Walden University
ScholarWorks
"&&%''#"&"#'#% '(&
"&&%''#"&"#'#% '(&
# '#"
Domestic Violence Recidivism: Restorative Justice
Intervention Programs for First-Time Domestic
Violence O!enders
Tamika L. Payne
Walden University
# #*'&"'#" *#%&' .$&&# %*#%&* "((&&%''#"&
%'#' %!"# #+#!!#"& %!"# #+"%!" (&'#!!#"&"'
("''')( '')#!$%')"&'#% '## #&#!!#"&
-&&&%''#"&%#(''#+#(#%%"#$"&&+' "&&%''#"&"#'#% '(&# '#"'# %#%&'&"
$'#%" (&#"" "&&%''#"&"#'#% '(&+"('#%,!"&'%'#%## %#%&#%!#%"#%!'#"$ &
#"'' # %#%&* "((
Walden University
College of Social and Behavioral Sciences
This is to certify that the doctoral dissertation by
Tamika Payne
has been found to be complete and satisfactory in all respects,
and that any and all revisions required by
the review committee have been made.
Review Committee
Dr. Barbara Benoliel, Committee Chairperson, Human Services Faculty
Dr. Tina Jaeckle, Committee Member, Human Services Faculty
Dr. Gregory Hickman, University Reviewer, Human Services Faculty
Chief Academic Officer
Eric Riedel, Ph.D.
Walden University
2017
Abstract
Domestic Violence Recidivism: Restorative Justice Intervention Programs for First-Time
Domestic Violence Offenders
by
Tamika L. Payne
MA, University of Cincinnati, 2010
BS, Old Dominion University 2007
Dissertation Submitted in Partial Fulfillment
of the Requirements for the Degree of
Doctor of Philosophy
Human Services
Walden University
June 2017
Abstract
Domestic violence impacts millions of Americans annually and, in spite of the use of
rehabilitative programs, recidivism in domestic violence continues to be more likely than
in any other offense. To date, batterer intervention programs (BIPs) have not proven to
be consistently impactful in reducing recidivism in cases of domestic violence. The
purpose of this quasi-experimental, quantitative study was to examine differences in
recidivism for first-time male domestic violence offenders who have participated in a BIP
and a more recently developed alternative: victim-offender mediation (VOM). The
theories of restorative justice and reintegrative shaming frame this study to determine if
offenders take accountability for their actions and face the victim in mediation, there can
be a reduction in recidivism. Archival data from records of first-time male, domestic
violence offenders, between the ages of 18 and 30, who participated in either a VOM or
BIP in a county in the Midwest were examined for recidivism 24-months
postintervention, and analyzed with an ANCOVA analysis while controlling for age. The
findings revealed no significant difference in recidivism for first-time male offenders 24-
months post participation in a BIP or a VOM intervention while controlling for age F
(1,109) =.081, p = .777. The findings provide support for the notion that restorative
justice interventions may be an additional intervention used in cases of domestic violence
deemed appropriate for the intervention. The findings from this study can add to the body
of research examining interventions to address the high recidivism in cases of domestic
violence, which impacts victims, offenders, and communities.
Domestic Violence Recidivism: Restorative Justice Intervention Programs for First-Time
Domestic Violence Offenders.
by
Tamika L. Payne
MS, University of Cincinnati, 2010
BS, Old Dominion University. 2007
Dissertation Submitted in Partial Fulfillment
of the Requirements for the Degree of
Doctor of Philosophy
Human Services
Walden University
June 2017
Dedication
I would like to dedicate this study to my grandmothers Faye Payne and Marian Williams
who have truly been examples of strength and courage for me throughout my life. I
would also like to dedicate this to my mother who has been my strength throughout the
journey. And last but not least to my husband, Melvin for the continual love and
encouragement.
i
Table of Contents
List of Tables .......................................................................................................................v
List of Figure...................................................................................................................... vi
Chapter 1: Introduction to the Study ....................................................................................1
Background ....................................................................................................................4
Problem Statement .........................................................................................................6
The Purpose of the Study ...............................................................................................9
Nature of the Study ........................................................................................................9
Research Question .......................................................................................................10
Hypotheses ...................................................................................................................10
Theoretical Framework ................................................................................................10
Definition of Key Terms ..............................................................................................15
Assumptions, Limitations and Delimitation ................................................................16
Significance of the Study .............................................................................................17
Social Change Implications .........................................................................................18
Summary ......................................................................................................................18
Chapter 2: Literature Review .............................................................................................20
Research Strategy.........................................................................................................20
Domestic Violence .......................................................................................................21
Prevalence of Domestic Violence Incidents in the United States ......................... 22
Victims .................................................................................................................. 23
Offenders............................................................................................................... 26
ii
Mental Health and Substance Abuse .................................................................... 28
Domestic Violence Laws .............................................................................................29
International .......................................................................................................... 29
Recidivism ...................................................................................................................32
RJI and Recidivism Demographics ....................................................................... 34
Interventions ................................................................................................................34
Schools of thought .......................................................................................................36
Theoretical Framework ................................................................................................41
Restorative Justice ................................................................................................ 41
Reintegrative Shaming .......................................................................................... 44
Summary ......................................................................................................................48
Chapter 3: Research Method ..............................................................................................50
Procedures ....................................................................................................................50
Research Design and Rationale ...................................................................................52
Methodology ................................................................................................................53
Data Analysis Plan .......................................................................................................55
Archival Data ...............................................................................................................55
Variables of the study ..................................................................................................56
Recidivism ............................................................................................................ 56
Victim-offender mediation.................................................................................... 57
Batterer intervention program ............................................................................... 57
Age ................................................................................................................... 57
iii
Operational Definition of Variables .............................................................................58
Research Question .......................................................................................................59
Hypotheses ...................................................................................................................59
Threats to validity and limitations ...............................................................................59
Ethical Assurances .......................................................................................................61
Assumptions .................................................................................................................61
Summary ......................................................................................................................62
Chapter 4: Results ..............................................................................................................63
Introduction ..................................................................................................................63
Data Collection ............................................................................................................63
Results ..........................................................................................................................64
Summary ......................................................................................................................68
Chapter 5: Discussion, Conclusion, and Recommendations .............................................69
Introduction ..................................................................................................................69
Interpretation of the Findings.......................................................................................69
Recidivism ............................................................................................................ 70
VOM ................................................................................................................... 71
BIP ................................................................................................................... 72
Age ................................................................................................................... 73
Theoretical Framework ................................................................................................74
Limitations of the Study...............................................................................................76
Discussion ....................................................................................................................78
iv
Recommendations ........................................................................................................80
Implications..................................................................................................................81
Conclusion ...................................................................................................................82
References ..........................................................................................................................83
v
List of Tables
Table 1. Levene’s Test of Equality of Error Variance………….......................................66
Table 2. Analysis of Covariance Summary………………………………..…………….67
vi
List of Figure
Figure 1. Estimated marginal means ................................................................................ 65
Figure 2. Normal Q-Q Plot RCDVSM for VOM ............................................................. 66
Figure 3. Normal Q-Q Plot for RCDVSM BIP................................................................. 66
1
Chapter 1: Introduction to the Study
Recidivism in cases of domestic violence in the United States occurs at a higher
rate than other violent crimes, despite the use of interventions such as protection orders,
probation, incarceration, and batterer intervention programs (Frantzen, Miguel, & Kwak,
2011; Mills, Barocas, & Ariel, 2013; Richards, Jennings, Tomsich, & Gover, 2014; Sloan,
Platt, Chepke, & Blevins, 2013). Furthermore, incidents of domestic violence are higher
in age groups 18-24, which correlates with statistics representing the prevalence of crime
by age (Breiding, Basile, Smith, Black, & Mahendra, 2015; Nelson, 2013). Domestic
violence offenders enter the criminal justice system once charged with the crime of
domestic violence (Frantzen et al., 2011).
Interventions for domestic violence offenders in situations deemed appropriate by
providers include victim–offender mediation (VOM), a restorative justice intervention
(RJI; Mills et al., 2013). Another intervention used was a batterer intervention program
(BIP; Mills et al., 2013). BIPs are the most commonly used programs for domestic
violence offenders; programs are modeled after the Duluth program developed in the
1980s by Ellen Pence and Michael Paymar (Pender, 2012). The BIP model was a group-
oriented therapeutic behavior modification treatment focusing on contributing factors of
domestic violence including anger and control (Pender, 2012). BIPs have been shown to
be effective in the reduction of recidivism as compared to traditional sanctions including
arrest, probation, and incarceration (Mills et al., 2013; Pender, 2012; Sherman & Harris,
2013).
2
VOM programs differ from BIPs, as VOM programs are a form of restorative
justice mediation between the victim and offender (Daly, 2012; Dhami, 2012; Gromet,
Okimoto, Wenzel, and Darley, 2012; Weeber, 2012). Present during the mediation session
in addition to the victim, offender, and mediator was also the prosecuting attorney, and
any other appropriate court officials (Daly, 2012; Dhami, 2012; Gromet, 2012; Weeber,
2012). VOMs are initiated through the request of the victim, and thereafter a certified
mediator or court official completes an evaluation to determine if the mediation was
appropriate (Daly, 2012; Dhami, 2012; Gromet, 2012; Weeber, 2012). Some factors that
are examined to determine appropriateness are the severity of the crime and willingness
of the offender to participate in the mediation (Daly, 2012; Dhami, 2012; Gromet, 2012;
Weeber, 2012). In some cases, VOM programs are combined with other treatments and
interventions to include BIPs, probation, and incarceration (Dhami, 2012). When
examining recidivism with the use of VOMs and BIPs, both programs have been shown
to reduce reconviction of domestic violence offenders (Mills et al., 2013; Sloan et al.,
2013). Despite these findings, VOMs are not as commonly used as BIPs as an
intervention in cases of domestic violence.
In research examining restorative justice (RJ) interventions such as VOM,
researchers found in two different research studies, that possible reasons for the limited
use of VOM were limited research on VOM outcomes within the criminal justice field,
the impact of VOM on victims, and legal professionals’ preference for punitive methods
(Gavrielides, 2015; Mills et al., 2013). Concerns regarding the victim arise because the
construct of VOM typically involves a session involving the victim, the offender, a
3
criminal justice representative, and a mediator to discuss the crime and come to an
agreed-upon resolution (Daly, 2012; Dhami, 2012; Gromet, 2012; Weeber, 2012).
Despite such reservations about the use of VOM in domestic violence cases, the
intervention has proved impactful in reducing recidivism in other violent crimes
(Mongold & Edwards, 2014). VOM was most commonly used as an intervention with
nonviolent offenses and with juveniles (Mongold & Edwards, 2014). Researchers have
recognized that RJIs are not appropriate in all cases of domestic violence, and should be
considered on a case by cases basis, taking into consideration the severity of the abuse
and victim willingness to participate (Gavrielides, 2015). In this study, I compared VOM
to BIP to examine the differences of each intervention on recidivism.
In a previous study, Nelson (2013) found that there was a correlation between
committing crimes and the age of the offender, which has also been examined regarding
domestic violence crimes. In research examining recidivism in cases of domestic
violence, Nelson (2013) found that convictions for crimes of domestic violence
significantly decreases after the age of 30. The researcher, Nelson (2013) concluded that
recidivism occurs because of aging, and that research should exclude age when
examining the effectiveness of intervention. In the study by Nelson (2013), the researcher
controlled for age in the comparison of recidivism between VOM and BIP interventions.
This chapter included an introduction to the problem of recidivism in domestic
violence cases. I provided an explanation of the background of the issue, the purpose of
the study, and hypothesized outcomes of this research. In this chapter I also included
definitions of terms used throughout the study, along with limitations, assumptions, and
4
social implications of the study. A detailed literature review and theoretical framework
are presented in Chapter 2. Chapter 3 includes a discussion of the methodology used in
the completion of this quantitative quasi-experimental study. I present the discussion of
the findings in Chapter 4, and present the conclusions in Chapter 5.
Background
Researchers Herman, Rotunda, Williamson, and Vodanovich (2014) estimated that
approximately 1.3 million women and 750 thousand men are affected by domestic
violence in the United States annually (Herman, Rotunda, Williamson, & Vodanovich,
2014). In 2012, there were 51,644 charges of domestic violence in the United States—an
increase of 12.6% from 2011 (Cucinelli, 2014). Nationally, from 2003 to 2012, domestic
violence accounted for 21% of all violent crimes (Truman & Morgan, 2014). With these
high rates, interventions are needed to provide protection for victims. Domestic violence
figures have remained high despite the use of various interventions to treat offenders.
The results of studies on domestic violence interventions Mills et al. (2013),
Pender, (2012), Sloan et al. (2013) have suggested that formal sanctions (e.g., restraining
orders, probation, and incarceration) are not effective in the reduction of recidivism.
Researchers Mill, Barocas, and Ariel have shown that BIPs are more efficient than formal
sanctions in reducing recidivism. In other findings researchers suggested that RJIs may
also be more effective in reducing recidivism than formal sanctions (Mills et al., 2013). In
a study comparing circles of peace (CPs; a form of VOM) and BIPs, researchers found
that in a 6-month period, CPs were more effective in reducing recidivism in offenders
(Mills et al., 2013). However, long-term results were not significant enough to indicate
5
that CPs were more effective, and there were questions about the effectiveness of other
RJIs (Mills et al., 2013).
BIPs and VOMs are two of four commonly used interventions used in response to
cases of domestic violence offenders, in addition to anger management and individual
therapy (Daly, 2012; Dhami, 2012; Gromet, 2012; Weeber, 2012). Offender participation
in a BIP is determined through assessment, addressing the nature of the offense and other
factors to determine appropriateness (Pender, 2012). There have been reductions in
recidivism among domestic violence offenders where BIPs have been used (Mills et al.,
2013).
In this study, I compared differences in recidivism rates for first-time male
offenders in VOM versus BIPs; I addressed questions from previous studies on the
effectiveness of the interventions as measured by reduction in recidivism. I examined
data ex post facto, with a particular focus on first-time male offenders between the ages
of 18 and 30, which is deemed to be the population with the highest recidivism rates in
cases of domestic violence (Renner, Whitney, & Vasquez, 2015; Richards et al., 2014;
Sutton, Simons, Wickrama, & Futris, 2014). The factors that were addressed in the
study—VOM, and BIPs—have not previously studied together in the context of
recidivism in first-time, male, domestic violence offenders. The aim of this study was to
determine if there was a significant difference in recidivism rates for first-time male
offenders enrolled in a VOM intervention or a BIP program without participation in
VOM.
6
Problem Statement
Despite the use of interventions in cases of domestic violence, recidivism remains
a concern (Apel, 2013; Herman et al., 2014; Mills et al., 2013). Apel (2013), in a study
examining formal sanctions’ effect toward criminal deterrence, researchers indicated that
formal sanctions such as probation, incarceration, and capital punishment do not have a
statistically significant effect in the deterrence of some crimes, including violent crimes
such as domestic violence. Further, formal sanctions within a punitive model of justice
can lead to additional harm and vengeance (Wozniak, 2014). In a study by Herman et al.
(2014), researchers found that, after completion of a BIP, approximately 37.4% of the
offenders reoffended. The researchers suggested that one reason for the program’s
ineffectiveness was that the BIP addressed the criminal and legal aspects of the crime,
rather than behaviors (Herman et al., 2014). RJ programs created an opportunity for
resolution between the victim and offender, through the examination of the motivation for
the crime and appropriate interventions (Morrison & Vaandering, 2012). In comparison,
RJI programs focused on offenders’ behaviors and their impact on the victim, offender,
and community (Morrison & Vaandering, 2012).
Despite findings from research studies were researchers have found significance
in reduction of recidivism for domestic violence offenders in RJI programs, during the
research period ending in 2012, RJI programs were not chosen as interventions for
domestic violence cases by judges, and lawyers n many criminal justice agencies in cases
of domestic violence (Alarid & Montemayor, 2012). A factor contributing to decreased
use of RJI programs has been mandatory arrest laws that mandate traditional punitive
7
legal actions for domestic violence offenders (Boal & Mankowski, 2014). As a result of
mandatory arrest laws, there has been an increase in BIPs, as these programs can be
mandated and monitored by probation officers (Boal & Mankowski, 2014). Although
mandatory arrest laws provided an immediate decrease in danger for victims, many
feminists argue that victims are placed at a greater risk of danger upon the offender's
release from jail (Munjal, 2012). Despite the increase in interventions and legal sanctions
in cases of domestic violence, recidivism rates for these crimes remain amongst the
highest for violent offenses (Alarid & Montemayor, 2012, Morrison & Vaandering,
2012). The high recidivism rates not only impact victims, but also the criminal justice
system with the financial cost, medical costs, and communities (Bell, Cattaneo,
Goodman, & Dutton, 2013; Juodis, Starzomski, Porter, & Woodworth, 2014).
Mills et al. (2013) suggested that there has been limited research examining
recidivism in domestic violence cases with the use of RJIs. In their study, Mills et al.
compared recidivism for offenders enrolled in either a BIP or an RJI CP over a 24-month-
period. CP interventions were group sessions involving victims, offenders, criminal
justice officials, and family members to address the issue of violence and find a
resolution (Mills et al., 2013). The participants in Mills et al.’s study were individuals
charged with misdemeanor domestic violence offenses; some had previously been
charged with a domestic violence offense.
In addition, there has been limited research examining recidivism while
controlling for age. In a study by Nelson (2013) examining recidivism and sentencing
practices, the researcher found that age was a significant factor in the reduction of
8
recidivism. Researchers found in previous studies that crime decreases with age (Breiding
et al., 2014; Nelson, 2013). In a study examining domestic violence, the researcher
concluded similar findings, that suggest that age should be a controlled variable when
examining if interventions are correlated to the reduction of recidivism (Nelson, 2013).
In the study by Richards et al., it was found that repeat domestic violence
offenders were more likely to reoffend than those charged for the first offense (Richards
et al., 2014). Over a 10-year period, multiple-time domestic violence offenders who were
charged with domestic violence were more likely to reoffend than first-time offenders
(Richards et al., 2014). One common theme remains: RJIs can be the most effective
response in certain cases of domestic violence (Gavrielides, 2015). Researchers
examining the use of RJIs found a reduction in recidivism for violent offenders who have
committed crimes similar to domestic violence, such as assault and battery (Morrison &
Vaandering, 2012).
Although the aforementioned research regarding recidivism in cases of domestic
violence contains significant findings, I have found limited research on recidivism rates
while controlling for age in cases of first-time male domestic violence offenders who
participated in the RJI VOM programs. Thus, further research was warranted on
recidivism in cases of domestic violence for first-time male offenders enrolled in RJI
programs in comparison with BIPs. The limited amount of research in this area highlights
the need for additional studies regarding recidivism in cases of domestic violence with
the use of RJIs (Sloan et al., 2013).
9
The Purpose of the Study
The purpose of this quasi-experimental quantitative study was to examine
differences in recidivism between offenders enrolled in a BIP and offenders enrolled in a
VOM program, while controlling for age. My goal was to examine the impact of
restorative justice interventions on domestic violence in a population that has been
deemed high risk for violent crimes. By examining the differences, the outcomes could
support the use of alternative interventions in cases of domestic violence. The
participants consisted of male offenders age 18-30 years. The timeframe of the
examination was a 2-year period following completion of the VOM or BIP program
(2012 through 2013). The results of this study could provide a better understanding of
strategies to address recidivism in cases of domestic violence.
Nature of the Study
The study involved quantitative, quasi-experimental analysis of archival data. A
quantitative, quasi-experimental design was the most appropriate for the proposed study,
as this method allows for examination of differences ex post facto for variables not
randomly assigned. The use of an ANCOVA analysis made it possible to examine the
differences between the dependent variable recidivism rates for two levels of the
independent variable, VOM and BIPs, in a region where both interventions are available
and in use, while controlling for the covariate, age, the second independent variable.
Using an existing database, I could track outcomes over a period of time.
10
Research Question
RQ: What is the difference between recidivism rates for offenders who have
participated in the restorative justice intervention VOM versus those who have
participated in a BIP for first-time male domestic violence offenders 24 months
postintervention?
Hypotheses
In this study, I examined the differences in recidivism for first-time male domestic
violence offenders in VOM versus BIP programs, while controlling for age. Recidivism
data were examined at 24 months’ post-intervention. The hypotheses are as follows:
Ho: µ1=µ2: There are no differences in recidivism rates between offenders
enrolled in VOM versus BIP at 24 months’ postintervention, while controlling for age,
H1: µ1 ≠ µ2: There are differences in recidivism rates between offenders enrolled
in VOM versus BIP at 24 months’ postintervention, while controlling for age.
Theoretical Framework
In this study, I examined recidivism for first-time male domestic violence
offenders who have participated in VOM, a restorative justice intervention program, as
well as a BIP. The theory of RJ relies upon an understanding of behavior through a
motivational and socialization viewpoint (Morrison & Vaandering, 2012). The basis of
the theory was that an individual’s motivation to commit crimes was contingent on how
committed they are to society (Morrison & Vaandering, 2012). Examples of commitment
included community, family, and friends, in addition to social norms. Several theories of
RJ posit different constructs about the criminal justice system (Daly, 2013; Kenny &
11
Leonard, 2014; Morrison & Vaandering, 2012; Weeber, 2012). The commonality among
the theories is the notion that social commitment plays a role in an individual’s
motivation not to commit crimes (Morrison & Vaandering, 2012).
The practice of RJ dates back to 68000 BC a time when human societies did not
have a formal criminal justice system (Dickerson-Gilmore, 2014). In societies where the
criminal justice system is structured around RJ principles, the victim and offender
worked together to address the issue, with the intended outcome of justice for the victim
and reduction in reoffending for the offender (Dickerson-Gilmore, 2014).
RJIs can be conducted in different ways to include family sessions, community
service, and victim and offender mediation (Beck, 2015; Gavrielides, 2015;
Laxminarayan & Woldhuis, 2015). VOM, one type of RJI, involves mediation between
the victim and offender to restore power to the victim through the victim’s participation
in court proceedings, and communication with the offender on punishment for the crimes
(Laxminarayan & Woldhuis, 2015). The anticipated outcome of recidivism was that there
would be a reduction in the likelihood of reoffending, as the offender would take
accountability for actions (Gavrielides, 2015; Laxminarayan & Woldhuis, 2015). Because
RJI did not encourage formal criminal justice procedures, VOM, like other RJ
approaches, was not widely used due to negative viewpoints of the nonpunitive
components of the interventions (Gavrielides, 2015; Laxminarayan & Woldhuis, 2015).
In cases of domestic violence, reservations concerning the use of VOM have been
based on limited research on the impact of VOM on victims (Gavrielides, 2015;
Laxminarayan & Woldhuis, 2015; Mills et al., 2013). Power and control dynamics that
12
are associated with the crime of domestic violence, and could be harmful to victims, was
determined by judges and lawyers to be another concern (Gavrielides, 2015;
Laxminarayan & Woldhuis, 2015; Mills et al., 2013). Because of the power and control
dynamics in cases of domestic violence, VOM may not be appropriate, as victims may
become revictimized due to the mediation session (Gavrielides, 2015; Laxminarayan &
Woldhuis, 2015; Mills et al., 2013). Because of the nature of domestic violence crimes,
the victim must be willing to participate and the environment must be safe for all parties
involved (Laxminarayan & Woldhuis, 2015). As a result of the limited use of VOM
programs, there have not been many opportunities to perform research studies on victims
or to examine the effects of VOM on offenders (Laxminarayan & Woldhuis, 2015).
Another reservation regarding the use of the VOM approach includes the
appropriateness of the use of RJIs for more violent crimes, including domestic violence.
Researchers have taken into consideration that RJIs are not appropriate to use in all cases
of domestic violence, and interventions such as BIPs are more appropriate when the
victim does not want to participate, and there has been a lethal level of violence
(Sherman, Strang, Mayo-Wilson, Woods, & Ariel, 2015). An example of inappropriate
cases determined by previous research are cases where there are significant violent
threats, and when the victim is not comfortable with the mediation (Sherman et al., 2015).
Despite these reservations, RJ theorists have proposed that the intervention is appropriate
in cases where a designated individual screened the victim and offender for suitability for
the intervention (Gavrielides, 2015; Laxminarayan & Woldhuis, 2015; Mills et al., 2013).
Researchers found RJIs to be more cost-effective than formal sanctions, which is an
13
additional benefit of using the intervention (Sherman et al., 2015). Researchers
determined cost by the nature of the crime and cost associated with recidivism (Sherman
et al., 2015).
To further explain recidivism, the theory of reintegrative shaming, developed by
Braithwaite (1989), may be applied to describe offenders’ behaviors. Reintegrative
shaming encompasses the idea that criminals who come to feel shame and remorse, and
are held accountable for actions by the members of their community who also care for
them, are less likely to engage in criminal behaviors than individuals who have received
punishment based on the crime and have reformed (Braithwaite, 1989; Mongold &
Edmonds, 2014). Developers of VOM interventions focused on the individual accepting
the criminal behaviors while also working toward reparations for actions, which was a
principle of reintegrative shaming (Dhami, 2012). VOM interventionist used theoretical
approaches to further understand the differences in the outcome between recidivism and
VOM as compared to BIPs.
Similar to RJ, researchers have used reintegrative shaming in cases of violence
and with juveniles (Mongold & Edwards, 2014). Reintegrative shaming focuses on the
crime, the use of shame by the people who the offender deems significant, such as their
family members, and reintegration to reduce crime (Braithwaite, 1989). Reintegrative
shaming theory suggests that once a crime has occurred, the offender should be held
accountable for behaviors; which should occur in a manner where respect is exhibited
(Braithwaite, 1989; Mongold & Edmonds, 2014). The intervention should lead to the
offender feeling shame and remorse for actions to those whom the offender holds dear,
14
then the offender should be openly reintegrated back into society (Braithwaite, 1989;
Mongold & Edmonds, 2014). Reintegrative shaming is a theory that focuses on the
importance of socialization, based on the position of disapproval of the actions associated
with rituals of forgiveness (Braithwaite, 1989). Mongold and Edmonds related
reintegrative shaming to VOM, as reintegrative shaming focuses on offenders
understanding the impact of their actions and making amends to society for wrongdoing
in reintegration efforts (Braithwaite, 1989; Mongold & Edmonds, 2014). Consideration,
according to theorist should be taken as every offender cannot be reintegrated back into
society (Mongold & Edmonds, 2014).
BIP and VOM interventions both focus on intervention in the behaviors of
offenders, with the goal of reducing recidivism (Dhami, 2012; Herman et al., 2014).
Interventions influenced by reintegrative shaming focus on reduction in recidivism by
addressing how individuals are socialized in society (Braithwaite, 1989). By combining
careful family focused shaming with reintegration, Braithwaite (1989) intended to have
individuals take accountability for crimes and learn that criminal behaviors do not meet
social norms. For persons who commit crimes of domestic violence, punitive shaming
can occur through measures such as arrest and filing of charges and this may lead to
anger and distrust, not reintegration. The difference between reintegrative shaming and
punitive shaming is that reintegrative shaming leads offenders to feel remorseful and
engaged in society, while punitive shaming can lead to anger and isolation (Braithwaite,
1989).
15
RJIs, in cases of domestic violence, may be effective in reducing recidivism when
used post-conviction (Miller & Iovanni, 2013). These findings not only suggest that
victims have more time to heal when RJ occurs post-conviction but also that offenders
may develop empathy and understanding of the crime (Miller & Iovanni, 2013). The
findings further suggest that as a result of reintegrative shaming (accountability),
offenders are better able to understand that domestic violence incidents do not meet
societal norms. Reintegrative shaming theory provides concepts to address recidivism
and socialization concerns about the crime of domestic violence (Braithwaite, 1989;
Dhami, 2012; Herman et al., 2014; Miller & Iovanni, 2013).
Definition of Key Terms
Batterer intervention program (BIP): A group intervention program for domestic
violence offenders that is mandated by the courts to address behaviors associated with
domestic violence (Mills et al., 2013). Programs are 26 weeks in length and include anger
management, and self-control techniques (Mills et al., 2013).
Domestic violence: An act of violence committed by an individual against an
intimate partner or family member (Breiding et al., 2014).
Male Offenders: Male individuals who have been charged with the crime of
domestic violence against an intimate partner during the reporting period as identified in
the study (Breiding et al., 2014).
Recidivism: A repeat offense resulting in a charge of domestic violence within the
24 months’ post-intervention (Herman et al., 2014).
16
Victim–offender mediation (VOM): A restorative justice intervention between the
victim, offender, and court officials in a structured environment to address the details of
the crime and agreed-upon sanctions (Laxminarayan & Woldhuis, 2015).
Assumptions, Limitations and Delimitation
In this study, I made several assumptions. One assumption was that the
participants would be representative of the population for the geographical location. For
example, I assumed that the participants of first-time male domestic violence offenders
enrolled in VOM and BIP would be similar to other first-time male offenders enrolled in
similar programs. Second, as archival data were collected, I assumed that the data were
accurate and of high quality. High quality data collected was collected to meet the
standards set for research and the requirements for the study. Thirdly, as I conducted an
ANCOVA analysis I assumed that there was linearity, homogeneity of variance, normal
distribution, and independence (Field, 2013).
The scope of this study is the examination of recidivism for offenders who
participated in a BIP or VOM program. The offenders were male, first-time offenders, in
the State of Ohio, who had been convicted of a crime of domestic violence and were
between the ages of 18 and 30. I collected data from the municipal court in Franklin
County, Ohio, including information from 2013. The design of the study was a quasi-
experimental analysis of archival data. I did not collect the data while participants were in
a BIP or VOM program. The limitations of the study include the chosen data collection
method, which may limit factors that might further explain recidivism. In addition, I did
not collect data while offenders participate in the programs, the data were secondary data
17
compiled by the Franklin County Municipal Court in Ohio, rather than by the researcher
for this research. Franklin County Municipal Court clerks collected the data, and I did not
have control over the criteria for participants placed in the programs. I also did not have
control on the accuracy of the information that is collected.
A delimitation of the study was controlling for the age of the participants; this
study excluded the population under 18 and over the age of 30. Another delimitation was
that the study included only first-time domestic violence offenders and did not include
repeat offenders. The sex of the offenders was male, which was also a delimitation.
Significance of the Study
In the United States, domestic violence affects thousands of families each year,
and the majority of domestic violence cases go unreported (Alarid & Montemayor, 2012).
There are interventions in place with the goals of decreasing recidivism among domestic
violence offenders and promoting the rehabilitation of these offenders, but recidivism
rates remain high (Mills et al., 2013). Mills et al. (2013), in a comparison of a BIP and
RJI CPs, found no significant difference in recidivism rates between the CP and BIP.
Mills et al. concluded that the RJI was effective in reducing recidivism for domestic
violence offenders at the same rate as the BIPs. The significant differences between RJIs
and BIPs are that RJIs are focused on victims, with offenders taking accountability for
their behaviors, and participants in VOM are specifically selected with a predisposition to
success (Androff, 2012; Laxminarayan & Woldhuis, 2015). This study could be added to
the body of literature examining interventions that have been found to be significant in
18
reducing recidivism, thus leading to additional treatment options for offenders of
domestic violence.
Social Change Implications
With statistics showing that domestic violence impacts millions within the United
States and even more internationally, examining ways to reduce recidivism may be
beneficial to victims and other individuals who are affected by this type of crime. In this
study, I sought to add to the current body of literature surrounding domestic violence and
VOM. As previously stated, there was research showing the outcomes of BIPs and their
impact on recidivism rates for domestic violence offenders. There was limited research
examining the use of VOMs in cases of domestic violence. The findings of this study
could be beneficial in the implementation of interventions in cases of domestic violence.
Summary
Recidivism among domestic violence offenders remained an issue impacting the
lives of victims. In addition, offenders are not receiving effective interventions for their
behaviors. BIP continues to be the intervention most commonly used in cases of domestic
violence, despite evidence showing that RJIs are effective in the reduction of recidivism.
The purpose of this quasi-experimental study was to examine any differences in
recidivism between first-time offenders enrolled in a BIP and first-time offenders enrolled
in a VOM program to support the use of additional interventions were determined to be
appropriate in cases of domestic violence.
In Chapter 2, I presented a discussion of literature within the field addressing RJ
and domestic violence, in addition to the theoretical framework of the study. In Chapter
19
3, I addressed the methodology used for this study. Lastly, I reported results in Chapter 4,
followed by a conclusion and summary in Chapter 5.
20
Chapter 2: Literature Review
The purpose of this study was to determine whether there is a difference in the
rate of recidivism between two interventions to domestic violence. Previous researchers
have addressed the prevalence of domestic violence, types of offenders, and effectiveness
of interventions (Apel, 2013; Dutton, 2012; Mills et al., 2013). Mills et al. (2013)
suggested that additional research needs to be conducted to determine whether RJI
programs can be as effective in reducing recidivism in domestic violence offenders as
BIPs have been shown to in previous research.
This chapter contains a literature review focused on domestic violence, BIPs, and
RJI programs. The first section includes background information about domestic
violence, including statistics on prevalence and definitions. In the second section, I
reviewed international, United States, and Ohio laws associated with domestic violence.
The third section includes the interventions explored in this study: VOM (Daly, 2012;
Dhami, 2012; Gromet, 2012; Mills et al., 2013; Weeber, 2012) and BIPs (Herman et al.,
2014; Mills et al., 2013; Pender, 2012). The fourth and final section contains the
theoretical framework for the study, which involves restorative justice and reintegrative
shaming.
Research Strategy
I conducted research for this literature review through an extensive search of
scholarly research and databases. Using the Walden University Library, I searched for
peer-reviewed articles through the following databases: SocINDEX, PsycINFO,
PsycARTICLES, ProQuest Criminal Justice, EBSCO, and Sage Premier.
21
In a comprehensive search, I used the following terms to identify studies
addressing the research problem: recidivism, domestic violence, retribution, intimate
partner violence, victim-offender mediation, batterer intervention program, the first-time
offender, crime, shame, reintegrative shaming, deterrence, and rearrest. The following
sections contain discussions explaining domestic violence, recidivism, and the theoretical
framework of the study.
Domestic Violence
The definition of domestic violence has changed over time in response to changes
in cultural norms and values. Information published by the Centers for Disease Control
and Prevention (CDC) define the term domestic violence as any harm caused through
physical, sexual, or psychological abuse brought against a current or former intimate
partner (Breiding et al., 2014). Domestic violence was also classified as intimate partner
violence (IPV). Domestic violence victimization is emotional, psychological, and
physical abuse, which has costs for all stakeholders (Pow et al., 2015). Between the years
1980 and 2008, one out of five homicides was the result of domestic violence (Pow et al.,
2015).
Leaving the relationship is not always a solution. When victims attempt to leave
domestic violence offenders, they may be at greater risk of physical harm, which can
even result in death (Pow et al., 2015). With These findings researchers further supported
the need for programs that seek to reduce recidivism in offenders. Most of the domestic
violence offenders who fatally injured victims have been charged with or convicted of a
prior domestic violence crime (Pow et al., 2015).
22
Domestic violence impacts the lives of millions of individuals, with an estimated
4.8 million women being victimized annually worldwide (Sloan et al., 2013). As a result
of these figures, numerous researchers are addressing the impact of domestic violence
and interventions to reduce violence and recidivism rates. As with any crime, there are
reparations for the crime that may restore safety and security to victims and the
community. Gromet (2012) proposed, in his study that satisfaction, or the victims’
feelings toward the interventions (including crimes of domestic violence), may restore
security to the victim and assist in the reduction of recidivism. In addition to these
outcomes, there was an implication for the need of additional research examining
effective interventions for offenders in crimes such as domestic violence (Gromet, 2012).
Prevalence of Domestic Violence Incidents in the United States
Domestic violence is prevalent in the United States, with one estimate indicating
that almost 60% of married women have been abused (Price, 2013). However, domestic
violence expands beyond the victimization of married individuals. There are an estimated
1.3 million women in the United States annually who report a history of victimization,
according to the National Coalition Against Domestic Violence in 2009 (Herman et al.,
2014). Because many crimes of domestic violence go unreported, these numbers have
been assumed to underrepresent the scope of the problem. These statistics showed that
there is a need to address the domestic violence endemic within the United States, as well
as internationally.
Domestic violence encompasses many acts of violence and intimidation,
including stalking, rape, and assault, as well as psychological abuse and intimidation
23
(Breiding et al., 2014). Breiding et al. (2014), found an estimated 18.3 million women
and 6.5 million men reported stalking in the year 2011. Almost 61% of female victims of
stalking estimated that the offenders were previous intimate partners, and 41% of men
reported that the offenders were intimate partners.
Rape is also considered a crime of domestic violence when it is committed by an
intimate partner. Through a national survey, 8.85% of women and .05% of men reported
being raped by an intimate partner over their lifetimes (Breiding et al., 2014). Statistics
from the same study also indicated that 31.5% of women and 27.5% of men reported
experiencing physical violence over their lifetime (Breiding et al., 2014). While the
number of incidents of reported physical abuse between men and women in the United
States is believed to be close in percentage to each other, there are disparities in reporting
between men and women. Statistics show that women report more abuse and that men are
charged with crimes of domestic violence significantly more often than women (Breiding
et al., 2014).
Victims
Domestic violence is classified as a gender-based crime due to the number of
incidences of violence reported by women (Richards et al., 2014). Its classification as a
gender-based crime does not rule out men experiencing violence, but because women
report abuse more often, there are more reports examining the impact of abuse against
women. Domestic violence can occur regardless of gender, race, and age. In some cases,
these characteristics may play a role in victimization, whereas some characteristics have
24
been found to have a statistically insignificant relationship to the occurrence of domestic
violence (Breiding et al., 2014; Price, 2013; Richards et al., 2014).
Gender. With one in four women being abused yearly, many of the resources for
victims of domestic violence are aimed at providing support and services for women
(Herman et al., 2014). The support for women in regards to domestic violence is
documented by research and gender-based laws that provide protection that meets the
specific needs of women. As a response to the overwhelming number of women reporting
abuse relative to the number of men reporting abuse, there are shelters and other
resources that target women. Though men are victimized, women are more likely to be
physically and psychologically injured by domestic violence than men (Straus &
Gozjolko, 2014).
Although most domestic violence services are geared toward women, it is
estimated that one in seven men are victimized yearly (Herman et al., 2014). This
statistic, though it indicates that domestic violence victimization is not as prevalent
among men as among women, is still significant in validating the need for services for
men. Disparities have been found in the reporting of domestic violence by men
(McKeown, 2014). McKeown (2014) who examined information gathered for use in
studies, as well as the data collected for statistical purposes from arrest records and
intervention programs indicated that there are more men than women among domestic
violence offenders.
Race. Domestic violence can impact men and women of various races. According
to a national study conducted by Breiding et al. (2014), women of multiracial ethnicity
25
were more likely than White and Black women to be victims of domestic violence crimes
of rape and violence. Regarding stalking, American Indian/Alaskan Native women were
more likely to be stalked than women of any other race (Breiding et al., 2014). In a
national study examining domestic violence demographics in youth ages 18-27, 19.15%
of the participants reported domestic violence by an intimate partner (Renner, Whitney, &
Vasquez, 2015). In the findings of the study researchers suggested that White women and
Black men between the ages of 18 and 27 reported higher instances of domestic violence
perpetration than other demographic groups (Renner et al., 2015). Renner et al. (2015)
stated that Black men and White women who reported perpetration of domestic violence
also reported a history of child abuse. Black women and White men reported lower
percentages of perpetration, as well as lower percentages of victimization (Breeding et
al., 2014; Renner et al., 2015). Despite lower percentages for domestic violence
perpetration and victimization, Black women and White men reported higher percentages
of alcoholism, which has been linked to domestic violence perpetration (Lipsky et al.,
2014; Renner et al., 2015).
Age. In a national study by Breiding et al. (2014), victims of domestic violence
were more prevalent in the 18-24 age group. The decrease in victimization directly
correlated with an increase in age. Researchers found that only 4.7% of victims reported
violence over the age of 45 (Breiding et al., 2014). Researchers illustrated that the
findings from this study that victimization decreases as age increases, which correlates
with patterns exhibited by offenders in crimes. Researchers found that as individual’s age,
they are less likely to commit crimes (Breiding et al., 2014).
26
Offenders
The characteristics of individuals can place them at higher risk of becoming an
offender or an abuser. Juodis, Starzomski, Porter, and Woodworth (2014) found that the
characteristics of offenders show trends related to age, race, and psychological and social
factors. When combined, these factors can impact the likelihood of recidivism and the
most appropriate type of intervention. These factors may also have an impact on
intervention programs that are designated for perpetrators of domestic violence.
Age. As with victims, the prevalence of offending decreases as age increases
(Nelson, 2013). There are differences found between women and men offenders that have
an impact on recidivism rates. For women who commit crimes of domestic violence,
evaluation of arrest records has shown that recidivism for crimes of domestic violence
decreases between the ages of 18 and 24 and further decreases in the late 40s (Nelson,
2013). Findings show that as women age, they are less likely to commit crimes of
domestic violence in addition to other violent crimes.
The statistics for men differ regarding the relationship between age and offending.
Between the ages of 18 and 28, there seems to be a slight decrease in recidivism for
women, but it is not significant enough to indicate a correlation between age and reduced
recidivism (Nelson, 2013). The researcher found in contrast to women, the rate of
recidivism in cases of domestic violence has been found to decrease for men over the age
of 28 (Nelson, 2013). Nelson (2013) reported that recidivism regarding domestic violence
for men decreases, in spite of recidivism for other violent crimes increasing for the male
27
population. Based on the findings from previous research, domestic violence offending is
highly correlated with age.
Race. There have been varied results in research concerning the race of domestic
violence offenders. Many researchers have determined that race is not a significant factor
in regard to domestic violence offending (Juodis et al., 2014; Straus & Gozjolko, 2014).
There are disparities found for race, when there is an addition of other factors such as
substance abuse and mental health issues, researchers found to have linkage to violence.
Gender. Though women are more likely to report abuse, in studies researchers
have found that men and women offend in equal amounts (Straus & Gozjolko, 2014).
Straus and Gozjolko (2014) indicated that while men and women offend at similar rates,
men are more likely to cause more physical harm than women. When there is physical
harm resulting in medical or legal intervention, there is more likely to be an arrest
(Bradley, 2015). When individuals are arrested, the statistics are gathered that show that
men are offending more than women. According to Straus and Gozjolko, however, men
who are convicted of crimes of domestic violence are not the only population that is in
need of interventions. There are disparities in what individuals consider to be domestic
violence. In regard to domestic violence, it is more acceptable for women to smack or hit
men than it is for men to strike women (Bradley, 2015). Bradley (2015) found that
despite the acceptance of this type of behavior, these actions are still considered domestic
violence.
28
Mental Health and Substance Abuse
Researchers have shown that there is a significant correlation between a history of
mental illness and substance abuse and perpetration of domestic violence (e.g., Juodis et
al., 2014; Lipsky et al., 2014; McKeown, 2014). Individuals who display symptoms of
psychopathy such as lack of remorse, guilt, and manipulative behaviors have been found
to be more physically aggressive and to have higher recidivism rates in cases of domestic
violence (Juodis et al., 2014). Women who had a history of mental illness or
posttraumatic stress as a result of the previous victimization were more likely to be
domestic violence offenders than women who did not have the same history (McKeown,
2014).
Substance abuse is also a contributing factor in domestic violence (Lipsky et al.,
2014). The researchers reported that binge drinking in Black and White women has a
direct correlation with domestic violence toward an intimate partner (Lipsky et al., 2014).
Furthermore, for Black men and women who have a history of familial violence in
childhood in combination with alcohol abuse, there is an increased likelihood of
victimizing an intimate partner (Lipsky et al., 2014).
Lipsky et al., (2014) found correlation between alcohol use and domestic violence
increased as the number of drinks a person consumes increases (Lipsky et al., 2014). The
use of other substances including cocaine and opiates have also been linked to domestic
violence (Lipsky et al., 2014). In populations of offenders, the researchers found that a
total of 64.9% of domestic violence offenders report drug use, and 75.7% specifically
report alcohol use (Juodis et al., 2014).
29
Domestic Violence Laws
International
Violence against women is a global issue that has been determined to be a human
rights violation by the Inter-American Court of Human Rights and the European Court of
Human Rights (Ramji- Nogales, 2014). In 1999, the United Nations declared that
domestic violence is an act of violence, and this declaration posed many changes for
various nations (Ramji- Nogales, 2014). One of the main challenges has been that
regarding domestic violence as a gender-based crime violates the norms of some cultures
(Qureshi, 2013). In some European cultures, for instance, submissiveness of women to
men is a norm, and recognition of domestic violence as a human rights issue threatens
cultural practices (Ramji- Nogales, 2014). These cultural issues notwithstanding,
domestic violence according to human rights advocates are considered a violation of
rights and many nations have made great strides since the initial recommendation for
classification of domestic violence as a human rights violation as presented by the United
Nations in 1999 (Qureshi, 2013).
Despite the classification of domestic violence as a human rights violation, some
countries have reported difficulties in the enforcement of anti-domestic-violence laws
(Qureshi, 2013). Difficulties have been presented since the initial recommendation in
1975, before becoming an official declaration by the United Nations (Qureshi, 2013).
Because of the challenges in defining domestic violence and agreement among nations as
a human rights violation, domestic violence was not recognized as a human rights
violation internationally until almost 21st century (Qureshi, 2013). Some of the concerns
30
presented at the United Nations Conferences were the language used in the definition of
domestic violence in the Human Rights Law, which has provided loopholes for countries
in developing laws that impact enforcement (Chaban, 2014).
The U.S. Federal Government. In 1994, The Violence Against Women Act
(VAWA) was passed, providing protection under the law for women impacted by
domestic violence (Weissmann, 2013). The VAWA creators recognized and provided
gender-based support for female victims of abuse at the hands of significant others (U.S.
Department of Justice [USDOJ], 2015). VAWA was reauthorized in 2013, signifying that
domestic violence was still a widespread issue within the United States (Weissmann,
2013). A component of the VAWA Lautenberg Amendment 18 (U.S. Cons. Art. CMXXII)
also impacts gun control for individuals convicted of domestic violence crimes (Price,
2014). According to the USDOJ (2015), domestic violence offenders cannot be in
possession of or obtain weapons in some cases. The possession of weapons includes the
shipping and handling of weapons and ammunition (USDOJ, 2015).
There are concerns with the federal laws regarding victims and offenders. For
instance, there is concern that limitations on gun possession for offenders will impact
judges’ decisions regarding whether to convict individuals of domestic violence crimes
(Price, 2014). The concerns surround judges handling of domestic violence cases, who
may view the loss of gun rights as not appropriate for the crime. These concerns arose
despite statistics showing that a victim of domestic violence is at greater risk of physical
harm, including harm inflicted through the use of a weapon, after the offender is
convicted of a crime (Sherman & Harris, 2013). Another concern relates to the possibility
31
of violating the rights of the victim if the victim does not want the offender to be
convicted or depends on the offender as a primary provider (Sherman & Harris, 2013).
With these concerns, judges and police officers continue to have discretion in the arrest
and conviction (Price, 2014).
The State of Ohio. According to Ohio law, domestic violence is defined as an act
of violence committed or threats to commit violence against a family or household
member (Ohio Government, 2010). Acts of violence in this context included sexual
assault, coercion, forceful detention, or any criminal activity with the intent to cause
bodily harm to a family or household member (Ohio Government, 2010). In the State of
Ohio, there are laws in place to provide protection specifically for victims of domestic
violence. Protection provided under Ohio law (Ohio Revised Codes §2919.25, §2919.26)
includes emergency protection orders which can be tracked nationwide (Ohio
Government, 2010).
Federal law is enforceable by individual states, which means that federal
mandates are applied in addition to a state's domestic violence laws (USDOJ, 2015). As a
result of federal laws being applicable throughout the United States, offenders are held to
the same sanctions throughout the country that were imposed in the original state of the
offense (USDOJ, 2015). As a result of such charges, many offenders are unable to obtain
employment in certain career fields.
Mandatory arrest laws are also applicable within the State of Ohio. Mandatory
arrest laws require that police officers arrest the violators of restraining orders in crimes
of domestic violence (Ohio Government, 2010). In the State of Ohio, offenders can be
32
arrested if they violate protection order as long as the officer believes that they have been
the primary aggressor (Ohio Government, 2010). Mandatory arrest laws in Ohio are
similar to policing agencies in the 21 other States with Mandatory arrest laws as officers
may exercise their discretion, after hearing about the incident from both parties, to
determine who should be arrested even if a protective order has not been granted to the
victim (Ohio Government, 2010). In some cases, if the officer cannot make a decision on
who was the primary aggressor, other parties can be arrested (domestic violence, 2016).
Mandatory arrest has posed some issues regarding advocacy for victims of domestic
violence. One of the concerns is the removal of victims’ rights (Sherman & Harris, 2013).
Mandatory arrest laws may impact the willingness of victims to call emergency
responders for assistance and create increased violence for victims (Price, 2014).
Recidivism
One of the problems in cases of domestic violence, as identified through previous
research, is recidivism (Gavrielides, 2015; Laxminarayan & Woldhuis, 2015; Mills et al.,
2013). In a quantitative study by Sloan et al. (2013) the researchers indicated that current
penalties are not effective in the reduction of recidivism. Through the use of arrest data
collected in North Carolina from 2007, the authors were able to ascertain whether
criminal sanctions were effective in reducing recidivism in domestic violence offenders
(Sloan et al., 2013). Sloan et al. suggested that current sanctions were not effective and
supported the need for further research on interventions for first-time domestic violence
offenders. The research by Nelson (2013) further supported the need for additional
research on domestic violence, in concluding that age is highly correlated with domestic
33
violence, and should be controlled when examining the impact of interventions on
recidivism. In addition, the researchers indicated the need for additional interventions
for offenders because current penalties for crimes of domestic violence have not had a
statistically significant impact in reducing recidivism.
In a quantitative study by Mills et al. (2013), they suggested that restorative
justice interventions are effective in the reduction of recidivism in offenders. The
participants in the study were offenders in Arizona who had been ordered to a RJI, or a
batter’s intervention program. Evidence from follow-up research at 6, 12, 18, and 24
months which showed that offenders had reduced recidivism rates as much, and at the 6-
month period more than offenders in BIPs (Mills et al., 2013). These findings by Mills et
al., (2013) (the limitations include small sample size, locale, and demographics of
offenders) supported claims that despite such programs’ impact on recidivism, there is a
need for research on RJI use in cases of domestic violence when examining recidivism
(Mills et al., 2013).
Researchers examined recidivism in cases of domestic violence offenders and
supported claim that there is a need for additional interventions to support reductions in
recidivism in domestic violence offenders (e.g., Frantzen et al., 2011; Mills et al., 2013;
Sloan et al., 2013). Frantzen et al. (2011) investigated offenders whose cases of domestic
violence were dismissed over a 2-year-period and found that prosecution impacts
recidivism. The handling of domestic violence cases through the criminal justice system,
and stigmatization offenders may receive by society was correlated with offender
recidivism (Frantzen et al., 2011). In order to reduce recidivism in domestic violence
34
offenders, there needs to be additional support from judges and prosecutors in mandating
participation in intervention programs (Frantzen et al., 2011).
RJI and Recidivism Demographics
Demographic characteristics play a role in recidivism. Younger, White men are
more likely to recidivate than men of other races in the same age group (Richards et al.,
2014). The likelihood of recidivism increases if an individual has a history of violent
crime offenses. Offenders who have already been convicted of a crime of domestic
violence according to Richards et al., were more likely to recidivate than individuals who
are first-time offenders (Richards et al., 2014).
Substance abuse is another factor in recidivism; individuals with a history of
substance abuse in combination with a history of violence, according to research have
been found more likely to recidivate than other offenders (Lipsky et al., 2014). The
relationship between the victim and offender also played a role in recidivism. In a
previous study, researchers found that offenders who are married to victims are less likely
to recidivate than nonmarried individuals (Richards et al., 2014). Subsequently, an
offender who shared a child with the victim but does not reside in the same home is at
higher risk of recidivism , as is an offender who was separated from the victim , than an
offender who cohabitates with,is married to,or is divorced from,the victim (Richards et
al., 2014).
Interventions
Domestic violence recidivism rates rank amongst the highest in violent crimes
(Herman et al., 2014). Historically interventions in case of domestic violence according
35
to study findings focused on punitive sanctions to include incarceration (Herman et al.,
2014). Based on the previous studies where researchers focused on recidivism, additional
research needs to be conducted on interventions to determine the effectiveness of
reducing recidivism in a case of domestic violence (Mills et al., 2013). Policing agencies
also need to be included in research as one of the reasons for the lack of use of RJI being
the perceived lack of support by prosecutors and judges, which impact perceived
legitimacy of intervention (Lee, Zhang, & Hoover, 2012).
Interventions are used in the United States, and internationally in cases of
domestic violence. A qualitative study by Boonsit, Piemyat, and Claassen (2012)
examined how cultural and international laws impact the implementation of interventions
in cases of domestic violence in Thailand. In the implications of the study, the
researchers suggested that there needs to be an additional evaluation of interventions to
determine the most efficient in the reduction of recidivism (Boonsit et al., 2012).
In a quantitative study by Gromet et al., (2012) the authors examined victim
satisfaction with RJI programs. The studies researchers examined victim’s satisfaction
with a level of RJI programs, with findings leading one to sugges that victims input in
cases where RJI programs are used, can be impactful on judges sentencing decisions and
address other areas of concern within the criminal justice system to include overcrowded
prisons in the United States (Gromet et al., 2012). Victim participation and previous
interventions research provided support for the use of RJI programs in cases of domestic
violence.
36
Schools of thought
In addition to the programs mentioned above, there are other schools of thought
to the perpetration of domestic violence. Abuse during childhood and adolescences have
been linked to domestic violence perpetration (Lohman, Neppl, Senia, & Schofield, 2013;
O’Leary, Tintle, & Bromet, 2014). It is believed that individuals can socially learn
behaviors, which can be predictive of their future behaviors, as theorized through the
social learning theory (Williams & McShane, 2004). In cases of domestic violence, in
previous studies researchers have found high correlations between domestic violence
perpetration and experience of childhood victimization (Lohman et al., 2013; O’Leary et
al., 2014).
Social Learning theorist proposed that domestic abuse is cyclical and
intergenerational (Lohman et al., 2013). Researchers have also found that abuse in
childhood has been linked to dating violence within traditional college age students 18-
24, which is the age bracket of a high incidence of domestic violence (Sutton et al.,
2014). In order to break the intergenerational cycle of abuse, there would need to be an
identification of children who are being abused physically or psychologically (O’Leary et
al., 2014). Many of these children are diagnosed in adolescence with substance abuse
disorders, and antisocial personality disorders, which has been linked to interpersonal
problems associated with domestic violence (Lawson & Brossart, 2013; Lohman et al.,
2013). According to O’Leary et al. (2014), abused children cases go unreported, making
early interventions more difficult. When problems are identified, approaches such as
individual therapy, and family therapy are intervention methods that could be used to
37
reduce future risk of domestic violence perpetration (Connors, Mills, & Gray, 2013).
Another approach was identifying offenders through assessments such as the Risk Need
Responsivity, which could provide evaluation and suggestions for interventions for
aggressive behaviors (Connors et al., 2013).
Though there is evidence to the social influence of domestic violence, there are
also schools of thought that suggest that propensity to violence may be a result of
development factors (O’Leary et al., 2014). The antisocial behaviors and substance abuse
diagnosis are identified as ill coping factors by theorist developed by children who
experience family violence (O’Leary et al., 2014). Substance abuse has been linked to
domestic violence and the earlier the onset the higher the risk for aggressive behaviors
(Juodis et al., 2014; Lipsky et al., 2014; O’Leary et al., 2014). Researchers found that a
history of mental health diagnosis such as antisocial personality disorder according to
findings from previous studies to be a contributing factor to violence in some offenders as
individuals with this diagnosis lack connection to others and remorse (Juodis et al.,2014;
Lipsky et al., 2014; O’Leary et al., 2014). Researchers have found that characteristics
such as lack of remorse and lack of social connectedness can impact treatment of
domestic violence offenders (Juodis et al., 2014; Lipsky et al., 2014; O’Leary et al.,
2014). The incidence of domestic violence, with the above factors according to previous
research findings, researchers suggested domestic violence prevalence is heighten when
considering the age of the offender. The prevalence of domestic violence has been found
to be highly correlating with age, and decreases at the age of 30 (Nelson, 2013). Age is
38
an additional factor which according to study findings impacted the prevalence of
domestic violence cases, and recidivism (Breiding et al., 2014; Nelson, 2013).
Anger management. Anger management is an intervention used with the goals
of prevention and reduction of recidivism in offenders. According to previous research
findings, an offender’s inability to effectively manage emotions contributing factors to
domestic violence (Shorey, Seavey, Quinn, & Cornelius, 2014; Whiting, Parker, &
Houghtaling, 2014). By providing interventions to address anger management, there can
be a reduction in domestic violence and recidivism rates. To determine the type of anger
management, program an offender may need they are assessed with scales, an example is
the Anger Management Scale (AMS) developed by Stith & Hamby in 2002 (Shorey et al.,
2014). Through assessment of college students with the AMS researchers found that
students who had high scores; indicated difficulty managing anger; were more likely to
perpetrate domestic violence on their partners (Shorey et al., 2014). In another study that
examined male perpetrators perspective on domestic violence, the male offenders
reported that anger problems, and inability to manage emotions as contributing factors to
domestic violence (Whiting et al., 2014). Anger management programs can be beneficial
as an early intervention, and reduction in recidivism tool.
Victim Offender Mediation. Despite research on RJIs in the use of cases of
violence crimes, there is limited research addressing its use in crimes of domestic
violence (Dhami, 2012; Mills et al., 2013). Many of the concerns with the use of VOM in
the use of cases of domestic violence is the possible revictimization of victims having to
address the issue of domestic violence (Mills et al., 2013). The use of VOM focuses on
39
the crime being between the victim and offender (Dhami, 2012). Though, there is a
mediator present, much of VOM happens between the offender and victim, with the
anticipated results that the offender takes accountability, and makes amends for actions
against the victim (Dhami, 2012). There is also anticipated outcome that the offender and
victim, in conjunction with the mediator and lawyer, agree upon a form of punishment
(Daly, 2012; Dhami, 2012; Gromet, 2012; Weeber 2012).
The use of VOM can assist the offender in understanding the offense, the impact
the crime has on stakeholders to include the victim and the community and to offender to
take accountability (Mongold & Edwards, 2014). One must also take into consideration
that VOM is not appropriate for every crime, and participation should be determined on a
case by case basis (Dhami, 2012). Moreover, with the use of VOM victims and
offenders can both tell their side, and come to a mutually agreed upon resolution for the
crimes (Mongold & Edwards, 2014). When RJI programs such as VOM are used, there
are additional benefits outside of the outcomes for victims and offenders. The first
benefit is the reduction of recidivism for nonviolent and violent offenders (Mongold &
Edwards, 2014). In cases of domestic violence researchers found to be a beneficial
factor as recidivism in domestic violence cases ranks amongst the highest within the
criminal justice system (Mills et al., 2013). The second benefit researchers found was that
there is decreased court cost as the cases are handled outside of the court system, saving
taxpayer dollars (Mongold & Edwards, 2014). Victims could also benefit from the use
of VOM. Typically, in cases of domestic violence with mandatory arrest laws, and
prosecution victims do not have a voice in decisions made which may impact their lives
40
(Mongold & Edwards, 2014). In cases where VOM are used, victims reported increased
satisfaction with the judicial system (Choi et al., 2012). The use of VOM could also be a
tool to increase victim participation in cases of domestic violence where victims are less
likely to be cooperative with the criminal justice process, than victims of other offenses
(Choi et al., 2012).
Batterer Intervention Program. According to Herman et al., (2014) BIPs were
designed as a response to court- mandated treatment programs for batterers. The most
popular BIP is the Duluth model developed in 1980 as a court mandated service by the
Domestic Abuse Intervention Program in Duluth, Minnesota (Pender, 2012). According
to a national survey over 53% of BIPs in the United States used the Duluth Model. The
difference between a VOM and a BIP is that BIPs are a behavior modification program
(Pender, 2012). The offender is placed in a program with other offenders ranging up to 52
weeks in which behaviors are addressed, often using the Cycle of Abuse wheel as a tool
(Herman et al., 2004). The Cycle of Abuse tool is used by program administrator to
explain to offenders why abuse and ways that offenders can prevent re-offending
(Herman et al., 2014). As with VOM, participants in BIPs are selected based on criteria
that includes nature of offense, repeat offending, and additional criteria as determined by
administrating programs (Mills et al., 2013).
According to findings from previous research there are some concerns with BIPs
effectiveness in reducing recidivism in domestic violence offenders (Herman et al., 2014;
Mills et al., 2013; Pender, 2012). Some of the concerns researchers found are that BIPs
main focus is to be used in conjunction with sanctions of the court, thus the programs
41
become legal based rather than treatment focused (Herman, et al., 2014). Implications of
the study by Herman et al. (2014) that examined recidivism in domestic violence
offenders over a nine-year follow-up it was found that the program was not significant in
reducing recidivism in domestic violence offenders. Subsequently, Pender (2012) found
that the lack of uniformity in the training of facilitators and programs is a contributing
factor to the lack of significance in the reduction of recidivism. BIP’s are the most
commonly used form of treatment for domestic violence offender, though it has been
found to reduce recidivism, differences in program administration, and legal basis of
some programs highlight that this may not be the most appropriate intervention in all
cases of domestic violence.
Theoretical Framework
The theoretical framework for this study is based on two theories; Restorative
Justice Theory and the Theory of Reintegrative Shaming (Braithwaite, 1989; Daly, 2014;
Kenny & Leonard, 2014; Mongold & Edwards, 2014; Morrison & Vaandering, 2012;
Weeber, 2012). Each theory provides the possible explanation for offender behaviors and
offers behavioral interventions that can be effective in the prevention of recidivism.
Restorative Justice
The theory of Restorative Justice (RJ) theorist explained behavior through a
motivational and socialization viewpoint (Daly, 2013; Kenny & Leonard; Morrison &
Vaandering, 2012; Weeber, 2012). The basis of the theory is that an individual’s
motivation to commit crimes is dependent on social commitment (Morrison &
Vaandering, 2012). The premise behind RJ is that if a person feels connected to a
42
community, they will be less likely to commit crimes, thus reducing recidivism rates with
the use of interventions (Weeber, 2012). The theory of RJ has been around for centuries
dating back to aboriginal tribes around 68000 BC (Kenny & Leonard, 2014). According
to previous research, researchers stated, historically, there has been accounts of Indian
tribes using RJ as a means of maintaining order and justice within tribes (Kenny &
Leonard, 2014). In a study conducted by Dickerson-Gilmore (2014), it was found that RJ
is still used in the communities, and there is reported greater satisfaction for victims, than
for traditional sanctions to include, probation, and incarceration. The idea of RJ was
popularized in the American criminal justice system in the 1960’s and 1970’s by radicals
against the use of formal sanctions to include incarceration and arrest (Kenny & Leonard,
2014). For RJ to be effective, the offender must be willing to participate in addressing
the harm caused by the crime (Weeber, 2012). Albert Eglash was one of the first
American theorists in the 20
th
century to re-coin the term RJ as a term to describe how
offenders can provide restitution to victims within the American Criminal Justice system
(Daly 2013).
Since the integration into the American criminal justice system of RJ by Eglash;
Randy Barnett, Howard Zehr and Nils Christie; have each added to the term creating new
practices for the use of RJ theory (Daly, 2013). Barnett proposed that RJ is a crime
between the victim and the offender and that reparation should be paid to the victim, and
not to the court or State (Daly, 2013). Zehr focused on the relationship, in addition to the
legal aspect, that is a proposed agreement to meet judicial and victim request (Daly,
2013). Despite the proposal by Zehr, in a study that examined satisfaction with the
43
combination of RJ and the judicial system it was found that individuals in an aboriginal
community reported less satisfaction than RJ as described by Barnett (Dickerson-
Gilmore, 2014). The theorist, Christie, focused on the refraining from the use of
punishment in RJ cases to concentrate on the offender making amends for actions (Daly,
2013).
Despite the varying theorist who have developed the RJ theory, there still lacks a
clear definition of what RJ is, and how it should be implemented in the criminal justice
system (Kenny & Leonard, 2014). Based on findings in previous research studies,
researchers stated that there are varying ways that criminal justice systems throughout the
world are implementing RJ in cases of domestic violence (Dickerson-Gilmore, 2014;
Gavrielides, 2015; Miller & Iovanni, 2013). According to Miller and Iovanni (2013), they
have found that placing the RJ process after completion of sanctions by the offender;
there gives times victim more time between the domestic violence incident and RJI to
“heal”. This practice contrasted with how some criminal justice systems use RJ, despite
the differences in implementation there is still reported satisfaction by victims (Miller &
Iovanni, 2013).
In cases of domestic violence RJ, according to researchers is used more
frequently within the last 20 years, but there is still research that needs to be conducted to
address the impact the use of the interventions has on victims. In a study by Gavrielides
(2015) that examined the use of RJ in cases of domestic violence highlights the limited
research examining the outcome for victims of domestic violence. Additionally, in a
study by Dickerson-Gilmore (2014) that examined outcomes for Aboriginal families, it
44
was found that there was limited research examining the outcomes for families when RJ
is used in conjunction with traditional criminal justice practices to include incarceration.
RJ is more widely used in juvenile justice case, and cases of petty crimes such as theft
(Weeber, 2012). According to Mills, Barocas, and Ariel (2013) the use of RJI could be
beneficial to giving power back to victims, and reducing recidivism in domestic violence
offenders. Despite the benefits, researchers questioned the safety and security of victims
both mentally and physically remains an issue (Dickerson-Gilmore, 2014; Gavrielides,
2015; Miller & Iovanni, 2013).
Reintegrative Shaming
The theory of Reintegrative Shaming was first coined by John Braithwaite in
1989 (Braithwaite, 1989). Braithwaite approached punishment through the aspect of
addressing the offender’s behaviors rather than the offender (Braithwaite, 1989).
According to Braithwaite the intended outcome is to reduce recidivism through
reestablishing ties to the community, and preventing crime through leading offenders to
feel shame for actions (Braithwaite, 1989). Reintegrative shaming is a combination of
labeling, subculture, control, learning, and opportunities theories combined, to explain
why crime occurs and how to prevent crime (Braithwaite, 1989). To better explain the
theories Braithwaite went on to further explain crime, shame, and reintegration and how
these terms relate to crime, and can the understanding can be used as tools to reduce
recidivism.
Crime. According to Braithwaite (1989), individuals are motivated to commit
crime through the need to fulfill selfish motivations, and what he refers to as hedonistic
45
pleasures. The motivation for the crime is says Braithwaite, is taken from the control
theory in which individuals who do not possess restraints whether external or internal,
they will be more apt to commit crimes (Braithwaite, 1989). Thus, these individuals will
have higher recidivism rates because of the lack of control in behaviors. Male
perpetrators of domestic violence have reported that inability to control situations,
emotions and anger as factors contributing to domestic violence (Whiting et al., 2014).
According to Braithwaite (1989), the individuals who fit into this category of offenders
will be adolescent to young adult men ages 15-25, who do not have social ties to include
marriage, children, and employment. These statistical findings were directly correlated
by researchers with male perpetrators of violence who have higher instances of
perpetration between the ages of 18-24, and non-married (Juodis et al., 2014; Nelson,
2013). Researcher also found that women between the same age group and those who are
married or have children are less likely to commit crimes than men (Braithwaite, 1989).
Shame. The use of shame according to Braithwaite (1989) was necessary to
prevent individuals from committing crimes, and reduce recidivism for offenders. For
shaming to be effective, there must be social norms in place, by which the community
would look down upon individuals for violating these norms (Mongold & Edwards,
2014). Braithwaite (1989) identified two types of shaming; stigmatizing and
reintegration. Stigmatizing shaming is a type of shaming in which the offender is
shunned from society for behaviors (Mongold & Edwards, 2014). This kind of shaming
can cause individuals to isolate themselves from society, making them more apt to break
social norms leading to criminal behaviors or recidivism (Braithwaite, 1989). According
46
to Haas (2012) historically stigmatization was the norm within society for examples the
use of the Scarlett letter to let others known that a woman engaged in adulterous
behaviors. In cases of domestic violence to decrease violence, there may be a need to
break social norms. In a study where researchers examined social norms found that some
cases of domestic violence were not reported as they were seen as norm social behaviors
(Witte & Mulla, 2013). The researchers also highlighted the impact of socialization and
acceptance on behaviors. There can be many contributing factors to the normalcy of
domestically abusive behaviors to include a history of abuse, and experience of abuse in
childhood according to findings from previous research (De Grace & Clark, 2012;
O’Leary et al., 2014). Within the current criminal justice systems, the stigmatizations,
reported by offenders are not used as often as a means of punishment. Though some
would argue that the use of Sex offender registry can be considered a form of
stigmatization (Braithwaite, 1989).
Reintegration. The solution to the harmful use of stigmatization shaming
according to Mongold and Edwards (2014) would be the implementation of reintegration
shaming in society. Reintegrative shaming is the idea that the offenders gain attachments
to community and have the ability through the connections to the society to understand
the harm that the criminal offense has caused (Mongold & Edwards, 2014). When the
offenders have an understanding of the offense before commission of the crime,
Braithwaite (1989) believed that the offender will be less likely to commit crimes. After
the offense, reintegration according to Braithwaite could reduce recidivism by facilitating
an environment for offenders taking responsibility for actions, through being held
47
accountable by individuals the offenders hold in regards (Braithwaite, 1989). In cases of
domestic violence, if offenders understand the harm that the offense has caused to the
victim and individuals who they hold dear, this can lead to a change in the behavior and a
need to be reintegrated into society (Braithwaite, 1989). RJIs such as Community
Conferences and Sentencing Circles according to researchers have been useful in the
reintegration of offenders into society (Mongold, & Edwards, 2014). These VOM
program administrators engaged the victim, offender, and other individuals who the
offender may hold in regard in a reconciliation process (Mongold, & Edwards, 2014). A
premise of reintegrative shaming is the offender has to have a relationship with the victim
and other impacted parties, where the offender holds the individuals in esteem, and will
feel remorseful for actions to these individuals for shaming to work (Braithwaite, 1989).
Through the use of RJ and reintegrative shaming concepts recidivism in domestic
violence according to researchers should decrease. Both of the theories authors
highlighted the need for the offender to have accountability for actions, while meeting the
needs of the criminal justice system, and needs of the victim (Braithwaite,1989; Daly,
2014; Kenny & Leonard,2014; Mongold & Edwards, 2014; Morrison & Vaandering,
2012; Weeber, 2012). With recidivism rates in domestic violence remaining high, there is
a need for changes in the handling of these cases. The RJ and reintegrative shaming
authors proposed addressing the needs of the offender, as well as the victim while still
meeting the needs of the community and criminal justice system. Despite the different
implementations of RJ in the criminal justice systems, researchers found that there are
48
perceived benefits in reduction of recidivism (Dickerson-Gilmore, 2014; Gavrielides,
2015; Miller & Iovanni, 2013).
Summary
The issue of domestic violence continues to be an epidemic despite the use of
interventions over the past few decades. Over the past few years, according to
researcher’s strides had been made in recognizing that domestic violence has impacted
the lives of women globally. Recidivism in domestic violence offenders remain high, and
many factors influenced the treatment of offenders, to include age of the offender.
Through the development of the literature review I concluded that RJ is not widely used
in the United States. The reasons according to researchers included lack of formalized
structure for implementation, and understanding of the impact RJ has on victims. The
use of formal criminal justice sanctions to include mandatory arrest laws, probation,
incarceration and enrollment in BIPs, according to these researchers has impacted the use
of RJ in cases of domestic violence. BIPs based on findings from previous studies
remained the most commonly used intervention, and there is evidence to support that
BIPs are effective in reducing recidivism.
Researchers concluded that there was a need for intervention in cases of domestic
violence to reduce the number of offenses that occur globally and provide protection for
victims. Despite the support for RJ, there were still reservations about the use of the
intervention. In the literature review I highlighted that lack of studies addressing the
impact the use of RJ interventions may have on victims. Despite research findings, in
49
which researchers have shown that RJ was effective in reducing recidivism in crimes,
there was limited research examining the effects RJ has on crimes of domestic violence.
The varying theories of RJ theorist offer a foundation of how to implement the
program in criminal justice systems. In combination with reintegrative shaming
practices, according to researchers, that have impacted offenders socialized behaviors,
and beliefs, recidivism in cases of domestic violence can be reduced. This research study
examined differences in recidivism in the first-time male offenders who are either
enrolled in a VOM or a BIP. From the previous research study findings, researchers have
supported the notion that both interventions have an impact on recidivism in cases of
domestic violence. RJ programs according to researchers focused on remorse, offender
accountability, and assisting offenders with reestablishing ties with the community to
reduce recidivism (Laxminarayan & Woldhuis, 2015). RJIs according to research
findings has been proven to be effective in the reduction of recidivism for juveniles and
has been a method used for centuries in Indian and Aboriginal communities to restore
peace (Daly, 2014; Dickerson-Gilmore, 2014; Gavrielides, 2015; Kenny & Leonard,
2014; Miller & Iovanni, 2013). Since RJ theory developers focused on offenders taking
accountability for their actions, and the documented findings from studies where
researchers examined effectiveness regarding recidivism, RJ could be an alternative
intervention in cases of BIP where deemed appropriate. The following chapter provides a
description of the study’s methods.
50
Chapter 3: Research Method
The problem that I addressed in this study was that, despite the use of
interventions for first-time offenders in cases of domestic violence, recidivism has not
significantly decreased. As a result, offenders are more likely to reoffend, placing their
significant others and communities at higher risk for victimization. The purpose of this
quasi-experimental study was to examine the differences in recidivism of offenders who
were enrolled in BIP and VOM programs while controlling for age. Through an
examination of the differences, the outcomes can support the use of alternative
interventions in cases of domestic violence.
In this chapter, I identified the research design and rationale for the study,
variables, hypothesis, data analysis plan, and the population and participants used to
conduct the study. I provided an explanation of ethical considerations, and threats to the
validity of the study.
Procedures
I compiled the data for this study from the municipal courts in the State of Ohio.
The information provided by court statisticians collected at the time of arrest and charge
of the crime of domestic violence. The information I collected included the names, date
of birth, gender, race, and locale in which the crime occurred. The information provided
by court statisticians was updated to reflect the disposition of the case to include charges
or dismissal of the case. In determining the appropriate treatment recommendation,
Judges in Ohio ordered screening for offenders, and made referrals based on
appropriateness in congruence with legislation (Taft & Bradley, 2003). The lethality
51
assessment is ordered by the court to assess domestic violence and the level of treatment
(Taft & Bradley, 2003). The lethality assessment by researchers was assessed for validity
in three independent studies, researchers suggested that the tool predicted reassault
(Roehl, O’Sullivan, Webster, & Campbell, 2005). Reliability was assessed through
comparing the finds to that of 15 other studies where researchers used the assessment and
findings state that reliability was .60 to .86 (Roehl et al, 2005). The findings by
researchers exhibited by test-retest reliability. The researchers published the information
through the data entry program to the municipal counties public website. I collected this
information from the municipal court and compiled by the data entry program to reflect
the period from 2013 to 2015, which was postintervention for the offenders used in the
study. The information I collected reflected the two-independent variables the first with
two levels; VOM and BIP programs, and the second which is the covariate AGE.
Individuals who participated in a VOM were placed in VOM and those who participated
in a BIP were placed in BIP in the SPSS program. I determined recidivism to be at least
one subsequent charge of domestic violence within the 24-month period after the initial
offense in 2013. I coded the recidivism data as nominal variables beginning at 0 for no
new offenses and 1 to reflect recidivism. I ran an ANCOVA analysis in SPSS to analyze
the data. I also ran assumption test to address homogeneity of variance, linearity, normal
distribution, and independence. I then coded the data to protect the personal information
of the offenders in accordance with ethical guidelines.
52
Research Design and Rationale
I used a quasi-experimental design to analyze recidivism for first-time domestic
violence offenders in a VOM program or BIP program. I selected the design to examine
the variability in recidivism due to the first-time offender’s interaction in a BIP or VOM
program.
The use of quasi-experimental designs in social research can allow the researcher
to examine variables that are unmanipulable (Frankfort-Nachmias & Nachmias, 2008).
The quasi-experimental design enables the researcher to manipulative the data, which can
influence data collection (Frankfort-Nachmias & Nachmias, 2008). The current study was
a quasi-experimental study with categorical variables, which according to Frankfort-
Nachmias and Nachmias (2008) required for comparative quasi-experimental studies.
There were two independent variables within the study. The first independent
variable had two levels, or conditions, which are groups for this study, and the other
independent variable was age, which was the covariate. The independent variable levels
in the study were VOM and BIP interventions. I defined the level VOM, or victim-
offender mediation, as an RJ intervention that involves the offender, victim, and court-
appointed officials agreeing to a set upon restitution for the crime. The level BIP, or
batterer intervention program, I defined as a course offered to address the defense of
domestic violence and is conducted in a group/classroom setting. I define the age of the
offenders was 18 to 30 years. The dependent variable was recidivism, which I defined as
a re-offense of a domestic violence crime. I examined recidivism in terms of differences
in first-time offenders in VOM and BIPs. Recidivism was scaled as a ratio variable. I
53
demonstrated reliability and validity in that the data were collected from the municipal
court in Ohio which reports date in accordance with the standardized format of the Court
Statistics Project by the Bureau of Justice Statistics. In addition, I collected the
information during this same time-period, and the charges were all domestic violence in
accordance with Ohio State Law. To provide additional evidence of validity, the validity
the process of records collecting was reviewed by the clerks of courts, and then
information was later analyzed by court statistician to examine accuracy of data
regularly. The process of record collecting was reviewed by the Supreme Court Records
Office annually, to adhere to standards of collection for State and Federal reports (The
Office of Criminal Justice Services, 2010). The Municipal Court administrators further
established identity in addition to social security information, birthdate, name, address,
and other demographic information were collected. The court officials provided a
disclaimer stating that the information was collected to the best of their knowledge and
procedures which is displayed on the court website. VOM and BIP variables will be
isolated, and cases where they are combined were not be used in the sample for this
study.
Methodology
In this quasi-experimental study, I analyzed recidivism in first-time domestic
violence offenders exclusively in a VOM or BIP program while controlling for age. The
sample I used in the study was from a population of domestic violence offenders. The
offenders selected from the population were first-time offenders charged with a crime of
domestic violence. The sample was a convenience sample, collected from a municipal
54
court. According to Hedt and Pagano (2011) the use of a convenience sample was
appropriate, as manipulation of variables was not a possibility for this study. The
researchers also stated that a convenience sample was inexpensive to collect and is
readily available saving time (Hedt & Pagano, 2011).
Domestic violence was defined by Ohio Revised Code §2919.25, as an intentional
act to cause physical harm or danger against a family member or household member
(Ohio Government, 2010). The source of the data was the municipal court in Franklin
County, Ohio and surrounding courts. The information administrators collected these data
from the courts from 2013 to 2015. To draw the sample, I pulled data for individuals,
ages 18-30, who participated in a VOM or a BIP in the year of 2013.
The data that I collected were archival data that were originally collected by the
clerk of courts at the municipal courts. According to Frankfort-Nachmias and Nachmias
(2008), this was a convenience under the nonprobability design, which is commonly used
in social science research.
In this study, I used a diverse population through determining sample size using
the G*power calculator. I chose the effect size of .05 as it was the effect size used in
similar studies where researchers analyzed recidivism and difference with interventions
(Mills et al., 2013; Pender, 2012; Sloan et al., 2013). To address type 1 errors, the sample
size was determined to be significant based on Cohen’s 1-β > .80 (Faul, Erdfelder, Lang,
& Buchner, 2009). A priori analysis was run with the effect size 0.40, α. = 05, 1-β = .80,
and df = 1, I determined that the sample size needed least 52 to be statistically significant
according to the G*power calculator for a two-tailed test.
55
Data Analysis Plan
I conducted the current study to examine whether there was a difference in
recidivism between VOM and BIP in first-time domestic violence offenders while
controlling for age. Based on the research questions and hypothesis, the most appropriate
test was ANCOVA, which can be used to test interaction and main effects of categorical
variables, while controlling for variables which covary with the dependent variable
(Bakker & Wicherts, 2014; Frankfort-Nachmias & Nachmias, 2008). The ANCOVA was
two-tailed, with alpha levels; the probability of rejecting the null hypothesis when it is
true; set at p < 0.05. I ran a the Levene’s test was run to test the between-group factors,
and to test that the assumption of homogeneity of regression is met (Frankfort-Nachmias
& Nachmias, 2008). I used SPSS software to analyze of the variables with the ANCOVA.
Archival Data
I designed the study to provide a comparison of the use of VOM and BIP and the
relation the programs have on recidivism on first-time domestic violence offenders.
Frankfort-Nachmias and Frankfort (2008) indicated that the use of archival data for
comparison may assist with generalization and increase understandings of outcomes. The
benefits of the use of archival data in this study were the financial benefits,
methodological reasons, and conceptual-substantive reasons that were identified as
primary factors for the use of archival data (Hedt & Pagona, 2011). The use of archival
data was the best choice for this study due to the nature of the study to include variables,
research questions, and theoretical framework.
56
Variables of the study
In this study, I examined the differences in the following dependent variable:
recidivism, and the two independent variables, the first with two levels: VOM and BIP,
and the second independent variable, that was the covariate age.
Recidivism
I described recidivism in the study as at least one charge of domestic violence by
the offender within the 24-month period after completion of a VOM or BIP program. I
determined that Recidivism was the dependent variable in the study, and was measured as
a ratio variable. I established reliability and validity in the collection of the data, that was
collected by trained court officials and compiled into the Case Management Data
Collection Program in Ohio Municipal Courts, and was reflective of the variables as
defined in this study. The definition aligned with the definition as set by the State of
Ohio as a re-offense of a crime which results in a new charge or conviction within a set
time of the first offense (The Office of Criminal Justice Services, 2010). The court
statistician examined the data for reliability through running a test through the Municipal
Court Case Management system that the court statistician examined the completeness of
information collected and this information was compared to previous data collected by
the statisticians working for the municipal courts. The court statistician determined Face
validity after the data were analyzed and examined against previously collected
recidivism data. The court statistician measured recidivism data were measured at the 1,
2, and 3 year periods after the initial offense (The Office of Criminal Justice Services,
2010). The court statistician scored the data as a “new offense” within the designated
57
time periods of 1, 2, and 3 year periods and scaled as a ratio variable (The Office of
Criminal Justice Services, 2010).
Victim-offender mediation
The State of Ohio Court administrators defined VOM as an intervention program
between the offender and victim and mediated by a certified mediator and court officials,
per Ohio State Law (Taft & Bradley, 2003). I measured the variable as a nominal
variable, and identified as VOM in the study. The court statistician established reliability
and validity in the collection of the data, that will have been collected by the data
program in Ohio Municipal courts, and will be reflective of the variables as defined in
this study.
Batterer intervention program
I defined BIP as a group intervention program for offenders to address issues
related to the domestic violence offense. I labeled BIP as a nominal variable and
identified as BIP in the study. I collected the data information from the Data Program for
Ohio Municipal courts. The court statistician established reliability and validity in the
collection of the data, that was collected by the data program in Ohio Municipal courts,
and was reflective of the variables as defined in this study.
Age
I defined age in the study as the age at the time of the first domestic violence
offense. I labeled age was an ordinal variable and identified as Age in the study. I
collected the data for age from the offense data from the Case Management Data Program
for Ohio Municipal courts. The court statistician established reliability and validity were
58
established in the collection process where data were examined for completeness and
consistency by the statisticians and trained staff in the Ohio Municipal Courts.
Operational Definition of Variables
I selected the variables to examine the differences in recidivism. The dependent
variable in the study was recidivism. I defined recidivism as the re-offense of a crime of
domestic violence as classified by Ohio Revised Code §2919.25, as an intentional act of
violence or threatening danger against a family member or household member. The
dependent variable was dichotomous with recidivism being yes or no in a 24-month
period. The first independent variable had two groups BIP and VOM. I categorized the
groups as Intervention, BIP group 1 and VOM group 2. The two interventions were
nominal variables. Age in the study was the second independent variable and covariate I
collected the age of the age the offender at the time of first offense. I scored the age of the
offender as age 18 to age 30, representing age at the time of the first offense. I collected
and determined age by the offender’s date of birth collected in the Ohio Case
Management Court Data System. My selected independent variable with two levels, and
covariate in the study were defined as:
1. Victim Offender Mediation (VOM) was an intervention program between the
offender and victim mediated by a certified mediator and court officials.
2. Batterer Intervention Program (BIP) was a group intervention program for
offenders to address issues related to the domestic violence offense.
3. Age that was determined at the time of the first-offense and will be
determined by birth date collected by the Municipal Court labeled as AGE.
59
My dependent variable was defined as:
1. I defined recidivism as a new conviction of domestic violence within the 24-
month period post-intervention labeled as RCDVSM.
Research Question
RQ 1: What is the difference between recidivism rates for offenders who have
participated in the restorative justice intervention VOM versus those who have
participated in a BIP for first-time domestic violence offenders 24 months’
postintervention while controlling for age?
Hypotheses
In this study, I examined differences in recidivism for first-time domestic violence
offenders in VOM versus BIP programs while controlling for age. Recidivism was
defined as at least one charge of domestic within a 24-month period post VOM and BIP
intervention labeled as RCDVSM. The age of the offender was the age determined by
birthdate at the time of the charge labeled as AGE. Age was the covariate in the study.
The hypothesis is as follows:
Ho: µ1=µ2: There are no differences in recidivism rates between offenders
enrolled in VOM versus BIP at 24 months’ postintervention, while controlling for age.
H1: µ1≠µ2: There are differences in recidivism rates between offenders enrolled
in VOM versus BIP at 24 months’ postintervention, while controlling for age.
Threats to validity and limitations
I determined that the quasi -experimental design was a valid and a reliable design
to examine the difference between recidivism and VOM and BIP interventions for first-
60
time domestic violence offenders. To minimize the threats to validity, I addressed the
rationale for the study, by the design of the study, and being preventative about threats
(Trochim & Donnelly, 2008). To address the rationale for the study research questions
and hypothesis I supported the rationale through the findings in previous research. I
addressed the gap in the literature and provided support through previous studies. I
determined the design of the study as appropriate based on previous studies where
researchers addressed recidivism.
Due to the nature of the study statistical regression, history, and mortality can
pose internal validity concerns when data were collected ex-post facto (Campbell &
Stanley, 1963). In addressing mortality, I understood that the study was conducted in the
State of Ohio and there was a possibility that during the two years’ postintervention an
offender could have relocated out of the state. Campbell and Stanley (1963) stated that
mortality can be a concern when working with groups, and can potentially skew analysis,
and should be acknowledged. I acknowledged that selection bias was a concern with
using secondary data. History can impact the outcome for the treatment, to include
treatment for previous crimes outside of domestic violence (Campbell & Stanley, 1963).
According to Campbell and Stanley (1963) statistical regression can occur when there are
extreme scores. In the case of this study, I examined two groups, and the scores could be
extreme in one group.
Throughout the study, I took appropriate measures in accordance with standards
set by the IRB to collect the data, and review of questions by peers to determine the
validity for the study. Also, I addressed bias, and the study was subjective, focusing on
61
the research question. Trochim and Donnelly (2008) indicated that analysis chosen for
the studies should reveal accurate results, and not desired. For this study, an ANCOVA
test was the statistical test that I used to determine the difference between variables.
Throughout the completion of this study, my process was evaluated to determine
appropriateness and accuracy.
Ethical Assurances
The first ethical assurance for this study was that I secured IRB approval before
the beginning of any data collection. I ethically guided this study by the Code of Ethics
for research and evaluation as set by the National Organization for Human Services
(NOHS) (National Organization…2015). To protect the confidentiality of the individuals
charged with crimes of domestic violence, I used no names, as well as any other
confidential information. The information I collected and data were secured in a locked
file cabinet. The data I collected was information used by the State of Ohio for benefit of
the State. In addition, the data analyst administrator monitored the use of the data in the
study in regards to the policies and procedures. I submitted a request the data analyst to
collect and release the data were made in correlation with IRB standards. I used a
password, for data that was processed and stored on the computer.
Assumptions
Through analyzing the variables with an ANCOVA, I assumed certain inferences
in explaining the differences in recidivism for first-time domestic violence offenders
enrolled in a BIP or VOM program. According to Field (2013) here was an assumption
that there was additivity and linearity, that Field implied there was a relationship between
62
the predicted variables. I created a scatterplot graph to demonstrate the linearity. I also
used the scatterplot to demonstrate normal distribution with the presence of a bell curve
within the graph. According to field (2013) the next assumption was that there is
homogeneity of variance between the variables that I analyzed with the Levene’s, which
test equality of error variances p >.05. The last assumption according to field (2013) was
that there was independence between means. According to Field (2013) independence
was assumed, as it was used to estimate standard error, that I used to determine
confidence, and statistical test.
Summary
To explore the differences in recidivism in first-time domestic violence offender, I
examined two variables, while controlling for age. Recidivism, in this study, I will label
as a dichotomous variable. I had two independent variables the first independent variable
that had two levels, which were the offenders’ participation in BIP or VOM programs and
the second independent variable was AGE which was the covariate. The individuals in
this study was first-time offenders charged with a crime of domestic violence in the State
of Ohio. To complete analysis of the variables, I collected secondary data from the
municipal court in Franklin County, Ohio including data from smaller surrounding courts.
In Chapter 4, I described the results from the study and statistical analyses. I presented
the conclusion and summary of the study in Chapter 5.
63
Chapter 4: Results
Introduction
The purpose of this quasi-experimental quantitative study was to examine
differences in recidivism between first-time domestic violence offenders enrolled in a BIP
and VOM program, while controlling for age. The research question was: what is the
difference between recidivism rates for offenders who have participated in the restorative
justice intervention VOM versus those who have participated in a BIP for first-time
domestic violence offenders 24 months postintervention while controlling for age? The
hypotheses were Ho: µ1=µ2: There are no differences in recidivism rates between
offenders enrolled in VOM versus BIP at 24 months’ postintervention, while controlling
for age, and H1: µ1≠µ2: There are differences in recidivism rates between offenders
enrolled in VOM versus BIP at 24 months’ postintervention, while controlling for age. I
categorized Recidivism as being charged with a crime of domestic violence at least one
time within the 24-months following the intervention (VOM or BIP). I controlled for the
age of the first-time offender at the time of offense in the study. This chapter was a
description of the data collection process, results of the study, and summary.
Data Collection
I collected archival data for this study from the Ohio Municipal Courts, that
included data from Franklin and surrounding counties. I collected data from timeframe
of February 13, 2017 through March 25, 2017. I found no discrepancies in the collection
process. The court statistician assisted me in the collection of the archival data. The
individuals in the participant population were first-time offenders who had been charged
64
with a crime of domestic violence in 2013. I determined the first-time offender status was
determined by having no previous charges for domestic violence in the State of Ohio. Out
of the 112 records there were 56 individuals who participated in VOM and 56 individuals
who participated in BIP. I collected age from the archival records. There was a total of
43 (38.5%) individuals age 18-24 years and 69 (61.5%)individuals age 25-30 years.
There was a total of 112 records used in the completion of this study. According
to the Ohio Bureau of Criminal Identification and Investigation (2014) there were a total
of 39,979 charges filed for domestic violence in Ohio in 2013. I drew the participants for
this study from a small portion of the State of Ohio Courts. The small sample was a
limitation for the study, and the sample cannot be considered an accurate representation
of the domestic violence population in the State of Ohio. I will describe this limitation in
more detail in Chapter 5.
Results
In the completion of the study, my goal was to examine whether there was a
difference in recidivism between first-time domestic violence offenders who participated
in a VOM or BIP program.
Hypothesis
Ho: µ1=µ2: There are no differences in recidivism rates between offenders
enrolled in VOM versus BIP at 24 months’ postintervention, while controlling for age.
H1: µ1≠µ2: There are differences in recidivism rates between offenders enrolled
in VOM versus BIP at 24 months’ postintervention, while controlling for age.
65
I performed an ANCOVA to analyze the variables. I performed assumption test
to make sure the testing procedures were accurate for an ANCOVA. The first assumption
was to test for linearity. I assumed Linearity based off the plot in Figure 1.
Figure 1. Estimated Marginal means
In the second assumption test I examined normal distribution. In Figures 2 and 3,
distribution can be seen in the Normal Q-Q plot of RCDVSM VOM and BIP.
66
Figure 2. Normal Q-Q Plot RCDVSM for VOM
Figure 3. Normal Q-Q Plot for RCDVSM BIP
The third test that I performed was Levene’s Test of Equality of Error Variance,
which I used to analyze homogeneity of variance. The result of Levene’s Test of
Equality of Error Variance on the dependent variable was a significance of .648 where p
67
> 0.05. As evidenced in Table 1, p > 0.05, thus, the data did not violate the assumption
of equality of error variances. In addition, I assumed the fourth assumption was
independence for the study.
Table 1
Levene’s Test of Equality of Error Variance
F df1 df2 Sig. Alpha
209 1 110 .648 .05
I analyzed the variables with an ANCOVA test between RCDVSM and domestic
violence intervention while controlling for age. In this analysis, the dependent variable
was RCDVSM, while there was one independent variable with two levels: VOM and
BIP. I categorized age as the covariate in the ANCOVA analysis and it was the second
independent variable. The outcomes of the ANCOVA analysis presented in Table 2, were
that there was no significant difference in RCDVSM for first-time male offenders 24-
months post participation in a BIP or a VOM intervention while controlling for age F
(1,109) =.081, p=.777. Thus, I would accept the null hypothesis, that there is no
significant difference between recidivism for first-time male offenders who participate in
a VOM or a BIP intervention.
68
Table 2.
Analysis of Covariance Summary
Source
Sum of
Squares
df
Mean
Square
F
Partial Eta
Squared
Intervention
.017
1
.017
.081
.001
Error
23.254
109
.213
Summary
I conducted a one-way ANCOVA to examine differences in recidivism in BIP and
VOM programs whilst controlling for age. I deemed the ANCOVA to be the most
appropriate analysis as age needed to be controlled for as it co-varied with recidivism. In
addition to analyzing the variables with the ANCOVA, I ran descriptive analyses to
address the assumptions. I performed a scatterplot, Levene’s test, and normality checks
and the assumptions were met. There was no significant difference in mean recidivism
[F (1,109) =.081, p=.777] between the interventions while adjusting for age. Meaning
that recidivism for first-time male offenders who participated in a VOM was not
significantly different than recidivism for first-time male offenders who participated in a
BIP. Section 4 included the results of the ANCOVA analysis of data for this study. In
Section 5 there, I will discuss the interpretation of the results, limitation of the study,
recommendations, and implications of the findings.
69
Chapter 5: Discussion, Conclusion, and Recommendations
Introduction
The purpose of this quasi-experimental, quantitative study was to examine
differences in recidivism for first-time male domestic violence offenders who have
participated in a BIP and a more recently developed alternative: VOM. The research
question was: what is the difference between recidivism rates for offenders who have
participated in the restorative justice intervention VOM versus those who have
participated in a BIP for first-time male domestic violence offenders 24 months’
postintervention while controlling for age? I categorized recidivism as being charged
with a crime of domestic violence at least one time within the 24-months following the
intervention (VOM or BIP). I controlled for age of the first-time offender at the time of
offense for in the study. I used an ANCOVA to analyze the variables, and findings were
that there was no statistical significant difference in recidivism for first-time male
offenders enrolled in a BIP or VOM program. This chapter includes a discussion on the
interpretation of the findings, limitations, recommendations, and implications for social
change.
Interpretation of the Findings
There was one research question addressed in this study: what is the difference
between recidivism rates for offenders who have participated in the restorative justice
intervention VOM versus those who have participated in a BIP for first-time male
domestic violence offenders 24 months’ postintervention while controlling for age? In
this section, I discuss the findings from the study and address each variable in the
70
research question. I also discuss how the findings contributed to the body of literature
examining domestic violence interventions and how it relates to previous research
findings.
Recidivism
I examined the differences in recidivism between VOM and BIP interventions. In
previous research studies researchers found that recidivism in cases of domestic violence
in the United States occurs at a higher rate than other violent crimes (Frantzen et al, 2011;
Mills et al, 2013; Richards et al, 2014; Sloan et al, 2013). I took findings from previous
studies in consideration in the use of interventions in cases of domestic violence,
including: protection orders, probation, incarceration, and batterer intervention programs.
In the implications of these studies researchers suggested that there was a need to
research alterantive interventions to use in cases of domestic violence to reduce
recidivism. According to Mills et al., (2013), Pender (2012), and Sherman and Harris,
(2013) the most commonly used intervention in cases of domestic violence were BIPs,
that research has shown to be effective in the reduction of recidivism as compared to
traditional sanctions including probation, arrest, and incarceration. There was limited
research examining recidivism in cases of domestic violence with the use of RJIs,
including VOMs. In one study, that compared recidivism in the use of a RJI Circles of
Peace and a BIP in cases of domestic violence, researchers found that there was no
significance in the reduction of recidivism between the two interventions (Mills et al.,
2013).
71
I defined recidivism as at least one charge of domestic violence postintervention;
VOM and BIP. I analyzed the variables with an ANCOVA. The results of the analysis,
that examined the differences in recidivism while controlling for age were F (1,109)
=.081, p=.777. The findings showed that there was no significant difference in recidivism
between first-time male offenders who participated in a VOM or BIP program. Thus, I
accepted the null hypothesis, that there is no significant difference between recidivism for
first-time offenders who participate in a VOM or BIP intervention. The results were
similar to previous results in a study completed by Mills, Baracos, and Ariel (2013), in
which the researchers examined recidivism for first-time offenders who participated in a
CP RJI program and BIP. The results of the study by Mills, Baracos, and Ariel (2013)
indicated that there was no significant difference in recidivism for offenders enrolled in a
RJI Circle of Peace program when compared to offenders enrolled in a BIP. The
implications of the Mills, Baracos, and Ariel (2013) study were that there were
interventions that were significant in reducing recidivism, and other RJI should be
explored. With this information added to the body of literature examining domestic
violence intervention, there will be a continued discussion on alternative interventions to
punitive sanctions such as restraining orders, probation, and incarceration.
VOM
According to Dhami (2012) VOM sessions should be focused on the crime being
between the victim and offender Additionally, researchers found that the use of RJIs has
resulted in the reduction of recidivism for nonviolent and violent offenders (Mongold &
Edwards, 2014).
72
VOM was one of the levels of the first independent variable in the study. I
defined VOM as an RJ intervention that involves the offender, victim, and court-
appointed officials agreeing to a set upon restitution for the crime. In the study, there was
a total of 56 offenders who participated in a VOM intervention program. Of the 56
individuals who participated in VOM, 16 (28.57%) were charged with a subsequent
crime of domestic violence within the 24-month period postintervention. 40 (71.43%)
were not charged with a crime of domestic violence within the 24-months
postintervention. I determined that the findings suggest that VOM is a factor in the
reduction of recidivism for domestic violence offenders. Since VOM has also been
linked to the reduction in nonviolent crimes, it would have been interesting to explore if
the offenders who completed VOM committed any non-violent crimes post completion of
the intervention.
BIP
While there are varying types of interventions used in cases of domestic violence,
according to researchers the most commonly used is the BIP. According to Herman et
al., (2014) BIPs were designed as a response to court- mandated treatment programs for
batters. Unlike in the VOM intervention, according to Herman et al, victims were not
included in the BIP intervention. The researchers stated that the offender was placed in a
program with other offenders where behaviors were addressed, where most frequently the
Duluth Model is used (Herman et al., 2004).
In the study by Pender 2012, judges assigned offenders to participate in BIPs after
a conviction of domestic violence. There was a total of 56 offenders who participated in a
73
BIP intervention. Of the 56, 39 (69.64%) were not charged with a subsequent crime of
domestic violence postintervention. A total of 17 (30.36%) offenders were charged with
a crime of domestic violence 24-months postintervention. All the offenders were
sentenced to probation in conjunction with the BIP intervention. This information is
congruent with previous research that addressed if BIPs could be considered treatment, as
the program is often linked with punitive sanctions (Herman et al., 2014; Mills et al.,
2013; Pender, 2012). In the pool of archival data there were more cases of offenders
who participated in a BIP than in VOM. This finding aligns with previous research
stating that BIP was the most commonly used intervention (Pender, 2012). Based on the
data, researchers have found that BIP has been shown to have some impact on recidivism
on cases of domestic violence, but researchers must keep in mind other factors which
may impact recidivism, to include probation (Herman et al., 2014; Mills et al., 2013;
Pender, 2012).
Age
Findings from previous research showed that between the age of 18 and 24 there
was a higher instance of violent offense among male perpetrators (Juodis et al., 2014;
Nelson, 2013). In a previous study, the researcher found that recidivism for men who
commit crimes of domestic violence decreased with age, when compared to other violent
crimes (Nelson, 2013). Based on the findings from previous research, and the
researcher’s findings that suggested that domestic violence offending is highly correlated
with age, it was appropriate for age to be a covariate in this study (Juodis et al., 2014;
Nelson, 2013).
74
In this study, there were a total of 29 (51.79%) offenders who participated in a
BIP between the ages of 18-24 and 27 (48.2%) in the age group of 25-30. In the BIP
there were a total of 10 (17.86%) offenders who were charged with a crime of domestic
violence 24-months postintervention as compared to the seven (12.5%) of offenders age
25-30. The age groups were almost similar in size, but there were slight differences in
the recidivism between the two groups. The total participant size may not be
representative of the population, which impacts whether these results can be generalized.
In regard to VOM participants, there were a total of 14 (25%) between the ages of
18-24 and 42 (75%) between the ages of 25-30. Of the offenders ages 18-24, three
(5.36%) reoffended within the 24-months postintervention compared to 13 (23.21%) in
ages 25-30. I noted that 42 (75%) of the offenders were in the 25-30 age group. There
was a considerable difference with the VOM participants’ ages than with BIP. Various
factors can be contributed to this difference, with the main being the type of courts. While
offenders can be sentenced by BIP intervention in either courts, VOM is only offender in
the Juvenile and Domestic Relations Court, which cases of divorce, child custody, and
parenting petitions are also handled. It was noted in previous research that another
characteristic of domestic violence offenders age 18-24 was that they were more likely to
be non-married with no children (Juodis et al., 2014; Nelson, 2013).
Theoretical Framework
There are many factors that affect whether first-time domestic violence offenders
will reoffend. One factor was how committed that individual is to society (Morrison &
Vaandering, 2012). The RJ theorists highlighted understanding of the offense and
75
determining the appropriate level of intervention (Gavrielides, 2015; Laxminarayan &
Woldhuis, 2015; Mills et al., 2013). In the State of Ohio, the use of interventions in
domestic violence were detailed in the Domestic Violence Bench Book (Taft & Bradley,
2003). I took special consideration in regard to the use of VOM, and it was primarily
used in juvenile and domestic relation courts throughout the State.
A component of RJI was the offender being committed to society, which includes
family and friends. Researchers suggested that despite domestic violence cases being
managed in the juvenile and domestic relations courts, there should be a focus on the
family in cases where VOM was used. The placement in the juvenile and domestic
relations court also related to the reintegrative shaming theory that focused on the shame
of an individual that the offender holds of high esteem, which could include family
members (Braithwaite, 1989). Braithwaite focused on the reduction in recidivism in
reintegrative shaming theory, which was the goal of both the VOM and BIP programs
(Braithwaite, 1989).
In the RJ and reintegrative shaming theories, the theorist highlighted that the use
of interventions should occur postconviction to allow victims time to heal, and offenders
time to gain empathy and remorse for the crimes (Braithwaite, 1989; Dhami, 2012;
Herman et al., 2014; Miller & Iovanni, 2013). Both interventions (VOM and BIP)
occurred postintervention. The difference was that VOM occurred at the request of the
victim or judge in conjunction with other proceedings including divorce or child custody
(Taft & Bradley, 2003). VOM can also be requested in the State of Ohio in cases of
domestic violence without a formal charge of domestic violence being filed (Taft &
76
Bradley, 2003). BIP programs, on the other hand, occurred while the individual was on
probation, in cases in which the information could be reported back to court regarding the
offender’s completion of the program (Taft & Bradley, 2003). According to Braithwaite,
when used, the BIP intervention could result in punitive shaming, which was not a
component of the reintegrative shaming theory (Braithwaite, 1989). These theories
helped further explain why RJIs need to be formatted differently than BIPs.
Limitations of the Study
As with any research design, there were limitations to the study regarding the use
of secondary data. The first limitation according to Frankfort-Nachmias and Nachmias
(2008) was that the sample of participants was derived from a nonprobability
(convenience) sample. The statistician collected the information in the municipal and
court of common pleas clerk of courts, and then placed in the court system. I could not
control for the completion of the demographics in the data.
I determined another limitation was that the data were collected from one court in
Ohio which has court data from smaller counties in the surrounding area. I also
determined that there were factors throughout the State of Ohio which may not have been
reflected in this data due to the small participant pool.
The design methodology selected was reliable and valid in examining differences
in first-time male domestic violence offenders in a VOM or BIP intervention. To address
validity, I examined if the data were collected using appropriate methods. I collaborated
with the statistician with the municipal court in the collection of data. The process for
collection of data included information being placed in the court information system by
77
the clerks of court, and the data processors and statiticians evaluated the data for
completion and accuracy. The data were also reliable, as I gathered the information
from a government agencies with policies in place for collection in accordance with the
Court Statistics Project by the Bureau of Justice Statistics, that was a standardized
method of collection used in the State of Ohio and nationwide.
There were also limitations that I previously highlighted in regard to internal
validity threats. Campbell and Stanley (1963) highlighted the importance of addressing
these threats in regard to establishing evidence of causality. The first internal validity
threat was history. I collected the data from a secondary source, and I was not able to
determine if the offenders had participated in any other form of treatment prior to the
intervention being analyzed in the study. I determined that history could have also been a
factor as recidivism was analyzed over a 2-year period after intervention. I examined the
data and was not able identify if the participants were enrolled in another intervention
program during that period.
The next internal validity threat I examined was statistical regression. Again, as
the data were secondary, I was not able to control for individuals who may be prone to
recidivism. While there were no extreme variables within this study, statistical regression
according to Stanley and Campbell (1963) is seen a threat to internal validity when
analyzing secondary data. The last threat to internal validity I examined was mortality.
According to Stanley and Campbell (1963) mortality refers to the loss of subjects within
the study. In regard to the use of secondary data for this study, I was not able to control or
account from the individuals selected not being in the study area 24-months
78
postintervention. While mortality is an internal validity threat in many studies, within the
quasi-experimental design according to Stanley and Campbell (1963) it could be a
concern as the researcher is not primarily collecting the data, thus cannot account for all
the participants throughout the study. According to Stanley and Campbell (1963) when
using secondary data, the researcher is dependent upon the primary source of collection
to address these concerns during the data collection process.
Discussion
Despite these limitations, I could control for age in the study. In regard to
recidivism for violent crimes, in implications from previous studies it was found that age
was a significant factor in recidivism for male perpetrators age 18-24 (Juodis et al., 2014;
Nelson, 2013). Based on these findings I determined that controlling for age in the study
was necessary to address the effects on the VOM and BIP interventions on recidivism. I
also determined that age also mitigated the statistical regression threat to validity. By
controlling for age, I reduced the causality of recidivism being linked to the age of the
offenders. Controlling for age proved to be a significant process for this study after
collecting the data I found that 50% of the individuals were age 18-24 and 50% were age
25-30 completed BIP, while 25% of the individuals were ages 18-24 who completed a
VOM while the other 75% were ages 25-30. I made age the covariate, so that I could
reduce the impact of age on the findings.
If I did not control for age in the study I determined that the data for offenders
between ages of 25-30 would have skewed the results, and there could have potentially
been a significantly lower level of recidivism for the VOM group as compared to the BIP
79
group. Thus, the findings from previous researchers that stated that recidivism is reduced
as age increases above the age of 25 in violent crimes, would have aligned with my
results (Juodis et al., 2014; Nelson, 2013).
As previously stated, another difference I determined was the demographics of
offenders which could account for the age differences in the VOM program as compared
to the BIP program. In the State of Ohio, cases were assigned to VOM in the juvenile and
domestic relations court, as compared to BIP being assigned in the juvenile and domestic
relations courts, and the municipal courts. Some of the key demographics of the offenders
tried in the juvenile and domestic relations court were that they were married and had
kids. Demographically, previous researchers found offenders ages 18-24 were non-
married without children (Juodis et al., 2014; Nelson, 2013). These findings from
previous studies could account for the difference in ages for the VOM and BIP participant
groups. These findings may be significant in other studies, since increase in age has been
correlated to the decrease in recidivism, in programs where VOM is linked strictly to
juvenile and domestic relations cases, there may be differences in age, that could account
for the decrease in recidivism.
I found the separation of the programs aligned with the concerns associated with
VOM programs. One of the concerns reported by researchers in previous studies
examining the limited use of VOM programs was safety. Taft and Brady stated that based
on the requirements by the State of Ohio the use of a VOM program should be used in
accordance to the lease minimal harm to the victim, and primarily used in cases of
domestic violence where there are children involved with the family. The location of the
80
VOM program in the Juvenile and Domestic Relations Court aligns with the criteria as
set in the State of Ohio Domestic Violence Bench book.
Recommendations
In implications from previous studies researchers suggested that there needed to
be additional research in regard to evaluating the use of RJIs in cases of domestic
violence (Gavrielides, 2015; Mills et al., 2013). Despite the findings that have shown that
RJIs are impactful in the reduction of recidivism in other violent crimes at the same level
as BIPs, I found in examination of research that there was still reluctance within the
criminal justice system in the use of RJIs.
In analyzing the results of the current study, in regard to previous studies, I would
recommend that “satisfaction” be address with the offenders “VOM or BIP programs. By
examining the offender’s viewpoint on the treatment, I believe there could be a better
understanding of why some offenders reoffend as compared to others. Also, I would
recommend that a study be completed with the offenders while they are in the programs,
and follow them over the 24-month period. By doing so, I could have accounted for
mortality in the study. Additionally, I would recommend examining a larger participant
pool, would allow for the study to be generalizable to a larger population.
In addition, I would recommend that further analysis should be done to examine
recidivism with the use of additional RJIs to include Family Group Counseling (FGC).
FGC is an extension of VOM (Hipple, Gruenewald, & McGarrell, 2015). While, VOM
includes the victim, offender, and court officials; FGC on the other hand included the
same individuals plus law enforcement and community partners (Hipple et al, 2015). In
81
addition, FGCs program administrators incorporated the theory of reintegrative shaming,
and was found to be significant in the reductions of recidivism in youth offenders (Hipple
et al, 2015). In the implications of the study researchers suggested additional research to
examine the impact of RJIs on recidivism. Lastly, I would recommend further research
examining judge’s decisions to sentence offenders to RJIs of BIP, could assist with a
better understanding of judge’s knowledge of the interventions.
Implications
The purpose of this research was to examine the impact of restorative justice
interventions on domestic violence in a population that has been deemed high risk for
violent crimes. In addition, I conducted the study to add to the body of research in regard
to domestic violence interventions. I found the outcomes of this research were congruent
with previous studies were researchers that analyzed similar variables. Significant
differences in recidivism have not been found, thus, justifying the need for additional
research on the use of RJIs in case of domestic violence. I determined the information
from this study can be useful within the judicial system, as it was found in previous
research that RJIs are not used due to lack of research examining the impact on
recidivism in domestic violence cases (Frantzen et al., 2011). Lastly, this study could lead
to additional research in examining requirements for participation in RJIs and BIPs.
Making sure offenders are placed in the most appropriate treatment, could not only
benefit offenders, but victims, and communities.
82
Conclusion
The responsibility of reducing recidivism lies not only with the offenders, but
with researchers, and interventionist. Through this research study I have added to the
body of literature examining recidivism in cases of domestic violence. It is important as
researchers that we are continuing to support a platform for future research. Domestic
violence continues to impact victims, offenders, and communities, and this study
highlighted that there are alternatives to current interventions. There is a need for
continued research as it could provide evidence and support for changes in sentencing
and treatment of offenders, as well as how to best address the need of victims in cases of
domestic violence.
83
References
Alarid, L. F., & Montemayor, C. D. (2012). Implementing restorative justice in police
departments. Police Practice & Research: An International Journal, 13(5), 450-
463. doi:10.1080/15614263.2011.607654
Androff, D. K. (2012). Narrative healing among victims of violence: the impact of the
Greensboro truth and reconciliation commission. Families In Society: The Journal
Of Contemporary Social Services, (1), 38. Retrieved from Ebscohost.
Apel, R. (2013). Sanctions, perceptions, and crime: Implications for criminal deterrence.
Journal of Quantitative Criminology, 29(1), 67-101. doi: 10.1007/s10940-012-
9170-9171.
Arias, E., Arce, R., & Vilariño, M. (2013). Batterer intervention programmes: A meta-
analytic review of effectiveness. Psychosocial Intervention / Intervencion
Psicosocial, 22(2), 153-160. doi: 10.5093/in2013a18
Bakker, M., & Wicherts, J. M. (2014). Outlier removal, sum scores, and the inflation of
the type I error rate in independent samples t-tests: The power of alternatives and
recommendations. Psychological Methods, 19(3), 409-427. doi:
10.1037/met0000014
Bell, M. E., Cattaneo, L. B., Goodman, L. A., & Dutton, M. A. (2013). Criminal case
outcomes, incarceration, and subsequent intimate partner violence. Journal of
Family Violence, 28(5), 489-502. doi: 10.1007/s10896-013-9515-z
Boal, A., & Mankowski, E. (2014). Barriers to compliance with oregon batterer
intervention program standards. Violence and Victims, 29(4), 607-619.
84
Boonsit, A., Piemyat, S., & Claassen, R. (2011). Restorative justice: Innovative conflict
resolution for family violence in Thailand. International Perspectives in
Victimology, 6(1), 61-64.
Bradley, M. S. (2015). Gender and the acceptance of retaliatory violence among
acquaintance, dating, and marital relationships. Violence and Victims, 30(3), 377-
392. doi:10.1891/0886-6708.VV-D-13-00169
Braithwaite, J. (1989). Crime, shame, and reintegration. Cambridgeshire, England:
Cambridge University Press.
Breiding, M., Smith, S., Basile, K., Walters, M., Chen, J., & Merrick, M. (2014).
Prevalence and characteristics of sexual violence, stalking, and intimate partner
violence victimization — National Intimate Partner and Sexual Violence Survey,
United States, 2011. Retrieved October 5, 2015, from
http://www.cdc.gov/mmwr/preview/mmwrhtml/ss6308a1.htm?s_cid=ss6308a1_e
Chaban, S. (2014). International human rights law and domestic violence: The
effectiveness of international human rights law by Ronagh J.A. McQuigg. Human
Rights Review, 15(1), 111-113. doi: 10.1007/s12142-014-0310-x
Campbell, D. T., & Stanley, J. C. (1963). Experimental and quasi-experimental designs
for research. Boston: Houghton Mifflin Company.
Connors, A. D., Mills, J. F., & Gray, A. L. (2013). Intimate partner violence intervention
for high-risk offenders. Psychological Services, 10(1), 12-23. doi:
10.1037/a0028979
85
Cucinnelli, K. (2014). 2013 annual report: Domestic and sexual violence in Virginia.
Retrieved November 5, 2015, from
http://www.oag.state.va.us/images/DomViolence/2013StatewideDVReport.pdf
Daly, K. (2013). The punishment debate in restorative justice. In J. Simon & R. Sparks
(Eds.), The SAGE handbook of punishment and society (pp. 356-375). London:
England. SAGE. doi:10.4135/9781446247624.n18
De Grace, A., BA, & Clarke, A., (2012). Promising practices in the prevention of intimate
partner violence among adolescents. Violence and Victims, 27(6), 849-59.
Retrieved from
http://search.proquest.com/docview/1243100841?accountid=14872
Dhami, M. (2012). Offer and Acceptance of Apology in Victim-Offender Mediation.
Critical Criminology, 20(1), 45-60. doi: 10.1007/s10612-011-9149-5
Dickson-Gilmore, J. (2014). Whither restorativeness? Restorative justice and the
challenge of intimate violence in Aboriginal communities. Canadian Journal of
Criminology and Criminal Justice, 56(4), 417-446. doi: 10.3138/cjccj. 2014. S02
Dutton, D. G. (2012). The prevention of intimate partner violence. Prevention Science,
13(4), 395-397. doi: 10.1007/s11121-012-0306-1
Faul, F., Erdfelder, E., Buchner, A., & Lang, A.-G. (2009). Statistical power analyses
using G*Power 3.1: Tests for correlation and regression analyses. Behavior
Research Methods, 41, 1149-1160.
Field, A. (2013). Discovering Statistics Using IBM SPSS Statistics (4th ed.). London:
Sage
86
Frankfort-Nachmias, C & Nachmias, D. (2008). Research methods in the social sciences.
New York, NY: Worth Publishers.
Frantzen, D., Miguel, C. S., & Kwak, D. (2011). Predicting case conviction and domestic
violence recidivism: Measuring the deterrent effects of conviction and protection
order violations. Violence and Victims, 26(4), 395-409. doi:10.1891/0886-
6708.26.4.395
Gavrielides, T. (2015). Is restorative justice appropriate for domestic violence cases?
Social Work Review / Revista De Asistenta Sociala, 14(4), 105-121.
Gromet D, Okimoto T, Wenzel M, Darley J. (2012) A victim-centered approach to
justice? Victim satisfaction effects on third-party punishments. Law and Human
Behavior 36(5):375-389. Available from: PsycARTICLES, Ipswich, MA.
Haas, K. (2012). Public shaming as punishment. In S. Barton-Bellessa
(Ed.), Encyclopedia of community corrections. 357-363. Thousand Oaks, CA:
SAGE Publications, Inc. doi:
http://dx.doi.org.ezp.waldenulibrary.org/10.4135/9781452218519.n145
Hedt, B. L., & Pagano, M. (2011). Health indicators: eliminating bias from convenience
sampling estimators. Statistics in Medicine, 30(5), 560-568.
Herman, K., Rotunda, R., Williamson, G., & Vodanovich, S. (2014). Outcomes from a
duluth model batterer intervention program at completion and long term follow-
up. Journal of Offender Rehabilitation, 53(1), 1-18.
doi:10.1080/10509674.2013.861316
87
Hipple, N. K., Gruenewald, J., & McGarrell, E. F. (2015). Restorativeness, Procedural
Justice, and Defiance as Long-Term Predictors of Reoffending of Participants in
Family Group Conferences. Criminal Justice and Behavior, 42(11), 1110-1127.
doi: 10.1177/0093854815601153
Intimate Partner Violence. (2015). Retrieved October 5, 2015, from
http://www.cdc.gov/violenceprevention/intimatepartnerviolence
Juodis, M., Starzomski, A., Porter, S., & Woodworth, M. (2014). What can be done about
high-risk perpetrators of domestic violence? Journal of Family Violence, 29(4),
381-390. doi: http://dx.doi.org/10.1007/s10896-014-9597-2
Kenny, P., & Leonard, L. (2014). Sustainability of Restorative Justice. Cambridge,
Massachusetts: Emerald Group Publishing Limited.
Lawson, D. M., & Brossart, D. F., (2013). Interpersonal problems and personality
features as mediators between attachment and intimate partner violence. Violence
and Victims, 28(3), 414-28. Retrieved from
http://search.proquest.com/docview/1368936818?accountid=14872
Lipsky, S., Kernic, M. A., Qiu, Q., Wright, C., & Hasin, D. S. (2014). A two-way street
for alcohol use and partner violence: Who's driving it? Journal of Family
Violence, 29(8), 815-828. doi: http://dx.doi.org/10.1007/s10896-014-9635-0
Lohman, B. J., Neppl, T. K., Senia, J. M., & Schofield, T. J. (2013). Understanding
adolescent and family influences on intimate partner psychological violence
during emerging adulthood and adulthood. Journal of Youth and Adolescence,
42(4), 500-17. doi: http://dx.doi.org/10.1007/s10964-013-9923-7
88
McKeown, A. (2014). Female offenders who commit domestic violence: aggression
characteristics and potential treatment pathways. Journal of Forensic Practice,
16(2), 127-138. doi: http://dx.doi.org/10.1108/JFP-01-2013-0007
McShane, M., & Williams III, F. (2004). Criminological theory (Fourth ed.). New York:
NY Pearson.
Miller, S. L., & Iovanni, L. (2013) Using Restorative justice for gendered violence:
success with a post-conviction model. Feminist Criminology, 8, 247-268.
Retrieved from
http://fcx.sagepub.com.ezp.waldenulibrary.org/content/8/4/247.full.pdf+html
Mills, L. G., Barocas, B., & Ariel, B. (2013). The next generation of court-mandated
domestic violence treatment: A comparison study of batterer intervention and
restorative justice programs. Journal of Experimental Criminology, 9(1), 65-90.
doi: 10.1007/s11292-012-9164-x
Mongold, J. L., & Edwards, B. D. (2014). Reintegrative shaming: Theory into practice.
Journal of Theoretical & Philosophical Criminology, 6(3), 205-212. Retrieved
from http://search.proquest.com/docview/1635433331?accountid=14872
Munjal, D. (2012). Intimate partner violence - is there a solution?. Duke Journal Of
Gender Law & Policy, 19347. Retrieved from Ebscohost.
Nelson, E. L. (2013). Domestic violence sentencing: Coefficient to a natural process that
already reduces recidivism simply as a function of aging. Crime Science, 2(1), 1-
33. doi: http://dx.doi.org/10.1186/2193-7680-2-9
89
Ohio Bureau of Criminal Identification and Investigation. (2014). Domestic Violence
Report. Retrieved from http://www.ohioattorneygeneral.gov/Law-
Enforcement/Services-for-Law-Enforcement/Domestic-Violence-
Reports/Domestic-Violence-Reports-2013/2013-Domestic-Violence-Incidents-by-
County-and-Age
Ohio Government. (2010). 2919.25 Domestic violence. Retrieved July 26, 2016, from
http://codes.ohio.gov/orc/2919.25
O’Leary, K. D., Tintle, N., & Bromet, E. (2014). Risk factors for physical violence
against partners in the U.S. Psychology of Violence, 4(1), 65-77. doi:
10.1037/a0034537
Pender, R. L. (2012). ASGW best practice guidelines: An evaluation of the duluth model.
Journal for Specialists in Group Work, 37(3), 218-231.
doi:10.1080/01933922.2011.632813
Pow, A. M., Murray, C. E., Flasch, Paulina, Doom, Elizabeth, & Snyder, M. (2015).
Learning from experience: A content analysis of domestic violence fatality review
team reports. Partner Abuse, 6(2), 197-216. Retrieved from
http://auth.waldenulibrary.org/ezpws.exe?url=/docview/1667926934?accountid=1
4872
Price, M. S. (2014). All's fair in love and war--or is it? Domestic violence and weapons
bans. American Journal of Family Law, 28(2), 79-86.
Qureshi, S. (2013). The emergence/extention of due diligence standard to assess the state
response towards violence against Women/Domestic violence. South Asian
90
Studies, 28(1), 55-66. Retrieved from
http://auth.waldenulibrary.org/ezpws.exe?url=/docview/1369670705?accountid=1
4872
Qureshi, S. (2013). The recognition of violence against women as a violation of human
rights in the united nations system. South Asian Studies, 28(1), 187-198.
Retrieved from
http://search.proquest.com/docview/1369670720?accountid=14872
Ramji-Nogales, J. (2014). Violence against women under international human rights
law. The American Journal of International Law, 108(2), 376-380.
Richards, T. N., Jennings, W. G., Tomsich, E., & Gover, A. (2014). A 10-year analysis of
rearrests among a cohort of domestic violence offenders. Violence and
victims, 29(6), 887-906.
Roehl, J., O'Sullivan, C., Webster, D., & Campbell, J. (2005). Intimate Partner Violence
Risk Assessment Validation Study, Final Report (Rep. No. 209731). Department
of Justice. Retrieved from https://www.ncjrs.gov/pdffiles1/nij/grants/209731.pdf.
Sloan, F. A., Platt, A. C., Chepke, L. M., & Blevins, C. E. (2013). Deterring domestic
violence: Do criminal sanctions reduce repeat offenses?. Journal of Risk and
Uncertainty, 46(1), 51-80. doi: 10.1007/s11166-012-9159-z
Sherman, L. W., & Harris, H. M. (2013). Increased homicide victimization of suspects
arrested for domestic assault: A 23-year follow-up of the Milwaukee domestic
violence experiment (MilDVE). Journal of Experimental Criminology, 9(4), 491-
514. doi: http://dx.doi.org/10.1007/s11292-013-9193-0
91
Sherman, L., Strang, H., Mayo-Wilson, E., Woods, D., & Ariel, B. (2015). Are restorative
justice conferences effective in reducing repeat offending? findings from a
Campbell systematic review. Journal of Quantitative Criminology, 31(1), 1-24.
doi: 10.1007/s10940-014-9222-9
Shorey, R. C., Seavey, A. E., Quinn, E., & Cornelius, T. L. (2014). Partner-specific anger
management as a mediator of the relation between mindfulness and female
perpetrated dating violence. Psychology of Violence, 4(1), 51-64. doi:
10.1037/a0033658
Straus, M. A., & Gozjolko, K. L. (2014). "Intimate terrorism" and gender differences in
injury of dating partners by male and female university students. Journal of
Family Violence, 29(1), 51-65. doi: http://dx.doi.org/10.1007/s10896-013-9560-7
Sutton, T. E., Simons, L. G., Wickrama, K. A. S., & Futris, T. (2014). The
intergenerational transmission of violence: Examining the mediating roles of
insecure attachment and destructive disagreement beliefs. Violence and Victims,
29(4), 670-87. Retrieved from
http://search.proquest.com/docview/1549953528?accountid=14872
Taft, B., & Bradley, J. (2003). The Ohio domestic violence bench book: A practical guide
to competence for judges & magistrates. Columbus: Family Violence Prevention
Center, Ohio Office of Criminal Justice Services.
The Office of Criminal Justice Services. (2010). Ohio Criminal Justice Statistics.
Retrieved from http://www.publicsafety.ohio.gov/links/ocjs_statistics.pdf
92
Trochim, W. M., & Donnelly, J. P. (2008). Research methods knowledge base. Mason,
OH: Atomic Dog/Cengage Learning.
U.S. Department of Justice. (2015). Federal Domestic Violence Laws. Retrieved October
5, 2015, from http://www.justice.gov/usao-wdtn/victim-witness-program/federal-
domestic-violence-laws
Weeber, S. (2012). Restorative justice. In S. Barton-Bellessa (Ed.), Encyclopedia of
community corrections. 395-396. Thousand Oaks, CA: SAGE Publications, Inc.
doi: http://dx.doi.org.ezp.waldenulibrary.org/10.4135/9781452218519.n159
Weissman, D. M. (2013). Law, social movements, and the political economy of domestic
violence. Duke Journal of Gender Law & Policy, 20(2), 221-254.
Whiting, J. B., Parker, T. G., & Houghtaling, A. W. (2014). Explanations of a violent
relationship: The male perpetrator's perspective. Journal of Family
Violence, 29(3), 277-286. doi: http://dx.doi.org/10.1007/s10896-014-9582-9
Witte, T. H., & Cousin Carmen and her daughter Amber 111 Michigan Ave. Pgh pa 15210, M. M.
(2013). Social norms for intimate partner violence. Violence and Victims, 28(6),
959-67. Retrieved from
http://search.proquest.com/docview/1470426180?accountid=14872
