LAW ENFORCEMENT ATTITUDES REGARDING ISSUES RELATED TO
A CITY AND COUNTY DEPARTMENT COMPARISON
B.A., University of Florida, 2005
A thesis submitted to the
University of Colorado Denver
in partial fulfillment
of the requirements for the degree of
Master of Criminal Justice
School of Public Affairs
This thesis for the Master of Criminal Justice
has been approved
Dagmar Pudrzynska (Master of Criminal Justice, School of Public Affairs)
Law Enforcement Attitudes Regarding Issues Related To Domestic Violence: A City and
County Department Comparison
Thesis directed by Associate Professor Dr. Angela R. Gover
Domestic violence is a serious crime that does not discriminate; it affects people in
every community, regardless of gender, race, class, or sexual orientation. Studies have
shown that over the years criminal justice agencies have changed their attitudes toward this
crime, taking it out of the privacy of the victims home and into the public sphere. Today there
are many services that domestic violence victims can turn to for help. Most of the time the
first organization contacted about a domestic violence incident is a police department. When
police officers respond to domestic violence cases there are a multitude of factors they have
to be ready to address, while at the same time instituting their departments policy regarding
this crime. It is important to examine attitudes of police officers about domestic violence,
since their attitudes will most likely influence the way in which officers respond to these
The main purpose of the study was to conduct a comparison between officer
attitudes about domestic violence from two agencies. The current study examines a sample
of 292 police officers from one city and one county law enforcement agency. Officers
responded to a self-report survey that asked questions about their response to domestic
violence in general, mandatory arrest, victim and offender behavior, officer frustrations, and
domestic violence dynamics. Univariate and bivariate analyses were conducted to examine
officer attitudes and compare them based on gender, race, rank, age, and experience. A
multivariate analysis was conducted using ordinary least squares regression. Findings show
that even though overall attitudes are more similar than different across the two types of
agencies, there are significant variations in officer attitudes based on demographic factors.
Study limitations, directions for future research propositions and policy implications are
I would like to thank my committee for their commitment to this project, especially my
chair, Dr. Angela R. Gover. Dr. Gover has provided me with incredible knowledge, support,
and understanding throughout the entire process. I also want to thank my husband, Alex,
who is my rock and always supports me in everything that I do. I would like to dedicate this
work to my Mom, Wteslawa, without whom I wouldnt be the person I am today.
TABLE OF CONTENTS
2. REVIEW OF RELEVANT LITERATURE.........................................6
Responding to Domestic Violence: A Historical Overview...........6
The Birth of Arrest as a Solution to Domestic Violence...........7
Mandatory Arrest: An Ongoing Controversy.........................9
An Overview of Police Attitudinal Studies........................12
Domestic Violence: A Real Crime..................................14
Arrest of Limited Value in Reducing Domestic Violence............15
Domestic Violence Victims Attitudes Toward the Police...........18
Consequences of Domestic Violence................................19
Officers Struggle to Understand Victims Actions and Attitudes.23
Police Attitudes and Actions Regarding Offenders.................25
Other Sources of Officer Frustration.............................27
The Need for More Training.......................................29
Other Findings of Attitudinal Studies............................31
Perceptions about Lethality....................................32
Influences of Demographic Characteristics on Police Attitudes..33
Comparison of Attitudes between City and County Officers ........35
Description of Agencies and Communities..........................37
Open Ended Questions........................................41
Race and Rank...............................................51
Age and Years of Experience.................................55
5. DISCUSSION AND CONCLUSIONS........................................64
Current Findings Within the Context of Prior Literature.......64
A. SURVEY INSTRUMENT.................................................73
B. INFORMED CONSENT FORM.............................................76
C. CORRELATION MATRIX................................................77
LIST OF TABLES
4.1 Descriptive Statistics...............................................43
4.2 Univariate Analysis..................................................46
4.3 T-Test for Attitudes and Gender......................................49
4.4 T-Test for Attitudes and Race........................................51
4.5 T-Test for Attitudes and Rank........................................53
4.6 Correlation for Attitudes, Age, and Experience.......................55
4.7 T-Test for Attitudes and Agency Type.................................57
4.8 OLS Regression for Attitudes and Agency Type.........................61
The abuse of women in intimate relationships is a pervasive and underreported
problem throughout the United States (Reitzel-Jaffe & Wolfe, 2001). It was not until the
1970s that domestic violence was acknowledged by the legal community through the work of
many womens advocates. In 1965 Congress passed laws prohibiting discrimination against
women in employment and requiring equal pay for work, but the traditional marriage contract
still remained in tact. In New York, in 1966, beating, treated as a part of cruel and inhumane
treatment, became grounds for divorce but the plaintiff was still required to establish that a
sufficient number of beatings had taken place (Martin, 1976). In 1972 womens advocates
and the Haven House in Pasadena, California, were the first to establish government funded
shelters for battered women, and in 1976 the National Organization for Women announced
the formation of a task force to examine the problem of battering, and at the same time
demanding research into this problem and money for shelters (Richie & Menard, n.d.). The
passage of the Family Violence Prevention and Services Act in 1984, through grassroots
lobbying efforts, earmarked federal funding for programs serving victims of domestic
violence. A few years later, in 1987, the National Coalition Against Domestic Violence
established the first national toll-free domestic violence hotline (Richie & Menard, n.d ). In
1994 Congress passed the Violence Against Women Act, a part of the Federal Crime Victims
Act, which funds services for victims of rape, domestic violence; allows women to seek civil
rights remedies for gender-related crimes; and provides training to increase police and court
officials sensitivity toward this issue (Heinemann, 1996).
The work that has been conducted over the past 40 years has led domestic violence
to be regarded as a public heath epidemic (Osofsky, 1999, p.33); this comes as no surprise
upon an examination of the shocking statistics that have emerged over the past quarter
century. Domestic violence is a crime that does not discriminate-it affects every social class,
race, age, and gender. According to Rodriguez, Bauer, McLoughlin, and Grumbach (1999),
domestic violence is estimated to occur in 4 to 6 million relationships each year in the United
States. Other research has estimated that nearly 25% of women have been a victim of
domestic violence at some point in their lives, and more than 40% of these women sustain
physical injuries (Tjaden & Thoennes, 2000). According to a recent study analyzing the
National Violence Against Women Survey, Miller (2006) found that approximately 19% of
women in the general population were victims of physical abuse at the hands of their
partners. In addition to physical assault, Miller (2006) also examined the prevalence of
stalking and rape in physically violent relationships and found that all three forms of violence
were highly associated. In addition to physical injuries being a common consequence of
domestic violence, mental health consequences are highly prevalent. Domestic violence
victims often suffer from symptoms such as depression, anxiety, lowered self-esteem, and
post-traumatic stress disorder. Even after escaping a domestic violence situation, survivors
may still be affected by such symptoms as flashbacks and suicidal ideation (Jones, 2000). In
other words, the effects of domestic violence do not just go away.
In addition to its highly prevalent nature, domestic violence is also considered to be
one of the most underreported crimes. Most domestic violence incidents are not reported to
the police (Tjaden & Thoennes, 2000). According to the National Violence Against Women
Survey, only about 25% of domestic violence assaults, 20% of domestic violence rapes or
sexual assaults, and 50% of domestic violence related stalking incidents are reported to law
enforcement agencies. The majority of victims who did not report their victimization to law
enforcement thought the police would not or could not do anything on their behalf (Tjaden &
Thoennes, 2000). Many people believe that available data greatly underestimate the true
magnitude of the problem. Overall, Edleson (1999) argues that the prevalence of domestic
violence has not changed as much as its visibility to the public, one way for domestic
violence to further increase its visibility is through contact with law enforcement. Police
officers play a crucial part in the domestic violence phenomenon. Many times the police are
the only ones that are notified if abuse is occurring in the home, they are the government
entity that the victim relies upon for help in this type of dangerous situation. It is extremely
critical to understand law enforcement officers' attitudes about domestic violence for the
development of continual training curricula regarding this prevalent problem.
The current research was modeled after Toon and Harts (2005) law enforcement
attitude study conducted among Arizona police officers. The aim of Toon and Harts (2005)
study was to report and analyze the experiences and opinions of police officers regarding
domestic violence, as a contribution to the ongoing debate about how the police should
respond to domestic violence calls. This study follows Toon and Harts (2005) model in
exploring police attitudes regarding domestic violence, using a sample of law enforcement
officers from one county and one city police department in a major metropolitan area of a
Western state. Researchers have noted that it is essential to understand how actual police
officers perceive their job and its difficulties, and by evaluating these attitudes administrators
are able to find ways to improve officers ability to respond effectively to incidents of domestic
violence (Johnson, 2004). Police officers play an extremely important role in domestic
violence incidents and many times have to master more than one course of action when
responding to these disputes. Studying police officer attitudes is crucial because they are
usually faced with the responsibility of diffusing domestic violence situations by protecting the
victim, keeping the peace, and enforcing the law. If officers have negative attitudes about
responding to domestic violence calls and domestic violence victims in general, this may
influence the outcome of the law enforcement response and victims inclination to seek help
though the justice system in the future (Logan, Shannon, & Walker, 2006).
Researching law enforcement attitudes is especially important to the development of
policy because exploring opinions can help the criminal justice system develop new ways to
improve current policies, since law enforcement officers are the systems first responders. It
is also important to remember that if someone disagrees with a policy, he or she may be
reluctant to follow and enforce its laws.
A thorough review of the social science literature identified a gap about direct
opinions of police officers regarding the treatment of domestic violence cases. The current
research follows Toon and Harts (2005) model and attempts to fill in the research gap by
directly soliciting the views of a large number of active-duty police officers. Generally, police
officers have historically held negative attitudes regarding domestic violence, victims, and law
enforcements responsibility to them. The view of domestic violence as a crime and accepted
responses to violations of domestic violence laws have changed over time. It is important to
continue monitoring the way that law enforcement officer attitudes are changing in order to
determine whether current policies are, in reality, being adopted and implemented. In
addition to partially replicating Toon and Harts (2005) research, this study contributes to the
law enforcement and domestic violence literature by comparing attitudes of police officers
from a county and city department. Although attitudinal comparisons have been made in the
criminal justice literature (i.e., rural vs. urban officer attitudes), this specific type of county
versus city department comparison has not been made. While this study is exploratory in
nature, it is important to examine whether or not there are differences among officers based
on their employment in two different geographically based departments.
The following chapter provides an in-depth review of the literature concerning law
enforcement attitudes regarding domestic violence. It focuses on the historical approach to
domestic violence, utilization of arrest in reducing domestic violence, victim-officer interaction,
victim attitudes toward police officers, and officer frustrations regarding discretion,
prosecution, and offenders. The third chapter discusses the methodology used to analyze
the data obtained from a city and county police department in a Western state. Chapter four
presents the results from the univariate, bivariate, and multivariate analyses. Chapter five
provides a discussion of this studys findings in terms of how the results relate to prior
research. In addition, chapter five also discusses the limitations of the current study and the
implications of the current findings on research, policy, and practice.
REVIEW OF RELEVANT LITERATURE
Responding to Domestic Violence Incidents: A Historical Overview
Prior to the 1980s, the law enforcement response to domestic violence was one of
annoyance, insignificance, and dismissal. This type of negative response began even before
the responding officer received the call. After calling 911, the first person that a victim would
have contact with would be the dispatcher. Prior to the 1980s, dispatchers treated domestic
disturbance calls as anything but serious violence. According to Pamas (1967), dispatchers
would commonly try and talk victims out of pursuing an arrest warrant. They would often fail
to ask for important details, such as whether a weapon was involved in the incident.
Dispatchers were instructed to classify domestic violence calls as batteries and
disturbances based on whether contact was made between the two parties (Pamas, 1967).
Some dispatchers believed that since most domestic violence cases did not make it to the
trial or sentencing stages, even after an arrest was made, they felt that there was no need to
burden their fellow officers with such calls (Pamas, 1967). Even the International Association
of Chiefs of Police took the position that when responding to domestic disputes, arrest should
be exercised as law enforcements last resort, with mediation being the first approach
Buzawa and Buzawa (2003) note that prior to the 1980s, only a small proportion of
intimate partner assaults resulted in the dispatch of police officers because the majority of
cases were diverted out of the system. It is clear that law enforcement agencies had their
own call screening protocol because they considered domestic violence calls to be a low
priority, which posed a severe challenge to an effective response to these (Buzawa &
Buzawa, 2003). Martin (1979) found that this type of call screening was sometimes used to
eliminate certain categories of calls, especially during peak time periods such as weekends
and nights. Since domestic violence was found to occur during these peak times, the way in
which calls were screened may have discouraged victims from contacting police in future
violent incidents, since many had already been redirected to social service agencies or were
told that the police could not assist them (Martin, 1979).
During the 1970s, views on law enforcement responses to domestic violence
incidents began to change. Domestic violence advocates, feminists, and even some law
enforcement officials began to think that the lackadaisical way in which police had been
responding to domestic violence calls needed to improve. During the early 1980s, policing
scholars Sherman and Berk conducted a study involving an experimental design that
addressed the issues about what needed to be done to improve the police response and
reduce the recidivism rates among domestic violence offenders. Results from their study
provided the initial push to respond to domestic violence cases as more than just menial
disturbances. Sherman and Berks (1984) findings led to changes in domestic violence
policies within law enforcement agencies throughout the United States.
The Birth of Arrest as a Solution to Domestic Violence
Over two decades ago, Sherman and Berk (1984) employed a classical experimental
design in their research, a remnant of an era before strict restrictions were placed on
researchers. They worked with a police agency in Minneapolis to determine which police
response (arrest, separation, or mediation) worked the best in reducing recidivism rates
among domestic violence offenders. Sherman and Berks (1984) sample consisted of
misdemeanor domestic assault cases where the victim and the offender were both present
when the police arrived. The cases also had to be ones in which the police officer had
probable cause to believe that a cohabitant or spouse had assaulted the victim within the last
four hours (Sherman, 1992). When police officers responded to these cases they had
colored cards with the three response choices on them, and they used them in the random
order in which they were organized. These cards determined whether the officer response
would be to arrest the offender, separate the parties, or provide mediation to the couple
(Sherman, 1992). Official results from Sherman and Berks (1984) study suggested that
arrest had the lowest recidivism (13%), compared to separation (26%), and mediation (19%).
The Minneapolis Domestic Violence Experiment influenced law enforcement
agencies across the country to move toward pro-arrest and mandatory arrest policies when
responding to domestic violence cases. In addition to critiques from scholars (Binder &
Meeker, 1993), Sherman and Berk (1984) noted that their experiment was far from being
representative of all areas and that replications of their research were necessary. They
proposed that in agencies similar to that of the Minneapolis Police Department, where the
presumption of arrest was favored when responding to domestic violence cases, arrests
should be made unless there were other factors clearly present that would suggest otherwise.
In the replication studies that followed, specifically the Spouse Assault Replication
Program founded by the National Institute of Justice, researchers concluded that their results
were inconsistent across the five cities documenting whether arrest was the best option for
reducing recidivism (Sherman, 1993). Only three out of the six replication sites found arrest
to have the largest deterrent effect. Policy makers and police departments across the
country began to doubt the use of arrest as the primary and mandatory choice for domestic
violence incidents. In response to these inconclusive findings Berk, Campbell, Klap, and
Western (1992) suggested that the findings from the original experiment do not provide a
sound rationale for abandoning arrest, even presumptory arrest, as a policy option (p. 706).
Following the replication studies, the National Institute of Justice funded reserach to
pool all of the data from the replication studies to produce a more powerful estimation of the
impact of arrest on domestic violence (Maxwell, Garner, & Fagan, 2001). The pooled
analysis findings showed, from the victim interviews, that arrest was more of a deterrent than
non-arrest (separation and mediation). Official results, based on police data, also showed
that arrest had the strongest deterrent effect on subsequent offending. These results were
significant while controlling for the differences across sites, the victim interview processes
across sites, and suspect characteristics (Maxwell et al., 2001). Although the pooled analysis
did not produce deterrent results that were as powerful as Sherman and Berks (1984)
original experiment, it still showed that arrest was the best option when striving to achieve
lower rates of recidivism.
As the research on police attitudes presented below will indicate, it seems that even
though science has shown arrest to have the strongest relationship with reduced recidivism
many officers believe that having mandatory arrest policies diminishes their power when
responding to domestic violence incidents by taking away one of their most valuable
resourcestheir discretion. Police officers are not the only ones that feel that mandatory
arrest may not be the best choice to make when responding to domestic violence incidents.
Mandatory arrest has been a controversial issue since its inception in many agencies and for
other reasons in addition to the loss of discretion. Several researchers suggest that
mandatory arrest policies may have a negative impact on domestic violence incidents in
terms of consequences experienced by victims.
Mandatory Arrest: An Ongoing Controversy
Although mandatory arrest is viewed by many as an appropriate response to
domestic violence by the criminal justice system, many people from the practice and research
communities recognize that there have been many unintended consequences for victims and
families involved in domestic violence incidents. As of the beginning of 2002, 23 states
implemented mandatory arrest laws for at least some assault and battery domestic violence
cases (Hirschel & Buzawa, 2002). As a consequence of the nationwide preference to use
mandatory and pro-arrest domestic violence policies, arrest rates for domestic violence
increased dramatically (Hirschel & Buzawa, 2002). When police utilize mandatory arrest laws
to detain both parties involved in an incident, they exercise what is termed dual arrest,
apprehending both as victim and offender. Policy makers anticipated an increase in arrest,
but what was less expected was a concurrent, and proportionally greater, increase in female
arrests. Not surprising, the increase in female arrests concerned practitioners and
researchers alike throughout the U.S. (Hirschel & Buzawa, 2002). Many scholars and
advocates noted that victims who have been unjustly arrested due to mandatory arrest
policies suffer a wide range of negative consequences. As a result of an unjust arrest,
victims may lose employment, experience financial hardship, lose custody of their children,
and be reluctant to call the police to report subsequent abuse despite a possible increase in
danger from the abuser (Hirschel & Buzawa, 2002).
According to White, Goldkamp, and Campbell (2005) proactive policies, such as
mandatory and pro-arrest policies, may have negative, unintended consequences for victims
involved in domestic violence disputes; especially the use of the judicial system as a weapon
that stigmatizes the victim. Chesney-Lind (2002) examined the impact of these laws on
womens arrests for domestic violence, particularly as aggressors. Specifically, Chesney-
Lind (2002) noted the increased numbers of women being arrested for domestic violence
following the implementation of mandatory and pro-arrest policies. For example, in Prince
William County, Maryland, the rates for arrests of females as domestic violence offenders
increased from 12.9% to 21 % over a three-year period of time (Chesney-Lind, 2002). In
other cities, such as Sacramento, the rise was even greater, with a 91% increase in women
arrested for domestic violence during a four-year time period. A California study conducted
by the Bureau of Criminal Information and Analysis (1999) showed that, over a ten year
period, the amount of female offenders arrested for domestic violence rose from 6% to
These statistics may point to the problem of police being unable to identify the
primary aggressor in domestic violence situations (Hirschel & Buzawa, 2002). The primary
aggressor in a domestic violence situation is the person who intentionally assaulted the other
party first. Comach, Chopyk, and Wood (2000) conducted a study in Canada, and reported
that in 35% of domestic violence cases involving women as aggressors, the accused women
had been the one to make the initial contact with the police. In an effort to prevent these
types of occurrences, many states have instituted laws against the use of dual arrest policies.
As of 2002, 24 states had laws that included either a primary or predominant aggressor
assessment which prohibited the police from arresting both parties in a domestic violence
case (Hirschel & Buzawa, 2002).
In addition to having general negative and unintended consequences, other
researchers note particular detrimental effects that these laws have had on minorities and
lower-class victims. Miller (1989) points out that these policies became even more
problematic when it was realized that they disproportionately affected women from minority
and lower socio-economic groups. She notes that both minority and lower-class women are
more likely to contact police in domestic violence cases. Chesney-Lind (2002) also agrees
that there is a racial disparity among women being arrested for domestic violence, where
African-American women and girls experience arrest rates approximately triple that of white
women and girls. Upper-class women, according to Miller (1989), have greater access to
resources that allow their private disputes with intimate partners to remain private. Therefore,
it is possible that minority and lower-class women are disproportionately affected by
mandatory and pro-arrest policies since they have fewer opportunities and alternatives
available to settle their disputes in private (Miller, 1989). These women also may be more
likely to experience more extreme economic consequences and therefore they may be more
likely to avoid calling the police if arrest is likely (Miller, 1989).
Miller and Meloy (2006) note that most practitioners and researchers agree that
battering must be explored and evaluated in a larger context by looking at the motivation,
meanings and consequences involved, not just at increased arrest rates for men and women.
In earlier research, Miller (1989) also pointed out the need for incorporating data that are
gathered from victims and professionals, such as shelter workers, hotline staff, legal
advocates, and counselors, in addition to law enforcement personnel and official statistics.
Miller (1989) states that responses to various criminal justice intervention strategies need to
be incorporated into empirical research in order to formulate ways to effectively intervene in
domestic violence cases when law enforcement is called upon. Chesney-Lind (2002)
stresses the point that even though domestic violence is a serious crime and social problem,
simplistic solutions (particularly ones that fail to address age, gender, and race inequality)
may result in crucial detrimental consequences for those who are at high risk of further
victimization. There is a continued need to assess the attitudes of those who are
implementing domestic violence policy, particularly the law enforcement officers who face
daily the decision of whether or not and whom to arrest.
An Overview of Police Attitudinal Studies
Several studies have examined police attitudes regarding law enforcement culture,
job satisfaction, and other issues that lead to officer frustration. It is important to keep these
studies in mind, as officer attitudes toward their job may affect the way they respond to
difficult calls such as domestic violence incidents. Yates and Pillais (1996) study on police
attitudes and community policing indicated that stress and job-induced strain were connected
to the officers attitudes about community policing. The more stress and strain that the
officers experienced the more likely they were to have negative attitudes toward community
policing. They also found that strain was influenced by the amount of support that the officers
received from administrators and governmental figures. The less support they received for
their community efforts, the more strain they experienced (Yates & Pillai, 1996). Findings
from this study suggest that when introducing a new philosophy into police practices, the new
approach needs to be supported by administration and government, and that levels of stress
and strain need to be monitored among police officers to ensure that attitudes about the new
approach remain positive. This line of thinking leads some to wonder how mandatory
policing policies were implemented as a method of policing and whether these policies
received widespread support by crucial members of government and other administrations.
Johnson (2004) conducted a qualitative study that focused on officer frustrations
when responding to domestic violence cases. He found that a significant source of officer
frustration was the way the police department, or certain key players (i.e., supervisors),
wanted officers to respond to domestic violence calls. Some officers in the study mentioned
that these differing types of attitudes from supervisors made it difficult for them to enforce the
domestic violence laws to their fullest extent. Again, these findings point to the importance of
obtaining widespread support when introducing new approaches into policing policy
response, not only from outside organizations but also within specific departments.
Paoline, Myers, and Worden (2000) conducted a study that examined law
enforcement culture, norms, and attitudes. In general, police culture focuses on crime
fighting and law enforcement, which potentially creates tensions between the community and
police and can lead to the abuse of power by the police. They also mention that the new
reforms in law, such as the introduction of community policing, disrupt accepted police culture
and are confronted with resistance by police officers. They hypothesized that even though
the police culture is present, that attitudes within it differ according to officer characteristics,
such as gender, race, education, and work experience. Paoline et al. (2000) note that even
before the introduction of community policing, as early as the 1970s, police attitudes
regarding the role of officers and the culture were not homogenous. In their findings, Paoline
et al. (2000) report that the notions of the traditional police culture were no longer as
prominently accepted among all the officers, as one would expect. They found that many
officers did not consider law enforcement to be their most important responsibility and many
others held a favorable outlook toward cooperation from the citizens. What was also
surprising was that the researchers failed to find differences in police attitudes based on
gender, race, education, or experience (Paoline et al., 2000). They conclude that police
culture may be capable of more change than it seems. In addition to general attitudes
shifting among police officers, attitudes and opinions regarding domestic violence have been
known to change over the years. It is crucial to continue investigating opinions among police
officers regarding these changing attitudes. The studies discussed below show a shift in a
positive direction among police officer attitudes regarding domestic violence overtime.
Domestic Violence: A Real Crime
Even with the negative and unintended consequences that some researchers have
empirically documented as being correlated with the implementation of mandatory arrest
policies, if it was not for this change in policy, domestic violence incidents may still be treated
as simple disturbances. A shift in perceptions has occurred, not only among civilians but also
among individuals who work in the field of law enforcement. It is therefore extremely vital to
continue exploring attitudes and document whether they are changing, as this information
may provide insight into improving police response to domestic violence incidents in the
Toon and Hart (2005) found that, as opposed to the historical view of domestic
violence calls, the majority of the police officers surveyed considered domestic violence to be
a major problem and a real crime. Toon and Harts (2005) findings confirmed Feders (1997)
earlier evidence that the officers no longer considered domestic violence to be a common
occurrence of family life. For example, 89% of the South Florida police officers surveyed
disagreed with the statement; Domestic calls are not real police work (Feder, 1997, p. 88).
Officers surveyed in this study showed strong support for police intervention as a response to
domestic violence incidents. Feder (1997) points out that these findings were not surprising
because this particular department viewed domestic violence as a serious issue. Supporting
Toon and Hart (2005) and Feder (1997), Robinson (2000) found that officers viewed arrest as
an appropriate police tactic, and viewed domestic violence as a public problem. Johnson
(2004) found that many officers in his study felt that domestic violence was a serious crime,
one which deserved to be treated just like any other violent crime. Johnson, Sigler, and
Crowleys (1994) earlier research showed that criminal justice personnel (including judges,
lawyers, and police officers) in Alabama thought that domestic violence is best treated as a
crime as opposed to a civil or social problem. Even though many police officers believe that
the appropriate response to domestic violence should involve legal authorities, many remain
unconvinced that arrest in particular is the best option in every situation. Sometimes officers
utilize discretion and use informal methods of social control when responding to domestic
violence incidents due to these beliefs (Stalans & Finn, 2006).
Arrest of Limited Value in Reducing Domestic Violence
Although there seems to be support for considering domestic violence a real crime
and responding to it through the use of the criminal justice system, there are studies that
show that police officers often believe that arrest is not the most appropriate response to
domestic violence incidents. One of Toon and Harts (2005) major findings illustrated that
many officers thought that arrest was of limited value when used as a response to domestic
violence. Toon and Hart (2005) are not alone in finding that these are common attitudes held
An early study conducted by Friday, Metzgar, and Walters (1991) found similar
opinions. Data were collected from law enforcement officers in a Midwestern community
about attitudes regarding their departments new pro-arrest policy for domestic violence
incidents. Overwhelming support was found for the policy itself, with 88% of the officers
approving of it. Friday et al. (1991) concluded that officers strongly supported the policy
because they perceived it as giving them more power when responding to domestic
situations than they had had before the law was enacted. This reflected officers feeling that
they were more capable when responding to domestic violence situations given the new
policy. Yet, even though support was found for the policy itself, the study also revealed that
officers did not view the policy as being effective in actually reducing domestic violence
incidents. Only 22% of the officers felt that the policy reduced the number of calls they were
responding to and 47% felt that it did not reduce the number of repeat visits to the same
households (Friday et al., 1991). One of the most important attitudinal findings was that
fewer than half (43%) of the officers felt that the new policy was impacting the number of
beatings that women experienced. In summary, the findings from this study provide an
example of a policy that in theory makes sense and is supported, but does not seem to be
having anticipated results.
Many officers believe that therapeutic treatment rather than punitive sanctions is
more beneficial for domestic violence offenders in terms of positive outcomes compared to
other types of offenders (Logan et al., 2006). Sinden and Stephens (1999) conducted
intensive interviews with police officers in New York about various domestic violence issues.
They found that even though 80% of the domestic incidents officers responded to involved
some sort of criminal conduct, the majority of the conduct responded to did not fall under
statutory requirements mandating arrest. In these cases, officers needed victims to press
charges or needed to secure witness testimony before exercising their formal powers. They
also found that most officers considered domestic violence incidents to be heterogeneous
events, and that officers made few generalizations about domestic violence. Yet, their study
revealed that the officers had doubts that arrest would result in many short and long term
deterrent effects. According to Blount, Yegidis, and Maheuxs (1992) study with police
officers in Florida approximately half of the officers surveyed were unsure about whether a
preferred arrest policy reduces subsequent violence.
Stalans and Finn (2006) obtained a mixed sample of seasoned police officers
(defined as those who had at least one year of service and responded to at least ten
domestic violence calls), novice police officers, and laypersons to examine attitudes toward
domestic violence using a survey that contained hypothetical incidents. They found that
laypersons and novice officers, compared to seasoned police officers, were more likely to use
informal methods of social control when responding to domestic violence cases involving no
visible injuries. Seasoned officers, however, were more likely to separate couples for the
night (Stalans & Finn, 2006). Seasoned officers also were more likely to arrest offenders if
there were visible injuries yet they were more likely to arrest both parties when there was
evidence of drinking among victims. Even though seasoned officers were more likely than
laypersons and novice officers to believe that arresting offenders was a fair option, the
sample as a whole rated overnight separation as being the fairest response to domestic
violence incidents, regardless of whether or not the victim sustained injuries.
Empirical literature has documented the numerous ways in which officers respond to
domestic violence incidents. The implementation of mandatory and pro-arrest laws provided
officers with a more powerful legal response to domestic violence incidents. Many officers,
however, have been unsure whether this response is most appropriate in all cases. Officers
often feel frustrated when encountering victims at the scene of domestic violence incidents.
Domestic violence is a complex crime and many times police officers fail to understand the
complicated dynamics of this unique form of violence. The following section addresses the
complexities of violent relationships. The next section explores attitudes victims hold toward
the police. This is important to discuss because many times victim attitudes fuel officer
attitudes, which in turn influences the way that officers respond at the scene of domestic
Domestic Violence Victims Attitudes Toward the Police
When a police officer arrives at the scene of a domestic violence incident he or she is
barraged with multiple scenarios and emotions, coming from victim and offender alike. As
previously noted, domestic violence is an incredibly complex crime involving unique dynamics
compared to other offenses and is one in which the victim may not always appear to act like a
typical victim of a violent crime. For these reasons the police have to be sensitive to these
types of emotions in order to fully understand the situation and respond appropriately.
Domestic violence is a particularly challenging phenomenon and is one that is particularly
difficult to understand since many times the parties involved state that they love each other
when, in reality, the dichotomy in the relationship is much more complicated because
violence is involved.
Domestic violence is caused by one partner in a relationship feeling entitled to have power
and control over his or her partner, and acting upon these feelings. Abusers attempt to
dominate their partners so that they gain all of the power in the relationship. Once this level of
power is achieved, abusers use violence in order to establish and maintain authority over
victims. In addition to physically abusing their partners, domestic violence offenders may
constantly criticize, belittle and humiliate their victims (National Center for Victims of Crime,
1999). Offenders often restrict victims social outlets, prohibiting the victim from maintaining
outside relationships with friends, family, and even co-workers. Domestic violence offenders
rarely are mentally ill, but instead are individuals who use abusive and manipulative
techniques to control their mates. Offenders also use their manipulative skills to
communicate with police officers and often attempt to make the victim appear hysterical and
delusional. During the law enforcement officers initial response to a domestic violence
incident victims may have a difficult time communicating with police officers because of the
psychological trauma endured due to the battering (NCVC, 1999). Again, it is for these
reasons that it is important for police officers to be knowledgeable about the complex
dynamics and nature of domestic violence.
Consequences of Domestic Violence
Domestic violence victims experience a variety of physical consequences. According
to the National Crime Victimization Survey, between 1993 and 2004 female victims of
domestic violence were injured 51% of the time during an incident (Catalano, 2006).
Although most of the injuries were classified as minor (43%), 5% of these victims sustained
serious injuries such as gunshot wounds, knife wounds, broken bones, being knocked
unconscious and various internal injuries. Only 19% of victims who were injured were
treated, and most of those were treated at home or at the scene of the incident (Catalano,
2006). This causes some concern because these statistics suggest that many injuries go
untreated, which may lead to further medical complications, and by not seeking immediate
medical treatment the incident has no chance of being initially reported by a health care
professional. According to a review of medial files from emergency room patients,
approximately 36% of female patients reported being injured by either a current or past
intimate partner (Rand, 1997). Coker, Smith, Bethea, King, and McKeown (2000) conducted
a study involving data collected from primary care doctors and found that half of the women
experienced domestic violence in their lifetime. Out of these women, 28% sustained injuries
that prevented them from working, 38% suffered from chronic neck or back pain, 37%
suffered from migraines, 32% experienced hypertension, 30% had a sexually transmitted
infection, and 33% suffered from gastric reflux as a result of domestic violence (Coker et al.,
2000). In addition to the few studies reviewed here, additional research has empirically
documented the short and long term physical consequences of domestic violence.
In addition to negative physical consequences of domestic violence, victims
commonly suffer from mental health consequences. Three psychological constructs most
commonly investigated in terms of their correlations to domestic violence include depression,
post-traumatic stress disorder (PTSD), and lowered self-esteem (Waldrop & Resick, 2004).
Depression occurs at high rates among victims of intimate partner violence compared to non-
domestic violence victims. Holzworth-Munroe, Smultzer, and Sandin (1997) found the
prevalence rate of depression among battered women to be between 50% and 83%.
Although prevalence estimates of PTSD vary among battered women, the trauma of
domestic violence can lead women to have one or all of the following symptoms: re-
experiencing the trauma, hyperarousal, and avoidance. One study found the rate of PTSD to
be 81% among physically abused women and 63% among women who had been emotionally
abused. A significant portion of the sample in this study were victims seeking treatment,
which accounts in part for the particularly high rate of PTSD and certainly should be
considered as a sample bias (Kemp, Green, Hovanitz, & Rawlings, 1995).
Although low self-esteem and depression are correlated, research has found that
self-esteem is related much more to abuse than to depressive symptomatology. Cascardi and
OLeary (1992) suggested that low self-esteem is a static construct and that depressive
symptoms tend to be dynamic. Another study comparing chronically battered women to
women who were no longer in abusive relationships found that women who were still involved
in a battering relationship had, on average, lower self-esteem than women who had
successfully ended their violent relationships. Low self-esteem can contribute to the cycle in
which women are less able to engage in active coping, therefore making victims more
vulnerable to subsequent abuse (Frisch & MacKenzie, 1991). Some have suggested that this
is the point in which victims feel helpless and isolated. Either by using psychological abuse
against the victim or by the victim experiencing negative psychological consequences from
physical violence, the offender maintains control over the victim, which makes it extremely
difficult for the victim to leave the relationship. Again, it is crucial that law enforcement
officers understand these complex dynamics of domestic violence, because a successful
intervention does not appear to be law enforcement officers simply telling victims to leave
In addition to contributing to the understanding of the dynamics of domestic violence,
studies on victim attitudes may help determine how to improve police officer responses to
domestic violence calls. Research on victim attitudes may also improve the law enforcement
image among victims, which was earlier discussed as a source of police officer frustration
especially when victims are not cooperating with an investigation. Apsler, Cummins, and Carl
(2003) conducted in-person and telephone interviews with domestic violence victims that
focused on attitudes regarding police officer assistance in their cases. Specifically, the
interviews examined whether victims wanted to obtain restraining orders, have the offenders
arrested, or needed counseling referrals. Findings indicated that the majority of the victims
felt that the police offered the appropriate level of assistance. On the other hand, when it
came to counseling, Apsler et al. (2003) found that 41% of victims who wanted help felt that
the police assisted them, while the remainder of the sample felt that the assistance they
needed was not provided in their case. This may indicate that law enforcement personnel
need to have more resources available to them to provide victim assistance when responding
to domestic violence calls, such as the accompaniment of a victim advocate. Apsler et al.
(2003) also examined helpfulness ratings of police assistance given by victims. Seventy-five
percent of victims considered the assistance to be very helpful" and 9% considered the
assistance not at all helpful. Eighty-one percent of the victims said that they would definitely
contact the police in the future for a similar incident and 6% said that they would definitely
not or probably not call the police again (Apsler et al., 2003, p. 1329). Women who felt
they would not call the police for assistance in the future attributed this to their fear of
retaliation or their fear that police presence would make the situation worse. When the
offender was arrested, despite victim objections, women reported to be less inclined to call
the police again for future incidents (Apsler et al., 2003). The point of mandatory and pro-
arrest policies was to be more punitive on domestic violence offenders. Unfortunately, some
victims, as a result of the new more punitive sanctions may now be less likely to call the
police in the future because of the situation they are faced with when their partner is arrested.
Victims are faced with having to choose between the lesser of two evils. This is a
complicated situation to understand and it is not surprising that without specialized domestic
violence training law enforcement officers would not understand the complicated nature of
this crime, and how domestic violence is very different from other crimes.
Russell and Light (2006) obtained a Canadian sample of victims and police officers to
examine how victim empowerment was connected to the efficacy of police interventions.
Victims for this study were recruited by victim services, and in most of the communities in
their sample (five out of six) the service providers contacted only those victims with whom
they had recent contact. The data from the officers and the victims were collected through
interviews. Findings indicated that police who believed that their function was separate from
other social service agencies, who believed that victims brought the abuse upon themselves,
and who believed that minimal responses were adequate in resolving domestic disputes were
regarded by victims as less empowering than police who viewed themselves as team
members, who did not blame victims, and who were seen to go the extra mile on behalf of
victims (Russell & Light, 2006, p. 391). Russell and Light (2006) suggest that there are
positive consequences from police officers working as a part of the community to resolve
these types of conflicts. Due to the complex nature of domestic violence incidents, it is
understandable that police officers may feel frustrated while interacting with victims of this
unique crime. Many times officers struggle to understand the actions and attitudes of victims,
and this often results in the officers forming negative and cynical attitudes toward victims.
These negative attitudes do not present the criminal justice response to this crime in a
Officers Struggle to Understand Victims' Actions and Attitudes
When police respond to a domestic violence incident, their main function is to resolve
the immediate conflict and help the victim. When victims behaviors are unsupportive of
further criminal justice system intervention, officers may be confused about the appropriate
response. Toon and Hart (2005) found that officers struggle to understand the attitudes and
actions of domestic violence victims. In order for victims to be treated fairly by the police
during the resolution of domestic violence incidents, it is important to understand the most
likely responses officers have to these cases. Johnson (2004) found that 38% of law
enforcement officers attributed victim behavior as being one of the major frustrations when
responding to domestic violence calls. Examples of victim behavior included not following
through with charges, refusing to sign a complaint, and recanting statements. However, of
the officers that cited victim behavior as a source of frustration, none placed blame on the
victim for being the cause of the domestic violence incident. At the same time, Johnson
(2004) found that some officers considered family issues, such as going to the same house
multiple times, to be their biggest frustration, which may reflect their lack of understanding of
that domestic violence is a cyclic crime, unlike other personal offenses they typically respond
In Sinden and Stephens (1999) study, officers reported having mixed feelings toward
domestic violence victims due to the way victims typically responded to their assistance. All
of the officers interviewed had experience working domestic violence cases that had an
absence of victim cooperation, either during the investigation or subsequent prosecution. All
of the officers interviewed reported experiencing at least one situation in which the victim
became verbally or physically abusive toward them at the scene (Sinden & Stephens, 1999).
Actions of the victim that are negatively aimed at the police officer can help explain why some
police officers become frustrated.
A part of police-victim interaction that is extremely important is in situations where the
victim is inappropriately arrested by the officer. If an officer develops negative attitudes about
the victim, there may be a chance that the officer arrests the victim on the scene due to
misconceptions about the case. Saunders (1995) study focused on reasons why police
officers arrest victims and characteristics of officers who did and did not arrest victims. The
sample consisted of police officers from agencies in Wisconsin. Officers were presented with
two vignettes, one in which a woman was beaten and the offender yelled at the officer, and
another in which a woman was beaten and continued arguing with the offender in the
presence of the officer. The officers surveyed were given ten response options representing
how likely it would be that they would respond in a certain way to these two vignettes.
Findings indicated that there was a 50% greater likelihood that the officers would arrest the
victim in the vignette where the victim and offender continue to argue. Also, of the officers
who said they would arrest the victim, almost all said it would be for disorderly conduct.
One officer said they would consider the continuing arguing to be battering (Saunders,
1995). This study also found that those officers who preferred to arrest the victim were more
likely to hold stereotypical beliefs about domestic violence. For example, one stereotypical
belief was believing that infidelity was a justification for this type of violence. These officers
were also less likely to believe that victims stay in violent relationships for practical reasons,
and officers who used dual arrest were more likely to believe that victims stay for
psychological reasons (Saunders, 1995).
With the combination of conflicting attitudes that collide between law enforcement
officers and victims in domestic violence cases it is clear that there are unanswered
questions in this line of research. In addition to understanding victim attitudes and actions,
police officers need to be prepared for the way in which offenders will act once they arrive on
the scene of domestic violence incidents.
Police Attitudes and Actions Regarding Offenders
When an officer arrives at the scene of a domestic violence call, he or she has to be
prepared to respond not only to the victim, but also to the offender and the way that these two
parties interact with one another. It is crucial to examine the way officers react to offender
actions and attitudes, because it is important that the actions of the perpetrator affect the
police as little as possible, given the manipulative nature of batterers. Buchbinder and
Eisikovits (2004) conducted a qualitative observation study of domestic violence offenders
and their attitudes and actions toward law enforcement officers at the scene of the incidents.
They observed that once the police arrive on the scene the offender would describe the
incident that led to the 911 call as absurd and unnecessary, making the victim seem irrational
for calling the police in the first place (Buchbinder & Eisikovits, 2004). Offenders also use
other techniques to normalize the situation before the police arrive, such as distancing
themselves from the incident, such as going outside and watering the lawn while they wait
for the police (Buchbinder & Eisikovits, 2004, p. 449). The researchers observations showed
that offenders attempt to seem rational, balanced, objective, and not dangerous, especially
during their first encounter with the police. An interaction with the police where batterers are
warned of the consequences of their actions, but end up not being punished, actually further
empowers the batter (Buchbinder & Eisikovits, 2004). It is crucial for police officers to be
prepared for the manipulative techniques that offenders notoriously use.
Other studies have found that the way in which the offender acts at the scene of the
incident can increase the odds of being arrested. Feder (1997, p. 90) found that police
officers were significantly more likely to arrest offenders who were drunk, in comparison to
the other scenarios they were presented with, which included: she disobeyed him, he was
laid off at his job, she nagged him incessantly, she was drunk, he just learned of her
sexual infidelity, and she insulted him in front of his friends. Yet, Feder (1997) found that in
all of the scenarios that were presented to officers, they were more likely not to arrest the
offender. This was a very surprising and concerning finding since this particular law
enforcement agency had a strong pro-arrest policy. Even with this policy in place, only 41%
of the officers said that they would arrest the offender in ail of the scenarios (Feder, 1997).
Logan et al. (2006) conducted a quantitative self-report study with a representative sample of
police officers. Their study focused mainly on officers attitudes toward offenders of certain
crimes and the sanctions for those crimes, and domestic violence was one of the crime types
examined. They did not find any differences in opinion based on officers gender or
education level. However, they did find differences based on time officers spent on the force.
Officers with three years of experience or less reported higher sanction ratings for domestic
violence offenders. Interestingly, Logan et al. (2005) suggest that as officers gain
experience, they are less likely to view arrest as an appropriate response to domestic
violence situations. Similar to Feder (1997), Logan et al. (2006) also found that officers rated
the use of sanctions higher for domestic violence offenders who use alcohol or drugs
compared to domestic violence offenders who did not use alcohol or drugs (with the
exception of substance abuse treatment). This indicates that whether the offender is under
the influence of alcohol is likely to have an impact on the chances of being arrested, which is
an extralegal factor that should not be taken into account when operating under a pro-arrest
Robinson (2000) conducted a study that examined change in officer attitudes after
the implementation of a mandatory arrest policy. The study compared attitudes of officers
who were employed prior to and after the implementation of a mandatory arrest policy and
the differences between their schemas when responding to domestic violence cases. Similar
to findings from Robinson and Chandek (2000), Robinson reported that regardless of the
officers schema category, when the suspect and the complainant were cohabiting the odds
of arrest were higher (Robinson, 2000). Older law enforcement officers may have
preconceived notions about men and women living together when they are not married.
These findings remind us that it is important to educate all law enforcement officers of all
ages in the overall dynamics of domestic violence including how offenders usually react to
police officers in order to prevent any type of manipulation that may occur at the scene. In
addition to feeling frustrated with victims and offenders, police officers experience frustration
from the other sources including the criminal justice system when intervening in domestic
violence cases. These systemic sources of officer frustration are discussed below.
Other Sources of Officer Frustration
Mandatory and pro-arrest laws provided officers with a major legal response to use
when responding to domestic violence cases. Yet, along with this additional legal response,
police officers reported experiencing significant frustration associated with a loss of one of
their most powerful weapons their discretion. Out of all of the actors in the criminal justice
system, police officers have the greatest opportunity to exercise discretionary judgment
(Wortley, 2003). By curbing their ability to use this discretion in cases that are already
complex, such as domestic violence cases, it was intended for offenders and victims to be
treated more equally. Unfortunately, this reasoning is not satisfactory to many officers who
respond to domestic violence cases. Toon and Hart (2005) reported that many officers felt
they needed more discretion to appropriately respond to domestic violence incidents. Since
officers do view these cases as heterogeneous, they feel they should have more types of
responses available to them. In studying police attitudes toward general discretion, Wortley
(2003) found that police officers who are more service oriented, as opposed to focusing only
on the law, view the use of discretion as an appropriate response to social problems. This is
a concerning finding, since the community policing philosophy has been implemented in
many departments across the U.S., yet at the same time these officers who are supposed to
be more service-oriented have to implement very strict laws, including mandatory arrest. It
seems many officers may feel internally conflicted when trying to figure out the best response
to a domestic violence situation. This becomes a source of frustration, especially when
combined with conflicting victim attitudes about arrest at the crime scene where emotions of
all parties involved tend to be rather high.
In addition to feeling frustration due to victim attitudes, discretionary policies, and
internal conflicts, police officers may often become frustrated with the way in which that the
rest of the criminal justice processes domestic violence cases. After interacting with the
victim and offender and deciding whom to arrest, there are times when these cases do not
proceed to the prosecution stage, which is an additional contributor to officer frustration.
Toon and Hart (2005) reported that officers are irritated due to the inaction of prosecutors. It
is understandable that some officers may feel this way, considering their effort in responding
to domestic violence cases. Toon and Hart (2005) reported that officers feel too few cases
are being prosecuted. This is problematic as it may dissuade officers from fully promoting the
criminal justice response to victims of domestic violence. If officers feel that the case is not
likely to be prosecuted, they may view their efforts in arresting the offender as a waste of
time. Johnsons (2004) study also found that 19% of the police officers surveyed were
frustrated with the court system and the specific ways in which cases processed after arrest
occurs. In Blount et al.'s (1992) research, even though officers were discouraged by their
supervisors from considering the possibility of prosecutorial inaction as a major factor in their
decision to arrest, 60% of the sample indicated that this was in fact a major factor in their
decision to make an arrest. It is problematic to simply instruct officers to disregard certain
possible consequences when police officers are the ones who are putting their efforts into the
case. Blount et al. (1992) also found that a vast majority of officers and supervisors agreed
that victim cooperation with the prosecutor is a major indicator of whether or not the
prosecutor will proceed with the case.
These deep rooted pessimistic attitudes about the likelihood of prosecution may also
impact victims in a negative manner. These negative expectations about prosecution may
cause the officer to be less likely to arrest the offender or act proactively at the scene. This
may make the officer act more negatively toward a victim, which in turn may promote
negative victim attitudes toward the officer. It is important to focus on the sources of officer
frustrations and explore ways in which they may be reduced. One approach to helping
officers process frustrations they experience when responding to domestic violence cases is
to continue to assess their attitudes and then to provide them with pertinent training as
identified by the research. This training would provide law enforcement officers with more
general knowledge about the dynamics of domestic violence and would be based on the idea
that further education would appropriately impact their response. The following section
discusses the need for and nature of domestic violence law enforcement training.
The Need for More Training
One of Toon and Harts (2005) major findings focused on the need for more police
training on how to respond to domestic violence calls. This finding is directly linked to
important policy implications. Domestic violence advocates and victims alike feel that officers
should be better trained about domestic violence issues. Therefore, if continued research
finds consensus among officers regarding the benefits of training that would help them better
understand cases, training policies have a better chance of being implemented. Gamer
(2005) found that police attitudes are capable of change, and training is one way to
accomplish this goal.
Johnson (2004) found that one of the biggest frustrations officers have with domestic
violence cases is figuring out how to respond in full accordance with the law, and figuring out
who the primary aggressor is. Logan et al. (2006) suggest that police officer training needs to
include more information about the fact that all forms of battering are violent crimes. Sinden
and Stephens (1999, p. 320) found that due to the confusing wording of the law the officers
reported a concern for their ability to determine when acts of physical violence (such as
striking, shoving, and kicking) produce sufficient physical harm to permit them to make an
arrest independently of other considerations, including the victims wishes. It is critical for
police to undergo training not only in how to respond to domestic violence cases, but also
how to properly comprehend the laws that these officers are charged with enforcing. It is
important that law enforcement officers understand the laws that they implement so that such
laws can be correctly enforced.
Saunders (1995) suggests that officers need more domestic violence training. He
suggests training that includes exercises such as role playing and hearing from a panel of
survivors. Garner (2005) points out that training which involves direct exercises is more
effective than indirect learning. Russell and Light (2006) note that police officers who learn to
respond proactively are more likely to have positive attitudes toward victims, and that this is
something that is likely to be influenced by training. In Blount et al.s (1992) study, 80% of
the sample reported attending domestic violence training, although this was significantly
different for supervisors and officers, with a smaller proportion of supervisors attending the
trainings. In addition to the importance of training itself, it is essential for domestic violence
training to be mandatory among all ranks, not just for incoming officers. Besides focusing on
knowledgeable training, officers need to have positive role models in their department,
Garner (2005) points out that one of the best ways to influence a persons behavior is
to alter your own. Novice police officers often learn how to respond to certain kinds of calls
on-the-job, and they model their behavior after the actions of their superiors. If incoming
police officers are trained to respond to domestic violence cases differently than seasoned
police officers, this may cause additional frustration. It is critical for all police officers to be
trained regularly on the changing dynamics of domestic violence and the responses to this
crime. In addition to focusing on victims, the criminal justice response, and officer frustration,
studies have explored officer attitudes as they relate to lethality, demographic characteristics
of the parties involved, as well as the officers themselves. These studies are discussed in
the sections below.
Other Findings of Attitudinal Studies
It is important to take into consideration findings from the studies discussed below, as
the survey that is being used for the current research covers similar topics. Gauging officer
attitudes on lethality and demographic differences is vital, as all of these factors are present
at the scene when an officer responds to a domestic violence call. The following section will
address police attitudes and ideas about the lethality of responding to domestic violence
scenes as well as officer attitudes about demographic characteristics, such as gender and
race, and the way that the officers own characteristics affect the attitudes they hold. It is
essential to look at these remaining factors that may influence a police officer's decision
making process, and the current study will attempt to do so.
Perceptions about Lethality
One type of situation in which the behavior of the offender can prove to be extremely
important is when the offender has a weapon. It is crucial for officers to know what type of
situation they are responding to and since domestic violence cases are so complex, it is
important to discuss the attitudes that officers have regarding the lethality of these types of
In addition to being manipulative, domestic violence offenders can often be very
dangerous. According to a nationwide study which examined every domestic violence
related officer murder, the first minute of an officer's presence on the scene of a domestic
violence call was found to be the most dangerous (Wynn, 2007). This study found that most
officers died while approaching the scene. In these cases 81% of the offenders were armed
and knew the officer was coming and 57% of the offenders were essentially suicidal (Wynn,
2007). Friday et al.'s (1991) study indicated that 78% of the officers they interviewed said
that they had been physically assaulted in the course of conducting an arrest in a domestic
violence call. Blount et als (1992) study also found that female and male officers reported a
concern for personal safety when responding to a domestic violence call, including
supervisors and line officers. In Sinden and Stephens (1999) study most of the officers they
interviewed perceived themselves as potentially at risk of harm when responding to domestic
incidents. However, none of the officers expressed a hesitance to respond to a domestic call
because of fear. Even though the officers report that they are not hesitant to respond to
these calls, it is not surprising that, if they believe that domestic violence cases are one of the
most dangerous incidents they can respond to, this may affect the way in which they react to
Having greater knowledge and statistics on lethality and overall dangerousness
regarding domestic situations, through training, may help officers respond in a more positive
manner to domestic violence incidents. Since they are not, in fact, the most dangerous type
of case to respond to, as some officers assume, it is clearly important to provide the correct
knowledge to these officers through such methods as training (Wynn, 2007). Once the
officers have a better idea of the reality of these situations, they will be able to respond in
more appropriate ways.
Influences of Demographic Characteristics on Police Attitudes
Everyone has preconceived notions regarding people and situations based on
demographic characteristics. A persons age, gender, race, sexual orientation, outward
appearance, and other factors may influence the way in which people interact with one
another. Police officers are trained to respond without prejudice, yet it is understandable
how, over the time of their service, they may develop specific ways of responding to certain
types of cases and individuals (including domestic violence victims and offenders) based on
their demographic characteristics. In addition, it is important to explore how the officers own
characteristics may affect the way that he or she responds to a case.
Research has consistently documented that women are more likely to be victims of
domestic violence. In addition, it is also known that females do not make up equal portions of
police officers in most departments throughout the country. Historically, women have been
excluded from the police profession and the traditional police culture overall (Paoline et al.,
2000). Consequently, one might expect that women would be less likely to internalize the
masculine, law enforcer type personality that male police officers often hold. Paoline et al.
(2000) also point out that female police officers might be more likely than male officers to hold
positive views of citizens, as they are viewed as being less confrontational and encounter
less resistance from citizens. Yet, after proposing these views, Paoline et al. (2000) did not
find significant differences in opinions and adaptations to the police culture between males
and females. They propose that perhaps because women that are attracted a career in law
enforcement are unrepresentative of women generally and that the beliefs that may be held
by women overall may not apply to police officers. On the other hand, some studies do find
differences in male and female officer attitudes and actions toward domestic disputes.
Robinson and Chandek (2000) found a gender difference in arrest decision-making, reporting
that female officers were less likely to make an arrest. These findings are extremely
important consider, since an officers gender should not influence how strictly they adhere to
In addition to gender, another clearly visible demographic factor is race. Several
studies have pointed to differences between the incidence of domestic violence among
different races. Rennison and Welchans (2000) study depicted Black women as being the
ones who suffer from the highest rates of domestic violence. Black females were victims of
domestic violence at a rate 35% higher than that of White females, and about 22 times the
rate of women of other races (Rennison & Welchans, 2000). Caetano, Field, Ramisetty-
Mikler, and McGrath (2005) examined the prevalence of domestic violence among couples in
the U.S. using data from 1995 and comparison data from 2000. They found domestic
violence prevalence estimates to be higher for Blacks and Hispanics than for Whites. They
also found that, while for Whites the prevalence rates decreased from 1995 to 2000, they
remained relatively stable for Blacks and Hispanics. They also report that Blacks were more
likely to report domestic violence in those two years than the other races (Caetano et al.,
With the differences being reported in prevalence rates, it is important that the race of
the parties involved and of the police officer does not affect the way that the domestic
violence situation is treated. Paoline et al. (2000) note that minority officers have not been
accepted into the police culture as readily as Whites, and therefore may be less likely to
embrace cultural valuesa positive thought considering that traditional police values diminish
the importance of domestic violence situations. Minority officers may also have more insight
and empathy regarding the perspective of minority communities, which is again positive due
to the prevalence statistics reported above. Paoline et al. (2000) suggest that minority
officers may also be less susceptible to stereotypes present inside the police culture and less
estranged from the minority communities, and thus may be more likely than their White
counterparts to see citizens as willing to cooperate.
Even though there is a possibility for these demographic factors to affect police
officers response tactics, other studies have shown more similarities than differences
(Robinson, 2000). Saunders (1995) found that officers' demographics, rank, years of service,
type of work before policing, or orientation to police work did not make a difference in their
likelihood to arrest. The current research is also exploring the way that police officers react to
the demographics of the parties involved in domestic violence and how their attitudes differ
according to the police officers own demographic characteristics.
Comparison of Attitudes between City and County Officers
In addition to exploring attitudinal differences of officers based on their demographic
characteristics, the current study investigates differences of opinion based on the type of
agency. Toon and Hart (2005) conducted a comparison between major urban police
agencies and other police agencies, which were mostly rural (p. 10). They found
differences in the volume of calls that officers responded to, with the officers in urban
agencies responding to a much higher volume. The officers in these urban areas showed
much higher degrees of frustration in responding to the calls. The urban officers were more
likely to agree that the best response to domestic violence cases in an informal one, and that
there should be a limit on how many times they respond to the same address (Toon & Hart,
2005). The urban officers also were more likely to agree that victims are often as responsible
for a domestic violence incident as the offender (Toon & Hart, 2005). Rural officers, in
contrast, held less traditional attitudes and believed that training would be helpful, that they
need more discretion, and that an arrest policy is the best legal approach to domestic
violence (Toon & Hart, 2005). One area that they failed to address, due to the content of
their data, was a comparison of attitudes between county and city police officers. It seems
that there is no research that has looked at police attitudes on domestic violence while
comparing an urban city police department to a suburban county department. Therefore, the
current research will provide novel insight into this type of attitudinal comparison. The current
research will conduct a comparison of police officer attitudes between a city and county
department, using well-sized samples from each locale.
The research design for this study involved a cross-sectional, convenience sample of
law enforcement officers from two agencies in a Western state. The data collected represent
self-reported information from police officers. The survey instrument used for this project was
constructed by Drs. Mary Dodge and Angela Gover, and was modeled after Toon and Harts
(2005) law enforcement survey (See Appendix A). The instrument consists of 28 questions
regarding attitudes on topics such as lethality, seriousness of domestic violence calls,
discretion, arrest as a solution, the need for training, suspect/victim behavior, officer
frustrations, and knowledge on domestic violence among minorities and same sex couples.
The survey was constructed specifically for police officers in order to assess their attitudes
about domestic violence. In addition to these questions, demographic and opened-ended
questions were asked. The survey took officers 15-20 minutes to complete. For the
purposes of confidentiality, the identities of the law enforcement agencies who participated in
this project will not be disclosed.
Description of Agencies and Communities
The county agency serves approximately 185,000 residents in the unincorporated
areas of the county. Both agencies have specialized in-house units, including a victim
services unit which provides victims with assigned advocates that stay in touch with them
after initial contact has been made with the police. The county agency employs
approximately 500 officers.1 The city agency employs about 1,500 officers, with the majority
being white (68%) and male (89%).
1 The demographic breakdown of for the county agency was not available.
The communities that the two agencies serve differ in various ways. The area which
the county agency serves has a lower percentage of minorities, approximately 1% Black
residents and 13% Hispanic residents, compared to that of the city agency which has
approximately 11% Black residents and 32% Hispanic residents (U.S. Census, 2000). The
median and per capita income of the area which the county agency serves are both higher
than the area the city agency serves, as are the house values and house ownership rates.
Also according to the 2000 Census, the county agency has a lower number of people per
square mile with approximately 683 persons, compared to 3,617 persons per square mile
compared to the geographic region served by the city police department.
Prior to the administration of the survey, approval was obtained from the Institutional
Review Board that the University of Colorado Denver. The overall project collected data from
six agencies in a Western state, but the current research focuses on the largest agencies
from which data were collected, one city and one county agency. The survey was
administered to police officers in two police departments, one county department and one city
department in a large metropolitan area. In the county police department, the surveys were
sent to a contact person who distributed them to officers during shift meetings, otherwise
known as roll-call, over a time period of approximately a week. Respondents were instructed
to complete and return the surveys to the departments contact person. The officers were
told by the contact person that the surveys were mandatory, which was a break in protocol.
In the city police department surveys were distributed to officers in the six precincts during
roll-call. The person in charge of roll-call distributed the survey to the police officers. The
police officers were able to turn in the surveys to a specified collection location at the agency
on their own time and the officers were not monitored during their completion of the survey.
An Informed Consent form was provided to participants that explained the purpose,
and the voluntary and confidential nature of the study. Yet, In the county agency the contact
person informed the officers that the survey was mandatory. In addition, the subjects were
told that they may not directly benefit from their participation, but that their participation would
give them the opportunity to provide input about how to improve the law enforcement
response to domestic violence. Subjects were also told that the minimal risks that they may
experience from their participation included feelings of discomfort and embarrassment about
domestic violence issues (See Appendix B for consent form). Respondents received no
compensation for their participation in the study.
Due to the nature of the survey questions, respondents were assured confidentiality
in several ways. Participants were instructed not to provide researchers with any personal
identifying information (such as name, badge number, etc.) and were given the opportunity to
complete the survey at a later time for privacy purposes. Both agencies identified a central
survey-drop location for officers who chose to complete their survey at a later time.
After data collection was completed, two graduate research assistants completed
survey coding and data entry. The data used for this study is a subset of data collected for a
project that involved a total of six law enforcement agencies in a Western state. The
estimated response rate for the city agency was 40%. The response rate for the county
department is estimated to be 100% by the departments contact person, as the officers were
told the survey was mandatory.
Building on the literature discussed in the previous chapter, the main research
question that is explored in this study is whether officer attitudes regarding the many facets of
domestic violence case responses differ according to the type of agency (i.e., city versus
county). Although there have been no previous studies comparing officer attitudes about
domestic violence in these two types of agencies, one may expect to see a difference
considering the make up of the departments and the differences in the communities that
these agencies serve. If differences are found according to agency type, the study will
attempt to determine what the sources of these differences are. The study also explores
attitudes about domestic violence based on demographic characteristics, including age,
gender, race, and years of experience on the force. Again, knowledge of attitudes will help
inform appropriate training protocols about domestic violence.
Demographic information was collected by asking the respondents their rank, gender,
age, and race/ethnicity. Rank was an open-ended question and was coded according to the
responses provided by the respondents. This variable (Rank) was then recoded to form a
dummy variable comparing line officers (0) with supervisors (1). Gender was measured as
either Male or Female (Male was coded as 1 and Female as 2). This variable was
recoded to a dummy variable with Males being coded as 1 and Females as 0. Age was
treated as an open-ended question with continuous answers. Race/ethnicity was measured
using the categories of White (1), Asian (2), African-American (3), Native American (4),
Hispanic (5), and Other (6). This variable was recoded into a dummy variable with
categories of White (consisting of the White category and coded as 1) and Non-White
(consisting of the remaining categories and coded as 0). The officers were also asked if their
agency had a domestic violence unit (1 = Yes, 2 = No), which was recoded to a dummy
variable with 0 and 1 responses. The number of years of experience at the current agency
and any previous agencies were both open-ended questions with continuous answers. The
officers were asked whether they have responded to domestic violence calls (Yes or No),
and whether their job involved any special domestic violence duties (Yes or No). These
two variables were recoded into dummy values. In addition, the samples from the two
agencies were coded differently using dummy codes in order to conduct a comparison of city
versus county agencies, with 0 representing the county agency and 1 representing the city
The 28 attitudinal questions about domestic violence included Likert-type response
options. Sample questions include: I am more likely to be injured during a DV call than any
other type of call, Too many DV calls are for verbal family arguments, I need more
freedom in deciding how to handle situations at DV calls, Arresting someone at a DV call
seldom helps reduce future DV incidents, Many DV victims could easily leave their
relationships, but dont. Respondents had the following answer choices: Strongly Agree,
Agree, Somewhat Agree, Somewhat Disagree, Disagree, and Strongly Disagree.
These responses were coded from 1 to 6, with lower scores indicating more agreement with
the attitudinal statements. In order to compare the univariate findings with Toon and Harts
(2005) findings the response options were recoded to form two categories, Agree and
Disagree, by combining the original six categories. The recoded responses were only used
in univariate analysis. Bivariate and multivariate analyses were conducted using the original
responses ranging from 1 to 6.
The final section of the survey asked officers about what type of special domestic
violence duties they performed (if any), and what they thought would help them respond to
domestic violence calls. In addition, officers were given the opportunity to add any comments
or thoughts they had about their experience with domestic violence calls. These qualitative
responses were used to provide examples of general officer attitudes.
For the purposes of the current research cases with missing data were deleted.
Univariate statistics, such descriptive statistics, were conducted on each of the 28 attitudinal
questions in order to look at each of the questions separately, using the recoded responses
(Agree and Disagree). In addition, the 28 attitudinal questions were analyzed according to
a comparison against the demographic variables in order to determine whether there was a
difference of opinion among officers based on gender, age, race, rank, and years of
experience. The analyses between attitudinal questions and dummy variables (rank, gender,
and race) were conducted using t-tests. Analyses between attitudinal questions and
continuous variables (age and years of experience) were examined using correlations. To
analyze the main research question, the survey responses were compared using the two
agency codes in order to determine whether there are attitudinal differences between officers
in city and county agencies. This issue was also examined with t-tests. Further analysis was
conducted if differences were found between the two agencies to determine the factors that
influence these differences. This analysis examined whether the attitudes that have
significant differences between the agencies would still be present once demographic make-
up of the two departments were controlled for (i.e., gender, race, age, rank and years of
experience). This final analysis was conducted using ordinary least squares multiple
As shown in Table 4.1, the overall sample consisted of 292 police officers. Fourteen
percent of the overall sample were employed by the county department (N=42) with the
remainder (N=250) employed by the city department. The mean age of the officers in the
study sample was 35.5 years with an average of 7.7 years of experience at the agency in
which they are currently employed. The majority of the sample consisted of officers who
were male (91%) and white (72%). Line officers made up 83.9% of the sample while
supervisors account for
Table 4.1 Descriptive statistics (N=292) _________________________________________________
Variable Total Sample (N=292) County Dept (N=42) City Dept (N=250)
Age (Mean Years) 35.5 (mean) 36.9 (mean) 35.3 (mean)
Current Agency Experience ( Mean Years) * 7.7 (mean) 10.6 (mean) 7.2 (mean)
Male 91.4% 83.3% 92.8%
Female 8.6% 16.7% 7.2%
White 71.6% 90.5% 68.4%
Non-White 28.4% 9.5% 31.6%
Supervisor 16.1% 4.8% 18.0%
Table 4.1 (Cont.)
Line Officer 83.9% 95.2% 82.0%
* p < .05
only 16%. The table also shows the demographic breakdown of the samples for the
individual agencies and notes whether the agencies are significantly different in terms of
these variables. The agencies differed in all of the measures except mean age. Officers
employed by the county agency had a significantly higher number of years of experience at
that agency (x=10.6) compared to the average number of years officers from the city
department had been working at that particular agency (x=7.2). The agencies also
significantly differed in their employee rates in terms of gender. According to those who
responded to the survey, only 17% from the county department were women, compared to
9% from the city agency. When examining the racial breakdown of the overall sample 72% of
officers who responded to the survey were white. When looking at the racial breakdown of
officers who responded to the survey by agency, 91% of the respondents from the county
agency were White compared to 69% from the city police department. Significant differences
were also found in officers who responded to the survey in terms of rank. The majority of
respondents from the county agency (95%) were line officers and the majority of respondents
from the city agency (82%) were line officers. Five percent of respondents from the county
department reported being supervisors and 16% of respondents from the city agency
reported being supervisors. All of these significant differences were accounted for in the final
Table 4.2 illustrates the overall sample responses to each of the 28 attitudinal
questions. When combining the data from the two agencies officers have mixed attitudes
regarding their consideration of domestic violence as a real crime. Most of the officers, on
the other hand, feel that domestic violence calls take too much of their time and effort and
they also show a high level of frustration with repeat calls to the same address. On the other
hand, most of the officers disagree with domestic violence cases being handled as a private
matter and a high percentage (66%) of the officers agreed with the statement Domestic
violence offenders must be arrested even when the victims dont feel it is necessary. This
provides evidence that many officers engage in practices that are statutorily and procedurally
proscribed and do not allow extralegal factors, such as the victims wishes, to influence their
The majority of the officers (68%) in the overall sample disagreed with using arrest
alone as a tool for responding to domestic violence. In general officers were undecided
about whether arresting an offender actually helps reduce the number of future incidents, with
40% of the officers agreeing to the statement and 60% disagreeing. An overwhelming
majority (92%) of the officers felt that too many of the domestic violence calls they respond to
are for verbal family arguments and were evenly split regarding whether an arrest should be
made only when there is clear evidence of injury. Although arrest was not a popular
response option for the officers in the sample, the results show that only 25% agreed with the
statement Its often best to arrest both parties in domestic violence calls, which is a positive
finding. As Table 4.2 shows, officers tend to have different attitudes about whether
prosecutors follow up in domestic violence cases.
The majority of the officers in the sample (87%) feel that they need more discretion in
responding to domestic violence cases. They also feel negatively about more training, as
only 28% of the officers agreed with the statement More training would help me assess
domestic violence scenes. The results regarding arrest decisions at the scene of the
domestic violence incident are also of concern, as the officers were almost evenly split
between disagreeing and agreeing with having difficulty identifying the primary aggressor and
identifying who to arrest overall. According to officers responses to the survey questions,
most are not influenced on the arrest decision making process by the behavior of the offender
as well as the presence of children. Officers were divided on the influence of victims
cooperative behavior. Only 8% said that they would be affected by the cooperativeness of
the suspect, 46% by the victims behavior, and 38% by the presence of children at the scene.
Probable cause identification seemed to be a problem as well, as 32% of the officers
considered it a difficult decision.
Table 4,2 Univariate analysis (N=292)___________________________________
Variable Agree Disagree
1. I am more likely to be injured during a DV call than any other type of call. 77.4% 22.6%
2. Too many DV calls are for verbal family arguments. 91.8% 8.2%
3.1 need more freedom in deciding how to handle situations at DV calls. 87.3% 12.7%
4. Its often best to arrest both parties in DV calls. 25.3 74.7%
5. Arresting someone at a DV call seldom helps reduce future DV incidents. 59.6% 40.4%
6. Police should arrest DV cases only when there is clear evidence of injury. 50.7% 49.3%
7. DV calls take too much of officers time and effort. 79.5% 20.5%
8. Identifying the primary aggressor at a DV call is difficult. 50.3% 49.7%
9. DV is best handled as a private matter, rather than by police. 11.3% 88.7%
10. Prosecutors usually follow up effectively in DV arrests. 49.0% 51.0%
11. Its often hard to know who to arrest in DV incidents. 40.1% 59.9%
12. lam less likely to make a DV arrest if the suspect is cooperative at the scene. 7.9% 92.1%
13. A mandatory arrest policy is the best approach to DV calls. 32.5% 67.5%
Table 42 (Cont.)
14. Many DV victims could easily leave their relationships, but dont. 72.9% 27.1%
15. A major problem with DV is repeat calls to the same address. 92.5% 7.5%
16. Substance and/or alcohol abuse is the main cause of DV. 79.1% 20.9%
17.1 am more likely to make DV arrests when children are present. 37.7% 62.3%
18. DV offenders must be arrested even when the victims dont feel it is necessary. 66.1% 33.9%
19. Its often hard to decide whether theres probable cause for arrest in DV cases. 31.8% 68.2%
20. Most DV incidents stem from abusers need for power and control over victims. 75.3% 24.7%
21.1 am more likely to make an arrest if the victim is cooperative at the scene. 46.2% 53.8%
22. Women are just as likely as men to engage in DV. 84.2% 15.8%
23. DV is higher among minorities compared to Whites. 28.4% 71.6%
24. Men abused by their partners are less likely than abused women to report DV. 78.4% 21.6%
25. DV laws should provide protections to same sex partners. 81.2% 18.8%
26. Minority victims are more likely to call the police than White victims. 23.6% 76.4%
27. More training would help me assess DV scenes. 27.7% 72.3%
28. Gays & lesbians DV occurs for the same reasons it does with heterosexuals. 81.5% 18.5%
The findings also show that in general, officers who responded to the survey have
mixed attitudes toward the misconceptions about the dynamics of domestic violence.
Twenty-seven percent of the sample disagreed with the statement that Many domestic
violence victims could easily leave their relationships, but dont. The majority of the officers
(75%) agreed with the notion that domestic violence stems from abusers need for power and
control over victims. Officers did have some progressive attitudes concerning domestic
violence laws, as 81% agreed that domestic violence laws should provide protections to
same sex partners and 82% agreed that domestic violence in gay and lesbian relationships
occurs for the same reasons as violence in heterosexual relationships.
More than half of the respondents (78%) thought that men abused by their partners
were less likely to report the incident to law enforcement. In general, officers did not perceive
minority victims to be more likely to call the police for domestic violence compared to White
victims. Only 16% of the respondents disagreed that women are just as likely to engage in
domestic violence as men while 84%of the respondents thought that women are just as likely
as men to engage in domestic violence.
Several bivariate analyses were conducted using the overall sample in order to
determine whether there were significant differences in officer opinion based on gender, race,
rank, age, years of experience, and agency type. The analyses were conducted using t-test
for the dichotomous variables and correlations for continuous variables. The results are
Table 4.3 illustrates the impact of gender on each of the attitudinal questions by
comparing mean responses from male and female respondents. As noted in the table,
significant differences in male and female responses were found in only three of the 28
questions. Response options were coded as: 1=Strongly Agree; 2=Agree; 3=Somewhat
Agree; 4=Somewhat Disagree; 5=Disagree; and 6=Strongly Disagree. Male respondents
were significantly more likely than female respondents to believe that they needed more
freedom in deciding how to handle situations at domestic violence calls. Females were
significantly more likely than males to disagree with the statement: Police should arrest DV
cases only when there is clear evidence of injury." The last significant difference between
male and female respondents was the statement DV offenders must be arrested even when
the victims dont feel it is necessary. Female respondents were significantly more likely than
male respondents to agree with this statement, which indicates that male officers are more
likely than female officers to allow the wishes of the victim to impact their actions.
Table 4.3 T-test for attitudes and gender (N=292)
Variable Means Means Difference
1. I am more likely to be injured during a DV call than any other type of call. 2.76 2.87 -0.11
2. Too many DV calls are for verbal family arguments. 2.52 2.25 0.27
3.1 need more freedom in deciding how to handle situations at DV calls. 2.92 2.18 0.74*
4. Its often best to arrest both parties in DV calls. 4.04 4.22 -0.18
5. Arresting someone at a DV call seldom helps reduce future DV incidents. 3.36 3.18 0.18
6. Police should arrest DV cases only when there is clear evidence of injury. 4.20 3.36 0.84*
7. DV calls take too much of officers time and effort. 2.88 2.47 0.41
8. Identifying the primary aggressor at a DV call is difficult. 3.48 3.68 -0.20
9. DV is best handled as a private matter, rather than by police. 4.72 4.70 0.02
10. Prosecutors usually follow up effectively in DV arrests. 3.80 3.62 0.17
Table 4.3 (Cont.)
11. Its often hard to know who to arrest in DV incidents. 3.60 3.88 -0.28
12. I am less likely to make a DV arrest if the suspect is cooperative at the scene. 4.96 4.76 0.20
13. A mandatory arrest policy is the best approach to DV calls. 3.68 4.05 -0.37
14. Many DV victims could easily leave their relationships, but dont. 2.76 2.87 -0.11
15. A major problem with DV is repeat calls to the same address. 2.04 2.14 -0.10
16. Substance and/or alcohol abuse is the main cause of DV. 2.56 2.75 -0.19
17. I am more likely to make DV arrests when children are present. 3.52 3.79 -0.27
18. DV offenders must be arrested even when the victims dont feel it is necessary. 2.52 3.07 -0.55*
19. Its often hard to decide whether theres probable cause for arrest in DV cases. 4.16 4.04 0.12
20. Most DV incidents stem from abusers need for power and control over victims. 2.84 2.90 -0.06
21. I am more likely to make an arrest if the victim is cooperative at the scene. 3.56 3.60 -0.04
22. Women are just as likely as men to engage in DV. 2.24 2.43 -0.19
23. DV is higher among minorities compared to Whites. 3.92 4.06 -0.14
24. Men abused by their partners are less likely than abused women to report DV. 2.12 2.59 -0.47
25. DV laws should provide protections to same sex partners. 2.20 2.64 -0.44
26. Minority victims are more likely to call the police than White victims. 4.00 4.16 -0.16
Table 4.3 (Cont.)
27. More training would help me assess DV scenes. 4.32 4.34 -0.02
28. Gays & lesbians DV occurs for the same reasons it does with heterosexuals. 2.24 2.66 -0.42
* p < .05
Race and Rank
In addition to gender, the analysis also included an analysis of respondents answers
according to the race of the respondent. White respondents were more likely than non-White
respondents to disagree with the statement I am more likely to make DV arrests when
children are present (See Table 4.4).
Table 4.4 T-test for attitudes and race (N=292)
Variable Means Means Difference
1. I am more likely to be injured during a DV call than any other type of call. 2.73 2.91 -0.17
2 Too many DV calls are for verbal family arguments. 2.08 2.34 -0.26
3. I need more freedom in deciding how to handle situations at DV calls. 2.13 2.29 -0.15
4. Its often best to arrest both parties in DV calls. 4.06 4.26 -0.20
5. Arresting someone at a DV call seldom helps reduce future DV incidents. 3.33 3.14 0.18
6. Police should arrest DV cases only when there is clear evidence of injury. 3.54 3.39 0.15
7. DV calls take too much of officers time and effort. 2.40 2.55 -0.15
Table 4.4 (Cont.)
8. Identifying the primary aggressor at a DV call is difficult. 3.83 3.60 0.23
9. DV is best handled as a private matter, rather than by police. 4.70 4.71 -0.01
10. Prosecutors usually follow up effectively in DV arrests. 3.59 3.66 -0.07
11. Its often hard to know who to arrest in DV incidents. 3.88 3.84 0.04
12.1 am less likely to make a DV arrest if the suspect is cooperative at the scene. 4.80 4.77 0.03
13. A mandatory arrest policy is the best approach to DV calls. 3.78 4.11 -0.33
14. Many DV victims could easily leave their relationships, but dont. 2.88 2.85 0.03
15. A major problem with DV is repeat calls to the same address. 2.11 2.14 -0.04
16. Substance and/or alcohol abuse is the main cause of DV. 2.84 2.69 0.15
17.1 am more likely to make DV arrests when children are present. 3.52 3.87 -0.35*
18. DV offenders must be arrested even when the victims dont feel it is necessary. 2.95 3.06 -0.11
19. Its often hard to decide whether theres probable cause for arrest in DV cases. 4.16 4.00 0.17
20. Most DV incidents stem from abusers need for power and control over victims. 2.86 2.91 -0.05
21. I am more likely to make an arrest if the victim is cooperative at the scene. 3.54 3.62 -0.08
22. Women are just as likely as men to engage in DV. 2.49 2.38 0.11
23. DV is higher among minorities compared to Whites. 4.02 4.05 -0.03
Table 4.4 (Cont.)
24. Men abused by their partners are less likely than abused women to report DV. 2.49 2.57 -0.08
25. DV laws should provide protections to same sex partners. 2.80 2.53 0.26
26. Minority victims are more likely to call the police than White victims. 4.06 4.19 -0.13
27. More training would help me assess DV scenes. 4.16 4.41 -0.25
28. Gays & lesbians DV occurs for the same reasons it does with heterosexuals. 2.70 2.60 0.11
* p < .05
Table 4.5 illustrates analysis conducted on each of the attitudinal questions based on
respondents being line officers versus supervisors. No significant differences in responses
were found depending on respondents rank.
Table 4.5 T-test for attitudes and rank (N=292)
Variable Means Means Difference
Line Officer Supervisor
1.1 am more likely to be injured during a DV call than any other type of call. 2.88 2.74 0.14
2. Too many DV calls are for verbal family arguments. 2.29 2.17 0.12
3. I need more freedom in deciding how to handle situations at DV calls. 2.24 2.28 -0.04
4. Its often best to arrest both parties in DV calls. 4.22 4.11 0.12
5. Arresting someone at a DV call seldom helps reduce future DV incidents. 3.16 3.36 -0.20
6. Police should arrest DV cases only when there is clear evidence of injury. 3.43 3.44 -0.02
Table 4.5 (Cont.)
7. DV calls take too much of officers time and effort. 2.55 2.28 0.27
8. Identifying the primary aggressor at a DV call is difficult. 3.68 3.57 0.11
9. DV is best handled as a private matter, rather than by police. 4.69 4.78 -0.10
10. Prosecutors usually follow up effectively in DV arrests. 3.60 3.83 -0.23
11. Its often hard to know who to arrest in DV incidents. 3.84 3.89 -0.05
12. lam less likely to make a DV arrest if the suspect is cooperative at the scene. 4.78 4.74 0.03
13. A mandatory arrest policy is the best approach to DV calls. 4.00 4.11 -0.10
14. Many DV victims could easily leave their relationships, but dont. 2.88 2.72 0.16
15. A major problem with DV is repeat calls to the same address. 2.11 2.26 -0.15
16. Substance and/or alcohol abuse is the main cause of DV. 2.71 2.85 -0.14
17.1 am more likely to make DV arrests when children are present. 3.72 4.04 -0.32
18. DV offenders must be arrested even when the victims dont feel it is necessary. 3.04 2.96 0.08
19. Its often hard to decide whether theres probable cause for arrest in DV cases. 4.04 4.09 -0.04
20. Most DV incidents stem from abusers need for power and control over victims. 2.84 3.17 -0.33
21. I am more likely to make an arrest if the victim is cooperative at the scene. 3.56 3.81 -0.25
22. Women are just as likely as men to engage in DV. 2.42 2.40 0.01
Table 4.5 (Cont.)
23. DV is higher among minorities compared to Whites. 4.01 4.21 -0.20
24. Men abused by their partners are less likely than abused women to report DV. 2.58 2.38 0.20
25. DV laws should provide protections to same sex partners. 2.58 2.72 -0.14
26. Minority victims are more likely to call the police than White victims. 4.14 4.21 -0.07
27. More training would help me assess DV scenes. 4.33 4.40 -0.08
28. Gays & lesbians DV occurs for the same reasons it does with heterosexuals. 2.64 2.51 0.13
* p < .05
Age and Years of Experience
Correlations were conducted between all of the attitudinal questions and the two
continuous variables: officer's age and years of experience at current agency. Table 4.6
illustrates these results. Since age and years of experience were the only two variables of
interest, the remainder of the matrix was omitted to conserve space. As shown in the table
there were only two questions to which the officers responses differed based on age.
Younger respondents were more likely to agree with the statements: I am more likely to
make DV arrests when children are present and I am more likely to make an arrest if the
victim is cooperative at the scene.
Table 4.6 Correlation for attitudes, age, and experience (N=292)________________________
Variable Age Experience
1. I am more likely to be injured during a DV call than any other type of call. -0.10 -0.08
2. Too many DV calls are for verbal family arguments. -0.07 0.07
Table 4.6 (Cont.)
3.1 need more freedom in deciding how to handle situations at DV calls. -0.04 -0.14*
4. Its often best to arrest both parties in DV calls. -0.10 -0.04
5. Arresting someone at a DV call seldom helps reduce future DV incidents. 0.11 0.16*
6. Police should arrest DV cases only when there is clear evidence of injury. -0.02 0.03
7. DV calls take too much of officers time and effort. -0.05 -0.06
8. Identifying the primary aggressor at a DV call is difficult. -0.04 -0.05
9. DV is best handled as a private matter, rather than by police. 0.04 0.00
10. Prosecutors usually follow up effectively in DV arrests. -0.03 -0.01
11. Its often hard to know who to arrest in DV incidents. -0.03 -0.08
12. lam less likely to make a DV arrest if the suspect is cooperative at the scene. 0.03 -0.01
13. A mandatory arrest policy is the best approach to DV calls. -0.03 -0.02
14. Many DV victims could easily leave their relationships, but dont. -0.07 0.02
15. A major problem with DV is repeat calls to the same address. 0.08 0.10
16. Substance and/or alcohol abuse is the main cause of DV. 0.01 0.10
17.1 am more likely to make DV arrests when children are present. 0.16* 0.17*
18. DV offenders must be arrested even when the victims dont feel it is necessary. -0.01 0.15*
19. Its often hard to decide whether theres probable cause for arrest in DV cases. -0.04 -0.04
20. Most DV incidents stem from abusers need for power and control over victims. 0.06 0.20*
Table 4.6 (Cont.)
21.1 am more likely to make an arrest if the victim is cooperative at the scene. 0.18* 0.06
22. Women are just as likely as men to engage in DV. 0.02 0.04
23. DV is higher among minorities compared to Whites. 0.01 -0.02
24. Men abused by their partners are less likely than abused women to report DV. -0.05 -0.02
25. DV laws should provide protections to same sex partners. 0.05 0.11
26. Minority victims are more likely to call the police than White victims. -0.02 -0.04
27. More training would help me assess DV scenes. -0.04 0.08
28. Gays & lesbians DV occurs for the same reasons it does with heterosexuals. 0.00 0.13*
* p < .05
The number of officers years of experience at their current agency was related to
differences in responses to six of the attitudinal questions significantly. More experienced
officers were more likely to agree with the statement: I need more freedom in deciding how
to handle situations at domestic violence calls. Officers with less experience were more
likely to agree with the following statements: Arresting someone at a domestic violence call
seldom helps reduce future domestic violence incidents, I am more likely to make domestic
violence arrests when children are present, Domestic violence offenders must be arrested
even when the victims don't feel it is necessary, Most domestic violence incidents stem from
abusers need for power and control over victims, and Gays and lesbians domestic violence
occurs for the same reasons it does with heterosexuals.
Bivariate analysis was conducted on the attitudinal questions based on type of
agency. Table 4.7 presents these findings. Agency type, city versus county, had an impact
on the responses to half (n=14) of the attitudinal statements, which is influence than any of
the other variables examined.
Table 4.7 T-test for attitudes and agency type (N=292)
Variable Means Means Difference
1.1 am more likely to be injured during a DV call than any other type of call. 2.52 2.92 -0.39
2. Too many DV calls are for verbal family arguments. 2.62 2.21 0.41*
3.1 need more freedom in deciding how to handle situations at DV calls. 2.45 2.21 0.24
4. Its often best to arrest both parties in DV calls. 4.24 4.20 0.04
5. Arresting someone at a DV call seldom helps reduce future DV incidents. 3.40 3.16 0.24
6. Police should arrest DV cases only when there is clear evidence of injury. 3.54 3.41 0.14
7. DV calls take too much of officers time and effort. 3.60 2.32 1.28*
8. Identifying the primary aggressor at a DV call is difficult. 3.19 3.74 -0.55*
9. DV is best handled as a private matter, rather than by police. 4.67 4.71 -0.05
10. Prosecutors usually follow up effectively in DV arrests. 3.36 3.68 -0.33
11. Its often hard to know who to arrest in DV incidents. 3.55 3.90 -0.36*
12. I am less likely to make a DV arrest if the suspect is cooperative at the scene. 4.88 4.76 0.12
13. A mandatory arrest policy is the best approach to DV calls. 3.67 4.08 -0.41
Table 4.7 (Cont.)
14. Many DV victims could easily leave their relationships, but dont. 2.76 2.87 -0.11
15. A major problem with DV is repeat calls to the same address. 2.48 2.08 0.40*
16. Substance and/or alcohol abuse is the main cause of DV. 2.71 2.74 -0.03
17.1 am more likely to make DV arrests when children are present. 4.31 3.68 0.63*
18. DV offenders must be arrested even when the victims dont feel it is necessary. 2.93 3.04 -0.12
19. Its often hard to decide whether theres probable cause for arrest in DV cases. 3.67 4.11 -0.45*
20. Most DV incidents stem from abusers need for power and control over victims. 3.24 2.84 0.40*
21.1 am more likely to make an arrest if the victim is cooperative at the scene. 4.02 3.53 0.50*
22. Women are just as likely as men to engage in DV. 2.02 2.48 -0.46*
23. DV is higher among minorities compared to Whites. 4.76 3.92 0.84*
24. Men abused by their partners are less likely than abused women to report DV. 2.76 2.52 0.25
25. DV laws should provide protections to same sex partners. 2.29 2.66 -0.37
26. Minority victims are more likely to call the police than White victims. 4.57 4.08 0.49*
27. More training would help me assess DV scenes. 3.81 4.43 -0.62*
28. Gays & lesbians DV occurs for the same reasons it does with heterosexuals. 2.24 2.69 -0.45*
* p < .05
Respondents from the county agency were significantly more likely than officers from
the city agency to agree that identifying a primary aggressor was difficult. While officers from
the county and city departments seemed neutral about knowing who to arrest in a domestic
violence situations, respondents from the county agency were significantly more likely to
agree with the statement: Its often hard to know who to arrest in domestic violence
incidents. This finding is consisted with another significant finding when comparing attitudes
held by law enforcement officers in two types of agencies. County respondents were
significantly more likely than city respondents to agree that it is often hard to decide whether
theres probably cause on domestic violence cases. Not surprising, based on the results just
presented, that county officers were significantly more likely to agree that more training would
help assess domestic violence scenes. Respondents from the city agency were significantly
more likely than respondents from the county agency to make a domestic violence arrest
when children are present and when the victim is cooperative at the scene. Respondents
from the city agency are significantly more likely than respondents from the county agency to
believe that domestic violence is higher among minorities and that minority victims are more
likely to call the police than White victims. Officers in the county agency were significantly
more likely than respondents from the city agency to agree to the statements: Women are
just as likely as men to engage in DV" and Gays and lesbians DV occurs for the same
reasons in does with heterosexuals.
The results suggest that city police officers may hold more frustration compared to
county officers. Officers employed by city agency were more likely to agree with the following
statements: Too many DV calls are for verbal family arguments, DV calls take too much of
officers time and effort, and A major problem with DV is repeat calls to the same address.
Results indicate overall attitudes about domestic violence vary by the type of agency
an officer is employed by. Findings also indicate that there are many attitudinal differences in
opinions about domestic violence based on demographic factors. To control for these
differences a multivariate analysis was conducted by estimating ordinary least squares
Each of the variables in which there was a significant difference present in attitudes
based on agency was analyzed individually. Using OLS regression, models were estimated
to see whether agency type still had an effect on the responses while controlling for gender,
race, rank, age, and years of experience. Even though the analyses were conducted
separately. Table 4.8 illustrates whether the effect of the type of agency still remained
significant for the 14 variables in which the differences were present in bivariate analyses.
Therefore, 14 models were estimated, with the dependent variables being attitudes that were
significantly different according to the t-tests. Independent variables included: gender, race,
rank, age, and years of experience at current agency.
Table 4.8 OLS regression for attitudes and agency type (N=292) ___________
Variable Agency Type B SE P-value
2. Too many DV calls are for verbal family arguments. -0.22 0.19 0.24
7. DV calls take too much of officers time and effort. -1.37 0.21 0.00*
8. Identifying the primary aggressor at a DV call is difficult. 0.54 0.20 0.01*
11. Its often hard to know who to arrest in DV incidents. 0.26 0.19 0.17
15. A major problem with DV is repeat calls to the same address. -0.42 0.17 0.01*
17.1 am more likely to make DV arrests when children are present. -0.61 0.22 0.01*
19. Its often hard to decide whether theres probable cause for arrest in DV cases. 0.45 0.19 0.02*
Table 4.8 (Cont.)
20. Most DV incidents stem from abusers need for power and control over victims. -0.29 0.21 0.17
21.1 am more likely to make an arrest if the victim is cooperative at the scene. -0.63 0.24 0.01*
22. Women are just as likely as men to engage in DV. 0.53 0.21 0.01*
23. DV is higher among minorities compared to Whites. -1.07 0.22 0.00*
26. Minority victims are more likely to call the police than White victims. -0.65 0.23 0.01*
27. More training would help me assess DV scenes. 0.89 0.22 0.00*
28. Gays & lesbians DV occurs for the same reasons it does with heterosexuals. 0.69 0.23 0.00*
* p < .05
The regression results show that, after controlling for the demographic measures
discussed earlier, 11 of the 14 variables originally showing an impact of agency type still
show a significant effect on attitudes. Officers in the county agency were still more likely than
officers in the city agency to agree that primary aggressor identification was difficult.
Respondents from the county department were also more likely to agree that women are as
likely to engage in domestic violence as men, and that domestic violence in homosexual
relationships occurs for the same reasons as domestic violence in heterosexual relationships.
Officers in the city agency were significantly more likely to disagree with difficulty determining
probable cause and the need for more training to help assess DV situations.
Officers in the city agency were significantly more likely than officers employed by the
county agency to agree that DV calls take too much time and effort and that repeat calls are a
major problem. City officers were significantly more likely to agree that the presence of
children influences their decision to arrest, compared to officers in the county department.
From the multivariate analysis, officers in both departments disagreed with the statements
DV is higher among minorities compared to Whites" and Minority victims are more likely to
call the police than White victims, yet county officers were significantly more likely to do so
than city officers. Overall, the results from the multivariate analysis expose interesting
differences in attitudes of officers from these two different departments.
DISCUSSION AND CONCLUSIONS
It is critical to consider law enforcement officer attitudes relating to domestic violence
incidents, victims, and offenders when focusing on the criminal justice systems response to
this issue. Attitudes among police officers about domestic violence have changed over the
years. Law enforcement attitudes have become less traditional, more accepting of domestic
violence as a real crime, and as a social problem that officers recognize that must be
addressed by the legal system, instead of remaining behind closed doors. The purpose of
the current research was to add to the existing research about law enforcement attitudes,
using self-report surveys, in an attempt to fill in gap in this area of research. This study
examined the attitudes of 292 police officers employed at a county and city department in a
Current Findings Within the Context of Prior Literature
Overall, officers in the overall sample supported treating domestic violence as a
violent crime, as illustrated by the general agreement in the findings that it should not be
handled as a private matter. These findings support prior research in which officers indicated
that domestic violence was a real crime and a major problem (Feder, 1997; Johnson, 2004;
Johnson et al., 1994; Toon & Hart, 2005). Officers in the overall sample, however, showed a
high level of frustration regarding the time it takes to respond to domestic violence calls and
the occurrence of repeat calls. Officers from the city department were significantly more
likely to feel this way compared to officers from the county department. This finding mirrors
that of Toon and Harts (2005) finding regarding officers from urban versus rural departments.
Looking at the census data for the respective areas that the two law enforcement agencies in
the sample serve, it is clear that the city officers work in a more populated area and
experience a higher volume of calls, which may in turn lead to this high level of frustration.
One of city department officers, when indicating what would help the officer respond to DV
calls, expressed his frustration with the time and paperwork domestic cases require, stating:
A reduction in the amount of paperwork and procedures required for routine domestic
violence calls. An average DV call takes two officers out of service for an excess of two
hours. As a result officers cannot effectively focus on other matters that require their
Even though officers from the city agency were more likely to agree with time
consumption and concerns about repeat calls, officers from the county agency also agreed
that repeat calls were problematic in their area. One officer from the county agency spoke to
this issue saying:
... [T]he cases where victim is not cooperative and repeat calls with same address,
making arrest and victim services is seemingly useless... Nothing can be told to
these victims to make them try to leave or want to leave the violence... Many
opportunities and funds are offered to help leave, but they turn their aggression to
the police they call.
In addition to feeling frustrated about procedures, officers seemed to have mixed
feelings regarding prosecutorial effectiveness, and some officers expressed their
disappointment and how it affected their work. One officer from the city department stated:
I feel that the DA's need to be more aggressive in prosecuting DV cases, despite
victims lack of cooperation. We see too many suspects back in the home the next
day or week because all charges have been dropped due to the victim wanting to
drop charges. If the courts dont tolerate a victim's lack of cooperation then the street
officer should not have to make a mandatory arrest when the victim is against it at
the time of the report. We are doing our job by completing very thorough
investigation on scene only to see the DAs not filing the case. It is very frustrating!
These feelings of frustration with the procedures that officers have to follow when
responding to domestic violence incidents are quite concerning. If an officer feels frustrated
about the overall issue of domestic violence call before he or she responds to a domestic
violence call, knowing the amount of frustration and paperwork it will require officers may be
more skeptical about the case before even arriving at the incident. Several officers from both
agencies cited less paperwork as a factor that would help in their response to domestic
violence cases. This type of frustration may be addressed by training the officers on the
complexity of domestic violence. A major component of this training would be to inform
officers of why the work that they do is so important to the victims in these cases. This
knowledge may help officers be more sympathetic toward domestic violence which hopefully
will make officers feel that the paperwork they fill out is worth the effort.
Officers in the sample were also unsure of whether arrest alone is an appropriate
response to domestic violence incidents. The majority of the officers in the sample disagreed
with the mandatory arrest policy being the best response to domestic violence cases, which
supports the findings of Toon and Harts (2005) study. Only 21% of the sample disagreed
with the statement Arresting someone at a DV call seldom helps reduce future DV incidents,
which supports Blount et al.s (1992) study and partially supports Friday et al.s (1991)
findings that officers supported the mandatory arrest policy, yet remained unconvinced that it
actually helped reduce the number of domestic violence incidents. Also, less experienced
officers were more likely to disagree with the idea of arrest reducing recidivism, which
supports Stalans arid Finns (2006) findings which showed that experienced officers were
more likely to think that arrest is a fair option. Although only 16% of the officers in the sample
disagreed that offenders should be arrested even when the victim does not feel it is
necessary, this suggests that even though officers may not agree with the policy, it appears
that the policy is enforced. Also, female officers were more likely to agree with this construct
compared to male officers, which may mean that these officers are more likely to enforce the
policy and not be affected by the victims behavior at the scene.
Victim cooperation was also examined, with the majority of the officers responding
neutrally to the issue, yet younger officers were more likely to arrest offenders if the victims
were cooperative at the scene while county older were less likely to do so. The cooperation
of the offender at the scene of the incident was also examined, with the majority of the
officers stating that they were indifferent to offender cooperation when determining whether
or not to make an arrest. Neither the behavior of the suspect nor the victim seemed to affect
the officers, some were more likely than others to make an arrest if children were present at
the scene of the incident. Although overall officers responded neutrally to this concept, Non-
White officers were more likely to agree with it, as were younger officers and officers with less
experience. In contrast, officers from the county agency were significantly less likely than city
officers to allow the presence of children to affect their decision to make an arrest. This
finding shows that novice and minority officers may be more sympathetic to the presence of
children at a domestic violence call, and may feel the need to enforce the law in order to
protect these children, all of which supports research conducted by Paoline et al. (2000).
Officer attitudes regarding the appropriateness of arrest as a response to domestic
violence may be connected to the fact that the majority of the responding officers in the
sample believed that too many domestic violence cases consist of verbal family arguments.
One of the city officers said that In most cases domestic violence is verbal and requires NO
police... Again, these findings point to the need for training on the dynamics of domestic
violence. Officers need to be informed that verbal abuse is abuse nonetheless, and many
times it may be just one of the incidents in the pattern of abuse, and may be the only incident
that actually will be reported to the police. This attitude is supported by the response to the
statement Police should arrest DV cases only when there is clear evidence of injury, to
which the sample was split almost evenly among the response options. Although female
officers in the overall sample were more likely to disagree with the above statement, it is
important for all officers to be educated about all aspects of domestic violence. Explaining
the agencys policies, and the reasons behind the policies, to all of the officers is an
extremely important step in gaining the officers support of the departments rules and
regulations. For example, one officer expressed his disagreement with dual arrest, saying
that The ability to arrest both parties sometimes would help him respond to DV cases and
that his agency does not allow it for some reason.
Overall, officers in the sample seemed to have mixed attitudes as to the prevalence
of domestic violence among different groups of people, including women and minorities. The
majority of officers agreed that men were less likely to report incidents of domestic violence
than females and they also felt that women were just as likely as men to engage in domestic
violence, with county officers being significantly more likely to express agreement with this
notion. Officers also expressed disagreement with the fact that domestic violence occurs
more among minorities and that minorities are more likely to report the violence, which goes
directly against Caetano et al.s (2005) and Rennison and Welchans (2000) findings. On
both of these concepts the whole sample expressed disagreement; yet county officers were
significantly more likely to disagree with these statements compared to city officers. County
officers may be more likely to disagree since the area which they serve is not highly
populated by Hispanics or African Americans. It is surprising that officers from the city
department, where the minority population constitutes a significant portion of the whole
population, would express this same attitude. Overall, the officers in the sample felt that
homosexual domestic violence occurs for the same reasons as heterosexual violence. A
significant difference in opinion was found, because even though both groups agreed with the
concept, county officers and less experienced officers were significantly more likely to agree
than city officers and officers with more experience. Also, officers in the entire sample felt
that domestic violence laws should provide protection for homosexual couples.
An area of major concern for responding law enforcement officers in the sample was
the need for more discretion, which confirms Toon and Harts (2005) findings. Overall, both
males and females agreed with the need for more discretion, yet males and officers with
more experience were significantly more likely to feel this way about discretion compared to
females and officers with fewer years of law enforcement experience. When answering the
question of what would help officers respond to domestic violence cases better, numerous
officers cited more freedom and more discretion as their answers. Here, again, the
incorporation of policy development is necessary, especially when looking at the qualitative
data. For example, one officer expressed his dislike of the mandatory arrest policy stating
that what would help him respond to domestic violence cases would be more latitude in
arresting the primary aggressor because sometimes I arrest someone only because I know
I will get in trouble if I dont. Also, the officers, especially officers from the city agency, do not
believe that training would assist them in responding to domestic violence cases better. It is
understandable that officers would want to be able to use discretion when responding to
domestic violence cases, as it is one of their most powerful weapons. Even though officers
did not express an overall need for more training, the results show that additional training is
necessary. In addition to the mixed attitudes discussed above, officers also seem confused
about the dynamics of domestic violence. Only 6% of the sample disagreed with the idea of
substance or alcohol abuse as a main cause of domestic violence. At the same time, 40% of
the officers agreed that power and control are the cause of domestic violence, with officers
who had less experience being more likely to agree with the notion of power and control
compared to officers who had more law enforcement experience at their current agency.
Another area of concern, especially since training does not seem to be viewed as
beneficial by the majority of responding officers, is their self-reported doubt in identifying the
primary aggressor and probable cause. Officers from both agencies reported neutral
responses to these concepts, with officers from the county department being more likely to
agree that they have difficulty identifying the primary aggressor. Officers in the city agency
were more likely to express that determining probable cause was not difficult, while county
officers reported neutral attitudes.
These findings are rather concerning because of the many areas in which responding
law enforcement officers seem unsure of themselves. Officers seem confused about the
complexities of domestic violence and what the appropriate response to this crime should be.
Training, even though its viewed as unnecessary, is highly needed for all law enforcement
officers that respond to domestic violence cases. The findings from the current research
show that when looking at the whole sample and comparing officers based on demographic
factors, there are more similarities in their attitudes than there are differences, which supports
empirical work conducted by Robinson (2000) and Saunders (1995). Yet, when the two
types of agencies are compared there are almost as many differences as there are
similarities, and this causes concern. The necessity for training exists not only to improve
officers knowledge about the dynamics of domestic violence, but also to standardize the
response that city and county agencies provide to domestic violence victims.
It is important to acknowledge several methodological limitations to this study. The
first limitation involved the sample size. The sample size from the city agency (N=250) is
substantial, but the number of officers who completed surveys from the county agency was
not as large (N=42). Also, the sample was neither random nor representative, and essentially
a convenience sample. Future research should gather a large sample of officers from city
and county agencies in one state that is random and representative, as it would make the
comparison among the officers and between the two agencies more accurate and allow for
the overall findings of the study to be generalizable.
Another limitation involves the method in which the surveys were administered by
both agencies. Officers in the county agency were told by the departmental contact that the
survey was mandatory, which was a break in our protocol and may have impacted the way in
which the officers responded. This also made the survey administration process different for
the two agencies, which further impacts the comparison between the agencies. Since the
officers were not monitored during the survey taking process, there is no way to ensure that
they did not discuss their answers with fellow officers and that these discussions did not
influence the way in which officers responded to questions.
This exploratory study suggests the need for future research to address interaction
effects. Since the two samples in this study differ significantly in the demographic
breakdown, it is important to consider that variations in the agencies institutional context may
mediate the impact of individual officer characteristics on their attitudes regarding domestic
violence. Comparisons based on race and gender, in particular, are of concern since
dominant police culture has historically consisted of White males.
A final limitation of this study involves a lack of qualitative research. Most of the
qualitative data gathered from the officers were negative toward domestic violence policies
and procedures. Perhaps conducting in-person interviews with a randomly selected sample
of law enforcement officers would produce a more representative spread of responses, as it
seems the responses provided by officers to the open-ended questions were given by officers
who needed an outlet to voice their opinions.
The findings of this study indicate that officers have mixed attitudes regarding the
dynamics of domestic violence, appropriate ways of responding to domestic violence, and
departmental procedures for doing so. Many of the responding officers in the sample were
highly frustrated regarding the amount of time and paperwork that is involved in responding to
domestic violence calls. Since this paperwork is a necessary step in processing cases
through the system, departments may want to invest in hiring special domestic violence
investigators that would only respond to these types of cases and therefore would lessen the
work of their fellow officers. Also, when the officers were qualitatively asked what would help
them improve their response to domestic violence incidents, a few officers mentioned
incorporating more resources for victims. Responses included dispatching a domestic
violence counselor to the scene with officers" and providing more information to victims to
enable them to get out of their present situations i.e., resources, shelters, finances, jobs,
education classes, etc. The advocate referenced to by these officers could be a part of a
special domestic violence unit at the police department, or could be associated with a victim
advocacy group or service providing agency in the community. Providing additional
resources to victims may help them cooperate with the subsequent prosecution of the
The findings of this study point to the need for more training, yet most officers
disagreed with this option actually helping them assess domestic violence incidents. Training
should not so casually be given up on. The results show that officers from two types of
agencies differ on several issues related to responding to domestic violence, which only
increases the need for a standardized policy regarding the appropriate response to this
complex crime. Even though every domestic violence incident is different, and involves
different offenders, victims, children, and extralegal factors, the decision on what to do should
not be left to an officer's discretion. Although attitudes regarding domestic violence are
changing, this study showed that officers are still not convinced about some of the important
causal factors and prevalence rates of this serious crime. Again, this supports the need for
training. Since domestic violence is such a prevalent crime, officers should go through
mandatory intense training sessions not only at the beginning of their careers but also on an
annual basis. The findings of this study also show the need for this intense mandatory
training to be standardized across the state. Officers in the same state should respond to
domestic violence cases in the same way. Future research should focus on conducting more
attitudinal comparison studies across departments. In addition to soliciting attitudes directly
from the officers, in-depth interviews with a random sample may add to the further
understanding of law enforcement officer attitudes regarding domestic violence.
Law Enforcement Survey on Domestic Violence Incidents
This survey is interested in your perspectives on domestic violence (DV) and policing. Your
answers and comments are confidential. If for any reason you have questions, please contact
Mary Dodge, Ph.D. or Angela Gover, Ph.D. at UCD&HSC at (303)556-5987. For questions about
your rights as a research subject you may contact the HSRC Administrator, 1380 Lawrence Street,
Suite 300, (303) 556-4060. Thank you for your cooperation and support.
Based on vour experience and opinion, please check your level of agreement with each of the
Stro ngly Agre e Agre e Some what Agree Some what Disagr ee Disag ree Stron giy Disag ree
1 am more likely to be injured during a DV call than any other type of call.
Too many DV calls are for verbal family arguments.
1 need more freedom in deciding how to handle situations at DV calls.
Its often best to arrest both parties in DV calls.
Arresting someone at a DV call seldom helps reduce future DV incidents.
Police should arrest DV cases only when there is clear evidence of injury.
DV calls take too much of officers time and effort.
Identifying the primary aggressor at a DV call is difficult.
DV is best handled as a private matter, rather than by police.
Prosecutors usually follow up effectively in DV arrests.
It's often hard to know who to arrest in DV incidents.
1 am less likely to make a DV arrest if the suspect is cooperative at the scene.
A mandatory arrest policy is the best approach to DV calls.
Many DV victims could easily leave their relationships, but don't.
A major problem with DV is repeat calls to the same address.
Substance and/or alcohol abuse is the main cause of DV.
1 am more likely to make DV arrests when children are present.
DV offenders must be arrested even when the victims dont feel it is necessary.
Its often hard to decide whether there's probable cause for arrest in DV cases.
Most DV incidents stem from abusers' need for power and control over victims.
1 am more likely to make an arrest if the victim is cooperative at the scene.
Women are just as likely as men to engage in DV.
DV is higher among minorities compared to Whites.
Men abused by their partners are less likely than abused women to report DV.
DV laws should provide protections to same sex partners.
Minority victims are more likely to call the police than White victims.
More training would help me assess DV scenes. n n
Gays & lesbians DV occurs for the same reasons it does with heterosexuals.
Please turn over and complete items on the other side.
Does your agency have a domestic violence unit? Yes L 1 No Don't Know
For statistical purposes only, please indicate the following
Gender: Male Female
Race/Ethnicity: White African American Hispanic
Asian Native American Other (please specify)_____________
How many years have you worked at your current agency?
How many years of experience in law enforcement at a previous agency?
Have you ever responded to a domestic violence (DV) call?
Does your job involve any special domestic violence duties?
If yes, please specify:
What would help you in dealing with domestic violence calls?
Please add any other comments or thoughts about your experience with domestic violence calls.
INFORMED CONSENT FORM
University of Colorado at Denver and Health Sciences Center
Mary Dodge, Ph D. Graduate School of Public Affairs
Campus Box 142, P.O. Box 173364
Denver, Colorado 80217-3364
Phone: 303-556-5987, Fax: 303-556-5971
Law Enforcement Survey on Domestic Violence Incidents Informed Consent
About the Study
This study is designed to explore the perceptions, opinions, and impact of training for law
enforcement officers who deal with domestic violence offenders and victims in the criminal
justice system. This survey is designed to explore your opinions and perspectives on policing
and domestic violence, so there are no right or wrong answers. We hope that the results of
the study will inform decisions around law enforcement training on domestic violence issues.
The survey will take approximately 15 to 20 minutes to complete.
The choice of whether to participate in this study is completely up to you. Your participation is completely
voluntary, and refusal to participate will involve no penalty to you. If you decide to participate in the
study, you have the right to withdraw your consent or discontinue participation at any time. You also
have the right to refuse to answer any question you do not wish to answer.
We guarantee that all of your answers and comments are confidential. Your responses to survey
questions will be protected according to professional standards established by the Institutional Review
Board at the UCD&HSC. Please do not include your name on the survey materials to help protect
confidentiality. The information you provide will only be reported in aggregate form. Access to the raw
data is limited to the researchers and graduate student assistants working on this project. The data will
be securely stored for a 3 year time period only.
Benefits and Risks to Participation
Although you may not directly benefit from completing this survey, this research will provide you with the
opportunity to have direct input on how to improve the response to and investigation of domestic violence
calls. The minimal risks to your participation may include feelings of discomfort or embarrassment about
domestic violence issues.
Researcher Contact Information
If you have any questions or concerns about the research study, please contact Mary Dodge,
Ph.D., Associate Professor at UCD&HSC by phone (303) 556-5987 or email:
. If you have questions about your rights as a participant, you may
contact the Human Subjects Research Committee Administrator, 1380 Lawrence Street, Suite
300, at (303)556-4060.
Thank you for support,
Mary Dodge, Ph.D.
Angela Gover, Ph.D.
CORRELATION MATRIX, MEANS, AND STANDARD DEVIATIONS FOR VARIABLES IN
9 R to CM Pi co 5= UP - cb CD cb ft s ft ft 3 CD a CD 5 a ft ft 5 ft St- ft ft 8 a a CM 8
3 3 a 3 Cs Sd a - a CM cq a ft ft a 'tr 3 8 CM cb Zb cb Pm 3 3 - 8 cb 8 ft s 3 r 8
3 El 3
3 8 & s a cb 3 Si a o cm R & ft ft 5 ft a 5 ?S Zb 8 a S CM a
3 3 CM o 8 s CM CD O) CD UD Pm ft a CP 3b CM 5 8 O) ft s 8 a a 5> CD P 8 UP UP CM cn
jÂ§ 5 a a CM (O CD C7> s r CD h- O Pm CO CO CD CTP CD - co CD JP CP O) CD ft ft - 8 & ft a S cr 8
3 a a a a a 1^4 Si s ft ft PL- ft a Op CD fq ft Zb CD ft a CM
3 & s 5 CD Sd P5 Jh CD Sd CM 3 cb Pi &> CM jr cn CD 3b SSr 3 UP 8 S OP 8
3 IS ft a a a a cb o- o 8 ft 8 5 8 ft ft CM UP CD 8 a a a CM Cr
3 3 rr *=t M to a a Zb 8 Zb a Cb ft a Zb Zb CM 5 IS CO CD ft CM a a 8 tit 8
9 ir> 03 CM a ft S CD o> pw ft s> CM s ft o ft 3 ft ft O) 8 a CO 8
m X & x ?M cm a CD a 8 ft CD 8 ft CD Ed CM 8 a ft CM CD 8 kir 125
M X 3 CD CO ft a CM p; ft 8 ft ft Pm ?T a Op h
es X 8 CD cm Cb ?b a ft P- *"1 25 CD 5 B O) CD ft ft a co CM 960
* X q a cz> o cq a Cb & cq cq CD ft ft a ft Zb CM a a CM a
3 a a cq 8 a a u> 04 lb oq UP 3 ft ft ft 5 s a 1.44
X co S s S 8 CO CD Pw cb 5 CO CD ft ft cb a IS: Â£ CD
a X a s 3 CD S lb CM CD S? a lb (D Pm 8 a a cb 8
N a 8 IS CD CD Cb IS a s CD 8 3
X o CD CD CD to 04 CD Cb
P X cq S UP CM S a P- rs a Zb a a a
8 8 a 3 ft - Op ft 8 250 8
a to CD 8 cq ft 05 Of Pm a cb e
= tb 5 Â§ ZS ri a R
CD ft 8
a. ft a Si Of UP
a |- q CD 1.00 Â£3 CM 8
o CO o CD Â£ CM R
** ft CD CD CM o2 CO UP
1 L X 3 >< >** O X x w X p X X X jf X QB X 0> X 3 5 3 >P >? J? J? >? 9 3 9 a. t>
* p < 0.05
p = Mean
a = Standard Deviation
xi = Age
x2 = Experience at Current Agency
x3= I am more likely to be injured during a DV call than any other type of call.
x4 = Too many DV calls are for verbal family arguments.
x5 = I need more freedom in deciding how to handle situations at DV calls.
x6 = Its often best to arrest both parties in DV calls.
x 7 = Arresting someone at a DV call seldom helps reduce future DV incidents.
x8 = Police should arrest DV cases only when there is clear evidence of injury.
x9 = DV calls take too much of officers time and effort,
x 10 = Identifying the primary aggressor at a DV call is difficult.
Xu = DV is best handled as a private matter, rather than by police,
x 12 = Prosecutors usually follow up effectively in DV arrests,
x 13 = Its often hard to know who to arrest in DV incidents.
x 14 = I am less likely to make a DV arrest if the suspect is cooperative at the scene,
x 15 = A mandatory arrest policy is the best approach to DV calls,
x i6 = Many DV victims could easily leave their relationships, but dont,
x 17 = A major problem with DV is repeat calls to the same address,
x is = Substance and/or alcohol abuse is the main cause of DV.
x 19 = I am more likely to make DV arrests when children are present.
x2o = DV offenders must be arrested even when the victims don't feel it is necessary.
x2i = Its often hard to decide whether theres probable cause for arrest in DV cases.
x22 = Most DV incidents stem from abusers need for power and control over victims.
x23 = I am more likely to make an arrest if the victim is cooperative at the scene.
x24= Women are just as likely as men to engage in DV.
x25 = DV is higher among minorities compared to Whites.
x 26 = Men abused by their partners are less likely than abused women to report DV
x27 = DV laws should provide protections to same sex partners.
x2B = Minority victims are more likely to call the police than White victims.
x2g = More training would help me assess DV scenes.
x 3o= Gays & lesbians DV occurs for the same reasons it does with heterosexuals.
Apsler, R., Cummins, M. R., & Carl, S. (2003). Perceptions of the police by female victims
of domestic partner violence. Violence Against Women, 9(11), 1318-1335.
Berk, R. A., Campbell, A., Klap R., & Western, B. (1992). The deterrent effect of arrest in
incidents of domestic violence: A Bayesian analysis of four field experiments.
American Sociological Review, 57(5), 698-708.
Binder, A., & Meeker, J. W. (1993). Implications of the failure to replicate the Minneapolis
experimental findings. American Sociological Review, 58(6), 886-888.
Blount, W. R., Yegidis, B. L., & Maheux, R. M. (1992). Police attitudes toward preferred
arrest: influences of rank and productivity. American Journal of Police, 11(3), 35-
Buchbinder, E., & Eisikovits, Z. (2004). Between normality and deviance: The breakdown
of batterers identity following police intervention. Journal of Interpersonal
Violence, 19(4), 443-467.
Bureau of Criminal Information and Analysis. (1999). Report on Arrests for Domestic
Violence in California. Sacramento: State of California, Criminal Justice Statistics
Buzawa, E. S., & Buzawa, C. G. (2003). Domestic violence: The criminal justice response
(3rd ed.) Thousand Oaks, CA: Sage Publications, Inc.
Caetano, R., Field, C. A., Ramisetty-Mikler, S., & McGrath, C. (2005). The 5-year course of
intimate partner violence among white, black, and Hispanic couples in the United
States. Journal of Interpersonal Violence, 20(9), 1039-1057.
Cascardi, M., & O'Leary, K. D. (1992) Depressive symptomatology, self-esteem, and self
blame in battered women. Journal of Family Violence, 7, 249-259.
Catalano, S. (2006). Intimate Partner Violence in the United States. Retrieved November
10, 2007, from: http://www.ojp.usdoj.gov/bjs/pub/pdf/ipvus.pdf.
Chesney-Lind, M. (2002). Criminalizing victimization: The unintended consequences of
pro-arrest policies for girls and women. Criminology and Public Policy, 2, 81-90.
Coker, A. L., Smith, P. H., Bethea, L., King, M. R., & McKeown, R.E. (2000). Physical
Health Consequences of Physical and Psychological Intimate Partner Violence.
Archives of Family Medicine, 9, 451-457.
Comach, E., Chopyk, V., & Wood, L. (2002). Mean Streets? Locations, Gender Dynamics,
and Patterns of Violent Crime in Winnipeg. Ottawa: Canadian Centre for Policy
Edleson, J. L. (1999). Childrens witnessing of adult domestic violence Journal of
Interpersonal Violence, 14(8), 839-870.
Feder, L. (1997). Domestic violence and police response in a pro-arrest jurisdiction.
Women & Criminal Justice, 8(4), 79-98.
Friday, P. C., Metzgar, S., & Walters, D. (1991). Policing domestic violence: Perceptions,
experience, and reality. Criminal Justice Review, 16(2), 198-213.
Frisch, M. B., & MacKenzie, C. J. (1991). A comparison of formerly and chronic battered
women on cognitive and situational dimensions. Psychotherapy, 28, 339-344.
Garner, R. (2005). Police attitudes: The impact of experience after training. Applied
Psychology in Criminal Justice, 1(1), 56-70.
Heinemann, S. (1996). Timelines of American Women's History. New York: Roundtable
Hirschel, D., & Buzawa, E. (2002). Understanding the context of dual arrest with directions
for future research. Violence Against Women, 8(12), 1449-1473.
Holzworth-Munroe, A., Smutzler, N., & Sandin, E. (1997). A brief review of the research
on husband violence: Part II: The psychological effects of husband violence on
battered women and their children. Aggression and Violent Behavior, 2(2), 179-
Johnson, I. M., Sigler, R. T., & Crowley, J. E. (1994). Domestic violence: A comparative
study of perceptions and attitudes toward domestic violence abuse cases among
social service and criminal justice professionals. Journal of Criminal Justice, 22(3),
Johnson, R. R. (2004). Police officer frustrations about handling domestic violence calls.
The Police Journal, 77, 207-219.
Jones, A. (2000). Next Time, Shell Be Dead: Battering & How To Stop It. Boston: Beacon
Kemp, A., Green, B. L., Hovanitz, C., & Rawlings, E. I. (1995). Incidence and correlates of
posttraumatic stress disorder in battered women. Journal of Interpersonal
Violence, 10(1), 43-55.
Logan, T., Shannon, L., & Walker, R. (2006). Police attitudes toward domestic violence
offenders. Journal of Interpersonal Violence, 21(10), 1365-1374.
Martin, D. (1976). Battered Wives. New York: Pocket Books.
Martin, D. (1979). What keeps a woman captive in a violent relationship?: The social
context of battering. In D. M. Moore (Ed.), Battered women (pp.33-57). Beverly
Hills, CA: Sage.
Maxwell, C. D., Gamer, J. H., & Fagan, J. A. (2001). Effects of arrest on intimate partner
violence: New evidence from the Spouse Assault Replication Program. National
Institute of justice Research in Brief (NCJ# 188199).
Miller, J. (2006). A specification of the types of intimate partner violence experienced by
women in the general population. Violence Against Women, 12(12), 1105-1131.
Miller, S. (1989). Unintended side effects of pro-arrest and their race and class implications
for battered women: A cautionary note. Criminal justice Policy Review, 3, 299-316.
Miller, S. L., & Meloy, M. L. (2006). Womens use of force: Voices of women arrested for
domestic violence. Violence Against Women, 12(1), 89-115.
National Center for Victims of Crime. (1999). Domestic violence. Retrieved May 1,2007,
Osofsky, J. D. (1999). The impact of violence on children. The Future of Children, 9(3), 33-
Paoline E. A., III. (2005). Job satisfaction, burnout, and perception of unfair treatment: The
relationship between race and police work. Police Quarterly, 8(4), 476-489.
Paoline, E. A., Ill, Myers, S. M., & Worden, R. E. (2000). Police culture, individualism, and
community policing: Evidence from two police departments. Justice Quarterly,
Parnas, R. I. (1967). The police response to the domestic disturbance. Wisconsin Law
Review, 2, 914-960.
Rand, M. (1997). Violence-related injuries treated in hospital emergency departments. U.S.
Department of Justice, Bureau of Justice Statistics (NCJ# 156921).
Reitzel-Jaffe, D., & Wolfe, D. A. (2001). Predictors of relationship abuse among young
men. Journal of Interpersonal Violence, 16(2), 99-115.
Rennison, M., &Welchans, W. (2000). Intimate Partner Violence. U.S. Department of
Justice, Bureau of Justice Statistics (NCJ# 178247).
Richie, B., & Menard, A. (n.d.). Timeline of milestones. Retrieved May 1,2007, from
Robinson, A. L. (2000). The effect of a domestic violence policy change on police offers
schemata. Criminal Justice and Behavior, 27(5), 600-624.
Robinson, A. L., & Chandek, M. S. (2000). The domestic violence arrest decision:
Examining demographic, attitudinal, and situational variables. Crime &
Delinquency, 46(1), 18-37.
Rodriguez, M. A., Bauer, H. M., McLoughlin E., & Grumbach K. (1999) Screening and
intervention for intimate partner abuse: Practices and attitudes of primary care
physicians. The Journal of the American Medical Association, 282, 468-474.
Russell, M., & Light, L. (2006). Police and victim perspectives on empowerment of
domestic violence victims. Police Quarterly, 9(4), 375-396.
Saunders, D. G. (1995). The tendency to arrest victims of domestic violence: A preliminary
analysis of officer characteristics. Journal of Interpersonal Violence, 10(2), 147-
Sherman, L. W. (1992). Policing domestic violence: Experiments and dilemmas. New York:
Sherman, L. W. (1993). Implications of a failure to read the literature. American
Sociological Review, 58(6), 888-889.
Sherman, L. W., & Berk, R. A. (1984). The specific deterrent effects of arrest for domestic
assault. American Sociological Review, 49, 261-272.
Sinden, P. G., & Stephens, B. J. (1999). Police perceptions of domestic violence: The
nexus of victim, perpetrator, event, self and law. Policing: An International Journal
of Police Strategies & Management, 22(3), 313-326.
Stalans, L. J., & Finn, M. A. (2006). Publics and police officers interpretation and handling
of domestic violence cases: Divergent realities. Journal of Interpersonal Violence,
Tjaden, P., & Thoennes N. (2000). Full report of the prevalence, incidence,
consequences of violence against women: Findings from the National Violence
Against Women Survey. U.S. Department of Justice (NCJ# 183781).
Toon, R, & Hart, B. (2005). Layers of meaning: Domestic violence and law enforcement
attitudes in Arizona. U. S. Department of Justice, Office on Violence Against
Women (NCJ# 212533).
United States Census Bureau. (2000). State and County QuickFacts. Retrieved November
11,2007, from: http://quickfacts.census.gov/qfd/sates.
Waldrop, A. E., & Resick, P. A. (2004). Coping among adult female victims of domestic
violence. Journal of Family Violence, 19(5), 291-302.
White, M. D., Goldkamp, J. S., & Campbell, S. P. (2005). Beyond mandatory arrest:
Developing a comprehensive response to domestic violence. Police Practice and
Research, 6(3), 261-278.
Wortley, R. K. (2003). Measuring police attitudes toward discretion. Criminal Justice and
Behavior, 30(5), 538-558.
Wynn, M. (2007). Domestic Violence Intervention and Investigation. Training conducted on
February 14, 2007, in Golden, Colorado.
Yates, D. L., & Pillai, V. K. (1996). Attitudes toward community policing: A causal analysis.
The Social Science Journal, 33(2), 193-209.