The author(s) shown below used Federal funds provided by the U.S.
Department of Justice and prepared the following final report:
Document Title: Intimate Partner Abuse in Divorce Mediation:
Outcomes from a Long-Term Multi-cultural
Study
Author: Connie J. A. Beck, Ph.D., Michele E. Walsh,
Ph.D., Mindy B. Mechanic, Ph.D., Aurelio Jose
Figueredo, Ph.D., Mei-Kuang Chen, M.A., M.S.
Document No.: 236868
Date Received: December 2011
Award Number: 2007-WG-BX-0028
This report has not been published by the U.S. Department of Justice.
To provide better customer service, NCJRS has made this Federally-
funded grant final report available electronically in addition to
traditional paper copies.
Opinions or points of view expressed are those
of the author(s) and do not necessarily reflect
the official position or policies of the U.S.
Department of Justice.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 1 2007-WG-BX-0028
Intimate Partner Abuse in Divorce Mediation
Outcomes from a Long-Term Multi-cultural Study
Connie J. A. Beck, PhD
1
Michele E. Walsh, PhD
2
Mindy B. Mechanic, PhD
3
Aurelio Jose Figueredo, PhD
1
Mei-Kuang Chen, MA, MS
1
1
School of Mind, Brain and Behavior
Department of Psychology
University of Arizona
Tucson, AZ 85721
2
John & Doris Norton School of Family and Consumer Sciences
Frances McClelland Institute for Children, Youth and Families
University of Arizona
Tucson, AZ 85721
3
Department of Psychology
California State University, Fullerton
April, 2011
Revised July, 2011
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 2 2007-WG-BX-0028
National Institutes of Justice, grant number 2007-WG-BX-0028
Acknowledgements
The study was a massive undertaking spanning nearly a decade. There are many people who
were indispensible in making the study possible. We would first like to thank Pima County
Superior Court Judges Karen Adam and Jan Kearney for recognizing the importance of this
study for family law in Pima County, Arizona, providing access to the databases and for their
support and guidance throughout the project. Without their support the project could never have
been completed.
We would next like to thank the current (Grace Hawkins) and former (Fred Mitchell) directors of
the Pima County Family Center of the Conciliation Court. Grace Hawkins was instrumental in
assisting the study by providing research office space, record storage space, access to computers
and court databases and for allowing the research team to spend years collecting data at their
offices. Fred Mitchell continued the work of former director Linda Kerr in identifying the need
for adequate screening for intimate partner abuse at the study site. Dr. Mitchell instituted a policy
requiring use of a written questionnaire screening measure, which became an integral part of the
data. Paul LaFrance, former office manager, identified variables within the mediation case files
and consulted on many aspects of the initial design of the study. Without this long-term support
the project would not have been possible. Thank you to the many mediators and office staff at
the Conciliation Court for answering our questions and providing assistance to us as we collected
the data. It is a rare public agency that welcomes this type of in-depth research and analysis of
their work.
We would also like to thank Sergeant William Briamonte of the Tucson Police Department and
Mr. Frank Gonzalez of the Pima County Sheriff’s office for working closely with the research
team and designing methods for data collection from area law enforcement databases. Our
colleague and friend Dr. Edward Anderson of the University of Texas at Austin provided vital
statistical consulting and conducted the latent class analyses in this study.
There are few studies which can be conducted without the extensive support of students. We
thank the many undergraduate students (Jessica Jorgensen Willis, Fairlee Fabrett, Merredith
Levenson, Stacey Brady, Whitney Kearney, Melissa Rubin, Ryan Davidson; Karey O’Hara
Brewster), Devon Harris and graduate students (Caitilin Taylor, Marieh Tanha and Melissa
Tehee) who contributed to collecting data, abstracting information from court records, coding
and entering data and conducting reliability checks. In particular, a heartfelt thanks to Karey
O’Hara Brewster and Ryan Davidson. Karey served mid-study as lead research keeping the
project organized and training other students in data collection. Ryan served as the lead research
assistant and liaison to the law enforcement agencies and kept the project going in its final
stages.
Thanks are also due to the National Institute of Justice (NIJ) for the financial support of this
project. Funding for mediation research is so difficult to secure yet so important for the lives of
divorcing parents and children, particularly those with intimate partner violence. We thank the
NIJ staff members who have worked with us over the years. Leora Rosen was instrumental in
stewarding us through the grant application and review process. Christine Crossland, Bethany
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 3 2007-WG-BX-0028
Backes and Bernie Auchter have served as program officers. A special thanks to Bernie Auchter
for his helpful suggestions, patience and generosity in providing important opportunities for us to
discuss the issues in this study and to present the research to a broad audience. Thanks are also
extended to the two reviewers of the draft final report. Both provided to us extremely important
comments and insights.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 4 2007-WG-BX-0028
Abstract
Despite three decades of scholarly research on numerous aspects of divorce mediation,
there is no comprehensive understanding of the short- and long-term outcomes for couples
legally ordered to mediation to resolve custody and parenting time disputes or for those using
free (or low cost) conciliation court mediation services to do so. Even less is known about the
use or effectiveness of court-mandated mediation services among couples alleging intimate
partner abuse (IPA). This study was funded by NIJ to address these gaps in the literature. Using
several archival court and law enforcement databases, we systematically documented actual
percentages of IPA in those participating in mediation, systematically analyzed mediator
practices addressing those IPA cases, and systematically assessed mediation outcomes, divorce
outcomes and post-decree outcomes for IPA cases.
To accomplish this we linked archival data from two court databases and two law
enforcement databases for a large matched sample (N=965) of couples involved in the divorce
process in one court-based mediation program in one jurisdiction. We first linked data produced
in business-as-usual, naturalistic clinical interviews used to screen parents for marital stressors
and IPA to questionnaire data also measuring specific IPA-related behaviors. We then linked this
IPA data to the mediator’s decisions concerning whether to identify a case as having IPA or not,
whether to proceed in mediation or to screen out IPA-identified cases, and whether to provide
special procedural accommodations for IPA-identified cases.
We then linked the IPA and mediator decision data to mediation outcome data from
mediation case files and to outcomes in final divorce decrees and parenting plans found in
Superior Court divorce files. We then linked these pre-divorce and divorce data to post-divorce,
longitudinal data concerning re-litigation of divorce-related issues in Superior Court and
longitudinal data concerning contacts with area law enforcement.
The results of this study provide strong empirical support for previous estimates that most
couples attending divorce mediation report some level of IPA. Mediators accurately identified
many but not all client self-identified cases of IPA. One third of the couples classified as non-
IPA reported at least one incident of threatened and escalated physical violence or sexual
intimidation, coercion or assault. Cases were rarely screened out of mediation (6%) and special
procedural accommodations were most often provided in cases where a parent called the
mediation service requesting the accommodations or reporting concerns about IPA and about
participating in mediation (84%). Calls to area law enforcement and orders of protection were
common (approximately 40% of couples for each category). While mediation agreements that
included restrictions on contact between parents or on parenting were rare, the victims of the
highest level of IPA often left mediation without agreements and returned to court, wherein they
obtained restrictions on contact between parents and/or restrictions on aspects of parenting at a
much higher rate than those appearing in mediation agreements.
Mediators are not judges and therefore, these results are to be expected. It is a rare abuser
who will voluntarily agree to terms that allow less control over contact with the victims and more
structured contact with the couple’s children. The majority of parents in the study returned to
court at some point to re-litigate divorce-related issues (62%); however, a small group of couples
(4.5%) who returned for a tremendous number of hearings (31% of total number of hearings for
all couples in study). The fact that parents reaching agreements are less likely to relitigate
provide significant support for the use of mediation programs.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 5 2007-WG-BX-0028
According to reporting by parents in this study, at least some form of IPA occurred in
over 90% of the cases and two thirds of the couples reported that either or both partners utilized
outside agency involvement from police, shelters, courts, or hospitals to handle the IPA. These
figures represent a tremendous amount of IPA in couples mandated to attend mediation. Thus, it
is essential that highly trained mediators who use standardized screening procedures and follow
program policies regarding how to handle IPA cases.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 6 2007-WG-BX-0028
Table of Contents
Intimate Partner Abuse in Divorce Mediation ................................................................................ 1
Abstract ....................................................................................................................................... 4
Acknowledgements ..................................................................................................................... 2
Table of Contents ........................................................................................................................ 6
List of Figures ............................................................................................................................. 9
List of Tables .............................................................................................................................. 9
Executive Summary ...................................................................................................................... 11
Introduction ............................................................................................................................... 11
Study Goals ............................................................................................................................... 11
Goal 1: Determine whether the mediation program accurately identifies couples with
self-identified IPA and assess whether these cases are treated differently. .......................... 12
Goal 2: Assess whether mediation agreements, divorce decrees and/or parenting plans
include safety measures in cases involving self-reported IPA. ............................................. 12
Goal 3: Assess the frequency with which mediation agreements/divorce decrees are re-
litigated over time. ................................................................................................................ 12
Goal 4: Test a multivariate conceptual model using variables that are hypothesized to
affect mediation, divorce case and post-divorce outcomes (see Cascade Model Figure 16).
12
Study Methods .......................................................................................................................... 12
Results and Conclusions ........................................................................................................... 13
Goal 1: Determine whether the mediation program accurately identifies couples with
self-identified IPA and assess whether these cases are treated differently. .......................... 13
Goal 2: Assess whether mediation agreements, divorce decrees and/or parenting plans
include safety measures in cases involving self-reported IPA. ............................................. 18
Goal 3: Assess the frequency with which mediation agreements/divorce decrees are re-
litigated over time. ................................................................................................................ 21
Goal 4 Results: Test a multivariate conceptual model using variables that are hypothesized
to affect mediation, divorce case and post-divorce outcomes (see Cascade Model Figure 16
............................................................................................................................................... 22
Recommendations for Research, Policy and Practice ............................................................... 28
Limitations ................................................................................................................................ 25
(1) Provide essential training in the dynamics of IPA .................................................... 28
(2) Use a structured, systematic approach to assessment of IPA in mediation .............. 28
(3) Automate screening measures ................................................................................... 29
(4) Assess Mediation special procedures to accommodate victims ............................... 29
(5) Acknowledge the limitations of the mediation process ............................................ 30
(6) Investigate hybrid mediation/arbitration mediation interventions ............................ 30
(7) Encourage legal findings of “domestic violence” in appropriate cases .................... 31
(8) Assess outcomes for children directly and prospectively ......................................... 31
(9) Assess the effectiveness of court-based programs for pre- and post-divorce “frequent
flyers” 32
Introduction ................................................................................................................................... 33
Review of Relevant Research and Background of Study ......................................................... 33
Statement of the Problem. ......................................................................................................... 33
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 7 2007-WG-BX-0028
What is mediation in this context? ........................................................................................ 33
Why did conciliation courts become the main services providers of mandated divorce
mediation? ............................................................................................................................. 35
What is intimate partner violence (IPA)? ............................................................................. 38
Why is mediation often legally mandatory? ......................................................................... 36
Why is mandatory mediation a potential problem for couples with IPA? ............................ 42
What is the prevalence of IPA in couples mandated to mediation? ..................................... 43
What dimensions of IPA are found in couples attending divorce mediation? ...................... 44
What are typologies of IPA and why might they be important in mediation? ..................... 46
What happens when IPA victims are identified in mediation? ............................................. 52
Should victims of IPA be screened out or accommodated? ................................................. 53
What happens when a case has both IPA and child abuse? .................................................. 55
How many people do not have attorneys and what happens in cases without attorneys? .... 58
Is post-separation IPA a problem? ........................................................................................ 60
Is there a risk that victims will mediate agreements that are unsafe? ................................... 62
NIJ Project Goals and Research Questions: .............................................................................. 66
Goal 1: Determine whether the mediation program accurately identifies couples with
self-identified IPA and assess whether these cases are treated differently. .......................... 66
Goal 2: Assess whether mediation agreements, divorce decrees and/or parenting plans
include safety measures in cases involving self-reported IPA. ............................................. 66
Goal 3: Assess the frequency with which mediation agreements/divorce decrees are re-
litigated over time. ................................................................................................................ 66
Goal 4: Test a multivariate conceptual model using variables that are hypothesized to
affect mediation, divorce case and post-divorce outcomes (see Cascade Model Figure 16).
66
Methods......................................................................................................................................... 67
Mediation Process ..................................................................................................................... 72
Participants. ............................................................................................................................... 67
Mediators .................................................................................................................................. 75
Security at Study Site. ........................................................................................................... 75
Mediator Training in Assessing IPA..................................................................................... 75
Data. .......................................................................................................................................... 76
Mediation Data. ..................................................................................................................... 76
Divorce File Data. ................................................................................................................. 80
Area Law Enforcement Data ................................................................................................ 81
Limited Jurisdiction Court Data............................................................................................ 82
Data-Analytic Strategy for Conceptual Model (Cascade Model) ............................................. 82
Marital stressors .................................................................................................................... 84
Interpersonal Abuse (IPA) .................................................................................................... 85
Child Maltreatment ............................................................................................................... 96
Mediator Determination of IPA Present/Absent ................................................................... 96
Mediation Special Procedures ............................................................................................... 97
Mediation Agreements. ......................................................................................................... 97
Orders of Protection. ............................................................................................................. 98
Court Legal Finding of “Domestic Violence.” ..................................................................... 98
Decree Restrictions in Parenting Plans. ................................................................................ 98
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 8 2007-WG-BX-0028
Terms of Parenting Agreements. .......................................................................................... 98
Number of Hearings Post-Decree. ........................................................................................ 99
Post-Decree Calls to the Law Enforcement. ......................................................................... 99
Number of Orders Post-Decree. .......................................................................................... 100
Results ......................................................................................................................................... 101
Statement of results ................................................................................................................. 101
Overview of IPA dimensions in the sample ....................................................................... 101
Goal 1: Determine whether the mediation program accurately identifies couples with
self-identified IPA and assess whether these cases are treated differently. To do so we: . 104
Goal 2: Assess whether mediation agreements, divorce decrees and/or parenting plans
include safety measures in cases involving self-reported IPA. ........................................... 116
Goal 3: Assess the frequency with which mediation agreements/divorce decrees are re-
litigated over time. .............................................................................................................. 124
Goal 4: Test a multivariate conceptual model using variables that are hypothesized to
affect mediation, divorce case and post-divorce outcomes (see Figure 16). ...................... 133
Discussion of the Findings ...................................................................................................... 145
Conclusions and Implications ..................................................................................................... 145
Goal 1: Determine whether the mediation program accurately identifies couples with
self-identified IPA and assess whether these cases are treated differently. ........................ 147
Goal 2: Assess whether mediation agreements, divorce decrees and/or parenting plans
include safety measures in cases involving self-reported IPA. ........................................... 160
Goal 3: Assess the frequency with which mediation agreements/divorce decrees are re-
litigated over time. .............................................................................................................. 169
Goal 4: Test a multivariate conceptual model using variables that are hypothesized to
affect mediation, divorce case and post-divorce outcomes (see Figure 16). ...................... 173
(1) Provide essential training in the dynamics of IPA .................................................. 187
Recommendations for Policy, Practice and Research ............................................................. 187
(2) Use a structured, systematic approach to assessment of IPA in mediation ............ 187
(3) Automate screening measures ................................................................................. 187
(4) Assess Mediation special procedures to accommodate victims ............................. 188
(5) Acknowledge the limitations of the mediation process .......................................... 189
(6) Investigate hybrid mediation/arbitration mediation interventions .......................... 189
(7) Encourage legal findings of “domestic violence” in appropriate cases .................. 190
(8) Assess outcomes for children directly and prospectively ....................................... 190
(9) Assess the effectiveness of court-based programs for pre- and post-divorce “frequent
flyers” 190
Limitations of the research. ..................................................................................................... 192
References ................................................................................................................................... 196
Implications for Future Research. ........................................................................................... 195
Dissemination of Research Findings .......................................................................................... 209
Publications ............................................................................................................................. 209
Publications in Progress and Submitted .................................................................................. 209
Attachments ................................................................................................................................ 213
Attachment A: Mediation Case File Data ............................................................................... 214
Attachment B: Premediation Interview Form (PMI) ............................................................ 229
Attachment C: Relationship Behavior Rating Scale (RBRS) ............................................... 229
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 9 2007-WG-BX-0028
Attachment D: Superior Court Divorce File Data ................................................................. 231
Attachment E: Data Collection Sheet for Law Enforcement Database .................................. 237
List of Figures
Figure 1: Case Flow Chart
Figure 2: Latent Classes and Mean of Continuous Measure of IPA
Figure 3: Latent Classes and Mean of Continuous Measure of IPA --No Coercive Controlling
Behaviors Included in Continuous Measure of IPA
Figure 4: Cases Identified by Mediator as IPA Present/Absent.
Figure 5: Reports of at Least One Incident of Either Threats of and/or Escalated Physical
Violence or Sexual Intimidation/Coercion/Assault in the Preceding 12 Months by
Mediator Identified/Not Identified IPV
Figure 6: T-scores for Client-Reported IPA by Whether IPA was identified by the Mediator
Figure 7: Relationship Among Calls to Law Enforcement at Different Time Periods
Figure 8: Orders of Protection by Client- or Mediator-Identified IPA
Figure 9: Mediation Accommodations Reported IPA
Figure 10: Mediation Agreement Type by T-score Mean of IPA
Figure 11: Legal Finding of Domestic Violence and Reported IPA
Figure 12: Custody Awards in Divorce Decrees by Reported IPA
Figure 13: Percentage of Post-Decree Hearings by Outcome of Mediation
Figure 14: Percentage of Post-Decree Orders by Outcome of Mediation
Figure 15: Percentage of Post-Decree Stipulations by Outcome of Mediation
Figure 16: Cascade Model
List of Tables
Table 1: Participant Demographics
Table 2: Constructs, Measured Variables, and Data Sources for Cascade Model
Table 3: Reliability Estimates for IPA Categories
Table 4: Correlations of Standardized IPA Subtypes and IPA Common Factor Score
Table 5: Raw Means and Standard Deviations of IPA reports by IPA Subscales
Table 6: Calls to Law Enforcement.
Table 7: Time Periods for Calls to Law Enforcement
Table 8: Correlation of Calls to Law Enforcement with Client-Reported IPA
Table 9: Average Calls to Law Enforcement by Whether the Mediator Identified IPA
Table 10: Orders or Protection by Mediator-Identified IPA
Table 11: Comparisons of Mediation Agreement by IPA T-Score
Table 12: Restrictions Placed on Parents or Parenting by Terms of Agreement
Table 13: Parenting Agreements: Physical and Legal Custody
Table 14: Frequencies of Decree Restrictions by Type
Table 15: Post-decree hearings, orders and stipulations
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 10 2007-WG-BX-0028
Table 16: Mean Differences among Post-Decree Hearings and Orders by Mediator-Identified
IPA
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 11 2007-WG-BX-0028
Executive Summary
Introduction
Mediation became popular in the United States in the 1980s as an alternative to the
typical dual attorney, adversarial model of negotiating custody and parenting time issues divorce
cases. It quickly gained tremendous popularity and now exists in some form (legally mandated,
at judicial discretion, or voluntary) in nearly every state in the United States as well as in Britain,
Canada and Australia.
There have been three decades of scholarly research on numerous aspects of divorce
mediation; however there remains no comprehensive understanding of the short- and long-term
outcomes for couples who are legally ordered to mediation and who use court-connected
programs to do so. In the case of couples alleging intimate partner abuse (IPA), even less is
known (Beck & Sales, 2001; Salem, 2009). This study was funded by NIJ to investigate a
naturalistic (as opposed to tightly controlled experimental) process and systematically document
actual percentages of IPA in mediation; to systematically analyze mediator practices addressing
IPA cases; and to assess mediation outcomes, divorce outcomes and post-decree outcomes for
both IPA and non-IPA cases.
Study Goals
The overall goal of this study was to enhance the effectiveness of screening and
management of cases with IPA by providing empirical data concerning outcomes in these cases
to public policy-makers within state legislatures, court systems, mediation programs and
professional organizations. Results of this study can be used to design targeted research to
answer specific questions arising from the study and to begin developing policies and practices
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 12 2007-WG-BX-0028
that will enhance the short- and long-term safety of IPA victims and their violence-exposed
children. The specific study goals were:
Goal 1: Determine whether the mediation program accurately identifies couples with
self-identified IPA and assess whether these cases are treated differently.
Goal 2: Assess whether mediation agreements, divorce decrees and/or parenting plans
include safety measures in cases involving self-reported IPA.
Goal 3: Assess the frequency with which mediation agreements/divorce decrees are re-
litigated over time.
Goal 4: Test a multivariate conceptual model using variables that are hypothesized to
affect mediation, divorce case and post-divorce outcomes (see Cascade Model Figure 16).
Study Methods
To accomplish these goals we conducted a study in Pima County, Arizona which linked
archival data from two court databases and two law enforcement databases for a large matched
sample (N=965) of couples involved in the divorce process. We first linked data produced in the
naturalistic clinical interviews used to screen parents for marital stressors and IPA in custody and
parenting time (i.e., weekly visitation, holidays and vacations) mediation to questionnaire data
also measuring specific IPA-related behaviors. We then linked this IPA data to the mediator’s
decisions concerning (1) whether to identify a case as having IPA or not; (2) whether to proceed
in mediation or to screen out IPA-identified cases; and (3) whether to provide special procedural
accommodations for IPA-identified cases.
We then linked the IPA data and mediator decision data to mediation outcome data from
in mediation case files and to outcomes in final divorce decrees and parenting plans found in
Superior Court divorce files. We then linked these pre-divorce and divorce data to post-divorce,
longitudinal data concerning re-litigation of divorce-related issues in Superior Court and
longitudinal data concerning contacts with area law enforcement.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 13 2007-WG-BX-0028
Based on previous work concerning IPA screening in mediation (Administrative Office
of the Courts [California], 2010, Ellis, 2006; Ellis & Stuckless, 2008; Newmark et al., 1995;
Pearson, 1997), we believe that mediation screening for IPA at the study site far exceeds the
standard for mediator screening practices employed in most jurisdictions. Screening at this site
included both a semi-structured clinical interview with each parent individually and a 41-item
behaviorally specific questionnaire covering a wide range of IPA-related behaviors. Therefore,
this study represents a “best case scenario” for screening practices and responding to IPA in the
mediation context at the time the study was conducted and likely currently as well.
Results and Conclusions
Goal 1: Determine whether the mediation program accurately identifies couples with
self-identified IPA and assess whether these cases are treated differently.
To put the findings regarding accurate identification by mediators in perspective, we first
assessed the total percentage and types of IPA reported in the sample and important demographic
differences between fathers and mothers.
A tremendous amount of IPA was reported by both parents in mediation. From the
data obtained in screening interviews with each parent individually, mediators identified 59%
(N=569) cases as having IPA. Analyzing data from the 41-item IPA screening questionnaire of
behaviorally-specific questions assessing a range of IPA-related behaviors that occurred in the
past 12 months, responses were categorized into the following subscale dimensions:
psychological abuse, coercive controlling behaviors, physical abuse, threatened and escalated
physical violence, and sexual intimidation/coercion/assault. Analysis of this client self-reported
data indicated that: 97% of both mothers and fathers reported at least one incident of
psychological abuse, coercive controlling behaviors; 58% of the mothers and 54% of the fathers
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 14 2007-WG-BX-0028
reported at least one incident of physical abuse; 62% of the mothers and 50% of the fathers
reported at least one incident of threatened and escalated physical violence; 56% of the mothers
and 30% of the fathers reported at least one incident of sexual intimidation/coercion/assault.
There were statistically significant differences between mothers’ and fathers’ reports on all
subscale dimensions with mothers reporting higher levels, except on the physical abuse (lower
levels of pushing and shoving) dimension where there was no difference between the parents’
reports.
Important disparities exist in incomes between divorcing parents. Upon entering
mediation, the mothers’ incomes in this study were approximately half of the fathers’ incomes.
One potential interpretation of this finding is that that mediation about custody and child support
is not an “even playing field” for both parties even before the issue of abuse is raised. This
marked sex difference in income graphically illustrates a highly gendered pattern in child care
and the patterns reflected in custody arrangements following separation. These patterns are often
inaccurately attributed to “gender bias” against men as opposed to reflecting the patterns
established in the family prior to divorce.
IPA does not end at separation and divorce. An important finding that is well
supported by the extant research on separation and divorce is that for some couples IPA did not
stop at separation or divorce. For a minority of couples, calls to police, orders of protection, and
court hearings regarding IPA were reported after separation and after the divorce was finalized.
Again, it is important that any agreements between the couples address the risk of continued
IPA.
An analysis of couple-level IPA data produced a five class typology; four classes
represented a clear victim and perpetrator.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 15 2007-WG-BX-0028
With the assistance of our colleague Dr. Edward Anderson from the University of Texas
at Austin, we investigated whether the data in this study would produce couple-level types of
IPA similar to those proposed in by Kelly and Johnson (2008) for this population. Extremely
simple definitions of Kelly and Johnson’s typology are as follows:
1. Coercive Controlling Violence (male perpetrated battering of female partner)
2. Violent Resistance (female partner resisting a coercive controlling male perpetrator)
3. Situational Couple Violence (symmetric low level physical abuse)
4. Separation-Instigated Violence (no prior history; either partner perpetrating)
5. Mutual Violent Control (both partners perpetrate coercive controlling violence)
Because the data in this study was extant archival court records, the long-term patterns of
IPA within couples could not be determined. Therefore, the categories of violent resistance and
separation-instigated violence were not able to be adequately assessed. We conducted a latent
class analysis of the 41-item, behaviorally specific questionnaire data using a pooled mother and
father mean (as opposed to gender-specific means) so that unit weights were equivalent across
parents. This analysis produced five types. Keeping with the definitions above and beginning
with the group reporting the lowest levels of IPA to the highest levels of IPA, the types were as
follows:
1. Mutually low (or no) IPA.
2. Lower Level Coercive Controlling Violence—father perpetrator (mother reports)
3. Coercive Controlling Violence—mother perpetrator (father reports)
4. Coercive Controlling Violence—father perpetrator (mother reports)
5. Mutual Violent Control—(father perpetrator (mother reports); mother perpetrator
(father reports).
A detailed explanation of these types is provided within the text of the report.
Mediators identified many, but not all, client self-identified cases of IPA. Mediators
used clinical judgment to sort cases and did not designate every case that reports any level of
IPA as one with IPA. However, there were a small number of cases that were classified as high
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 16 2007-WG-BX-0028
level IPA that were not identified by the mediators as such. One third of the couples classified as
non-IPA by the mediator reported at least one incident of threatened and escalated violence or
sexual intimidation/coercion/assault. For those fathers and mothers completing the behaviorally
specific questionnaire and combining all dimensions for a total score for each parent, 49 mothers
(n=887; 6%) and 37 (n=866; 4%) fathers reported victimization that was frequent and/or severe
enough to categorize their abuse as two standard deviations above the sample mean. Mediators
classified as IPA all of these cases as having IPA except six mother victim cases and two father
victim cases. Basing classification of cases on a nonsystematic, semi-structured clinical
interview data may not be the most effective method of assessing IPA. Research is mixed
concerning whether people find it easier to report IPA behavior in face-to-face interviews or on
written questionnaires. For the present, a prudent method may be to use both methods.
Even using both methods of assessment, mediators must be well trained in identifying
IPA to be in a position to make very difficult, nuanced decisions concerning proceeding,
accommodating or screening cases out of mediation.
Cases are rarely screened out of mediation. Seven percent of cases were screened out
of mediation. There was a statistically significant difference in mother but not father total IPA
scores of those screened out and those that remained in mediation. For those parents with IPA
scores two standard deviations above the mean, 25% of the mothers and 9% of the fathers were
screened out.
Mediators must strike a delicate balance between empowering parents to represent their
own interests in mediation sessions and determining the best method of protecting victims and
violence-exposed children. The choice is simple if the victim does not want to proceed; the
decision becomes more complex if the victim wants to move ahead. Allowing a victim a chance
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 17 2007-WG-BX-0028
to proceed gives the victim a safe forum to have a voice with her or his abuser and provides the
mediator a chance to observe the dynamics of the couple and to evaluate the fairness of the
process. And, screening cases out of mediation immediately after the individual interview may
be risky. The mediator terminating mediation after the initial session may better protect victims
than screening them out initially.
Special mediation procedures (safety accommodations) were provided most often
for parents who telephoned the mediation service expressing safety concerns or requesting
accommodations. Accommodations to the mediation procedure included: parents required to
leave at separate times; security escort to car; separate waiting rooms; screening on different
days; shuttle mediation, and an experienced team of mediators. Approximately 19% of the
couples were provided at least one of these procedural accommodations. Accommodations were
provided for couples with higher levels of IPA; however, fewer than half of the couples reporting
IPA at levels two standard deviations above the mean were provided accommodations.
Mediators were more likely to provide accommodations if they had identified IPA present (28%)
versus absent (6%). Mediators were statistically significantly more likely to provide
accommodations or to screen the couple out if a parent phoned the mediation service ahead of
time with concerns about mediating or requesting accommodations (84%; N=31).
There are important issues in considering these findings. First, screening for IPA was
ongoing throughout the mediation and if it appeared accommodations were in order they were
provided. Second, the data concerning accommodations likely represents a serious underestimate
of actual accommodations provided. There was no systematic documentation of accommodations
for the study site kept in mediation case files. And third, advocates working with victims
involved in the divorce process should remind their clients to telephone the mediation service if
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 18 2007-WG-BX-0028
they have concerns about mediating. Mediators in this study were extremely responsive to these
calls. However, even without calls, accommodations were consistently provided.
Calls to area law enforcement and orders of protection are common. Couples in the
study entered mediation between May, 1998 and October, 2000 and data was collected from area
law enforcement for these cases beginning two years prior to filing for divorce up February,
2007. In nearly 40% of couples, at least one parent telephoned area law enforcement at least once
during the study period. The range in the number of calls was 0-21. In addition, 42% of the
couples in the study reported receiving an order of protection at some point in the relationship.
There were high levels of agreement between parents concerning whether law enforcement had
been called, whether arrests had been made or whether orders of protection were in place at the
time. Those couples with the highest number of calls to law enforcement fell into the mediator
IPA-identified group as did those couples with orders of protection.
Data concerning calls to area law enforcement also likely represent a serious
underestimate of the actual calls made by the couples. Data was collected from the two law
enforcement agencies serving the location directly surrounding the study site. Any calls placed to
agencies beyond these were not included. In addition, only calls placed within the limited study
period were counted.
Goal 2: Assess whether mediation agreements, divorce decrees and/or parenting plans
include safety measures in cases involving self-reported IPA.
Levels of reported IPA influenced whether couple were likely to reach agreements.
In mediation, 56% of the couples reached a full agreement on all issues (legal decision-making
custody; physical custody and residential arrangements, holiday and vacation schedules); 22%
reached a partial agreements, and 22% failed to reach an agreement. Those couples who failed to
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 19 2007-WG-BX-0028
reach agreement had the highest levels of reported IPA, followed by those that reached partial
agreements. Those who reached full agreement reported the lowest IPA scores.
Mediation agreements and divorce decrees rarely have restrictions on parenting
(e.g., supervised parenting time) or on contact between parents (e.g., written contact only).
Of the couples who reached any kind of mediation agreement (partial or full) in mediation, 6%
included some type of restriction on contact between parents or restriction on parenting. Cases
that had restrictions reported higher levels of IPA. However, for those cases with the highest
levels of IPA reported (two standard deviations above the mean) there were no restrictions
placed on 9 cases (out of 19 cases identified in that range). In terms of decrees, 12% contained
restrictions. Cases with restrictions in place tended to have higher levels of reported IPA. In
addition, a greater percentage of cases at the highest levels of IPA (two standard deviations or
higher) had restrictions, as compared to those at the lower levels.
While restrictions in mediation agreements were rare, it appears that victims with the
highest levels of IPA are typically leaving mediation without agreements. These couples returned
to court and obtained restrictions at a much higher rate than those found in couples who reached
mediation agreements. There are two important issues that relate to restrictions. First, it is
important for mediators, advocates and lawyers working with victims to remind victims that, in
most jurisdictions, they are required to attend the orientation screening sessions and negotiate in
good faith, but they are not required to reach an agreement in mediation. And second, the
limitations of the mediation process must understood and acknowledged. Mediators are not
judges and cannot require abusers agree to or abide by restrictions on contact between parents or
restrictions on parenting even if it would be in the best interests of the children to do so. If the
goal is having appropriate restrictions in mediation agreements for couples with IPA then use of
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 20 2007-WG-BX-0028
other processes are likely necessary, such as a hybrid mediation/arbitration intervention which
provides the arbitrator with legal authority and control over the terms of agreements.
Investigations of hybrid mediation/arbitration interventions have been conducted over the
past two decades (Colon, Moon, & Yee Ng, 2002; Loschelder, & Trotschel, 2010;
McGullicuddy, Welton, & Pruitt, 1987; Ross, Brantmeier, & Ciriacks, 2002). The essential
elements of these interventions are that clients to a mediation attempt to mediate an agreement. If
the they do not reach an agreement, the mediator then becomes an arbitrator and issues a binding
decision (med/arb(same)) or the mediators steps aside and a new person becomes the arbitrator
and issues a binding decision (med/arb(diff)). These models were empirically investigated in a
community mediation center (McGillicuddy et al., 1987). When compared to traditional
mediation (if no agreement clients return to court) and med/arb(diff), mediation clients in the
med/arb(same) condition engaged in more problem solving and were less hostile and
competitive. The important aspect of these interventions in the high-IPA divorce mediation
context is that if a couple is determined to have IPA in the relationship and the abuser will not
agree to safety precautions in the mediation agreement, the arbitrator has the authority to issue a
binding decision which includes these terms. Prior to adoption of these models it is extremely
important that they be carefully empirically investigated and that the safety of victims in the
short- and long-term are the uppermost consideration.
Legal findings of “domestic violence” were rare. Only 2% of the cases in the sample
had a formal legal finding labeled domestic violence in the divorce file. Couples with a legal
finding of domestic violence reported higher overall IPA scores compared to those who did not
have the legal finding. However, among couples reporting the highest levels of IPA (greater than
two standard deviations above the mean), only five couples had this finding. Because it is
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 21 2007-WG-BX-0028
necessary for there to be a clear victim and perpetrator, none of the seven couples who reported
high mutual IPA received a legal finding of domestic violence. It is unclear why this legal
remedy is not being used more frequently. It could be because the evidence required to meet the
legal burden is too difficult to obtain or is too risky for the victims to pursue.
The most common parenting agreement in divorce decrees, regardless of reported
level of IPA, was mother as the primary physical custodian with joint legal custody.
Mothers were awarded primary custody (physical and/or legal) in 70% of the cases; fathers were
awarded primary custody in 9% of the cases and joint custody was awarded in 21% of the cases.
In comparing decree outcomes for those who reached full agreement in mediation to those who
failed to reach agreement in mediation, mothers in full agreement cases were awarded primary
custody in 69% (full agreement) versus 71% (no agreement ) cases. Fathers were awarded
primary custody in 8% (full agreement) versus 11% (no agreement) cases. Joint custody was
awarded in 23% (full agreement) versus 18% (no agreement) cases. The lowest level of reported
IPA was found in joint custody cases.
Interestingly, those couples falling in the Mutually Violent Control type for which
decrees were entered (N=26), mothers were awarded primary custody in 22 cases (85%); joint
custody was awarded in three cases (11%); and, fathers were awarded primary physical custody
and/or legal custody in one case (4%).
Goal 3: Assess the frequency with which mediation agreements/divorce decrees are
re-litigated over time.
The majority of parents return to court to re-litigate divorce issues; however, there
is a small subset of “frequent flyers.” Sixty-two percent of the couples returned to court for at
least one hearing after the divorce was finalized. The range in the number of hearings was 0-31
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 22 2007-WG-BX-0028
hearings. Approximately half of those couples who did return returned between 1-5 times (52%).
There was a subset of couples (4.5%; N=31 who accounted for 31% of the total number of
hearings requested in the sample. In the last decade, court-based programs have been developed
targeting this group of couples. It is important that continued efforts to validate the effectiveness
of these programs be supported.
Parents reaching agreement in mediation are significantly less significantly likely to
re-litigate divorce-related issues post-divorce. Those couples who reached full agreement in
mediation were significantly less likely to return to court to re-litigate aspects of their divorce.
This finding is particularly important for policy-makers considering whether to develop (or fund)
mediation programs. In this study, well-trained mediators were given a wide latitude to provide
the number of sessions necessary for couples to successfully negotiate an agreement, thus using
fewer court resources post-divorce.
Goal 4 Results: Test a multivariate conceptual model using variables that are
hypothesized to affect mediation, divorce case and post-divorce outcomes (see Cascade
Model Figure 16
To better understand the dynamic nature of the divorce process and the many interrelated
variables, we used a multivariate analytic approach. We began by aggregating theoretically
relevant constructs across an entire system to guide our inquiry into those that lead to outcomes
associated with mediation, the divorce and the post-divorce legal processes. We then created a
theoretically-specified causal order for the constructs and then ran a series of multiple regression
models in the context of a Cascade Model (Figure 16). Specific hypotheses were tested relating
to the relationship between the constructs, which were categorized into three temporally related
sets of variables. The first set included variables that occurred prior to mediation (marital
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 23 2007-WG-BX-0028
stressors, father and mother perpetrated coercive controlling behaviors and IPA); the next set of
variables was those that occurred as a result of mediation (mediator identification of IPA,
mediator providing special procedures/accommodations, and whether an agreement was
reached); the third set of variables included those that occurred as a result of the final divorce
(court finding of IPA, restrictions in decree on the abusing parent, order of protection, child
maltreatment and custody decree favoring mothers). The final set of variables related to post-
decree outcomes that were possible after the divorce process (post decree hearings, orders and
calls to law enforcement). Selected findings that extend findings noted above will be reviewed
below as the model.
Decades of research have documented the emotional chaos and financial hardship
experienced by families going through a divorce. Results indicate that marital stress influenced
mothers’ and fathers’ perpetration of IPA; however, it also influences fathers’ use of IPA and
court orders placing restrictions on contact between parents and on parenting in divorce decrees.
Marital stressors negatively influenced custody agreements favoring mothers. Although the
majority of mothers are awarded primary legal and/or physical custody, when mothers’ marital
stressors are higher, the courts appear to respond by more frequently awarding fathers custody.
IPA is defined as the mean unit weighted total scores for reported physical abuse,
threatened and escalated physical violence; sexual intimidation/coercion/ assault. IPA
influenced many constructs in the model. Fathers’ and mothers’ perpetration of IPA influenced
the mediator’s identification of a case as having IPA; however, beyond these relationships,
mothers’ and fathers’ IPA influenced downstream constructs in different ways.
When fathers’ perpetrate higher levels of IPA, the couple is less likely to reach an
agreement in mediation, and .there is a significantly greater chance that an order of protection
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 24 2007-WG-BX-0028
will be issued. There is also a greater chance that restrictions will be included in divorce decrees
and that custody agreements will favor mothers. When mothers perpetrate higher levels of IPA,
courts are more likely to issue a legal finding of domestic violence in the case, and there will
likely an order of protection and child maltreatment reports in the case. It appears that in mother-
perpetrator cases, fathers take the legal step of obtaining a legal finding of domestic violence and
orders of protection for himself and the children.
Child maltreatment also emerged with interesting relationships to mothers’ perpetration
of coercive controlling behaviors, IPA, divorce and post-divorce outcomes. It appears that when
mothers’ perpetration of coercive controlling behaviors is higher fewer reports of child
maltreatment are made. However, mothers’ perpetration of IPA is associated with high reports of
child maltreatment. The chance for a custody award favoring the mother is then significantly
reduced and there are more frequent post-divorce calls to law enforcement.
Coercive Controlling Behaviors emerged as an important influence on downstream
divorce and post-divorce constructs both directly and indirectly through IPA total scores. In a
prior analysis of this data we hypothesized, and then confirmed using structural equations
modeling, that control is the motivator for other dimensions of IPA (Tanha et al., 2009). In other
words, when coercive controlling behaviors fail to control the other parent, physical dimensions
of IPA (physical abuse, threatened and escalated physical violence; sexual
intimidation/coercion/ assault) are then enlisted. Thus, it appears that when mothers’ and fathers’
attempts to control the other parent failed, both tended to escalate into other, more severe forms
of IPA.
However, looking at the relationships of coercive controlling behaviors to other
important variables, the use of coercive controlling behaviors by mothers against fathers
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 25 2007-WG-BX-0028
additionally influenced other constructs in the model. The higher the levels of mothers’ coercive
controlling behaviors, the lower the chance of a child maltreatment report, order of protection or
an agreement in mediation.
In summary, one reason for conflicting findings with respect to screening,
documentation, triage, and monitoring of IPA cases in mediation in the literature is likely the
lack of consensus on how to identify IPA; what to do about it when it is detected; and agency
policy differences on how to handle IPA cases. The results of this study provide strong
empirical support for previous estimates that most couples attending divorce mediation report
some level of IPA. Given that couples mandated to mediation are there precisely because they
cannot agree on significant aspects of the divorce, this result was expected. Few mothers and
fathers reported no IPA of any kind (only four couples reported that there were no incidents of
any sort of IPA in the preceding 12 months prior to mediation). The parents reported at least
some form of IPA occurred in over 90% of the cases and two thirds of the couples reported that
either or both partners utilized outside agency involvement from police, shelters, courts, or
hospitals to handle the IPA. These figures represent a tremendous amount of IPA in couples
mandated to attend mediation and that the level was likely more serious that “common couple
violence” identified in previous national random sample surveys (Kelly & Johnson, 2008).
Limitations
Archival data from court records and area law enforcement records are an unobtrusive
and efficient means of gathering important data. There are, however, limitations to using these
type of official records to collect data. The most important limitation is that the data available is
limited to what the records contain. Records are often incomplete, files can be misplaced and
documents intended for inclusion in the files can be lost or misfiled. The quality of the records is
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 26 2007-WG-BX-0028
not under the control of the researcher (Jewkes, 2008). Access to the records can be limited and
refusal to allow access or provide certain portions of the data can occur. More specific to this
study, these data are from only divorcing couples who had disputes about custody and/or
parenting time of their children from one jurisdiction and one court-sponsored, free mediation
program. There is no way to empirically identify which couples chose this program over private,
fee-for-service mediator or mediation program. It is likely that those couples with the most
disposable income did so. Because this is a study specifically of divorcing couples with children
referred to mediation and not all divorcing couples, couples without children and/or those
couples with children but who were able to work out agreements concerning custody and/or
parenting time were also not included in this study. Because of the specifics of this sample, there
are limits to the generalizability of the findings.
It is important to note several issues relevant to the screening procedure and the use of the
screening forms by the mediators. The semi-structured interview data used in this study represent
the individual determinations by mediators as to what seemed important enough to note in order
for them to determine the best course of action for the case. In addition, the behaviorally-specific
questionnaire included in the file (the RBRS) assessed IPA in the prior 12 months only. Thus, the
RBRS did not capture the full extent of IPA that may have existed in the entire relationship nor
did it document ongoing IPA ongoing in the relationship. The RBRS also did not assess for who
the primary or predominant aggressor was or the relationship context in which any of these
behaviors occurred. Thus, whether the acts were self-defense or abusive/violent acts cannot be
distinguished. Future research needs to include structured, standardized, comprehensive
assessment measures of IPA in addition to in-depth interviews with the clients to gain a full
understanding of the the specific context and impact of the violence in the family. It is extremely
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 27 2007-WG-BX-0028
important for future research that a clear picture of the history of IPA in the relationship can be
understood and the victim and abuser can be correctly identified so that the court system can
respond appropriately to protect the victim and the violence-exposed children.
As noted above, the data in this study concerning accommodations provided to couples in
mediation is likely a serious underestimates of the actual number of accommodations provided.
This study was a naturalistic evaluation of one county’s practices therefore data used in this
study was not designed for specifically research purposes but instead to conduct day-to-day
operations in the mediation program. While some accommodations provided clients were
documented, not all of the accommodations were clearly documented in the files. There has been
two decades of scholarly discussion of the possibility of special procedures used in mediation to
accommodate couples with IPA, but there has yet been no systematic empirical research
investigating what accommodations are provided and when they are provided and why they are
provided. This study represents one of the first attempts to document the use of special
accommodations for couples with IPA. Future prospective research is needed which carefully
defines special procedures/accommodations, assigns couples to specific accommodations based
on hypothesized needs, then documents defined outcomes associated with the couples in the
different accommodation groups (Ellis, 2008). This research would be helpful in understanding
how accommodations affect case processing and case outcomes.
Data in this study concerning law enforcement is also likely a serious underestimate of all
the calls placed to law enforcement by couples in the study. Because only local area law
enforcement (police and sheriff) calls were gathered in this study, calls placed to jurisdictions
outside the local area were not captured.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 28 2007-WG-BX-0028
Despite the limitations of this naturalistic study, these data are unprecedented in the field
of divorce mediation research. To date, virtually all of the published samples used to study
mediation included small samples, limited data, or low response rates (Beck & Sales, 2001). In
this study, the sample included all parents (N=965 couples) who were disputing custody and
parenting time issues in a pending divorce and were mandated by a local court rule to attend
mediation during the targeted period and did so for a first-time mediation through a court-
sponsored, free mediation program.
Recommendations for Research, Policy and Practice
(1) Provide essential training in the dynamics of IPA
Mediators, judges and court staff working directly with potential victims of IPA need to
be well trained in the dynamics of IPA and how it may manifest the context of a divorce.
(2) Use a structured, systematic approach to assessment of IPA in mediation
Using a structured, systematic, multi-method approach is likely to be considerably more
effective than using only a semi-structured clinical interview in reducing the rate of false
negatives in the IPA screening process. We are strongly recommending that mediators adopt a
standardized approach to administration of both semi-structured clinical interviews and
standardized questionnaires and ask all questions about IPA of all parents.
Mediation practitioners, scholars and violence experts also need to work together to
continue developing a more reliable, valid, approach to screening for all dimensions of IPA in
mediation including stalking. Because coercive controlling behaviors were important indicators
separate from the other dimensions of IPA, continued focus on developing adequate screening
measures for both mothers and fathers is needed.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 29 2007-WG-BX-0028
(3) Automate screening measures
Development of an automated version of a comprehensive IPA screening measure that
quickly combines data from both members of a couple for mediator review and decision-making
is critical for an accurate assessment of IPA. Although assessment can be automated relatively
easily, the complex decisions mediators must make for a particular couple in determining
whether to move forward with mediation or not, or to provide an accommodation or not, requires
careful thought and analysis of the benefits and risks in consultation with the reported victim.
(4) Assess Mediation special procedures to accommodate victims
Prospective research using systematic and structured assessment of IPA which clearly
documents the reasons for dispositions of cases with documented IPA histories is needed so that
effective policies to screen clients out or provide accommodations can be developed. An
accommodation which might be helpful to empirically investigate for reported victims of IPA
wishing to continue in mediation is for mediators to pre-arrange with the victim a sign (e.g., hand
signal or word) that the victim can use in the event the she or he feels unsafe or afraid. If used the
mediator will know to stop and possibly caucus with the parents individually to insure the victim
wishes to continue. Some abusers convey threats only the victim will understand and it is
important for a victim to be able to signal a mediator when this occurs. Perhaps researchers
working with practitioners can design a study where mediators are included in developing a
standardized set of accommodation practices along with hypotheses that identify which
accommodations are best implemented in cases involving different types of IPA. Carefully
defined and collected outcomes can then measure the effectiveness of the accommodations
provided, thereby empirically investigating if the process can be improved and appropriate
accommodations instituted across cases.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 30 2007-WG-BX-0028
Mediation program personnel, lawyers and advocates working with potential victims
should strongly encourage clients who are concerned about IPA or who want specific
accommodations in the mediation process to call and request such accommodations or express
concerns about mediating.
(5) Acknowledge the limitations of the mediation process
Mediators are not judges and cannot order that certain conditions to be included in
agreements (such as protections for children or restrictions on contact between parents) or that
the conditions be followed even if such requirements would be in the best interests of the
children or the adults. In the majority of jurisdictions across the United States, mediators instead
facilitate negotiations between parents and each parent has a right to decide what they will and
will not agree to. If both parents do not agree to a specific condition, it will not be included in
any resulting agreement. If parents do not agree on any issues then no parenting agreement will
result. Holding mediators and mediation programs responsible for parenting agreements, freely
negotiated between the parties, that do not include safety precautions flies in the face of what
mediation in this and many other jurisdictions are designed to accomplish. If a mediator believes
that one of the parties cannot adequately represent their interests or the agreement negotiated
between the parents will not be in the best interests of the child[ren] is terminate mediation, the
resource the mediator has is to terminate mediation without an agreement. This decision is
difficult, complex and must be made carefully and in consultation with the party having trouble
representing her or his interests.
(6) Investigate hybrid mediation/arbitration mediation interventions
If policy-makers wish to ensure that couples with IPA negotiate mediation agreements
which include safety restrictions, they must consider empirically investigating a separate process
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 31 2007-WG-BX-0028
for these couples. The hybrid mediation/arbitration intervention may be particularly important to
empirically investigate with pro se couples with high IPA. Currently, victims without legal
representation who are unable to negotiate safe parenting plans in mediation currently must
return to court and present her or his case to the judge.
(7) Encourage legal findings of “domestic violence” in appropriate cases
The results of this study indicate that a parent with a specific legal finding of domestic
violence in their case is more likely to have safety restrictions on contact between abuser and
victim and on parenting (e.g., orders of protection; orders restricting method of contact between
parents; neutral exchanges of the children at a specific public location, a family member’s home,
or a judicially approved service provider) in the divorce decrees. Lawyers, advocates for victims
and mediators should carefully consider, in consultation with victims, recommending that
victims obtain court orders with legal findings of domestic violence. This legal finding also
provides important information to subsequent judges assigned to the case.
(8) Assess outcomes for children directly and prospectively
Children are often caught in the cross-fire of parental IPA and so the co-occurrence of
IPA with child maltreatment is high. Broader assessment and evaluation of child maltreatment
and IPA (looking beyond physical forms of both) is essential to truly understand if mediation or
any other court program is benefitting children. Psychological abuse can be just as devastating as
other forms of abuse (Coker, Smith, Bethea, King, & McKeown, 2000; O’Leary, 1999; Stark,
2007; Theran, Sullivan, Bogat, & Stewart, 2006). Because virtually all cases in the study
involved some form of psychological abuse, mediators should pay particular attention to
assisting the parents to understand and remediate the impact of such abuse on the members of the
family, particularly the children.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 32 2007-WG-BX-0028
(9) Assess the effectiveness of court-based programs for pre- and post-divorce
“frequent flyers”
A small group of families were found to require a tremendous number of court hearings
post-divorce. As a result, parenting coordinator programs are developing rapidly across the
United States with very little empirical basis of support. Courts need to empirically evaluate the
effectiveness of court-based parenting coordinator and case management processes to determine
whether these programs are having the desired benefits for families.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 33 2007-WG-BX-0028
Introduction
Statement of the Problem.
Despite three decades of research on numerous aspects of divorce mediation, there is still
no comprehensive understanding of the short- and long-term outcomes for families legally
ordered to mediation to resolve custody and parenting time disputes and who use the free (or low
cost) conciliation court mediation services to do so. This is particularly true for those families
alleging intimate partner abuse (IPA) (Beck & Sales, 2001; Salem, 2009). In 2007, the National
Institute of Justice funded this project to fill this gap in understanding.
Review of Relevant Research and Background of Study
What is mediation in this context?
Mediation is defined as a task-oriented, time-limited, alternative dispute resolution
process (Folberg, 1982). It is important to remember that mediation does not refer to just one
process. The practice can vary along several important dimensions (Beck & Sales, 2001). For
example, models of mediation can vary significantly in terms of the mediators’ role. Mediators
can be either neutral facilitators of a process or can be active participants who, if the parties do
not reach an agreement, can make recommendations to courts regarding specific terms of
agreements. The means by which cases are referred to mediation can vary from a court merely
mentioning that mediation is a possibility to a legal mandate requiring that all cases involving a
dispute over particular issues be mediated. The criteria for excluding cases may also vary from
no cases being excluded, to all cases with IPA being excluded. In some states only issues of
custody and parenting time are mediated while in other states all issues related to a divorce are
mediated (Beck & Sales, 2001). Although the process has changed significantly over the last
decade (Salem, 2007; Saposnek, 2004), at the time this study was conducted there were some
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 34 2007-WG-BX-0028
important similarities among court-sponsored, mandatory mediation programs focusing on
custody and parenting time issues that were conducted by mental health-trained mediators (Beck
& Sales, 2001).
Mediation was originally offered as an alternative to the traditional dual attorney,
adversarial model of representation and litigation. The process of mediation typically began with
the parties meeting together in an office separate from courtrooms, as it was intended to be less
formal than a courtroom. A mediator (or sometimes a mediator team) served as a neutral who
would allow the parties to “tell their stories.” This process provided the “day in court” that was
believed to be missing for many litigants who had previously used the traditional adversarial
process (Beck & Sales, 2001).
The mediator’s role was to assist the parents in isolating those issues that were in dispute
and considering options and alternatives. In doing so, the mediator’s responsibility was to
empower the parties, encouraging them to negotiate and compromise so that they could make
decisions in the best interest of their families. Mediation operated under the assumption that a
highly skilled mediator would attempt to maintain a fair and balanced dispute resolution process
as they assisted the parties to explore and work through various options. The mediator’s role was
not to dictate solutions or pressure couples into developing settlement agreements, but rather to
facilitate a balanced, fair, decision-making process, in which both parties’ stories were heard and
a fair agreement could be reached. By allowing the parties to “tell their stories” and providing a
neutral, less adversarial process for parental self-determination, it was believed the mediation
process would provide not only access to justice, which is felt to be missing in traditional
litigation, but a better, more self-determined model justice for couples as well (Beck & Sales,
2001).
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 35 2007-WG-BX-0028
The eventual goal of mediation was to reach a consensual settlement agreement regarding
the disputed issues. A fundamental tenet of mediation was the belief that if the parties worked
out their own settlement agreement they would be more satisfied and consequently, more likely
to abide by the agreement in the long-term. Interestingly, it was (and still is) strongly believed by
some that even participating in the mediation process without reaching an agreement provided
significant benefits to couples (Edwards, Baron, & Ferrick, 2008; Ferrick, Baron, & Edwards,
2008; Folberg & Taylor, 1984). Note that the research and scholarship is mixed on supporting
this assumption (Schraufnagel & Li, 2010). Although there was some support (Walker, 2010),
more recent research challenges this assumption (Dingwall, 2010).
With increasing interest in mediation from scholars and practitioners, and the possibility
of increased compliance with custody agreements, an important issue for the courts was how to
provide mediation services at low (or no) cost for those who had been historically excluded from
the court system (lower socioeconomic groups, women and minorities). Couples separating and
struggling financially would likely be less able to afford an additional legal process to obtain a
parenting plan as well as the divorce. The question courts were asking was who would be best
situated to provide mediation services, particularly for clients of low socioeconomic means?
Why did conciliation courts become the main services providers of mandated divorce
mediation?
Prior to the 1970s, conciliation courts were developed and staffed by mental health
professionals to provide “conciliation” (marital) counseling with the goal of resolving conflicts
and encouraging parents to remain married. The 1970s shift toward no-fault divorce mirrored a
simultaneous cultural change in focus away from trying to keep couples married to helping
families amicably resolve divorce disputes. These changes provided the courts with an important
reason to keep conciliation courts intact and provided mental health professionals at conciliation
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 36 2007-WG-BX-0028
courts with an opportunity to maintain employment by expanding their professional services
(Morrill, 2008). Conciliation court staffs were already employees of the court trained in
providing therapy and interventions directed toward conflict resolution. Judges who wanted these
cases out of their courtrooms began experimenting with mediation by referring particularly
difficult, high conflict cases to conciliation court therapists with directions to assist the couples
in reaching an agreement. Over time, conciliation court mental health practitioners then became
the logical service providers of the newly developing mediation intervention (Morrill, 2008).
The zeal with which the proponents embraced mediation led to an arrayof changes in how
divorce disputes were processed nationwide, and, now, around the world (Dingwall, 2010;
Morrill, 2008). Historically, conciliation courts conduct the vast majority of custody and
parenting time (visitation) mediations in major urban areas across the United States, particularly
for those clients of modest economic means (Pearson, Ring, & Milne, 1983). This is particularly
true in states and/or jurisdictions within states wherein it is legally mandated for couples to
participate in mediation when disputing custody and parenting time of their children. In the
United States, the pattern of conciliation courts providing mediation services has continued. In
the current study, participants were mandated by a local court rule of practice (Pima County
Super. Ct. Local Rule 8.7, 1998) to attempt to mediate their dispute, and they were referred to the
local conciliation court for services.
Why is mediation often legally mandatory?
Although professionals were enthusiastic about the potential benefits of mediation, both
private programs and free, public, court-based programs had ongoing difficulties attracting
clients. Under a voluntary scheme, many couples refused to try mediation when it was offered
(Pearson, Thoennes, & Vanderkooi, 1982). In fact, one court-based mediation service instituted
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 37 2007-WG-BX-0028
mandatory diversion to the program in order to generate enough cases to conduct an early study
concerning the impact of mediation on disputing couples and their children (Cohen, 1982;
Pearson et al., 1983). This is consistent with what is currently occurring in the United Kingdom
(Dingwall, 2010) and Australia (Rhoades, 2010). A very small proportion of clients who are
provided information about mediation was willing to actually try it (National Audit Office, 2007;
Walker, 2001).
In the United States, based on limited initial empirical evidence, strongly held beliefs that
mediation was likely to lead to better outcomes for children, and the knowledge that most clients
would not voluntarily attend mediation when it was offered, many states and local jurisdictions
passed legislation mandating participation in mediation for certain classes of cases (custody or
parenting time disputes). The first mandatory mediation statute was passed in 1981 in California.
In the last three decades states have increasingly moved to mandatory statutory schemes for
custody and parenting time disputes. In the most recent review of divorce mediation statutes, it
was found that mediation in some form (mandated by law, at judicial discretion, or voluntary)
now exists in nearly every state in the United States (Murphy & Rubinson, 2005; Tondo,
Coronel, & Drucker, 2001). Prior to setting a date for trial, 12 states mandate that couples attend
mediation if there is a dispute about custody and/or parenting time of their children (Johnson,
Saccuzzo, & Koen, 2005). In another 33 states, judges are granted the power to order (mandate)
couples to attend mediation. In the remainder of the states, there exists a variety of local court
rules of practice including some mandating attendance, some leaving the decision to judicial
discretion, and others offering mediation on a voluntary basis only (Johnson et al., 2005). In
some states, all disputed divorce issues can be mediated (Alaska, Kansas), whereas other states
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 38 2007-WG-BX-0028
limit the issues that can be mediated to custody and parenting time only (Arizona, Nevada)
(Beck, Walsh, Mechanic & Taylor, 2010).
Over the last three decades, mediation has become embedded in the court system, in a
process often referred to as “institutionalized” (Morrill, 2008). Institutionalization in this context
means that mediation has become an integral part of the court process. As originally envisioned,
mediation was intended to be a voluntary alternative to court processes (Morrill, 2008; Salem,
2009; Saposnek, 2004). Today in many jurisdictions, disputing couples can no longer schedule a
court hearing without first either participating in mediation or attending the initial
orientation/screening session and requesting to be excused. In some jurisdiction, mediation is
also offered at no or very low cost by court-financed programs. As more and more couples are
now compelled to attempt this institutionalized process by law, one issue that remains hotly
debated is how to screen for, and what to do with, mediation clients with intimate partner abuse
(IPA) in the relationship. It is extremely important that courts do not place victims at further risk
of violence by forcing them to participate in a legal process that does not take into account
dangerous interpersonal dynamics. As noted above, participants in the current study were
mandated by law to attempt mediation.
What is intimate partner violence (IPA)?
IPA is a heterogeneous construct comprising several dimensions; a fact that has, at times
hampered efforts to measure it accurately (Ellis & Stuckless, 2006; Holtzworth-Munroe, 2005;
Jacobson & Gottman, 1998; Johnson, 2008; McCloskey, Figueredo, & Koss, 1995; Stark, 2007).
Early efforts to measure the construct identified three dimensions of “conflict tactics” couples in
abusive relationships engaged in: reasoning, verbal aggression and physical violence (Straus,
1979). Note that only two of these scales, verbal aggression and physical violence, were thought
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 39 2007-WG-BX-0028
to reflect IPA. Straus et al., (1996) later revised and refined the measurement of these dimensions
in order to inquire about both one’s victimization and perpetration of: physical assault,
psychological aggression; sexual coercion and physical injury; and non-aggressive tactics, such
as negotiation. As efforts to measure IPA continued to evolve, researchers began to
conceptualize IPA as a multidimensional construct rather than focusing primarily on physical
assault as its defining feature.
Today, the most commonly recognized dimensions of IPA include: psychological/
emotional abuse, physical aggression, sexual assault/coercion, and stalking. Researchers from
diverse disciplines have investigated several, if not all, of these dimensions of IPA in their work
(Campbell & Johnston, 1986; Johnson, 2006; Saltzman, Fanslow, McMahon, & Shelley, 1999;
Straus, Hamby, Boney-McCoy, & Sugarman, 1996). McCloskey, Figueredo and Koss, (1995)
found strong evidence for four dimensions of IPA: verbal, physical, sexual, and escalated (life-
threatening) physical violence. McCloskey and colleagues (1993) asked 28 behaviorally specific
questions pertaining to abuse over the past 12 months. Participants were recruited from three
sources: 1) a stratified sample of 67 women who had sought temporary residence at a battered
women’s shelter; (2) 103 community volunteers responding to posters displayed throughout the
community, that sought women volunteers who had been “physically abused by a partner” in the
past year; and, (3) a representative community sample consisting of 195 women recruited for a
“family study” screened only for the presence of children of a particular age group. Covariance
matrices were used to group the items into rationally-derived dimensions of IPA and then the
items were tested for convergent validity. Factors representing the different dimensions were
then tested for convergent validity using structural equations modeling software (EQS; Bentler,
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 40 2007-WG-BX-0028
1989). The factors representing various IPA dimensions were then constructed using
confirmatory factor analysis. Results supported the four-dimension factor structure of IPA.
In addition, McCloskey and her colleagues’ support for the four-dimension factor
structure was also found using two different reporters of IPA--the mothers’ and the children’s’
reports (partner against mother; partner against child). In later studies, these dimensions were
again empirically validated in cross-cultural samples from the United States, Spain and Mexico,
(Figueredo et al., 2001). These four dimensions of IPA also map onto standardized measures of
IPA (Straus et al., 1996). The dimensions of IPA used in the current study are those in the
McCloskey, et al. (1995) study with updated labels (psychological abuse, physical abuse,
threatened and escalated physical violence and sexual intimidation/coercion/assault) with the
addition of a fifth dimension coercive controlling behaviors (see below).
Substantiation of reports of IPA in Divorce Context.
Recently, research, court and social policies and screening practices for IPA have been
plagued by a contentious, impassioned debate concerning the veracity of reports of IPA
victimization in the context of divorce. On one side of the debate, victim advocates, certain legal
scholars and some social science researchers argue that if IPA is reported to attorneys, judges,
mediators and court personnel, the information is generally ignored, minimized or dismissed
(Bala, 2008; Cobb, 1997; Grillo, 1991; Hart, 1990; Johnson, 2005a; Johnson, 2005b). On the
other side of the debate, fathers’ rights advocates, some social science researchers, and some
legal scholars argue that false allegations of IPA are widespread and are generally committed by
mothers against fathers to gain unfair advantage in divorce cases (Baskerville, 2007; Crowley,
2008; Dutton, 2005a; Dutton, 2005b, Dutton, 2006; Dutton, Hamel, & Aaronson, 2010). Extant
empirical research has not supported this claim (Johnston et al., 2005; Thoennes & Tjaden,
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 41 2007-WG-BX-0028
1990). Women in custody disputes are no more likely to allege unsubstantiated IPA than are men
(Johnston et al., 2005). An important implication from this debate, however, is the necessity for
structured, comprehensive screening of IPA and the importance of independent evidence
supporting allegations that are made in the context of divorce. When investigators have reviewed
divorce cases, independent evidence corroborating IPA allegations in the divorce case files
(substantiation) is rarely found (Johnston et al., 2005; Moloney, 2008). To fill this gap in the
literature, the current study, reviewed mediation case files, divorce case files and area law
enforcement databases for independent evidence corroborating IPA allegations made by the
parties.
Sex differences in reports of IPA.
In addition to the impassioned debates concerning the veracity of self-reports of IPA and
the general lack of independent forensic evidence to support allegations of IPA in court files,
there is an additional debate in the literature concerning sex differences in IPA victimization.
There is strong empirical support for some basic findings concerning the nature of sex
differences in IPA perpetration and victimization (Kelly & Johnson, 2008; Salem & Dunford-
Jackson, 2008a; Salem & Dunford-Jackson, 2008b).
In the context of heterosexual relationships, both men and women perpetrate some forms
of IPA. Research indicates that both men and women can be aggressors against equally abusive,
less abusive and more abusive partners. Women, however, generally do not perpetrate at the
same rate nor does it result in the same severity of consequences. Both men and women can be
fearful and can be injured by their partners, but again, generally not with the same interpersonal
dynamics, frequency or consequences. Data suggest that particular dimensions of IPA often
referred to as ‘minor violence’ or ‘common couple violence’ (e.g., pushing and shoving) tend to
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 42 2007-WG-BX-0028
be perpetrated at roughly equal rates by men and women, whereas, other dimensions of IPA
(e.g., sexual abuse and assault, physical violence leading to injury) tend to be predominantly
male perpetrated. The extent of sex differences is also vociferously debated and will not be
reviewed here (Archer, 2000a; Archer, 2000b; Dutton, 2005a; Dutton, 2005b; Dutton, 2006;
Felson & Cares, 2005; Hamby, 2009; Hamby, 2005; Johnson, 2005a; Johnson, 2005b; O'Leary,
2000; Stark, 2007; White, Smith, Koss, & Figuerdo, 2000). An important issue arising from this
debate and for the current study was that both fathers and mothers married to each other were
carefully evaluated using the same instruments that measure IPA. Dyadic data of IPA of this type
has been rarely collected due to the significant risk for victims who report abusers’ behaviors in
ongoing relationships. Because the data was collected as part of a confidential, routine screening
for all divorcing couples in the context of mediation it thus provided safety for the victims and
invaluable information on relationship patterns of IPA for these couples.
In addition, victims most often experience multiple dimensions of IPA as opposed to
experiencing one dimension in isolation; however, with notable exceptions (Johnson, 2006;
O’Leary, 2000; Stark, 2006) these dimensions of behaviors have generally been studied as
separate dimensions (Saltzman et al., 1999). The current study assessed multiple dimensions of
IPA, their interrelatedness, and did so for both mothers and fathers that are legally married to
each other.
Why is mandatory mediation a potential problem for couples with IPA?
Although divorce mediation has been widely adopted, it is by no means seen as a panacea
for all divorcing couples (Beck & Sales, 2001; Cobb, 1997; Dingwall, 2010; Trinder, Firth, &
Jenks, 2010). In fact, strong arguments have been raised from practitioners and scholars from
many disciplines against mandatory implementation, particularly when IPA is alleged (Beck &
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 43 2007-WG-BX-0028
Sales, 2001; National Council of Juvenile and Family Court Judges, 2008; Grillo, 1991; Johnson
et al., 2005; Moloney, 2008). When mediation is mandated for a couple for which IPA has been
documented in the relationship, decades of research indicates that several concerns about IPA
should be paramount (Beck & Sales, 2001; Fischer, Vidmar, & Ellis, 1993b; Grillo, 1991; Hart,
1990; Maxwell, 1999; Thoennes, Salem, & Pearson, 1995):
1) Can mediators accurately identify it?;
2) Once identified, how do mediators respond to IPA?;
3) Are victims protected during joint sessions?;
4) Is the process truly voluntary and fair?;
5) Are the resulting agreements fair and equitable?; and,
6) Have safety precautions been built in for victims participating in the process?
Below, we review the relevant empirical work addressing these issues and discuss how
the current study built upon the extant base of knowledge.
What is the prevalence of IPA in couples mandated to mediation?
As a comparison, in representative samples of the general United States population, using
behaviorally specific questions related to partner rape, physical assault and stalking, prevalence
rates of 25% for women and 7.6% of men were found (Tjaden & Thoennes, 2000). Rates of
partner rape, physical assault and stalking for the prior 12 months were 1.5% for women and
0.9% for men (Tjaden & Thoennes, 2000). In the mediation context, from their experience
directing programs mediation directors estimated the proportion of divorcing couples alleging
some dimension of IPA in the history of the relationship to be between 40% and 80% (Pearson,
1997). A more recent study assessed a representative sample of mediation clients in California
during one week in June 2008 (81% response rate), asking directly if “physical violence” had
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 44 2007-WG-BX-0028
occurred in the relationship. Overall 52% of families reported physical violence had occurred
(47% of mothers and 27% of fathers) (Administrative Office of the Courts [California], 2010).
The majority of reports were for physical violence that had occurred more than one year before
(60% of mothers and 58% of fathers); however, 36% of mothers and 35% of fathers reported
violence within the prior 12 months.
Recent international studies in both the United Kingdom (Moloney et al., 2007; Walker,
1989) and Australia (Keys Young, 1996) also confirm high rates of IPA in mediation samples.
These rates were much higher than those found in representative sample of the general United
States population noted above. We extended existing mediation research in the current study in a
couple of important ways. We assessed IPA using behaviorally specific questions as reported by
mothers and fathers who were married to each other and expanded the content of the questions
asked to include previously excluded dimensions (sexual intimidation/coercion/assault, and
coercive controlling behaviors).
What dimensions of IPA are found in couples attending divorce mediation?
Until the last decade, few studies assessed rates of IPA in mediation samples. In those
few studies that did, very different methods of assessment and definitions were used, making it
difficult to compare and contrast findings. One early study used a brief version of the Conflict
Tactics Scale (Straus, 1979), and addressed both physical and emotional abuse (Mathis &
Tanner, 1998). In this study the couples were divided into two groups. The “violent” group
included at least one partner who reported physically violent behaviors ranging from minor acts
of aggression, such as pushing or shoving to much more serious acts, such as use of a weapon.
Interestingly the “nonviolent” group included at least one partner who reported behaviors
ranging from yelling and swearing to smashing property (non-person-directed physical violence).
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 45 2007-WG-BX-0028
The majority of the couples were placed in the “violent” group (60%), and half of those (51%)
had used extreme tactics (beating or using weapons).
Researchers conducting another study created an inclusive category labeled “abused,”
which was defined as those reporting any occurrence of physical abuse (slapping, grabbing,
shoving, kicking, punching); severe physical abuse (beating, choking) or use of weapons (to
threaten or injure); or two or more incidents of intimidation (threats, stalking, telephone
harassment). This “abused” category included 80% of the total sample of women and 72% of the
men (Newmark, Harrell, & Salem, 1995). The most common dimension of abuse reported was
intimidation (74% of the women and 61% of the men). Physical abuse was reported by 68% of
the women and by 55% of the men. Alarmingly, severe abuse was reported by 38% of the
women and by 20% of the men; threats with a weapon were reported by 18% of the women and
by 15% of the men. The women in the “abused” group also scored significantly lower on
measures of empowerment (both personal and in relation to working with the court system)
compared to non-abused women. M men in the “abused” group scored significantly lower than
non-abused men in personal empowerment, but not in terms of working with the court system.
Two conclusions can be drawn from this study; a tremendous amount of IPA was reported by
these families in roughly equal percentages by men and women; and, women reported greater
disadvantage in navigating the court system.
In 1999, as part of the Statewide Uniform Statistical Reporting System, the California
Administrative Office of the Courts examined the prevalence of IPA among a representative
sample of parents in court-based child custody mediation cases (Administrative Office of the
Courts [California], 2002). Results indicated that 76% of families reported IPA (66% physical
abuse, 55% physical violence, 48% threats of violence, 44% had orders of protection, 13%
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 46 2007-WG-BX-0028
sexual assault). As with the prior studies, very few families reported only one dimension of
violent behavior. The most common combination of violent behaviors found in 26% of the
families was pushing, grabbing, shoving, throwing things, slapping, kicking, biting, or hitting.
The 2008 update to this study indicated that more than half of the families reported a history of
physical violence and 15% reported having a current restraining order in place (Administrative
Office of the Courts [California], 2010). This work addressed two important issues that had been
left out of most prior investigations—assessment of sexual assault and questions concerning
restraining orders either in the past or present. Sexual assault is the most under assessed
dimension of violence in the married population in general (Bergen, 2005) and in the mediation
context specifically.
To extend this work, the current study investigated a mediation sample and asked
behaviorally-specific questions including sexual forms of IPA and asked about restraining
orders, but did so of fathers and mothers married to each other. This was significant because
assessing each parent in isolation does not provide an accurate picture of the dynamics IPA in the
relationship. Assessing IPA dyadically is particularly important in a divorcing population
because mutual allegations of IPA are common and the safety of victims and violence-exposed
children are at stake (Kelly & Johnson, 2008). Courts must decide what is in the “best interests
of the child[ren]” and without a clear and accurate understanding of the IPA dynamics within a
couple, this decision is rendered impossible.
What are typologies of IPA and why might they be important in mediation?
One particular IPA typology (Kelly & Johnson, 2008) has gained attention recently and
has been applied in the mediation context, making it especially relevant here. Johnson and Kelly
(2008), elaborating on Johnson’s prior work (Johnson, 2006), describe five important couple-
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 47 2007-WG-BX-0028
level types of IPA: Coercive Controlling Violence, Violent Resistance, Situational Couple
Violence, Separation-Instigated Violence, and Mutually Violent Control.
The first type, labeled Coercive Controlling Violence, is similar to the “culture of
violence” type investigated by Fischer, Vidmar and Ellis (1993). This pattern is likely the one
most members of the general public think of when they hear terms such as “wife beating,”
“battering,” “spousal abuse” or “domestic violence” (Kelly & Johnson, 2008). Coercive
controlling violence is a pattern of violence, control and manipulation, that includes one partner
controlling the other partner’s actions, relationships, and activities. This type of abuse often
includes surveillance, and the victim is often punished when she or he fails to follow the rules
established by the coercive party (Kelly & Johnson, 2008).
Using Johnson’s (2006) conceptualization, Stark (2007) further discusses what he calls
coercive control as a pattern of violence, intimidation, isolation, and control in which the main
goal is to restrict the other person’s liberties. The perpetrator exerts control through acts of
physical abuse or violence; through sexual assault, intimidation, or coercion; or verbal threats of
serious violence. Where there has been a history of severe physical violence, nonviolent control
tactics may be effective in controlling the victim. Thus, coercive controlling violence may not
necessarily be accompanied by high levels of physical violence, but instead may exist alone
(Kelly & Johnson, 2008). Stark argues that researchers should examine levels of coercive control
in a relationship in order to understand the context and meaning of IPA. This type of IPA, while
sometimes initiated by women, is thought to be perpetrated predominantly against women by
men (Johnson, 2006). This type of violence is most commonly reported among victims in law
enforcement, hospital and shelter samples (Graham-Kevan & Archer, 2003; Graham-Kevan &
Archer, 2008; Johnson & Leone, 2005; Johnson, 2006). It has been linked to lethal or near-
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 48 2007-WG-BX-0028
lethal violence against the victim (Campbell et al., 2003); to post-separation violence (Hardesty,
Khaw, Chung, & Martin, 2008) and to post-divorce problems with parenting (Dutton &
Goodman, 2005; Hardesty & Chung, 2006).
A second type of IPA is Violent Resistance (resistance to a violent, coercively controlling
partner). This type of IPA has been described as the use of defensive violence in response to a
coercively controlling partner. In this context, IPA behaviors are an attempt to stop the violence;
defend oneself against an attack; or to physically defend others, such as children or pets (Kelly &
Johnson, 2008). Many battered women’s advocates and researchers have categorized all female-
initiated IPA as this type (Female Resistence, Resistive/Reactive Violence, Self-Defense), (Kelly
& Johnson, 2008; Pence, Dasgupta, Taylor, & Praxis International, 2003; Walker, 1984; Yllo &
Bograd, 1988). This type of violence garners the most media attention particularly in the most
severe cases such as when a woman murders her abusive partner (Kelly & Johnson, 2008).
Interestingly, in research, few factors differentiated battered women convicted of homicide from
battered women who did not murder their partners. What did differentiate the battered women
who did not kill their partners from the battered women who did was the nature and type of
behavior by the abusers (Browne, 1987; Kelly & Johnson, 2008). Abusers who were killed were
reported to have attacked their partners more frequently; attacked them with more severe
injuries; sexually assaulted them more often; and make more frequent death threats against them
(Browne, 1987; Kelly & Johnson, 2008). The murders often occurred in the context of
separation, when the battered women were attempting to leave the abuser (Kelly & Johnson,
2008).
A third type of IPA is Situational Couple Violence. This type of IPA does not reflect core
issues of power and control, nor is fear characteristic (Johnson & Leone, 2005; Johnson, 1995).
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 49 2007-WG-BX-0028
Compared to men involved in Coercive Controlling Violence, misogynist attitudes are not
characteristic of men involved in this type of IPA (Holtzworth-Munroe, Meehan, Herron,
Rehman, & Stuart, 2000; Kelly & Johnson, 2008). In this type of violence IPA is contextually
the result of “situations” or arguments between partners that escalate into physical violence
(Kelly & Johnson, 2008). One or both partners are unable to manage their anger and/or have
poor conflict resolution skills. It is thought to occur more frequently than the other types and to
involve more minor forms of violence (pushing, shoving, grabbing) (Kelly & Johnson, 2008).
When assessed via large population surveys, perpetration by men and women were reported at
similar rates (Kelly & Johnson, 2008; Kwong, Bartholomew, & Dutton, 1999). Nearly always
absent from this group are severe violence and core issues surrounding power and control.
The fourth type of IPA is Separation-Instigated Violence. This type of IPA occurs upon
separation among couples with no prior history of IPA (Kelly & Johnson, 2008). It is believed to
likely be triggered by traumatic experiences related to the separation. For example, it can be
triggered when a parent comes home from work and, with no prior warning, finds the family
home empty. Similarly, this type of IPA can be triggered when a prominent professional or
political figure is humiliated by being served divorce papers by a process server or when a souse
comes home and finds her or his partner with a lover (Kelly & Johnson, 2008). This type of IPA
can include stalking; threats with weapons; destroying cherished property; throwing objects at a
partner; and sideswiping or ramming a partner’s car (Kelly & Johnson, 2008). Perpetrators of
this type of IPA generally admit their behaviors, and orders of protection tend to be effective in
keeping the spouses apart. Nonetheless, it ought to be noted that this form of abuse can be
extremely dangerous and potentially lethal (Dutton, 2007; Holtzworth-Munroe et al., 2000;
Jacobson & Gottman, 1998; Kelly & Johnson, 2008). This type of IPA is initiated by both men
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 50 2007-WG-BX-0028
and women and tends to be uncharacteristic of the parties and unexpected. In a study of high
conflict parents disputing custody, about a fifth of the sample indicated the violence began after
separation (Johnston & Campbell, 1993).
The fifth type is Mutual Violent Control. This type of IPA is described as occurring
between two coercively controlling and violent partners. By definition, this type is initiated by
both men and women (Johnson, 2006; Hardesty & Ganong, 2006). Johnson found few couples
meeting this criteria and little is known about its frequency, features, or consequences (Kelly &
Johnson, 2008). This is an important type to investigate in the mediation context in that little is
known about this type of IPA and the consequences for children could be severe (Kelly &
Johnson, 2008).
One qualitative study assessed divorcing couples in relation to the above typology,
however, did not find evidence of all five types (Hardesty & Ganong, 2006). This study relied
upon a sample of women attending a court-mandated parent education program for all divorcing
parents in two counties in the central United States. Results found that women reported one of
only three patterns of IPA (Hardesty & Ganong, 2006). The first pattern included chronic verbal,
emotional/ psychological, sexual, physical abuse, physical violence, and threats of violence
accompanied by coercive controlling behaviors, which began early in the relationship and
continued throughout the marriage. This type was similar to that of Coercive Controlling
Violence noted above.
The second pattern included episodic physical abuse with coercive controlling behaviors.
The central dynamic in this type was control; however, the physical abuse was infrequent and
generally not severe. This pattern is not found in Johnson and Kelly’s original typology as
described above. The third pattern included episodic physical abuse but was not accompanied by
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 51 2007-WG-BX-0028
a pattern of coercive control. The physical abuse generally occurred in the context of specific
arguments often related to money or drug use. This type of IPA is similar to that of Situational
Couple Violence. It is unclear why the other types of IPA proposed by Johnson (2006) were not
found in this sample. It could be that the sample was restricted, since all women in the study
were screened for experiencing physical abuse and/or violence prior to or after the divorce.
While the types of IPA described above have been discussed in the context of mediation
samples, it has yet to be investigated empirically with a mediation population. Therefore, it is not
yet known if the frequency, severity, context, or type of IPA found in mediation couples
resembles that seen in large-scale national surveys of the community (Situational Couple
Violence) or the type of violence more commonly found in shelter, agency, hospital or law
enforcement surveys (Coercive Controlling Violence) (Kelly & Johnson, 2008), or whether IPA
is more similar to that found in the sample of women attending court-mandated parent education
(Hardesty & Ganong, 2006) (chronic IPA using all types plus coercive controlling behaviors;
episodic physical abuse with coercive controlling behaviors; physical abuse only). We theorized
that the current study sample will more closely resemble those of other help-seeking samples
than that of nationally representative samples, and would report higher rates of IPA that included
higher risk for lethal or near lethal outcomes.
As noted above, IPA is commonly reported in mediation samples, with a recently
documented rate of 52% when clients were asked directly about physical violence
(Administrative Office of the Courts [California], 2010). Interestingly, in 47% of the cases in one
mediation study, neither parent discussed IPA in either separate screening interviews or
separation sessions before or during mediation (Administrative Office of the Courts [California],
2010). This finding could mean several different things. It could be that the individuals did not
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 52 2007-WG-BX-0028
perceive the IPA as serious enough to bring up because it may have occurred in the distant past;
have been mutually perpetrated; not have been thought to be important enough to disrupt
negotiations in mediation; or, because the victims were so fearful of negative consequences of
revealing IPA that they chose not to disclose it (Kelly & Johnson, 2008). Alternatively, it is
possible that the mediator failed to sensitively screen for its presence, thus limiting detection
options. The current research clarified the types of IPA that are characteristic of a sample of
couples in mediation using a sample that are married to each other. Also included in the study
were variables addressing the short and long term outcomes associated with IPA. The findings
from this current study are particularly important for mediators working in the field, as previous
small, qualitative studies found that specific types of IPA were related to ongoing problems with
post-separation parenting (Hardesty & Chung, 2006).
What happens when IPA victims are identified in mediation?
Regardless of the dimensions of IPA or the types reported, if mediators do not “call” or
expressly identify a case as having IPA, then no opportunities exist to screen cases out or to
provide procedural accommodations. Although it is clear that safety concerns for families with
IPA should be addressed, there has been sparse research on mediator screening, documenting,
and responding to IPA, including whether and how safety measures are included in mediation
agreements.
The work of researchers Johnson, Saccuzzo and Koen (2005) indicates that even if IPA is
reported and documented, it does not necessarily mean that a case will be flagged and treated
uniquely due to the presence of IPA. It is thus the mediator’s “calling” the case as involving IPA
that is paramount in determining how the case is processed, and what if any accommodations are
implemented. Johnson et al. (2005) assessed recommendations from the mediator contained in
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 53 2007-WG-BX-0028
reports to the court in cases in which couples were unable to reach agreements via mediation.
Results suggested that in 57 % of the cases with clear indicators of IPA (IPA alleged or
temporary restraining orders recorded on the screening form or in the case file) mediators failed
to comment on IPA in their recommendations to the court.
In contrast to the Johnson et al. study, other studies have found that IPA was screened
and monitored more consistently, resulting in differences in case processing, and in the rates of
agreements among couples with and without IPA (Mathis & Tanner, 1998; Tishler,
Bartholomae, Katz, & Landry-Meyer, 2004; Wissler, 1999). One reason for conflicting findings
with respect to screening, documentation, triage, and monitoring of IPA cases in mediation is
likely the lack of consensus on how to identify IPA; what to do about it when it is detected; and
agency policy differences on how to handle IPA cases. This current study identified cases with
IPA that were accommodated; not accommodated; and screened out, and identified long-term
outcomes for each group.
Should victims of IPA be screened out or accommodated?
Because of safety concerns for IPA victims, some commentators have argued that
couples with identified IPA histories ought to be screened out of mediation altogether (Howe &
McIsacc, 2008; Johnson et al., 2005). Others argue that the use of mediation enhances victim
safety due to case processing at a more leisurely pace compared to court hearings as well as
enhanced opportunities to develop mediation agreements to protect victims (safe exchanges of
the children, supervised visitation) (Edwards et al., 2008; Pearson, 1997). Thus, some
commentators argue that mediation procedures can and should be modified to accommodate
certain couples with IPA (Edwards et al., 2008; Ellis, 2008; Erickson & McKnight, 1990;
Rimelspach, 2001). The caveat is that adequate methods of assessment must be used and couples
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 54 2007-WG-BX-0028
with IPA must be assigned to specialized safety-enhanced mediation procedures that are matched
to their level of assessed risk (Ellis, 2008). In addition, any mediation agreements produced must
include safety restrictions for the victim and violence-exposed children.
Studies of IPA victim’s responses to mediation are mixed. In one study approximately
50% of women reported violence in the marriage (Pearson, 1993). These women also reported
that the violence impaired their ability to communicate with the ex-spouse on an equal basis
(Pearson, 1991; Pearson 1993). Of this 50%, approximately 33% also reported they felt they had
less bargaining power in mediation (Pearson, 1991; Pearson, 1993). It is unclear whether victim
perception of having reduced bargaining power led to any capitulation to unsafe or unfair terms
of the agreement. A second study found that female abuse victims felt less empowered in
mediation compared to non-abuse victims (Newmark et al., 1995). However, victim responses to
survey items suggested that victims believed they could state their needs and stand up for
themselves despite their awareness of negative repercussions for doing so (Newmark et al.,
1995). Abuse victims of both sexes stated that their partner had controlled decision-making in
the relationship and believed that their partner might physically harm them in the next six
months. Victims of IPA also did not believe that the negotiations would lead to an acceptable
agreement (Newmark et al., 1995). Thus, it appears that abuse victims indicated that they are
fearful and often unable to meet a core requirement of mediation--speaking up for themselves.
As a result, abusers likely controlled the negotiation process and the agreement ultimately
reached.
In another study, however, the Alaska Judicial Council (1992) found that 61% of parents
requesting mediation were ineligible due to a statutory requirement excluding all cases involving
“domestic violence.” The report stated that many (exact percentage not given) victims believed
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 55 2007-WG-BX-0028
they should have been given the choice to participate or not in mediation rather than being
automatically excluded from mediation (Alaska Judicial Council, 1992). Interestingly, these
abuse victims did not mediate because they were screened out so it is unclear what percentage
would have reported disempowerment and capitulation to unfair agreements had they actually
been able to mediate. This study highlights the importance of listening to victims concerns about
IPA and allowing them a strong voice in the decision to proceed with or without mediation.
As a matter of practice, there are contested custody and visitation cases with IPA that are
being mediated. Researchers found that only 5-6% of mediation cases are screened out for IPA
even though 40-80% of participants report experiencing IPA (Pearson, 1997). In the current
study, we compared the percentage of cases reporting IPA with the percentage of cases screened
out. We evaluated whether the cases that were screened out had particular characteristics that
were readily identifiable. And, we assessed in detail the dimensions of IPA (coercive controlling
behaviors, psychological abuse, physical abuse, threatened and escalated physical violence, and
sexual intimidation/coercion/assault) and the types of IPA (situational couple violence, coercive
controlling violence, violent resistance, separation-instigated violence, mutually violent control)
(Kelly & Johnson, 2008) among those screened out and those who proceeded with mediation.
What happens when a case has both IPA and child abuse?
Estimates of the co-occurrence of IPA and child abuse range from 30-60% (Appel &
Holden, 1998; Casanueva, Martin & Runyan, 2009); Knickerbocker, Heyman, Slep, Jouriles, &
McDonald, 2007; Malik, Silverman, Wang, & Janczewski, 2008; Zolotor, Theodore, Coyne-
Beasley, & Runyan, 2007). One recent, methodologically rigorous study estimated the co-
occurrence of partner and child-targeted aggression as being 45% (Slep & O'Leary, 2005).
Discussion concerning the possible causes or mechanisms underlying this co-occurrence will not
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 56 2007-WG-BX-0028
be discussed; however, it is important to note that in the context of a divorce, allegations of co-
occurring IPA and child abuse are extremely common (Moloney et al., 2007).
Under nearly all circumstances, parents have a fundamental right to parent (Troxel v.
Granville, 2000; Spitko, 2006), necessitating momentous legal hurdles (strict scrutiny) to
overcome whenever that right is questioned for any reason (Spitko, 2006). However, when one
partner commits IPA against the other, the sanctity of this fundamental right has been questioned
(Stein, 1998; Haddix, 1996). Nonetheless, in many cases parents who have brutalized their
partners have still maintained custody and unsupervised parenting time of their children (White,
2005). In some states, however, views toward the fundamental right to parent appear to be
changing, particularly in the context of IPA. For example, in 40% of the states, when a legal
finding of IPA is made by the court, there is a rebuttable presumption against joint custody and
the perpetrator must put forth a legal argument as to why the court should consider joint custody
in the case instead of presumptively dismissing it (Dunford-Jackson, 2004).
In recent years, some states have expanded their statutory definitions of child abuse to
include children’s exposure to IPA by a parent (Goodmark, 2010). Consequently, if children are
exposed to IPA or its aftermath, (broken property or physical injuries), they may be identified as
abused and subject to removal if the parental IPA victim fails to leave the abuser and fails to
obtain an order of protection keeping the abuser from seeing the children (Goodmark, 2010).
Victims of IPA whose children are also abused and who do not leave the abuser can be charged
with failure to protect the children from the abuser, and those children removed by child
protection agencies (Goodmark, 2010). If a victim of IPA flees the home and the abusive partner
without the children in order to protect herself from danger, charges of child abandonment can
follow (Magen, 1999). Thus, there are formidable systemic conflicts between the child protection
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 57 2007-WG-BX-0028
system and those processes that are intended to protect battered women from harm by their
partners, including but not limited to family law remedies (Buckley, Whelan, & Carr, 2011). A
battered woman reporting IPA and/or child abuse in the divorce context (moving to a shelter,
obtaining an order of protection, litigating to keep abusers away from the children) may be
perceived as obstructing the relationship between the abusive parent and children due to
vindictiveness or for reasons of secondary gain (Baskerville, 2007; Buckley et al., 2011;
Crowley, 2008; Dutton, 2005a; Dutton, 2005b, Dutton, 2006; Dutton, Hamel, & Aaronson,
2010). In some states “friendly parent” provisions are built into state statutes (Ariz. Stat. Ann.
§25-403.03). These provisions require the judge determining custody to consider which parent is
more willing to facilitate ongoing involvement of the other parent. Without the additional legal
hurdle of formally requesting and having the court issue a legal finding of domestic violence in
the relationship, these “friendly parent” provisions in state statutes can be extremely risky for
victims in cases with co-occurring IPA and child maltreatment.
Thus, the paradox for IPA victims and their violence-exposed children is this: while
victims are expected to protect the children by leaving the abuser, “i.e., the leave ultimatum”
(Douglas & Walsh, 2010), that same behavior may lead to allegations of having left the abuser is
a tactical form of manipulation in the subsequent divorce proceeding (Adam, 2010; Buckley et
al., 2011). To combat this problem there are jurisdictions with unified family courts and the goal
of “one judge one family,” which can eliminate much of this problem. There are also
jurisdictions without unified family courts but with pioneering judges who consolidate cases
when families have orders or cases in both juvenile and family courts (Adam, 2011).
Unfortunately, unified family courts and pioneering judges consolidating juvenile and family law
cases are not common. The majority of courts in the United States are designated by subject
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 58 2007-WG-BX-0028
matter (criminal, family, probate, immigration, mental health) and in so doing operate as “silos”
that are devoid of knowledge or contact with other court systems (Adam, 2010). While courts
may operate in subject matter silos, families do not live in subject matter silos (Adam, 2010).
Particularly with respect to issues relating to children, overlapping jurisdictional problems
between the court silos creates situations wherein judges unknowingly issue incomplete or
contradictory orders across court systems. This current study investigated the short- and long-
term outcomes of cases with IPA and reported child maltreatment. We theorized that IPA and
reported child maltreatment would co-occur, and when it did mothers would have less favorable
custody and parenting time outcomes in court.
How many people do not have attorneys and what happens in cases without attorneys?
Little research has been conducted concerning the effects of self-representation (pro se
representation) in mediation cases. Most research to date documents the percentages of pro se
cases but does not investigate the effects it has on the process or outcome of mediation. For
example, in a recent study in a large representative sample of mediation cases in California, the
California Administrative Office of the Courts (2010) found the majority of clients were self-
represented. Likely due to several interrelated factors (the economic downturn in the United
States economy, rising rates of unemployment and underemployment, rising cost of attorney
representation), the proportion of cases involving at least one pro se party increased steadily over
time, from 52% in 1991 to 75% of cases in 2008 (Administrative Office of the Courts
[California], 2010).
Although not specifically focusing on mediation, our prior work with divorce cases
generally sheds some light on the difficulties pro se litigants face (Sales, Beck, & Haan, 1993a).
In divorce cases filed in one large urban county in Arizona in 1990, both parties were pro se in
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 59 2007-WG-BX-0028
52% of the cases and in 88% of the cases at least one party was pro se (Sales, Beck, & Haan,
1993b). Analyzing differences in divorce cases with and without attorney representation (Sales et
al., 1993a), we found that litigants without attorneys have much more difficulty negotiating the
divorce process. Often litigants without attorneys failed to complete the necessary paperwork
and failed to make the proper arguments, in the correct form, to support their positions.
It is likely that IPA victims without attorneys are less likely to understand the
significance of identifying IPA early in the process of divorce or to know the specific legal
procedure for requesting a judicial finding of IPA. The benefit of a specific legal finding of IPA
is that the legal finding then triggers a rebuttable presumption the judge must consider in
awarding custody. As noted above, in approximately 40% of the states, there is now a rebuttable
presumption built into many statutes that presumes it is not in the best interest of children for a
parent who perpetrates IPA to be awarded sole or joint custody (Dunford-Jackson, 2004). A
study comparing states with and without this rebuttable presumption found that judges were
significantly more likely to assign structured parenting time for abusers (69%) in states with this
presumption, compared to states without it (54%) (Morrill, Dai, Dunn, Sung, & Smith, 2005).
Without an attorney, it is unlikely that victims would know how to present their cases in
such a manner that would require judges to consider whether the presumption of joint or sole
custody had been rebutted. Even if victims begin a case with an attorney, it is likely that at some
point they will represent themselves due to the high cost of attorney representation. For example,
a recent study in New York State found that 97% of parents are unrepresented in child support
matters in New York City and 95% are unrepresented in the remaining counties in the state (Task
force to Expand Access to Civil Legal Services in New York, 2010).
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 60 2007-WG-BX-0028
In addition, attorneys are generally more informed about divorce than laypersons, and are
aware of the special procedures that can be offered to victims of IPA in mediation. Clients
without attorneys are less likely to know to request special procedures in mediation. In this
current study, we assessed the mediation and divorce process and outcomes of cases with and
without attorney representation. We focused on those cases where there was a legal finding of
IPA documented. In the study jurisdiction, the state statute states that when there is a finding of
IPA, there is a rebuttable presumption against sole legal custody being awarded to the abuser
(Ariz. Stat. Ann. §25-403.03).
Is post-separation IPA a problem?
One widely accepted myth about IPA, that is, by one party exiting an abusive
relationship, the violence will cease, he victim (and children) will be safe. Research indicates,
however, that leaving not only fails to stop the violence, but separation poses a significant risk
of escalated, often lethal violence (American Psychological Association, 1996; Saunders, 2007;
Tjaden & Thoennes, 2000). Decades of research documents that the time immediately preceding
and following separation—a time when mediation could be occurring—is the most dangerous for
IPA victims, frequently leading to escalated and, at times, lethal acts of violence (Campbell et
al., 2003; Hardesty & Chung, 2006; Mahoney, 1991; Tjaden & Thoennes, 2000). One study
found that more than a third of the 135 victims studied were re-assaulted within two years after
leaving their abuser (Fleury, Sullivan, & Bybee, 2000). However, the time immediately
surrounding the separation is not the only time post-separation occurs. In a large representative
sample of Canadian citizens, 40% of women and 32% of men reported post-separation violence
(Hotton, 2001). Of those reporting post-separation violence, in 39% of the cases the violence
reportedly began after separation; in 24% of the cases the violence became more serious post-
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 61 2007-WG-BX-0028
separation; and in 37% of the cases the violence reportedly remained the same. When violence
occurred post-separation, 55% of women reported these incidents to the law enforcement
whereas only 30% of men did so.
A large representative sample of citizens in the United States found that married women
who were separated from their husbands were nearly four times more likely to report
victimization by their husbands (raped, physical assaulted, stalked) that were women who lived
with their husbands (20% versus 5.4%) (Tjaden & Thonnes, 2000). This increased rate of
victimization also holds true for married men living apart from their wives, however
victimization for men was reported at a significantly lower rate overall (7% versus 2.4%). When
the women in the study were asked about when the violence occurred, 24.7% of the rape victims,
18.2% of the physical assault victims and 36.4% of the stalking victims indicated that the
victimization occurred both during and after the relationship ended. For 6.3% of the rape victims,
4.2% of the physical assault victims and 42.8% of the stalking victims, the violence began after
the relationship ended. In Arizona specifically, an analysis of homicides in 2004 was conducted
(Bileski, 2008). Overall, 20% of the homicides were linked to domestic violence. Of these
domestic violence homicides, 18% were committed by a husband, 7% were committed by a
wife/Ex-wife and 16% were committed by a significant other (no sex of perpetrator or victim
designated) (Bileski, 2008). Unfortunately there is no data concerning whether the victims were
in the process of divorce at the time they were murdered.
Children can get caught in the crossfire of IPA, becoming direct or indirect targets of
abuse, jeopardizing their short- and long-term physical and emotional health and safety. As noted
by Mechanic, Weaver and Resick (2002) IPA can continue for many years post-separation for
some victims. One of the many ways abusers continue to abuse victims is take them to court
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 62 2007-WG-BX-0028
repeatedly in order to gain increased parenting time, harass the victim, and exhaust their finances
(Hardesty, 2002; Miller & Smolter, 2011; NCJFCJ, 2008; Pollet, 2011). In the context of
mediation, an important consideration for agreements then is to build in arrangements for
providing safety for victims and children so that there are reduced opportunities for post-
separation violence (supervised exchanges, supervised parenting time were needed, orders of
protection) and reduced opportunities for court litigation as a means of harassment and stalking.
The current study assessed post-divorce IPA by linking calls to area law enforcement reports at
three stages: prior to mediation, during the divorce process, and post-divorce. The study also
assessed the number of, type of and outcomes hearings requested post-divorce. In so doing, the
degree of post-divorce IPA and court-based harassment was measured.
Is there a risk that victims will mediate agreements that are unsafe?
A frequent research finding is that there is no difference in provisions in agreements
between couples who report IPA and those who do not (Moloney et al., 2007). Additionally, for
those couples reaching an agreement, there is no difference in the nature of the agreements
between couples with and without reported IPA. Tishler et al. (2004) found that 30% of couples
reached an agreement in mediation, but there was no statistically significant difference in
agreement rates between couples with IPA and without IPA , even though fewer couples with
IPA than without IPA reached agreements (22% versus 34%). Wissler (1999a, 1999b) found the
likelihood of reaching an agreement in mediation was unaffected by whether the case involved
violence, the frequency or recency of violence, or whether the victim required medical care. The
likelihood of reaching an agreement was also unrelated to whether one party had expressed fears
or concerns about the other party’s violence prior to entering mediation (Wissler, 1999). Mathis
and Tanner (1998) found that 54% of cases reaching full agreements in mediation were
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 63 2007-WG-BX-0028
categorized as “violent.” Of the agreements negotiated for the “violent” group, 57% agreed to
joint custody with standard parenting time or agreed to split custody with equal primary
residence and parenting time (Mathis & Tanner, 1998). These findings are potentially troubling
in several ways. Some people consider mediation a “success” if any agreement is reached,
regardless of the agreement and regardless of the context of the relationship between the parents.
First, after a couple separates children can become a means for abusers to maintain
control over estranged spouses, potentially harming both the estranged spouse and the children.
Second, threats to pursue or challenge custody, abduct the children, or falsely report child abuse
can financially drain victims via lengthy custody battles, thus effectively coercing them into
accepting unsafe terms of custody agreements that maintain unwanted perpetrator-victim contact
(Arendell, 1995; Hardesty & Ganong, 2006). If victims accept agreements that include joint-
decision making (joint legal custody), unsupervised exchange of children, and no restrictions on
contact between abuser and victim, then post-divorce abuse seems likely to continue.
Unfortunately, to date there has been no methodologically sound, longitudinal study assessing
either the effectiveness of current IPA screening in mediation or the long term consequences of
decisions made in mediation and thereafter.
Outside corroboration of reports of IPA seem to be the key for factor victims to obtain
agreements with safety restrictions. One study found that in cases where IPA was identified,
agreements negotiated gave more control to victims such as sole custody (Mathis & Tanner,
1998). In two more recent studies, it was only when outside evidence of IPA was supplied that
IPA allegations impacted the agreements that were made (Kernic et al., 2005; Moloney, 2008).
In the Kernic et al study, when outside evidence was included in the divorce file, the courts
restricted the abuser’s access to the children in 71% of the cases, compared to only 17% of cases
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 64 2007-WG-BX-0028
that lacked independent corroboration. Thus, external corroboration of IPA appears to be critical
in influencing the terms of mediated cases with IPA. Supervised visitation was a common
restriction that occurred in 26% of the cases with outside evidence, but only 5% of those without
it. In the Moloney (2008) study, there was a smaller likelihood of overnight stays for children
when allegations of IPA were buttressed by external evidence. Alarmingly, even with court
ordered supervised exchange and/or visitation programs, post-separation abuse can continue
(Parker, Rogers, Collins, & Edleson, 2008).
A qualitative study of post-separation co-parenting found that particular types of IPA
were indicative of post-separation abuse. Specifically, former husbands who had difficulty
differentiating the parenting relationship from the marital relationship and who had a history of
using coercive controlling violence against their former wives continued using the children and
the courts to exert control over their wives (Hardesty et al., 2008). Victims reported that
unsupervised exchanges of the children were a common pathway whereby abusers attempted to
reassert control over their former partners (Hardesty et al., 2008). In contrast, former husbands
who did not have difficulty separating the parental role from the marital role and who exhibited
Situational Couple Violence during the marriage did not continue to use the children or the
courts to exert control. Thus, typological variation in the nature and dynamics of IPA were
integral to predicting post-separation violence and parenting.
To assess safety factors in mediation agreements and divorce decrees and external
corroboration of IPA, we reviewed IPA reported in mediation and the specifics of any
agreements made in mediation including restrictions on contact between parents, orders of
protection and restrictions on parenting time. For those couples who did not reach an agreement
in mediation, the terms of parenting agreements at the time of divorce were analyzed. As one
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 65 2007-WG-BX-0028
method of assessing external evidence of IPA, calls to area law enforcement were time-stamped
to the dates couples filed for divorce, during the divorce process and post-divorce.
To complicate matters, alternatives to mediation for victims of IPA are scarce. The
current market price for attorney representation precludes many couples from securing legal
advocacy, resulting in more frequent pro se cases (Herman, 2006; Sales et al., 1993). Courts
have struggled to accommodate pro se litigants and some have developed self-help centers that
assist litigants to obtain forms, legal information and sometimes advice. Yet these centers may
not be adequate for litigants with complex cases (Mather, 2003), or for those who are victims of
IPA.
Interestingly, some IPA victims, even those identified by mediators, request to proceed in
mediation despite IPA. As noted above, in one study 68% of potential mediation couples were
screened out due to IPA. Some IPA victims reported the abuse was irrelevant and expressed the
desire to proceed, arguing that mediation afforded them at least some assistance in negotiating
(Alaska Judicial Council, 1992). It is likely that victims also request to proceed in mediation
because the alternative negotiating in court before a judge alone, without attorney representation.
As noted above, our prior work indicates that pro se litigants face significant barriers when
attempting to navigate the legal divorce process, which was designed for legal professionals.
Because mediation involving couples with IPA is occurring, attention must be paid to how these
cases are being handled. The current study identified the dimensions, types and frequencies of
IPA occurring among couples in mediation. The study then tracked cases with and without IPA
over time to better understand the short- and long-term outcomes of these cases.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 66 2007-WG-BX-0028
NIJ Project Goals and Research Questions:
The overall goal of this study was to enhance the effectiveness of screening and careful
handling of cases with IPA that are mandated to attend mediation, by providing empirical data
concerning long-term outcomes of these cases to public policy-makers within state legislatures,
court systems, mediation programs and professional organizations. Results of this study can be
used to develop policies and practices that enhance the short- and long-term safety of IPA
victims and children. To do so, this NIJ funded project tracked a large (N=965) sample of
mediation couples mandated to attend mediation through a court-connected conciliation program
for a period of 29 months using multiple official databases and records. Behaviorally specific
IPA data gathered in the mediation screening process was subsequently linked to long-term
outcomes recorded in superior and limited jurisdiction courts and area law enforcement
databases.
There four goals for this project were to:
Goal 1: Determine whether the mediation program accurately identifies couples with
self-identified IPA and assess whether these cases are treated differently.
Goal 2: Assess whether mediation agreements, divorce decrees and/or parenting plans
include safety measures in cases involving self-reported IPA.
Goal 3: Assess the frequency with which mediation agreements/divorce decrees are re-
litigated over time.
Goal 4: Test a multivariate conceptual model using variables that are hypothesized to
affect mediation, divorce case and post-divorce outcomes (see Cascade Model Figure 16).
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 67 2007-WG-BX-0028
Methods
Participants.
Participants in the present study consisted of all couples who attended the court-
connected mediation program in Pima County (Tucson), Arizona between May 1998 and
October 2000 to resolve custody and parenting time disputes. The sample was limited to those
couples attending mediation for the first time, as a result of a pending divorce (N=1015 cases;
2030 individual participants). Excluded from this analysis were parents returning to mediation
for a second attempt at pre-divorce mediation or to renegotiate issues post-divorce; couples who
were never married; and, grandparents negotiating with parents to see their grandchildren. At
follow-up, 50 cases were found to fall within the excluded categories above and were thus
dropped from the analyses.
Case variables for the Superior Court database collected early in the study (N=271) were
not usable in the analyses concerning post decree outcomes. The original data collection
instrument was found to be too complicated and time consuming; therefore, a modified version
was developed to facilitate data collection for the remainder of the subjects (N=694).
Unfortunately, data collected early did not precisely match the variables collected in the
remainder of the study. Thus, the final sample was N=965 cases for the mediation case data. The
final sample was N=694 for the Superior Court case data (see Figure 1). All of the cases in the
Superior Court case data appear in the mediation case data.
Using a one-way ANOVA testing if there were significant differences between the earlier
and later samples on important demographic variables (age, marital length, number of children,
number of prior marriages, employment status, amount of income, ethnicity, education, Spanish-
speaking, mediator’s call of IPA) yielded only two small yet statistically significant differences
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 68 2007-WG-BX-0028
between the groups. The length of marriages, on average, for the latter group (n=694) were
longer (F=6.381, df=1, 861, p=.012), with a mean difference in years of 0.84 years. The second
difference between the groups (F=8.815, df=1, 963, p=.003) was that on average the latter group
(n=694) had a greater number of children, yet the mean difference (0.20) was very small.
Figure1. Case Flow Chart. Representation of the cases excluded from in sample.
The average age of participants was 35 years for mothers and 37 years for fathers (see
Table 1). Generally, both partners reported at least part time employment, although more fathers
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 69 2007-WG-BX-0028
(89%) than mothers were employed (79%). On average, fathers’ yearly income was
approximately double that of mothers ($26,082 versus $13,956), and fathers reported a wider
range of incomes than their spouses (fathers $0 – 228,000; mothers $0 – 109, 992). For 25% of
the couples the disparity in income between the fathers and the mothers was $5,300 or less. For
50% of the couples, the disparity was $13,800 or less. In 75% of the couples the disparity was
approximately $25,200 or less.
To compute the number of families in the sample living at or below poverty level, each
mother’s and father’s reported income was summed to determine the family income. Because the
parents participated in mediation between 1998 – 2000, these incomes were then compared to the
2000 U.S. Department of Health and Human Services federal poverty guidelines for the
appropriate “family unit” size and corresponding dollar figure for the poverty level of the family
(U.S. Department of Health and Human Services, 2000). Ten percent of the families in the
sample fell below the 2000 Federal poverty level. In the 2000 U.S. Census Bureau figures for
Pima County, Arizona (U.S. Census Bureau, 2000) for all families 10.5% were at or below the
poverty level; 16.4% of families with related children under 18 years old in the home and 20.6%
of families with related children under 5 years old.
This was the first marriage for most of these couples (fathers 86% versus mothers 84%)
and the average length of marriage was 9 years (range 0-39 years). Children ranged in age from
infants to 18 years of age, with a mean age of 8 years. The number of children in the family
ranged from 1-6 with a mean number of 2 children per marriage. The educational level of the
participants was categorized into four non-overlapping levels (0-12 but not a high school
graduate, high school graduate, some college, college graduate). The mean level of education for
mothers was 2.59 and 2.63 for fathers representing high school graduate to some college. There
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 70 2007-WG-BX-0028
was no significant difference between these means (paired T=0.35; p=.298). Sixty five percent
of both mothers and fathers fell into this combined group. Fathers' educational levels were
correlated with mothers' educational levels (Kendall tau=0.38, p<.05). Eighty seven percent of
the fathers and mothers were within one educational level of each other. The participants were
predominantly non-Hispanic White (61% of mothers and fathers) and Hispanic (29% of mothers
and 27% of fathers).
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 71 2007-WG-BX-0028
Table 1.
Participant Demographics
Range
Mean age fathers 37 19 -- 71
Mean age mothers 35 17 -- 58
Father have previous marriages 13.70% 0 -- 4
Mother have previous marriages 15.40% 0 -- 3
Mean children per household 2 1 -- 6
Education Sample Local Area 2000 Census
Father
0 – 12 9.50% 16.60%
High school graduate 37.90% 23.30%
Some college 26.60% 26.70%
College graduate (AA, BA/BS or above) 20.00% 33.50%
Not Answered 5.90%
Mother
0-12 11.40%
High school graduate 35.00%
Some college 29.90%
College Graduate (AA, BA/BS or above) 17.70%
Not answered 5.90%
Household or Family Income Sample
Father’s reported yearly income $31,825.95
Mother’s reported yearly income $16,707.97
Father employed 89.20%
Mother employed 79.50%
Ethnicity Sample Local Area 2000 Census
Father
White 60.90% 75.10%
Hispanic 26.70% 29.30%
African American 2.80% 3.00%
Asian American 0.90% 2.00%
Native American 0.90% 3.20%
Bi-racial 0.30% 3.20%
Other 0.70% 13.30%
Not Answered 6.70%
Mother
White 60.60%
Hispanic 29.40%
African American 1.60%
Asian American 1.00%
Native American 0.60%
Bi-racial 0.60%
Other 0.10%
Not Answered 6.0%
Language Sample
Mother English speaking 94.40%
Mother Spanish speaking 5.60%
Mother speaks both Spanish and English 0.00%
Father English speaking 95.00%
Father Spanish speaking 4.80%
Father English and Spanish speaking 0.20%
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 72 2007-WG-BX-0028
Mediation Process
In the study jurisdiction, couples who were divorcing and disputing custody or parenting
time (or both) were mandated by a local court rule to attempt to mediate custody and parenting
time disputes. This was done either by establishing temporary orders prior to the court setting a
date for a trial for a divorce or prior to modifying a decree post-divorce (Pima County Super. Ct.
Local Rule 8.7, 1998). The parents were either referred to the court-sponsored mediation service
or could stipulate to participate in private, fee-for-service mediation.
The process of mediation at the study site, the court-sponsored mediation program, began
by the court referring a case to the mediation program. A date was then set for mediation and
each parent was sent a packet of materials to complete and send back to the mediation service or
bring with them to the first meeting. This packet included extensive questions concerning
demographic information, issues that were in dispute, and a proposal to resolve disputed issues.
The first meeting at the mediation program occurred partially in a group format
(orientation), partially with each parent individually (screening) and when determined to be
appropriate with the couple together (mediation). This entire process lasted up to three and a half
hours. First, couples were ordered to appear on a particular day at the courthouse where they
were checked in and then met briefly with a family court judge. Judges on the family court bench
each took turns conducting this brief orientation. The role of the judges was to encourage the
couples to participate in mediation and make decisions for their families, outline the benefits of
doing so and to cite risks of not doing so.
To accommodate IPA victims who may not want to walk in a group with their abuser,
mediation staff accompanied the couples back to the court-connected mediation program offices
in two separate groups. Once the groups returned, the couples were provided a detailed
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 73 2007-WG-BX-0028
orientation to mediation by one of the mediators (mediators rotated this responsibility). Topics
included the emotional nature of divorce for every family member, the negative effects of
conflict on children, introduction to legal terms and the mediation process. The mediator also
covered the basic ground rules of the mediation (listening without interrupting, no threatening
behaviors, and stating limits of confidentiality). Couples were also told that while they were
court ordered to attend the orientation session they were not ordered to reach an agreement. The
couples were also shown a film reinforcing some of these topics.
Once the film was complete, the couple was separated and each parent met individually
with a mediator who conducted a semi-structured clinical interview. It is important to note that
this orientation session was for the mediation process only. This study occurred prior to the law
instituting four hours mandatory parent education for all couples divorcing with children in
Arizona. The orientation described above is not mandatory parent education, which is also
currently conducted at the study site.
An individual portion of this first meeting was a private interview with each parent. The
purpose of the individual, private interview was to determine if mediation was appropriate and if
so the best way to proceed. Issues covered included: if the parent felt comfortable
agreeing/disagreeing with their spouse; if the parent felt that they could advocate for themselves;
major mental health issues; current substance abuse; significant child abuse or IPA; fear of the
other spouse; current law enforcement or child protective services involvement.
Once the screening was complete, the mediator (or mediators in the case of a mediator
team) made a determination as to whether or not to proceed with mediation. For the couples
proceeding, after the individual screening sessions, the couple generally came together for a joint
approximately two hour session.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 74 2007-WG-BX-0028
The individual mediators were given wide latitude in determining the process (whether to
provide individual caucus sessions, whether to include family members in sessions or whether to
interview the children as part of the mediation process).
The facilitative approach was the overarching model of the mediation service. In this
approach the mediators controlled the process of mediation but did not direct clients to make an
agreement or dictate the terms that should be included in an agreement, if any agreement was
reached. While mediators could suggest terms, they did not require specific terms be included in
the parents’ agreements. There were no restrictions placed on the number of mediation; in
practice, however, the range of the number of sessions held was one to eight, with an average of
two (SD=1). Three quarters of the couples received two sessions or less; 92% received three
sessions or less.
If the couple came to an agreement on any parenting time or custody-related issue in
dispute, the mediator drafted the agreement. The couple either stayed until the draft was
complete and reviewed and signed it, or returned another day to review and sign it. If the couple
were pro se, the signed agreement was then sent directly to the court for the assigned judge’s
review and order. If no agreement was reached, the mediation service notified the court. It was
then up to the parties (if pro se) or attorneys to determine the next step. No further
recommendations or information were provided to the court by the mediator. If there was one or
more attorneys involved in the case, the agreement was sent to the attorney[s]. The attorneys
then offered advice to their clients concerning the agreement and whatever the final form of
agreement the attorneys were responsible for drafting it and sending it to the court. Once the
court received a completed agreement, the judge reviewed it. The judge could either modify it or
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 75 2007-WG-BX-0028
make it an order of the court in its original form. Couples only returned to mediation prior to a
decree if they were court ordered to do so.
After the first mediation was completed, 133 couples (13.8%) returned for at least one
more attempt to resolve issues through mediation. And, 24 of the 965 couples in the sample
(2.5%) returned for two or more attempts to resolve disputed issues through mediation. Each
attempt resulted in one or more mediation sessions.
Mediators
All cases from all mediators employed at the mediation service (n=11) during the study
period were included in the sample. Nine of the mediators were full-time employees; two were
doctoral-level student interns. All had at least master’s level training, and one had both a
master’s degree and a law degree. Seven had over 10 years’ experience, one had six years, and
one had two years of experience. There were four males and seven females. Mediators at the
study site received extensive in-house training and supervision prior to mediating cases on their
own. Depending on prior expertise in mediation, the training included mediating with
experienced mediators for three to six months.
Security at Study Site.
The study site had a full-time security guard stationed in the large waiting room at all
times. The guard responded to any requests from the mediators for additional security and
escorted some parents to their cars or bus stops.
Mediator Training in Assessing IPA
All mediators had extensive training in assessing IPA. In addition to the mandatory
training, many of the mediators periodically conducted trainings in IPA awareness for local
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 76 2007-WG-BX-0028
professionals and partnered with local divorce attorneys to conduct information sessions for
victims at local shelters.
Mediator Screening for IPA
Screening for IPA was a two-step process. First, superior court divorce files were
reviewed collectively by the mediators in a group prior to meeting with the parents. Initial
decisions were made concerning whether a mediator team was needed as opposed to a single
mediator. Cases with significant conflict were often assigned to a mediator team or an especially
experienced mediator. There were also discussions concerning whether the parents would need
additional safety precautions while mediating based on allegations or evidence of IPA.
Interviews by research staff with the mediators indicated that these decisions were based on
divorce file information indicating a history of restraining orders, alleged abuse, or high levels of
conflict. Second, the mediators conducted semi-structured interviews with each parent
individually (see Premediation Interview Form below).
Data.
Data examined for this study were extant archival data from several sources: 1) court-
connected mediation service in Pima County, Arizona (Family Center of the Conciliation Court);
2) Domestic Relations Division of the Pima County Superior court files; 3) Tucson Police and
Pima County Sherriff’s Department records, and 4) publically available data from limited
jurisdiction courts in Arizona.
Mediation Data.
A portion of these data were abstracted from existing documents assembled for the
mediation service. Temporary staff hired by the Conciliation Court collected and entered specific
variables directly into an ACCESS database referenced by research case identification numbers.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 77 2007-WG-BX-0028
The mediation program permanent staff also provided additional key variables from database
maintained by the program. After being stripped of identifiers the data was provided to the
research staff (see Attachment A). The databases were then exported into a statistical package for
data analysis (Statistical Package for the Social Sciences, 18). Data outside the study period were
used to create the coding scheme for this study. Several key documents used for data collection
are detailed below.
Pre-mediation Interview Form (PMI)
For stage two IPA screening, mediators conducted a semi-structured clinical interview
(Pre-Mediation Interview (PMI). The interview was administered individually with each parent
in a private meeting before mediation sessions began. Specific information gathered by
mediators on the Pre-Mediation Interview (PMI) related to mental and physical health concerns
and needs of parents and children, concerns surrounding substance use or abuse, concerns about
safety and IPA and any outside support sought (hospital, shelter, police, child protective services
(see Attachment B).
It is important to note several issues relevant to the screening procedure and the use of the
screening forms by the mediators. First, the PMI Forms were a portion of the mediation
program’s routine screening procedures and were not designed specifically for research
purposes. Instead, these PMI forms were provided as a tool for the mediators to use in recording
factors relevant to their decisions about whether to or how to continue mediation. Therefore,
these data represent the individual determinations by mediators as to what was important to them
to note in working on the case. The PMI data were coded as mother or father reporting these
items as occurring in the family as opposed to reporting being the perpetrator or the victim. In
addition, some of the mediators noted timeframes of specific acts, whereas others did not;
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 78 2007-WG-BX-0028
however, interviews by the research staff with the mediators concerning their assessment of IPA
indicated mediators were more focused on current or recent IPA, rather than distal acts.
To capture PMI form data, research staff photocopied the forms. All identifying
information from the forms was redacted and the redacted form was recopied. The mediation
program maintained the original redacted copies in their files. Pre-Mediation Interview forms for
cases outside the study period were also copied, redacted, recopied and used to refine the
database and develop the coding scheme for these variables. The specific items on this form (see
Attachment B) were dichotomized and coded as present or absent. Three raters were trained on
iterative blocks of 20 cases until the interclass correlation coefficient of agreement exceeded .80
on each item. Final inter-rater reliability was calculated as a series of intra-class correlations by
item across the final set of 20 cases for all three coders. Agreement was high across raters for the
31 items, with perfect agreement on 19 items, and a range of interclass correlations .82-.99 on
the others (median=1.0; mean across all items for all coders=0.97). For the final sample of
training cases, the senior rater’s ratings were used for the final code values in the dataset.
Relationship Behavior Rating Scale (RBRS)
After the group orientation, the parents were informed that the office was collecting some
data. They were given the Relationship Behavior Rating Scale (RBRS) and were asked to please
fill it out and return it to the mediator in the private session. The parents were told that the
information contained in the RBRS would not be used by the mediator in their case.
The RBRS was a 41-item revised version of the Partner Abuse Scales (Attala, Hudson, &
McSweeney, 1994) (see Attachment C). Subscales consisted of the Partner Abuse Scale:
Physical (PASPH) and the Partner Abuse Scale: Non-physical (PASNP). The RBRS was
shortened slightly and reworded to be comprehensible to participants at all levels of education.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 79 2007-WG-BX-0028
Participants were asked to rate how often they experienced different forms of abuse, as
perpetrated by their spouse, on a scale of 0–6 (0 = none of the time, 1 = very rarely, 2 = a little of
the time, 3 = some of the time, 4 = a lot of the time, 5 = most of the time, 6 = all of the time).
Once completed, the mediator typically placed the assessment measure in the file without
looking at it and proceeded to conduct a semi-structured clinical interview screening blind to
parent’s scores on the RBRS (see Attachment C). Recently, the RBRS was successfully validated
against the original scales (Beck, Menke, Brewster, & Figueredo, 2009).
Of the total sample (n=965), only 97 (10%) fathers and 77 (8%) mothers did not complete
any of the items on the RBRS. An additional 5 fathers and 7 mothers partially completed the
instrument.
Mediator Checklist.
The mediator checklist detailed mediator notes regarding screening outcomes and
accommodations provided. It included reasons for screening cases out of mediation and listed
additional categories of family problem identified through the screening process (child abuse,
refusal to negotiate, mental health issues, substance abuse, incompetence). If the mediators
decided not to continue mediation due to IPA or any other reason, this was noted on the PMI and
Mediator Checklist forms.
Contact Record.
The contact record detailed mediators’ determinations regarding IPA. In addition, this
record included demographic information (i.e., participant ages, income, previous marriages,
education, and ethnicity).
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 80 2007-WG-BX-0028
Mediation agreements.
Research assistants carefully read each mediation agreement and typed verbatim EXCEL
spreadsheet entries for any language in the agreements relating to physical or legal custody,
parenting time, and any restrictions on parenting or contact between the parents. One research
assistant then content analyzed the data entries and classified them into the following categories:
legal custody (no agreement, sole to mother, sole to father, joint); physical custody (no
agreement, sole to mother, sole to father, joint (could be less that 50/50), split—children divided
between parents or split 50/50 time between parents); restrictions on contact between the parents
(orders of protection; orders restricting method of contact between parents; neutral exchanges of
the children at a specific public location, a family member’s home, or a judicially approved
service provider). The first author then reviewed all coded data against the original language to
provide a reliability check prior to further data analysis.
Divorce File Data.
These data were abstracted from existing documents assembled in the Pima County
Superior Court divorce files. Research staff traveled to the court-connection mediation service to
review and code key variables from an on-line publicly available divorce file database (see
Attachment D). These data were entered directly into an ACCESS database referenced by
research case identification numbers. Several key documents used for data collection are detailed
below.
Divorce Decrees and Parenting Agreements.
Research staff carefully read each divorce decree and parenting agreement and typed
verbatim EXCEL spreadsheet entries for any language in the decree/parenting agreement relating
to allegations of or legal findings of IPA, physical or legal custody, parenting time, and any
restrictions on parenting or contact between the parents. One research assistant then content
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 81 2007-WG-BX-0028
analyzed the data entries and classified them into the following categories: legal custody (no
agreement, sole to mother, sole to father, joint); physical custody (no agreement, sole to mother,
sole to father, joint (could be less that 50/50), split—children divided between parents or split
50/50 time between parents); restrictions on contact between the parents (orders of protection;
orders restricting method of contact between parents; neutral exchanges of the children at a
specific public location, a family member’s home, or a judicially approved service provider). The
first author then reviewed all coded data against the original language to provide a reliability
check prior to further data analysis.
Area Law Enforcement Data
For the two years prior to the first contact with the mediation service, through January
2007, IPA-related incidents and/or arrests recorded in the area law enforcement records were
collected. Calls that involved both the mother and father as either suspect or victim were
collected. Calls with either the mother or father that were listed as something other than the
victim or the suspect were not collected and calls with the children listed as victims were also not
collected (see Attachment E).
Law enforcement data was collected by representatives of each department (Records
Department at the Pima County Sheriff’s Office and representatives of the Domestic Violence
Unit at the Tucson Police Department). Research staff created a worksheet for the representatives
to accurately and efficiently record the calls made by one member of the couple. The
representatives of these agencies filled out the date, type of crime, suspect (father or mother),
victim (father or mother), and whether or not an arrest was made. Once all data on the calls was
collected, research staff entered them into a database.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 82 2007-WG-BX-0028
Limited Jurisdiction Court Data
There was a central repository for many of the courts serving Arizona. The administrator
of the repository was provided names and research identification numbers for the study subjects.
The administrator then sent data appearing in the repository for the designated study period for
the subjects by research identification number only. Unfortunately Pima County Superior Court
was not participating in the data collection efforts during the period of the study. The result was
that any information that was handled by benches other than the Family Relations Bench in Pima
County was not available. Data received from the repository was extremely limited and of little
significance and is therefore not included in the data analyses.
Data-Analytic Strategy for Conceptual Model (Cascade Model)
Previous studies on IPA and mediation have been limited by use of bivariate analyses.
Although bivariate analyses are useful for initial descriptive purposes, the dynamic nature of
mediation and court proceedings and the many interrelated variables requires the additional use
of multivariate analytic approaches. By aggregating theoretically relevant variables across the
entire system, we can disentangle direct and indirect effects of couple, mediator, and procedural
characteristics used in mediation with outcomes from mediation, child maltreatment, divorce,
and post-divorce. We thus developed a conceptual model to guide our inquiry into constructs
leading to mediation, divorce and post-divorce outcomes. Table 2 includes the model constructs,
variable names used in the model, measured variables and data sources.
Table 2.
Constructs, Measured Variables, and Data Sources for Conceptual Model (Cascade Model).
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 83 2007-WG-BX-0028
Construct Variable Name in
Cascade Model
Measured variables Data Sources
Marital
stressors
Stress
Standardized sums of: mental health concerns, physical health limitations or
concerns, hospitalizations, use of medications, substance abuse/use concerns,
use of counseling/therapy, standardized inverse of poverty index (higher
score equals more poverty)
Mediation case
files,
Pre-Mediation
Semi-
structured
Interview
(PMI)
Father
Perpetrator’s
coercive
controlling
behavior
(mother
reporting)
ZFCC
Standardized mean of Father Perpetrator’s coercive controlling behavior
(mother reporting)
Relationship
Behavior
Rating Scale
(RBRS)
Mother
Perpetrator’s
coercive
controlling
behavior
(father
reporting)
ZMCC
Standardized mean of Mother Perpetrator’s coercive controlling behavior
(father reporting)
Relationship
Behavior
Rating Scale
(RBRS)
Father
Perpetrator’s
Interpersonal
Abuse
(mother
reporting)
ZFIPA
Unit-weighted factor score (standardized grand mean of standardized
dimensions) (psychological abuse, coercive controlling behaviors, physical
abuse, threaten and escalated physical violence, sexual intimidation/
coercion/assault)
Relationship
Behavior
Rating Scale
(RBRS)
Mother
Perpetrator’s
Interpersonal
Abuse (father
reporting)
ZMIPA
Unit-weighted factor score (standardized grand mean of standardized
dimensions) (psychological abuse, coercive controlling behaviors, physical
abuse, threaten and escalated physical violence, sexual intimidation/
coercion/assault )
Relationship
Behavior
Rating Scale
(RBRS)
Reported
Child
maltreatment
RCM Single dichotomous item regarding Child Protective Service involvement
with the family as reported by either mother or father
Pre-Mediation
Semi-
structured
Interview
Mediator call
on domestic
violence
present/absen
t
MEDIPA Single dichotomous item regarding whether DV is absent or present Pre-Mediation
Semi-
structured
Interview
Mediation
special
procedures
(safety
accommoda-
tions
MEDPROC Single dichotomous item indicating if special procedures (accommodations
for safety) were used (shuttle mediation, separate screening days, separate
waiting areas, mediation separate days, escort to car, experienced mediator
team, parents required to leave mediation site at separate times)
Mediation case
files
Mediation
Agreements
MEDAGREE Whether there was an agreement reached: None, partial agreement reached,
full agreement reached (0-2)
Mediation case
files
Orders of
Protection
ORDPROT Single item indicating if Orders of Protection found in case and whether it
was merely mentioned without any additional evidence of if there was
substantiation included (specific Orders of Protection Case Numbers or
specific language in mediation or parenting agreement citing language in
Order of Protection)
Mediation case
files and
Superior Court
divorce files
Court Legal
Finding of
Domestic
Violence
COURTIPA Single dichotomous item indicating if legal finding of “domestic violence” in
superior court divorce file (separate motion/order or incorporated into
divorce decree)
Superior Court
divorce files
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 84 2007-WG-BX-0028
Court-
Decreed
Restrictions
COURTRE Sum of restrictions on drop off/pickup, supervised parenting time,
supervised exchanges, if any
Superior Court
divorce files
Court
Custody/Pare
nting Time
Terms of
Decree
Favoring
Mothers
COURTCU Decrees coded for both physical (residential) and legal (decision-making)
custody as follows: -3=father primary physical, sole legal; -2 father primary
physical, joint legal, -1 father joint physical, sole legal; 0=joint physical and
legal; +1=mother joint physical sole legal; +2=mother primary physical,
joint legal; +3=mother primarily physical and sole legal
Superior Court
divorce files
Post-Divorce
Calls to Law
Enforcement
POSTCPW Total number of calls per week from post-divorce to 1/31/2007 between
couple for assault, trespassing, harassment, phone harassment, stalking,
interfering with custody, child abuse, custodial violation, threats, criminal
damage, keeping the peace, disturbing the peace, order of protection,
violation of order of protection, other miscellaneous domestic violence call
Law
enforcement
records for
local police
and sheriff
Post-Divorce
Number of
Hearings
HEARINGS Sum of all the hearings held post-divorce for any reason. Superior Court
divorce files
Post-Divorce
Number of
Decree
Orders
COURTORD Sum of all the orders issued post-divorce for any reason. Superior Court
divorce files
Marital stressors
The marital stressors construct was created using data gathered by mediators in semi-
structured clinical interviews (PMI) with each parent individually. Included were risk factors
determined by the mediation program to be important in determining whether and/or how to
proceed with a case. These factors included: mental health concerns, physical health limitations
or concerns, hospitalizations, use of medications, use of counseling/therapy, and substance
abuse/use concerns. If either the mother or father reported that a factor was present in the
family, the factor was scored as present (1).
At a statistical level, due to the unreliability and likely multicollinearity of individual,
dichotomously scored-items, it was decided that the marital stressor construct would be most
reliably represented in aggregate rather than at the individual item level. Precedent exists in the
methodological literature for the use of this measurement approach (Bollen & Lennox, 1991;
Diamantopoulos & Winklhofer, 2001; Little, Cunningham, Shahar, & Widaman, 2002). Thus,
the factors were summed and standardized to create one indicator.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 85 2007-WG-BX-0028
In addition, reports of mother and father income were summed to produce a family
income variable. This total was divided by the Census 2000 poverty threshold value based on
the family size, then the inverse was taken (after adding 1 to correct for those who reported a 0
income) and the total standardized to produce another marital stress indicator, with a higher
score representing more poverty. The mean of these two indices was taken and re-standardized
to produce an overall marital stress index. Due to the mediation data begin gathered between
1998-2000, the 2000 Census data was used for calculating poverty level.
Interpersonal Abuse (IPA)
Based on prior research (Figueredo & McCloskey, 1993; Tanha, Beck, Figueredo, &
Raghavan, 2010) and factor analyses of the data, the participants’ responses were empirically
clustered into five theoretically-derived dimensions of IPA: coercive controlling behaviors;
1
psychological abuse; physical abuse; threatened and escalated physical violence; and sexual
coercion/intimidation/assault. Preliminary assessment of the RBRS indicated sound
psychometric properties (Beck, Menke, O’Hara Breweter & Figueredo, 2009). Estimates of
internal consistency were generated for each domain using Cronbach’s coefficient alpha.
Examples of each dimension of IPA and the reliability estimates for each dimension, for both
husbands and wives, are given in Table 3.
1
We revised the language used in previously published papers (see Dissemination of Research Section) for this
dimension. In previous papers we referred to this dimension as “coercive control.” We are revising the label for this
to “coercive controlling behaviors.” We made this change to distinguish our use of this dimension from Johnson’s
(2006) “coercive controlling violence,” and Stark’s (2007) “coercive control.” Both scholars inspired our interest in
this construct; however, we do not include physical and sexual violence in our dimension. Our use of the term
“coercive controlling behaviors” is more similar to the construct “coercive control” described by Dutton, Goodman
& Schmidt (2006), albeit much more limited in scope (see Table 3).
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 86 2007-WG-BX-0028
Table 3.
Reliability Estimates for IPA Dimensions
IPA Dimension
# of
Items
Illustrative Items
Cronbach’s
Alpha (father
perp from
mother reports)
Cronbach’s
Alpha (mother
perp from father
reports)
Coercive Controlling Behaviors 10 Did not want me to have
male/female friends; demanded
that I obey him/her; controlled
my coming and going
.84 .80
Psychological Abuse 7 puts me down; insulted or
shamed me in front of others
.91 .90
Physical Abuse 5 pushed or shoved; hit or
punched; bit or scratched
.83 .87
Threatened and Escalated Physical
Violence
12 threatened me with or used a
weapon against me; broke one
or more of my bones; choke,
strangle, suffocate
.87 .79
Sexual Intimidation/Coercion/
Assault
6 demanded sex acts I did not
want; insisted on sex whether I
wanted it or not; physically
forced sex
.83 .66
The items in each subscale were aggregated by taking their mean to provide a score for
the average frequency with which both fathers and mothers reported experiencing each
dimension of victimization in the past 12 months. Furthermore, because our earlier research
demonstrated that four of the dimensions of abuse (psychological abuse; physical abuse;
threatened and escalated physical violence; and sexual intimidation/coercion/assault) could be
seen as indicators of a latent factor representing the overall level of victimization (Tahna et al.,
2010), we also estimated a unit-weighted factor score that was the mean of the subscales
(Gorsuch, 1983; Figueredo, McKnight, McKnight, & Sidani, 2000).
The correlations among the four dimensions of abuse and the overall level of
victimization were high, supporting this approach (see Table 4).
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 87 2007-WG-BX-0028
Table 4.
Correlations of Standardized IPA Dimensions and IPA Common Factor Score
IPA Dimensions
Psychological Physical Escalated Sexual
Father Perpetrator (mother reporting)
Psychological Abuse
Physical Abuse .528
Threatened and Escalated Physical Violence .562 .840
Sexual Intimidation/Coercion/Assault .448 .494 .496
IPA Common Factor .776 .871 .887 .747
Mother Perpetrator(father reporting)
Psychological Abuse
Physical Abuse .500
Threatened and Escalated Physical Violence .493 .713
Sexual Intimidation/Coercion/Assault .339 .404 .429
IPA Common Factor .747 .838 .844 .695
Note: N=863-866. All correlations significant at p<.001
Based on theoretical considerations and prior work (Tanha et al., 2009), coercive
controlling behaviors was considered a separate construct. We hypothesized, and confirmed
using structural equations modeling, that control is the motivator for the other dimensions of
IPA, with control of the victim as the goal. It appears that when coercive controlling behaviors
fail, physical dimensions of IPA are then used in service of the goal of control.
Considerations about the Measurement of IPA
In the study of any psychological trait, there has always been a certain degree of tension
between those researchers who favor using latent dimensions, which are continuous variables,
and those researchers who prefer typologies, which are discrete latent classes. Mayr (1976)
described how this tension extends beyond the behavioral science and is generalized to the
biological sciences as a whole, distinguishing typological thinking from populational thinking.
Some statisticians have argued that dichotomizing (splitting into two categories) a continuous
variable simply cannot be justified. The serious consequence of this dichotomizing is a
degradation of measurement. Approximately one-fifth to two-thirds of the variance that could
possibly be accounted for on the original variables is lost, which is equivalent to discarding one-
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 88 2007-WG-BX-0028
third to two-thirds of the sample (Cohen, 1983). Others (Meehl, 1965) have long argued that it
might be possible to justify the construction of latent taxa (meaningful categories) quantitatively
if the existence of discrete groups is in fact suggested by the data. There is some agreement that
the dichotomizing, or polytomizing (splitting into multiple categories), of our measurement is
something that needs to be empirically, and not just philosophically, justified, with specific
recourse to the patterns that are observed (or not) in the data.
In the case of the measurement of IPV, many researchers have favored a continuous and
dimensional form of measurement and others have favored the development of a typology. Those
who favor continuous measurement will also often use Confirmatory Factor-Analytic (CFA)
methods to construct a single latent variable underlying all the different behavioral
“manifestations” of IPA. Those who favor discrete measurement will instead often use cluster-
analytic or Latent Class-Analytic (LCA) methods to construct a typology that sorts the different
“forms” of IPA into different combinations, sometimes organized by the different levels with
which each of the specific behaviors encompassed by the overarching concept of IPA is engaged
in by the men and the women within each couple.
The statistical problem in making such a determination of optimal measurement is that a
natural typology can sometimes generate what appears to be a continuous dimension, and a
natural continuous dimension can sometimes generate what appears to be a typology. One
therefore needs to develop and test a generative causal theory regarding the specific relationship
that one expects might exist between the appearance and the presumably underlying reality.
Although much work remains to be done on this thorny issue, we examined one such
theoretical point of view with our data. We used a pragmatic approach to determine which
representation of the data was actually more explanatory in accounting for the observed variance
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 89 2007-WG-BX-0028
in our measures. We proposed that the single latent dimension might be the underlying causal
process at work, and that differences along this latent dimension gave rise to the various
combinations of IPA by fathers and mothers proposed for the mediation population by Kelly &
Johnson (2008) (Johnson, 2006). As described in detail above, the Kelly and Johnson typology
was:
1. Coercive Controlling Violence
2. Violent Resistance
3. Situational Couple Violence
4. Separation-Instigated Violence
5. Mutual Violent Control
In collaboration with our colleague Dr. Edward Anderson from the University of Texas at
Austin, for the current data we partially replicated this taxonomic scheme using Latent Class
Analysis (LCA).
Latent Classes
Latent Class Analysis (LCA) (McCutcheon, 1987; Muthen & Muthen, 2001) was used to
identify types of couples that are similar in their responses to the RBRS. LCA has been used in a
wide variety of substantive research areas (Hagenaars & McCutcheon, 2002; Nylund,
Asparouhov, & Muthen, 2007). We limited the LCA to those couples in which both partners had
at least partial completion of the RBRS. In 54 couples, both wives’ and husbands’ reports were
missing; 23 couples were missing the wives’ reports, and 43 were missing the husbands’ reports.
The sample was thus reduced to N = 845 (87% of the original 965 couples). Of the 845 couples
with both partners providing data, 12 couples had one member providing only partial data; these
couples were included in the LCA.
To facilitate comparisons across ratings from women and men, standardized scores were
computed for each RBRS subscale based on the pooled mean and standard deviation from
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 90 2007-WG-BX-0028
women’s and men’s reports. Further, because the distributions for the variables assessing
physical abuse, threatened and escalated physical violence, and sexual intimidation/coercion/
assault were highly skewed, a constant value of .5 was added to each score and scores were then
log-transformed.
We first conducted the LCA on the complete sample of N = 845. We began with a model
estimating two latent classes, increasing the number of latent classes in subsequent models until
there was evidence that increasing the number of latent classes further would result in
deterioration of fit (Nylund et al., 2007). To test the robustness of the final solution, the sample
was split into 50% random subsamples, and the LCA was conducted on both subsamples to
determine whether the solution replicated. Reliability was evaluated by comparing the
classification results of the split-half and combined latent class solutions.
Although we obtained five latent classes, the classes were somewhat different than those
types of IPA proposed by Kelly and Johnson (2008) and Hardesty and Ganong (2006). Our
classes were:
6. Mutually low (or no) IPA.
7. Lower Level Coercive Controlling Violence—father perpetrator (mother reports)
8. Coercive Controlling Violence—mother perpetrator (father reports)
9. Coercive Controlling Violence—father perpetrator (mother reports)
10. Mutual Violent Control—(father perpetrator (mother reports); mother perpetrator
(father reports).
To better understand these classes we identify the perpetrator of the IPA behaviors that
were elevated in the different classes. We want to emphasize that data collected in the study was
victimization data. In other words, mothers reported fathers’ perpetration of specific behaviors;
fathers’ reported mothers’ perpetration specific behaviors.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 91 2007-WG-BX-0028
Mutually low (or no) IPA.
This first type comprised 37% of the sample. Both mothers’ and fathers’ IPA behaviors
were below the mean (pooled mean of mothers and fathers) on all dimensions of IPA (coercive
controlling behaviors, physical abuse, threatened and escalated physical violence, and sexual
intimidation/coercion/assault).This is not to say that there was no IPA in all of these
relationships. The LCA only indicated that the reported levels were below the mean reported
levels for mothers’ and fathers’.
Lower level coercive controlling violence—father perpetrator (mother reports).
The second type comprised 29% of the sample. The elevations on the dimensions of IPA
reported were for father perpetrator on coercive controlling behaviors, psychological abuse, and
elevations but lower elevations on threatened and escalated physical violence and sexual
intimidation/coercion/assault. Interestingly for both mother and father, there were no elevations
on the physical abuse dimension. Reports of perpetration on all other dimensions for both
mothers and fathers were again below the mean. This type did not appear in the typology by
Kelly and Johnson (2008). It did, however, appear in the Hardesty and Ganong (2006) study.
Coercive controlling violence—mother perpetrator (father reports).
The third type comprised 17% of the sample. There were elevations on all dimensions of
mother and father perpetrator IPA; however, the reported levels for father perpetrator were near
the mean while the mother perpetrator elevations were not. The only dimension of IPA below the
mean was mother perpetrator of sexual intimidation/coercion/assault. This type replicates the
type of a similar name proposed by Kelly and Johnson (2008) with the caveat that the higher
elevations are for mother perpetrators (as reported by fathers).
Coercive controlling violence—father perpetrator (mother reports).
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 92 2007-WG-BX-0028
The fourth type comprised 13% of the sample. There were very high elevations on all
dimensions of father perpetrated IPA reported by the mothers. Interestingly, mother perpetrated
physical abuse was also elevated, albeit at a much lower level than father reportedly perpetrated.
And, mother perpetrated sexual intimidation/coercion/assault was well below the mean. This
type replicates the type of a similar name proposed by Kelly and Johnson (2008).
Mutually violent control (mother reports of father perpetration; father reports of mother
perpetration).
The fifth type comprised 4% of the sample. There were extremely high elevations on all
dimensions of IPA reported by both mothers and fathers. This type replicates the type of a
similar name proposed by Kelly and Johnson (2008).
An Alternative Causal Hypothesis
Two alternative construals of the typological scheme produced by LCA were therefore
tested and compared, both to each other and to a dimensional representations. These analyses
were based on the hypothesis that these five taxa could be rank-ordered by degree of severity of
the IPA expressed, as assessed for the couple, and that the progression through each of the
behavioral types was driven by this severity of IPA dimension.
Of course, it is trivial and tautological to say that types of behavior are generated by
severities of behavior, per se. We presumed that the severity dimension was more than a mere
description of behavior, but instead represented the degree of aggressiveness that either or both
members of the couple may possess as a latent psychological trait or behavioral disposition. This
is similar to the approach used in Item Response Theory (IRT), where the number of correct
responses on a test is used to infer the degree of latent ability (in this case aggressiveness) that
each individual has in the content area being assessed. In IRT, this latent ability dimension is
referred to as Theta (θ).
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 93 2007-WG-BX-0028
In IRT, the proportion of correct responses on any single item across a sample of test-
takers is used to infer the item difficulty, where the items having a very low proportion of correct
responses in the sample is presumed to have a higher difficulty than the items having a very high
proportion of correct responses. This is because most of the test-takers can answer the “easy” (in
this case lower levels of IPA) questions correctly, whereas few of the test-takers can answer the
“difficult” (in this case high levels of IPA) questions correctly. This can then be used to relate the
item difficulties to θ, where it only takes a certain level of θ (latent ability) to answer the “easy”
questions correctly, whereas it only a higher level of θ (latent ability) to answer the “difficult”
questions correctly.
Of course, in IPA research, we are hardly comparing either individuals or couples in their
abilities to perform the behavior, but rather in their dispositions to do so. We expect that
individuals or couples with lower levels of aggressive disposition (θ) will disproportionately
perform behaviors that are lower in severity, whereas individuals or couples with higher levels of
aggressive disposition (θ) will disproportionately perform behaviors that are higher in severity.
Simply put, it takes less latent aggressiveness to verbally insult a partner than to assault one with
a deadly weapon, such as a firearm (White et al., 2000).
This causal theory implies that the membership of individuals or couples in behaviorally-
defined IPA types is a straightforward reflection of the relative severities of the specific IPA
behaviors defining each of the types. Engaging in these behaviors of different degrees of
assignable severity is ultimately a reflection of the relative degrees of latent dispositional
aggressiveness characterizing the individuals or couples so classified. The testable hypothesis
that follows logically and inevitably from this causal theory is that the taxa produced by LCA, if
theoretically rank-ordered on their relative degrees of severity, should be highly correlated with
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 94 2007-WG-BX-0028
the scores estimated by a Confirmatory Factor Analysis for the members of each taxon on the
latent continuous dimension of aggressive disposition.
We performed these analyses and found that this was indeed the case, which constituted
tentative support for our hypothesis (see Figure 2). Of course, we have conducted only this one
test of this particular causal hypothesis and much further work remains to be done to be able to
draw more definitive conclusions regarding the optimal measurement of IPA. Nevertheless,
similar results were obtained whether the five latent classes were compared to the unitary
common factor constructed for IPA, or to each of the five component subscales of the IPA factor
(coercive controlling behaviors, psychological abuse, physical abuse, threatened and escalated
violence, and sexual intimidation/coercion/assault).The latter correlations were slightly lower
due to a couple of possible reasons: lower frequencies of occurrence of the behaviors; or, either
the unitary common IPA factor or the five component subscales were disaggregated according to
the sex of the perpetrator.
We also found that these relations did not change appreciably, whether or not coercive
controlling behaviors was included or excluded from the factor model (Figure 3). We did this
additional sensitivity analysis because some previous research had indicated that coercive
controlling behaviors might not really be an indicator of IPA per se, but rather the principal
motivating factor behind it (Figueredo & McCloskey, 1993; Figueredo et al., 2001; Figueredo et
al., 2009; Tanha et al., 2010). We thus chose this way to represent it in Goal 4, the Cascade
Model.
Furthermore, the correlations among the dimensions of IPA perpetrated by the fathers
and mothers within each couple were lower than has typically been reported. For example, the
structural relation among the unitary common IPA factors between partners was only β=.33
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 95 2007-WG-BX-0028
(p<.0001), whereas in other samples it had been found to be as high as β=.54 (p<.0001), as
reported by women, and β=.63 (p<.0001), as reported by men (Figueredo et al., 2009). However,
the previous findings were from intact couples rather than those in the process of separation and
divorce, as is true in the current sample.
Therefore, we have opted to use the continuous and unitary common factor we
constructed for IPA, disaggregated by the sex of the perpetrator, and including the coercive
controlling behaviors subscale as a separate predictor, in our Cascade Models, as the
provisionally more parsimonious representation.
Figure2.Latent classes and the unit-weighted factor score (mean of the subscales) comprising
the continuous measure of IPA (with coercive controlling behaviors included).
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 96 2007-WG-BX-0028
Figure3. Latent classes and the unit-weighted factor score (mean of the subscales) comprising
the continous measure of IPA – no coercive controlling behaviors included in continuous
measure of IPA.
Child Maltreatment
Child abuse and neglect are extremely important in terms of divorce-related decision
making. Because of the archival nature of the data, we were unable to investigate fully the
nature, if any, of child abuse allegations in the sample. We did, however, want to put into the
model a measure of reported child maltreatment. One variable available in the data was an
indication of Child Protective Service involvement with the families. Therefore, included in the
model was a single dichotomous item regarding reported child maltreatment as reported by either
mother or father.
Mediator Determination of IPA Present/Absent
Accurately identifying a case as having IPA in mediation is essential to determining if
accommodations to the process are desired or necessary to protect the victims’ safety and to
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 97 2007-WG-BX-0028
including safety precautions in agreements if any are reached. The data collected included a
single dichotomous item wherein mediators identified a case as IPA present or absent.
Mediation Special Procedures
Special procedures have been suggested as a way to safely accommodate couples with
IPA in their relationship in a mediation context. There were a number of accommodations
available to mediation couples at the study site. Those included: providing a parent an escort to
their transportation, providing parents separate rooms and conducting “shuttle” mediation
sessions on separate days, mediating with an experienced team of mediators as opposed to just
one mediator, providing separating waiting rooms for parents waiting to mediate, and requiring
one parent to stay behind allowing the other parent to reach transportation and leave the area
before the second parent was allowed to leave the mediation program site. It was possible that
parents could be provided one or more of these accommodations. The model included one single
dichotomous item indicating if special procedures were used in any of the mediation sessions
attended by the parents.
Mediation Agreements.
A frequent finding in the literature is that there are no differences in levels of IPA
between those couples who reach agreement and those who do not. To better understand the
influences on a couple’s decision to reach an agreement or not, we included in the model a
variable indicating whether there was no agreement, a partial agreement or a full agreement (0-
2).
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 98 2007-WG-BX-0028
Orders of Protection.
Although controversial, orders of protection have been found to be an important method
by which victims can protect themselves from an abusive and/or violent partner (Logan, 2009).
Therefore, we included this variable in the model.
Court Legal Finding of “Domestic Violence.”
Many state statutes, including the study site, include a provision stating that if the court
makes a finding that there is significant domestic violence in the relationship, joint custody will
not be awarded (Ariz. Stat. Ann. §25-403.03).And, the court is to consider evidence of domestic
violence as being contrary to the “best interests of the child”—the legal standard for determining
custody. In cases where domestic violence is present the court is to consider the safety and well-
being of the child and the victim as primary. We therefore included in the model a variable
indicating if the court had made such a legal finding.
Decree Restrictions in Parenting Plans.
Research has indicated that post-divorce, the structure of parenting agreements can be
associated with continued IPA. It is during the times couples meet face-to-face to handle child-
related activities that IPA occurs. It is argued that perpetrators use these times to gain access to
the victim to continue the abuse. We thus included in the model is a variable summing the
number of restrictions on contact between the parents and/or parents and children found in
parenting agreements (orders of protection; orders restricting method of contact between parents;
neutral exchanges of the children at a specific public location, a family member’s home, or a
judicially approved service provider).
Terms of Parenting Agreements.
Research has also indicated that joint physical and/or legal custody, which requires close
contact and negotiation, is not advisable when there is IPA in the relationship. We thus included
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 99 2007-WG-BX-0028
in the model a variable capturing the range of options available for physical and/or legal custody
from that which would completely favor the mother (primary physical, sole legal) to that which
would equally and completely favor the father.
Post-Decree Calls to the Law Enforcement.
Calling the law enforcement to intervene in a difficult family situation is another means
for victims to protect themselves from an abuser. Thus, we included in the model a variable
summing the number of calls made to law enforcement per week from the date of the divorce
decree to January 31, 2007 (the beginning of the grant period). While all contacts with the law
enforcement are important, the local law enforcement agencies required we limit the types of
calls so that the data collection would be manageable in the time frame needed and with the
personnel available. The types of calls included in the variable included only those occurring
between the couple and related to IPA or problems with custody and/or parenting time issues
(assault, trespassing, harassment, phone harassment, stalking, interfering with custody, child
abuse, custodial violation, threats, criminal damage, keeping the peace, disturbing the peace,
order of protection, violation of order of protection, and other miscellaneous domestic violence
calls).
Number of Hearings Post-Decree.
Research indicates that continued use of the legal system to further litigate settled issues
or raise new issues is one method by which an abuser can continue to harass, intimidate and
drain financial resources from a victim. An important assumption underlying mediation is that
when negotiate an agreement in mediation, they are less likely to require post-divorce litigation.
Thus, included in the model was a variable representing the sum of all hearings held post-
divorce.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 100 2007-WG-BX-0028
Number of Orders Post-Decree.
Although issues can be raised post decree, judges can dismiss the issues raised. As a
measure of post-decree issues of substance, we included in the model a variable summing the
number of post-decree judicial decisions.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 101 2007-WG-BX-0028
Results
Statement of results
We first systematically examined the bivariate relationships between IPA outcomes
throughout the mediation, divorce and post-divorce processes perpetrated by the father (as
reported by the mother), and perpetrated by the mother (as reported by the father). We then built
a multivariate Cascade Model to examine the direct and indirect effects of IPA as well as the
intermediate societal responses on various outcomes.
Overview of IPA dimensions in the sample
Table 5 provides the raw means and standard deviations of IPA on the following
dimensions: coercive controlling behaviors; psychological abuse; physical abuse; threatened
and escalated physical violence; and sexual/intimidation/coercion/assault; and the IPA Common
Factor. IPA perpetrated by both the father (as reported by the mother), and by the mother (as
reported by the father) are presented. Mothers reported more frequent victimization across all
dimensions of IPA compared to fathers, with the exception of the physical abuse dimension,
where there was no difference between mothers’ and fathers’ reports (Table 5). Psychological
abuse was the most frequently reported dimension of IPA with mothers rating the average
frequency just slightly more than ‘some of the time” (3.04) and fathers reporting the average
frequency between ‘a little of the time” and ”some of the time” (2.53).
2
The average frequency of mothers’ and fathers’ reports on the other IPA (coercive
controlling behaviors; psychological abuse; threatened and escalated physical violence; and
sexual/intimidation/coercion/assault) contributing to the IPA Common Factor was fairly low
2
Recall that participants were asked to rate how often they had experienced a number of different forms of abuse on a scale of 0-
6 (0=none of the time, 1=very rarely, 2=a little of the time, 3=some of the time, 4=a lot of the time, 5=most of the time, 6=all of
the time). The subscales represent the means of those items, theoretically combined by sub-dimension.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 102 2007-WG-BX-0028
(with a mean less than 1, or ‘very rarely”). However, well over half of the mothers’ reported that
fathers’ perpetrated at least one incident on each other dimension of IPA (coercive controlling
behaviors; psychological abuse; threatened and escalated physical violence; and
sexual/intimidation/coercion/assault), and that it occurred in the 12 months prior to the
mediation. Approximately half the fathers also reported an incident on the dimension threatened
and escalated physical violence and approximately one third of the fathers reported at least one
incident of sexual intimidation/coercion/assault in the prior 12 months. Only four couples in the
sample reported no incidents of any sort of IPA in the 12 preceding months.
For the ease of interpretation, the Common Factor IPA scores are presented as
standardized as T-scores in the analyses below. This allows for comparison between the father’s
and mother’s reports to account for the differences in reporting. For both sexes, the mean is
standardized at 50 with a standard deviation of 10.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 103 2007-WG-BX-0028
Table 5.
Raw Means and Standard Deviations of IPA reports by IPA Subscales.
IPA subscales and
common factor
Father perpetrator (mother
report)
Mother perpetrator (father
report)
Mothers
reporting at
least one
incident in
preceding 12
months (n)
Mean
1
(sd)
Fathers
reporting at
least one
incident in
preceding 12
months (n)
Mean
1
(sd)
Paired
t-test
(df)
Coercive
Controlling
Behavior
97.6%
(814)
2.38
(1.39)
97.4%
(812)
1.94
(1.14)
8.53*
(841)
Psychological
Abuse
98.2%
(819)
3.04
(1.51)
97.1%
(810)
2.53
(1.40)
7.79*
(844)
Physical Abuse
57.7%
(481)
.60
(.89)
54.4%
(454)
.58
(.91)
.54
(834)
Threatened and
Escalated Violence
61.5%
(513)
.42
(.66)
50.1%
(418)
.21
(.41)
8.97
*
(835)
Sexual/Intimidation/
Coercion/Assault
55.6%
(463)
.62
(.93)
29.6%
(247)
.19
(.46)
12.74
*
(836)
Common Factor
IPA
98.4%
(821)
1.17
(.80)
97.4%
(812)
.88
(.64)
10.15
*
(833)
*p<.001; numbers in parentheses are numbers of participants.
1
Mean of items scored 0-6 (0=none of the time, 1=very rarely, 2=a little of the time, 3=some of the time, 4=a lot of the time,
5=most of the time, 6=all of the time).
As noted above, we developed four Goals that guided the current research. Following are
the study results which addressed each of the proposed Goals.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 104 2007-WG-BX-0028
Goal 1: Determine whether the mediation program accurately identifies couples with
self-identified IPA and assess whether these cases are treated differently. To do so we:
1. Evaluated whether couple level questionnaire reports of IPA corresponded to
mediator determinations of IPA-present or absent based on semi-structured
clinical interviews;
Following a semi-structured clinical interview with each of the parents individually,
mediators identified over half of the couples in the sample having IPA in the relationship (59%;
N=569). Client self-identified IPA was related to mediator’s determination of IPA, whether
based on the mothers’ reports of fathers’ perpetration (F=140.90 (1,885), p<.001) or father’s
reports of mothers’ perpetration (F=70.59 (1,864), p<.001). Cases identified by the mediator as
involving IPA had levels of both father and mother IPA above the mean on scaled scores (see
Figure 4).
53.05
45.50
52.28
46.7
30.00
40.00
50.00
60.00
70.00
findingofIPAbythe
mediator
nofindingofIPAby
themediator
LevelofReportedIPA(Tscore)
FatherIPAperpetration
Tscore
MotherIPA
perpetrationTscore
Figure4. Cases identified by the mediator as IPA present/absent.
There were also a number of cases exhibiting non-negligible levels of IPA that were
incorrectly identified by mediators (i.e., false negatives). Figure 5 provides the proportion of
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 105 2007-WG-BX-0028
couples reporting at least one incident of either threatened an escalated violence or sexual
intimidation/coercion/assault in the preceding 12 months, by whether or not mediators identified
IPA in their relationship. Approximately one-third of the couples classified as non-IPA by the
mediator reported at least one incident of threatened and escalated violence or sexual
intimidation/coercion/assault on the questionnaire.
Figure5. Reports of at least one incident of either threatened and escalated physical violence or
sexual intimidation/coercion/assault in the preceding 12 months by mediator identified/not
identified IPA.
0
10
20
30
40
50
60
70
80
90
100
reportedatleastone
incidentofescalated
violence
reportedatleastone
incidentofsexual
intim/coer/assault
reportedatleastone
incidentofescalated
violence
reportedatleastone
incidentofsexual
intim/coer/assault
mediatorIPAyes mediatorIPAno
Fatherperpetrated
Motherperpetrated
Figure 6 illustrates the range of the T-scores for parent-reported IPA by whether or not
IPA was identified by the mediator. The line represents scores 2 standard deviations above the
mean IPA score (70). Cases marked by ‘1’ were provided safety accommodations during the
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 106 2007-WG-BX-0028
mediation due to safety concerns (see below). For those fathers and mothers completing the
behaviorally specific questionnaire and combining all dimensions for a total score for each
parent, 49 mothers (n=887; 6%) reported victimization that was frequent and/or severe enough
categorize their abuse as at or above 2 standard deviations above the sample mean; 37 fathers
(n=866; 4%) reported experiencing a comparably severe/frequent level of IPA. Although the
majority of severe IPA cases were accurately identified by mediators as presenting IPA, and all 7
of the cases where couples reported mutual high levels of abuse were also identified, 6 of the
father-perpetrated cases (12%) and 2 of the mother-perpetrated cases (5%) were not identified as
having IPA by the mediator.
Figure6. T-scores for client-reported IPA by whether IPA was identified or not by the mediator.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 107 2007-WG-BX-0028
The decision to screen couples out of mediation was left to the mediator. Sixty-two
couples were screened out of mediation (6%). Thirty-two couples (3%) were screened out of the
mediation process after it had begun. Thirty couples (3%) were screened out at the intake
interview.
Of the 30 couples screened out at intake, 27 couples (3%) were screened out of the
mediation process because of IPA concerns identified by the mediator (though, four of these
couples were not initially identified by the mediator as having IPA). Of the 27 couples screened
out at intake for IPA, the mothers who were screened out reported higher levels of victimization
on average (m=59.02) compared to mothers who remained in mediation (m=49.77) (F=18.74 (1,
855), p<.001). However, there was not a statistically significant difference between victimization
reports of fathers who were screened out (m=49.81) and those who remained in mediation
(m=50.0) (F=.01(1,864), p=.932). Among those with the highest levels of IPA (> 2 standard
deviations above the mean), only 25% (N=12) of the father perpetrated cases were screened out;
only 9% (N=3) of the highest level mother-perpetrated cases were screened out.
Couples who were screened out of mediation, regardless of whether they were screened
out for IPA at intake of after mediation began for IPA (N=50) or other reasons (N=12, total
N=62), were not included in the subsequent analysis of mediation outcomes.
2. Determined the strength of the relationship between mediator determinations of
IPA and two external sources of data-- law enforcement records of family
conflict, abuse, and violence and Superior Court Divorce case files.
Data collected from area law enforcement (sheriff and police) detailed in Table 6.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 108 2007-WG-BX-0028
Table 6.
Calls to Law Enforcement
Type of Call
Number of
Calls
Mother
Victim
Father
Victim
Unknown
Victim
Other Miscellaneous DV 234 163 58 1
Interference With Custody 160 62 88 10
Assault 157 105 63 0
Order of Protection 154 123 20 11
Keeping the Peace 97 34 22 41
Violation of Order of Protection 56 30 26 0
Criminal Damage 54 36 19 0
Threats 34 23 11 0
Disturbing the Peace 32 21 7 4
Harassment 21 10 10 1
Trespassing 17 11 6 0
Phone Harassment 17 13 4 0
Child Abuse 11 6 0 5
Stalking 1 0 1 0
Totals 1045 637 335 73
Of the total number of calls with a specific alleged victim, in the nearly double the
number of calls was made with the mother victim (61%) as compared to the father (32%). Seven
percent of the cases the victim was not indicated in the law enforcement records. There data
available regarding three types of calls could be reliably identified as likely related to family
conflict: those concerning partners; those concerning custody; and, those concerning property
damage. In nearly 40% (39.2%; N=378) of the families, at least one member of the couple made
a call to the law enforcement at some point during one of the three time periods addressed. The
three time frames were: calls recorded up to two years prior to filing the divorce petition (before
petition); calls recorded between the filing of the petition and the divorce decree (during
divorce); and those recorded between the time of the decree and the end of the study period
(post-divorce). Calling the law enforcement before the petition was correlated with calling
during divorce, and calling during divorce was correlated with calling post-divorce at p<.001
(see Figure 7).
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 109 2007-WG-BX-0028
Figure7. Relationship among calls to law enforcement at different time periods.
The range of calls per couple varied considerably from 0-7 before the petition, 0-10
during the divorce and 0-21 after the divorce. The majority of families did not seek law
enforcement intervention at any period; however, there were a minority of couples who sought
multiple interventions by area law enforcement (see Table 7).
Table 7.
Time Periods for Calls to Law Enforcement
Range
0 Calls 1 Call >1Call
Calls before petition 0-7 84%
(812)
11%
(107)
5%
(46)
Calls during divorce 0-10 83.5%
(806)
8.4%
(81)
8.1%
(78)
Calls after divorce 0-21 85%
(817)
8%
(78)
7%
(70)
Note: numbers in parentheses are the number of couples.
The breakdown of when calls were made was nearly evenly distributed over the time
periods: 16% of cases (153) had a record of any calls before petition, 16.5% (159) during the
divorce process, and 15% (148) afterwards. Levels of father-perpetrated IPA and of mother-
perpetrated IPA were both correlated with law enforcement reports at all three time points,
though not strongly (see Table 8).
Calls
before
petition
Calls
during
divorce
Calls
post-
divorce
.15 .31
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 110 2007-WG-BX-0028
Table 8.
Correlation of Calls to Law Enforcement with Client-Reported IPA
T score of father
perpetrated violence
(N=887)
T score of mother
perpetrated violence
(N = 866)
number of calls before petition .135
**
.098
**
number of calls during the divorce process .155
**
.150
**
number of calls after decree date .094
**
.066
**Correlation is significant at the 0.01 level (2-tailed).
In addition, those couples who mediators identified as having IPA in the relationship had
a higher average number of law enforcement calls at all three time points compared to those not
identified as having IPA (see Table 9). Those with the highest numbers of calls all fell within
the IPA-identified group.
Table 9.
Average Calls to Law Enforcement by Whether the Mediator Identified IPA
N M SD Minimum Maximum
number of calls before petition
F=30.29 (1,963)
p<.001
No IPA 396 .0985 .41843 .00 4.00
IPA 569 .3322 .76938 .00 7.00
Total 965 .2363 .65859 .00 7.00
number of calls during the divorce
process
F=18.36 (1,963)
p<.001
No IPA 396 .1818 .65779 .00 8.00
IPA 569 .4728 1.23464 .00 10.00
Total 965 .3534 1.04688 .00 10.00
number of calls after decree date
F=11.94 (1,963)
p=.001
No IPA 396 .1919 .77853 .00 9.00
IPA 569 .4868 1.56893 .00 21.00
Total 965 .3658 1.31141 .00 21.00
We compared data obtained by the mediator in a semi-structured clinical interview
concerning whether police were called and/or arrests were made at any time in the relationship to
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 111 2007-WG-BX-0028
data obtained from area law enforcement. Although fewer cases appeared in law enforcement
databases than were reported in interviews, mothers’ and fathers’ reports of law enforcement
involvement were highly correlated with each other in both the mediation context (r=.76;
N=934) and in the law enforcement data (r=.367; for mothers; r=.384 for fathers).
We also examined the number of couples reporting orders of protection. In the context of
the mediation interview, 42% (N=421) of cases reported receiving an order of protection.
Slightly more mothers reported receiving orders of protection at some time in the relationship
37% (N=353) compared to fathers 36% (N=345); however, there was a strong relationship
between mothers and fathers reporting orders of protection (r=.84; p<.001). Of the total cases
with self-reported orders of protection, 56% had no further substantiation of the order in the legal
files. In 25% of the cases (N=106) there was further substantiation of the order found in the case
files reported by either mother or father.
We also examined whether orders of protection were more likely to be found among
those reporting higher levels of IPA and among mediator identified cases of IPA. Both types of
IPA indicator (parent-identified and mediator-identified), predicted having an order in the
mediation case file. For parent-identified IPA, there were differences in IPA levels among those
with and without orders of protection for both father (F=69.82 (2, 884), p<.001) and mother-
perpetrated IPA (F=21.35 (2,863), p<.001). There was also a statistically significant difference in
mean levels of IPA between those with unsubstantiated versus those with substantiated
protection orders (p<.001) for father- but not mother-perpetrated (see Figure 8).
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 112 2007-WG-BX-0028
46.96
53.28
56.74
48.21
52.00
53.84
30.00
40.00
50.00
60.00
70.00
OOPnot
mentioned
OOPmentioned,
butnot
substantiated
OOPmentioned
and
substantiated
FatherIPAperpetrationT
score
MotherIPAperpetrationT
score
Figure8. Orders of Protection by Client-Identified IPA
The proportion of cases with orders of protection also differed depending on whether or
not the mediator identified IPA (χ
2
=127.09, df=2, p<.001). There was, however, no difference
between whether the order of protection was substantiated or not (p=.132) (see Table 10).
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 113 2007-WG-BX-0028
Table 10.
Orders or Protection by Mediator-Identified IPA
Mediators call of DV
No Yes
Order of protection not mentioned
308 (78%) 236 (42%)
Order of protection mentioned but not substantiated 72 (18%) 243 (43%)
Order of protection mentioned and substantiated 16 ( 4%) 90 (16%)
Total 88/396 333/569
3. Determined the strength of the relationship between mediator determinations of IPA-
present or absent and implementation of special procedures (considered safety
accommodations) in mediation.
It is important to note that there was no systematic data collection by the mediation
program staff concerning the number of or types of accommodations provided to couples
participating in mediation. Interviews with mediators indicated that several mediators considered
accommodations to be “mediation techniques” as opposed to safety accommodations. Therefore,
the following results concerning safety accommodations are likely to seriously underestimate of
the actual number of accommodations provided to couples during this period.
With that caveat in mind, there were 6 different types of accommodations provided to
couples (parents leave at different times; security escort to car; separate waiting rooms; screening
on different days; shuttle mediation; experienced team of mediators). Approximately one fifth of
couples (18.5%; N=179) were provided safety accommodations. Seventy-four percent of the
couples provided accommodations were given one accommodation whereas the remaining 26%
were provided more than one accommodation. Among the 75% of couples receiving one
accommodation, the most common type provided (64%; N=114) was to require one parent leave
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 114 2007-WG-BX-0028
the mediation program offices ahead of the other parent, separated by a period of time. The next
most common accommodation provided (3%; N=6) was mediation services by an experienced
male and female mediator team; followed by (2%; N=4) mothers provided a security escort to
her car; (2%; N=3) couples were provided shuttle mediation; (2%; N=3) couples were provided
screening on separate days; and 1%; (N=2) couples were provided separate waiting rooms.
The remaining 26% of the couples were provided more than one type of accommodation
(22% were provided two and 4% were provided three). Accommodations tended to be made for
those couples reporting higher levels of IPA, whether perpetrated by the father (F=91.064
(1,885), p<.001) or the mother (F=28.71 (1, 864), p<.001). However, fewer than half of the
couples reporting the highest levels of IPA (70 or higher) were provided accommodations: 21
(43%) of those experiencing high levels of father-perpetrated IPA; 16 (43%) of those
experiencing high levels of mother-perpetrated IPA; and only two (29%) of the seven couples
who reported mutually high levels of IPA.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 115 2007-WG-BX-0028
56.34
48.53
53.78
49.15
30
40
50
60
70
mediationaccomodations nomediation
accomodations
FatherIPAperpetrationT
score
MotherIPAperpetration
Tscore
Figure9. Mediation Accommodations by Reported IPA.
Mediators were more likely to provide accommodations in cases where they had
identified IPA present (28%; N=157) compared to cases identified as IPA absent (6%; N=22)
(χ
2
=75.050, df=1, p<.001). In addition, if one of the parents called the mediation program offices
and expressed concerns about coming to mediation or made requests to change the mediation
process (N=31), mediators were significantly more likely to provide those accommodations
(81%; N=25) (χ
2
=81.742, df=1, p<.001).
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 116 2007-WG-BX-0028
Goal 2: Assess whether mediation agreements, divorce decrees and/or parenting
plans include safety measures in cases involving self-reported IPA.
1. Determined the percentage of mediation agreements, divorce decrees/parenting plans that
include safety factors.
Within the initial sample of 965 couples, 117 couples (12%) did not continue with
mediation. Of the 117 cases not continuing, 68 cases (7%) were found inappropriate to continue
either due to IPA or for other reasons; 40 (4%) cases settled on their own during the time
mediation was occurring; 9 (1%) cases were closed by the mediation program during the time of
mediation due to couples either not appearing for initial mediation or failing to attend scheduled
mediation sessions; and, 18 (2%) of cases did not have a mediation outcome noted in the file. Of
the remaining 830 couples (86%), 466 (56%) reached a full agreement on all issues (legal and
physical custody; holiday and vacation schedules), 184 (22%) reached a partial agreement, and
180 (22%) did not reach an agreement.
The level of reported IPA influence whether an agreement was reached in mediation.
Regardless of whether the mother or father reported it, couples who failed to reach any
agreement had the highest overall IPA scores, followed by those who reached only partial
agreement; those who reached full agreement reporting the lowest IPA scores (Mothers
reporting: F=43.57 (2,766), p<.001; Fathers reporting: F=16.27 (2,751), p<.001) (see Figure 10).
However, post hoc comparisons indicated that there was no statistically significant difference in
the level of mother-perpetrated IPA between couples who reached full agreement and those who
only reached partial agreement.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 117 2007-WG-BX-0028
54.61
49.47
47.25
53.68
49.71
48.36
30.00
40.00
50.00
60.00
70.00
noagreementreached partialagreementreached fullagreementreached
LevelofReportedVictimization
FatherperpetrationmeanTscore
MotherperpetrationmeanTscore
Figure10. Mediation Agreement Type by T-score Mean of IPA Score.
Couples reporting the highest level of IPA (those at or above 2 standard deviations above
the mean), were substantially more likely to fail to reach an agreement, compared to those
reporting average or lower IPA scores, especially among those who reported father-perpetrated
IPA (see Table 11).
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 118 2007-WG-BX-0028
Table 11.
Comparisons of Mediation Agreement by IPA T-Score
Level of IPA No agreement
reached
Partial agreement
reached
Full agreement
reached
Father-
perpetrated IPA
Average or lower
(IPA <=50)
14%
(69)
23%
(115)
63%
(316)
High
(IPA>=70)
60%
(20)
27%
(9)
12%
(4)
Mother-
perpetrated IPA
Average or less
(IPA <=50)
15%
(76)
22%
(110)
62%
(303)
High
(IPA>=70)
34%
(12)
17%
(6)
49%
(17)
*Note: numbers in parentheses are numbers of participants.
Mediation agreements that included restrictions placed on contact between parents or
protections for the children were rare (Table 12).
Table 12.
Restrictions Placed on Parents or Parenting by Terms of Agreement
Type of Restriction Mentioned in Agreement Partial Full Total
No Restrictions 145 398 543
Keep parents apart (Orders of Protection and/or
Supervised Exchanges) 9 14 23
Keep parents apart and Father Supervised 1 1 2
Father Supervised Only 1 5 6
Mother Supervised Only 1 0 1
Other 1 1 2
Total 158 419 34
Note: Total N reduced due to missing data.
Of the 650 couples who reached any kind of agreement (partial or full) in mediation, 543
had information recorded in the case files on whether restrictions were in place. Of these, only 34
(6%) included some type of protection or restriction on contact. The cases that included
restrictions had more IPA as reported by both mothers and fathers (Mothers reporting: F=11.483
(1,552), p<.001; Fathers reporting: F=10.04 (1,552) p<.001). However, there were no
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 119 2007-WG-BX-0028
restrictions placed on the eight cases of high father-perpetrated IPA (2 standard deviations or
greater about the mean IPA level), and restrictions were placed on only one case involving the
highest levels of mother-perpetrated abuse (out of 19 cases identified in that range).
Two percent of the cases in the sample (N=22) had a legal finding of domestic violence
in the divorce file. Couples with a legal finding of domestic violence reported higher overall IPA
scores compared to those without (F=6.3 (1,882), p=.012, father-perpetrated IPA; F=6.81
(1,861), p=.009 mother perpetrated) (see Figure 11). However, among couples reporting the
highest levels of IPA (greater than 2 standard deviations above the mean), only three couples
(6%) reporting high father-perpetrated IPA received a legal finding of domestic violence, and
only three couples (8%) reporting high mother-perpetrated violence did. None of the seven
couples who reported high mutual IPA received a legal finding of domestic violence.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 120 2007-WG-BX-0028
30.00
40.00
50.00
60.00
70.00
decreefindingof
domesticviolence
nodecreefindingof
domesticviolence
FatherIPAperpetrationT
score
MotherIPAperpetration
Tscore
Figure11. Legal Finding of Domestic Violence and Reported IPA.
To calculate custody in divorce decrees we created a continuum of possible responses
from those favoring fathers to those favoring mothers (see Table 13).
Table 13.
Parenting Agreements: Physical and Legal Custody.
Physical and Legal Custody by Type Frequency
Father primary physical, sole legal 28
Father primary physical, joint legal 46
Joint physical and legal 180
Mother joint physical, sole legal 7
Mother primary physical and joint legal 401
Mother primary physical and sole legal 199
Total 861
*Note: Outcomes for the N=104 cases not included in these analyses include: dismissed, parents
reconciled, state has jurisdiction of children or other.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 121 2007-WG-BX-0028
Among those for whom custody was recorded in the divorce decree (N=861), mothers were
awarded primary custody (physical and/or legal) in 70% of the cases (N=607), and fathers were
awarded primary physical and/or legal custody in 9% (N=74); joint physical and legal custody
was awarded in 21% of the cases (N=97).
Divorce decree outcomes for full agreement cases in mediation were compared to cases
where there was no agreement reached in mediation. For the 466 cases where a full mediation
agreement was reached, mothers were awarded primary custody (physical and/or legal) in 69%
of the cases (N=290), whereas fathers were awarded primary physical and/or legal custody in 8%
(N=33); joint physical and legal custody was awarded in 23% of the cases (N=97).
Of the 180 of couples failing to reach any agreement in mediation, 158 had information
about the status of custody recorded about in the divorce case file. Mothers were awarded
primary custody (physical and/or legal) in 71% of the cases (N=112), and fathers were awarded
primary physical and/or legal custody in 11% (N=17); joint physical and legal custody was
awarded in 18% of the cases (N=29).
For all cases combined, the level of IPA significantly differed across the custody
arrangement groups. For both reports of father and mother perpetrated IPA, the lowest level of
reported IPA were among those awarded joint custody. Considering custody arrangements
ordered in the decree of just those couples classified by the latent class analysis as Mutually
Violent Control and for which decrees were entered (N=26); mother was awarded primary
physical and/or legal custody in 22 cases (85%); joint custody was awarded in three cases (11%);
and, father was awarded primary physical and/or legal custody in one case (4%).
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 122 2007-WG-BX-0028
Figure12. Custody Awards in Divorce Decree by Reported IPA.
47.30
50.44
50.95
48.25
49.63
53.38
30.00
40.00
50.00
60.00
70.00
JointCustod
y
MotherPrimary
Custody(physical
and/orlegal)
FatherPrimary
Custody(physical
and/orlegal)
FatherIPAperpetrationT
score
MotherIPAperpetrationT
score
Decree restrictions included restrictions on seeing the child; on drop off and pick up
locations; on the amount of and structure of parenting time; and, on supervision of exchanges or
supervision of parenting time. Restrictions were rarely implemented in decrees. Overall, 89%
(853) of cases had no restrictions, 11% (91) had one, and only 17 (2%) cases had more than one
restriction in place. Frequencies of each type of restriction found in the decree are shown in
Table 14.
Table 14.
Frequencies of Decree Restrictions by Type
Yes (n) No (n)
Restrictions on who sees child 0.4% (4) 99.6%(958)
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 123 2007-WG-BX-0028
Restrictions on drop off and pick up location 4% (38) 96% (924)
Restriction on time of parenting time 4% (37) 96% (925)
Supervised exchanges 2% (18) 98% (944)
Supervised parenting time 4% (38) 96% (924)
Cases with restrictions in place tended to have higher average levels of reported father-
perpetrated IPA (54.97) compared to those without restrictions (49.38) (F=27.8(1, 882), p<.001),
as well as higher average levels of reported mother-perpetrated IPA (53.93 vs. 49.50, F=17.319
(1, 861), p<.001). In addition, a greater percentage of cases at the highest levels of IPA (2
standard deviations or higher) had restrictions, as compared to those at the lower levels. Ten
percent at the lower levels of father-perpetrated IPA had restrictions in place, compared with
29% at the highest level (
χ
2
= 15.275, df=1, p<.001). Eleven percent of cases at the lower levels of
mother-perpetrated IPA had restrictions, compared with 30% at the highest level (χ
2
=12.97, df=1,
p<.001).
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 124 2007-WG-BX-0028
Goal 3: Assess the frequency with which mediation agreements/divorce decrees are re-
litigated over time.
1. Assessed specific types of hearings in superior court post-divorce and the orders emanating
from the hearings.
Hearings
As noted above, case variables for the Superior Court database collected early in the
study (N=271) were not usable in these analyses. For the cases wherein we collected usable data
on post-decree events in the case (N=694), the majority of cases (62%; N=432) had at least one
hearing after the divorce was finalized. The range in the number of hearings per case varied from
0-31 hearings (mean=2.08; SD=3.455). Thirty-eight percent of the cases (N=262) cases had no
post-decree hearings; 52% (N=358) had between 1–5 hearings; 8% (N=53) had between 6-10
hearings; 3% (N=21) had 11 or more hearings.
There was a subset of cases requesting a disproportionate number of hearings. Thirty one
cases (4.5%) accounted for 444 (31%) of the total number of hearings (N=1441).
Few cases were denied hearings when requested. Only 32 (5%) of the cases were denied
post-decree hearings, with a range of 0-3 hearings denied (mean = .07). Twenty-one cases (3%;
N=694) had one request for a hearing denied; 8 cases (1%; N=694) had two requests for hearings
denied and 3 (0.4%; N=694) cases had three requests for hearings denied.
It was not always clear who was bringing the action; therefore, Table 15 includes a
column for Hearings wherein the initiator of the action was the Court or it was Unknown who
was making the request. Over one-third of the hearings were in regard to child support, either
modifying or enforcing it (see Table 15). Of the total number of hearings for modifying child
support, the majority (55%) were requested by fathers followed by mothers (35%). For the
remaining 11% the requestor of the action was unknown. Of the total number of hearings for
child support enforcement, approximately half was requested by the Court or an Unknown
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 125 2007-WG-BX-0028
source (49%) and a little less than half by mothers (46%). A small percentage was requested by
fathers (5%).
Approximately one quarter of hearings was for modifying child custody or parenting
time. Of these, fathers (46%) requested slightly more than mothers (42%) with the remaining
requested by the Court or an Unknown source (11%). Six percent of the hearings were in regard
to IPA, child abuse, or harassment. Mothers were overwhelmingly more likely to request these
hearings (70% were requested by mothers). The remainder was nearly evenly split between
fathers (16%) and the Court/Unknown source (14%). There were also 230 post-decree hearings
(16%) regarding other issues, such as property (refinancing or sale of house; family photos),
retirement accounts or taxes, attorney withdrawals, disagreements about medical
decisions/payments and other miscellaneous requests. A total of 278 review hearings were also
held post-decree, with a range of 0-18 (mean = .40) (See Table 14). Of the 278 review hearings,
81 were counted in the specific categories outlined in the table.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 126 2007-WG-BX-0028
Table 15.
Number of post-decree hearings, orders and stipulations.
Type Hearings
Hearings
Mother
Requested
Hearings
Father
Requested
Hearings
Court or
Unknown
Requested
Orders Stips
Modifying Child Support 376 129 205 42 429 54
Enforcement Child Support 158 72 8 78
Included
Above
Modification Spousal
Maintenance.
78 16 50 8 59 8
Modification Child Custody 82 32 43 1 77 38
Modification Parenting Time 247 95 126 26 268 55
IPA, Child Abuse, Harassment 81 57 13 11 67
Other Issues 230 230 158 54
Review Hearings Not In Above 189 189
Totals 1441 401 445 585 1058 209
Orders
The majority of cases (59%; N=694) had at least one order issued in their case after the
divorce was finalized. The number of orders ranged from 0-19 (mean=1.52; SD=2.134). Forty-
one percent (N=282) of the cases had no post-decree orders; 53% (N=369) had between 1 and 5;
6% (N=38) had between 6-10; and 1% (N=5) had 11 or more post-decree order. In total there
were 1058 orders issued in N=412 cases.
Unlike the case of post-decree hearings, there was not a distinct group of families who
received a disproportionate number of orders. The highest-volume of post-decree hearings was
for 8 cases (1%; N=694) who had over 10 post decree orders accounting for 6.5% (N=96) of the
total number of hearings (N=1058).
There were 429 post-divorce orders modifying child support (one modification N=210;
two modifications N=64; three modifications N=18; four modifications N=8; five modifications
N=1). There were 354 post-divorce orders/notices regarding withholding income for child
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 127 2007-WG-BX-0028
support (one modification N=22; two modifications N=22; three modifications N=21; four
modifications N=12; five modifications N=25; six modifications N=7; seven or eight
modifications N=4; nine modifications N=1; twelve modifications N=2; seventeen modifications
N=1).There were 268 post-divorce orders modifying parenting time (one modification N=113;
two modifications N=26; three modifications N=14; four modifications N=5; five modifications
N=4; six through eight modifications N=1). There were 77 post-divorce orders modifying child
custody (one modification N=56; two modifications N=9; three modifications N=1). There were
59 post-divorce orders modifying spousal maintenance (one modification N=31; two
modifications N=8; three modifications N=4). There were 67 post-divorce orders regarding IPA,
Child Abuse or Harassment (one order N=41; two orders N=9; three orders and five orders
respectively N=1). There were 158 post-divorce orders for other miscellaneous reasons (one
order N=65; two orders N=19; three orders N=11; four orders N=1; six orders respectively N=3)
(see Table 15).
There was an additional category of post-decree interactions with the court that were not
included in the post-divorce court orders but remain important in terms of burden on the courts.
To reduce conflict between parents concerning the timing or the amount paid for child support
each month, courts often garnish the wages of the parent responsible for paying child support. In
this way the payee does not need to write checks and send them. Payment is handled through the
court. There are a variety of reasons why the courts would issue an order or a notice to withhold
income for child support. For example, if the amount of child support changes or a parents
change jobs, an order or notice would be issued. There were data available 446 cases (64%;
N=694). For those cases, there were 354 orders and/or notices were issued regarding withholding
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 128 2007-WG-BX-0028
income for child support issued in 101 cases (82 cases had between 1-5 orders/notices; 16 cases
had between 6-10 orders/notices; and 3 cases had 11 or more notices/orders).
Stipulations
Stipulations are an agreement between the parties to partially or fully settle issues in the
case before the court. They are not, however, a separate legal action. Some courts hold special
sessions for filing stipulations some do not. In this study, there were 209 (30%; N=694) cases
with stipulated agreements on various post-decree issues. The number of stipulations per case
ranged from 0-7 (mean=0.30; SD=0.784). Fifty-nine percent (N=567) of the cases had no post-
decree stipulations; 11% (N=79) had between 1–5 stipulations; 8% (N=53) had between 6-10
stipulations; 3% (N=21) had 11 or more stipulations. There were a total of 54 stipulations to
changes in child support, 8 stipulations to changes in spousal maintenance, 38 stipulations to
changes in child custody and 55 stipulations to changes in parenting time. There were also 54
stipulations regarding other issues. These issues included stipulations regarding use of the
Judicial Supervision Program (for supervised parenting time), referring a child to counseling, a
child’s school (location, category, and tuition), retirement benefits, and payment of taxes (see
Table 15).
IPA and Hearings and Orders
Although both level of father-perpetrated IPA (r=.129, p=.001) and level of mother-
perpetrated IPA (r=.101, p=.01) predicted the number of hearings held post-decree, it was a
weak relationship. Among the different types of orders, only mother-perpetrated IPA weakly
predicted the number of parenting time orders (r=.09, p=.03). The average number of hearings
was also higher for those cases that had been identified as having IPA by the mediator than for
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 129 2007-WG-BX-0028
those not so identified (see Table 16). In addition, mediator-identified IPA cases had higher
mean number of parenting time orders, and numbers of orders regarding IPA or harassment.
Table 16.
Mean Differences Among Post-Decree Hearings and Orders by Mediator-Identified IPA
N M SD Minimum Maximum
Number of hearings
F=5.07 (1, 692)
p=.025
No IPA 274 1.71 2.975 0 20
IPA 420 2.31 3.719 0 31
Number of parenting
time orders
F=13.07 (1, 692)
p<.001
No IPA 274 .23 .545 0 3
IPA 420 .49 1.067 0 8
Number of orders regarding IPA
or Harassment
F=8.44 (1,692)
p=.004
No IPA 274 .04 .222 0 2
IPA 420 .13 .463 0 5
Post-Decree Hearings and Orders by Mediation Outcome
We also investigated whether the couples returned to court to revise the agreements
negotiated in mediation. Specifically, we investigated whether there was a significant difference
in the number of post-decree hearings and post-decree orders based on the outcome of mediation
(no agreement, partial or full agreement). Results indicated that there was a significant difference
between these three groups (F=13.979, df=2, p<.001). To identify which groups were
significantly different, the post-hoc test (Student-Newman-Keuls) found a significant difference
between the full agreement group and the other two groups (partial and no agreement) but there
was no significant difference between the latter two groups. Because the distribution was highly
skewed, the analysis was rerun using a nonparametric test (Kruskal-Wallis), which confirmed the
results of the original significance test (χ
2
=32.318, df=2, p<.001). Figure 13 details the
percentages of couples in the different groups that had post-decree hearings. The percentages of
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 130 2007-WG-BX-0028
each mediation outcome (no agreement, partial agreement and full agreement) individually sum
to 100%. The total number of cases able to be classified in this analysis was 590 (N=694). The
number of hearings in each mediation outcome category are: no agreement N=131, partial
agreement N=139; and, full agreement N=320.
Figure 13:
Percentages of Post-Decree Hearings by Outcome of Mediation.
5
10
15
20
25
30
35
40
45
50
0123456+
PercentofCases
NumberofPostDecreeHearings
PostDecreeHearings
No
Partial
Full
To explore whether there was a significant difference in the number of post-decree
orders, a similar set of analyses was run. Results indicated that there was a significant difference
between the three groups (no agreement, partial or full agreement) (F=6.860, df=2, p=.00). The
post-hoc test (Student-Newman-Keuls) found a significant difference between the full agreement
group and the other two groups (partial and no agreement) but there was no significant difference
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 131 2007-WG-BX-0028
between the latter two groups. The nonparametric test (Kruskal-Wallis) again confirmed the
results of the ANOVA (χ
2
=13.382, df=2, p=.001). Figure 14 details the percentages of couples in
the different groups that had post-decree orders. Again, the percentages of each mediation
outcome (no agreement, partial agreement and full agreement) sum to 100%. The percentages of
each mediation outcome (no agreement, partial agreement and full agreement) individually sum
to 100%. The total number of cases able to be classified in this analysis was 590 (N=694).
Figure 14:
Percentages of Post-Decree Orders by Outcome of Mediation.
5
10
15
20
25
30
35
40
45
50
0123456+
PercentofCases
NumberofPostDecreeOrders
PostDecreeOrders
No
Partial
Full
We also investigated whether there was a significant difference in the number of post-
decree stipulations between the three groups of mediation outcomes (no agreement, partial and
full agreement). The results indicated that there was also a significant difference between the
three groups (no agreement, partial or full agreement) (F=7.543, df=2, p=.001) post-hoc test
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 132 2007-WG-BX-0028
(Student-Newman-Keuls) found a significant difference between the partial agreement group and
the other two groups (full agreement and no agreement). There was no significant difference
between the latter two groups. The nonparametric test (Kruskal-Wallis) again confirmed the
results of the original significance test (χ
2
=9.177, df=2, p=.01). Figure 15 details the percentages
of couples in the different groups that had post-decree orders. Again, the percentages of each
mediation outcome (no agreement, partial agreement and full agreement) sum to 100%.
Figure 15
Percentages of Post-Decree Stipulations by Outcome of Mediation.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 133 2007-WG-BX-0028
Goal 4: Test a multivariate conceptual model using variables that are hypothesized to
affect mediation, divorce case and post-divorce outcomes (see Figure 16).
The Measurement Models
Multivariate statistical analyses can be roughly grouped according to two distinct
functions: (1) Measurement Models, which are used to aggregate multiple convergent indicators
to indirectly assess latent theoretical constructs; and (2) Structural Models, which are used to test
hypotheses regarding the possible causal relations that may exist among these latent constructs.
Some multivariate methods address one or the other of these goals exclusively, whereas others
(such as Factor Analytic Structural Equation Model) are designed to address both at once. Either
way, it is widely recommended that the measurement model be fully developed before a
structural model is attempted. This is because it is not a logically or statistically valid procedure
to test the hypothesized causal relations among latent constructs that are not yet adequately
measured.
Our first step was therefore to develop an adequate multivariate measurement model for
our major latent construct of interest: Intimate Partner Abuse (IPA) (see Figure 16). We carefully
considered, compared, and contrasted various different approaches to the measurement of IPA
before proceeding to use this construct to predict the various outcomes that IPA was theoretically
expected to influence:
(1) Mediation Outcomes:
(Identification of IPA by the mediator (MEDIPA); special procedures provided to the
couple in mediation (MEDPROC); mediation agreements negotiated by the couple
(MEDAGREE));
(2) Child maltreatment outcomes
(Child Protective Service Involvement with the family (RCM));
(3) Divorce Outcomes
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 134 2007-WG-BX-0028
(Legal findings of “domestic violence” in a case (COURTIPA); restrictions placed on the
offender parent in relation to contact with the abuser and/or children (COURTRES);
custody and parenting time arrangements ordered in the decree which favor mothers
(COURTCUS)); and
(4) Law-Enforcement Outcomes (calls to the law enforcement post-divorce (POSTCPW));
(5) Post-Divorce Outcomes.
(hearings post-divorce (HEARINGS); court orders post-divorce (COURTORD).
Two different measurement models were developed for modeling IPA using the RBRS
data that rely on a psychometric, dimensional approach. The RBRS has five validated subscale
dimensions: (1) Coercive Controlling Behaviors; (2) Psychological Abuse, (3) Physical Abuse,
(4) Threatened and Escalated Violence, and (5) Sexual Intimidation/Coercion/Assault (Beck et
al., 2009). Recently, Tanha et al. (2010) used Confirmatory Factor Analysis (CFA) to model the
last four of these subscales as convergent indicators of a single latent common IPA factor;
Structural Equations Modeling (SEM) was then used to estimate the role of Coercive Controlling
Behaviors as a common causal influence on all of them, rather than as a direct indicator of IPA.
Another approach is to use Coercive Controlling Behaviors as an additional indicator variable by
virtue of its high correlations to all of the others, and to the latent common factor itself (see
Tanha et al, 2010). Because the addition of Coercive Controlling Behaviors subscale did not
appreciably affect the measurement model for IPA, the former approach was used in the final
structural model in order to test its possible causal role in the etiology of IPA, as predicted by
theory.
The Structural Model
The structural model tested, relating the IPA constructs to the various outcome variables,
was structured as a pattern of regressions referred to as a cascade model in cognitive psychology
(Mouyi, 2006; Demetriou, Christou, Spanoudis, & Platsidou, 2002). This procedure is
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 135 2007-WG-BX-0028
conceptually equivalent to a Sequential Canonical Analysis (Figueredo & Gorsuch, 2007;
Gorsuch & Figueredo, 1991), which controls statistically for any indirect effects of the predictors
through the causally prior criterion variables.
In a cascade model, a series of multiple regressions is performed in which the multiple
criterion variables are analyzed sequentially according to a hypothesized causal order. Because
these criterion variables are expected to causally influence each other, they are entered
sequentially into a system of multiple regression equations with each hierarchically prior
criterion variable entered as the first predictor for the next. Thus, each successive criterion
variable is predicted from an initial predictor variable, each time entering the immediately
preceding criterion variable as the first predictor, then entering all the ordered predictors from
the previous regression equation. Thus, each successive regression enters all of the preceding
criterion variables in reverse causal order, to statistically control for any indirect effects that
might be transmitted through them. Within this analytical scheme, the estimated effect of each
predictor is limited to its direct effect on each of the successive criterion variables. The general
format for this system of multiple regressions is therefore as follows:
Y
4
= β
1
X
1
+ β
2
X
2
+ β
3
X
3
Y
5
= β
4
Y
4
+ β
1
X
1
+ β
2
X
2
+ β
3
X
3
Y
6
= β
5
Y
5
+ β
4
Y
4
+ β
1
X
1
+ β
2
X
2
+ β
3
X
3
Where X
1
, X
2
and X
3
are the ordered predictor variables and Y
4
, Y
5
, and Y
6
are the
ordered criterion variables, numbered consecutively after the predictors to avoid confusion
among the subscripts.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 136 2007-WG-BX-0028
Analogous to a Sequential Canonical Analysis (SEQCA), this kind of cascade model has
been proposed to serve as an exploratory form of path analysis (Figueredo & Gorsuch, 2007)
where the exact model specification cannot be completely predicted by existing theory, making
formal SEM unsuitable for the task because SEM requires a complete model specification based
on strong a priori theory. Instead, this kind of exploratory model merely requires the tentative
specification of an expected causal order among the variables, with all of the exact causal
linkages not completely pre-specified. A cascade model therefore provides a theoretically-
guided exploration rather than a formal and confirmatory test of a priori theory. A cascade
model avoids compromising the purely confirmatory nature of the SEM tests of whole-model
goodness-of-fit.
The Cascade Model: Model Specification
The Cascade Model for the current data were entered in the following hypothesized
causal order:
1. Marital Stressors
2. Father Perpetration of Coercive Controlling Behaviors (as reported by mother)
3. Mother Perpetration of Coercive Controlling Behaviors (as reported by father)
4. Father Perpetration of IPA
5. Mother Perpetration of IPA
6. Reported Child Maltreatment (reports by either parent that CPS involved)
7. Mediator’s Determination of IPA present/absent
8. Mediation Safety Accommodations
9. Mediation Agreements
10. Orders of Protection
11. Court Legal Finding of IPA
12. Court-Decreed Restrictions
13. Court Custody Decrees Favoring Mothers
14. Post-Divorce Calls to Law Enforcement
15. Post-Divorce Number of Hearings
16. Post-Divorce Number of Decree Orders
Missing Data
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 137 2007-WG-BX-0028
To estimate the magnitudes of indirect effects within the system of regression equations
by the multiplication of sequential path coefficients, it was necessary to base all of these
estimates on the same sample of couples. Thus, we began with the N=694 cases wherein we had
mediation and Superior Court post-decree data. Because each multiple regression deleted cases
for which data were missing on any of the variables included in the equation, we first created a
working subset of the data for which all such missing data were eliminated. The reduced sample
size was N=458 usable cases. Although this is normally considered a more than adequate sample
size, as even large sample statistics require 300 – 500 cases, we were concerned about the
potential bias involved in eliminating such a large number of cases due to missing data.
We therefore created a dichotomous variable, called MISSING, to distinguish between
the “listwise deleted” and “listwise retained” subsamples, as described in Figueredo, McKnight,
McKnight, and Sidani (2007). Each of the variables used in the Cascade Model was then tested
for significant differences between these two subsamples to determine if there were any
systematic biases in the data that were missing. Only one such variable was found to have a
significant difference (p<.05) between the “listwise deleted” and “listwise retained” subsamples,
and that variable was Marital Stressors. This effect indicated that the level of Marital Stressors
was higher for couples for which data were missing on one or more of the Cascade Model
variables: F(1,963)= 12.92 (p=0.0003). Although statistically significant due to the large sample
size and consequent power of the test, the squared multiple correlation of that effect only
accounted for 1.32% of the variance. We therefore felt quite comfortable about proceeding with
the structural analyses without the application of any missing data imputation method, most of
which do not work very well with more than 20% of the original sample missing and especially
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 138 2007-WG-BX-0028
when the missing data are biased (Figueredo et al., 2000; McKnight, McKnight, Sidani, &
Figueredo, 2007).
Results of Multiple Regression Models.
The following are the results of the system of multiple regression models run within the
context of a Cascade Model, listed by criterion variable, in the theoretically-specified causal
order, and reporting standardized partial regression weights, tests of statistical significance, and
probabilities under the null hypothesis. As per APA Style, only the statistical results for the
significant predictors are reported (see Figure 16). Included within these results are the specific
hypotheses developed in construction of the model.
It is extremely important to note that we had to choose specific language to describe the
series of regressions (victim or perpetrator). To better understand who is performing the
behaviors in these regressions we identify the perpetrator of the IPA behaviors.
To simplify discussion of the results in the conclusion section of the paper for the Goal 4
hypotheses, we indicated which of the following equations addressed the Goal 4 Hypotheses in
the original grant proposal. We then refer to these equations in the conclusion section of the
paper.
Intimate Partner Violence in Mandatory Divorce Mediation
Figure 16. The Cascade Model.
Beck, Walsh, Mechanic, Figueredo & Chen 139 2007-WG-BX-0028
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 140 2007-WG-BX-0028
Equation 1:
As expected, Father Perpetration of Coercive Controlling Behaviors (as reported by mother)
(ZFCC) was significantly and positively predicted by Marital Stressors (β=0.23, t
456
=4.96,
p<.0001).
Equation 2:
Mother Perpetration of Coercive Controlling Behaviors (as reported by father)
(ZMCC) was significantly and positively predicted both by Father’s Coercive controlling
behaviors (β=0.16, t
455
=3.36, p<.0008) and by Marital Stressors (β=0.13, t
455
=2.80, p<.0053).
Equation 3:
Father Perpetration of IPA (as reported by the mother) (ZFIPA) was significantly and
positively predicted both by Father’s Coercive controlling behaviors (β=0.70, t
454
=21.06,
p<.0001) and by Marital Stressors (β=0.07, t
454
=1.98, p<.0480), but not by the Mother’s
Coercive controlling behaviors.
Equation 4:
Mother Perpetration of IPA (as reported by the father) (ZMIPA) was significantly and
positively predicted both by Mother’s Coercive controlling behaviors (β=0.61, t
453
=17.40,
p<.0001) and by Father’s Coercive controlling behaviors (β=0.11, t
453
=2.20, p<.0284), but not by
either the Father’s IPA or Marital Stressors.
Equation 5:
Reported Child Maltreatment (CPS involvement as reported by either mother or father)
(RCM) was significantly and positively predicted by Mother’s IPA (β=0.22, t
452
=3.59, p<.0004)
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 141 2007-WG-BX-0028
but was significantly and negatively predicted by Mother’s Coercive controlling behaviors (β=-
0.16, t
452
=-2.64, p<.0086), but not by the Father’s IPA, Father’s Coercive controlling behaviors,
or Marital Stressors.
Equation 6:
The Mediator’s Report of IPA (MEDIPA) was significantly and positively predicted by
Mother’s IPA (β=0.19, t
451
=3.23, p==.0013) and by the Father’s IPA (β=0.29, t
451
=4.62,
p=.0001), as reported by their partners, but not by Reported Child Maltreatment, Mother’s
Coercive controlling behaviors, Father’s Coercive controlling behaviors, or Marital Stressors.
Equation 7:
Mediation Special Procedures/Safety Accommodations (MEDPROC) were significantly
and positively predicted by the Mediator’s identification of a case as having IPA (β=0.20,
t
450
=4.14, p<.0001) and by the Mother’s IPA (β=0.14, t
450
=2.26, p=0242), as reported by the
Father, but not by Reported Child Maltreatment, Father’s IPA, Mother’s Coercive controlling
behaviors, Father’s Coercive controlling behaviors, or Marital Stressors.
Equation 8:
Mediation Agreements (MEDAGREE) were significantly and negatively predicted by
Mediation Special Procedures/Safety Accommodations (β=-0.24, t
449
=-5.19, p=.0001) and by the
Father’s IPA (β=-0.16, t
449
=-2.40, p=.0167), Mother’s Coercive controlling behaviors (β=-0.17,
t
449
=-3.00, p=.0028), but not by the Mediator’s Report of IPA, Mother’s IPA, Reported Child
Maltreatment, Father’s Coercive controlling behaviors or Marital Stressors.
Equation 9:
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 142 2007-WG-BX-0028
Orders of Protection(ORDPROT) were significantly and positively predicted by
Mediation Safety Accommodations (β=0.26, t
448
=5.19, p=.0001), ), by Mediator’s Report of IPA
(β=0.10, t
448
=2.03, p=.0427), by the Mother’s IPA (β=0.15, t
448
=2.54, p=.0114), by the Father’s
IPA (β=0.14, t
448
=2.15, p=.0323), and were significantly and negatively predicted by the
Mother’s Coercive controlling behaviors (β=-0.15, t
448
=-2.57, p=.0105), but not by Mediation
Agreements, Reported Child Maltreatment, Father’s Coercive controlling behaviors, or Marital
Stressors.
Equation 10:
Court Legal Findings of IPA (COURTIPA) were significantly and positively predicted
by Orders of Protection (β=0.23, t
447
=4.61, p=.0001), Reported Child Maltreatment (β=0.10,
t
447
=2.17, p=.0302), and by the Mother’s IPA (β=0.19, t
447
=2.91, p=.0038), but not by Mediation
Agreements, Mediation Safety Accommodations, Mediator’s Report of IPA, Father’s IPA,
Mother’s Coercive controlling behaviors, Father’s Coercive controlling behaviors, or Marital
Stressors.
Equation 11:
Court-Decreed Restrictions (COURTRES) were significantly and positively predicted by
Court Findings of IPA (β=0.10, t
446
=2.23, p=.0259), by Orders of Protection (β=0.11, t
446
=2.26,
p=.0242), and by Marital Stressors (β=0.14, t
446
=3.01, p=.0028), but not by the Mediation
Agreements, Mediation Safety Accommodations, Mediator’s Report of IPA, Reported Child
Maltreatment, Mother’s IPA, Father’s IPA, Mother’s Coercive controlling behaviors, or Father’s
Coercive controlling behaviors.
Equation 12:
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 143 2007-WG-BX-0028
Court Custody Decrees Favoring Mothers (COURTCUS) were significantly and
negatively predicted by Reported Child Maltreatment (β=-0.17, t
445
=-3.52, p=.0005), were
significantly and positively predicted by Father’s IPA (β=0.15, t
445
=2.15, p=.0322) as reported
by the mother, and were significantly and negatively predicted by Marital Stressors (β=-0.11,
t
445
=-2.23, p=.0262), but not by the Court-Decreed Restrictions, Court Findings of IPA,
Substantiated Orders of Protection, Mediation Agreements, Mediation Safety Accommodations,
Mediator’s Report of IPA, Mother’s IPA, Mother’s Coercive controlling behaviors, or Father’s
Coercive controlling behaviors.
Equation 13:
Post-Divorce Calls to Law Enforcement (POSTCPW) were significantly and positively
predicted by Mediator’s Report of IPA (β=0.14, t
444
=2.61, p=.0095), and by Reported Child
Maltreatment (β=0.12, t
444
=2.45, p=.0146), but not by the Court Custody Decrees Favoring
Mothers, Court-Decreed Restrictions, Court Findings of IPA, Orders of Protection, Mediation
Agreements, Mediation Safety Accommodations, Mediator’s Report of IPA, Mother’s IPA,
Father’s IPA, Mother’s Coercive controlling behaviors, Father’s Coercive controlling behaviors,
or Marital Stressors.
Equation 14:
Post-Divorce Number of Hearings (HEARINGS) were significantly and negatively
predicted by Court Findings of IPA (β=-0.13, t
443
=-2.66, p=.0082) and significantly and
negatively predicted by Mediation Agreements (β=-0.22, t
443
=-4.47, p=.0001), and were
significantly and positively predicted by Reported Child Maltreatment (β=0.10, t
443
=2.15,
p=.0324), but not by the Post-Divorce Calls to Law Enforcement, Court Custody Decrees
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 144 2007-WG-BX-0028
Favoring Mothers, Court-Decreed Restrictions, Substantiated Orders of Protection, Mediation
Safety Accommodations, Mediator’s Report of IPA, Mother’s IPA, Father’s IPA, Mother’s
Coercive controlling behaviors, Father’s Coercive controlling behaviors, or Marital Stressors
Equation 15:
Post-Divorce Number of Decree Orders (ORDERS) were significantly and positively
predicted by Post-Divorce Number of Hearings (β=0.69, t
442
=19.22, p=.0001) and were
significantly and negatively predicted by Court Custody Decrees Favoring Mothers (β=-0.07,
t
442
=-2.05, p=.0412), but not by the Post-Divorce Calls to Law Enforcement, Court-Decreed
Restrictions, Court Findings of IPA, Substantiated Orders of Protection, Mediation Agreements,
Mediation Safety Accommodations, Mediator’s Report of IPA, Reported Child Maltreatment,
Mother’s IPA, Father’s IPA, Mother’s Coercive controlling behaviors, Father’s Coercive
controlling behaviors, or Marital Stressors.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 145 2007-WG-BX-0028
Conclusions and Implications
We realize that we cannot infer causality from correlational data or really any data;
however, to be socially useful we will discuss conclusions, implications and policy
recommendations as if these data represent actual influences. Only in this way can we then
discuss possible adjustments to the mediation enterprise that could produce better outcomes for
both the couples involved in the system and the court system processing the cases.
Discussion of the Findings
There has been three decades of scholarly research on numerous aspects of divorce
mediation; however no comprehensive understanding of the short- and long-term outcomes for
couples legally ordered to mediation to resolve custody and parenting time. In the case couples
alleging intimate partner abuse (IPA), even less is known (Beck & Sales, 2001; Salem, 2009).
This study was funded by NIJ to systematically document actual percentages of IPA in
mediation; to systematically analyze mediator practices addressing those IPA cases; and to assess
mediation outcomes, divorce outcomes and post-decree outcomes for IPA cases. To do this we
linked naturalistic clinical interview data used to screen for marital stressors and IPA and
behaviorally-specific IPA questionnaire data from a large couple-matched sample to mediator’s
decisions concerning whether to proceed with mediation as usual, to screen out IPA-identified
cases; or, to provide special procedural accommodations for IPA-identified cases. We then
linked all these IPA data to mediation outcomes from in mediation case files and to decisions
concerning the final parenting plans found in Superior Court divorce files. We then linked these
data to longitudinal data concerning re-litigation of divorce-related issues in superior court and
longitudinal data concerning contacts with area law enforcement.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 146 2007-WG-BX-0028
Based on previous work concerning IPA screening in mediation (Administrative Office
of the Courts [California], 2010, Ellis, 2006; Ellis & Stuckless, 2008; Newmark et al., 1995;
Pearson, 1997), we believe that mediation screening for IPA at the study site far exceeds the
standard for mediator screening practices employed in most jurisdictions. Therefore, this study
represents a “best case scenario” for screening practices and responding to IPA in the mediation
context at the time the study was conducted, and likely currently as well.
The results of this study provide strong empirical support for previous estimates that most
couples attending divorce mediation report some level of IPA. Given that couples mandated to
mediation are there precisely because they cannot agree on significant aspects of the divorce, this
result was expected. Few mothers and fathers reported no IPA of any kind; only four couples
reported that there were no incidents of any sort of IPA in the preceding 12 months prior to
mediation. The parents reported at least some form of IPA occurred in over 90% of the cases and
two thirds of the couples reported that either or both partners utilized outside agency
involvement from law enforcement, shelters, courts, or hospitals to handle the IPA (Beck et al.,
2010). These figures represent a tremendous amount of IPA in couples mandated to attend
mediation and that the level was likely more serious that “common couple violence” identified in
previous national random sample surveys (Kelly & Johnson, 2008).
Important disparities were also found in incomes between divorcing parents. Upon
entering mediation, the mothers’ incomes in this study were approximately half of the fathers’
incomes. One potential interpretation of this finding is that that mediation about custody and
child support is not an “even playing field” for both parties even before the issue of abuse is
raised. This marked sex difference in income graphically illustrates a highly gendered pattern in
child care and the patterns reflected in custody arrangements following separation. These
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 147 2007-WG-BX-0028
patterns are often inaccurately attributed to “gender bias” against men as opposed to reflecting
the patterns established in the family prior to divorce.
Goal 1: Determine whether the mediation program accurately identifies couples with
self-identified IPA and assess whether these cases are treated differently.
Two methods were used to address this question. The first was a written questionnaire
with behaviorally-specific questions and the second was a semi-structured clinical interview
performed by the mediator. The latter provided the basis for the mediators’ decisions to
determine a case as having IPA. The questionnaire data provided a clearer understanding of the
specific behavioral dimensions of reported IPA behaviors for research purposes, while the semi-
structured clinical interview provided a more detailed understanding of the broad context of the
relationship including substance use and abuse, medical or mental health issues, and reported
child maltreatment.
Mothers reported significantly higher frequencies of nearly all dimensions of IPA than
did fathers on the questionnaires. In terms of severe and potentially life-threatening violence,
well over half of the mothers reported at least one incident of threatened and escalated physical
violence and sexual intimidation/coercion/assault in the previous year. Interestingly, half of all
fathers also reported at least one incident of threatened and escalated physical violence from
their partners but significantly less frequently reported incidents of sexual intimidation/coercion/
assault than did mothers. The average frequency of mothers reporting father-perpetrated sexual
intimidation/coercion/assault and threatened and escalated physical violence was “very rarely;”
however, any reports of physically forced sex, broken bones, or use of weapons in the previous
12 months is alarming, particularly given these mothers and fathers are negotiating long-term,
difficult-to-modify legal agreements with their abusers and were likely no longer living together.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 148 2007-WG-BX-0028
The only dimension of IPA that was not statistically significantly different for fathers and
mothers and that was the milder forms of physical acts (physical abuse). This finding appears to
initially support previous research suggesting fathers and mothers both engage in milder forms of
physical aggression, such as pushing and shoving (and other less severe violence, such as hitting
or biting) with equal frequency. Critically important is the finding that other more life-
threatening dimensions of IPA (threatened and escalated physical violence; sexual
intimidation/coercion/assault) are perpetrated much more frequently by fathers against mothers
(Kelly & Johnson, 2008). This vital distinction between lower level acts of physical abuse and
life-threatening, injurious physical acts of violence has not often been made in previous literature
on violence (Archer, 2000a; Dutton & Nicholls, 2005; Felson & Outlaw, 2007; Straus, 2008).
Analyzing the data taking into consideration both parents in the relationships and all
dimensions of IPA presents a different picture. The latent classes produced from this data did not
produce a class representing situational couple violence (Kelly & Johnson, 2008). Instead, of the
five latent classes produced from this data, four classes clearly indicated a perpetrator and a
victim in the relationship. One class represented couples with mutually low (or no) abuse or
violence. In other words, a class with only elevations on physical abuse or in combination with
psychological abuse for both members of the couple did not emerge from this data. It may be that
because the participants in this study were all divorcing and were in conflict over custody and/or
parenting time, and were reporting high levels of IPA, thus, there was a restricted range of IPA
behaviors in these couple relationships compared to that generally reported among representative
community samples. Whereas situational couple violence is common in nationally representative
samples (Johnson, 2006), it does not appear to be common in this divorcing sample of couples,
all of whom, by definition, are in conflict over parenting issues and in divorce mediation. These
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 149 2007-WG-BX-0028
findings support the growing argument about the gendered nature of IPA perpetration as largely
a function of the nature of the type of sample from which participants are recruited, with
common couple violence much more likely to be found among normative, compared to ‘clinical’
samples of any sort.
The four IPA classes that emerged from data in this study included Coercive Controlling
Violence—father perpetrator, Coercive Controlling Violence—mother perpetrator, Mutual
Violent Control and Lower Level Coercive Controlling Violence—father perpetrator. The two
latter latent class types are very interesting. Although Kelly and Johnson (2008) described the
Mutual Violent Control type in their work, this is the first study that actually empirically
validated that this class exists, albeit a small one. The class contained only 4% of the eligible
couples (N=32). Continued investigation and cross-validation of the latent classes which
emerged from this study is important as several types of couple-level IPA has severe risks for
child maltreatment post-divorce. A deeper understanding of couple-level IPA can provide
information for constructing safer family structures and parenting arrangements post-divorce
(Kelly & Johnson, 2008).
One other interesting class that emerged from this data was the Lower Level Coercive
Controlling Violence—father perpetrator. The elevations on the IPA subscale dimensions were
clearly lower than those found in the Coercive Controlling Violence—father perpetrator class,
and this class did not have elevations on physical abuse. This Lower Level class may represent a
formerly Coercively Controlling Violence—father perpetrator class; however, because the
participants in the study were all divorcing and no longer living together the opportunities
available for the father to perpetrate other forms of abuse were reduced. Because this Lower
Level class represents 35% of the sample, it is important to cross-validate these findings using a
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 150 2007-WG-BX-0028
separate sample of abusive but intact or recently separated couples to support these findings.
Future research should also include in-depth interviews with both parents to fully understand the
nature, frequency, context and consequences of IPA throughout the entire relationship.
In terms of mediator identification of IPA, using clinical interview data, mediators
identified well over half of couples as presenting IPA (59%) and, in general, the mediators
tended to successfully identify the couples with the highest levels of IPA. In comparing clinical
interview data to behaviorally specific questions in a questionnaire format, however, about one
third of those not identified with IPA by the mediator had reported on the questionnaire at least
one incident of either threatened and escalated violence or sexual intimidation/coercion/assault,
suggesting some degree of false negative classification by the mediators.
In terms of the couples with the very highest levels of reported IPA, mediators correctly
identified the seven cases where couples reported mutually high levels of violence; however,
they failed to identify eight of these cases (six couples where the mother reported father as the
perpetrator of high violence, and two where the father reported the mother as the perpetrator of
the high violence). Interestingly, although not identified as a case with IPA by the mediator, one
of the high father IPA cases was nevertheless screened out for IPA. False negatives are of
particular concern in this context, as failing to accurately identify IPA when it is present, is likely
more detrimental than inaccurately identifying IPA as present when it is not. In the first case,
potential safety is compromised, as well as the potential fairness of the mediation process itself,
and in the latter, extra attention or accommodations might be provided when they are
unnecessary, without creating the potential for any harm. Thus, in the context of mediation, it’s
better to over- than under-identify IPA among high conflict, divorcing couples.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 151 2007-WG-BX-0028
It is important to understand how these cases failed to be identified by the mediator. In
the current naturalistic study, methods of IPA assessment were self-report, one for the data
obtained via a face-to-face semi-structured clinical interview and the other via written responses
to questions presented in questionnaires. The mediators at the study site did not use the
questionnaire data to make IPA determinations but instead relied solely on the semi-structured
clinical interviews to determine IPA.
Although intuitively attractive to practitioners, reliance on semi-structured clinical
interviews that have not been empirically validated may not accurately identify some high
violence couples. The semi-structured clinical interview (PMI) was used unsystematically and in
a non-standardized fashion, with much mediator discretion about whether, how, and what
questions were asked; therefore, some cases were missed – false negatives. It is important to
note, however, that there were no false positives (cases identified by mediators as IPA that were
not identified as such on other measures), again a concern of less significance than that of false
negative detection, given the potential consequences of under- versus over-detection of IPA.
Research is mixed in terms of whether individuals are more likely to disclose sensitive
information, such as IPA, in the context of an interview or a questionnaire format. Consequently,
it seems prudent to employ both methods, and to encourage mediators to use questionnaire data
in their assessment of IPA, and to adopt a standardized approach to administration, in which
ALL questions about IPA are asked of ALL individuals, irrespective of whether or not any red
flags, or suspicions are raised about IPA by the couples or on questionnaire data. The model of
universal screening is likely to be considerably more effective in reducing the rate of false
negatives in the IPA screening process.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 152 2007-WG-BX-0028
Although not yet well validated, several instruments have been developed for this
purpose (Ellis & Stuckless, 2006; Holtzworth-Munroe, Beck & Applegate, 2010; Michigan
Supreme Court, 2006; Putz, Ballard, Holtzworth-Munro, Applegate, & Beck, 2009). Universal,
comprehensive screening for IPA is currently rare (Holtzworth-Munroe, et al., 2010) but gaining
popularity. The Ontario Canada Court of Justice and the Superior Court of Justice recently
provided extensive guidelines for required IPA screening for all potential clients to be provided
family mediation in these courts (Ontario Attorney General, 2011).
Clinically this finding also lends some support to the victim advocate’s position that in
general, mothers do not tend to exaggerate the levels of IPA in interviews with the mediator to
gain unfair advantage in determinations of custody. If this were the case it would be expected
that the IPA-related information provided in face-to-face interview with mediators would be at
least equally or more severe than that reported on the written questionnaires.
It is crucially important that mediators accurately identify IPA because without
identification of IPA, a) the cases will not be screened out of mediation when appropriate, and/or
b) the mediation process will likely not be modified to accommodate potential victims. This
study confirms previous mediation program director estimates that rarely are cases screened out
of mediation (Pearson, 1997). In this study, although 59% of the cases were identified by the
mediator has having IPA, only 3% of the cases were screened out due to IPA concerns identified
by the mediator prior to mediation. Although few cases were screened out, the mothers (but not
fathers) who were screened out reported higher levels of victimization than those that remained
in mediation. Interestingly, four couples who were not initially identified by the mediator as
having IPA were screened out later. In addition, 3% of the cases were screened out for other
reasons.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 153 2007-WG-BX-0028
There are several possible reasons for this low percentage of cases screened out. Judges
find high level IPA cases disturbing and often want the assistance of mental health professionals
in interviewing litigants and negotiating agreements that are in the best interests in the children
involved. Litigation is not necessarily the best method of obtaining in-depth, accurate
information concerning relationship dynamics, particularly if the couples do not have attorneys
and are trying to navigate the process alone. Although not all, many mediators do screen for IPA
and are able to interview each parent individually to learn about relationship dynamics. With the
couple dynamics in mind, mediators can then assist couples in negotiating agreements.
An additional pressure for mediators to process cases may come from the mediation
program administration as one of the measures of job performance is the numbers of cases
processed and the numbers of agreements obtained.
The victims may also have requested to proceed in mediation even in the face of high
levels of IPA. Some victims of IPA want to continue in mediation believing that mediation may
be their only, or best, option for working out a parenting agreement with their abuser. Without
attorneys, victims are left to negotiate directly with their abusers without professional assistance
of any kind. The victims may also have not believed the IPA to be serious or to be a factor in
negotiating an agreement. This possibility is consistent with research on battered women
indicating their tendency to minimize the severity and impact of violence in their lives. In
considering the effects of high IPA and that very few of cases are being screened out of
mediation, it is important to assess whether victims want to proceed. Previous research indicated
that many victims of IPA want to be given the choice to mediate (Alaska Judicial Council, 1992).
In this context, respecting victims’ wishes to proceed (or not to proceed) in mediation is an
important consideration, and fits with the feminist model of empowering victims to make their
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 154 2007-WG-BX-0028
own choices, in part, as a way of reconstituting their sense of personal agency and control, which
were diminished as a result of IPA perpetrated against them.
While it is important to consider victim input into decisions about mediation, it may also
be the case that they are uninformed about the potential threats to safety of the negotiated
agreements without specific restrictions on contact between parents and parenting (orders of
protection; orders restricting method of contact between parents; neutral exchanges of the
children at a specific public location, a family member’s home, or a judicially approved service
provider), and threats to the fairness of the process itself that might result from proceeding with
mediation in spite of IPA history. In this context, it might be important to educate victims about
these considerations, so that they are able to make a truly ‘informed’ and empowered decision. In
the context of this naturalistic study, it may be that mediators allowed victims the opportunity to
attempt to mediate if they wished to do so. In subsequent studies, it will be important to
interview couples mandated to mediation and mediators providing these services to learn reasons
for proceeding in the face of high levels of IPA.
A second important aspect of mediators identifying cases as having IPA is whether
victims went on to negotiate a potentially unsafe agreement in mediation. On a hopeful note,
victims of IPA may be able to safely proceed in mediation if the process is modified to
accommodate them with special procedures for safety. The mediators in the study were sensitive
to client reports of IPA. Overall, approximately one fifth of the couples in this sample were
provided some type of accommodation to promote safety for the victims in the mediation
process. The accommodations varied widely; some were focused on ensuring that victims were
not put at greater risk by attending mediation. For example, accommodations were provided to
ensure victims were able to leave the program offices and get to transportation without worrying
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 155 2007-WG-BX-0028
about whether the abuser was going to harass or intimidate them along the way (requiring
parents to leave separately; providing a security escort to the car). Other accommodations were
focused on ensuring victims were not harassed or intimidated while at the mediation program
offices (separate waiting rooms; screenings on separate days; shuttle mediation; mediation
sessions on separate days; and, an experienced male and female mediation team providing
services).
Reasons for the inconsistency in the provision of accommodations are unclear although it
is likely due to the fact that this was a naturalistic study of records from one jurisdiction, which
were not developed for research purposes but instead for mediators to use in conducting their
work. If the same results are found in other jurisdictions, it might have something to do with
mediator unfamiliarity with some accommodations or the necessity of providing them. Perhaps
researchers working with practitioners can design a study where mediators are provided with a
standardized set of accommodation practices along with hypotheses for which are best to
implement in cases involving different types of IPA, thereby empirically investigating if the
process can be improved by which accommodations are instituted across cases.
An important finding is that safety accommodations were significantly more likely to be
provided for parents who telephoned the mediation program office and expressed concerns about
mediating or made specific requests concerning the particular accommodations which would be
helpful. Accommodations were also more likely to be instituted in cases identified by the
mediator as having IPA, but there was a weak correlation between overall level of IPA and
mediators’ providing accommodations. There are several reasons for this weakened correlation.
In interviews with the mediators we found that not all the mediators considered the
special procedures identified in the literature and identified for this study as accommodations for
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 156 2007-WG-BX-0028
safety (shuttle mediation, separate screening days, separate waiting rooms, mediation on separate
days, escorts to car, experienced mediator team, parents required to leave at different times) as
such but instead considered them as “mediation techniques.” Therefore, for a number of
mediators, they did not note in the medication case file when these “techniques” were used. The
data in this study then represents an underestimate of the actual number of accommodations
provided. As noted above, future research will need to be conducted prospectively with carefully
defined definitions of accommodations provided and training in standardizing documentation of
these procedures so that all the accommodations provided are captured in the data.
These findings suggest that couples referred to mediation ought to be encouraged to
discuss with mediators IPA-related issues and to request specific accommodations to enhance
safety. IPA-related issues cannot be addressed without adequate information from the victim to
the mediator. In addition, documenting accommodations when provided will assist agencies in
tracking when these are provided and the circumstances of the cases.
One method of determining whether IPA is serious enough that outside agencies are
brought into assist is to assess calls to the local area law enforcement. A tremendous number of
parents made calls to area law enforcement agencies to request assistance in managing IPA in the
relationship. In the pre-mediation screening interview, approximately half of both parents
indicated to the mediators that police had been called for IPA or a parent had been arrested for
IPA at some point in the relationship (50% fathers and 54% reported police were called). This
high rate of calls was substantiated in local area law enforcement records, thus providing
consensual validation for the self-reported data on this variable. As noted above, using a much
more restricted timeframe (two years prior to filing for divorce up to February 1, 2007), law
enforcement records document that 40% of the families in the study had made calls.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 157 2007-WG-BX-0028
Interestingly, the calls did not come at the same point in the divorce process. The calls came in
about equal percentages in different timeframes (16% before filing for divorce, 16.5% during the
divorce process and 15 % post-divorce).
It appears that at all steps in the divorcing process, a number of parents use law
enforcement as an intervention in managing family conflict, and only a minority rely upon it at
all stages of the process. And, it appears likely that calls to law enforcement are occurring in
couples with no IPA and a range of levels of IPA. The types of calls assessed included a broad
range yet related to IPA and children (assault, child abuse, violation of order of protection,
custodial violation, trespassing, harassment, phone harassment, interfering with custody, keeping
the peace). Nearly a quarter of the couples in this study also mentioned having an order of
protection. It appears from the findings that a significant number of parents who are nonviolent
or who perpetrate lower levels of IPA, likely commit a variety of crimes against their co-parent
during the divorce process. This suggests that the stressful nature of the divorce process,
particularly among high conflict couples, may result in new cases of IPA among those who had
not been previously violent, and may exacerbate the violence among those with prior histories of
violence. Consequently many of the victims appear to turn to area law enforcement to respond to
the conflict or to enforce an order of protection.
Several important conclusions can be drawn from these findings. Neither parent is
underreporting the number of calls to law enforcement when questioned in individual face-to-
face interviews with mediators. In addition, parents generally agree about the rates of law
enforcement involvement with the family. And, cases that mediators identified as having IPA
were the most frequent callers. This is not surprising as outside agency involvement with the
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 158 2007-WG-BX-0028
family (including shelters, hospitals, and orders of protection) was a significant factor driving
mediators identifying a case as having IPA (Beck et al., 2010).
Although less common than calling area law enforcement, 42% of the cases in this study
reported having orders of protection at some point in the relationship. Cases identified as having
IPA by the mediator were found among those reporting orders of protection. Interestingly, the
level of substantiation of the order of protection (mentioned only or mentioned and
substantiated) did not matter for mediators in identifying a case as having IPA. Unfortunately,
the available data did not report on rates of violation of such orders, thus it is impossible to
gauge their effectiveness in limiting or preventing IPA in this sample. Future research needs to
track violations of orders of protection in divorce cases.
In summary, there is strong empirical evidence that most couples attending divorce
mediation report some form of IPA and, second that few cases are actually screened out of
mediation. It may be true that “common couple violence” (Kelly & Johnson, 2008) is the most
common type of IPA in the general population; however, for many couples in this sample the
IPA appears to be much more serious. This finding strongly suggests that there is a need to
carefully screen for and document all levels and types of IPA.
Mediators identify many fewer cases as having IPA than couples identify themselves.
The lower number of mediator-identified cases of IPA relative to parent-identified cases of IPA
reflects a clinical decision by mediators to identify cases wherein more dimensions of IPA, and
higher severity within the dimensions, were designated IPA as compared to when any dimension
of abuse was reported. Thus, mediators are using clinical judgment to sort cases and do not
designate every case that reports any abuse as one with IPA. Unfortunately, there were a small
number of high level IPA cases that were not identified as such by the mediators. Basing
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 159 2007-WG-BX-0028
classification of cases on nonsystematic, semi-structured clinical interviews may not be the most
effective method of assessing IPA. This suggests the need for more systematic assessment and
management of IPA in the mediation context.
Whereas some argue that mothers exaggerate the seriousness of IPA to gain unfair
advantage, at least in terms of reports of area law enforcement calls and/or arrests and reports of
orders of protection, parents substantially agree when reporting these to mediators in semi-
structured clinical interviews. Thus, it does not appear that one parent is exaggerating the
seriousness of IPA and/or outside agency involvement with the family to mediators.
Mediators were sensitive to both mothers’ and fathers’ reports of IPA when deciding on
whether to provide safety accommodations for couples. When clients telephoned the mediation
program offices with concerns or specific requests for accommodations, mediators responded at
an even higher rate than when mediators themselves identified the need. Although mediators in
this study responded by modifying the mediation process to accommodate many of these
couples, the rates reported are likely an underestimate of those actually provided. For mediation
programs tracking this important information, documenting when accommodations are provided
and for what reason will be critical.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 160 2007-WG-BX-0028
Goal 2: Assess whether mediation agreements, divorce decrees and/or parenting plans
include safety measures in cases involving self-reported IPA.
If cases are not screened out and the couple has IPA in the relationship, and an agreement
is reached, an important consideration is the adequacy of agreements in providing protections for
children (parenting time supervision) and safety restrictions (orders of protection, limits on
contact, neutral exchanges of children in public places or third party exchanges) for adult
victims. In this study there were reports of frequent psychological abuse and, rarer still but
present, reports of physical abuse, sexual intimidation/coercion/assault, and threatened and
escalated physical violence. While full or partial agreements were reached in many cases,
provisions for supervised parenting time and restrictions on contact between parents (orders of
protection; orders restricting method of contact between parents; neutral exchanges of the
children at a specific public location, a family member’s home, or a judicially approved service
provider) were virtually absent in the entire sample of cases.
This finding could lead to speculation that the mediation did not protect these victims and
their children or effectively discriminate among various dimensions of IPA. This speculation is
buffeted by the finding noted above that, regardless of whether it was the mother or father
reporting victimization, couples who later failed to reach any agreement and were transferred
back to the traditional litigation process were found to have previously reported the highest
overall IPA scores (2 standard deviations above the mean), followed by those who reached only
partial agreement. The lowest IPA scores were reported by those who ultimately reached full
agreements in mediation. Thus, those couples of most concern, in terms of IPA severity did not
reach an agreement in mediation and returned to court. For these couples with high IPA, it would
be helpful if there was a the requirement of representation in the adversarial process.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 161 2007-WG-BX-0028
Allowing victims the opportunity to choose to proceed (or not to proceed) in mediation is
an important consideration. Proponents of mediation have long argued that providing a safe,
neutral and controlled environment for victims allows victims to have a voice and express their
views (Duryee, 1995; Erickson & McKnight, 1990; Girdner, 1990; Kelly, 1995; Landau, 1995;
Magana & Taylor, 1993; Marthaler, 1989; Newmark et al., 1995; Pearson & Thoennes, 1984;
Vincent, 1995; Yellott, 1990). On the contrary, opponents of mediation have argued that most
often victims’ voices are ignored or marginalized in mediation thus recreating the imbalance of
power existing in the relationship (Bryan, 1992; Cobb, 1997; Grillo, 1991). Staying sensitive to
the desires of victims of IPA without forcing decisions to proceed or not to proceed is a delicate
balance all mediators must face. It may be that some of the victims in this study agreed to enter
mediation because they wanted to have a voice and express their views; but, they were
empowered enough to refuse to agree to proposals that would not adequately protect them and
their IPA-exposed children.
For example, if the abusive parent would not agree to supervised exchanges, restrictions
on contact between parents or supervised parenting time, the victim may have exercised the
option of not making a mediation agreement and moving back into traditional court processes.
As a result, victims of the highest levels of IPA might have wanted to participate in mediation
but not to be forced to make an agreement that was not suitable for their family. It is important
that future research gather data empirically sound data concerning why victims of IPA choose to
proceed or not in mediation. This data could assist mediators in further refining the mediation
process to safely allow victims a voice.
In interviews with mediators, we learned that in some cases mediators protect victims by
moving ahead initially with mediation and then terminating the mediation within the first
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 162 2007-WG-BX-0028
session. By doing so, mediators take responsibility for terminating mediation as opposed to
terminating mediation based on what the mediator was told by the victim in the screening
interview. Several mediators at the study site believed terminating mediation after only the
private screening interview might subject the victim and violence-exposed children to additional
violence for revealing the IPA to the mediator. An important point is that the actions a mediator
can take to protect victims may not be captured in statistics concerning the number of cases
screened out after the first individual interview with the parents. Future research will need to
design prospective studies to better capture mediator and victim decision-making concerning
whether to proceed and why, and methods used to protect victims in specific cases when
decisions to proceed or terminate the sessions are made. Only with this rich information can we
truly understand the difficult balance mediators must take in working with couples with IPA.
Another possible form of safety precaution that could be included in the divorce decree
was to legally recognize that a relationship had a history of IPA. Unfortunately only 2% of the
cases included a legal finding of domestic violence in the divorce file and among some of the
cases of extremely high IPA there were no legal findings in the divorce file. In the study
jurisdiction the legal finding of “significant domestic violence” (Domestic Violence and Child
abuse; Ariz. Stat. Ann. §25-403.03) has important legal ramifications. The finding precludes a
joint custody award. In addition, the finding creates a rebuttable presumption that an award of
custody to the parent who has committed the act of IPA is contrary to the best interests of the
child. The perpetrating parent then also has the burden of proving to the court that parenting time
will not endanger the children and the court can place restrictions on contact between the
perpetrating parent and the children (orders of protection; orders restricting method of contact
between parents; neutral exchanges of the children at a specific public location, a family
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 163 2007-WG-BX-0028
member’s home, or a judicially approved service provider, prohibit overnight parenting, require
a bond for the safe return of the child, limit use of alcohol or controlled substances during
parenting time).
A complicating factor for victims in this legal analysis is that the statute regulating this
finding (Ariz. Stat. Ann. § 25-403) states that the presumption does not apply if both parents
have committed an act of domestic violence (i.e., causes or attempts to cause sexual assault or
serious physical injury, apprehension of imminent serious physical injury, behaviors for which
the court could issue an order of protection). Clear evidence of one parent as the abuser and the
other parent having committed no defined acts are necessary for the statute to govern. While this
legal finding could be extremely important for victims in assisting them in obtaining decrees
which include safety precautions, it could also be difficult to qualify under it files cross or
counter pleadings against the victim and the court finds that both committed domestic violence
and issues a dual order or counter orders of protection. It is therefore important for advocates for
victims, lawyers representing victims and policy makers to strongly encourage victims who
would quality under this statute to seek these legal findings so that statutory safety precautions
are enlisted to help victims obtaining decrees that include safety precautions. An issue for further
debate is if statute is useful in the event that victims’ defend themselves or their violence-
exposed children physically. Are the victims then no longer considered victims under the statute?
One troubling finding that highlights the need for continued vigilance is that two cases
with a legal finding of domestic violence had parenting plans with joint legal and physical
custody of the children. It is unclear how this happened. It may be that both had a legal finding
of IPA. There could have been some extenuating circumstances. Why this parenting arrangement
in these two cases is unclear.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 164 2007-WG-BX-0028
When agreements were made in mediation, by far the most frequent agreement was
primary physical custody to the mother with parents sharing joint legal custody. In fact in
general, the specific mediation agreements as to physical and legal custody did not significantly
differ according to the level of reported overall IPA. Mothers were the most frequent primary
residential custodians in mediation agreements no matter the level of IPA was reported (Beck,
Walsh & Weston, 2009).
Mirroring the custody arrangements made in mediation, mothers were most commonly
awarded primary physical and/or legal custody (70%) while fathers were awarded primary
physical and/or legal in 8% of the cases. Joint custody was awarded in 18% of the cases. The
courts were sensitive to IPA in making these arrangements as joint custody was rarer in either
mother-perpetrated or father-perpetrated cases of IPA, or when the reports of IPA were
combined for a family level of IPA.
Although some couples with high IPA reached an agreement, an important question is
what happens when couples with higher levels of IPA do not reach an agreement in mediation.
Interestingly, the final parenting plans in these cases were very similar results to those
arrangements that were reached in mediation. Mothers were most commonly awarded primary
physical and/or legal custody (71%) while fathers were awarded primary physical and/or legal in
11% of the cases. Joint custody was awarded in 18% of the cases.
The encouraging news is that for couples who did not reach an agreement in mediation,
safety restrictions were put into place in the parenting plans at the final divorce at a much higher
rate than security restrictions negotiated in mediation (18% at time of divorce versus 6% in
mediation) and those appearing generally in cases at the time of the divorce (7%). It appears that
some victims whose abuser would not agree to safety precautions in mediation, chose to not
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 165 2007-WG-BX-0028
make an agreement in mediation, were able to return to court and obtain safety restrictions from
the courts at the time of divorce.
Mothers were by far the most frequent primary residential and/or legal custodians both in
mediation and in final divorce decree arrangements. This arrangement continued to even the
highest level of combined mother and father IPA (2 standard deviations above the mean for both
mothers and fathers). When taking a somewhat less severe classification of mutual IPA (those
families classified as Mutually Violent Control in the latent class analysis) the majority of
couples did not reach agreement in mediation but returned to the court for resolution of the
disputed issues. For those who did remain in mediation, mother custody was the most frequent
outcome. This was also true for court awards of custody. Over half of the couples’ awards of
custody were for mother primary custody. Somewhat surprisingly, in court ordered awards 15%
were joint custody awards.
In summary, rarely are cases screened out of mediation after the initial meeting with the
parents individually, rarely do abusers agree to restrictions placed on parenting or contact
between parents in mediation agreements, rarely are restrictions placed on abusers in divorce
decrees, and rarely is a legal finding of domestic violence made in cases with any level of IPA. A
complex analysis of the findings indicates that screening cases out of mediation at an initial
meeting may be risky. Mediators are in the difficult position of choosing whether to provide a
victim an opportunity to choose whether she/he would like to proceed and altering the
procedures to make the mediation as comfortable as possible for the victim, or making the choice
for the victim of whether proceed or not to proceed in mediation. While the choice is simple if
the victim does not want to proceed; the choice becomes much more complex if the victim does
want to move ahead. Allowing a victim to choose could be empowering and could provide a
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 166 2007-WG-BX-0028
forum for the victim to safely have a voice with their abuser. Allowing the mediation to proceed
also gives mediators a chance to observe the dynamics of the couple and evaluate the fairness of
the process while it is in process. In addition, a mediator terminating the mediation after the
initial session may better protect the victims than screening them out initially.
The critical piece in this analysis is that mediators must be well trained in identifying IPA
to make these nuanced decisions at every point in the mediation process. Future research must
focus on assessing how IPA impacts the negotiation process for victims. In the meantime,
because virtually all cases involve some form of psychological abuse and coercive controlling
behaviors, mediators may want to pay particular attention to focusing on the parents’
understanding and then remediating the impact of such abuse on all members of the family,
particularly the children. Psychological abuse can be just as devastating as other forms of abuse
(O’Leary, 1999; Stark, 2007; Theran, Sullivan, Bogat, & Stewart, 2006).
While restrictions are rarely placed on contact between parents and on parenting in
general in agreements in mediation, it also appears that victims of the highest levels of IPA are
leaving mediation without agreements. These couples then go on to court and have restrictions
included in their divorce decrees at a much higher rate as compared to restrictions in mediated
agreements and for couples with lower rates of IPA in mediation. It is possible, however, that
standardized, multi-modal screening for IPA would result in more accurate identification and
dispositions of these high risk couples early on in the process, before they go through mediation
and fail to reach an agreement.
As was done in this jurisdiction, it is important for mediators, advocates and lawyers
working with victims to remind victims in most mandatory mediation jurisdictions they are
required to attend the orientation screening session but they are not required to reach
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 167 2007-WG-BX-0028
agreements, particularly agreements that are unsafe. It is important that the limitations of the
mediation process are acknowledged. This reminder should not be buried in a long list of
potential benefits of mediation but should instead be an important focus of the orientation.
Because mediators are not judges they cannot order that specific terms (such as protections for
children or restrictions on contact between parents) be included in mediation agreements even if
it would be in the bests interests of the children and the adult victim (Beck, Walsh & Weston,
2009). Although mediators in this study allowed couples to negotiate agreements that did not
often include specific restrictions, couples with the highest levels of IPA did not reach
agreements and, for many of these couples, the divorce decrees included restrictions. If
advocates is for victims and policy-makers have as a goal the desire for safety restrictions to be
included in mediation agreements then a process other than traditional mediation will likely be
required. Hybrid models of mediation may be an option to consider, particularly for couples with
high levels of IPA and who do not have attorneys to assist in framing the issues and negotiating
the terms of divorce decrees and parenting plans if mediation fails.
Mediator/arbitrator hybrid models of mediation have been used in community mediation
and results of early studies investigating these models found they produced effective negotiation
behaviors during mediation (Pruitt, 1995; McGillicuddy, Welton & Pruitt, 1987). Researchers
investigated three conditions:
(1). Straight mediation-- mediator did not issue any decisions following unsuccessful
resolution by the clients (med only);
(2) Mediation/arbitration-- mediator did issue a legally-binding decision if the clients
did not reach agreement (med/arb[same]); and
(3) Mediation/arbitration—a different mediator issued a legally-binding decision if the
clients were unable to reach an agreement (med/arb[differ]) .
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 168 2007-WG-BX-0028
The results of the study found that in med/arb(same), disputants engaged in more
problem solving, were less hostile and less competitive. Somewhat surprisingly, in this condition
the mediators/arbitrators were also viewed as less forceful, and the clients saw themselves as
more involved in the decisions. And, mediators in both mediation only and med/arb(differ)
conditions were less involved than those in med/arb(same) group (Beck & Sales, 2001).
Results of studies investigating similar hybrid models in the divorce mediation context
indicated that satisfaction rates did not differ between those jurisdictions in California where
mediators make recommendations to the court in unsuccessful mediation, and thus resemble the
med/arb(same) model, and those mediations where the process is confidential, which resembles
the more traditional mediation process (Kelly, 1996;Depner, Cannata, & Ricci,1995). Future
research would, however, need to carefully empirically investigate the short- and long-term
outcomes for self-representing, high IPA couples assigned these hybrid models as compared to
mediation as usual. The potential benefits of such models are that critical IPA screening
information provided in mediation could be incorporated into agreements to protect victims and
violence-exposed children in the volatile years post-divorce.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 169 2007-WG-BX-0028
Goal 3: Assess the frequency with which mediation agreements/divorce decrees are
re-litigated over time.
In the five years post-divorce, the majority of families returned to court at some point for
hearings, orders and/or stipulations. The good news is that regardless of the legal procedure the
family participated in (hearing, order, or stipulation) the majority of families returned between
one and five times in the five year period. The bad news is that, a small subset of families was
responsible for consuming a tremendous number of court resources; 31 families were responsible
for 31% of the hearings in the five years of the study. For these families, continued court
involvement was the hallmark of their post-divorce life. Depending on the circumstances of the
cases, excessive involvement of the courts in families is being discussed and recognized as a
form of economic abuse and stalking (Miller & Smolter, 2011; NCJFCJ, 2008; Pollet, 2011).
Data collection for this study was conducted prior to the current efforts of courts to
manage these high volume cases. Recently courts have developed Parenting Coordinator
Programs to address the needs of these high conflict, high volume users of court processes
(AFCC Task Force on Parenting Coordination; 2005; Beck, Putterman, Sbarra & Mehl, 2008;
Coates, Jones, Bushard, Deutsch, Hicks & Stahl, et al, 2003; Hayes, 2010; Henry, Fieldstone &
Bohac, 2009). These programs enlist the assistance of a parenting coordinator to quickly respond
to and resolve many of the issues that would have required couples return to court. The proposed
benefit to the couple and the children is that issues get resolved much more quickly and do not
linger and escalate. The benefit for the court systems is the reduction in returns to court.
Although few have been evaluated for effectiveness, those that have been evaluated have shown
some promise (Beck et al., 2009; Henry et al., 2009). Continued court-based efforts to develop
case management processes for these high conflict families and empirically evaluate their
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 170 2007-WG-BX-0028
effectiveness are essential for reducing the burden these high users place on their families and
the courts.
Since all of the participants in this study had disputes concerning the future of their
children; it is not surprising that the majority of the post-decree hearings, stipulations and orders
post-decree were child-related (child support, child custody, or parenting time). Issues related to
child support required the most court resources. Fathers were overwhelmingly more likely to
request hearings to modify child support while in contrast mothers were only as likely as the
Court or an Unknown source to request hearings to enforce child support orders. Fathers are
somewhat more likely to request hearings to modify parenting time. Fathers are also
overwhelmingly more likely to request modification of spousal maintenance while mothers are
overwhelming likely to request hearings for issues related to IPA, child abuse and harassment.
Thus, it appears fathers request the court modify financial obligations and mothers request the
courts enforce fathers’ financial obligations. It is not the case that mothers are more often
instigating changes in financial arrangements but mothers and sometimes courts and presumably
welfare agencies are more often requesting obligations to be met by fathers.
Contempt actions for enforcement of child support orders are often brought by the court
(if the payor’s wages are being garnished and paid through the courts) or by the department of
public welfare (if the recipient or any children are receiving public assistance). When these
actions are taken over the objections of the mothers, it could place the victim at increased risk. A
national survey of child welfare agencies found that less than half of the agencies had written
policies and procedures for addressing the issue of IPA in families seeking their services (Hazen,
Connelly, Edleson, Kelleher, Landsverk, Coben et al, 2006; Yearwood, Dowd & Starsoneck,
2008) and another study found that only slightly more than one half of the surveyed welfare
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 171 2007-WG-BX-0028
agency caseworkers reported they always screened families for IPA, with the remaining
caseworkers indicating they sometimes screen for IPA (Bourassa, Lavergne, Damant, Lessard &
Turcotte, 200; Yearwood et al., 2008). These results are concerning in that another study of 1082
applicants for public assistance in four different welfare offices found that 40% of the applicants
for public assistance confided that they were currently experiencing domestic violence (Pearson,
Thoennes & Griswold, 1999; Yearwood et al., 2008).
At times victims will specifically not seek child support or enforce child support orders.
A study found that 7% of female victims of IPA in Colorado applied for a waiver releasing the
other parent from paying child support (Pearson & Griswold, 1997; Yearwood et al., 2008). Of
these women, 76% viewed the abuser as still dangerous and believed that seeking child support
would make the situation even more dangerous; 38% had a confidential address; and, 72%
moved to avoid the abuser (Pearson & Griswold, 1997; Yearwood et al., 2008). The complexity
of post-divorce parental relationships needs further investigation particularly concerning
financial relationships as they place the most significant burden on the post-divorce custodial
parents and may place victims of IPA and their IPA-exposed children to increased risk of future
violence.
In this study fathers were predominantly the parent paying child support and spousal
maintenance as mothers were overwhelmingly the primary physical custodians of the children.
An important question for future research is to investigate why fathers’ were overwhelmingly the
instigators of legal disputes concerning financial arrangements. A simple answer is that the laws
governing child support payments do not take into consideration important costs for fathers
resulting in economic hardships for fathers (Braver, Shapiro & Goodman, 2006). Another
possibility is that physical custody arrangements change after the divorce leaving fathers with
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 172 2007-WG-BX-0028
more time with their children and more costs associated with the more time. It could also be due
to changes in the fathers’ financial status, changes in child development and child needs with
respect to outside activities.
An important consideration is that continued legal wrangling over child support and/or
spousal maintenance could be a method of harassing and abusing an ex-spouse through the legal
system (Miller & Smolter, 2011; NCJFCJ, 2008; Pollet, 2011). Mothers do overwhelmingly
request hearings for IPA, child abuse and harassment; however IPA measured at the time of
mediation had was only a weak influence on post-decree hearings and orders. It may be that IPA
changes over time with increases in IPA for some families.
The good news for mediation programs is that mediation does appear to have a positive
impact on reducing post-divorce litigation. Those couples who were able to reach a full
agreement in mediation required significantly fewer post-decree hearings and orders.
Unfortunately, not all couples are able to reach a full agreement and reaching a partial agreement
did not have the same benefits in reductions in post-decree litigation. Couples with only partial
agreements often continue to return to court for additional stipulations post decree. Stipulations,
however, indicate some degree of cooperation and an ability to resolve issues without hearings
and thus place less of a burden on the courts.
The finding that mediation reduces post-decree litigation is important for policy makers
considering whether to develop or fund mediation programs. In the study jurisdiction, well-
trained mediators were given wide latitude to provide as many sessions with couples as needed
and were able to assist many couples in reaching agreements. Over half of the couples entering
mediation were able to negotiate a full agreement. And, in doing so, these couples required fewer
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 173 2007-WG-BX-0028
court resources post-divorce. For the couples unable to reach agreement in mediation, mediation
added a potentially time consuming step to the divorce process.
An important finding that is well supported by the extant research on separation and
divorce is that for some couples IPA did not stop at separation or divorce. For a minority of
couples, calls to law enforcement, orders of protection, and court hearings regarding IPA were
reported after separation and after the divorce was finalized. Again, it is important that any
agreements between the couples address the risk of continued IPA.
Goal 4: Test a multivariate conceptual model using variables that are hypothesized to
affect mediation, divorce case and post-divorce outcomes (see Figure 16).
To better understand the dynamic nature of the divorce process and the many interrelated
variables, we also used a multivariate analytic approach. We began by aggregating theoretically
relevant constructs across an entire system to guide our inquiry into those leading to outcomes
associated with mediation, the divorce and the post-divorce legal processes. We then created a
theoretically-specified causal order for the constructs and then ran a series of multiple regression
models in the context of a Cascade Model (Figure 16). Beginning at the far left of Figure 16, 15
specific hypotheses were tested relating to the relationship between the constructs (see pp. 113-
118). These can be categorized into three temporally related sets of variables.
The first set was variables that occurred prior to mediation (marital stressors, father and
mother perpetrated coercive controlling behaviors and IPA) and we hypothesized these variables
had a strong impact on mediation outcomes as well as downstream outcomes associated with the
divorce and post-divorce. The next set of variables was those that occurred as a result of
mediation (mediator call of IPA, mediator providing special procedures/accommodations, and
whether an agreement was reached). The third set of variables was those that occurred as a result
of the final divorce (court finding of IPA, restrictions in decree on the abusing parent, order of
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 174 2007-WG-BX-0028
protection, child maltreatment and custody decree favoring mothers). The final set of variables
related to post-decree outcomes that were possible after the divorce process (post decree
hearings, orders and calls to law enforcement); however, fortunately not all couples had post-
decree interactions with the area law enforcement or the courts.
Prior to Mediation
The first four hypotheses concern the relationships between marital stressors, coercive
controlling behaviors and parental IPA. We hypothesized that marital stressors are an important
trigger forms of coercive controlling behaviors. Marital stressors assessed in this study included
mental and physical health concerns, use of counseling/therapy, use of medications, substance
abuse/use concerns and a measure of poverty. Decades of research has documented the
emotional chaos and financial hardship experienced by families going through a divorce
(Ahrons, 1994; Emery, 1994; Fine & Harvey, 2006; Hetherington & Kelly, 2002). Divorce is an
emotionally difficult and stressful time on any couple, even the most physically and mentally
healthy and financially secure. We hypothesized that those couples experiencing additional
stressors noted above could be triggered to use coercive controlling behaviors or IPA. The
results indicated that this was the case. Marital stressors influenced fathers’ use of coercive
controlling behaviors. For mothers, both fathers’ use of coercive controlling behaviors and
marital stressors influenced mothers’ use of coercive controlling behaviors. Although significant,
the data was limited to those stressors documented in case files. To more fully understand the
relationship between marital stressors and coercive controlling behaviors, future research needs
to investigate more completely distal and proximate marital stressors that are found to trigger
instances of IPA (Finkel, DeWall, Slotter, Oaten & Foshee, 2009) and the impact of these
triggers on couples in the context of an ongoing divorce.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 175 2007-WG-BX-0028
The analyses then focused on three hypotheses concerning the relationships between IPA
(psychological abuse, physical abuse, threatened and escalated physical violence and sexual
intimidation/coercion/assault), marital stressors and coercive controlling behaviors. Prior
research indicated a strong relationship between coercive controlling behaviors, and the other
dimensions of IPA for this population (Kelly & Johnson, 2008). There were significant
relationships between these variables; however, there was a difference in what influenced
fathers’ and mothers’ use of IPA.
Fathers’ use of IPA was strongly influenced by his coercive controlling behaviors
towards the mother and marital stressors but not by mothers’ use of coercive controlling
behaviors towards him. At least for fathers in this sample, it appears that when coercive
controlling behaviors failed to achieve the goal of controlling the mother additional dimensions
of IPA were then used. Mothers’ use of IPA is strongly influenced by her use of coercive
controlling behaviors and the fathers’ use of coercive controlling behaviors but not his use of
IPA or marital stressors. From these results it appears that mothers’ struggle for control turns into
IPA whereas for fathers marital stress drives his use of coercive controlling behaviors. If these
coercive controlling behaviors fail he uses IPA to deal with the marital stress and control the
relationship with the mother. In our prior work we hypothesized, and then confirmed using
structural equations modeling, that control is the motivator for other dimensions of IPA (Tanha
et al., 2009). In other words, when coercive controlling behaviors fail, physical dimensions of
IPA (physical abuse, threatened and escalated physical violence; sexual intimidation/coercion/
assault) are then used in service of the goal of control. It appears that attempts at control did fail
for both mothers and fathers and both moved into using other, more severe forms of IPA.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 176 2007-WG-BX-0028
We then moved to the hypothesis concerning the effects of child maltreatment on
mediation outcomes, divorce and post-divorce outcomes. We hypothesized that child
maltreatment would co-occur with IPA and that mothers would have less favorable custody and
parenting time outcomes when this occurred. Reported child maltreatment was influenced by
mothers’ use of IPA against the father but was significantly and negatively influenced by
mothers’ use of coercive controlling behaviors. Reported child maltreatment was not, however,
influenced by of fathers’ use of either IPA or coercive controlling behaviors or marital stressors.
These findings are interesting in that child maltreatment is extremely important for
making custody decisions post-divorce. There are very few studies investigating co-occurrence
of IPA and child maltreatment, particularly in the context of reports and re-reports to Child
Protective Services. The ones that do exist generally focus only on patterns of physical abuse
between both parents and children (Jouriles et al., 2008; Mahoney, Donnelly, Boxer, & Lewis,
2003; Slep & O’Leary, 2005). Two studies using small samples found the most common form of
co-occurrence was for both spouses to engage in IPA and for one or both parents to engage in
severe aggression toward the child (Jouriles, et al., 2008). And, in these prior studies mothers
were about as likely as fathers to use severe aggression toward the child. It is interesting that in
this study we found that the predictive association was more on mothers’ maltreatment of
children as opposed to fathers’.
A unique feature of this study is that we were able to assess a broader range of parental
IPA behaviors. By including coercive controlling behaviors we found there was a protective
effect of a mother who uses coercive controlling behaviors against the father on the reports of
child maltreatment. From these findings it appears that if mothers’ coercive controlling
behaviors of fathers is effective the mothers do not move on to physical dimensions of IPA with
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 177 2007-WG-BX-0028
either the fathers or the children. Alternative explanations for the relationship between mothers
perpetration of both IPA and child maltreatment are that (1) fathers made retaliatory and possibly
false reports of child abuse against the mother and (2) that mothers disciplined the child harshly
knowing that if she did not the father would do so far more harshly. One unpublished study from
a hospital-based program at Boston’s Children’s Hospital found that 16 months after children
were released from the hospital 85% of the battered women were no longer living with their
batterer, the child abuse stopped and only 3 of the 500 children studied were subsequently placed
in foster care (Taylor, 1998). It appears that separating the parents was instrumental in stopping
the child maltreatment regardless of whether it was perpetrated by the mothers or the fathers.
Future research needs to continue this finer grained analysis of the relationship between
different types of IPA and coercive controlling behaviors and child maltreatment. Focusing
solely on the co-occurrence of physical forms of IPA and child maltreatment also neglects to
take into consideration the serious and lasting impact of the more psychologically oriented forms
of IPA and child maltreatment (O’Leary, 1999; Stark, 2007; Theran et al., 2006) and the
complex relationship between the two.
Mediation Outcomes
The next four hypotheses concern the relationships between the measures of IPV,
coercive controlling behaviors, and mediation outcomes (whether a mediator determines a case
to include IPA, whether the mediator offers the couple special mediation procedures to
accommodate IPA, and whether couples make an agreement in mediation). We hypothesized that
both mothers’ and fathers’ IPA and coercive controlling behaviors would predict whether a
mediator would determine that a case has IPA. We also hypothesized that couples’ IPA and
coercive controlling behaviors and whether couples were determined by the mediator to have
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 178 2007-WG-BX-0028
IPA would predict whether a mediator would offer a couple special mediation procedures to
accommodate IPA. And, we hypothesized that mediator identified IPA, special accommodations
and the parents’ IPA and coercive controlling behaviors would predict whether a couple
negotiated an agreement in mediation. These hypotheses were supported in interesting ways.
The mediator’s determination that a case had IPA was influenced by both the mothers’
and fathers’ reported levels of IPV. This is important in that this finding indicates that mediators
are sensitive to both mother and fathers’ reports of IPA. This mediator IPA determination is
extremely important for the couple in that it opens door for the mediator consider a host of
options to protect the safety of the victim and violence-exposed children, in the short and long-
term. The mediator can suggest procedural accommodations to ensure the victim feels safe
enough to represent their own interests during the mediation process.
To ensure the seriousness of the IPA is documented, is assessed in terms of effects on
parenting, and that it comes before the judge to consider in determining custody arrangements,
the mediator can recommend the victim obtain an assessment of IPA. These assessments can be
obtained at relatively low cost from a domestic violence advocate agency and sometimes well-
trained domestic violence units of local law enforcement agencies. These agencies can also assist
the victim in deciding how best to proceed in terms of obtaining orders of protection and other
safety measures to protect themselves and their IPA-exposed children. If the victim has
economic resources, a mediator can recommend the victim request a custody evaluation from an
evaluator who is well trained in detecting IPA and making appropriate recommendations to
protect the safety of the victim and the IPA-exposed children. Suggestions to contact custody
evaluators and law enforcement agencies are predicated on the critical assumption that both are
well-trained in the accurate assessment and detection of IPA. Many custody evaluators are not
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 179 2007-WG-BX-0028
well-trained and thus can create conditions that lead to continued victimization (Dalton, Drozd,
& Wong, 2006).
Having a determination that a couple had IPA in the relationship significantly influenced
the decision of the mediator to provide special accommodations for the couple in mediation.
Mediators’ provision of special procedures was also significantly influenced by mothers IPA
against fathers but not fathers IPA against mothers. And, neither parents’ coercive controlling
behaviors was independently influenced by special procedures. It was hypothesized that coercive
controlling behaviors, would be independently detrimental to victims’ confidence in negotiating
with an abuser. This was not found in this data. The reasons for this lack of direct relationship
could be explained by the indirect relationships coercive controlling behaviors has through IPA.
Both parents’ IPA was highly influenced by their perpetration of coercive controlling behaviors.
It may also be that victims’ who experience coercive controlling behaviors in the absence of
physical forms of IPA did not request special procedures be provided. It may also be that victims
of coercive controlling behaviors do not feel at a disadvantage in negotiating in mediation.
Future research needs to continue developing measures of coercive controlling behaviors. The
measure used in this study was a first effort at investigating how these particular behaviors
impacted mediation. There may be other important items that need to be included for both
mothers and fathers.
The reasons for mediators’ particular sensitivity to fathers’ victimization in relation to
providing special procedures is interesting. For decades victims’ advocates have argued that
mothers’ victimization does not make a significant impact on mediator decision-making;
however, this is not quite the entire story in this study. Mediators were sensitive to mothers’
victimization in identifying a case as having IPA but were more sensitive to fathers in provided
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 180 2007-WG-BX-0028
special accommodations during the process of mediation. It may be that mediators were
concerned that mothers’ IPA would lead to incidents of IPA immediately following the
mediation sessions. The most common accommodations provided were for the parents to leave
the mediation site at separate times or for one parent to be escorted to their car. With mothers
high on IPA the mediator may have felt the couple at higher risk of incidents of IPA on the way
out of the offices. It could also be that there was more evidence in the form of law enforcement
reports etc. to substantiate claims made by father. Attempting to rationally negotiate the future
structure of your family at the same time both parents are going through immense emotional
turmoil is an unnatural process and can cause anger and increase the changes of overt hostility.
We hypothesized that couples provided special procedures and with high levels of IPA
and coercive controlling behaviors would not reach an agreement in mediation. This hypothesis
was partially supported. Making an agreement in mediation was significantly and negatively
related to a mediator providing special procedures in mediation, significantly and negatively
influenced by fathers’ IPA and significantly and negatively influenced by mothers’ coercive
controlling behaviors. In other words, if a couple needs a mediator to provide special
accommodations to participate in mediation, there is less likelihood of the couple actually
reaching an agreement. There are some practitioners and scholars that vehemently argue that
couples benefit from mediation whether the reach an agreement or not. The argument is that the
mediation process itself can provide significant benefit for couples in learning how to negotiate
and focusing them on the needs of their children. There has been less discussion concerning how
to operationalize potential outcomes to capture and quantify of this benefit. Assessing the costs
and benefits of providing accommodations is essential.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 181 2007-WG-BX-0028
Beyond special accommodations, couples wherein fathers’ do not perpetrate IPA and
mothers do not perpetrate coercively controlling behaviors were more likely to reach an
agreement. Fathers who feel controlled and manipulated by mothers and mothers who are
victims of IPV likely do not trust the other parent to negotiate fair agreements in mediation or do
not feel safe in negotiating with the other parent. Rather, couples with these issues prefer to
return to court for awards of custody and/or parenting time.
Divorce Outcomes
The next four hypotheses concern the relationships between the measures of IPV,
coercive controlling behaviors, mediation outcomes and divorce outcomes (whether a case has a
legal finding of IPA, whether a divorce decree has restrictions placed on parenting or contact
between parents, whether there is an order of protection and whether the custody agreement
favors mothers). We hypothesized that fathers and mothers’ IPA and coercive controlling
behaviors, and child maltreatment would be influenced by the case having orders of protection.
This hypothesis was partially supported. Mediators’ determinations of IPA, mediators’ providing
special procedures for the mediation sessions, fathers’ IPA and mothers IPA were all positively
and significantly influenced by whether a case would have an order of protection. These results
are not surprising because our prior analysis of this data indicated that an important factor in
mediators making a determination that a case has IPA was evidence of outside agency
involvement with the family (Beck et al., 2010). What was interesting was the influence of
mothers’ coercive controlling behaviors negatively on whether a case would have an order of
protection.
This pattern of findings mirrors the relationship between mother’s coercive controlling
behaviors and reports of child maltreatment. It may be that if mothers perpetrate IPA the reaction
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 182 2007-WG-BX-0028
of the legal system is very strong and orders of protection are issued. However, if mothers
exhibit coercive controlling behaviors toward the fathers and it is successful in controlling the
relationship, the mothers then do not move on to committing IPA nor do they raise the concerns
of the legal system. These behaviors then actually protect themselves and their children.
Defining and investigating empirically coercive controlling behaviors in the context of
mediation is just beginning. It may be that other items need to be included to fully understand the
relationship between coercive controlling behaviors and orders of protection. It will be important
to continue this research and continue to develop measures that accurately capture the construct
for both mothers and fathers. It may be that the measure used in this study did not adequately
capture mother-perpetrated coercive controlling behaviors. Preliminary evidence from a small
scale study using the same measure of coercive controlling behaviors that was used in this study
indicated that at least for a sample of men in same sex relationships, the coercive controlling
behaviors take different forms (Raghavan & Beck, 2010).
We hypothesized that a legal finding of IPA in a case would be related to fathers and
mothers IPA and coercive controlling behaviors as well as a case having an order of protection
and child maltreatment. A potential safeguard for victims in a family with IPA is for the victim
to request the court to issue an order formalizing a finding of domestic violence in the family. In
the study jurisdiction, if such a finding is ordered, special conditions were placed on who could
be awarded custody and the restrictions that could be placed on the abuser. This hypothesis was
only partially supported. Results indicated that mothers’ IPA, reported child maltreatment and
orders of protection all significantly and positively influenced whether a case would have a legal
finding of domestic violence in the case. Fathers’ IPA did not have an influence.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 183 2007-WG-BX-0028
One potential reason for these findings is that there is a popular stereotype of mothers
who are involved in custody disputes. The stereotype is that mothers make false claims about
IPA or exaggerate IPA that is there, and make false claims about child maltreatment as a way to
gain unfair advantage in a divorce (Mastracci, 2007). Divorce attorneys representing mothers
may be hesitant to request a legal finding of domestic violence for fear of playing into this
stereotype and reducing their chances for a good outcome for clients who are mothers. Careful
analysis of court records is needed to determine which parent petitioned the court for the legal
finding.
Another potential reason is that it is mothers who are maltreating the children and
perpetrating IPA against fathers and it is these behaviors that drive a legal finding of domestic
violence in a family. Careful future research documenting the specific types of IPA in a
relationship, the specific types of maltreatment against the child, and the specific perpetrators of
both is needed to determine which of these scenarios are true.
Restrictions placed on abusers’ access to victims and interactions with violence-exposed
children are important safeguards to preventing future IPA. We hypothesized that restrictions
placed on parenting and contact between a couple would be determined by a court finding of
IPA, mothers and fathers’ IPA and coercive controlling behaviors, orders of protection and child
maltreatment. This hypothesis was partially supported. A court finding of IPA, marital stressors,
fathers’ IPA and orders of protection all significantly and positively influenced whether a decree
would include restrictions on contact and parenting. It is interesting that while mothers’ IPA was
related to a legal finding of IPA in a relationship, it was the fathers’ IPA that directly influenced
restrictions being placed on parenting agreements. It is important for future research to
investigate reasons attorneys seek legal findings of IPA. It may be that attorneys have
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 184 2007-WG-BX-0028
determined that fathers’ will have a better chance of obtaining custody if they request a legal
finding of IPA but not mothers.
Interestingly, marital stressors and orders of protection also influence restrictions in the
decree. Items included in the marital stressors constructs may drive this finding. Substance abuse
and mental health concerns, along with child maltreatment, are extremely important factors
judges consider in ordering restrictions on parenting.
We coded the custody agreement variable in such a way as to capture the range of legal
and physical custody options possible. We coded the custody agreement variable such that a
positive number would reflect an arrangement favoring the mother (sole legal and physical
custody to mother) and at the opposite end of the spectrum a negative relationship would indicate
the custody arrangement favoring the fathers. Custody decrees favoring mothers were
significantly and negatively influenced by reported child maltreatment and marital stressors, and
significantly and positively influenced by fathers’ IPA. This finding is interesting in that it
suggests that courts are extremely sensitive to mothers’ child maltreatment and possibly marital
stressors relating to the mother (mental health and/or substance abuse concerns). Because
mothers are overwhelmingly the primary residential parents in this study, courts are seriously
considering if mothers can be fit parents.
Post-Decree Outcomes
Continuing conflict between parents after a divorce has been finalized has well
established detrimental effects the parents and children as well as places burdens on the law
enforcement agencies and courts responsible for assisting parents in resolving conflicts. The
remaining three hypotheses concern the variables which influence continued involvement with
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 185 2007-WG-BX-0028
the legal system. Two sources of continued involvement were assessed: calls to area law
enforcement and court hearings/orders post-divorce.
The majority of parents returned to court after the divorce was finalized and nearly as
many had interactions with local law enforcement. Forecasting who might need post-decree
intervention is important. With this information mediators and judges could discuss how
procedures might be modified to help reduce the possibility of conflict post-divorce and thus the
need for continued court and law enforcement involvement. A measure of post-divorce conflict
is whether one or both of the parents made calls to the area law enforcement. We hypothesized
that calls to area law enforcement would be influenced by parental IPA and coercive controlling
behaviors, outcomes from mediation and the divorce process. Interestingly, the two important
predictors of post-decree calls to area law enforcement were the mediator identification of a case
as having IPA and reported child maltreatment. In our previous work we identified that a major
influence on mediator identifying a case as having IPA was outside agency involvement
including shelters, medical (hospital), police and criminal justice systems (order of protection)
(Beck et al., 2010). It is interesting that two of these variables collected at the time of mediation
(calls to police and child maltreatment) are then predictive of post-divorce calls to law
enforcement. First, the finding speaks to the importance of mediators including these questions
on any screening. And, second, it speaks to the importance of judges’ structuring agreements to
protect the safety of victims and maltreated children possibly through restrictions being placed
on abusing parents. Because parents are no longer living together, the opportunity for IPA is
reduced, although not eliminated due to shared custody or visitation agreements, but not so for
child maltreatment or for parents using the children as pawns, or as another way to control the
other parent (i.e., emotional abuse).
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 186 2007-WG-BX-0028
We hypothesized that the number of post-decree court hearings would influenced by
parental IPA, mediation and divorce outcomes. The number of court hearings post-divorce was
significantly and negatively influenced by whether a couple reached an agreement in mediation
and whether the court made a legal finding that the case involved IPA, and was significantly and
positively influenced by reported child maltreatment. These findings are exciting for mediation
programs in that if couples are able to reach a full agreement in mediation, there is a reduction in
post-decree litigation.
The remaining findings indicate that multiple problem families (IPA and child
maltreatment) continue to request court intervention to resolve issues in the family. Court-based
programs (such as parenting coordinator programs) have come into existence since the data was
collected for this study. These programs are specifically designed to assist multiple problem
families quickly resolve the majority of disputes without the need for court hearings (Beck,
Putterman, Sbarra & Mehl, 2008). Continued evaluation of these programs including direct
assessment of the children are needed to determine the usefulness of them in decreasing
relitigation.
The number of court orders was significantly and positively influenced by the number of
post-decree hearings and significantly and negatively influenced by court custody awards in
decrees favoring mothers. It appears that mothers who received unfavorable awards continued to
work toward more involvement with their children post-divorce.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 187 2007-WG-BX-0028
Recommendations for Policy, Practice and Research
Making decisions about who should be and who should not be in mediation is not as clear
as one would hope (Beck & Frost, 2006; Beck et al., 2010). Decisions concerning couples with
IPA in mediation are extremely complex issues that must include input from victims, mediators,
court administrators and policy-makers. From the results of this study, we are advocating:
(1) Provide essential training in the dynamics of IPA
Mediators, judges and court staff working directly with potential victims of IPA need to
be well trained in the dynamics of IPA and how it may manifest the context of a divorce.
(2) Use a structured, systematic approach to assessment of IPA in mediation
Using a structured, systematic, multi-method approach is likely to be considerably more
effective than using only a semi-structured clinical interview in reducing the rate of false
negatives in the IPA screening process. We are strongly recommending that mediators adopt a
standardized approach to administration of both semi-structured clinical interviews and
standardized questionnaires and ask all questions about IPA of all parents.
Mediation practitioners, scholars and violence experts also need to work together to
continue developing a more reliable, valid, approach to screening for all dimensions of IPA in
mediation including stalking. Because coercive controlling behaviors were important indicators
separate from the other dimensions of IPA, continued focus on developing adequate screening
measures for both mothers and fathers is needed.
(3) Automate screening measures
Development of an automated version of a comprehensive IPA screening measure that
quickly combines data from both members of a couple for mediator review and decision-making
is critical for an accurate assessment of IPA. Although assessment can be automated relatively
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 188 2007-WG-BX-0028
easily, the complex decisions mediators must make for a particular couple in determining
whether to move forward with mediation or not, or to provide an accommodation or not, requires
careful thought and analysis of the benefits and risks in consultation with the reported victim.
(4) Assess Mediation special procedures to accommodate victims
Prospective research using systematic and structured assessment of IPA which clearly
documents the reasons for dispositions of cases with documented IPA histories is needed so that
effective policies to screen clients out or provide accommodations can be developed. An
accommodation which might be helpful to empirically investigate for reported victims of IPA
wishing to continue in mediation is for mediators to pre-arrange with the victim a sign (e.g., hand
signal or word) that the victim can use in the event the she or he feels unsafe or afraid. If used the
mediator will know to stop and possibly caucus with the parents individually to insure the victim
wishes to continue. Some abusers convey threats only the victim will understand and it is
important for a victim to be able to signal a mediator when this occurs. Perhaps researchers
working with practitioners can design a study where mediators are included in developing a
standardized set of accommodation practices along with hypotheses that identify which
accommodations are best implemented in cases involving different types of IPA. Carefully
defined and collected outcomes can then measure the effectiveness of the accommodations
provided, thereby empirically investigating if the process can be improved and appropriate
accommodations instituted across cases.
Mediation program personnel, lawyers and advocates working with potential victims
should strongly encourage clients who are concerned about IPA or who want specific
accommodations in the mediation process to call and request such accommodations or express
concerns about mediating.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 189 2007-WG-BX-0028
(5) Acknowledge the limitations of the mediation process
Mediators are not judges and cannot order that certain conditions to be included in
agreements (such as protections for children or restrictions on contact between parents) or that
the conditions be followed even if such requirements would be in the best interests of the
children or the adults. In the majority of jurisdictions across the United States, mediators instead
facilitate negotiations between parents and each parent has a right to decide what they will and
will not agree to. If both parents do not agree to a specific condition, it will not be included in
any resulting agreement. If parents do not agree on any issues then no parenting agreement will
result. Holding mediators and mediation programs responsible for parenting agreements, freely
negotiated between the parties, that do not include safety precautions flies in the face of what
mediation in this and many other jurisdictions are designed to accomplish. If a mediator believes
that one of the parties cannot adequately represent their interests or the agreement negotiated
between the parents will not be in the best interests of the child[ren] is terminate mediation, the
resource the mediator has is to terminate mediation without an agreement. This decision is
difficult, complex and must be made carefully and in consultation with the party having trouble
representing her or his interests.
(6) Investigate hybrid mediation/arbitration mediation interventions
If policy-makers wish to ensure that couples with IPA negotiate mediation agreements
which include safety restrictions, they must consider empirically investigating a separate process
for these couples. The hybrid mediation/arbitration intervention may be particularly important to
empirically investigate with pro se couples with high IPA. Currently, victims without legal
representation who are unable to negotiate safe parenting plans in mediation currently must
return to court and present her or his case to the judge.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 190 2007-WG-BX-0028
(7) Encourage legal findings of “domestic violence” in appropriate cases
The results of this study indicate that a parent with a legal finding of domestic violence in
their case is more likely to have safety restrictions on contact between abuser and victim and on
parenting (e.g., orders of protection; orders restricting method of contact between parents;
neutral exchanges of the children at a specific public location, a family member’s home, or a
judicially approved service provider) in the divorce decrees. Lawyers, advocates for victims and
mediators should carefully consider, in consultation with victims, recommending that victims
obtain court orders with legal findings of domestic violence.
(8) Assess outcomes for children directly and prospectively
Children are often caught in the cross-fire of parental IPA and so the co-occurrence of
IPA with child maltreatment is high. Broader assessment and evaluation of child maltreatment
and IPA (looking beyond physical forms of both) is essential to truly understand if mediation or
any other court program is benefitting children. Psychological abuse can be just as devastating as
other forms of abuse (Coker, Smith, Bethea, King, & McKeown, 2000; O’Leary, 1999; Stark,
2007; Theran, Sullivan, Bogat, & Stewart, 2006). Because virtually all cases in the study
involved some form of psychological abuse, mediators should pay particular attention to
assisting the parents to understand and remediate the impact of such abuse on the members of the
family, particularly the children.
(9) Assess the effectiveness of court-based programs for pre- and post-divorce
“frequent flyers”
A small group of families were found to require a tremendous number of court hearings
post-divorce. As a result, parenting coordinator programs are developing rapidly across the
United States with very little empirical basis of support. Courts need to empirically evaluate the
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 191 2007-WG-BX-0028
effectiveness of court-based parenting coordinator and case management processes to determine
whether these programs are having the desired benefits for families.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 192 2007-WG-BX-0028
Limitations of the research.
Archival data from court records and area law enforcement records are an unobtrusive
and efficient means of gathering important data. Data collection in this manner is not traumatic
for the subjects and it does not disrupt the ongoing court and law enforcement processes.
Archival data is also relatively cheaper, quicker and easier to assess large numbers of cases. And,
in some cases the information cannot be assessed in any way (Jewkes, 2008). There are,
however, limitations to using official records to collect data. The data available is limited to what
the records contain. Records are often incomplete, files can be misfiled and documents intended
for the files can be misfiled. The quality of the records is not under the control of the researcher
(Jewkes, 2008). Access to the records can be limited and refusal to allow access or provide
certain portions of the data can occur.
More specific to this study, these data are from only divorcing couples who had disputes
about custody and/or parenting time of their children from one jurisdiction and one court-
sponsored, free mediation program. There is no way to empirically identify which couples chose
this program over private, fee-for-service mediator or mediation program. It is likely that those
couples with the most disposable income did so. Because this is a study specifically of divorcing
couples with children referred to mediation and not all divorcing couples, couples without
children and/or those couples with children but who were able to work out agreements
concerning custody and/or parenting time were also not included in this study. Because of the
specifics of this sample, there are limits to the generalizability of the findings.
It is important to note several issues relevant to the screening procedure and the use of the
screening forms by the mediators. As noted above, these forms were a portion of the mediation
program’s routine screening procedures and were not designed specifically for research
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 193 2007-WG-BX-0028
purposes. These forms were provided as a tool for the mediators to use in recording factors
relevant to their decisions about whether to and how to continue mediation. Therefore, these data
represent the individual determinations by mediators as to what was important to them to note.
In addition, the behaviorally-specific questionnaire included in the file (the RBRS)
looked at IPA in the past 12 months only. Thus, it did not capture the full extent of IPA that
existed in the entire relationship nor did it document specifically current IPA ongoing in the
relationship. The RBRS also did not assess either who the primary or predominant aggressor was
or the relationship context in which any of these behaviors occurred. Whether the acts were self-
defense or abusive/violent acts cannot be distinguished using the RBRS. Future research needs to
include structured, standardized assessment measures of IPA in addition to in-depth interviews
with the clients to gain a full understanding of the important areas and the specific contexts and
impact of the violence in the family. It is extremely important for future research that a clear
picture of the history of the IPA in the relationship can be understood, the victim and abuser can
be correctly identified so that the court system can respond appropriately to protect the victim
and the violence-exposed children.
The data in this study concerning accommodations is likely a serious underestimate of the
actual number of accommodations provided. Some mediators at the study site did not document
these accommodations because they believed they were mediation techniques as opposed to
accommodations. There has been two decades of scholarly discussion of the possibility of special
procedures used in mediation to accommodate couples with IPA (Duryee, 1995; Erickson &
McKnight, 1990; Girdner, 1990; Landau, 1995; Magnana & Taylor, 1993; Marthaler, 1989;
Newmark et al., 1995; Vincent, 1995; Yellott, 1990), but there has yet been no systematic
empirical research investigating what accommodations are provided and when they are provided
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 194 2007-WG-BX-0028
and why they are provided. This study represents one of the first attempts to document the use of
special accommodations for couples with IPA. This study was a naturalistic evaluation of one
county’s practices. Future prospective research is needed first which carefully defines special
procedures/accommodations, assigns couples to specific accommodations based on hypothesized
needs, then documents defined outcomes associated with the couples in the different
accommodation groups (Ellis, 2008). This research would be helpful in understanding how
accommodations affect case processing and case outcomes.
In assessing independent evidence of IPA allegations, a strong word of caution is
necessary about the law enforcement data in this study. It is extremely important to note that the
time period from two years prior to filing a divorce decree to January 31, 2007 could represent a
serious underestimate of the total calls to all law enforcement for the couple. These data are only
those calls that were made to the two local law enforcement agencies at the study site. Calls
could have been placed to agencies outside the local area and would not be reported in these
data. Families often relocate during and after a divorce. Those that relocated temporarily at any
point or relocated permanently and called law enforcement assistance in the new location would
not be represented in these data. In addition, due to limited law enforcement agency resources to
collect data for this study, these data represent only calls where both parents are involved. If the
mother and father were not suspect and victim in a call, the call was not included in the data.
Thus, these data likely underestimate law enforcement contact in this sample.
Although limitations exist, these data are unprecedented in the field of divorce mediation
research. To date, virtually all of the published samples used to study mediation included small
samples, limited data, or low response rates (Beck & Sales, 2001). In this study, the sample
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 195 2007-WG-BX-0028
included all parents (N=965 couples) mandated to attend mediation and did so for a first time
mediation for a pending divorce through a court-sponsored, free mediation program.
Implications for Future Research.
This study provides valuation information that can improve future research, practice and
evaluation of IPA screening practices in mediation. The design of the study can serve as a model
that can be emulated in other jurisdictions. The valuable methodological innovations included in
this study are numerous and potentially include: cross checking with criminal justice data; cross
checking with civil court data (orders of protection); cross checking with service agency data
(Child Protective Service and area shelters); efforts to empirically assess actual practices in
mediation in IPA cases; collection of empirical data on the outcomes of mediation and court
cases where custody is disputed and IPA is reported; use of the mediation context as an
opportunity to systematically screen for IPA and as a resource for research in a high-risk yet non-
shelter sample.
Future studies could easily overcome the limitations of this current study
(decontextualized counts of behaviorally-specific acts) by including questionnaire items and/or
in-depth interviews concerning the history of the relationship and context surrounding the IPA
drawn from the state of the art research on lethality assessment and measurement of IPA, much
of which has been supported by NIJ. Future studies could also assess more systematically shelter
and other protective service use to screen for high risk cases and cases likely to result in ongoing
conflict and risk of harm. Future studies could also more carefully assess short- and long-term
well-being and safety outcomes for children and parents in addition to outcomes of cases in
courts and law enforcement records.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 196 2007-WG-BX-0028
References
Adam, K., S. (2010). Consolidating cases in non-unified family courts. Juvenile and Family
Justice Today, Fall 2010, 17-21.
Adams, M., & Coltrane, S. (2007). Framing divorce reform: Media, morality, and the politics of
family. Family Process, 46(1), 17-34.
Administrative Office of the Courts. (2002). Domestic violence in court-based child custody
mediation cases in California. Center for Families and Children of the Courts: Judicial
Council of California.
Administrative Office of the Courts. (2010). Snapshot study 2008: Summary findings. San
Francisco, California: Administrative Office of the Courts, Judicial Council of California.
Ahrons, C. R. (1994). The good divorce. New York: Harper Perennial.
Alaska Judicial Council. (1992). Alaska child visitation mediation pilot project: Report to the
legislature. Anchorage, AK: Alaska Judicial Council.
American Psychological Association, W., DC. (1996). Violence and the family: Report of the
American Psychological Association Presidential Task Force on Violence and the Family.
American Psychological Association, 750 First Street, NE, Washington, DC 20002-4262.
Ansara, D. L., & Hindin, M. J. (2010). Exploring gender differences in patterns of intimate
partner violence in Canada: A latent class approach. Journal of Epidemiology & Community
Health, 64, 849-854.
Appel, A. E., & Holden, G. W. (1998). Parenting: Correlates and outcomes - the co-occurrence
of spouse and physical child abuse: A review and appraisal. Journal of Family Psychology:
JFP: Journal of the Division of Family Psychology of the American Psychological
Association (Division 43., 12(4), 578.
Applewhite, A. (2003). In L. Ganong (Ed.), Divorce reform won't lower the divorce rate. Los
Angeles, CA US: Roxbury Publishing Co.
Archer J. (2000a). Sex differences in aggression between heterosexual partners: A meta-analytic
review. Psychological Bulletin, 126(5), 651-80.
Archer, J. (2000b). Articles - sex differences in physical aggression to partners: A reply to Frieze
(2000), O'Leary (2000), and White, Smith, Koss, and Figueredo (2000). Psychological
Bulletin, 126(5), 697.
Arendell, T. (1995). Fathers and divorce. Thousand Oaks, CA: Sage.
Arizona Revised Statutes. Custody; Best Interests of Child, Ariz. Stat. Ann. § 25-403, (2007).
Attala, J. M., Hudson, W. W., & McSweeney, M. (1994). A partial validation of two short-form
partner abuse scales. Women & Health, 21(2/3), 125.
Bala, N. (2008). An historical perspective on family violence and child abuse: Comment on
Moloney et al, 'allegations of family violence,' 12 June 2007. Journal of Family Studies,
14(2-3), 271-278.
Baskerville, S. (2007). Taken into custody: The war against fathers, marriage, and the family.
Nashville, TN: Cumberland House.
Beck, C. J., & Sales, B. D. (2001). Family mediation : Facts, myths, and future prospects.
Washington, DC: American Psychological Association.
Beck, C., Menke, J. M., Brewster, K. O., & Figueredo, A. J. (2009). Validation of a measure of
intimate partner abuse with couples participating in divorce mediation. Journal of Divorce
& Remarriage, 50(5), 295-308.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 197 2007-WG-BX-0028
Beck, C. J. A., Walsh, M. E., Mechanic, M. B., & Taylor, C. S. (2010). Mediator assessment,
documentation, and disposition of child custody cases involving intimate partner abuse: A
naturalistic evaluation of one county’s practices. Law and Human Behavior, 34(3), 227-
240.
Bentler, P. M. (1989). EQS structural equations program manual. Los Angeles, Calif.: BMDP
Statistical Software.
Bergen, R. K. (2005). Studying wife rape: Reflections on the past, present and future. Violence
Against Women, 10, 1407-1416.
Bileski, M. (2008). Arizona Homicide Study. Phoenix, AZ: Statistical Analysis Center, Arizona
Criminal Justice Commission. Retrieved 7/12/2011 from
http://www.jrsa.org/events/conference/presentations-08/Matthew_Bileski.pdf.
Bollen, K., & Lennox, R. (1991). Conventional wisdom on measurement: A structural equation
perspective. Psychological Bulletin, 110(2), 305-314.
Bourassa, C., Lavergne, C., Damant, D., Lessard, G., & Turcotte, P. (2006). Awareness and
detection of the co-occurrence of interparental violence and child abuse: Child welfare
worker’s perspective. Children and Youth Services Review, 28(11), 1312-1328.
Braver, S. L., Shapiro, J. R., & Goodman, M. R. (2006). Consequences of divorce for parents. In
Mark A. Fine and John H. Harvey (Eds.). Handbook of Divorce and Relationship
Dissolution, 313-338. Mahwah, NJ: Lawrence Erlbaum Associates.
Browne, A. (1987). When battered women kill. New York: Free Press.
Bryan, B. (1992). Killing us softly: Divorce mediation and the politics of power. Buffalo Law
Review , 40, 441-523.
Buckley, H., Whelan, S., & Carr, N. (2011). "Like waking up in a Franz Kafka novel:' service
users' experiences of the child protection system when domestic violence and acrimonious
separations are involved. Children and Youth Services Review, 33, 126-133.
Campbell, J. C., Webster, D., Koziol-McLain, J., Block, C., Campbell, D., Curry, M. A.,…
Laughon, K. (2003). Risk factors for femicide in abusive relationships: Results from a
multicase case control study. American Journal of Public Health, 93(7), 1089-1097.
Campbell, L. E., & Johnston, J. R. (1986). Impasse-directed mediation with high conflict
families in custody disputes. Behavioral Sciences & the Law, 4(2), 217-241.
Casanueva, C., Martin, S. L., & Runyan, D. (2009). Repeated reports for child maltreatment
among intimate partner violence victims: Findings from the National Survey of Children
and Adolescent Well-Being. Child Abuse & Neglect, 33(2009), 84-93.
Cauble, A. E., Thoennes, N., Pearson, J., & Appleford, R. (1985). A case study: Custody
resolution counseling in Hennepin County, Minnesota. Conciliation Courts Review, 23(2),
27-35.
Cobb, S. (1997). The domestication of violence in mediation. Law & Society Review, 31(3), 397-
440.
Cohen, J. (1983). The cost of dichotimization. Applied Psychological Measurement, 7(3), 249-
253.
Cohen , S. (1982). The diversion study: A preliminary report. Unpublished manuscript.
Coker, A. L., Smith, P. H., Bethea, L., King, M. R.,& McKeown, R. E. (2000). Physical health
consequences of physical and psychological intimate partner violence. Archives of Family
Medicine, 9, 1-7.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 198 2007-WG-BX-0028
Coogler, O. J. (1978). Structured mediation in divorce settlement: A handbook for marital
mediators. Lexington, MA: D.C. Health.
Coulson, R. (1969). Family arbitration-an exercise in sensitivity. Family Law Quarterly, 3, 22-
30.
Crowley, J. E. (2008). Defiant dads: Fathers’ rights activists in America. Ithaca, NY: Cornell
University Press.
Dalton, D., Drozd, L. M., & Wong, F. Q. F (2006). Navigating Custody & Visitation Evaluations
in Cases with Domestic Violence: A Judge’s Guide. State Justice Institute and the National
council of Juvenile and Family Court Judges. Retrieved July 27, 2011, from
http://www.ncjfcj.org/images/stories/dept/fvd/pdf/navigating_cust.pdf.
Danzig, R. (1973). Toward the creation of a complementary, decentralized system of criminal
justice. Stanford Law Review, 26, 1-54.
Depner, C. E., Cannata, K., & Ricci, I. (1995). Report 4: Mediated agreements on child custody
and visitation: 1991 California Family Court Services Snapshot Study. Family and
Conciliation Courts Review, 33(1), 87-109.
Demetriou, A., Christou, C., Spanoudis, G., Platsidou, M., Fischer, K. W., & Dawson, T. L.
(2002). The development of mental processing: Efficiency, working memory, and thinking.
Monographs – Society for Research in Child Development, 67(268), ALL.
Diamantopoulos, A., & Winklhofer, H. M. (2001). Index construction with formative indicators:
An alternative to scale development. Journal of Marketing Research, 38(2), 269-277.
Dingwall, R. (2010). Divorce mediation: Should we change our mind? Journal of Social Welfare
& Family Law, 32(2), 107-117.
Douglas, H., & Walsh, T. (2010). Mothers, domestic violence, and child protection: Toward
collaboration and engagement. Violence Against Women, 16(5), 537-542.
Duryee, M. A. (1995). Guidelines for family court services intervention when there are
allegations of domestic violence. Family and Conciliation Courts Review, 33, 79-86.
Dunford-Jackson, B. L. (2004). The role of family courts in domestic violence: The U.S.
experience. In P. Jaffe G., L. L. Baker & A. J. Cunningham (Eds.), Protecting children from
domestic violence: Strategies for community intervention (pp. 8-28). New York: Guilford.
Dutton, D. (2005a). Domestic abuse assessment in child custody disputes beware the domestic
violence research paradigm. Journal of Child Custody, 2(4), 23-42.
Dutton, D. (2005b). On comparing apples with apples deemed nonexistent A reply to Johnson.
Journal of Child Custody, 2(4), 53-63.
Dutton, D. (2006). A briefer reply to Johnson re-affirming the necessity of a gender-neutral
approach to custody evaluations. Journal of Child Custody, 3(1), 67-69.
Dutton, D. G. (2007). The abusive personality: Violence and control in intimate relationships
(2nd ed.). New York, NY US: Guilford Press.
Dutton D. G., Hamel J., & Aaronson J. (2010). The gender paradigm in family court processes:
Re-balancing the scales of justice from biased social science. Journal of Child Custody,
7(1), 1-31.
Dutton, D. G., & Nicholls, T. L. (2005). The gender paradigm in domestic violence research and
theory: Part 1 – The conflict of theory and data. Aggression and Violent Behavior, 10, 680-
714.
Dutton, M., & Goodman, L. (2005). Coercion in intimate partner violence: Toward a new
conceptualization. Sex Roles, 52(11-12), 11-12.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 199 2007-WG-BX-0028
Edwards, L., Baron, S., & Ferrick, G. (2008). A comment on William J. Howe and Hugh
McIssaac's article "finding the balance" published in the January 2008 issue of family court
review. Family Court Review, 46(4), 586-591.
Ellis, D. (2008). Divorce and family court: What can be done about domestic violence? Family
Court Review, 46(3), 531-536.
Ellis, D., & Stuckless, N. (2006). Domestic violence, DOVE, and divorce mediation. Family
Court Review, 44(4), 658-671.
Emery, R. E. (1994). Renegotiating family relationships: Divorce, child custody, and mediation.
New York: Guildord Press.
Erickson, S. K., & McKnight, M. S. (1990). Mediating spousal abuse divorces. Conflict
Resolution Quarterly, 7(4), 377-388.
Felson, R. B., & Cares, A. C. (2005). Gender and the seriousness of assaults on intimate partners
and other victims. Journal of Marriage and Family, 67(5), 1182-1195.
Felson, R. B., & Outlaw, M. C. (2007). The control motive and marital behavior. Violence and
Victims, 22, 387-407.
Finkel, E. J., DeWall, C. N., Slotter, E. B., Oaten, M., & Foshee, V. A. (2009). Self-regulatory
failure and intimate partner violence perpetration. Journal of Personality and Social
Psychology,l 97(3), 483-499.
Ferrick, G., Baron, S., & Edwards, L. (2008). Surreply. Family Court Review, 46(4), 595-597.
Figueredo, A. J., & McCloskey, L. A. (1993). Sex, money, and paternity: The evolutionary
psychology of domestic violence. Ethology and Sociobiology, 14(6), 353.
Figueredo, A. J., McKnight, P. E., McKnight, K. M., & Sidani, S. (2000). Multivariate modeling
of missing data within and across assessment waves. Addiction, 95(11s3), 361-380.
Figueredo, A. J., Corral-Verdugo, V., Frías-Armenta, M., Bachar, K. J., White, J., McNeill, P.
L., Kirsner, B. R., Castell-Ruiz, I. (2001). Blood, solidarity, status, and honor: The sexual
balance of power and spousal abuse in Sonora, Mexico. Evolution and Human Behavior,
22(5), 295-328.
Figueredo, A. J., & Gorsuch, R. L. (2007). Assortative mating in the jewel wasp. 2: Sequential
canonical analysis as an exploratory form of path analysis. Journal of the Arizona-Nevada
Academy of Science, 39(2), 59-64.
Figueredo, A. J., Montero-Rojas, E., Frías-Armenta, M., & Corral-Verdugo, V. (2009).
Individual differences and social contexts: The absence of family deterrence of spousal
abuse in San José, Costa Rica. Journal of Social, Evolutionary & Cultural Psychology, 3(1),
29-48.
Fischer, K., Vidmar, N., & Ellis, R. (1993a). The culture of battering and the role of mediation in
domestic violence cases. Southern Methodist University Law Review, 34, 161-182.
Fleury, R. E., Sullivan, C. M., & Bybee, D. I. (2000). When ending the relationship does not end
the violence: Women's experiences of violence by former partners. Violence Against
Women, 6(12), 1363-1383.
Folberg, J. (1983). A mediation overview: History and dimensions of practice. Mediation
Quarterly, 1, 3-13.
Folberg, J. (1982). Divorce mediation—A workable alternative. In J. Davidson, L. Ray, & R.
Horowitz (Eds.), Alternative means of family dispute resolution, 11-41. Washington, DC:
American Bar Association.
Folberg, J., & Milne, A. (1988). Divorce mediation: Theory and practice. New York: Guilford.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 200 2007-WG-BX-0028
Folberg, J., & Taylor, A. (1984). Mediation : A comprehensive guide to resolving conflicts
without litigation. San Francisco: Jossey-Bass.
Frank, J. (1969). American law: The case for radical reform. New York: Macmillan.
Fuller, L. L. (1970). Mediation-its forms and functions. Southern California Law Review, 40,
305-339.
Furstenberg, F. F., Jr. (1994). History and current status of divorce in the United States. The
Future of Children, 4(1), 29-43.
Galanter, M. (1983). Reading the landscape of disputes: What we know and don't know (and
think we know) about our allegedly contentious and litigious society. UCLA Law Review,
31, 4-71.
Gerber, R. J. (1990). Recommendation on domestic relations reform. Arizona Law Review, 32, 9-
19.
Girdner, L. K. (1990). Mediation triage: Screening for spouse abuse in divorce mediation.
Conflict Resolution Quarterly, 7(4), 365-376.
Goodmark, L. (2010). Mothers, domestic violence, and child protection: An American legal
perspective. Violence Against Women, 16(5), 524-529.
Gorsuch, R. L., & Figueredo, A. J. (1991). Sequential canonical analysis as an exploratory form
of path analysis. Paper presented at the American Evaluation Association Conference,
Chicago, Illinois.
Graham-Kevan, N., & Archer, J. (2003). Physical aggression and control in heterosexual
relationships: The effect of sampling. Violence and Victims, 18(2), 181.
Graham-Kevan, N., & Archer, J. (2008). Does controlling behavior predict physical aggression
and violence to partners? Journal of Family Violence, 23(7), 539-548.
Grillo, T. (1991). The mediation alternative: Process dangers for women. Yale Law Journal, 100,
1545-1610.
Haddix, A. (1996). Unseen victims: Acknowledging the effects of domestic violence on children
through statutory termination of parental rights. California Law Review, 84(3), 757.
Hagenaars, J. A., & McCutcheon, A. L. (2002). Applied latent class analysis. Cambridge; New
York: Cambridge University Press.
Hamby, S. L. (2005). Measuring gender differences in partner violence: Implications from
research on other forms of violent and socially undesirable behavior. Sex Roles, 52(11-12),
725-742.
Hamby S. (2009). The gender debate about intimate partner violence: Solutions and dead ends.
Psychological Trauma: Theory, Research, Practice, and Policy, 1(1), 24-34.
Hardesty, J. L. (2002). Separation assault in the context of post-divorce parenting: An integrative
review of the literature. Violence Against Women, 8(5), 597-625.
Hardesty, J. L., & Chung, G. H. (2006). Intimate partner violence, parental divorce, and child
custody: Directions for intervention and future research. Family Relations, 14, 353-379.
Hardesty, J., L., & Ganong, L., H. (2006). How women make custody decisions and mange co-
parenting with abusive former husbands. Journal of Social and Personal Relationships,
23(4), 543-563.
Hardesty J. L., Khaw L., Chung G.H., & Martin J.M. (2008). Coparenting relationships after
divorce: Variations by type of marital violence and fathers' role differentiation. Family
Relations, 57(4), 479-491.
Hart, B. J. (1990). Gentle jeopardy: The further endangerment of battered women and children in
custody mediation. Mediation Quarterly, 8, 317-330.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 201 2007-WG-BX-0028
Haynes, J. M. (1981). Divorce mediation: A practical guide for therapists and counselors. New
York: Springer.
Hazen, A. L., Connelly, C. d., Edleson, J. L., Kelleher, K. J., Landsverk, J. A., Coben, J. H.,
Barth, R. P., McGeehan, J., Rolls, J. A., & Nuszkowski, M. A. (2006). Assessment of
intimate partner violence by child welfare services. Children and Youth Services Review,
29(4), 490-500.
Hetherington, E. M., & Kelly, J. (2002). For better or worse: Divorce reconsidered. New York:
Norton.
Herman, M. (2006). Self-representation: Pro se statistics. Retrieved 2/11/2011, from
http://www.ncsconline.org/wc/publications/memos/prosestatsmemo.htm.
Holtzworth-Munroe A., Meehan J. C., Herron K., Rehman U., & Stuart G. L. (2000). Testing the
Holtzworth-Munroe and Stuart (1994) batterer typology. Journal of Consulting and Clinical
Psychology, 68(6), 1000-19.
Holtzworth-Munroe, A. (2005). Female perpetration of physical aggression against an intimate
partner: A controversial new topic of study. Violence and Victims, 20(2), 251-259.
Hotton, T. (2001). Spousal violence after marital separation. Juristat: Canadian Centre for
Justice Statistics, 21(7), 1-19.
Howe, W., J., & McIsacc, H. (2008). Finding the balance: Ethical challenges and best practices
for lawyers representing parents when the interests of children are at stake. Family Court
Review, 46(1), 78-90.
Irving, H. H., & Benjamin, M. (1983). Outcome effectiveness of conciliation counseling: An
empirical study. Conciliation Courts Review, 21(2), 61-70.
Jacobson, N. S., & Gottman, J. (1998). When men batter women. New York: Simon & Schuster.
Jewkes, R. (2008). Research to understand integrated medico-legal responses. In C. Ajema and
M. Valai (Compilers), Workshop Report on Medico-Legal Responses to Sexual Violence,
June 2-6, Nairobi, Kenya. Retrieved 4/20/2011 from
http://www.svri.org/2008MLworkshop.pdf.
Johnson, M. P. (1995). Patriarchal terrorism and common couple violence: Two forms of
violence against women. Journal of Marriage and the Family, 57(2), 283-294.
Johnson, M. P. (2005a). Apples and oranges in child custody disputes: Intimate terrorism vs.
situational couple violence. Journal of Child Custody, 2, 43-52.
Johnson, M. P. (2005b). A brief reply to Dutton. Journal of Child Custody, 2(4), 65-68.
Johnson, M. P. (2006). Conflict and control: Gender symmetry and asymmetry in domestic
violence. Violence Against Women, 12(11), 1003-1018.
Johnson, M. P. (2008). A typology of domestic violence: Intimate terrorism, violent resistance,
and situational couple violence. Boston: Northeastern University Press.
Johnson, M. P., & Leone, J. M. (2005). The differential effects of intimate terrorism and
situational couple violence. Journal of Family Issues, 26(3), 322-349.
Johnson, N. E., Saccuzzo, D. P., & Koen, W. J. (2005). Child custody mediation in cases of
domestic violence: Empirical evidence of a failure to protect. Violence Against Women,
11(8), 1022-1053.
Johnston J. R., & Campbell L. E. (1993). A clinical typology of interparental violence in
disputed-custody divorces. The American Journal of Orthopsychiatry, 63(2), 190-9.
Johnston, J. R., Lee, S., Olesen, N. W., & Walters, M. G. (2005). Allegations and substantiations
of abuse in custody-disputing families. Family Court Review, 43(2), 283-294.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 202 2007-WG-BX-0028
Jouriles, E. N., McDonald, R., Slep, A. M. S., Heyman, R. E., & Garrido, E. (2008). Child abuse
in the context of domestic violence: Prevalence, explanations, and practice implications.
Violence and Victims, 23(2), 221-235.
Katz, S. N. (1994). Historical perspective and current trends in the legal process of divorce. The
Future of Children, 4(1), 44-62.
Kelly, J. B. (1995). Power imbalance in divorce and interpersonal mediation: Assessment and
intervention. Conflict Resolution Quarterly, 13(2), 85-98,
Kelly, J. B., & Johnson, M. P. (2008). Differentiation among types of intimate partner violence:
Research update and implications for interventions. Family Court Review, 46, 476-499.
Kernic, M. A., Monary-Ernsdorff, D., Koepsell, J. K., & Holt, V. L. (2005). Children in the
crossfire: Child custody determinations among couples with a history of intimate partner
violence. Violence Against Women, 11(8), 991-1021.
Keys Young (Firm), & Australia. Office of Legal Aid and Family Services. (1996). Research
evaluation of family mediation practice and the issue of violence. Canberra: Attorney-
General's Department.
Knickerbocker, L., Heyman, R. E., Slep, A. M. S., Jouriles, E. N., & McDonald, R. (2007). Co-
occurrence of child and partner maltreatment definitions, prevalence, theory, and
implications for assessment. European Psychologist, 12(1), 36-44.
Kwong, M. J., Bartholomew, K., & Dutton, D. G. (1999). Gender differences in patterns of
relationship violence in Alberta. Canadian Journal of Behavioural Science., 31(3), 150-
160.
Landau, B. (1995). The Toronto Forum on Women Abuse: The process and the outcome. Family
and Conciliation Courts Review, 33, 63-78.
Little, T. D., Cunningham, W. A., Shahar, G., & Widaman, K. F. (2002). To parcel or not to
parcel: Exploring the question, weighing the merits. Structural Equation Modeling, 9(2),
151-173.
Local Rules of Practice: Arizona Superior Court in Pima County, Article 8. Family Law Cases,
Section 8.7: Conciliation Court Services: Mediation of Custody and Parenting Time
Disputes, (1999).
Logan, TK, Walker, R., Hoyt, W., & Faragher, T (2009). The Kentucky civil protective order
study: A rural and urban multiple perspective study of protective order violation
consequences, responses, and costs. NIJ Report #2005-WG-BX-008.
Magana, H. A., & Taylor, N. (1993). 1992 Meyer Elkin Essay Contest Winner: Child custody
mediation and spouse abuse: A descriptive study of a protocol. Family and Conciliation
Courts Review, 31(1), 50-64.
Magen, R. H. (1999). In the best interests of battered women: Reconceptualizing allegations of
failure to protect. Child Maltreatment, 4(2), 127-135.
Mahoney, A., Donnelly, W., Boxer, P., & Lewis, T. (2003). Marital and severe parent-to-
adolescent physical aggression in clinic-referred families: Mother and adolescent reports on
co-occcurrence and links to child behavior problems. Journal of Family Psychology, 17(1-
3).
Mahoney, M. R. (1991). Legal images of battered women: Redefining the issue of separation.
Michigan Law Review, 90, 1-94.3-19.
Malik, N., Silverman, J., Wang, K., & Janczewski, C. (2008). Domestic violence and
dependency courts. Journal of Interpersonal Violence, 23(7), 956-980.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 203 2007-WG-BX-0028
Marthaler, D. (1989). Successful mediation with abusive couples. Conflict Resolution Quarterly,
Spring(23), 53-66.
Mastracci, M. A. (2007). Domestic violence laws often abused to gain child custody. Retrieved
8/15/2008 from http://www.divorcewithoutdishonor.com/category/abuse-issues/.
Mather, L. (2003). Changing patterns of legal representation in divorce: From lawyers to pro se.
Journal of Law and Society, 30(1), 137-155.
Mathis, R. D., & Tanner, Z. (1998). Effects of unscreened spouse violence on mediated
agreements. The American Journal of Family Therapy, 26(3), 251.
Maxwell, J. P. (1999). Mandatory mediation of custody in the face of domestic violence:
Suggestions for courts and mediators. Family & Conciliation Courts Review, 37(3), 335-
355.
Mayr, E. (1976). Evolution and the Diversity of Life. Cambridge: Belknap Press of Harvard
University Press.
McCloskey, L. A., Figueredo, A. J., & Koss, M. P. (1995). The effects of systemic family
violence on children's mental health. Child Development, 66(5), 1239-1261.
McCutcheon, A. L., (1987). Latent class analysis. Newbury Park: Sage Publications.
McGillicuddy, N. B., Welton, G. L. & Pruitt, D. G. (1987). Third party intervention: A field
experiment comparing three different models. Journal of Personality and Social
Psychology, 53, 104-112.
McKnight, P. E., McKnight, K. M., Sidani, S., & Figueredo, A. J. (2007). Missing data: A gentle
introduction. New York, NY US: Guilford Press.
Mechanic, M. B., Weaver, T. L., & Resick, P. A. (2002). Intimate partner violence and stalking
behavior: Exploration of patterns and correlates in a sample of acutely battered women. In
R. D. Maiuro (Ed.), Stalking: Perspectives on victims and perpetrators. (pp. 62-88). New
York, NY US: Springer Publishing Co.
Meehl, P. E., (1965). Detecting latent clinical taxa by fallible quantitative indicators lacking on
accepted criterion (Report No. PR-65-2). Minneapolis: University of Minnesota Research
Laboratories of the Department of Psychiatry.
Michigan Supreme Court. (2006). Domestic violence and child abuse/neglect screening for
domestic relations mediation: Model screening protocol. Retrieved 7/12/2011 from
http://courts.michigan.gov/scao/resources/standards/odr/dvprotocol.pdf
Miller, S. L., & Smolter, N. L. (2011). “Paper abuse”: When all else fails, batterers use
procedural stalking. Violence Against Women 17(5), 637-650.
Mnookin, R., & Kornhauser, L. (1979). Bargaining in the shadow of the law: The case of
divorce. Yale Law Review, 88, 950-997.
Moloney, L., & Australian Institute of Family Studies. (2007). Allegations of family violence and
child abuse in family law children's proceedings: A pre-reform exploratory study.
Melbourne, Vic: Australian Institute of Family Studies.
Moloney, L. (2008). Violence allegations in parenting disputes: Reflections on court-based
decision making before and after the 2006 Australian family law reforms. Journal of Family
Studies, 14(2-3), 254-270.
Morrill, A. C., Dai, J., Dunn, S., Sung, I., & Smith, K. (2005). Child custody and visitation
decisions when the father has perpetrated violence against the mother. Violence Against
Women, 11, 1076-1107.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 204 2007-WG-BX-0028
Morrill, C. (2008). Institutional change through interstitial emergence: The growth of alternative
dispute resolution in American law, 1965-95. University of California, Berkley: Working
paper, Center for the Study of Law and Society.
Mouyi , A. (2006). Untangling the cognitive processes web. Paper. Seventh Annual Conference
of the International Society for Intelligence Research, San Francisco, California.
Murphy, J. C., & Rubinson, R. (2005). Domestic violence and mediation: Responding to the
challenges of crafting effective screens. Family Law Quarterly, 39(1), 53.
Muthen, L. K., & Muthen, B. O. (2001). Mplus : Statistical analysis with latent variables :
User's guide. Los Angeles, CA: Muthén & Muthén.
National Audit Office (2007). (2007). Legal aid and mediation for people involved in family
breakdown. (National Audit Office No. HC 256). London: TSO.
National Council of Juvenile and Family Court Judges (2008). A judicial guide to child safety in
custody cases. Reno, NV: NCJFCJ.
National Council of Juvenile and Family Court Judges (2008). Batterer Manipulation of the
Courts to Further Their Abuse, and Remedies for Judges, Synergy 12(1), 12-13. Retrieved
July 27, 2011, from
http://www.ncjfcj.org/images/stories/dept/fvd/pdf/synergy%2012%20no.1.pdf.
Newmark, L., Harrell, A., & Salem, P. (1995). Domestic violence and empowerment in custody
and visitation cases. Family & Conciliation Courts Review, 33(1), 30-62.
Nock, S. L., Sanchez, L. A., & Wright, J. D. (2008). Covenant marriage: The movement to
reclaim tradition in America. New Brunswick, NJ: Rutgers University Press.
Nylund, K. L., Asparouhov, T., & Muthen, B. O. (2007). Deciding on the number of classes in
latent class analysis and growth mixture modeling: A Monte Carlo simulation study.
Structural Equation Modeling: A Multidisciplinary Journal, 14(4), 535-569.
O'Leary, K. D. (2000). Are women really more aggressive than men in intimate relationships?
Comment on Archer (2000). Psychological Bulletin, 126(5), 685-689.
O’Leary, K. D. (1999). Psychological abuse: A variable deserving critical attention in domestic
violence. Violence and Victims, 14(1), 3–23.
Ontario Court of Justice and Superior Court of Justice (2011), Request for Proposals for Family
Mediation and Information Services, OSS_00166368 Issued February 18, 2011. Queen’s
Printer for Ontario.
Parker, T., Rogers, K., Collins, M., & Edleson, J. L. (2008). Danger zone: Battered mothers and
their families in supervised visitation. Violence Against Women, 14(11), 1313-1325.
Pearson, J. (1991). The equity of mediated divorce agreements. Mediation Quarterly, 9(2), 179-
197.
Pearson, J. (1993). Ten myths about family law. Family Law Quarterly, 27(2), 279-299.
Pearson, J. (1997). Mediating when domestic violence is a factor: Policies and practices in court-
based divorce mediation programs. Mediation Quarterly, 14(4), 319-336.
Pearson, J., & Griswold, E. A. (1997). Child support and domestic violence: A preliminary look
at client experiences with good cause exemptions to child support cooperation
requirements. The National Center for Policy Research. Retrieved on July 27, 2011, from
http://cospl.coalliance.org/fez/eserv/co:1128/hu6182d711997internet.pdf.
Pearson, J., Ring, M. L., & Milne, A. (1983). A portrait of divorce mediation services in the
public and private sector. Conciliation Courts Review, 21(1), 1-24.
Pearson, J., Thoennes, N., & Griswold, E. A. (2000). Child support and domestic violence: the
victims speak out. Violence against Women, 5(4), 427-448.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 205 2007-WG-BX-0028
Pearson, J., Thoennes, N., & Vanderkooi, L. (1982). The decision to mediate: Profiles of
individuals who accept and reject the opportunity to mediate contested custody and
visitation issues. Journal of Divorce, 6, 17-35.
Pence, E., Dasgupta, S. D., Taylor, T., & Praxis International (2003). Building safety for battered
women and their children into the child protection system: Written by Ellen Pence and
Terri Taylor; compiled by Shamita Das Dasgupta. Duluth, Minnesota: Praxis International.
Pollet, S. L. (2011). Economic Abuse: The Unseen Side of Domestic Violence, NYSBA Journal,
February, 40-44.
Pruitt, D. G. (1995). Process and outcome in community mediation. Negotiation Journal,
October, 365-377.
Putz, J. W., Ballard, R. H., Holtzworth-Munroe, A., Applegate, A. G., & Beck, C. J. A. (2009).
Intimate partner violence in divorce mediation: Detection and screening. Poster presented at
the 43rd annual convention of the Association for Behavioral and Cognitive Therapy
(ABCT), New York.
Raghavan, C. & Beck, C. J. A. (2010). What about coercion in same sex relationships?
Presentation to the American Psychology-Law Society Annual Conference, Vancouver,
BC Canada, March 19-20, 1010.
Rhoades, H. (2010). Mandatory mediation of family disputes: Reflections from Australia.
Journal of Social Welfare & Family Law, 32(2), 183-194.
Rimelspach, R. L. (2001). Mediating family disputes in a world with domestic violence: How to
devise a safe and effective court-connected mediation program. Ohio State Journal on
Dispute Resolution, 17, 95-112.
Saks, M. J. (1992). Do we really know anything about the behavior of the tort litigation system.
and why not? University of Pennsylvania Law Review, 140(4), 1147-1292.
Salem, P. (2009). The emergence of triage in family court services: The beginning of the end for
mandatory mediation? Family Court Review, 47(3), 371-388.
Salem, P., & Dunford-Jackson, B. (2008a). Beyond politics and positions: A call for
collaboration between family court and domestic violence professionals. Family Court
Review, 46(3), 437-453.
Salem, P., & Dunford-Jackson, B. L. (2008b). Beyond politics and positions: A call for
collaboration between family court and domestic violence professionals. Juvenile and
Family Court Journal, 59(3), 19-37.
Sales, B., Beck, C. J. A., & Haan, R. E. (1993a). Self-representation in divorce cases. Chicago:
American Bar Association.
Sales, B. D., Beck, C. J. A., & Haan, R. K. (1993b). Is self-representation a reasonable
alternative to attorney representation in divorce cases? Saint Louis University Law Journal,
37(3), 553-605.
Saltzman, L. E., Fanslow, J. L., McMahon, P. M., & Shelley, G. A. (1999). Intimate partner
violence surveillance: Uniform definitions and recommended data elements (version 1.0).
Atlanta, GA: Centers for Disease Control.
Sander, F. E. A. (1976). Varieties of dispute processing, the pound conference. Federal Rules
Decisions, 70, 111-134.
Saunders, D. G. (2007). Child custody and visitation decisions in domestic violence cases: Legal
trends, risk factors, and safety concerns (Revised 2007). Harrisburg, PA:
NRCDV/VAW.net. Retrieved 7/12/ 2011, from http://www.vawnet.org/applied-research-
papers/print-document.php?doc_id=371.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 206 2007-WG-BX-0028
Saposnek, D. T. (1983). Mediating child custody disputes. San Francisco, CA: Josey Bass.
Saposnek, D. T. (2004). Commentary: The future of the history of family mediation research.
Conflict Resolution Quarterly, 22(1-2), 37-53.
Sbarra, D. A., & Emery, R. E. (2005). Co-parenting conflict, nonacceptance, and depression
among divorced adults: Results from a 12-year follow-up study of child custody
mediation using multiple imputation. American Journal of Orthopsychiatry, 75, 22–37.
Schraufnagel, S., & Li, Q. (2010). Testing the suitability of mediation of child support orders in
title IV-D cases. Research on Social Work Practice, 20(2), 212-222.
Slep, A. M. S., & O'Leary, S. G. (2005). Parent and partner violence in families with young
children: Rates, patterns, and connections. Journal of Consulting and Clinical Psychology,
73(3), 435-444.
Spitko, G., E. (2006). Constitutional function of biological paternity: Evidence of the biological
mother's consent to the biological father's co-parenting of her child. Arizona Law Review,
48, 97-147.
Stark, E. (2006). Commentary on Johnson's 'Conflict and control: Gender symmetry and
asymmetry in domestic violence'. Violence Against Women, 12(11), 1019-1025.
Stark, E. (2007). Coercive control: How men entrap women in personal life. New York: Oxford
University Press.
Stein, H. R. (1998). Massachusetts' "Lizzie's Law": protection for children or violation of
parents' constitutional rights?. Boston University Law Review, 78, 5, 1547-1582.
Straus, M. A. (1979). Measuring intrafamily conflict and violence: The conflict tactics (CT)
scales. Journal of Marriage and the Family, 41(1), 75-88.
Straus, M. A. (2008). Dominance and symmetry in partner violence by male and female
university students in 32 countries. Children and Youth Services Review, 30, 252-275.
Straus, M. A., Hamby, S. L., Boney-McCoy, S., & Sugarman, D. B. (1996). The revised tactics
scales (CTS2): Development and preliminary psychometric data. Journal of Family
Issues, 17(3), 283-316.
Tanha, M., Beck, C. J. A., Figueredo, A., & Raghavan, C. (2010). Sex differences in intimate
partner violence and the use of coercive control as a motivational factor for intimate partner
violence. Journal of Interpersonal Violence, 25(10), 1836-1854.
Task Force to Expand Access to Civil Legal Services in New York (2010). Report to the Chief
Judge of the State of New York. United court system of the State of New York. Retrieved
July 27, 2011, from http://www.courts.state.ny.us/ip/access-civil-legal-services/PDF/CLS-
TaskForceREPORT.pdf.
Taylor, R. B. (1998). Preventing violence against women and children. New York: Milbank
Memorial fund. Retrieved 07/27/2011, from http://www.milbank.org/reports/violence.html
.
Theran, S. A., Sullivan, C. M., Bogat, G. A., & Stewart, C. S. (2006). Abusive partners and ex-
partners: Understanding the effects of relationship to the abuser on women’s well-being.
Violence against Women, 12(10), 950–969.
Thoennes, N., Salem, P., & Pearson, J. (1995). Mediation and domestic violence: Current
policies and practices. Family & Conciliation Courts Review, 33(1), 6-29.
Tishler, C. L., Bartholomae, S., Katz, B. L., & Landry-Meyer, L. (2004). Is domestic violence
relevant?: An exploratory analysis of couples referred for mediation in family court.
Journal of Interpersonal Violence, 19(9), 1042-1062.
Tjaden, P. G., Thoennes, N., National Institute of Justice (U.S.), & Centers for Disease Control
and Prevention (U.S.). (2000). Extent, nature, and consequences of intimate partner
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 207 2007-WG-BX-0028
violence. Washington, DC: U.S. Dept. of Justice, Office of Justice Programs, National
Institute of Justice.
Tondo, C., Coronel, R., & Drucker, B. (2001). Mediation trends: A survey of the states. Family
Court Review, 39(4), 431-453.
Trinder, L., Firth, A., & Jenks, C. (2010). 'So presumably things have moved on since then?' the
management of risk allegations in child contact dispute resolution. International Journal of
Law, Policy and Family, 24(1), 29-53.
Troxel v. Grandville, 2000. 530 U.S. 57.
U.S. Census Bureau (2000). Census 2000 profiles for Pima County, Arizona. Retrieved
02/21/2011, from censtats.census.gov/data/AZ/05004019.pdf
U.S. Department of Health and Human Services. (2010). The 2000 HHS poverty guidelines.
Retrieved 02/21/2011, from aspe.hhs.gov/poverty/00poverty.htm
Vincent, M. (1995). Mandatory mediation of custody disputes: Criticism, legislation and support.
Vermont Law Review, 20, 255-297.
Walker, J., Great Britain, & University of Newcastle upon Tyne. (2001). Information meetings
and associated provisions within the Family Law Act 1996: Final evaluation report.
Newcastle upon Tyne: Newcastle Centre for Family Studies.
Walker, J. (2010). Family mediation: The rhetoric, the reality and the evidence. Tidsskrift for
Norsk Psykologforening, Journal of the Norwegian Psychological Association, 47(8), 676-
687.
Walker, L. E. (1984). The battered woman syndrome. New York: Springer.
Walker, J. A. (1989). Family conciliation in Great Britain: From research to practice to research.
Mediation Quarterly, 24, 29-54.
Warshak, R. A. (2010). Divorce poison: How to protect your family from bad-mouthing and
brainwashing. New York: Harper.
Weitzman, L. J. (1988). Women and children last: The social and economic consequences of
divorce law reforms. In M. S. S.M. Dornbusch (Ed.), Feminism, children, and the new
families (pp. 212-248).
White, J. W., Smith, P. H., Koss, M. P., & Figueredo, A. J. (2000). Intimate partner aggression-
what have we learned? comment on Archer (2000). Psychological Bulletin, 126(5), 690-
696.
White, P., L. (2005). You may never see your child again: Adjusting the batterer's visitation
rights to protect children from the future abuse. American University Journal of Gender,
Social Policy & the Law, 1-17.
Wissler, R. (1999a). Study suggests domestic violence does not affect settlement. Dispute
Resolution Magazine, 6 (Fall), 29.
Wissler, R. L., & State Justice Institute (U.S.). (1999b). Trapping the data: An assessment of
domestic relations mediation in Maine and Ohio courts. Columbus, OH: Supreme Court of
Ohio.
Yearwood, D. L., dowd, K., & Starsoneck, L. (2008). Exploring alternatives for improving child
support application filing rates within the context of civil protection order cases. Children
and Youth Services Review, 30, 821-833.
Yellott, A. W. (1990). Mediation and domestic violence: A call for collaboration. Conflict
Resolution Quarterly, 8(1), 39-50.
Yllo, K., & Bograd, M. L. (1988). Feminist perspectives on wife abuse. Newbury Park, Calif.:
Sage Publications.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 208 2007-WG-BX-0028
Zolotor, A. J., Theodore, A. D., Coyne-Beasley, T., & Runyan, D. K. (2007). Intimate partner
violence and child maltreatment: Overlapping risk. Brief Treatment and Crisis Intervention,
7(4), 305-321.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 209 2007-WG-BX-0028
Dissemination of Research Findings
Publications
Beck, C. J. A., Walsh, M. E., & Weston, R. (2009). The difficult balance of safety versus access:
Mediation agreements of families with intimate partner abuse. Invited paper. Special
Issue of Family Court Review, 47(3), 401-415.
Beck, C. J. A., Menke, J. M., O’Hara Brewster, K., & Figueredo, A. J. (2009). The Relationship
Behavior Rating Scale: Validation and expansion of a revised measure of intimate partner
violence. Journal of Divorce and Remarriage, 50(1/2), 295-308.
Beck, C. J. A., Walsh, M. E. & Mechanic, M. B., & Taylor, C. S. (2009). Intimate partner
violence screening and accommodations provided for safety in mandatory divorce
mediation. Law and Human Behavior, 34, 227-240.
Tanha, M., Beck, C. J. A., Raghavan, C., & Figueredo, A. J. (2010). Coercive Control as a
Motivational Factor for Intimate Partner Violence. Journal of Interpersonal Violence,
25(10), 1836-1854.
Beck, C. J. A., & Raghavan, C. (2010). Coercive control measures couple power dynamics in
custody mediation. Family Court Review, 48(3), 555-565.
Beck, C. J. A, Walsh, M. E., Ballard, R. H., Holtzworth-Monroe, A., Applegate, A., & Putz, J.
W. (2010) Divorce mediation with and without legal representation: A focus on intimate
partner violence and abuse. Family Court Review, 48(4), 631-645.
Holtzworth-Monroe, A., Beck, C. J. A, & Applegate, A. (2010). The Mediator’s Assessment of
Safety Issues and Concerns (MASIC): A screening interview for intimate partner
violence and abuse available in the public domain. Family Court Review, 48(4), 646-662.
Ballard, R. H., Holtzworth-Munroe, A., Applegate, A. G., & Beck, C. J. A. (2011). Detecting
intimate partner violence in family and divorce mediation: A randomized trial of intimate
partner violence screening. Psychology, Public Policy, and Law, 17(2), 241-263.
Thesis
Perkins, A.B. (2010). Spousal rape in divorcing couples: An exploration of spousal rape through
the report of couples in divorce mediation (Unpublished Master's Thesis). University of
Arizona, AZ.
Publications in Progress and Submitted
Beck, C. J. A. & Walsh, M. E. (in preparation). Understanding gender differences in parenting
time in the context of intimate partner abuse.
Beck, C. J. A., Anderson, E., Walsh, M. E. & Benjamin, G. A. H. (in preparation) Latent class
analysis of couple level intimate partner violence in couples mandated to attend divorce
mediation.
Perkins, A.B., Becker, J.V., Beck, C.J.A., Jacobs, W.J., Tehee, M. (in preparation). Spousal rape
and interpersonal violence and abuse.
Perkins, A.B., Becker, J.V., Beck, C.J.A., Jacobs, W.J., Tehee, M. (in preparation). Spousal rape
in divorce mediation: Interactions with mediators and law enforcement.
Davidson, R. D. (in preparation). Divorcing Couples: Self-Reports of Intimate Partner Abuse and
Law Enforcement Reports of Involvement. Undergraduate Honors Thesis.
Tehee, M. & Beck, C. J. A. (in preparation). Gender symmetry in domestic violence depends on
the definition and the research methods used to measure it (invited chapter). In C.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 210 2007-WG-BX-0028
Raghavan Domestic Violence: Quantitative and Qualitative Analysis in Dialogue. North
Eastern Publishers, North Eastern Series on Gender, Crime, and Law.
Presentations
Beck, C. J. A., (2007). IPV Assessment, Screening and Accommodations: Implications for
Court-Based Divorce Mediation. Presentation to Indiana University School of Law and
Indiana University Department of Psychology and Brain Sciences, Working Conference,
For the Sake of the Children: Advances in Family Dispute Resolution, November 15,
2007.
Beck, C. J. A., Mechanic, M., Walsh, M., & Taylor, C. S. (2008). IPV Assessment, Screening
and Accommodations: Implications for Court-Based Divorce Mediation. Presentation to
the Arizona Legislative Committee on the Impact of Domestic Violence and the Courts,
Phoenix, Arizona: February 12, 2008.
Beck, C. J. A., Mechanic, M., Walsh, M., Taylor, C. S. (2008). IPV Assessment in the Context
of Divorce Mediation. Presentation to the American Psychology-Law Society Annual
Conference, St. Petersburg, Florida: March 7, 2008.
Beck, C. J. A. (2008). IPV Assessment, Screening and Accommodations: Implications for
Court-Based Mediation. Presentation to the Arizona Legislative Committee on the
Impact of Domestic Violence and the Courts, Phoenix, Arizona, February 12, 2008.
Frost, L., Beck, C. J. A., & Hedeen, T. (2008). Mediation with Clients with Mental
Impairments. Presentation at the Association for Conflict Resolution National
Conference, Austin, Texas, September 25, 2008.
Raghavan, C., & Beck, C. J. A. (2009). Preliminary examination of risk of femicide in a sample
of non-abused divorcing women. Understanding and Preventing the Murder of Women in
Intimate Partner Violence Conference. New York, NY, November 7, 2008.
Beck, C. J. A., Walsh, M. Mechanic, M., Taylor, C. S. (2009). A Naturalistic Evaluation of
Mediator Assessment, Documentation, and Disposition of Child Custody Cases Involving
IPV. Presentation to the American Psychology-Law Society Annual Conference, San
Antonio, TX, March 5, 2009.
Beck, Walsh, Mechanic and Tehee (2009). Mediator identification and response to IPV in
couples participating in divorce mediation. Association of Family and Conciliation
Courts, May New Orleans. AFCC 46
th
Annual Conference in New Orleans, May 27-30,
2009.
Beck, C. J. A., (2009). Intimate Partner Abuse in Couples Mandated to Attend Divorce
Mediation. Presented at the University of Texas at Austin SPAM Group Meeting,
September 29, 2009.
Beck, C. J. A., Walsh, M. E., Weston, R., & Mechanic, M. B. (2009). Analysis of Divorce
Mediation Agreements of Families Reporting Intimate Partner Abuse/Violence. Paper
presented at a symposium at the 43
rd
annual convention of the Association for Behavioral
and Cognitive Therapy (ABCT). New York, November, 2009.
Holtzworth-Munroe, A., Arany, J.G., Putz, J.W., Ballard, R.H., Applegate, A.G., & Beck, C.J.A.
(2009). The impact of intimate partner violence on divorce mediation outcomes:
Implications for children’s vulnerability. Paper presented at a symposium at the 43
rd
annual convention of the Association for Behavioral and Cognitive Therapy (ABCT).
New York, November 2009.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 211 2007-WG-BX-0028
Beck, C. J. A., & Raghavan, C. (2010). Divorce mediation: The importance of assessing coercive
control. Presentation to the American Psychology-Law Society Annual Conference,
Vancouver, BC Canada, March 19-20, 1010.
Raghavan, C. & Beck, C. J. A. (2010). What about coercion in same sex relationships?
Presentation to the American Psychology-Law Society Annual Conference, Vancouver,
BC Canada, March 19-20, 1010.
Applegate, A. G., Ballard, R. H., Holtzworth-Munroe, A., Putz, J. W., & Beck, C. J. A. (2010).
Assessing and managing cases with intimate partner violence at a law school family
mediation clinic. Presentation to the American Psychology-Law Society Annual
Conference, Vancouver, BC Canada, March 19-20, 2010.
Beck, C. J. A., Walsh, M. E., Ballard, R.H, Holtzworth-Munroe, A., Applegate, A.G., Putz, J.
W., & Anderson, E. R. (2010). Divorce Mediation & Legal Representation: A Focus on
Intimate Partner Violence and Abuse. Paper presented at a workshop at the 47th Annual
Conference of the Association of Family and Conciliation Courts in Denver, CO, June 2-
5, 2010.
Ballard, R.H, Pettersen, M.M., Putz, J.W, Applegate, A.G., Beck, C.J.A., & Holtzworth-Munroe,
A. (2010). Legal representation in mediation: party fear, concern, preparation and
satisfaction. Paper presented at a workshop at the 47th Annual Conference of the
Association of Family and Conciliation Courts in Denver, CO, June 2-5, 2010.
Raghavan, C., and Beck, C. J. A. (2010). Coercive Control in Abusive Relationships.
Presentation to the Ninth Biennial International Conference, John Jay College of
Criminal Justice, Marrakesh, Morocco, June, 2-5, 2010.
Beck, C. J. A. & Anderson, Edward R. (2010). Why Smart People Disagree about Asymmetry
and Symmetry in Perpetration of IPV. Invited Presentation at the National Institute of
Justice 15
th
Anniversary Conference: Domestic Violence Research 15 Years After the
Violence Against Women Act, Washington, DC. June, 2010.
Posters
Davidson, R. D., Tehee M., Perkins A., Beck, C. J. A (2009). Divorcing Couples: Self-Reports of
Intimate Partner Abuse and Law Enforcement Reports of Involvement. Presented at
University of Arizona’s Graduate and Professional Student Council’s Student Showcase,
Tucson, Arizona, November 6, 2009.
Putz, J.W., Ballard, R.H., Holtzworth-Munroe, A., Applegate, A.G., & Beck, C.J.A. (2009).
Intimate Partner Abuse in divorce mediation: Detection and screening. Poster presented
at the 43
rd
annual convention of the Association for Behavioral and Cognitive Therapy
(ABCT). New York, NY. November 22, 2009.
Beck, C. J. A., Tehee M., Davidson, R. D., Perkins A. (2010). Divorcing Couples: Self-Reports
of Intimate Partner Abuse and Law Enforcement Reports of Involvement. Presented at
National Summit on Interpersonal Violence and Abuse Across the Lifespan: Forging a
Shared Agenda, Dallas, Texas, February 24, 2010.
Davidson, R. D., Tehee M., Perkins A., Beck, C. J. A. (2010). Divorcing Couples: Self-Reports
of Intimate Partner Abuse and Law Enforcement Reports of Involvement. Presented at
American Psychology-Law Society Annual Conference, Vancouver, British Columbia,
Canada, March 19, 2010.
Davidson, R. D., Tehee M., Perkins A., Beck, C. J. A. (2010). Divorcing Couples: Self-Reports
of Intimate Partner Abuse and Law Enforcement Reports of Involvement. Presented at
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 212 2007-WG-BX-0028
University of Arizona’s College of Science Mind and Brain Lecture Series, Tucson,
Arizona, March 2, 24, & 30, 2010.
Davidson, R. D., Tehee M., Perkins A., Beck, C. J. A. (2010) Divorcing Couples: Self-Reports of
Intimate Partner Abuse and Law Enforcement Reports of Involvement. Presented at
University of California – Los Angeles Psychology Undergraduate Research Conference,
Los Angeles, California, May 14, 2010.
Davidson, R. D., Tehee M., Perkins A., Beck, C. J. A. (2010). Divorcing Couples: Self-Reports
of Intimate Partner Abuse and Law Enforcement Reports of Involvement. Presented at
Arizona Psychological Association Annual Convention, Tucson, Arizona, October 15-16,
2010.
Davidson R. D., Tehee M., Perkins A., Beck, C. J. A. (to be presented, 2011). Divorcing
Couples: Self-Reports of Intimate Partner Abuse and Law Enforcement Reports of
Involvement. Presented at University of Arizona Honors College Forum, Tucson,
Arizona, April 29, 2011.
Davidson R. D., Chen, M., Beck, C. J. A., Edais, A. (to be presented, 2011). Divorcing Couples:
Reasons for Returns to Mediation. Presented at Association of Family and Conciliation
Courts Annual Conference, Orlando, Florida, June 1-4, 2011.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 213 2007-WG-BX-0028
Attachments
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 214 2007-WG-BX-0028
Attachment A: Mediation Case File Data
Demographics
Mediation Order Date
Mediation Start Date
Mediation End Date
RESEARCH_RECORD_ID
(SPSS label: casenum)
h_CLIENT_AGE_AT_PETITION_DATE: Husband’s age at the petition date
(SPSS label name: hpetage)
h_ETHNICITY: Husband’s Ethnicity
(SPSS label: hethnic)
h_EDUCATION: Husband’s Level of Education
(SPSS label: hedu)
h_OCCUPATION: Husband’s Occupation
(SPSS label: hoccup)
h_GROSS_INCOME: Husband’s Gross Income
(SPSS label: hgrinc)
h_RELIGION: Husband’s Religion
(SPSS label: hrelig)
h_MARRIAGE_COUNT: Number of marriages the husband has had
(SPSS label: hmarriag)
h_AGREEMENT_OUTCOME: The agreement outcome of the case
(SPSS label: hagree)
h_DOMESTIC_VIOLENCE: Husband: Was domestic Violence Present
(SPSS label: hdv)
h_CHILDREN_COUNT: Count of the number of children the father has
(SPSS label: hchildnm)
h_INDIVIDUAL_SESSION_COUNT: Count of the number of individual therapy sessions for the father
(SPSS label: hindsenm)
h_JOINT_SESSION_COUNT: Count of the number of joint therapy sessions
(SPSS label: hjntsenm)
h_CONF_HELD_COUNT: Count of the number of mediation sessions held
(SPSS label: hconfnum)
w_CLIENT_AGE_AT_PETITION_DATE: Wife’s age at the petition date
(SPSS label name: wpetage)
w_ETHNICITY: Wife’s Ethnicity
(SPSS label: wethnic)
w_EDUCATION: Wife’s Level of Education
(SPSS label: wedu)
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 215 2007-WG-BX-0028
w_OCCUPATION: Wife’s Occupation
(SPSS label: woccup)
w_GROSS_INCOME: Wife’s Gross Income
(SPSS label: wgrinc)
w_RELIGION: Wife’s Religion
(SPSS label: wrelig)
w_MARRIAGE_COUNT: Number of marriages the Wife has had
(SPSS label: wmarriag)
w_AGREEMENT_OUTCOME: The agreement outcome of the case
(SPSS label: wagree)
w_DOMESTIC_VIOLENCE: Wife: Was domestic Violence Present
(SPSS label: wdv)
w_CHILDREN_COUNT: Count of the number of children the Wife has
(SPSS label: wchildnm)
w_INDIVIDUAL_SESSION_COUNT: Count of the number of individual therapy sessions for the mother
(SPSS label: windsenm)
w_JOINT_SESSION_COUNT: Count of the number of joint therapy sessions
(SPSS label: wjntsenm)
w_CONF_HELD_COUNT: Count of the number of mediation sessions held
(SPSS label: wconfnum)
Table 1
Referral Source: Referral Source to the FCCC (Family Center for the Conciliation Court)
(SPSS label name: referral)
MO Requested by: Mediation was requested by
(SPSS label name: moreqby)
MO Requested by other, desc: Mediation was requested by other, description
(SPSS label name: moreqoth)
Case Open Day Count: How many days the case was open
(SPSS label name: codaycnt)
M Representation: Mother’s representation
(SPSS label name: mrepres)
F Representation: Father’s representation
(SPSS label name: frepres)
Mediator Staff: Who was the mediator assigned to the case
(SPSS label name: mediator)
M Spanish: Is the mother Spanish speaking
(SPSS label name: mspanish)
F Spanish: Is the father Spanish speaking
(SPSS label name: fspanish)
M SSAS Discussed Concerns: Did the mother discuss concerns with the support or administrative staff (e.g. security)
(SPSS label name: mssasdc)
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 216 2007-WG-BX-0028
M SSAS Accompanied: Was the mother accompanied to her car, etc. by support or administrative staff (e.g. security)
(SPSS label name: mssasacc)
M SSAS Separate Waiting Area: Did the mother need a separate waiting area
(SPSS label name: mssasswa)
M SSAS Separate Orientation: Did the mother need a separate orientation
(SPSS label name: mssasso)
M SSAS Other Desc: What were the mother’s other concerns that she discussed with support or administrative staff
(SPSS label name: mssasoth)
M SSAS Referred for Further Review: Was the mother referred for further review (form is located in the front office)
(SPSS label name: mssasrfr)
F SSAS Discussed Concerns: Did the father discuss concerns with the support or administrative staff (e.g. security)
(SPSS label name: fssasdc)
F SSAS Accompanied: Was the father accompanied to his car, etc. by support or administrative staff (e.g. security)
(SPSS label name: fssasacc)
F SSAS Separate Waiting Area: Did the father need a separate waiting area
(SPSS label name: fssasswa)
F SSAS Separate Orientation: Did the father need a separate orientation
(SPSS label name: fssasso)
F SSAS Other Desc: What were the father’s other concerns that he discussed with support or administrative staff
(SPSS label name: fssasoth)
F SSAS Referred for Futher Review: Was the father referred for further review (form is located in the front office)
(SPSS label name: fssasrfr)
M MS Accompanied: Did the mediator recommend the mother be accompanied
(SPSS label name: mmsacc)
M MS Separate Waiting Area: Did the mediator recommend a separate waiting area for the mother
(SPSS label name: mmsswa)
M MS Other Desc: Mediator’s other recommendations for the mother, description
(SPSS label name: mmsoth)
F MS Accompanied: Did the mediator recommend the father be accompanied
(SPSS label name: fmsacc)
F MS Separate Waiting Area: Did the mediator recommend a separate waiting area
(SPSS label name: fmsswa)
F MS Other Desc: Mediator’s other recommendations for the father, description
(SPSS label name: fmsoth)
M CRC Security Escort Count: Contact Record Categories, count of the number of escorts needed for the mother
(SPSS label name: mcrcsec)
F CRC Security Escort Count: Contact Record Categories, count of the number of escorts needed for the father
(SPSS label name: fcrcsec)
U CRC Security Escort Count: Contact Record Categories, count of escorts needed for unknown
(SPSS label name: ucrcsec)
CRC Leave Separately Count: Contact record categories, count of the number of times the couples left separately
(SPSS label name: crclvsep)
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 217 2007-WG-BX-0028
CRC Security Other Desc: Contact record categories, security description
(SPSS label name: crcsoth)
M Order of Protection: Did the mother have an order of protection
(SPSS label name: morofpro)
M Order of Protection Desc: Order of protection for mother, description
(SPSS label name: morofprd)
F Order of Protection: Did the father have an order of protection
(SPSS label name: forofpro)
F Order of Protection Desc: Order of protection for father, description
(SPSS label name: forofprd)
Critical Incident Report: Was a critical incident report included in the file
(SPSS label name: crincrep)
M Internal Referral: Did the mediator refer the mother for counseling
(SPSS label name: minref)
M internal Referral Reason: The reason the mother was referred by the mediator
(SPSS label name: minrefr)
M internal Referral Counselor Meeting: Was mother was referred to a meeting with a counselor
(SPSS label name: minrefcm)
M internal Referral Separate File: Did the mediator make a separate file for the mother
(SPSS label name: minrefsf)
M internal Referral DV: Was there a referral for the mother because of evidence of domestic violence
(SPSS label name: minrefdv)
M internal Referral Prior Case: Did the mother have a prior case
(SPSS label name: minrefpc)
M Internal Referral Comment: The mediators comments re: the internal referral
(SPSS label name: minrefco)
F Internal Referral: Did the mediator refer the father for counseling
(SPSS label name: finref)
F internal Referral Reason: The reason the father was referred by the mediator
(SPSS label name: finrefr)
F internal Referral Counselor Meeting: Was the father referred to a meeting with a counselor
(SPSS label name: finrefcm)
F internal Referral Separate File: Did the mediator make a separate file for the father
(SPSS label name: finrefsf)
F internal Referral DV: Was there a referral for the father because of evidence of domestic violence
(SPSS label name: finrefdv)
F internal Referral Prior Case: Did the father have a prior case
(SPSS label name: finrefpc)
F internal Referral Comment: The mediators comments re: the internal referral
(SPSS label name: finrefco)
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 218 2007-WG-BX-0028
SUB MO Recom: What does the mediator recommend re: subsequent mediation
(SPSS label name: morecom)
SUB MO Recom Desc: Subsequent mediation recommendation, description
(SPSS label name: morecomd)
MO Outcome: Mediation outcome
(SPSS label name: moutcome)
Agreement DV Language: In the agreement is there language to indicate domestic violence (e.g. supervised visitation,
limited phone calls and visits, etc.)
(SPSS label name: agdvlang)
Agreement DV Language Desc: Language in the agreement indicating domestic violence, description
(SPSS label name: agdvlgd)
Agreement Child Maltreatment: Is there language in the agreement to indicate child maltreatment
(SPSS label name: agchlmal)
Agreement Child Maltreatment Desc: Language in the agreement indicating child maltreatment, description
(SPSS label name: agchmald)
Agreement Other Special Language: Was there other special language in the agreement to indicate a problem
(SPSS label name: agothsl)
Agreement Other Special Language Desc: Special language in the agreement, description
(SPSS label name: agothsld)
Table 4
M Months since last together: How many months did the mother report passed since the couple was last together
(SPSS label name: mmslatog)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #1: Resolved things peacefully with me.
(SPSS label name: mrbrs1)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #2: Did not want me to have male/female friends.
(SPSS label name: mrbrs2)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #3: Became very upset if my chores were not done
when he/she thought they should be.
(SPSS label name: mrbrs3)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #4: Demanded that I obey him/her.
(SPSS label name: mrbrs4)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #5: Put me down.
(SPSS label name: mrbrs5)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #6: Became angry if I said he/she was drinking too
much or using drugs.
(SPSS label name: mrbrs6)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #7: Demanded that I perform sex acts that I did not
want to.
(SPSS label name: mrbrs7)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #8: Told me I was ugly or unattractive.
(SPSS label name: mrbrs8)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #9: Told me I could not manage or take care of
myself without him/her.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 219 2007-WG-BX-0028
(SPSS label name: mrbrs9)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #10: Insulted or shamed me in front of others.
(SPSS label name: mrbrs10)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #11: Became very angry if I disagreed with his/her
point of view.
(SPSS label name: mrbrs11)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #12: Controlled how much money I could have or
how I spent it.
(SPSS label name: mrbrs12)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #13: Controlled my coming and going.
(SPSS label name: mrbrs13)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #14: Felt that I should not work or go to school.
(SPSS label name: mrbrs14)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #15: Insisted on sex whether I wanted it or not.
(SPSS label name: mrbrs15)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #16: Screamed or yelled at me.
(SPSS label name: mrbrs16)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #17: Had no respect for my feelings.
(SPSS label name: mrbrs17)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #18: Acted like a bully toward me.
(SPSS label name: mrbrs18)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #19: Treated me like I was stupid or dumb.
(SPSS label name: mrbrs19)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #20: Was mean or rude to me. (SPSS label name:
mrbrs20)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #21: Physically forced me to have sex.
(SPSS label name: mrbrs21)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #22: Pushed or shoved me around.
(SPSS label name: mrbrs22)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #23: Hit or punched me.
(SPSS label name: mrbrs23)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #24: Threatened me with or used a weapon against
me.
(SPSS label name: mrbrs24)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #25: Hurt me so badly I had to seek medical help.
(SPSS label name: mrbrs25)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #26: Made me afraid for my life.
(SPSS label name: mrbrs26)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #27: Throw me around the room.
(SPSS label name: mrbrs27)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #28: Acted like he/she would like to kill me.
(SPSS label name: mrbrs28)
1 = None of the time
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 220 2007-WG-BX-0028
Mother’s Relationship Behavior Rating Scale (Self-report) Question #29: Threatened to cut or stab me with a knife or
other sharp object.
(SPSS label name: mrbrs29)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #30: Tried to choke or strangle me.
(SPSS label name: mrbrs30)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #31: Kicked or stomped me.
(SPSS label name: mrbrs31)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #32: Threw objects at me.
(SPSS label name: mrbrs32)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #33: Bit or scratched me.
(SPSS label name: mrbrs33)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #34: Violently pinched or twisted my skin.
(SPSS label name: mrbrs34)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #35: Hurt me while we were having sex.
(SPSS label name: mrbrs35)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #36: Injured my genitals.
(SPSS label name: mrbrs36)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #37: Injured my breasts.
(SPSS label name: mrbrs37)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #38: Tried to suffocate me.
(SPSS label name: mrbrs38)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #39: Poked or jabbed me with pointed objects.
(SPSS label name: mrbrs39)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #40: Broke one or more of my bones.
(SPSS label name: mrbrs40)
Mother’s Relationship Behavior Rating Scale (Self-report) Question #41: Kicked my face and head.
(SPSS label name: mrbrs41)
F Months since last together: How many months did the father report passed since the couple was last together
(SPSS label name: fmslatog)
Father’s Relationship Behavior Rating Scale (Self-report) Question #1: Resolved things peacefully with me.
(SPSS label name: frbrs1)
Father’s Relationship Behavior Rating Scale (Self-report) Question #2: Did not want me to have male/female friends.
(SPSS label name: frbrs2)
Father’s Relationship Behavior Rating Scale (Self-report) Question #3: Became very upset if my chores were not done
when he/she thought they should be.
(SPSS label name: frbrs3)
Father’s Relationship Behavior Rating Scale (Self-report) Question #4: Demanded that I obey him/her.
(SPSS label name: frbrs4)
Father’s Relationship Behavior Rating Scale (Self-report) Question #5: Put me down.
(SPSS label name: frbrs5)
Father’s Relationship Behavior Rating Scale (Self-report) Question #6: Became angry if I said he/she was drinking too
much or using drugs.
(SPSS label name: frbrs6)
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 221 2007-WG-BX-0028
Father’s Relationship Behavior Rating Scale (Self-report) Question #7: Demanded that I perform sex acts that I did not
want to.
(SPSS label name: frbrs7)
Father’s Relationship Behavior Rating Scale (Self-report) Question #8: Told me I was ugly or unattractive.
(SPSS label name: frbrs8)
Father’s Relationship Behavior Rating Scale (Self-report) Question #9: Told me I could not manage or take care of myself
without him/her.
(SPSS label name: frbrs9)
Father’s Relationship Behavior Rating Scale (Self-report) Question #10: Insulted or shamed me in front of others.
(SPSS label name: frbrs10)
Father’s Relationship Behavior Rating Scale (Self-report) Question #11: Became very angry if I disagreed with his/her
point of view.
(SPSS label name: frbrs11)
Father’s Relationship Behavior Rating Scale (Self-report) Question #12: Controlled how much money I could have or
how I spent it.
(SPSS label name: frbrs12)
Father’s Relationship Behavior Rating Scale (Self-report) Question #13: Controlled my coming and going.
(SPSS label name: frbrs13)
Father’s Relationship Behavior Rating Scale (Self-report) Question #14: Felt that I should not work or go to school.
(SPSS label name: frbrs14)
Father’s Relationship Behavior Rating Scale (Self-report) Question #15: Insisted on sex whether I wanted it or not.
(SPSS label name: frbrs15)
Father’s Relationship Behavior Rating Scale (Self-report) Question #16: Screamed or yelled at me.
(SPSS label name: frbrs16)
Father’s Relationship Behavior Rating Scale (Self-report) Question #17: Had no respect for my feelings.
(SPSS label name: frbrs17)
Father’s Relationship Behavior Rating Scale (Self-report) Question #18: Acted like a bully toward me.
(SPSS label name: frbrs18)
Father’s Relationship Behavior Rating Scale (Self-report) Question #19: Treated me like I was stupid or dumb.
(SPSS label name: frbrs19)
Father’s Relationship Behavior Rating Scale (Self-report) Question #20: Was mean or rude to me. (SPSS label name:
frbrs20)
Father’s Relationship Behavior Rating Scale (Self-report) Question #21: Physically forced me to have sex.
(SPSS label name: frbrs21)
Father’s Relationship Behavior Rating Scale (Self-report) Question #22: Pushed or shoved me around.
(SPSS label name: frbrs22)
Father’s Relationship Behavior Rating Scale (Self-report) Question #23: Hit or punched me.
(SPSS label name: frbrs23)
Father’s Relationship Behavior Rating Scale (Self-report) Question #24: Threatened me with or used a weapon against
me.
(SPSS label name: frbrs24)
Father’s Relationship Behavior Rating Scale (Self-report) Question #25: Hurt me so badly I had to seek medical help.
(SPSS label name: frbrs25)
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 222 2007-WG-BX-0028
Father’s Relationship Behavior Rating Scale (Self-report) Question #26: Made me afraid for my life.
(SPSS label name: frbrs26)
Father’s Relationship Behavior Rating Scale (Self-report) Question #27: Throw me around the room.
(SPSS label name: frbrs27)
Father’s Relationship Behavior Rating Scale (Self-report) Question #28: Acted like he/she would like to kill me.
(SPSS label name: frbrs28)
Father’s Relationship Behavior Rating Scale (Self-report) Question #29: Threatened to cut or stab me with a knife or
other sharp object.
(SPSS label name: frbrs29)
Father’s Relationship Behavior Rating Scale (Self-report) Question #30: Tried to choke or strangle me.
(SPSS label name: frbrs30)
Father’s Relationship Behavior Rating Scale (Self-report) Question #31: Kicked or stomped me.
(SPSS label name: frbrs31)
Father’s Relationship Behavior Rating Scale (Self-report) Question #32: Threw objects at me.
(SPSS label name: frbrs32)
Father’s Relationship Behavior Rating Scale (Self-report) Question #33: Bit or scratched me.
(SPSS label name: frbrs33)
Father’s Relationship Behavior Rating Scale (Self-report) Question #34: Violently pinched or twisted my skin.
(SPSS label name: frbrs34)
Father’s Relationship Behavior Rating Scale (Self-report) Question #35: Hurt me while we were having sex.
(SPSS label name: frbrs35)
Father’s Relationship Behavior Rating Scale (Self-report) Question #36: Injured my genitals.
(SPSS label name: frbrs36)
Father’s Relationship Behavior Rating Scale (Self-report) Question #37: Injured my breasts.
(SPSS label name: frbrs37)
Father’s Relationship Behavior Rating Scale (Self-report) Question #38: Tried to suffocate me.
(SPSS label name: frbrs38)
Father’s Relationship Behavior Rating Scale (Self-report) Question #39: Poked or jabbed me with pointed objects.
(SPSS label name: frbrs39)
Father’s Relationship Behavior Rating Scale (Self-report) Question #40: Broke one or more of my bones.
(SPSS label name: frbrs40)
Father’s Relationship Behavior Rating Scale (Self-report) Question #41: Kicked my face and head.
(SPSS label name: frbrs41)
Table 8
Critical Incident Reporting Count: Count of the number of critical incident reporting forms filled out
(SPSS label name: circount)
CIRF ITEM01: Critical Incident Reporting Form: Suicidal/homicidal threats and ideation
(SPSS label name: cirf1)
CIRF ITEM01 DESC: Critical Incident Reporting Form: Description of suicidal/homicidal threats and ideation
(SPSS label name: cirf1d)
CIRF ITEM02: Critical Incident Reporting Form: Subpoena for documentation and/or testimony of
counselor/mediator/staff member
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 223 2007-WG-BX-0028
(SPSS label name: cirf2)
CIRF ITEM02 DESC: Critical Incident Reporting Form: Description of the subpoena for documentation and/or
testimony of counselor/mediator/staff member
(SPSS label name: cirf2d)
CIRF ITEM03: Critical Incident Reporting Form: Information concerning an event which occurs before/after a session
but which indicates potential for harm, or other security issue
(SPSS label name: cirf3)
CIRF ITEM03 DESC: Critical Incident Reporting Form: Description of the information concerning an event which
occurs before/after a session but which indicates potential for harm, or other security issue
(SPSS label name: cirf3d)
CIRF ITEM04: Critical Incident Reporting Form: Reports that an attorney has recommended that a client not cooperate
with a counselor/mediator
(SPSS label name: cirf4)
CIRF ITEM04 DESC: Critical Incident Reporting Form: Description of the reports that an attorney has recommended
that a client not cooperate with a counselor/mediator
(SPSS label name: cirf4d)
CIRF ITEM05: Critical Incident Reporting Form: Reference to serious domestic violence (e.g. current/chronic assault,
harassment, serious threats of harm or death)
(SPSS label name: cirf5)
CIRF ITEM05 DESC: Critical Incident Reporting Form: Description of a reference to serious domestic violence (e.g.
current/chronic assault, harassment, serious threats of harm or death)
(SPSS label name: cirf5d)
CIRF ITEM06: Critical Incident Reporting Form: Accommodation for disability or language
(SPSS label name: cirf6)
CIRF ITEM06 DESC: Critical Incident Reporting Form: Description of accommodation for disability (specify) or
language (specify)
(SPSS label name: cirf6d)
CIRF ITEM07: Critical Incident Reporting Form: Physical or sexual abuse of a child
(SPSS label name: cirf7)
CIRF ITEM07 DESC: Critical Incident Reporting Form: Description of physical or sexual abuse of a child
(SPSS label name: cirf7d)
CIRF ITEM08: Critical Incident Reporting Form: Physical altercation in the office
(SPSS label name: cirf8)
CIRF ITEM08 DESC: Critical Incident Reporting Form: Description of any physical altercation in the office
(SPSS label name: cirf8d)
CIRF ITEM09: Critical Incident Reporting Form: Potential for complaint
(SPSS label name: cirf9)
CIRF ITEM09 DESC: Critical Incident Reporting Form: Description of any potential for complaint
(SPSS label name: cirf9d)
CIRF ITEM10: Critical Incident Reporting Form: Threat to abduct a child
(SPSS label name: cirf10)
CIRF ITEM10 DESC: Critical Incident Reporting Form: Description of any threat to abduct a child
(SPSS label name: cirf10d)
CIRF ITEM11: Critical Incident Reporting Form: Injury
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 224 2007-WG-BX-0028
(SPSS label name: cirf11)
CIRF ITEM11 DESC: Critical Incident Reporting Form: Description of any injury
(SPSS label name: cirf11d)
CIRF ITEM12: Critical Incident Reporting Form: Threats of harm or death to FCCC staff
(SPSS label name: cirf12)
CIRF ITEM12 DESC: Critical Incident Reporting Form: Description of any threats of harm or death to FCCC staff
(SPSS label name: cirf12d)
CIRF ITEM13: Critical Incident Reporting Form: Bizarre or unusual behavior
(SPSS label name: cirf13)
CIRF ITEM13 DESC: Critical Incident Reporting Form: Description of any bizarre or unusual behavior
(SPSS label name: cirf13d)
CIRF ITEM14: Critical Incident Reporting Form: Client leaves session without permission of the counselor/mediator
(SPSS label name: cirf14)
CIRF ITEM14 DESC: Critical Incident Reporting Form: Description of client leaving session without permission of the
counselor/mediator
(SPSS label name: cirf14d)
CIRF DESC: Critical Incident Reporting Form: Description of the incident(s) [Time, place, circumstance, witnesses]
(SPSS label name: cirfdesc)
CIRF STAFF RESPONS: Critical Incident Reporting Form: Description of the staff’s response to the incident
(SPSS label name: cirfsres)
CIRF DIRECTOR CONSULTED: Critical Incident Reporting Form: Was the FCCC Director consulted
(SPSS label name: cirfdirc)
CIRF STAFF CONSULTED: Critical Incident Reporting Form: Was the FCCC staff consulted
(SPSS label name: cirfscon)
CIRF SECURITY ESCORT MOTHER: Critical Incident Reporting Form: Was a security escort requested/provided for
the mother
(SPSS label name: cirfsem)
CIRF SECURITY ESCORT FATHER: Critical Incident Reporting Form: Was a security escort requested/provided for
the father
(SPSS label name: cirfsef)
CIRF SECURITY LEAVE SEPARATELY: Critical Incident Reporting Form: Were the clients instructed to leave
separately
(SPSS label name: cirfslvs)
CIRF SECURITY OTHER DESC: Critical Incident Reporting Form: Description of other security issues
(SPSS label name: cirfsoth)
Pre-mediation Screening Form
Subsequent Mediation: Is there a subsequent mediation PMI
(SPSS label name: submed)
Father reporting own major health problems
(SPSS label name: ffmajhea)
Father reporting mother's major health problems
(SPSS label name: fmmajhea)
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 225 2007-WG-BX-0028
Father reporting child(ren)'s major health problems
(SPSS label name: fcmajhea)
Father reporting own hospitalization for mental reasons
(SPSS label name: ffhosmen)
Father reporting mother's hospitalization for mental reasons
(SPSS label name: fmhosmen)
Father reporting child(ren)’s hospitalization for mental reasons
(SPSS label name: fchosmen)
Father reporting own hospitalization for physical reasons
(SPSS label name: ffhosphy)
Father reporting mother's hospitalization for physical reasons
(SPSS label name: fmhosphy)
Father reporting child(ren)'s hospitalization for physical reasons
(SPSS label name: fchosphy)
Father reporting own use of medicine
(SPSS label name: ffmeds)
Father reporting mother's use of medicine
(SPSS label name: fmmeds)
Father reporting child(ren)'s use of medicine
(SPSS label name: fcmeds)
Father reporting therapy
(SPSS label name: fthpy)
Father reporting own substance abuse in the past
(SPSS label name: ffsapast)
Father reporting mother’s substance abuse in the past
(SPSS label name: fmsapast)
Father reporting own substance abuse in the present
(SPSS label name: ffsapres)
Father reporting mother's substance abuse in the present
(SPSS label name: fmsapres)
Father reporting his fears
(SPSS label name: ffears)
Father reporting his concerns
(SPSS label name: fconcern)
Father reporting threats
(SPSS label name: fthreat)
Father reporting being coerced
(SPSS label name: fcoerce)
Father reporting harassment
(SPSS label name: fharass)
Father reporting intimidation
(SPSS label name: fintim)
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 226 2007-WG-BX-0028
Father reporting property damage
(SPSS label name: fproprty)
Father reporting physical abuse
(SPSS label name: fphyscl)
Father reporting the use of weapons
(SPSS label name: fweapon)
Father reporting sexual intimidation
(SPSS label name: fsexintm)
Father reporting sexual assualt
(SPSS label name: fsexaslt)
Father reporting sexual manipulation/control
(SPSS label name: fsexmnct)
Father reporting false allegations of sexual intimidation, sex. Assault, and man./control
(SPSS label name: fsexfalg)
Father reporting orders of protection in the present
(SPSS label name: foppres)
Father reporting orders of protection in the past
(SPSS label name: foppast)
Father reporting police being called
(SPSS label name: fpolcal)
Father reporting arrests for domestic violence
(SPSS label name: farstdv)
Father reporting either calls or arrests
(SPSS label name: fcalarr)
Father reporting going to the hospital because of DV
(SPSS label name: fgohosp)
Father reporting considering going to the hospital because of DV
(SPSS label name: fcnsdhsp)
Father reporting either going or considering going to the hospital because of DV
(SPSS label name: fgocnsd)
Father reporting going to the shelter
(SPSS label name: fshelter)
Father reporting CPS involvement
(SPSS label name: fcps)
Mother reporting father’s major health problems
(SPSS label name: mfmajhea)
Mother reporting own major health problems
(SPSS label name: mmmajhea)
Mother reporting child(ren)'s major health problems
(SPSS label name: mcmajhea)
Mother reporting father’s hospitalization for mental reasons
(SPSS label name: mfhosmen)
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 227 2007-WG-BX-0028
Mother reporting own hospitalization for mental reasons
(SPSS label name: mmhosmen)
Mother reporting child(ren)’s hospitalization for mental reasons
(SPSS label name: mchosmen)
Mother reporting father’s hospitalization for physical reasons
(SPSS label name: mfhosphy)
Mother reporting own hospitalization for physical reasons
(SPSS label name: mmhosphy)
Mother reporting child(ren)'s hospitalization for physical reasons
(SPSS label name: mchosphy)
Mother reporting father’s use of medicine
(SPSS label name: mfmeds)
Mother reporting own use of medicine
(SPSS label name: mmmeds)
Mother reporting child(ren)'s use of medicine
(SPSS label name: mcmeds)
Mother reporting therapy
(SPSS label name: mthpy)
Mother reporting father’s substance abuse in the past
(SPSS label name: mfsapast)
Mother reporting own substance abuse in the past
(SPSS label name: mmsapast)
Mother reporting father’s substance abuse in the present
(SPSS label name: mfsapres)
Mother reporting own substance abuse in the present
(SPSS label name: mmsapres)
Mother reporting her fears
(SPSS label name: mfears)
Mother reporting her concerns
(SPSS label name: mconcern)
Mother reporting threats
(SPSS label name: mthreat)
Mother reporting being coerced
(SPSS label name: mcoerce)
Mother reporting harassment
(SPSS label name: mharass)
Mother reporting intimidation
(SPSS label name: mintim)
Mother reporting property damage
(SPSS label name: mproprty)
Mother reporting physical abuse
(SPSS label name: mphyscl)
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 228 2007-WG-BX-0028
Mother reporting the use of weapons
(SPSS label name: mweapon)
Mother reporting sexual intimidation
(SPSS label name: msexintm)
Mother reporting sexual assault
(SPSS label name: msexaslt)
Mother reporting sexual manipulation/control
(SPSS label name: msexmnct)
Mother reporting false allegations of sexual intimidation, sex. Assualt, and man./control
(SPSS label name: msexfalg)
Mother reporting orders of protection in the present
(SPSS label name: moppres)
Mother reporting orders of protection in the past
(SPSS label name: moppast)
Mother reporting police being called
(SPSS label name: mpolcal)
Mother reporting arrests for domestic violence
(SPSS label name: marstdv)
Mother reporting either calls or arrests
(SPSS label name: mcalarr)
Mother reporting going to the hospital because of DV
(SPSS label name: mgohosp)
Mother reporting considering going to the hospital because of DV
(SPSS label name: mcnsdhsp)
Mother reporting either going or considering going to the hospital because of DV
(SPSS label name: mgocnsd)
Mother reporting going to the shelter
(SPSS label name: mshelter)
Mother reporting CPS involvement
(SPSS label name: mcps)
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 229 2007-WG-BX-0028
Attachment B: Premediation Interview Form (PMI)
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 230 2007-WG-BX-0028
Attachment C: Relationship Behavior Rating Scale (RBRS)
Copyrighted, please see first author for details on obtaining
permission to use.
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 231 2007-WG-BX-0028
Attachment D: Superior Court Divorce File Data
1. Date of Marriage
2. Mother’s DOB
3. Mother’s INS status
4. Mother’s Language
5. Mother’s Employment
6. Mother’s Income – NOTES option if income changes drastically
7. Is Mother Receiving Aid
8. Mother’s Representation
9. Number of changes in mother’s representation
10. Father’s DOB
11. Father’s INS Status
12. Father’s Language
13. Father’s Employment
14. Father’s Income – NOTES option if income changes drastically
15. Is Father Receiving Aid
16. Father’s Representation
17. Number of changes in father’s representation
18. Was there a prior or current custody issue in juvenile court re these children?
Children Information
19. Is mother pregnant?
20. Child 1 Initial
21. Child 1 Sex
22. Child 1 DOB
23. Child 1 place of residence at time of filing
24. Child 2 Initial
25. Child 2 Sex
26. Child 2 DOB
27. Child 2 place of residence at time of filing
28. Child 3 Initial
29. Child 3 Sex
30. Child 3 DOB
31. Child 3 place of residence at time of filing
32. Child 4 Initial
33. Child 4 Sex
34. Child 4 DOB
35. Child 4 place of residence at time of filing
36. Child 5 Initial
37. Child 5 Sex
38. Child 5 DOB
39. Child 5 place of residence at time of filing
40. Child 6 Initial
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 232 2007-WG-BX-0028
41. Child 6 Sex
42. Child 6 DOB
43. Child 6 place of residence at time of filing
Court Code Book Table 1
44. Case Number
45. Coder ID
Petition and Response Information
46. Date of Petition
47. Who is the Petitioner?
48. Did Petitioner Request a Confidential Address?
49. Did Petitioner mention an OOP?
50. Did Petitioner file a child affidavit?
51. Date of Response
52. Who was the Respondent?
53. Did the respondent request a confidential address?
54. Did the respondent file a child affidavit?
Requests
55. Petitioner’s Request for Legal Custody
56. Petitioner’s Request for Physical Custody
57. Did Petitioner Request to Move Out of State?
58. Respondent’s Request for Legal Custody
59. Respondent’s Request for Physical Custody
60. Did Respondent Request to Move Out of State?
61. Did Petitioner make special requests for visitation?
62. Did Respondent make special requests for visitation?
63. Did Petitioner make an allegation of DV?
64. Did Respondent make an allegation of DV?
65. Is there child abuse alleged in petition or response?
66. Did mother make any special requests for father?
67. Did father make any special request for mother?
68. Did father request a paternity test?
69. Did mother request Spousal Maintenance?
70. Did mother request payment of attorney’s fee?
71. Did mother request child support?
72. Did father request spousal maintenance?
73. Did father request payment of attorney’s fees?
74. Did father request child support?
Temporary Orders (“Pendente Lite”)
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 233 2007-WG-BX-0028
75. Case Number
76. Coder ID
Temporary Custody
77. Was there an informal interim order for temporary custody filed?
78. Did mother request temporary custody of the children?
79. Were there allegations that temporary custody be assigned
80. Did father request temporary custody of the children?
81. Were father allegations that temporary custody be assigned
82. Was temporary custody awarded?
83. Date temporary custody was awarded
84. Is the temporary custody a change from when petition was filed?
85. Was there a hearing or stipulated agreement on temporary orders?
86. Was there any review hearings ordered?
Temporary Visitation
87. Was temporary visitation requested by Mother?
88. Was temporary visitation requested by Father?
89. Was temporary visitation awarded?
90. Is anyone restricted from seeing the child?
91. Are there restrictions or specifications on drop off and pick up?
92. Are the visits supervised?
93. Are the exchanges supervised?
Temporary Child Support
94. Was temporary child support requested by mother?
95. Was temporary child support requested by father?
96. Was temporary child support ordered?
Temporary Use of Residence
97. Was temporary use of the residence requested by mother?
98. Was temporary use of the residence requested by father?
99. Was temporary use of the residence awarded?
Temporary Division of Income
100. Was temporary division of income requested by mother?
101. Was temporary division of income requested by father?
102. Was temporary division of income awarded?
Temporary Spousal Maintenance
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 234 2007-WG-BX-0028
103. Was temporary spousal maintenance requested by mother?
104. Was temporary spousal maintenance requested by father?
105. Was temporary spousal maintenance awarded?
Temporary Payment of Attorney’s Fees
106. Was temporary payment of attorney’s fees requested by mother?
107. Was temporary payment of attorney’s fees requested by father?
108. Was temporary payment of attorney’s fees awarded?
Child Abuse / Domestic Violence
109. Is child abuse alleged in temporary orders?
110. Are there allegations of DV in temporary orders?
Post Decree
Demographics
PostCaseID Case Number
MemplCh Employment Changes Mother
FemplCh Employment Changes Father
MlivesCh Address Changes Mother
FlivesCh Address Changes Father
JudgeCh Official Judge Changes
Substance Abuse
SArestr Was there restrictions set on alcohol or drugs?
Supervision of Parenting time or Exchanges
SuperEx Were there supervised exchanges?
SuperPT Were there supervised visits?
TxPT Were there therapuetic visits?
Guardian Ad Litem
GAL Was a GAL appointed?
Mediation
Mediate Did the parties go to mediation?
MedOutcom Outcome of mediation
Hearings
HearTot Number of Total Hearings
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 235 2007-WG-BX-0028
HrDenyTot Number of Denied Hearings
HrRevTot Number of Review Hearings
HrOthTot Number of Other Court Initiated Hearings
Modification of Support Requests
MModCSRq Number of Child Support Hearings Mother Requested
FModCSRq
Number of Child Support Modification Hearings Father
Requested
UknCSRq Number of Child Support Hearings Requested by Unknown
CSModOrd Number of Child Support Modifcation Orders
Enforcement of Support Requests
MEnfCSRq
Number of Child Support Enforcement Hearings Mother
Requested
FEnfCSRq
Number of Child Support Enforcement Hearings Father
Requested
StateCSRq
Number of Child Support Enforcement Hearings Requested by
the State
CSEnfOrd
Number of Order/Notices to Withhold Income for Child
Support?
Arrests for Support
ArrestCS Were arrest warrants issued For child support?
Maintenance
MModSMRq Number of Spousal Maintenance Hearings Mother Requsted
FModSMRq Number of Spousal Maintenance Hearings Father Requested
UknSMRq
Number of Spousal Maintenance Hearings Requested by
Unknown
SMOrd Number of Spousal Maintenance Orders
Modification of Custody
MModCCRq Number of Custody Hearings Mother Requested
FModCCRq Number of Custody Hearings Father Requested
UknCCRq Number of Custody Hearings Requested by Unknown
CCModOrd Number of Custody Orders
Parenting Time
MModPTRq Number of Parenting Time Hearings Mother Requested
FModPTRq Number of Parenting Time Hearings Father Requested
UknPTRq Number of Parenting Time Hearings Requested by Unknown
PTModOrd Number of Parenting Time Orders
DV/Abuse
MDVHearRq Number of DV/Abuse Hearings Mother Requested
FDVHearRq Number of DV/Abuse Hearings Father Requested
UnkDVHRq Number of DV/Abuse Hearings Requested by Unknown
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 236 2007-WG-BX-0028
DVOrders Number of orders regarding DV, Abuse or Harrassment
Orders re: Contempt
Mcontemp Number of times Mother was held in contempt
Fcontemp Number of times Father was held in contempt
Awards of Fees/Judgments
MfeeJudg Number of times Mother was awarded Fees/judgements
FfeeJudg Number of times Father was awarded fees/judgements
Other Issues
HearOthr Number of Hearings re other issues
NoteOthr Notes on Other Hearings
OrderOthr Number of Orders re Other Issues
NoteOrdOth Notes on Orders re Other Issues
Stipulations
StipCS Stipulations re Child Custody Modification
StipSM Stipulations re Spousal Maintenance
StipCC Stipulations re Child Custody
StipPT Stipulations re Parenting Time
StipOthr Stipulations re Other
StipNotes Notes Other Stipulations
Intimate Partner Violence in Mandatory Divorce Mediation
Beck, Walsh, Mechanic, Figueredo & Chen 237 2007-WG-BX-0028
Attachment E: Data Collection Sheet for Law Enforcement Database
