violence, perhaps drawing out more from those interviewed.

Even the lower numbers of the NCVS suggest that intimate partner violence in the United States is extensive. However, NCVS trend data through 2001 shows that partner violence between current and former intimates has declined significantly. From 1993 through 2001, the rate of reported intimate violence dropped by about 50 percent in the United States.[7]From 1994 through 2001, the rate of every major violent and property crime declined by similar percentages.[8], It is unknown whether domestic violence is paralleling these declines for the same or different reasons.

† From 1994 through 2001, the rate of every major violent and property crime also steeply declined: homicide/manslaughter (down 40 percent); rape/sexual assault (down 56 percent); robbery (down 53 percent); aggravated assault (down 56 percent); simple assault (down 46 percent); household burglary (down 51 percent); motor vehicle theft (down 52 percent); theft (down 47 percent). (Rennison, 2001).

Domestic violence homicides have declined in similar proportions as well. In the United States, there were about half the number of intimate partner homicides (spouses, ex-spouses, boyfriends, and girlfriends) in 2002 as there were in 1976 with the largest portion of the decline in male victims (see Figure 1).[9]

Some commentators suggest that the decline in homicides may be evidence that abused women have developed legitimate ways to leave their relationships (e.g., divorce, shelters, police, and courts). The reasons for the decline may be even more complex because there is wide variation by race, not just by gender. Between 1976 and 2002, the number of black male victims of intimate partner homicide fell by 81 percent as compared to 56 percent for white males. The number of black female victims of intimate partner homicide fell 49 percent as compared to 9 percent for white females.[10]

Homicides of intimates by genderof victim, 1976-2002 (Figure 1)

Homicides of intimates by gender of victim, 1976-2002
Fox and Zawitz (2002)

Women as Offenders

There is a robust debate among researchers about the level of relationship violence women are responsible for and the extent to which it is in self-defense or fighting back.[11]The NCVS and other studies have found that women are the victims in as much as 85 percent of domestic violence incidents.[12]However, there are also research findings that women in heterosexual relationships have the same, if not higher, rates of relationship violence as men.[13]Generally, studies about domestic violence fall into two categories: family conflict studies and crime victimization studies. Those that tend to show high rates of violence by women (or rates higher than men) are family conflict studies and contain questions about family conflicts and disputes and responses to these, including physical responses. These studies use a family conflict assessment tool. Those studies that show that male assaultive behavior predominates in domestic violence are criminal victimization surveys and/or studies that rely on the counting of crime reports.[14]

Critics suggest that studies finding about equal rates of violence by women in relationships are misleading because they fail to place the violence in context (Dekeseredy et al. 1997); in other words, there is a difference between someone who uses violence to fight back or defend oneself and someone who initiates an unprovoked assault. Also, the physical differences between some women and their male partners may make comparisons between equivalent types of violence (slapping, kicking, punching, hitting) less meaningful, particularly because many studies show that violence by women is less likely to result in injury. Researchers agree that women suffer the lion’s share of injuries from domestic violence.[15]

Women living as partners with other women report lower rates of violence (11 percent) compared to women who live with or were married to men (30 percent).[16]About 8 percent of men living with or married to women report that they were physically abused by the women. About 15 percent of men cohabitating with men reported victimization by a male partner. These data suggest that men are engaged in more relationship violence.

Harms Caused by Domestic Violence

Domestic violence can include murder, rape, sexual assault, robbery, and aggravated or simple assault. In addition to the physical harm victims suffer, domestic violence results in emotional harm to victims, their children, other family members, friends, neighbors, and co-workers. Victims and their children experience the brunt of the psychological trauma of abuse, suffering anxiety, stress, sleep deprivation, loss of confidence, social isolation, and fear.[17]Batterers frequently also subject their victims to harassment (such as annoying or threatening phone calls), vandalism, trespassing, stalking, criminal mischief, theft, and burglary.[18]

Domestic violence also has economic costs. Victims may lose their jobs because of absenteeism related to the violence, and may even lose their homes because of loss of income. Some domestic violence victims must rely on shelters or depend on others to house them, and others become part of a community’s homeless population, increasing their risk for other types of victimization. Medical expenses to treat injuries, particularly of uninsured victims, create additional financial burdens, either for the victims or for the public.

† In the United States estimates of the percentage of homelessness among women resulting from domestic violence vary, but it may be upwards of 20 percent (National Coalition for the Homeless, 2004). In the United Kingdom, about 16 percent of the homeless to whom local authorities provided housing were victims of domestic violence (Office of the Deputy Prime Minister, n.d.). A recent report from Australia found that domestic violence is a major contributing factor to homelessness. The study also found that more than one-third of those accessing government housing assistance for homelessness were women escaping domestic violence, and two-thirds of the children in the housing program were those who accompanied a female parent or guardian escaping domestic violence (Australian Institute of Health and Welfare, 2005).

Theories About Domestic Violence

Theories about why men or women batter and why some people are reluctant to end abusive relationships may seem abstract to police practitioners, but the theories have important implications for how police might effectively respond to the problem.

Why Some Men Batter

Generally, four theories explain battering in intimate relationships.[19]

Psychological theory. Battering is the result of childhood abuse, a personality trait (such as the need to control), a personality disturbance (such as borderline personality), psychopathology (such as anti-social personality), or a psychological disorder or problem (such as post traumatic stress, poor impulse control, low self-esteem, or substance abuse).

Sociological theory. Sociological theories vary but usually contain some suggestion that intimate violence is the result of learned behavior. One sociological theory suggests that violence is learned within a family, and a partner-victim stays caught up in a cycle of violence and forgiveness. If the victim does not leave, the batterer views the violence as a way to produce positive results. Children of these family members may learn the behavior from their parents (boys may develop into batterers and girls may become battering victims). A different sociological theory suggests that lower income subcultures will show higher rates of intimate abuse, as violence may be a more acceptable form of settling disputes in such subcultures. A variant on this theory is that violence is inherent in all social systems and people with resources (financial, social contacts, prestige) use these to control family members, while those without resort to violence and threats to accomplish this goal.

Feminist or societal-structural theory. According to this theory, male intimates who use violence do so to control and limit the independence of women partners. Societal traditions of male dominance support and sustain inequities in relationships.

Violent individuals theory. For many years it was assumed that domestic batterers were a special group, that while they assaulted their current or former intimates they were not violent in the outside world. There is cause to question how fully this describes batterers. Although the full extent of violence batterers perpetrate is unknown, there is evidence that many batterers are violent beyond domestic violence, and many have prior criminal records for violent and non-violent behavior.[20]This suggests that domestic violence batterers are less unique and are more accurately viewed as violent criminals, not solely as domestic batterers. There may be a group of batterers who are violent only to their current or former intimates and engage in no other violent and non-violent criminal behavior, but this group may be small compared to the more common type of batterer.[21]

Why Some Women Batter

Some women batter their current or former intimates. Less is known about women who use violence in relationships, particularly the extent to which it may be in self-defense, to fight back, or to ward off anticipated violence. When asked in a national survey if they used violence in their relationships, many Canadian college women said they did. However, the majority of these women said it was in self-defense or to fight back and that the more they were victimized the more they fought back.[22]One researcher suggests that women should be discouraged from engaging in minor violence because it places them at risk for retaliation from men and men are more likely to be able to inflict injury.[23]

Clearly, there are women who use violence in relationships provocatively outside the context of fighting back or self-defense. The extent of this problem, as we noted earlier, remains unknown but is ripe for additional study.

The theories explaining male violence cited earlier may also have some relevance for women, although the picture is not clear.

Why Some Women are Reluctant to End Abusive Relationships

Police commonly express frustration that many of the battered women they deal with do not leave their batterers. Although many women do leave physically abusive relationships, others remain even after police intervene. There is no reliable information about the percentage of women who stay in physically abusive relationships. Researchers offer a number of explanations for the resistance by some to leave an abuser.[24]

Cycle of violence. Three cyclical phases in physically abusive intimate relationships keep a woman in the relationship: 1) a tension-building phase that includes minor physical and verbal abuse, 2) an acute battering phase, and 3) a makeup or honeymoon phase. The honeymoon phase lulls an abused woman into staying and the cycle repeats itself.

Battered woman syndrome. A woman is so fearful from experiencing cycles of violence that she no longer believes escape is possible.

Stockholm syndrome. A battered woman is essentially a hostage to her batterer. She develops a bond with and shows support for and kindness to her captor, perhaps because of her isolation from and deprivation of more normal relationships.

Traumatic bonding theory. A battered woman experienced unhealthy or anxious attachments to her parents who abused or neglected her. The woman develops unhealthy attachments in her adult relationships and accepts intermittent violence from her intimate partner. She believes the affection and claims of remorse that follow because she needs positive acceptance from and bonding with the batterer.

Psychological entrapment theory. A woman feels she has invested so much in the relationship, she is willing to tolerate the battering to save it.

Multifactor ecological perspective. Staying in physically abusive relationships is the result of a combination of factors, including family history, personal relationships, societal norms, and social and cultural factors.

Factors Contributing to Domestic Violence

Understanding the risk factors associated with domestic violence will help you frame some of your own local analysis questions, determine good effectiveness measures, recognize key intervention points, and select appropriate responses. Risk factors do not automatically mean that a person will become a domestic violence victim or an offender. Also, although some risk factors are stronger than others, it is difficult to compare risk factor findings across studies because of methodological differences between studies.


The female age group at highest risk for domestic violence victimization is 16 to 24.[25] Among one segment of this high-risk age group—undergraduate college students—22 percent of female respondents in a Canadian study reported domestic violence victimization, and 14 percent of male respondents reported physically assaulting their dating partners in the year before the survey.[26] And although the victimization of teen girls is estimated to be high, it is difficult to “…untangle defensive responses from acts of initial violence against a dating partner.”[27]

Socioeconomic Status

Although domestic violence occurs across income brackets, it is most frequently reported by the poor who more often rely on the police for dispute resolution. Victimization surveys indicate that lower-income women are, in fact, more frequently victims of domestic violence than wealthier women. Women with family incomes less than $7,500 are five times more likely to be victims of violence by an intimate than women with family annual incomes between $50,000 and $74,000.[28]

Although the poorest women are the most victimized by domestic violence,[29] one study also found that women receiving government income support payments through Aid for Families with Dependent Children (AFDC) were three times more likely to have experienced physical aggression by a current or former partner during the previous year than non-AFDC supported women.[30]


Overall, in the United States, blacks experience higher rates of victimization than other groups: black females experience intimate violence at a rate 35 percent higher than that of white females, and black males experience intimate violence at a rate about 62 percent higher than that of white males and about two and a half times the rate of men of other races.[31] Other survey research, more inclusive of additional racial groups, finds that American Indian/Alaskan Native women experience significantly higher rates of physical abuse as well.[32],

† It is unclear how much of the differences in victimization rates by race is the result of willingness to reveal victimization to survey interviewers (Tjaden and Thoennes, 2000).

Repeat Victimization

Domestic violence, generally, has high levels of repeat calls for police service.[33] For instance, police data in West Yorkshire (United Kingdom) showed that 42 percent of domestic violence incidents within one year were repeat offenses, and one-third of domestic violence offenders were responsible for two-thirds of all domestic violence incidents reported to the police. It is likely that some victims of domestic violence experience physical assault only once and others experience it repeatedly[34] over a period as short as 12 months.[35] British research suggests that the highest risk period for further assault is within the first four weeks of the last assault.[36]

Incarceration of Offenders

Offenders convicted of domestic violence account for about 25 percent of violent offenders in local jails and 7 percent of violent offenders in state prisons.[37] Many of those convicted of domestic violence have a prior conviction history: more than 70 percent of offenders in jail for domestic violence have prior convictions for other crimes, not necessarily domestic violence.[38]

Termination of the Relationship

Although there is a popular conception that the risk of domestic violence increases when a couple separates, in fact, most assaults occur during a relationship rather than after it is over.[39] However, still unknown is whether the severity (as opposed to the frequency) of violence increases once a battered woman leaves.


Contrary to popular belief, pregnant women are no more likely than non-pregnant women to be victims of domestic violence.[40] In fact, some women get a reprieve from violence during pregnancy. The risk of abuse during pregnancy is greatest for women who experienced physical abuse before the pregnancy.[41] Some additional factors increase the risk during pregnancy: being young and poor and if the pregnancy was unintended.[42] Physical abuse during the pregnancy can result in pre-term delivery, low birth weight, birth defects, miscarriage, and fetal death.[43]

Multiple Risk Factors for Women and Men

Being young, black, low-income, divorced or separated, a resident of rental housing, and a resident of an urban area have all been associated with higher rates of domestic violence victimization among women.[44] For male victims, the patterns were nearly identical: being young, black, divorced or separated, or a resident of rental housing.[45] In New Zealand, a highly respected study found that the strongest predictor for committing partner violence among the many risk factors in childhood and adolescence is a history of aggressive delinquency before age 15.[46]

The study also found that committing partner violence is strongly linked to cohabitation at a young age; a variety of mental illnesses; a background of family adversity; dropping out of school; juvenile aggression; conviction for other types of crime, especially violent crime; drug abuse; long-term unemployment; and parenthood at a young age.[47]

Other Risk Factors

Several other risk factors emerge from research:

Recently, there is much discussion among police about the link between pet abuse and domestic violence. Although some overlap is likely, particularly under the theory that many batterers are generally violent, not enough is known because of the types of studies undertaken. Some small surveys of domestic violence shelter residents suggest that some women might have left their abuser sooner but they worried about their pet’s safety.[51]

Finally, although alcohol and drug use do not cause intimate partner battering, the risk of victim injury increases if a batterer is using alcohol or drugs.[52]

Understanding Your Local Problem

Notwithstanding its decline over the last decade, domestic violence stubbornly remains a frequent call for police, and efforts to further reduce it require general and specific information about the nature of the problem. You must combine the basic facts with a more specific understanding of your local problem. Analyzing the local problem carefully will help you design a more effective response strategy.


In addition to criminal justice agencies, the following groups have an interest in the domestic violence problem and ought to be considered for the contribution they might make to gathering information about the problem and responding to it:

Asking the Right Questions

The following are some critical questions you should ask in analyzing your community’s domestic violence problem, even if the answers are not always readily available. Your answers to these and other questions will help you choose the most appropriate set of responses later on.

Aggregate data is an important source for analyzing your domestic violence problem, but it is useful only if domestic violence incidents are properly investigated and documented. It is important for investigating officers to understand the context and history of domestic assaults to determine if the incident is part of a series of abuse the victim has sustained and if it’s likely to recur or escalate to more serious violence. For instance, in assessing individual incidents it is important to find out how long the abuse has been occurring, the frequency of the abuse, if the abuse is escalating, specific threats (even threats of suicide), whether threats can be carried out or there is an indication that they will be carried out, and whether victimization also involves other criminal behavior (i.e., harassing phone calls, vandalism, theft, burglary). You should analyze a variety of data sources such as calls for police service relating to domestic disputes, offense/incident reports of domestic violence, and databases from domestic abuse social service agencies.



† You should review one year’s worth of domestic violence cases from two years prior. For instance, if the current calendar year is 2006, review the results of domestic assault cases from the calendar year 2004. The two-year gap allows agencies to follow offenders from arrest to sentencing and even the participation in and completion of treatment.



Current Responses

Measuring Your Effectiveness

Measurement allows you to determine to what degree your efforts have succeeded, and suggests how you might modify your responses if they are not producing the intended results. You should take measures of your problem before you implement responses, to determine how serious the problem is, and after you implement them, to determine whether they have been effective. (For more detailed guidance on measuring effectiveness, see the companion guide to this series, Assessing Responses to Problems: An Introductory Guide for Police Problem-Solvers.)

Impact measures gauge the degree to which you reduced the harms caused by the problem. Process measures gauge the degree to which you implemented responses as planned. A good assessment employs both impact and process measures.

Impact Measures

The following are potentially useful measures of the effectiveness of responses to domestic violence:

Process Measures

The following are potentially useful measures of the extent to which you implemented your planned responses:

Responses to the Problem of Domestic Violence

Your analysis of your local problem should give you a better understanding of the patterns of domestic violence cases and calls in your jurisdiction. Once you have analyzed your local problem or important aspects of it and established a baseline for measuring effectiveness, you should consider possible responses to address the problem.

These strategies are drawn from a variety of sources, including descriptive materials, research studies, and police reports. It is critical that you tailor responses based on reliable analysis. In most cases, an effective strategy will involve implementing several different responses. Law enforcement responses alone are seldom effective in reducing or solving the problem. Do not limit yourself to considering what police can do: carefully consider others in your community who share responsibility for the problem and can help police better respond to it. In some cases, the responsibility of responding may need to shift toward those who have the capacity to implement more effective responses. (For more detailed information on shifting and sharing responsibility, see Response Guide No. 3, Shifting and Sharing Responsibility for Public Safety Problems).

General Considerations for an Effective Response Strategy

  1. Implementing a comprehensive and collaborative response strategy. Comprehensive and collaborative approaches to reducing domestic violence are more likely to succeed than piecemeal approaches, but they require significant commitment from all participating. One research team wrote, “[a]n offense as complex as domestic violence is unlikely to be prevented by a single measure.”[53] A thorough assessment of the current overall response to domestic violence can identify critical gaps in strategies, resources, and response protocols.

    Although some communities have adopted a more integrated approach engaging advocates, police, and the criminal justice system, for the most part, recidivism remains high. In the small studies of these integrated domestic violence approaches, there is evidence that victim satisfaction is high but insufficient evidence that recidivism and revictimization rates have decreased. One commentary suggests that the small core of persistent batterers (who are violent toward others as well as their intimates) are perhaps resistant to even highly coordinated efforts.[54]

    † For a fuller description of the different types of integrated approaches adopted and their studies, see Buzawa and Buzawa, chapter 15 (2003).

    To improve the likelihood that a comprehensive approach reduces recidivism and victimization requires a continuum of responses depending on the most reliable research and covering the different points in time most important to reducing domestic abuse: before an incident to keep it from occurring, during an incident to stop the immediate violence, and after an incident to reduce or prevent revictimization. It involves responses that focus on victims and potential victims and strategies that focus on offenders and potential offenders. As well, it involves the improved identification and reporting of cases of abuse between current and former intimates and dating partners.

    The matrix below may help you organize the strategic focus, goals, and timing of your responses to domestic violence. This section is followed by information on the impact of specific responses to domestic violence strategies. Many of the responses will require the support and collaboration of other governmental agencies and community nonprofits.

Figure 2. Matrix of Responses to Domestic Violence
Strategic Focus Strategic Times for Responses Goal Police Role Other Agencies, Organizations, Group
At-risk population Before incidents Prevention; persuade those at risk that, if abused, call the police Alert and educate at-risk victim population; educate/warn at-risk offending population Public health organizations; domestic violence coalitions; schools and educators; medical professionals
Peers and neighbors of at-risk individuals Ongoing Getting peers and neighbors to call the police if they learn of domestic abuse Educate these groups about the importance of calling the police to reduce the violence Public health organizations; domestic violence coalitions; educators
Injured women and men During medical care Screen the injured for domestic violence; raise awareness of available services; provide medical care Engage the medical profession and link medical professionals with appropriate referral organizations Medical professionals
Individual incident During Violence cessation Stop the violence; identify primary aggressor; accurately identify abuse history Medical and public health professionals
Immediately after incident After; ongoing Prevent revictimization Assist with victim safety; develop tailored strategies for victim and offender based on risk/physical violence history; increase focus on high-risk offenders; ensure victim is linked with needed resources; increase focus on high-risk victims; ongoing monitoring Domestic violence victim advocates, victims’ friends and family, shelters, victim services, criminal justice system, treatment services
  1. Educating collaborative partners. Each partner in a domestic violence reduction collaborative brings a unique perspective and body of knowledge. It is important for members of domestic violence collaborations to operate with precise and accurate information about what does and does not work and about the dimensions of the problem in your community. Do not assume that other professional groups participating in the collaborative have the most up-to-date knowledge about the problem or are following prescribed protocols.
  2. Tailoring the police response on the basis of offender and victim risk. Some researchers advocate a graded response to domestic violence. They view batterers along a continuum—some are easily dissuaded from rebattering, others require increased actions—and a graded or tiered approach to control offender behavior can be effective. For instance, we know that a percentage of batterers is deterred from rebattering simply by having the police called on them, so encouraging reporting is essential; others may be deterred with the additional application of a restraining order. More is required to keep other batterers from rebattering, which suggests a need to refine assessments about who these batterers are.[55] Some researchers also suggest a graded approach to victim safety for similar reasons: some victims are at low risk of being revictimized and some are at higher risk.

    Graded approaches to both victim and offender can be effective ways to reduce revictimization. British researchers conducted several evaluations and found that significant reductions in repeat victimization are achievable.[56] The key to reduction is that, at each level, police (and others) must focus equal attention on the victim and the offender. The victim and the offender must know about the actions police have and will take in relation to each other. Graded approaches must be applied quickly because the highest risk period for further assault is within the first four weeks of the last assault.[57]

    Under a graded approach victims are assigned to one of three follow-up response levels based on the following:

    • the harm the abuser caused or threatened
    • the victim’s account of previous incidents not reported to the police
    • all incidents recorded in police databases involving the same victim with the same offender
    • the offender’s previous violence to other victims.[58]

    Incidents other than domestic violence are reviewed because they may be indicators of escalating aggressiveness (Hanmer et al. 1999).

    Each of the three levels to which a victim is assigned requires some follow-up. There is in an increase in the variety and intensity of safety measures used to protect the victim and to restrict the offender from rebattering if the offender’s violence and criminal history indicates he is at an increased risk to re-offend.[59] Evaluations of graded approaches found reduced domestic violence calls and increased time intervals between violence.[60] The methods to protect the victim or deter the offender can include a variety of situational crime prevention opportunity blocking mechanisms, such as the following:

    • increased police surveillance of victims’ homes
    • greater coordination with other parts of the criminal justice system
    • pendant alarms for at-risk victims
    • video cameras placed in high-risk victims’ homes
    • cocoon watch over victims
    • target hardening of victims’ vulnerable properties
    • police watch of offenders
    • police opposition to bail

    †† With a victim’s permission, neighbors, relatives, friends, or all three are asked to look out for the victim and immediately call if the offender returns.

    • electronic ankle bracelet monitoring of high-risk released offenders
    • alarm-activated recording devices with two-way speech capability (allowing victims to speak directly to the police, and vice versa).

    † To view the 25 techniques of situational crime prevention, see www.popcenter.org/25techniques/ in the Center for Problem-Oriented Policing. Click on a particular technique to view its description.

    Tailoring police responses to particular offenders based on the seriousness and frequency of their offenses has been successfully applied in the context of conventional crime and may be as useful for dealing with domestic batterers. This is likely because much domestic violence is committed by repeat and chronic offenders who may be particularly vulnerable because they often are under various forms of legal supervision due to past offending.[61]

    †† See Kennedy, Waring, and Piehl (2001) for a description of the pulling levers/focused deterrence application in the homicide context, and Spelman (1990) for a discussion of repeat offender programs. See Kennedy (2002) for a discussion of the application of the approach to domestic violence.

    According to Buzawa and Buzawa, “The criminal justice system must develop the capabilities to identify those batterers for whom normal deterrence can be effective, perhaps the majority in terms of numbers of incidents… [i]t should also be able to differentiate, segregate, and incapacitate batterers who must be deterred by special approaches.”[62]

Specific Responses to Domestic Violence

  1. Educating potential victims and offenders. Some police agencies participate in domestic violence awareness campaigns and school programming, such as classroom instruction to teens about dating violence and ways to handle conflict. Domestic violence prevention messages may target the general population or specific populations. For example, campaigns may be designed to encourage victim reporting, deter potential offenders, or raise the consciousness of potential witnesses of abuse (neighbors, friends, relatives). However, the effect of these prevention strategies is unknown.[63] For instance, few of the programs developed to reduce teen dating violence have been evaluated, and of those that have, there have been mixed results.[64] Although some report an increase in knowledge in the targeted population and greater familiarity with available resources to help victims, this does not necessarily translate into a reduction in the incidence level of dating violence.

    † The Lancashire (United Kingdom) Police Constabulary placed messages about domestic violence on police vehicles, beer glass coasters in bars, utility bills, and lampposts, and used radio advertising to increase awareness of domestic violence.

    As a rule, prevention is more likely to work if highly targeted. General campaigns are not typically effective. Highly targeted campaigns that focus on a specific target group or geographic area can have some impact. Offender-oriented campaigns, which are designed to raise potential offenders’ perceptions that there will be meaningful consequences to battering, are more likely to be effective than campaigns that appeal to potential offenders’ morals.

    † For more in-depth information about prevention campaigns and the conditions under which they are most likely to be effective, see Response Guide No. 5, Crime Prevention Publicity Campaigns.

    Prevention efforts targeting potential victims should focus on those at higher risk, such as young women ages 16 to 24, as they experience the highest rates of intimate violence. Special efforts should be made to reach the poorest women in this age group as they are at an even higher risk. In addition, some recent immigrant communities, depending on the laws and privileges in the home country, may show a high level of domestic abuse, particularly if there is a lack of familiarity with assault laws in the adopted country. One of the reasons crime prevention campaigns have had limited success is that potential victims do not see themselves as such; victim-oriented prevention campaigns must overcome this threshold issue.

    For a review of the research pertaining to domestic violence and immigrant populations, see Buzawa and Buzawa (2003).

    See “Overcoming Language Barriers: Solutions for Law Enforcement” PDF

  2. Encouraging domestic violence victims and witnesses to call the police. Police and other members of a domestic violence reduction collaborative should encourage people to call the police if they are victims of, witnesses to, or know a victim of domestic violence. Prevention and education efforts should include this as a core message. A study of more than 2,500 domestic violence victims concluded that calling the police had a strong deterrent effect on revictimization, even when the police did not make an arrest, when the offender had a prior history of violence against the victim, and when the assault was sexual. Calling the police was beneficial even when the violence was severe. In addition, offender retaliation did not appear to be more likely even when a victim rather than a third party called police.[65] ,

    † The study, which examined felony and misdemeanor violence, male and female offenders, and couples in different types of relationships, tracked victims for three years.

  3. Encouraging other professionals to screen for domestic violence victimization and make appropriate referrals. The American Medical Association adopted domestic violence screening and referral guidelines for medical practitioners.[66] Physicians should screen injured women patients to determine if domestic violence was the cause of the injury. Medical professionals should also discuss domestic violence with pregnant patients during prenatal checkups. Physicians’ documentation of specific incidents of domestic abuse can be critical to the successful prosecutions of batterers.

    † The American Academy of Pediatrics, the American Medical Association, and the American College of Obstetricians and Gynecologists all endorse screening.

    †† See Isaac and Enos (2001) for guidelines for proper medical documentation of battered patients.

    In spite of these professional recommendations, most physicians are reluctant to routinely screen women for domestic violence, citing a lack of training in how to conduct screenings and insufficient knowledge of appropriate responses and referrals when a patient discloses domestic violence.[67]

  4. Providing victims with emergency protection and services after an assault. Battered women’s shelters protect women from further harm after an assault, sometimes on referral from the police and sometimes not. Typical services include a domestic violence hotline, temporary housing, information and referrals to other social services, safety planning, victim advocacy for emergency benefits or at court proceedings, and referrals for legal services.

    The first shelter for battered women and their children opened in London in 1972. There are currently more than 2,000 shelters in the United States.[68] In most communities, shelters raise community funds for operation; in some communities, police contribute a portion of their budget to aid in shelter operation. Shelters often rely on volunteers and a few paid personnel to provide round-the-clock assistance to battered women and their families. Little is known about the number of repeat victims served annually, the length of average stays, or the effectiveness of shelter services in preventing subsequent violence.

    Although there are confidentiality issues to resolve or respect, police should seek to exchange information with domestic violence victim service providers as much as possible to learn more about the domestic violence victim population, some of whom do not seek out police assistance. This information exchange can aid in identifying the highest risk victims and offenders, targeting prevention efforts, designing safety plans, and learning more about the community’s offenders.

    † More than 60 U.S. law schools offer student advocacy services for domestic violence victims at court proceedings (Roberts, 2002b).

    Recently emerging are family justice centers, which house domestic violence victim services in one location to increase victim survival, independence, and recovery. Formerly, victim services were scattered in different places, sometimes at opposite ends of cities. If victims followed up with these fragmented services, they too often experienced the frustration of retelling the story of violence to every individual provider.

    † The San Diego Family Justice Center provides victims with advocacy, childcare, clothing, counseling, court support, deaf/hard-of-hearing assistance, emergency housing, food, forensic documentation of injuries, housing for pets, internet access, law enforcement, legal assistance, locksmith services, medical services, military assistance, phones, phone cards, restraining orders, support groups, safety planning, spiritual support, transportation, and victim compensation (San Diego Family Justice Center).

  5. Assessing the threat of repeat victimization. Gathering accurate information about past abuse, including unreported incidents, is critical in assessing a victim’s current risk and tailoring appropriate offender interventions. Although there is no foolproof profile of a repeat batterer or repeat victim, having a complete and accurate picture of both the victim’s and the offender’s history of abuse is useful for predicting future risk. You may need to gather records from other jurisdictions where the victim or offender lived. You should also be alert for other related behaviors such as threats of violence, harassment, trespassing, vandalism, stalking, protective order violations, and prior use of a gun, as these behaviors offer clues as to whether abusive behavior is chronic and/ or escalating.

    You should supplement official records with the victim’s personal knowledge about the offender. Keep in mind that for a variety of social and psychological reasons, victims may be reluctant to reveal the extent of the battering, particularly to police.

    Police may consider soliciting the assistance of trained medical professionals to help determine a victim’s abuse history. A study of one initiative in which a doctor and a nurse (or paramedic) accompanied police on domestic violence calls found victims revealing much more about the extent to which their partner battered them than police typically elicit, suggesting that victims may feel more comfortable reporting repeat victimization to medical professionals. Even with the high levels of repeat victimization uncovered, few victims had sought counseling, shelter, or medical treatment for the prior assaults.[69]

    There is great interest in developing an assessment instrument police can use to predict and help prevent domestic violence homicides. Although some such instruments exist, they tend to over-predict lethality. This is because only a very small portion of domestic violence victims are murdered and distinguishing between victims who will be murdered and those who will not remains elusive. For instance, even though offender unemployment is a risk factor, the vast majority of unemployed abusers do not murder their current or former intimates. Even when you combine unemployment with other risks it does not give you the profile of a murderer, but someone who is at an increased risk of battering.

  6. Arresting offenders. Many U.S. police agencies adopted pro-arrest or mandatory arrest domestic violence policies in the 1980s and early 1990s. Propelling these policies were:
    • legal decisions establishing civil liability against the police for failure to protect women victims of domestic violence[70]
    • the women’s movement’s advocacy and activism on behalf of domestic violence victims. The women’s movement challenged the view of domestic violence as a family problem that could or should be handled privately or differently from stranger assault
    • widespread dissemination of pro-arrest results of a misdemeanor domestic assault research study.[71] ,

      † More than 300 newspapers reported the results of the study, unprecedented coverage for that time (Fagan, 1996).

    Generally, pro-arrest laws and policies apply not only to spouses, but to unmarried partners, former intimates, and persons who had or raised a child together. In many jurisdictions the laws or policies apply to both heterosexual and homosexual relationships.

    Police interventions in domestic violence incidents have expanded beyond merely separating and counseling the parties; they’ve become full-blown criminal investigations in which witnesses are interviewed, neighbors are canvassed, injuries are photographed, physical evidence is collected, future threats are assessed, and victims are referred to follow-up protective services and helped to plan for their future safety. In addition, some states permit police to seize firearms from alleged batterers,[72] and federal laws generally prohibit convicted misdemeanant batterers or those against whom there is a valid order of protection from possessing a gun.[73] All U.S. states now permit police to make warrantless arrests for both misdemeanor and felony assaults.[74]

    The highly influential 1980s study of police interventions in domestic violence incidents in Minneapolis found deterrence value in arrest in misdemeanor domestic assaults, as compared to two other interventions — separation of parties or mediation of the dispute at the scene.[75] However, less well-reported replication studies in the late 1980s produced mixed results.[76] A more recent analysis combining five of the replication studies concluded there is only some modest deterrent effect from arrest.[77] However, even this modest effect should be viewed with caution for several reasons:

    • These studies considered only misdemeanor, not felony, domestic assaults.
    • Offenders’ employment status appears to be an important variable: unemployed offenders have less at stake (they will not lose their jobs) if they re-offend and thus are less deterred by arrest.[78]
    • A prior arrest record for any crime, and intoxication at the time of the incident, increase the risk that batterers will re-offend after arrest.[79]
    • A small group of offenders, perhaps fewer than 10 percent, appear to continue battering regardless of the intervention, including arrest.[80] ,

    A more recent study of victims of both misdemeanor and felony assaults concluded the following:

    • calling the police was a strong deterrent to repeat battering
    • the effects of arrest were small and statistically insignificant but could not be ruled out completely
    • if arrest does have an effect, it is likely to be modest, particularly in comparison to the effect of calling the police.[81] ,

    Victim advocacy groups have generally not been swayed by findings of little or modest effect of arrest. For many advocates, batterer arrest is seen as an important symbol of a woman’s legal right to be free of intimate partner violence[82] and, moreover, argue that police continue to arrest other types of offenders without strong evidence of its effectiveness. Arrest is believed to be an important message to children that abuse of their parent is illegal, and perhaps also a deterrent to male children as they become men. This belief has not yet been properly studied, however.

    † Nearly ten percent of the more than 3,000 studied battered repeatedly regardless of the intervention. This group of 250 batterers accounted for 7,380 battering incidents in the six months after the initial intervention. Interventions, even arrest, did not deter this small but violent group.

    †† The researchers found that batterers who battered again had an increased likelihood of battering a third time, and offenders who were under the influence of alcohol or drugs at the time of the incident were more likely to re-offend. The researchers did not find that marital status, poverty, race, education or gender improved the effect of arrest, but because they did not have access to employment information, they could not rule out that arrest deters employed offenders but not unemployed offenders.

    ††† Victims of domestic abuse have also called for police agencies to monitor, address, and more appropriately sanction officers engaging in domestic assault. Although some police agencies have been responsive, additional efforts are required for victims to have higher levels of confidence in police agencies’ domestic violence response policies.

  7. Issuing and enforcing restraining orders. Restraining orders (also known as “stay away” or protective orders) are intended to prevent offenders from further harassing, threatening or contacting the victims. Courts have made restraining orders widely available to domestic violence victims, whether or not they file a police report. Courts may issue a temporary (time-limited) restraining order even when the “party being restrained” is not present or represented. Protective relief may be temporary or permanent. Violation of these orders is now a criminal offense in all U.S. states.[83]

    Domestic violence restraining orders are frequently violated although some offenders may be deterred by them. Some research findings suggest that a victim is more likely to seek a protective order if the partner had a criminal history of violent offending, which may be why so many orders are violated; those with robust abuse histories may be the least likely to be deterred by written limits so police are advised that more must be done in these cases.

    † In most jurisdictions these are obtained from civil court; however, some jurisdictions also grant concurrent jurisdiction to criminal court. Criminal courts can also issue these once a criminal proceeding begins.

    †† Civil restraining orders were in fact developed to counter the reluctance of police, prosecutors, and criminal courts to treat domestic violence as a serious criminal matter (Buzawa and Buzawa, 2003).

    ††† For a good review of the research about protective orders, see Buzawa and Buzawa (2003).

  8. Aggressively pursuing criminal prosecution of severe domestic violence cases and publicizing convictions. Police pro-arrest and mandatory arrest policies have generated significantly larger caseloads for prosecutors. Similarly, prosecution policies against dropping charges (“no-drop”) even when the victim expresses such a desire (the functional equivalent of “mandatory arrest” for police) has further strained prosecutorial resources.[84] , Although such police and prosecution policies can have the beneficial effect of reducing an offender’s urge to retaliate against the victim because responsibility for the prosecution is no longer in the victim’s hands, it is not yet clear whether such policies have limited further violence or have had the unintended consequence of discouraging some victims from calling police in the first instance.[85]

    † The rationale for “no-drop” policies is that the state has an independent interest in seeing domestic violence offenders prosecuted because of the harm caused to victims, victims’ children, and potential victims.

    †† Victim recanting remains high. As a result, many prosecutors rely on physical evidence such as photographs and medical reports of victims’ injuries and out-ofcourt statements (e.g., 911 call tapes of in-progress assaults) to counter uncooperative and fearful victims. For an excellent review of prosecutorial response to the increased numbers of domestic violence cases and studies of prosecutorial case screening practices, see Buzawa and Buzawa chapter 11 (2003).

    “No-drop” policies have some drawbacks. Victim discretion is further reduced, case backlogs increase, and time to disposition is lengthened, which can strain resources devoted to pretrial victim safety.[86] The increase in prosecutorial workload can force prosecutors to trade off the prosecution of other crimes for misdemeanor battering. It is now apparent that prosecutors in “no-drop” jurisdictions rarely prosecute all cases; they retain some level of discretion in case filing decisions, typically at the intake and case screening point.[87] Police must therefore help prosecutors identify the most severe and chronic cases from among the many arrests and encourage prosecutors to prosecute such cases vigorously, make special efforts to protect the victim, and publicize convictions so as to maximize the general deterrent effect.

  9. Establishing special domestic violence courts. There are more than 200 domestic violence courts in the United States, and a growing number in the United Kingdom as well. The proliferation of these courts is part of a wider trend toward specialty courts: drug court, mental health court, drunk driving court, etc. Advocates for specialty courts believe they result in improved outcomes: an increase in specialty knowledge critical to case handling (including the dynamics of the underlying crime/behavior, whether it is battering, drinking, or schizophrenia, depending upon the court), timely attention to the case, and a concentration of appropriate resources that traditional courts do not have that can lead to more effective case handling.

    See Problem-Specific Guide No. 42, Witness Intimidation, for further information on measures to protect victim-witnesses.

    Some of these courts are termed “problem-solving” courts, but the term in this context should not be confused with its meaning in problem-oriented policing. Problem-solving in the court context refers to solving the individual defendant’s personal problems that contribute to his offending rather than to addressing an aggregation of incidents. See National Center for State Courts (n.d.).

    Typically in specialty courts a single judge works with a community team to develop a case plan for the defendant and uses pending criminal sanctions to compel a defendant’s compliance with treatment. The judge monitors compliance and imposes criminal sanctions if the defendant fails to keep to the case plan.

    Early evaluations of domestic violence courts generally report on how these courts handle their workload, victim satisfaction, and issues of implementation. It remains unclear if these courts impact recidivism. Researchers who examined these courts in New York describe some of the more important unresolved issues:

    Many domestic violence advocates are hesitant to embrace the idea that domestic violence courts are “problem-solving courts.” There are substantial differences between domestic violence courts and other problem-solving courts. Many of these differences stem from how success is measured and to whom services are offered. Drug courts can easily look to see whether defendants are successfully completing their court-mandated drug-treatment programs. But domestic violence courts are not targeted at “rehabilitating” defendants. Indeed, services are offered primarily to help victims achieve independence. The primary “service” offered to defendants is batterers programs. But in New York, batterers programs are used by domestic violence courts primarily as a monitoring tool rather than as a therapeutic device. This approach is based on the research about batterers programs, which is extremely mixed. It is unclear whether these programs have any impact at all in deterring further violence.[88]
  10. Providing treatment for batterers. Some batterer treatment programs are voluntary; others are court-mandated. In some jurisdictions, prosecutors recommend these programs as part of pre-trial diversion; in others they are part of court-ordered mandatory sentencing. Many states now mandate batterer treatment.[89]

    Batterer treatment programs may take a variety of forms. Many offer group treatment with a focus on anger management. Others include individual assessments and individual counseling, and substance abuse and/or mental health treatment.[90]

    Unfortunately, few batterer treatment programs have undergone thorough evaluation, and those that have show a mix of positive and negative results.[91] Court-mandated treatment is more likely to result in batterers completing programs,[92] “but there is little evidence to support the effectiveness of one batterer program over another in reducing recidivism.”[93]

    The quality of the evaluations of batterer treatment programs have improved over time but continue to encounter both methodological and programmatic challenges as illustrated by two recent studies of batterer treatment programs, one in Broward County, Florida and the other in Brooklyn, New York.[94] In both, offender treatment was based on the Duluth treatment model, which is the most commonly used. The Broward evaluation found that treatment attendance did not reduce battering, but that offenders who were married, employed, or homeowners were less likely to batter again (that is, these offenders had a “stake in conformity”). Also, younger men, particularly those with no stable residence, were more likely to rebatter.[95] In Brooklyn, the evaluation showed minor improvement for some of the batterers

    † The Duluth model suggests that batterers seek to control their partners (or ex-partners) and this must change for batterers’ behavior to change (Pence and Paymar, 1993). The model “helps offenders to understand how their socialized beliefs about male dominance impede intimacy; that violence is intentional and a choice designed to control their intimate partner; that the effects of abusive behavior damage the family; and that everyone has the ability to change” (Minnesota Program Development, Inc.).

    (that is, some reduction in the number of battering incidents for those attending a 26-week treatment program rather than the same program condensed into an eight-week schedule). In neither case were batterers’ attitudes toward domestic abuse changed. Even these evaluation results are not fully reliable because both studies experienced data collection challenges as a result of a high drop-out rate by offenders, difficulty finding relocated victims for follow-up interviews, and inadequate offender attitude assessment tools. In addition, judges sometimes overrode random assignment of batterers, thereby tainting the makeup of the different groups studied.[96] Evaluations of other types of treatment programs, including cognitive-behavior therapy (another widely used approach), have also suffered from similar methodological flaws.

    Several experts suggest that greater refinement in assigning batterers to appropriate programs could improve results. The most chronic batterers should receive the most intensive treatment.[97] A “one-size fits-all” approach to batterer intervention cannot accommodate the diverse population of batterers entering the criminal justice system.”[98] The different types of batterers—family-only, one who is generally violent even to others, dysphoric (mood-disordered)/borderline—may require tailored treatment.[99]

    Experts recommend that treatment programs be designed around explicit theories. In other words, each intervention proposed should have a specific underlying theory. Outcomes expected from each of the interventions should be clearly defined and then evaluated for short-and long-term impact. Designing treatment programs that fit this model requires close collaboration between service providers and researchers. [100] In addition, the timing of treatment may be an important element to success.

    “Counseling ideally would begin almost immediately after a violent episode, when the offender feels most remorseful, most frightened of the criminal justice system, and most receptive to demands for change.”[101]

Responses With Limited Effectiveness

  1. Arresting both parties in a domestic violence incident. Arresting both participants in a domestic violence incident (so-called “dual arrest”) under the principle of mutual combat is ineffective toward interrupting the pattern of violence between the two. In the context of the longer history of the relationship—as opposed to the one incident—there is nearly always one primary aggressor. When police respond to a domestic violence call, self-defense may look like “mutual combat” and only detailed interviews of the parties (their prior abuse history) and witnesses may reveal the primary aggressor. In addition, at the scene of domestic abuse, “a victim may feel safe to express anger against the batterer in the presence of the police … giving the impression that they are the perpetrators.”[102] As a result, more than 20 states have enacted primary aggressor or “predominant aggressor” laws. Even these might not be enough to discourage dual arrest practices unless the law recognizes the importance of ascertaining the pattern/history and not just the aggression within a single incident. To complicate matters, there are couples in which both partners are violent. These are more likely to entail “recurrent acts of minor violence initiated by either party, but the type of violence generally seen by police (and in shelter and clinical samples) is more likely to involve serious and frequent beatings, as well as the terrorizing of women.”

† For a good discussion of “dual arrest” research and the complexities within the primary aggressor issue, see Buzawa and Buzawa (2003).

Summary of Responses

The table below summarizes the responses to Domestic Violence, the mechanism by which they are intended to work, the conditions under which they ought to work best, and some factors you should consider before implementing a particular response. It is critical that you tailor responses to local circumstances, and that you can justify each response based on reliable analysis. In most cases, an effective strategy will involve implementing several different responses. Law enforcement responses alone are seldom effective in reducing or solving the problem.

General Considerations for an Effective Response Strategy
# Response How It Works Works Best If… Considerations
1 Implementing a comprehensive and collabo rative response strategy Addresses both victimization and offending; identifies gaps in strategies, resources, and response protocols … the collaborative does an appraisal ofthe community’s response to domestic violence to identify what is and isn’t working and gaps Group should be educated about what works in reducing domestic violence victimization and revictimization and the limitations of some approaches; group commits to ongoing evaluation of efforts; collaboration with a university researcher may be useful; will probably require a champion who pursues a collaborative response strategy
2 Educating collaborative partners Increases likelihood of adoption ofproven effective responses …collaborative partners commit to relying on facts and research, rather than anecdotes Requires high level of coordination
3 Tailoring the police response on the basis of offender and victim risk Applies the most appropriate type and level of response to the particular victim and offender …offender is told about the measures police put in place; graded responses are applied quickly because the highest risk period for further assault is within the first four weeks of the last assault Accurate victimization and offending information is needed to select the most appropriate level of response
Specific Responses to Domestic Violence
# Response How It Works Works Best If… Considerations
4 Educating potential victims and offenders Encourages victim reporting, de-motivates potential offenders, or raises the consciousness of potential witnesses to abuse …efforts are highly targeted and focused on a geographic area or certain high-risk groups If evaluation mechanisms are not put in place, the campaign, which can be costly, will remain of unknown value
5 Encouraging victims and witnesses to call the police Deters potential and actual offenders …at-risk populations and their peers and neighbors believe that calling the police will be effective Hard core batterers are not likely to be deterred just by calling, so more must be done
6 Encouraging other professionals to screen for domestic violence victimization and make appropriate referrals Increases likelihood of effective intervention in abusive relationships …doctors have adequate training Requires active participation of community’s medical profession
7 Providing victims with emergency protection and services after an assault Provides safe place for victims; improves information sharing between police and victim service providers; informs police about high-risk victims and offenders; links victims with other essential services …there is a belief that each service provider, including the police, has a common interest in ensuring victim safety and de-motivating the offender May require extensive discussions by parties to define roles, responsibilities, and limits of partnership; collaboration requires agreement about confidentiality issues
8 Assessing the threat of repeat victimization Determines need for immediate protection ofvictim and apprehension ofoffender …officers/collaborators trained to assess revictimization threats Requires training and timely and accurate intelligence information
9 Arresting offenders Incapacitates offender during high-risk periods and deters potential and actual offenders …a graded response to battering is adopted depending on the likelihood of rebattering; used with situational crime prevention opportunity blocking framework Under some conditions arrest may increase risk of revictimization; some offenders undeterred by arrest
10 Issuing and enforcing restraining orders Removes excuses for offender and victim to come into contact with one another …police recognize that defiance of a restraining order may be an indicator of future risk to the victim Police see violation of a restraining order as the need for a victim safety plan, and the adoption of a graded response to both victim and offender depending on the circumstances
11 Aggressively pursuing criminal prosecution ofsevere domestic violence cases and publicizing convictions Incapacitates offender and deters potential and actual offenders …police and prosecutors can agree beforehand what constitutes chronic and severe offenders; used as part of pulling levers/focused deterrence approach or other graded responses to batterers Requires proper evidence collection; should also include coordination and prosecutor participation around victim safety
12 Establishing special domestic violence courts Enhances judges’ knowledge ofparticular victims and offenders and ability to monitor compliance with court orders …courts participate in an evaluation to improve knowledge about recidivism reduction; courts can discern between batterers who can be deterred and those who can’t May require extra court resources
13 Providing treatment for batterers Reduces batterers’ propensity for violence …treatment is court-ordered; treatment has proven effectiveness and is tailored to a specific type ofbatterer Requires up-todate knowledge ofeffectiveness of different treatment approaches
Responses With Limited Effectiveness
# Response How It Works Works Best If… Considerations
14 Arresting both parties in a domestic violence incident Incapacitates both parties during high-risk period   Consumes scarce jail and court resources and not generally recommended for reasons discussed earlier in this guide


[1] Rennison (2003)[Full Text].

[2] Tjaden and Thoennes (2000)[Full Text]; Rennison and Welchans (2000)[Full Text].

[3] Tjaden and Thoennes (2000)[Full Text]; Rennison and Welchans (2000)[Full Text].

[4] Tjaden and Thoennes (2000)[Full Text]; Greenfeld et al. (1998)[Full Text].

[5] For an excellent discussion of the challenges these different definitions cause in making findings across research studies and surveys, see Mears (2003) and Buzawa and Buzawa (2003).

[6] Tjaden and Thoennes (2000)[Full Text].

[7] Rennison (2003)[Full Text].

[8] Rennison (2001)[Full Text].

[9] Fox and Zawitz (2002)[Full Text].

[10] Fox and Zawitz (2002)[Full Text].

[11] See Straus (2005) and Buzawa and Buzawa (2003) for a review.

[12] Rennison and Welchans (2000)[Full Text].

[13] Straus (2005); Moffit and Caspi (1999)[Full Text].

[14] Straus (2005).

[15] Tjaden and Thoennes (2000)[Full Text]; Straus (2005).

[16] Tjaden and Thoennes (2000)[Full Text].

[17] Carrington and Phillips (2003).

[18] Charlotte-Mecklenburg Police Department (2002).

[19] For a summary, see Brewster (2002).

[20] Hotaling et al. (1989).

[21] For a thorough review of the research in support of this theory, see Kennedy (2002). Also see Hotaling et al. (1989) for the results of several studies and Buzawa and Buzawa (2003) for a general description of study findings about this topic.

[22] Dekeseredy et al. (1997).

[23] Straus (2005).

[24] For a summary, see Brewster (2002).

[25] Greenfeld et al. (1998)[Full Text]; Rennison and Welchans (2000)[Full Text].

[26] Dekeseredy et al. (1997).

[27] Hickman et al. (2004). For a detailed discussion of the challenges of measuring teen dating violence and the results of different studies that have attempted to, see Hickman et al. (2004).

[28] Rennison and Welchans (2000)[Full Text].

[29] Rennison and Welchans (2000)[Full Text].

[30] Lloyd (1998).

[31] Rennison and Welchans (2000)[Full Text].

[32] Tjaden and Thoennes (2000)[Full Text].

[33] Hanmer et al. (1999)[Full Text].

[34] Tjaden and Thoennes (2000)[Full Text].

[35] Hanmer et al. (1999)[Full Text].

[36] Hanmer et al. (1999)[Full Text]. Also see Buzawa and Buzawa (2003) describing a study one of the authors conducted with others in Quincy District Court (Massachusetts) finding a similar pattern.

[37] Greenfeld et al. (1998)[Full Text].

[38] Greenfeld et al. (1998)[Full Text].

[39] Tjaden and Thoennes (2000)[Full Text].

[40] Jasinski (2001a); Jasinski (2001b); for a summary of the research related to pregnancy and domestic violence, see U.S. General Accounting Office (2002)[Full Text].

[41] Jasinski (2001a); Jasinski (2001b); for a summary of the research related to pregnancy and domestic violence, see U.S. General Accounting Office (2002)[Full Text].

[42] Summary of research on pregnancy and domestic violence, U.S. General Accounting Office (2002)[Full Text].

[43] Jasinski (2001b).

[44] Rennison and Welchans (2000)[Full Text].

[45] Rennison and Welchans (2000)[Full Text].

[46] Moffitt and Caspi (1999)[Full Text].

[47] Moffitt and Caspi (1999)[Full Text].

[48] Tjaden and Thoennes (2000)[Full Text].

[49] Tjaden and Thoennes (2000)[Full Text].

[50] Tjaden and Thoennes (2000)[Full Text].

[51] Ascione (2004).

[52] Tjaden and Thoennes (2000)[Full Text].

[53] Lloyd et al. (1994)[Full Text].

[54] Buzawa and Buzawa (2003).

[55] Buzawa and Buzawa (1996).

[56] Hanmer et al (1999)[Full Text]; Anderson et al (1995)[Full Text]; Bridgeman and Hobbs (1997)[Full Text]; Chenery et al (1997)[Full Text]; Lloyd et al (1994)[Full Text]; Pease (1998)[Full Text]. For the application of a graded approach to reducing repeat domestic violence victimization in the United States, see the description of the Fremont Police Department approach in Sampson and Scott (2000) [Full Text].

[57] Hanmer et al. (1999)[Full Text]. Also see Buzawa and Buzawa (2003) describing a study one of the authors conducted with others in Quincy District Court (MA) finding a similar pattern.

[58] Hanmer et al. (1999)[Full Text].

[59] Hanmer et al. (1999)[Full Text].

[60] Hanmer et al. (1999)[Full Text].

[61] Kennedy (2002).

[62] Buzawa and Buzawa (1996).

[63] U.S. General Accounting Office (2002)[Full Text].

[64] Hickman et al. (2004).

[65] Felson, Ackerman, and Gallagher (2005).

[66] See adopted recommendations in American Medical Association Data on Violence Between Intimates (2000).

[67] For a summary of the research on physician screening for domestic violence, see U.S. General Accounting Office (2002)[Full Text].

[68] Roberts (2002a).

[69] Brookoff et al. (1997).

[70] See Bruno v. Codd 47 N.Y. 2d 582, 393 N.E. 2d 976, 419 N.Y.S. 2d 901 [1979]; and Thurman v. City of Torrington 595 F. Supp 1521 [1984].

[71] Sherman and Berk (1984).

[72] Roberts and Kurst-Swanger (2002).

[73] Bureau of Alcohol, Tobacco, Firearms and Explosives (2004) [Portable Document FormatMisdemeanor Crimes of Domestic Violence Full Text ],[Portable Document FormatProtection Orders Full Text ] .

[74] Roberts (2002a).

[75] Sherman and Berk (1984).

[76] For a discussion of the replication studies, see Gelles (1996) and Fagan (1996)[Full Text].

[77] Maxwell, Garner and Fagan (2001)[Full Text].

[78] Maxwell, Garner and Fagan (2001)[Full Text].

[79] Maxwell, Garner and Fagan (2001)[Full Text].

[80] Maxwell, Garner and Fagan (2001)[Full Text].

[81] Felson, Ackerman, and Gallagher (2005).

[82] Loue (2001).

[83] Buzawa and Buzawa (2003).

[84] Loue (2001).

[85] Buzawa and Buzawa (2003).

[86] Buzawa and Buzawa (2003) discuss the empirical research from two mandatory prosecution jurisdictions and the arguments for and the drawbacks to “no drop” policies.

[87] Smith et al. (2001)[Full Text]; Buzawa and Buzawa (2003).

[88] Mazur and Aldrich (2003).

[89] Buzawa and Buzawa (2003).

[90] Brewster (2002).

[91] Brewster (2002).

[92] Brewster (2002).

[93] U.S. General Accounting Office (2002)[Full Text].

[94] Jackson (2003)[Full Text].

[95] Jackson (2003)[Full Text]; Feder and Forde (2003)[Full Text].

[96] Jackson et al. (2003)[Full Text].

[97] Fagan (1996)[Full Text]; Buzawa and Buzawa (1996).

[98] Healey et al. (1998)[Full Text].

[99] Fagan (1996)[Full Text]; also see Healey et al. (1998)[Full Text] and Buzawa and Buzawa (2003).

[100] Kruttschnitt et al. (2003).

[101] Buzawa and Buzawa (2003).

[102] Healey et al. (1998)[Full Text].


American Medical Association Data on Violence Between Intimates (2000). Available at www.amaassn.org/ama/pub/category/13577.html. Adopted recommendations available at www.amaassn.org/ama/ pub/category/13577.html#RECOMMENDATION.

Anderson, D., S. Chenery, and K. Pease (1995). Biting Back: Tackling Repeat Burglary and Car Crime. Crime Detection and Prevention Series, Paper 58. London: Home Office. [Full Text]

Ascione, F. (2004). Children and Animals: Exploring the Roots of Kindness and Cruelty. West Lafayette (Indiana): Purdue University Press.

Australian Institute of Health and Welfare (2005). "Female SAAP Clients and Children Escaping Domestic and Family Violence 2003-2004." Bulletin No. 30, AIHW Cat. No., AUS 64. Canberra. [Full Text]

Brewster, M. (2002). “Domestic Violence Theories, Research, and Practice Implications.” In A. Roberts (ed.), Handbook of Domestic Violence Intervention Strategies. Oxford and New York: Oxford University Press.

Bridgeman, C., and L. Hobbs (1997). Preventing Repeat Victimisation: The Police Officers’ Guide. Police Research Group. London: Home Office. [Full Text]

Brookoff, D., K. O’Brien, C. Cook, T. Thompson, and C. Williams (1997). “Characteristics of Participants in Domestic Violence: Assessment at the Scene of Domestic Assault.” JAMA: The Journal of the American Medical Association, 277(17):1369-1373.

Bruno v. Codd, 47 N.Y. 2d 582, 393 N.E. 2d 976, 419 N.Y.S. 2d 901 (1979).

Bureau of Alcohol, Tobacco, Firearms, and Explosives (2004). Relevant regulations related to 18 U.S.C. 922(g) and (n), 27 CFR 178.32(a) and (b), and 924(a)(24). [Misdemeanor Crimes of Domestic Violence Full Text ],[Protection Orders Full Text ]

Buzawa, E., and C. Buzawa. (2003). Domestic Violence: The Criminal Justice Response (3rd ed). Thousand Oaks, London, and New Delhi: SAGE Publications.

——— (1996). Do Arrests and Restraining Orders Work? Thousand Oaks, London, and New Delhi: SAGE Publications.

Carrington, K., and J. Phillips (2003). "Domestic Violence in Australia—an Overview of the Issues, E-Brief." Available at www.aph.gov.au/library/intguide/SP/Dom_violence. htm.

Charlotte-Mecklenburg (North Carolina) Police Department (2002). "Baker One Domestic Violence Intervention Project: Improving Response to Chronic Domestic Violence Victims." Finalist for the Herman Goldstein Award for Excellence in Problem-Oriented Policing. Available at www.popcenter.org/library/ goldstein/2002/02-09(F).pdf.

Chenery, S., J. Holt, and K. Pease (1997). Biting Back II: Reducing Repeat Victimisation in Huddersfield. Crime Detection and Prevention Series, Paper 82. London: Home Office. [Full Text]

Dekeseredy, W., D. Saunders, M. Schwartz, and S. Alvi (1997). “The Meanings and Motives for Women’s Use of Violence in Canadian College Dating Relationships: Results from a National Survey.” Sociological Spectrum Vol. 17(2):199-222.

Fagan, J. (1996). The Criminalization of Domestic Violence: Promises and Limits. National Institute of Justice Research Report. [Full Text]

Feder, L., and D. Forde (2003). “The Broward Experiment.” In Jackson, S., L. Feder, D. Forde, R. Davis, C. Maxwell, and B. Taylor (eds.), Batterer Intervention Programs: Where Do We Go From Here?. National Institute of Justice, Special Report. Washington, D.C.: U.S. Department of Justice, National Institute of Justice. [Full Text]

Felson, R., J. Ackerman, and C. Gallagher (2005). “Police Intervention and the Repeat of Domestic Assault.” Criminology, Vol. 43(3):563-588.

Fox, A., and M. Zawitz (2002). “Homicide Trends in the United States.” Washington, D.C.: U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics. [Full Text]

Fridell, L., and A. Pate (1997). “‘Death on Patrol’ Killings of American Law Enforcement Officers.” In R. Dunham and G. Alpert (eds.), Critical Issues in Policing (3rd ed.), Waveland Press.

Gelles, R. (1996). “Constraints Against Family Violence: How Well Do They Work?” In Buzawa, E. and C. Buzawa (eds.), Do Arrests and Restraining Orders Work? Thousand Oaks, London, New Delhi: SAGE Publications.

Greenfeld, L., M. Rand, D. Craven, P. Klaus, C. Perkins, C. Ringel, G. Warchol, C. Matson, and J. Fox (1998). Violence by Intimates: Analysis of Data on Crimes by Current or Former Spouses, Boyfriends, and Girlfriends. Bureau of Justice Statistics Factbook, Washington, D.C.: U.S. Department of Justice, Bureau of Justice Statistics. [Full Text]

Hanmer, J., S. Griffiths, and D. Jerwood (1999). Arresting Evidence: Domestic Violence and Repeat Victimization. Police Research Series, Paper 104. London: Home Office, Police Research Group. [Full Text]

Healey, K., C. Smith, with C. O’Sullivan (1998). Batterer Intervention: Program Approaches and Criminal Justice Strategies. Washington, D.C.: U.S. Department of Justice, National Institute of Justice. [Full Text]

Hickman, L., L. Jaycox, and J. Aranoff (2004). “Dating Violence among Adolescents: Prevalence, Gender Distribution, and Prevention Program Effectiveness.” Trauma, Violence, & Abuse, Vol. 5, No. 2. RAND Reprint, originally published by SAGE Publications.

Hotaling, G., M. Straus, and A. Lincoln (1989). “Intrafamily Violence, and Crime and Violence Outside the Family.” In O. Lloyd and M. Tonry (eds.), Family Violence. University of Chicago Press.

Isaac, N., and V. Pualani Enos (2001).  “Documenting Domestic Violence: How Health Care Providers Can Help Victims.” Research in Brief. Washington, D.C.: U.S.  National Institute of Justice. [Full Text]

Jackson, S. (2003). “Analyzing the Studies.” In Jackson, S., L. Feder, D. Forde, R. Davis, C. Maxwell, and B. Taylor, Batterer Intervention Programs: Where Do We Go From Here?  Special Report. Washington, D.C.: U.S. National Institute of Justice. [Full Text]

Jackson, S., L. Feder, D. Forde, R. Davis, C. Maxwell, and B. Taylor (2003). Batterer Intervention Programs: Where Do We Go From Here? National Institute of Justice, Special Report. Washington, D.C.: U.S. Department of Justice, National Institute of Justice. [Full Text]

Jasinski, J. (2001a). “Pregnancy and Violence Against Women: An Analysis of Longitudinal Data.” Journal of Interpersonal Violence 16(7):712-733.

——— (2001b). “Pregnancy, Stress and Wife Assault: Ethnic Differences in Prevalence, Severity, and Onset in a National Sample.” Violence and Victims 16(3):219-232.

Kennedy, D. (2002). “Controlling Domestic Violence Offenders.” Paper prepared for the Hewlett-Family Violence Prevention Fund. Available from author.

Kennedy, D., E. Waring, and A. Piehl, (2001). “Problemoriented Policing, Deterrence, and Youth Violence: An Evaluation of Boston’s Operation Ceasefire.” Journal of Research in Crime and Delinquency 38(3):195-225.

Kruttschnitt, C., B. McLaughlin, and C. Petrie (2003). Advancing the Federal Research Agenda on Violence Against Women. Washington, D.C.:  National Academies Press. Available at books.nap.edu/openbook/0309091098/html/index.html

Lloyd, S. (1998). “Domestic Violence and Women’s Employment.” NU Policy Research, electronic journal of the Institute of Policy Research at Northwestern University, available at www.northwestern.edu/ipr/ publications/nupr/nuprv03n1/lloyd.html.

Lloyd, S., G. Farrell, and K. Pease (1994). Preventing Repeated Domestic Violence: A Demonstration Project in Merseyside. Police Research Group, Crime Prevention Unit, Paper 49. London: Home Office. [Full Text]

Loue, S. (2001). Intimate Partner Violence: Societal, Medical, Legal, and Individual Responses. New York: Kluwer Academic/Plenum Publishers.

Maxwell, C., J. Garner, and J. Fagan (2001). "The Effects of Arrest on Intimate Partner Violence: New Evidence from the Spouse Assault Replication Program." Research in Brief.  Washington, D.C.: U.S. National Institute of Justice. [Full Text]

Mazur, R., and L. Aldrich (2003). “What Makes a Domestic Violence Court Work.” Judges’ Journal 42(2):5-9, 41-42.

Mears, D. (2003). “Research and Interventions to Reduce Domestic Violence Revictimization.” Trauma, Violence, & Abuse 4(2):127-147.

Minnesota Program Development, Inc. (n.d.). “The Duluth Model.” Available at www.duluth-model.org/recentresearch.htm (accessed June 23, 2006).

Moffitt, T., and A, Caspi (1999). “Findings about Partner Violence from the Dunedin Multidisciplinary Health and Development Study.”  Research in Brief. Washington, D.C.: U.S. National Institute of Justice. [Full Text]

National Center for State Courts (n.d.). “Problem Solving Courts FAQs.” Available at www.ncsconline.org/WC/FAQs/SpeProFAQ.htm (accessed June 29, 2006).

National Coalition for the Homeless (2004). “Who is Homeless?” National Coalition for the Homeless, Fact Sheet No. 3. [Full Text]

Office of the Deputy Prime Minister (n.d.). “Homelessness Statistics: September 2002 and Domestic Violence.” Available at www.communities.gov.uk/index.asp?id=1149672 (accessed August 22, 2006).

Pease, K. (1998). Repeat Victimisation: Taking Stock. Crime Detection and Prevention Series, Paper 90. London: Home Office Police Research Group. [Full Text]

Pence, E., and M. Paymar (1993). “Domestic Violence Information Manual: The Duluth Domestic Abuse Intervention Project.” Springer Publishing Company, Inc. Available at www.eurowrc.org/05.education/education_en/12.edu_en.htm

Rennison, C., and S. Welchans (2000). “Intimate Partner Violence.” Washington, D.C.: U.S. Department of Justice, Bureau of Justice Statistics. [Full Text]

——— (2001, revised 09/18/02). “National Crime Victimization Survey. Criminal Victimization 2001: Changes 2000-2001 with Trends 1993-2001.” Washington, D.C.: U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics. [Full Text]

Rennison, C., and S. Welchans (2000). “Intimate Partner Violence.” Washington, D.C.: U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics. [Full Text]

Roberts, A. (2002a). Handbook of Domestic Violence Intervention Strategies. Oxford and New York: Oxford University Press.

——— (2002b). “Police Response to Battered Women.” In A. Roberts (ed.), Handbook of Domestic Violence Intervention Strategies. Oxford and New York: Oxford University Press.

Roberts, A. (2002a). Handbook of Domestic Violence Intervention Strategies: Policies, Programs, and Legal Remedies.  New York: Oxford University Press.

Sampson, R., and M. Scott (2000). “Domestic Violence in Fremont, Calif.”  In Tackling Crime and Other Public Safety Problems: Case Studies in Problem-Solving. Washington, D.C.: U.S. Department of Justice, Office of Community Oriented Policing Services, [Full Text]

San Diego Family Justice Center. www.familyjusticecenter.org (accessed June 27, 2006).

Sherman, L., and R. Berk (1984). “The Specific Deterrent Effects of Arrest for Domestic Assault.” American Sociological Review 49(2):261-272.

Smith, B., R. Davis, L. Nickles, and H. Davies (2001). Evaluation of Efforts to Implement No-Drop Policies: Two Central Values in Conflict.  Unpublished Final Report. [Full Text]

Spelman, W. (1990). Repeat Offender Programs for Law Enforcement. Washington, D.C.: Police Executive Research Forum.

Straus, M. (2005). “Women’s Violence Towards Men is a Serious Social Problem.” In D. Loeske, R. Gelles, and M. Cavanaugh (eds.), Current Controversies on Family Violence (2nd ed.). Thousand Oaks, Calif.: Sage.

Thurman v. City of Torrington, 595 F. Supp 1521 (1984).

Tjaden, P., and N. Thoennes (2000). Extent, Nature, and Consequences of Intimate Partner Violence. Washington, D.C.: U.S. National Institute of Justice and Centers for Disease Control.[Full Text]

U.S. General Accounting Office (2002). Violence Against Women: Data on Pregnant Victims and Effectiveness of Prevention Strategies are Limited. Report to the Honorable Eleanor Holmes Norton, House of Representatives. GAO-02-530. Washington, D.C.: GAO. [Full Text]

Related POP Projects


The quality and focus of these submissions vary considerably. With the exception of those submissions selected as winners or finalists, these documents are unedited and are reproduced in the condition in which they were submitted. They may nevertheless contain useful information or may report innovative projects.

Aiming to Reduce Repeat Victimisation for Domestic Violence, Lancashire Constabulary (Lancashire, UK), 2000

Baker One Domestic Violence Intervention Project [Goldstein Award Finalist], Charlotte-Mecklenburg Police Department (NC, US), 2002

CMHAPD Police Assisted Referral Program, Cuyahoga Metropolitan Housing Authority Police Department (OH, US), 2011

Chronic Domestic Disturbance, Port St. Lucie Police Department (FL, US), 1995

Domestic Conflict Unit, Calgary Police Service (AB, CA), 1998

Domestic Violence Family Justice Center, San Diego Police Department (CA, US), 2003

Domestic Violence Re-victimization Prevention [Goldstein Award Finalist], Fremont Police Department (CA, US), 1997

Family Violence Followup Teams [Goldstein Award Finalist], Edmonton Police Service (AB, CA), 1994

Finding Hope: Bringing Transformation to Families of Domestic Violence, Citrus Heights Police Department (CA, US), 2011

Foyle Domestic Violence Protocol, Foyle District Command Police (Londonderry, Northern Ireland, UK), 2007

Hollywood Area Domestic Violence, Los Angeles Police Department (CA, US), 1996

Increasing Reporting of Domestic Violence, Devon and Cornwall Constabulary (Exeter, UK), 1999

Know Your Role, Cleveland Police Department (Middlesbrough, UK), 2007

Project Recover, Illinois State Police (IL, US), 1998