University of Southern California Dissertations and Theses

Prison of Peace in a jail setting: a randomized trial of a conflict resolution program for incarcerated women

(USC Thesis Other)

Prison of Peace in a jail setting: a randomized trial of a conflict resolution program for incarcerated women

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Running Head: PRISON OF PEACE RANDOMIZED TRIAL i

PRISON OF PEACE IN A JAIL SETTING:
A RANDOMIZED TRIAL OF A CONFLICT RESOLUTION PROGRAM FOR
INCARCERATED WOMEN

BY
Marie L. Gillespie, M.A.

___________________________________________________

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PSYCHOLOGY)
December 2018
PRISON OF PEACE RANDOMIZED TRIAL ii

Acknowledgements
I would like to dedicate this work to my late mother, Sophie. She was an incredible
woman who bred in me an unquenchable thirst for knowledge and sense of curiosity, and a
desire to help people whom society have deemed unworthy of compassion.
I could not have achieved a single accomplishment in my graduate journey without the
unwavering support and incredible guidance of my advisor, Dr. Stan Huey. Thank you for the
countless hours you dedicated to the development of my writing skills. Thank you for
challenging my mind, pushing me to question everything and use objective cultural lenses,
expanding my knowledge base beyond your own expertise, empowering me to effect change
within underrepresented communities, and prioritizing my wellbeing above all else. In our
current political culture, I find myself especially lucky to have a mentor who empowers his
female students to pursue family development in concert with academic achievement. I hope to
always be able to emulate your integrity and analytic ardor.
I would also like to thank my esteemed committee members: Drs. Gayla Margolin,
Richard John, and Avelardo Valdez. Your commitment to my research training has been greatly
appreciated and your expertise in various fields has undeniably strengthened every section of the
manuscript herein. Several faculty members should also be recognized for their excellent
mentorship, which has significantly shaped my career trajectory: Drs. Jasmine Tehrani, Elyn
Saks, Karen Hennigan, Gerald Davison, and Jesse Graham.
I would especially like to acknowledge the people who made this dissertation possible.
To Laurel Kaufer – thank you for the incredible woman, lawyer, teacher, and force of nature that
you are. The time and dedication you give incarcerated populations, and the perseverance you
exhibit when working within the political frameworks of correctional facilities is truly inspiring.
PRISON OF PEACE RANDOMIZED TRIAL iii

To the staff at the correctional facility– thank you for allowing me access to the women housed
at your jail, for being so kind to an outsider like me, and for helping me navigate an intricate
study design in a fast-paced environment.
To Gabrielle Lewine – thank you for your constant support from the beginning to the end
of this project. From helping at the jail, to attending every coding meetings, and providing
manuscript edits, you are the best lab-mate anyone could ask for. A warm thanks to the other
graduate students who helped me collect data at the jail: Miriam Rubenson and Brandie
Boghosian. Having you by my side provided me with enough confidence, strength, and energy
during those 12-hour days. Thank you to my undergraduate research assistants –Hailey
Konovalov and Madison Burger – for your data entry and for your keen eyes and sharp minds
when coding nearly 400 inmate responses to conflict situations. I would also like to thank Dr.
Julie Slayton and Sarah Luery at the USC IRB for helping me craft a protocol that was accepted
on the first submission.
I would especially like to acknowledge the 75 women who participated in the study.
Thank you for your genuine responses to personal questions and for entrusting me with your
information. I sincerely hope that research like this can help you lead more meaningful lives.
A thousand thanks to my cohort-mates: Kelly Miller, Justin Hummer, and Rubin
Khoddam. I wouldn’t trade the three of you for the world – I consider you to be the most brilliant
minds and kindest hearts in my life. We are forever bound by our USC roots and I have no doubt
that we will each impact the world in our own hilarious, passionate, and intellectual ways. I
would also like to acknowledge my Huey lab-mates; from those who have already graduated –
Caitlin Smith Sayegh and Eddie Jones – to those who have seen me through this dissertation
project – Gabby and Miriam, and Nina Jhaveri, Crystal Wang, and Sylvanna Vargas. Thank you
PRISON OF PEACE RANDOMIZED TRIAL iv

for helping me think through problems with myriad lenses, perspectives, and voices, and for the
laughter, friendship, and commiseration.
Finally, I would like to thank my husband, Jess, and my 18-month-old son, Luke. To
Luke, thank you for an easy pregnancy, which enabled me to collect data at the jail up until two
months before your birth, and for being such an easy baby, which allowed me to collect data
again three months after you were born. To Jess, thank you for your patience, kindness, love,
support, and belief in my abilities to balance motherhood and academics. Thank you for sticking
by the less-desirable versions of me and allowing me to write for days on end by feeding me and
taking care of our son. You are the reason I am here, living the dream and hoping to someday
change the world.

PRISON OF PEACE RANDOMIZED TRIAL v

Table of Contents
Acknowledgements ii
Table of Contents v
List of Tables vi
List of Figures vii
List of Appendices viii
Abstract i
Background and Significance 1
Method 16
Results 42
Discussion 66
Conclusion 78
References 80

PRISON OF PEACE RANDOMIZED TRIAL vi

List of Tables
Table 1: Baseline characteristics 43

Table 2: GEE Model Estimates of the Relationship Between 48
Prison of Peace and Pure Control Participants and
Outcome Variables Across Time

Table 3: GEE Model Estimates of the Relationship Between 49
Prison of Peace and Contamination Control Participants and
Outcome Variables Across Time
Table 4: GEE Model Estimates of the Relationship Between 50
Contamination Control and Pure Control Participants and
Outcome Variables Across Time
Table 5: Unadjusted Means and Standard Deviations of Significant 51
GEE Interactions of Outcomes and Conditions

Table 6: GEE Model Estimates of the Relationship Between 60
Prison of Peace Completers and Pure Control Participants and
Outcome Variables Across Time

Table 7: GEE Model Estimates of the Relationship Between 61
Prison of Peace Completers and Contamination Control Participants
and Outcome Variables Across Time
Table 8: GEE Model Estimates of the Relationship Between 62
Contamination Control and Pure Control Participants and
Outcome Variables Across Time (PoP Completers in the model)
Table 9: Binary Logistic Regression of Dichotomous Variables 66

PRISON OF PEACE RANDOMIZED TRIAL vii

List of Figures
Figure 1: CONSORT flow diagram 24
Figure 2: Distribution of T1 incident reports by condition for the full sample 46
Figure 3: Distribution of T2 incident reports by condition for the full sample 46
Figure 4: Distribution of T3 incident reports by condition for the full sample 47
Figure 5: Perceived hostility of conflict situation T1-T3 53
Figure 6: Hostility of solutions generated T1-T3 53
Figure 7: Conflict scale means T1-T3 54
Figure 8: Anger means T1-T3 55
Figure 9: Depression means T1-T3 56
Figure 10: Parenting stress means T1-T3 57
Figure 11: Incident report ratios T1-T3 58
Figure 12: Non-hostile incident report ratios T1-T3 58
Figure 13: Depression means with PoP completers T1-T3 63
Figure 14: Parenting stress means with PoP completers T1-T3 64
Figure 15: Hostile incident report ratios with PoP completers T1-T3 65
PRISON OF PEACE RANDOMIZED TRIAL viii

List of Appendices
APPENDIX A: Detailed CONSORT Flow Diagram 92

APPENDIX B: Assessment Battery Questionnaires 93

APPENDIX C: Coding Manual for Conflict Resolution Measure Vignettes 123

APPENDIX D: Conflict Resolution Measure Vignette Codes by Vignette Type 146
and Time Point

APPENDIX E: Intraclass Correlation Coefficients (ICCs) for Conflict Resolution 150
Measure Composite Scores

APPENDIX F: Arrest Descriptions at T1 for Participant Sample 151

APPENDIX G: Significant Interactions Between Missing Data 152
and Baseline Variables

APPENDIX H: Adverse Childhood Experiences (ACE) Questionnaire 153
Item Endorsement

APPENDIX I: Tests of Differences on Demographic and Correctional Data 154

APPENDIX J: Tests of Differences on Outcome Variables at Baseline 156

APPENDIX K: Arrest and Incident Report (IRTS) Counts 158

APPENDIX L: Means, Standard Deviations, and Sample Sizes 159
for Outcome Variables

APPENDIX M: Unadjusted Means and Standard Deviations of Non-Significant 161
GEE Interactions of Outcomes and Conditions

APPENDIX N: Tests of Differences Between Prison of Peace Completers (N=16) 163
and Non-completers (N=9)

PRISON OF PEACE RANDOMIZED TRIAL i

Abstract
Prisoner misconduct (e.g., rule violations, fighting) among incarcerated women leads to
significant physical, psychological, and economic costs (Steiner & Wooldredge, 2014a). Several
behavioral interventions targeting mental health problems (e.g., trauma, substance abuse) are
available for the growing population of female offenders, but very few studies use strong designs
and there are no published randomized trials of programs targeting institutional misconduct. Our
evaluation tested the efficacy of a 7-week conflict resolution program, Prison of Peace (PoP;
Kaufer & Noll, 2014), among 75 incarcerated women at a Southern California jail. Using a
quasi-experimental design, participants in a high-risk unit were randomly assigned to PoP or
Contamination Control, and inmates housed in a separate unit were randomly selected to be
participants in a Pure Control group. Inmates were assessed at pre- and post-intervention, and at
6-month follow-up. As expected, ITT analyses showed that PoP participants exhibited greater
decreases in hostile attitudes and significant increases in conflict resolution skills compared to
controls. PoP was also effective at reducing self-reported anger and depression compared to
controls at post-intervention. Unexpectedly, however, PoP participants exhibited greater
increases in official misconduct than controls at follow-up. Discussion points include the
potential iatrogenic effects of the intervention due to setting limitations (i.e., jail constraints, lack
of opportunities for skills practice), as well as possible effects of deputy attitudes on biased
reporting. Results suggest that this prison-based program, while effective at improving conflict
resolution skills and self-reports of well-being, may be ineffective at decreasing misconduct in a
jail setting.
PRISON OF PEACE RANDOMIZED TRIAL
1

Background and Significance
Prisoner misconduct has increased steadily since the early 1990s and has been equally
observed in men and women (Chamberlain, 2012; Steiner & Wooldredge, 2014a). Rising
criminogenic need (i.e., risk factors; Andrews & Bonta, 1994) among prisoners (e.g., antisocial
peers, lack of problem-solving skills, substance use) has been linked to this growth in
misconduct and, as such, it may be that inmates with the greatest unmet treatment needs are at
greatest risk for committing rule violations (Andrews, Zinger, Hoge, Bonta, Gendreau, & Cullen,
1990; Chamberlain, 2012). Thus, there is a clear need for incarceration-based programming
aimed at reducing misconduct.
Regarding female offenders, correctional interventions predominantly focus on substance
abuse, parenting skills, trauma, depression, and health (Bartlett, Jhanji, White, Harty, Scammell,
& Allen, 2015; Tripodi, Bledsoe, Kim, & Bender, 2011). However, very few programs target
misconduct that occurs during incarceration. Within the male-dominant literature, most studies
examine the effects of anger management programs on anger and various mental health problems
(Auty et al., 2017; Beck & Fernandez, 1998) and one conflict resolution program targets
misconduct (Walrath, 2001). Violence prevention program evaluations that include female
offenders either do not include measures of misconduct (Kubiak, Kim, Fedock, & Bybee, 2015)
or have been limited to small juvenile samples (Goldstein, Dovidio, Kalbeitzer, Weil, &
Strachan, 2007) or non-randomized studies (Eamon Munchua, & Reddon, 2001; Messina,
Burdon, & Prendergast, 2006). Consequently, studies with strong research designs are needed to
evaluate programs aimed at reducing misconduct for the overlooked yet growing (Zeng, 2016)
group of incarcerated women.

PRISON OF PEACE RANDOMIZED TRIAL
2

Prisoner Misconduct
Prisoner misconduct, including rule violations, theft, destruction of property, threats, and
assault, has been identified as a key index of institutional maladjustment (Steiner & Wooldredge,
2014b; Van Voorhis, Wright, Salisbury, & Bauman, 2010). Self-report data indicate that close to
16% of correctional officers have been assaulted by inmates (Duhart, 2001) and that
approximately 20% of prison inmates are victims of violence (Schenk & Fremouw, 2012; Wolff,
Blitz, Shi, Siegel, & Bachman, 2007). Prisoner misconduct has costs that go beyond physical
injuries and loss of privileges for inmates. High conflict also impacts correctional officers’ well-
being, safety, stress, and turnover rates, which in turn increases institutional expenditures
(Welsh, McGrain, Salamatin, & Zajac, 2007). Moreover, inmates involved in misconduct while
incarcerated are more likely to recidivate and return to prison or jail (Trulson, DeLisi, &
Marquart, 2011).
Rates of misconduct at the county jail level are rarely researched or made available to the
public, but it is estimated that serious infractions (e.g., fighting) occur less frequently in county
jails than state facilities (Austin, Naro-Ware, Ocker, Harris, & Allen, 2012). Nonetheless, these
dangerous, costly, and pervasive problems are ripe for research and intervention development.
Furthermore, female offenders are mostly excluded from this literature, and thus the
generalizability of prison violence findings has come into question (Steiner & Wooldredge,
2014b).
Incarcerated Women
In Los Angeles County, approximately 13% of the jail population is female (California
Sentencing Institute, 2016), which is similar to the U.S. average (i.e., 14.5%; Zeng, 2016).
National averages indicate that the male jail population decreased by 3% between 2010 and
PRISON OF PEACE RANDOMIZED TRIAL
3

2014, while the female population increased approximately 18% (Minton & Zeng 2015). This
growth in the female inmate population began in the 1980s and has largely been attributed to
policy changes related to the increased criminalization of drug-related behavior. Indeed, one
third to one half of incarcerated women are serving time for drug-related crimes or were under
the influence of drugs at the time of their arrest (Messina et al., 2006). Additionally, the passing
of the Public Safety Realignment Act in 2011 (Assembly Bill 109; AB 109 Cal. Stat.)
significantly increased the overall jail population in California (Zeng, 2016), as it mandated that
offenders sentenced to non-serious and non-violent crimes be incarcerated in county jails instead
of state prisons. Accordingly, the number of female offenders housed in California jails greatly
increased.
Unfortunately, there is a dearth of literature on jail-based misconduct among female
inmates. One study examining the predictive validity of a violence-risk measure on future jail
misconduct reported no gender differences in rates of misconduct when controlling for number
of days incarcerated (Hastings, Krishnan, Tangney, & Stuewig, 2011). With regard to official
reports of prison misconduct, findings are mixed when examining gender differences (Cao,
Zhao, & Van Dine, 1997; Craddock, 1996; Marcum, Hilinski-rosick, & Freiburger, 2014; Pierce,
Freiburger, Chapin, Epling, & Madden, 2018). However, some recent research suggests that
male and female inmates commit both violent and non-violent rule violations at similar rates
(Steiner & Wooldredge, 2014a) or even that women commit higher rates of overall misconduct
(Marcum et al., 2014). Despite the pervasive exclusion of female prisoners from intervention
studies on misconduct and aggression, correlational researchers have identified several factors
associated with women’s incarceration-based offending. Imprisoned women who are young,
non-white, and have lower levels of education have the highest rates of institutional infractions
PRISON OF PEACE RANDOMIZED TRIAL
4

(Gover, Perez, & Jennings, 2008; Sorenson, Wrinkle, & Gutierrez, 1998). Further, research by
Wright, Salisbury, and Van Voorhis (2007) on the gender-responsive treatment needs of female
offenders shows that anxiety, depression, trauma history, and unsupportive relationships are
highly correlated with misconduct. Lastly, women’s length of stay has been found to be
differentially predictive of institutional infractions compared to men, indicating that long-term
female inmates violate rules at a higher rate than short-stay offenders (Casey-Acevedo &
Bakken, 2001; Gover et al., 2008; Pierce et al., 2018; Thompson & Loper, 2005).
The latter finding may indicate that women in prison have greater difficulty adjusting to
extended periods of incarceration than men. Indeed, female offenders are more likely to
experience mental health problems, particularly depression and anxiety, compared to their male
counterparts, both in prison (73% vs. 55%, respectively) and in jail (75% vs. 63%, respectively;
James & Glaze, 2006). Women incarcerated in jails are more likely to be unemployed, using
drugs, or receiving public assistance at the time they were arrested compared to males sent to jail
(Swavola, Riley, & Subramanian, 2016). Additionally, the majority of female offenders have
minor children that depend on them; the number of children with mothers in prison has more
than doubled since 1991, a growth of 131% for women compared to 77% for men (Glaze &
Maruschak, 2010). Separations from children likely exacerbate parental stress and psychological
distress for incarcerated women (Gover et al., 2008; Loper & Tuerk, 2011; Steiner &
Wooldredge, 2014b). The unique stressors and needs of female offenders, which likely
contribute to institutional misconduct and later recidivism, suggest that evidence-based programs
may be particularly crucial for this growing incarcerated population (Wright et al., 2007).

PRISON OF PEACE RANDOMIZED TRIAL
5

Correctional-Based Interventions for Offenders
In a meta-analysis that included 58 interventions for offenders, Landenberger and Lipsey
(2005) found that CBT-based programs were effective at reducing recidivism, especially those
that included elements of interpersonal problem solving, anger control, victim impact, and
behavior modification. Additional reviews presented similar positive effects of such programs
(Lipsey & Cullen, 2007; Pearson, Lipton, Cleland, & Yee. 2002), particularly for behavioral
interventions that addressed the criminogenic needs of higher risk offenders (Andrews & Bonta,
2003). The development of incarceration-based programs has steadily increased over the past
decade (Chamberlain, 2012) as policy makers attempt to bridge the gap between research and
practice (Gendreau & Keyes, 2001). Indeed, federal incentives have been offered to ensure that
state- and county-level institutions adopt evidence-based rehabilitation practices for the
imprisoned population (Feinstein, 2011).
Empirically-based criminogenic factors are the focal point of many interventions,
whereby program developers aim to target factors that have been commonly linked to recidivism
(Sorbello, Eccleston, Ward, & Jones, 2002; Tripodi et al., 2011). Given the aforementioned
statistics on incarceration rates for substance-using women, it is not surprising that substance use
is a critical variable linked to women’s criminal behavior, and therefore a common intervention
target. Interestingly, the primary outcome of interest in most prison-based substance abuse
intervention programs is reincarceration rather than ongoing sobriety (Tripodi et al., 2011). Other
programs target offenders’ psychological, psychosocial, and physical well-being, and aim to
improve overall functioning during and after incarceration (Ward & Stewart, 2003). Although
reduced re-offending may be a collateral gain from programs that target depression, anger, self-
esteem, sexual abuse trauma, and health risks, the primary goal is to improve current institutional
PRISON OF PEACE RANDOMIZED TRIAL
6

adjustment and general welfare. Therefore, the main outcomes measured in these programs are
typically variables related to mental health, life-skill development, and substance use (Tripodi et
al., 2011).
In recent years, the emergence of parenting programs has also been observed in prisons
and jails (Green, Miranda, Daroowalla, & Siddique, 2005; Kennon, Mackintosh, & Meyers,
2009). Parenting interventions aim to reduce parenting stress and institutional violations, and
improve emotion regulation and communication skills, with programs developed by Loper and
colleagues showing promising results in these domains (Houck & Loper, 2002; Loper, Carlson,
Levitt, & Scheffel, 2009; Loper, Thompson, Tuerk, Daoust, Kramer, & Komarovskaya, 2007;
Loper & Tuerk, 2011). Such interventions also aim to enhance parental motivations to refrain
from criminal behavior by strengthening family bonds and environmental supports in preparation
for community re-entry (Gonzalez, Romero, & Cerbana, 2007). Researchers have stressed the
need for evaluators to comprehensively assess the full impact of correctional programs by
including measures of psychological functioning and institutional behaviors (Tripodi et al.,
2011).
Interventions for female offenders. A recent systematic review and meta-analysis by
Bartlett and colleagues (2015) focused on mental health interventions for female offenders. The
researchers included international studies and programs conducted in both community and
correctional settings. Most studies included substance-abuse and trauma-focused intervention
programs. Medium effect sizes were found for interventions targeting depression and post-
traumatic symptoms, but no significant effects were found for global assessments of mental
health. Bartlett’s work highlights the need for robust intervention research methods: only 11 of
the 17 studies included in the review were eligible for the meta-analysis due to lack of
PRISON OF PEACE RANDOMIZED TRIAL
7

comparison groups, and very few were randomized trials. Additional barriers to effective
evaluations were small study samples and lack of follow-up measures (Bartlett et al., 2015).
These represent persistent challenges in the research literature on prison-based interventions
(French & Gendreau, 2006; Gendreau, Goggin, & Cullen, 1999).
Tripodi and colleagues (2011) broadened the scope of their review by including all types
of interventions for female offenders. However, they narrowed their search to U.S.-based studies
conducted after 1988 in correctional settings, thereby focusing their efforts on programs
developed after the dramatic increase in female incarcerations. They included 24 studies, with
the majority focused on mental health interventions, and a few programs targeting HIV
prevention or parenting skills. Analyzing studies aimed at reducing recidivism (n=6), the authors
found that women who participated in substance-abuse treatment programs were 45% less likely
to re-offend after their release from prison compared to women who did not participate in these
treatments. For studies aiming to improve inmate well-being, CBT, psychoeducational, and
trauma-based programs affected the greatest change in self-reported psychological functioning
(i.e., depression and anxiety) and hostility. However, and similarly to the findings by Bartlett et
al. (2015), most interventions did not include comparison groups (Tripodi et al., 2011).
Interventions for Prisoner Misconduct
Several studies have examined the effects of incarceration-based interventions on
prisoner adjustment and related misconduct (Auty, Cope, & Liebling, 2017; French & Gendreau,
2006; Welsh et al., 2007), with most results showing reductions in disruptive institutional
behavior and recidivism (Prendergast, Hall, Wexler, Melnick, & Yao, 2004; Wilson, Gallagher,
& MacKenzie, 2000). As such, the most common management recommendation for the
maintenance of safety and stability in prisons and jails is the broad implementation of treatment
PRISON OF PEACE RANDOMIZED TRIAL
8

programs (Gendreau & Keyes, 2001). French and Gendreau (2006) examined 68 studies that
used randomized or comparison control groups in their meta-analysis of prison-based treatments,
which included educational, vocational, behavioral, and therapeutic programs targeting various
problems. The authors found that behavioral interventions were generally effective at reducing
institutional misconduct in addition to the intervention’s focal problem. Interestingly, many of
these programs did not target aggression or misconduct directly. The programs that were most
effective at curtailing disruptive behavior were also associated with greater reductions in re-
offending once prisoners were released. Consistent with other meta-analytic and systematic
reviews, French and Gendreau (2006) noted the paucity of robust methodology in the broad
intervention literature, and they emphasized the need for on-site program evaluations that utilize
randomization procedures and a comparison group.
With regard to interventions that specifically target institutional violence, CBT-based
anger management programs for male offenders are most commonly found in the literature.
While many used randomization procedures and comparison groups, the majority of outcome
measures were limited to post-intervention self-reports of aggression and anger or did not
include official records of prisoner misconduct (Gaertner, 1984; Napolitano, 1992; Smith &
Beckner, 1993; Stermac, 1986). Nonetheless, most studies reported significant treatment gains
for participants that completed anger management programs, with some studies showing
reductions in observed aggressive behavior or decreases in official incident reports (Beck &
Fernandez, 1998; McDougall & Boddis, 1991).
Auty and colleagues (2017) recently conducted a systematic review of violence-reduction
programs in forensic settings (e.g., prisons, halfway houses, hospitals). Their findings, based on
21 randomized and non-randomized studies published in the last 20 years, generally mirror
PRISON OF PEACE RANDOMIZED TRIAL
9

previous research showing that CBT approaches are most effective. The authors noted an
increase in the sophistication of statistical methods used and reported mixed results for the
efficacy of the programs. Most interventions used life skills, anger management, substance abuse
treatment, and social learning principles. Only three randomized trials included official incident
reports from correctional institutions to measure misconduct (Armstrong, 2002; Hogan, Lambert,
& Barton-Bellessa, 2012; Lambert, Hogan, Barton, and Stevenson, 2007). Relevant to the current
evaluation, Auty et al. (2017) highlighted the need for gender-specific interventions and greater
female inclusion.
Two incarceration-based conflict resolution programs were found in the literature and
both used a peer-facilitated intervention model. The Insight Prison Project (IPP; Silva &
Hartney, 2012) was implemented and evaluated with 45 inmates serving life sentences at San
Quentin State Prison in California. Due to restrictions by the California Department of
Corrections and Rehabilitation (CDCR) at the time, researchers did not obtain approval to use
comparison groups or pre-post testing, or to collect criminal history data; therefore, quantitative
findings were very limited. IPP included a 24-week CBT-based violence prevention course in
which peer instructors (i.e., inmates who previously completed the course) co-facilitated classes
and mentored newer inmates. Based on self-report measures and interviews, participants who
remained in the program longer than six months exhibited more positive outcomes than
participants who dropped out prior to six months. Moreover, longer program participation was
associated with improvements in self-esteem, aggression, hopelessness, problem-solving skills,
social support, and increases in positive, non-aggressive relationships (Silva & Hartney, 2012).
Walrath (2001) developed a program that most closely resembles the Prison of Peace
model in our current study. The Alternatives to Violence Project (AVP) is an inmate-run
PRISON OF PEACE RANDOMIZED TRIAL
10

intervention that teaches offenders to take responsibility for their actions and to find alternative
options to fight-or-flight responses to conflict through interactive activities. The program
consists of 15 day-long group sessions, whereby inmates receive conflict resolution training and
eventually become trainers themselves. Although AVP has been implemented in various
correctional facilities since the 1980s, Walrath’s study was the first systematic evaluation of this
intervention. Fifty-three male non-randomized inmates participated in AVP and 41 general
population inmates were used as a comparison group at a medium-security correctional facility
in Maryland. At 6-month follow-up, AVP participants showed significant reductions in anger,
and self-reported violent and non-violent confrontations compared to controls (Walrath, 2001).
This study provides some support for the effectiveness of a peer inmate-facilitated conflict
resolution program. However, a gap in the literature remains with respect to female offenders
and programs that measure both recidivism and archival records of prisoner misconduct.
Interventions for female misconduct. To the best of our knowledge, there are no
published, randomized controlled trials of incarceration-based programs specifically aimed at
reducing institutional misconduct for adult female offenders. However, one randomized trial
examined the mental health impact of Beyond Violence (BV; Covington, 2013), an intervention
program for violent female prisoners (Kubiak et al., 2015). BV was specifically developed for
the unique needs of female inmates and addresses the interplay between violence, trauma,
substance abuse, and personal relationships. The authors randomized 22 inmates to the 20-
session BV program or to a treatment-as-usual intervention for violent offenders at the facility.
Although women in both programs improved on mental health domains, BV participants
exhibited greater decreases in self-reported anger, aggression, and anxiety at post-intervention.
Unfortunately, Kubiak and colleagues did not collect archival data and did not assess for
PRISON OF PEACE RANDOMIZED TRIAL
11

misconduct. The BV program was subsequently modified to include peer educators and was
examined with 62 women serving long-term sentences for violent crimes in two California
prisons (Messina et al., 2016). Results generally replicated the Kubiak et al. (2015) findings and
importantly highlighted the cost-effectiveness of using peer teachers. However, no comparison
group was used in the study and, again, misconduct was not measured.
Two additional studies examining the effectiveness of violence-reduction programs for
female inmates were found in the literature. Goldstein and colleagues (2007) conducted a pilot
study of an anger management program for 12 girls in a juvenile justice facility. Participants
were randomized to the 18-session program or to treatment as usual, and five participants
completed post-intervention assessments. Medium to large effect sizes were found in the
treatment group for pre- to post-intervention self-reported anger and aggression, but there were
no statistically significant differences between groups. Similarly, Eamon and colleagues (2001)
used a comparison group in their evaluation of an anger-management program for 33 female
inmates in a Canadian prison. The CBT-based 12-week program was more effective than control
at reducing the number of institutional charges at post-intervention. Although the small sample
of participants was not randomized to conditions, this is one of the only published evaluations to
include archival records of prison misconduct for female offenders.
In their study examining misconduct risk for 272 incarcerated women, Wright and
colleagues (2007) highlighted the importance of considering additional “gender-responsive”
needs, including relationships, self-efficacy, parenting, and communication, when placing
inmates into custody classifications (i.e., risk level and program needs). The authors found that
high levels of relationship conflict, miscommunication, and dysfunction, both with other inmates
and non-incarcerated family and friends, led to institutional misconduct over time. This is
PRISON OF PEACE RANDOMIZED TRIAL
12

consistent with longstanding research showing that female distress, conflict, and aggression are
particularly influenced by interpersonal relationship difficulties (Gilligan, 1986; Miller, 1976;
Wright et al., 2007). Therefore, interventions that aim to strengthen female offenders’
interpersonal relationships, communication skills, and conflict-resolution problems may be
effective at curtailing incidents of institutional misconduct.
Background on Prison of Peace
Prison of Peace (PoP) was developed in 2010 by Laurel Kaufer and Doug Noll, both
lawyers and professional mediators, in response to requests for a conflict resolution program
from inmates serving life sentences at Valley State Prison for Women (VSPW) in Chowchilla,
California. Kaufer, Noll, and Mayer (2014) describe the conceptualization of PoP through a
Prisoner Facilitated Mediation (PFM) model, whereby inmates are trained to resolve conflicts
and serve as neutral mediators between two parties. As with more traditional school-based peer-
mediation interventions (Kaya, Blake, & Chan, 2015), inmates are assumed to benefit from PFM
when they mediate or are mediated by peers that can relate to their problems and environmental
stressors.
Kaufer and colleagues cited work by Phillips and Cooney (2005) on inmate conflict,
which showed that prison fights became violent less frequently when untrained third-parties
intervened in a nonpartisan manner (20%) instead of a biased manner (77%). The method of
training inmates to non-violently resolve conflicts without staff intervention can empower
offenders to become less impulsive and more self-efficacious individuals. PoP enables inmates to
improve their prison cultures by teaching them empathy, communication, relationship, and
conflict-resolution skills. The program developers manualized their intervention using Essential
Problem Solving Skill principles drawn from the book People Skills (Bolton, 1986),
PRISON OF PEACE RANDOMIZED TRIAL
13

neuroscience research on empathy and emotions by Lieberman and colleagues (2007), as well as
decades of their own personal mediation experience (Kaufer et al., 2014).
Training program: levels. At the time of the current study (August 2016), PoP’s
training program consisted of two workshops, Peacemaker and Mediator, and four progressive
levels within each workshop (Student, Mentor, Rising Trainer, and Trainer). Interactive
workshops are implemented in group sessions of 15 to 25 inmates. First, the Peacemaker
workshop teaches offenders basic listening, communication, and personal problem-solving skills,
which help inmates identify and manage strong emotions (their own and those of others). As
“homework,” participants are required to engage in conversations with others outside the
classroom, during which they practice PoP skills and write about their experiences. They are also
required to organize and lead a certain number of “Peace Circles,” in which they conduct highly
structured group discussions with other inmates in their incarcerated community. This first
Student level, led by PoP trainers, includes a minimum of 32 hours of classroom time over the
course of eight to 18 weeks, depending on institutional scheduling opportunities.
Second, the Mediator workshop provides training in basic mediation skills, through
which inmates learn techniques for engaging as impartial intermediaries in conflict situations.
Participants prepare, organize, and conduct mediation sessions within their living spaces as
homework assignments. Focus is placed on identifying and de-escalating strong emotion, the
sharing of each participant’s personal experience of the conflict, identifying the interests of the
parties, and guiding them in identifying their own options for mutually acceptable conflict
resolution. This introduces them to a process for rationally interpreting behaviors, learning to
make goal-directed decisions, providing choices to the parties involved, and maintaining
confidentiality within mediation sessions.
PRISON OF PEACE RANDOMIZED TRIAL
14

The Mentor level of each workshop enables those who have successfully completed that
course, whether Peacemaker or Mediator, to guide new students and assist trainers in conducting
classes. In addition to serving as peer-mentors, inmates at the Mentor level are encouraged to
constantly refresh their own skills (e.g., review materials, repeat homework assignments, and
conduct more Peace Circles). After successful completion, inmates are eligible to become a
Rising Trainer and subsequently indicate whether they are interested in becoming licensed PoP
trainers. This level is comprised of an intensive training program, followed by the team-teaching
of three cohorts, in which inmates are observed by program developers. During this stage, Rising
Trainers lead their own Peacemaker or Mediator courses, assist individuals at the Mentor level,
review all written assignments, and manage all in-session activities.
The entire PoP program, from Peacemaker through Mediator and Trainer certification,
can be completed in as little as 18 months and has taken as long as four years depending on
various factors (e.g., administrative support, space availability, scheduling opportunities).
Through this progressive program structure, PoP is designed to become self-sustaining within a
correctional setting. The current study examined the effects of the Student level of the first
Peacemaker workshop.
Impact. To date, PoP developers have directly trained over 1,000 inmates in 11
correctional facilities in California (Laurel Kaufer personal communication, May 14, 2018). The
Department of Corrections is currently funding the program in six institutions from April 2017
through October 2019. Due to the sustainability model of this program, over 20,000 inmates
have been impacted by PoP, either through enrollment in the peer-led courses or through
participation in Peace Circles and mediations led by PoP students and alumni. Kaufer and Noll
conducted qualitative evaluations of the program using pre- and post-intervention surveys with
PRISON OF PEACE RANDOMIZED TRIAL
15

PoP inmates. Results indicated that participants showed improvements in communication,
listening, problem-solving skills, and managing strong emotions. Moreover, most inmates
reported positive changes in their overall prison environments (Kaufer et al., 2014).
In addition to improvements in prison climate, PoP has the potential to impact other life
domains. For instance, the skills taught in PoP could enhance performance and engagement in
other jail programs. Indeed, some studies have shown that enrollment in mediation or conflict
resolution programs versus no participation led to higher achievements in other academic areas
(e.g., Stevahn, Johnson, Johnson, & Schultz, 2002). While these results seem promising, the
comprehensive effects of PoP have yet to be empirically and systematically evaluated with
regard to prisoner misconduct, re-arrest, and collateral benefits.
Current Study
This study is the first randomized trial testing the efficacy of a conflict resolution
program for incarcerated women in a jail setting. We hypothesized that participants enrolled in
Prison of Peace would exhibit greater reductions in institutional misconduct and re-arrest rates
and greater improvements in conflict resolution skills than participants in the two control groups.
Additionally, we hypothesized that PoP would provide collateral benefits for program
participants, such as improvements in psychological functioning, interpersonal relationships, and
emotion regulation. Lastly, we expected that control participants housed in the same module as
PoP enrollees would vicariously benefit from the program through their involvement in Peace
Circles compared to control participants housed in a separate module.

PRISON OF PEACE RANDOMIZED TRIAL
16

Method
Evaluation Setting
This was the first implementation of Prison of Peace in a jail and the first formal
evaluation of the program. Although PoP was initially designed for and implemented with
women serving life in prison, the program expanded over the last few years to include inmates
housed at an all-female jail in Southern California. Given institutional flexibility, access, and
approval, this female jail population was used in the current study. The facility utilized a gender-
responsive model and provided several programs, including academic, employment, substance
abuse, parenting, life skills, CBT-based mental health, and health programs. Although these
interventions target many important life domains, PoP was the first manualized conflict
resolution program offered at this facility.
According to jail officials, there were close to 1,952 women incarcerated at the time of
recruitment and 214 official incidents of inmate-on-inmate violence from January to September
2015 (personal communication, October 26, 2015). Moreover, the monetary cost of officer use of
force, which is often necessary in situations that involve misconduct, totaled $178,600 over the
same 8-month period. With nearly six violent incidents occurring every week, staff were eager to
accommodate research initiatives aimed at evaluating a potentially impactful program.
Design and Procedures
All procedures for Human Subjects research were approved in May 2016 by the
institutional review board at the University of Southern California (USC). Discussions of study
procedures with jail personnel occurred between July 2015 and July 2016, and official approval
was provided in July 2016 by the jail facility’s research committee.

PRISON OF PEACE RANDOMIZED TRIAL
17

Housing modules. Jail officials agreed to freeze all movement of study participants for
the duration of the 7-week intervention to reduce contamination between groups. Researchers
were provided access to the two high-risk, long-term (e.g., AB-109 inmates) housing modules
(Mod A and Mod B). “Inmate security level” was a designation provided by jail officials and
indicated risk of violence and/or escape for each inmate (i.e., 1=low risk to 8=high risk). Our two
modules housed offenders who had the highest security levels at the jail and were not eligible to
receive programming. Despite these similarities, the modules differed structurally. Mod A was
often referred to as “the worst of the worst” by jail staff, and deputies kept the 96 inmates on
lockdown in their cells for 23 hours a day. They were allowed to “program” (i.e., eat, use the
phone, shower, walk in an enclosed space with a window, or order commissary
1
) in small groups
for one hour a day. One deputy managed the module and was rotated out approximately every
other day. Mod B housed 110 high-risk inmates at the time of recruitment and they generally had
more freedom of movement within their module. Due to overcrowding, day beds were placed in
the common area and two deputies were simultaneously on duty. Both housing units had access
to a 25-person classroom and jail officials agreed to random assignment of the experimental
module between Mod A and B.
Recruitment. All women housed in the aforementioned housing units (N=180 present at
recruitment) were asked to sit in the common area of their modules and voluntarily listen to an
announcement describing Prison of Peace. Given that all inmates could not be out of their cells at
the same time for safety purposes, scripted recruitment speeches were made to several groups.
Three female advanced graduate students completed the recruitment process. The purpose of the

1
Commissary in this context refers to a store within the correctional facility from which inmates can use money
from their jail accounts to purchase goods (e.g., hygiene products, food, drinks, supplies, etc.).
PRISON OF PEACE RANDOMIZED TRIAL
18

study was introduced to the inmates and randomization procedures were explained; it was
emphasized that each participant would have an equal chance of being assigned to PoP or control
groups. Consent, questionnaires, and confidentiality were also explained to the inmates. To
ensure that PoP involvement was the primary motivation for study participation, the monetary
incentive (i.e., $10 per assessment battery) was not mentioned during the recruitment speech.
The following summary statement was made:
To recap, we’re doing this study to understand the effects of Prison of Peace. If you want to
be in the study, you’re agreeing to fill out questionnaires three different times over the
period of about nine months. If you agree to participate, you agree to be randomly assigned
to Prison of Peace or not, like flipping a coin with an equal chance of being selected. All
your information will be kept completely secret and no one from the jail will have access to
your answers. Your name won’t be on any of the questionnaires. You can also choose to
give us access to your arrest records as a separate agreement.
The co-founder/CEO and co-developer of the PoP program, Laurel Kaufer, was also the
program instructor for the current study. She subsequently presented video testimonials
2
from
PoP graduates and described the details of the intervention program. A PoP graduate housed at
the jail was assigned as Ms. Kaufer’s assistant and discussed her own experience with the
program. Individuals who were interested in participating in the study were asked to read an
informational flier and to indicate that they met all inclusion criteria: (1) fluently read, speak, and
write in English; and (2) be present for the duration of the program (i.e., not scheduled to be in
trial, released or transferred to prison; agree to remain in module). Names, booking numbers, and
signatures were collected on sign-up sheets. Copies of the consent form were handed out and
inmates interested in the program were asked to carefully read the document in their cells while
waiting to be called to individually meet with researchers.

2
See http://www.prisonofpeace.org/
PRISON OF PEACE RANDOMIZED TRIAL
19

Prior to consent procedures, booking numbers were run through the jail computer system
to verify that interested inmates met inclusion criterion number two. After 14 inmates were
screened out, 53 sign-up sheets were counted for Mod A and 38 for Mod B. Because not enough
interest was gathered in the latter module, randomization was compromised for the selection of
the experimental module and Mod A was chosen to receive the intervention program. The three
researchers then individually met with interested inmates and consent procedures were
completed the following day.
Consent. Prior to signing the consent form, inmates were asked to indicate their
understanding of the study procedures (i.e., through general discussion or by going over the main
points of each section of the consent form) and were encouraged to ask questions about the
program and/or the research components. In addition to signing the consent form, participants
were asked to sign an optional release of information to allow researchers access to their official
records. Participants had to sign/initial on three separate lines to allow access to their (1) booking
numbers that would enable us to track them through the jail system, (2) past and future incident
reports that would enable us to assess misconduct at the jail, and (3) past and future arrest
records from the jail database for the duration of the study. It was emphasized that enrollment in
the study would not be conditional on signing this separate release and that the program
instructor would not know whether or not access to archival records had been provided. A data-
log template displaying only assessment numbers and coded variable names was provided to
inmates who asked how their information would be retained.
Random assignment. After consent was obtained, random assignment was conducted.
A fourth researcher who had no contact with the jail used a random number generator and
followed a two-step process. First, 25 of the 38 interested inmates in Mod B were randomly
PRISON OF PEACE RANDOMIZED TRIAL
20

selected to create the Pure Control group. Second, for Mod A, 50 inmates were randomly
selected from the group of 53 sign-up sheets and participants were randomly assigned to PoP or
Contamination Control groups. Thus, the three study conditions included: 1) Prison of Peace
intervention (PoP), 2) Pure Control (PC), and 3) Contamination Control (CC). Because PoP
participants are required to conduct weekly “Peace Circles” whereby they engage in mediation
with other inmates on their unit, we expected strong crossover effects among Contamination
Controls housed in the same module as PoP participants, but no crossover effects among Pure
Controls housed in a separate module. Group assignments (PoP, CC, or PC) were placed in a
sealed envelope that would be provided to participants directly following pre-intervention
assessments. This enabled researchers at the jail to remain blinded vis à vis participant condition.
The 16 women who did not get randomly selected for the study (three from A and 13 from B)
were individually notified.
Our consent form indicated that inmates who were not selected to receive the program
may have an opportunity to participate in PoP after the study was completed, provided that the
program was still being offered at the jail. It further indicated that Ms. Kaufer would have the list
of names of women interested in PoP and would give them priority in signing up if they were
still being housed at the facility and interested. Since the study completion, however, Ms. Kaufer
has not returned to this jail to provide the program. Given the tentative language in our consent
form and given that the researchers had no bearing on Ms. Kaufer’s future decisions to
implement her program at the facility, Contamination Control and Pure Control participants were
not considered “wait-list” but rather “treatment as usual” controls.
Assessments procedures. Assessment batteries were completed the week prior to the
intervention start date (T1), one week following intervention completion (i.e., approximately 8
PRISON OF PEACE RANDOMIZED TRIAL
21

weeks later; T2), and 6 months following intervention completion (T3). For participants housed
at the jail at the time, T1 assessments were administered in three separate groups given the
limitations of classroom space, and T2 and T3 assessments were administered in several small
groups or individually at the jail depending on the location of the inmates within the facility.
Three researchers were present at all times to answer questions, hand out deidentified assessment
packets to the appropriate inmates, and provide participants with their assignment envelopes and
notice-of-payment envelopes. Administration was standardized and scripted directions were
provided in the same way to all groups of participants. The women were encouraged to raise
their hands to ask questions, for example if they needed words clarified. Participants were told
they could refuse to complete any of the assessments without consequences from jail officials.
Participants were asked to fill out a “community contact sheet” to provide us with
addresses and phone numbers in the event of early release. Self-report questionnaires took
approximately 70 minutes to complete and included demographic and personal characteristics,
arrest history, conflict resolution and problem-solving skills, psychological functioning, self-
efficacy, friend and family relationships and parenting stress, perspective taking, empathy, and
anger. Directly following data collection at each time point, $10 cash payments were made to
each participant’s inmate account at the jail cashier’s office. Receipts were promptly deidentified
and all data was transported back to USC in a locked portable filing case.
Variations in assessments procedures and attrition. Several deviations were made from
the primary group assessment procedures, most of which occurred at T2 and T3. One elderly
participant was verbally administered the assessments at T1 and T2 (released before T3) due to
poor graphomotor skills/writing ability. Reasons for participants not completing assessments at
the jail included long-term solitary confinement (n=2 at T2), refusal to participate (n=2 at T3),
PRISON OF PEACE RANDOMIZED TRIAL
22

and hostile behavior/refusal from jail staff (n=1 at T3). When participants were not present on
the day of the assessments (i.e., due to court dates, illness, short-term solitary confinement), a
researcher returned to the jail within one week to administer individual batteries. An evaluator
was allowed to administer an assessment to one participant in solitary confinement at T3.
Participants who had been released or transferred to prison were contacted via phone and
mailings. At post-intervention, all but one participant who had been transferred to prison
completed and returned the assessment battery (n=9). Of the 12 participants released by T2,
seven were reached by phone and mailed assessments, and five were unreachable (e.g.,
disconnected phones; voicemails never returned); only one mailed her assessment back.
Participants at the jail completed T2 assessments on 10/07/2016 and, with the exception of two
participants, most assessments mailed out were sent between 10/04/2016 and 10/20/2016 and
were completed within two weeks of the mailing date. Given that it took some time to locate and
speak with released participants, mailings to those women were sent out at a later date. Payments
were made through electronic transfers using jpay.com (prison) and cash mailed out (released)
the day assessments were received in the mail, and a follow-up letter was sent thanking the
participants for their time. In one instance, a participant was transferred to an out-of-state prison
with stricter guidelines regarding mailings to inmates. A chaplain at the prison was contacted and
agreed to transfer the mailing to the inmate; the completed assessment was returned within two
weeks.
At the 6-month follow-up, 11 of the 17 participants in prison returned their completed
assessments. Of the 31 released participants, two of whom had been transferred to prison and
subsequently released to the community, 13 were reached by phone and mailed assessments, and
18 were unreachable (i.e., disconnected phones, phone number of family member who was not in
PRISON OF PEACE RANDOMIZED TRIAL
23

contact with participant, messages left but no returned calls, no phone number provided, no reply
to email/mailing); seven mailed back their completed assessments. Payment procedures were
similar to those at T2. Participants at the jail completed T3 assessments on 04/07/2017. For those
no longer housed at the facility, most assessments mailed out were sent between 04/03/2017 and
04/19/2017 and were completed within three weeks of the mailing date.
A Consolidated Standards of Reporting (CONSORT) flow diagram of each step of the
study procedures, and participant attrition data, are presented below (Figure 1). A more detailed
flow diagram, including reasons for incomplete assessments, can be found in Appendix A.

PRISON OF PEACE RANDOMIZED TRIAL
24

Figure 1: CONSORT flow diagram

Attended Recruitment Presentation
Mod A: (n=80)
Mod B: (n=100)
Excluded (n= 89)
 Did not meet inclusion criteria (n= 14)
 Declined to participate (n= 75)
Analyzed (n= 75)

Completed T3 Assessments (n= 13)

Completed T3 Assessments
 Mod A (n= 14)
 Mod B (n= 12)

Allocated to Control Groups (n= 50)
Completed T1 Assessments
 Mod A (n= 22)
 Mod B (n= 25)

Allocation & T1
T 3 Follow-Up
Enrollment
Signed Consent Forms
 Mod A (n= 53)
 Mod B (n= 38)
Completed T2 Assessments
 Mod A (n= 19)
 Mod B (n= 20)

T2 Post-Intervention
Analysis
Randomly Selected & Randomized (n= 75)
Not Selected for the study (n=16)
 Mod A:
Prison of Peace (n=25)
Contamination Control (n=25)
 Mod B
Pure Control (n=25)

Completed T2 Assessments (n= 20)

Allocated to Prison of Peace (n= 25)
Completed T1 Assessments (n= 24)
 Completed the program (n= 16)
 Did not complete the program (n=9)

PRISON OF PEACE RANDOMIZED TRIAL
25

Confidentiality. Each assessment page and each folder containing the assessment
packages were labeled with a unique assessment number for each inmate. The list of participants
and corresponding assessment numbers were kept in a locked file cabinet in a locked office and
in a password-protected electronic document. This list was only removed from the university
when researchers visited the jail to collect assessment data. All computer files with identifiable
data were password protected. A certificate of confidentiality was obtained from the U.S.
Department of Health and Human Services in May 2016 (CC-MH-16-203; exp. 12/31/2019) to
protect participants who report engaging in criminal behavior or illegal substance use; therefore,
any incriminating information collected as part of this proposed study is not subject to subpoena
in court. In addition, researchers did not inform jail staff about individual inmates’ decisions to
enroll in the study, their responses to questionnaires, or the condition to which they were
assigned. However, it should be noted that deputies had the responsibility to transfer inmates
from their cells to the PoP classroom, and therefore were likely to know who was in the program.
Lastly, the PoP instructor did not have access to participants’ assessment information and was
not informed about individual inmates’ decisions to consent to releases of information.
Intervention Condition
The PoP Peacemaker course (Student level) was implemented by Laurel Kaufer over a
period of seven weeks and consisted of four-hour group sessions held twice a week in a
classroom located within the jail. A PoP graduate housed in another module at the jail was
allowed to serve as Ms. Kaufer’s assistant during the classes. The first author and the program
instructor were only in contact when issues occurred relating to inmate movement (e.g., three
PoP participants were moved to another module and it had to be arranged for them to be escorted
to class) or program fidelity (e.g., finding ways to allow inmates to complete between-session
PRISON OF PEACE RANDOMIZED TRIAL
26

exercises during lockdown). The program consisted of the following eight modules (Kaufer &
Noll, 2014), which each incorporated role plays, worksheets, and various interactive small group
exercises to help inmates connect their own experiences to the lessons being taught:
1. Restorative Justice: Introduction to Justice Systems. The first module teaches the
concept of restorative justice and guides offenders in understanding the basic human
needs of all victims. This step engages the women to adopt perspective-taking and
empathy skills, and to appreciate the consequences of their actions.
2. Listening Skills. Participants learn how to actively listen, reflect to clarify and
demonstrate understanding of what they have heard, build trust, create connections,
and show compassion. This module adopts concepts from the book People Skills by
Robert Bolton (1986).
3. Agreements. The goal of this module is to help inmates recognize the difference
between assumptions, expectations, and agreements, and to understand the rationale
for obtaining and maintaining agreements with others. Participants learn to develop
negotiation skills and accountability, and to prevent reactive and passive-aggressive
behaviors.
4. Results-based Listening. This module focuses on developing more effective
interpersonal relations and conducting goal-oriented conversations. The active
listener learns to guide others in understanding their own problems, goals and
possible solutions in order to prevent the onset of conflict in their lives. Participants
are taught to conduct conversations in a step-wise process: one person talks about
their concern; the other person actively listens, clearly identifies and states a problem,
reframes the problem, and finally transitions into problem-solving.
PRISON OF PEACE RANDOMIZED TRIAL
27

5. Peace Circles. The inmates are taught to organize structured, values-based
discussion groups where they can share their individual perspectives on topics related
to personal values. Peace Circles aim to foster understanding and empathy, deepen
community bonds, and promote collaboration, respect, dignity and equality.
6. Managing Strong Emotions. Participants are introduced to the concept of identifying,
managing, and expressing strong emotions, and to the cycle of emotion which leads
to behaviors and consequences. They learn to be self-aware by recognizing emotional
triggers and physical reactions, to acknowledge and validate the emotional
experiences of others and to understand the difference between subjective and
objective perception of events.
7. Using Peace Circles for Moral Re-engagement. This module guides participants in
interacting with other inmates who are morally disconnected by teaching the
principles of moral disengagement (i.e., methods used to reduce personal
accountability for one’s actions). Through this process, the women are also asked to
identify the moral disengagement that contributed to their own criminal behavior,
such as dehumanization, hatred, or peer-pressure.
8. Using Peace Circles for Conflict Resolution. Participants incorporate the skills they
have learned in the program into their Peace Circles and create a space for safe and
respectful listening within their communities. Treatment termination and graduation
occurs during this final phase.
As part of the weekly homework and skill practice assignments required to complete the
program, all participants were asked to conduct five Peace Circles, observe two Peace Circles
conducted by others, and write short reports about their experiences and observations. Given that
PRISON OF PEACE RANDOMIZED TRIAL
28

the women in Mod A were placed on lockdown in their cells for 23 hours a day, Ms. Kaufer
visited the jail on non-class days and pulled participants out of their cells to provide some
opportunities to practice the skills they had learned. At the end of the program, “graduates”
received a certificate of completion. Of note, all inmates at the jail, including participants housed
in Mod A, had access to Alcoholics Anonymous and church meetings.
Treatment as usual. Participants assigned to the two control groups took part in daily
activities as usual. Although Mod A and B were not allowed to partake in formal programs,
throughout the nine-month study period, various meetings were available to all inmates at the
facility (e.g., Alcoholics Anonymous, church group, parenting, life skills and GED classes). As
mentioned previously, Contamination Control participants housed in Mod A were locked in their
cells 23 hours of the day and Pure Control participants housed in Mod B had greater freedom of
movement through the common area of their module during the day.
Measures
All measures administered to participants are presented in Appendix B. Demographic
information and other historical information was completed at T1. The Adverse Childhood
Experiences (ACE) measure was administered at T2. Archival arrest and incident report data
were obtained at T3. Ms. Kaufer provided a log of class attendance and program engagement
following PoP completion, and all participants were asked to indicate their involvement with any
programs at T2 and T3. All remaining self-report measures were completed by participants at
pre-intervention, post-intervention, and 6-month follow-up. Parenting questions were only
completed by inmates that indicated they had a biological or step child of any age. Although the
following questionnaires have not been normed with incarcerated populations, some have been
used with prison inmates, and all items are estimated to range from a third-grade to a sixth-grade
PRISON OF PEACE RANDOMIZED TRIAL
29

reading level (https://readability-score.com). The average reading abilities of prison and jail
inmates in California are at the seventh-grade level (Petersila, 2006). Therefore, it was expected
that most participants had little difficulty completing the assessment battery.
Demographic information and personal characteristics. Participant age, ethnicity,
primary language, country of origin, relationship status, religious affiliation, education level,
household income, family composition and history, parenthood status, and history of substance
abuse, mental illness, and employment information were collected through self-report at T1.
Additionally, arrest history, family history of incarceration, and experience with current and
previous incarceration-based education programs were collected.
Program and location questionnaires. At T2 and T3, participants were asked to indicate
if they had participated in any programs at the jail (e.g., Alcoholics Anonymous, parenting
classes) and what, if any, exposure they had with Prison of Peace (e.g., “Since your last set of
questionnaires, have you been involved in a Peace Circle (participants sit in a circle and use a
talking piece to start a discussion about a specific topic)?”). Participants were asked to indicate if
they had been enrolled in PoP, and if they had led, participated in, or observed a Peace Circle,
and to indicate if their experience was better, worse, the same, or different than typical
conversations. Responses to this questionnaire were not used in final analyses, as several
inconsistencies were discovered with regard to group assignment. For example, several
participants in the Pure Control group, housed in Mod B, indicated that they had attended PoP
classes and led Peace Circles.
At T3, participants were asked to indicate their incarceration status (i.e., transfers to
prison, releases to the community, re-arrests) and movement within the jail (i.e., housing
modules they stayed in; placement in solitary confinement) over the last six months. This self-
PRISON OF PEACE RANDOMIZED TRIAL
30

report data was initially collected in the event that we would not be able to collect archival data
from jail officials.
Adverse childhood experiences. The Adverse Childhood Experiences (ACE; Felitti et
al., 1998) questionnaire was added to the T2 assessment battery to assess for history of family
hardship/conflicts. This is a 10-item yes/no questionnaire asking participants to answer questions
about incidents of abuse, neglect, and adversity in the first 18 years of their life. The typical use
of this questionnaire has been in large scale HMO studies and has been widely implemented with
various populations. Our participants were provided with specific directions before completing
this assessment. Due to the potentially triggering and sensitive nature of the items, the inmates
were told that they did not have to answer any questions they were uncomfortable answering and
that they could take breaks when answering questions. Participants were also explicitly asked to
only circle “yes” or “no” and not to write anything on their paper; this was done to prevent the
potential issue of child abuse reporting in the event that a perpetrator’s name was disclosed in
writing.
Conflict resolution skills. Conflict resolution skills were measured by using a vignette-
based assessment created for this study. Questions were developed and answers were coded
based on the rating system used for the Interpersonal Problem-solving Analysis (IPA; Marsh,
1982) measure, which reflects the problem-solving framework outlined by D’Zurilla and
Goldfried (1971). The IPA has displayed adequate test-retest reliability ranging from 0.75 to 0.93
for each rating measure (Slaby & Guerra, 1988) and has most commonly been used with
aggressive adolescents and delinquent youth (Foglia, 1996; Prince, 1994). The IPA (Marsh,
1982) was originally used by presenting children with a dilemma and asking them to provide
responses to the hypothetical situation.
PRISON OF PEACE RANDOMIZED TRIAL
31

The vignettes for the current study were created in collaboration with Laurel Kaufer and a
Prison of Peace graduate who had previously been incarcerated at the jail in both Mod A and B.
As with the original measure by Marsh, our participants were asked to read two to three conflict
scenarios and asked to: 1) define the problem; 2) describe the perceived goal of the situation; 3)
generate several solutions; and 4) select the best solution to achieve the perceived goal.
Scenarios were developed using common conflicts and problem behaviors experienced at the
jail. The same two vignettes (Phone and Shower) were used for pre-intervention and follow-up
assessments, as the elapsed time period (8 months) appeared sufficiently lengthy to elicit a
change in responses. Three vignettes were provided for the post-intervention assessment. Two of
the vignettes (Commissary and Hot Water) mirrored the conflicts provided in the baseline
vignettes (i.e., someone cutting the participant in line). A third ambiguous vignette (Coffee) was
added to pull for themes of “jumping to hostile conclusions” when it was unclear who was at
fault for a lost item. The vignettes are presented below:
PHONE: An inmate has been on the phone for half an hour. You’re next in line and you
need to call your family. The cop
3
announced “your 15 minutes are up” and you ask nicely
for your turn. The inmate ignores you the first time you ask. The second time you ask, she
says “I’m not getting off this phone.”

SHOWER: It’s shower time and you were able to put your stuff on the 4th chair in line to
hold your spot. You run back to your bunk to grab something and come back. Your stuff’s
been moved three chairs back and another girl is standing by the 4th chair with her stuff.

COMMISSARY: You’re in line to order your commissary and your program time is about
to run out. The inmate in front of you is taking forever ordering her stuff. You listen closely
to her order and realize she is placing orders for other inmates as well. (similar to phone
vignette)

HOT WATER: It’s program time and you get in line to get hot water. An inmate cuts in a
few people in front of you. Right before it’s your turn, the hot water runs out. They put the

3
Although the official term for officers assigned to the modules is “deputy” or “officer,” the slang term “cop” was
provided by a former jail inmate as the term most often used by inmates to indicate the deputy in charge.
PRISON OF PEACE RANDOMIZED TRIAL
32

hot pot away because the person in front of you got the last cup of hot water. (similar to
shower vignette)

COFFEE: You just got your bag of commissary and you’re looking forward to making a
meal. You go to your bag during program time and find that half of your bag of your coffee
is gone. You look across the room and see your bunkie sipping on a cup of coffee.
(ambiguous situation)

For each vignette, participants were asked the same four question and the following codes
were derived from the answers:
1. What is the problem in this situation? Describe what you think is wrong.
• Hostility v. Need
• Empathy/Positivity (present/absent)
• Attribution of Blame (Coffee vignette only; bunkie v. someone else)

2. What is your goal in this situation? What do you want/need?
• Conflict Definition: Hostility v. Need
• Empathy/Positivity (present/absent)
• Conflict Avoidance (present/absent)

3. Come up with at least 3 solutions to this problem.
• Alternate Solution Thinking: Hostile v. Non-hostile v. Avoidant
• Three Solutions Provided (adherence check; present/absent)

4. Choose the best solution to achieve your goal and explain why you chose it:
• Effective v. Ineffective
• Solution Type: Hostile v. Non-hostile v. Avoidant
• Best Solution Provided (adherence check; present/absent)

Across all time points, 391 vignettes were coded and 3,820 codes were generated. Two
undergraduate research assistants were trained over a period of two weeks and a detailed coding
protocol was developed and followed (Appendix C). All vignettes were double coded by the RAs
and disagreements were discussed in weekly coding meetings. Percentage agreement remained
between 82-97% for individual codes (e.g., Conflict Definition) and 87-90% for codes within
each individual vignette. Overall percentage agreement was 88%.
PRISON OF PEACE RANDOMIZED TRIAL
33

Common disagreements in coding variables were related to the language used in answers
provided and the tone each rater would attribute to an answer. For example, profanity was often
used, but did not necessarily indicate hostility. Other ambiguous content included the distinction
between “tell” and “ask.” For example, solutions for the phone vignette, “tell her to get her ass
off the phone” and “ask her to get her ass off the phone” yielded different ratings for hostility.
Raters also disagreed when participants expressed the possibility of hostility but provided an
adequate solution to prevent themselves from becoming hostile (e.g., “walk away, otherwise I’ll
get angry”). Disagreements also arose from complex answers that included a non-hostile solution
followed by a solution that would clearly lead to hostility (e.g., “ask her to please move and then
put my stuff back in front of her”). A gold standard rating was provided by the first author and
another graduate student to settle disagreements and the codebook was revised accordingly to
clarify any misinterpretations of directions provided. Consensus ratings were generated as
outcome variables.
Participant answers that did not adequately address the questions were coded as
“invalid.” These responses included expressing wishful thinking instead of providing a solution
(e.g., “we should all get more time to shower”), providing a solution when asked to generate a
goal for the situation (e.g., “I would just jump in front of her in the shower line”), random
comments (e.g., “the Lord is my savior and I know I’ll make it home someday”) and lack of
effort in providing an answer (e.g., “I don’t really care”). These invalid codes were not
calculated in the hostility indices for the outcome variables. Individual vignette codes by vignette
type and at each time point are provided in Appendix D.
Composite variables were created to capture three main outcomes. Mean perceived
hostility was created by taking the mean of “Conflict Definition” and “Goal Orientation” for all
PRISON OF PEACE RANDOMIZED TRIAL
34

vignettes at each assessment period. Both codes aim to discern whether participants were
focusing on hostility (e.g., the problem is I’m being disrespected; my goal is to beat that girl up
to show her respect) and need (e.g., the problem is I need to call my family; my goal is to have
enough time to call home). Mean hostility of solutions generated was created by taking the mean
of “Alternate Solution Thinking” and “Solution Type” for all vignettes at each assessment
period. Both codes identify whether the participant used a majority of hostile or non-hostile
solutions to address the conflict. Mean Effectiveness was created by taking the mean of the
“Effectiveness” code (i.e., was the best solution effective in obtaining the goal in question and
was it non-hostile) for all vignettes at each assessment period.
These three composites were assessed for interrater reliability. Intraclass correlation
coefficients (ICCs) using a consistency definition and two-way mixed model were calculated and
the convention developed by Cicchetti (1994) was adopted (i.e., below .40 = poor, .40 to .59 =
fair, .60 to .74 = good, and .75 to 1.00 = excellent). T1 to T3 ICCs were good to excellent for
Perceived Hostility (range = .735 - .881), fair to excellent for Hostility of Solutions Generated
(range = .579 - .809), and good to excellent for Effectiveness of Solutions (range = .672 - .753;
Appendix E).
Inmate conflict. Inmate conflict was measured at each time point using an adapted
version of the Conflict Scale. This 6-item questionnaire was developed by Macpherson (1986) as
part of a dissertation project examining the effects of an anger management program for
incarcerated men. Participants rated items on a 6-point scale ranging from “never” to “daily”
with regard to the number of times they were involved in various conflict situations in the past
two months (e.g., “I have been in a situation where some physical harm was done to someone”).
Types of conflicts ranged from basic disagreements to fighting and the use of weapons. Although
PRISON OF PEACE RANDOMIZED TRIAL
35

the Conflict Scale has not yet been validated (Macpherson, 1986), it is the only measure we
could identify that assesses the frequency of conflict incidents with other inmates and staff.
It should be noted that the format of the scale administered to participants at T1 was
subsequently modified because several mistakes were found in answers provided. For example,
when answering “In the last 2 months, I have been in a discussion with some disagreements,”
several participants wrote in the number of times this situation occurred, rather than using the
scale provided at the top of the page. When the two research assistants entering the data agreed
on the clarity of inmate responses (e.g., wrote in “1-2 times” or “zero” when zero was not a
response option), answers were modified to fit the scale. However, some responses had to be
invalidated. The scale was therefore modified for T2 and T3 assessments to include the rating
scale under each item of the questionnaire for participants to circle.
Anger. Anger and physical aggression were assessed using the Buss-Perry Aggression
Questionnaire-Short form (BPAQ-S; Buss & Perry, 1992). This measure, which consists of five
of the 29 original items, is rated on a 7-point scale ranging from “extremely untrue of me” to
“extremely true of me” (e.g., “if I am provoked enough, I may hit another person”). The original
measure displayed good internal consistency (α =.89; Buss & Perry, 1992).
Psychological functioning. Depression symptoms were measured using the Patient
Health Questionnaire-9 (PHQ-9; Kroenke, Spitzer, & Williams, 2001). This measure consists of
nine items that assess symptoms of depression experienced during the past two weeks (e.g.,
“little interest or pleasure in doing things”). The items are rated on a 4-point scale, with higher
ratings indicating higher frequency of symptoms. The PHQ-9 has been validated as a measure
with adequate internal consistency (α=.89; Kroenke et al., 2001) and with repeated
PRISON OF PEACE RANDOMIZED TRIAL
36

administrations reliably indicating improvements or worsening of depressive symptoms. The
suicide item was removed from the PHQ-9 as there was no clear site policy for reporting
procedures that would have maintained confidentiality among all inmates.
Anxiety symptoms were assessed using the 7-item Generalized Anxiety Disorder scale
(GAD-7; Löwe, Kroenke, Spitzer & Williams, 2006). This questionnaire measures symptoms of
generalized anxiety experienced during the past two weeks (e.g., “worrying too much about
different things”). It is rated on a 4-point scale indicating frequency of experienced symptoms.
The GAD-7 demonstrates high internal consistency (α =.92), good test-retest reliability
(intraclass correlation=0.83), as well as concurrent validity with alternate anxiety self-report
measures (Ruiz, Zamorano, García-Campayo, Pardo, Freire, & Rejas, 2011).
Both measures provide empirically-derived cut-off scores indicating severity of symptom
presentation, which can be used to determine clinically significant changes in symptom range
(i.e., 0-4=minimal to 20-27=severe symptoms; Löwe et al., 2006; Kroenke et al., 2001)
Parenting stress. Participants were asked to complete the Parental Stress Scale (PSS;
Berry & Jones, 1995) if they considered themselves parents. Inmates were asked to rate 18 items
relating to the positive (e.g., emotional benefits) and negative (e.g., demands on resources)
themes of parenthood on a 5-point scale (e.g., “The major source of stress in my life is my
child(ren)”). The PSS has adequate internal consistency (α =.83) and shows good sensitivity to
changes in stress levels for parents who are using external sources of support (Berry & Jones,
1995).
Perspective taking, personal distress, and empathy. The Interpersonal Reactivity
Index (IRI; Davis, 1980) is a 28-item measure answered on a 5-point scale ranging from
“describes me well” to “does not describe me well” (e.g., “I sometimes find it difficult to see
PRISON OF PEACE RANDOMIZED TRIAL
37

things from the "other guy's" point of view”). Three of the IRI’s four subscales were used for the
current study and measured perspective taking, empathic concern, and person distress (i.e.,
unease in tense interpersonal settings). The IRI has previously been used with offender
populations (Lauterbach & Hosser, 2007; Rong, Sun, Huang, Cai, & Li, 2010). This measure has
displayed good test-retest reliability for a 2-month period (intraclass correlations ranged from
0.62 to 0.71), adequate internal consistency (alpha ranged between 0.71 and 0.77), and good
concurrent validity with measures of social functioning, self-esteem, emotionality, sensitivity to
others, and empathy
4
.
Self-efficacy. The General Self-Efficacy Scale (GSE; Schwarzer & Jerusalem, 1995) is a
10-item measure rated on a 5-point scale ranging from “never” to “very often.” The scale
assesses participants’ optimistic self-beliefs and their ability to set goals, problem solve, and
exert control over their own behavior (e.g., “I can solve most problems if I try hard enough”) and
it exhibits adequate internal consistency (α=.80; Schwarzer, 1993).
Friend and family relationships. At each time point, participants were asked to indicate
the frequency (i.e., 1=daily to 6=never) and type (i.e., phone, mail, visits) of contact they had
with family members, children, and friends over the past two months. Family contact survey
questions were partially adapted from the Survey of Inmates in State Correctional Facilities
obtained from the Bureau of Justice Statistics (2004).
Archival data: inmate misconduct. Institutional misconduct was assessed for inmates
who signed the additional release of information. Official inmate incident reports (IRTS) were

4
Due to comprehension concerns for a few items and to maintain rating scale consistency across measures, a
modified version of the IRI was administered (e.g., item 6 “In emergency situations, I feel apprehensive and ill-at-
ease” was changed to “In emergency situations, I feel anxious and uneasy”).
PRISON OF PEACE RANDOMIZED TRIAL
38

collected from jail personnel at T3 and the number of disciplinary violations were aggregated for
each time period: pre-intervention, post-intervention, and follow-up. Pre-intervention was limited
to six months prior to PoP start date, post-intervention included the 7-week PoP period, and
follow-up included the six months after intervention completion. Behaviors described in incident
reports ranged from refusing to follow orders from deputies to violent altercations. IRTS were
assessed for level of hostility exhibited.
Non-hostile incident reports included possession of contraband (e.g., drugs, hoarding of
extra sheets, toothbrushes), unauthorized roaming, and insubordination with no indication of
hostility (e.g., inmate was told to return to her cell, initially refused, but eventually complied
without incident). Other examples of insubordination were inappropriate behavior with another
inmate (e.g., sexual activity), speaking to another inmate against regulation, refusing direct
orders from deputies, being described as “having an attitude” or “yelling
5
,” and taking/using
items without authorization. Hostile incident reports included any type of physical altercation
(e.g., fighting, resisting) or verbal aggression (e.g., yelling profanities, threats). When incident
reports indicated that no offense had occurred (e.g., “inmate requested extra food. She was told
no extra food would be allowed”), it was not coded as an IRTS in our analyses.
A ratio variable was calculated so participants who had the opportunity to be written up
for institutional misconduct (i.e., incarcerated) could be validly compared to participants who did
not have the opportunity to be written up (i.e., released/transferred to prison). Number of IRTS
were therefore divided by the number of days an inmate was housed at the jail at each time point.
For example, for the pre-intervention period of 02/16/2016 to 08/15/2016 (181 days), if an

5
Yelling with no other indication of hostility was not coded as hostile. Due to the distance between the inmate cells
and the deputy’s station, inmates often needed to yell to be heard by deputies.
PRISON OF PEACE RANDOMIZED TRIAL
39

inmate with three IRTS was at the facility and then released from jail on 03/01, re-arrested on
03/30, released on 04/15, and re-arrested again on 06/15, they would have a total of three IRTS
per their 90 days at the jail, for a ratio of .033333.
Archival data: arrest. Arrest records were gathered for inmates who consented to the
additional release. Arrest descriptions for participant offenses at recruitment are presented in
Appendix F. Archival records were obtained from the official database at the jail, and re-arrest
rates were analyzed for released participants. As with IRTS data, arrests were aggregated for
each time period: pre-intervention, post-intervention, and follow-up. Pre-intervention was limited
to six months prior to PoP start date, post-intervention included the 7-week PoP period, and
follow-up included the six months post-completion. A ratio variable was calculated by dividing
the number of arrests by the number of days an inmate was in the community at each time point.
For example, for the pre-intervention period of 02/16/2016 to 08/15/2016 (181 days), if an
inmate was released from jail on 03/01, re-arrested on 03/30, released on 04/15, and re-arrested
again on 06/15, they would have a total of 2 arrests per their 90 days in the community and a
ratio of .022222. As the number of released participants was equivalent across conditions, an
additional dichotomous variable was calculated to assess whether participants were ever re-
arrested after T1 (yes/at least one arrest=1; no/no re-arrests=0).
Re-offending. A dichotomous re-offending composite variable was calculated to assess
whether participants had any counts of re-arrest and/or any incident reports on their records after
T1 (at least one arrest or one IRTS=1; no re-arrest/no IRTS=0).
Program completion. Program completion was measured for PoP participants at T2 by
collecting attendance and Peace Circle/homework completion data from Ms. Kaufer’s records.
PRISON OF PEACE RANDOMIZED TRIAL
40

Of the 25 women randomly assigned to Prison of Peace, 16 attended at least half of the classes
and were therefore considered “completers.” Reasons for program dropout included transfer to
prison, release to the community, transfer to another module within the jail, refusal to participate,
or having missed more than half of the 14 classes due to various reasons (e.g., placed in solitary
confinement, illness). Completers attended an average of 12 classes, whereas non-completers
attended an average of two classes. Although it was an official graduation requirement to
conduct a minimum of five Peace Circles and observe a minimum of two Peace Circles, the 23-
hour lockdown procedure in the experimental module made these tasks difficult. On average,
PoP completers conducted 2.5 Peace Circles (range 0-5) and observed an average of 1.1 Peace
Circles (range 0-2).
6

Analyses
Statistical analyses were conducted using IBM SPSS Statistics 24. All assessment and
participant data were double-entered into SPSS, and qualitative data (i.e., conflict resolution
measure vignette responses) were double coded with high percentage agreement (88%). Power
analyses (i.e., G*Power 3.1; Faul, Erdfelder, Lang, & Buchner, 2007) for participants in the three
conditions (N=75) indicated adequate statistical power (i.e., power= .80) at the moderate to large
effect size level, but less than adequate power at the small effect size level.
Program effectiveness was tested using generalized estimating equations (GEE; Liang &
Zeger, 1986; Twisk, 2013). GEE is a linear predictor of all types of variables and can be
understood as a type of mixed model which utilizes a working correlation matrix (to correct for

6
Per the instructor, only five inmates earned a Peacemaker Certificate, as determined by their near-perfect
attendance, completion of two to five Peace Circles and homework assignments, and “appropriate in-class
behavior.” Five additional inmates attended classes and completed most homework assignments, but only received a
Certificate of Completion due to their in-class behavior (i.e., “they were not in any way able to behave like
peacemakers;” L. Kaufer, personal communication, February 13, 2018). The last six inmates did not receive
certificates due to homework incompletion and lack of engagement in the classes.
PRISON OF PEACE RANDOMIZED TRIAL
41

dependency of observation/within-subject correlations) to provide marginalized estimated means
for missing data. Further, GEE allows for the use of all data (i.e., does not use listwise deletion
of cases) and applies a robust standard of error. Multivariate linear regression models were fitted
to compare condition (PoP, Pure Control, and Contamination Control groups) on all outcomes
over time, controlling for T1 outcomes and including a condition by time interaction term.
Maximum likelihood estimations were used and an independent correlation structure was
applied. Baseline anxiety scores, inmate security level at the jail, ethnicity, and parenting status
were associated with attrition patterns and therefore identified a priori as potential covariates and
added to the model. Unstandardized regression coefficients and betas were calculated to reflect
rate-of-change differences (i.e., slopes) from T1 to T2 and T1 to T3 for each dyad (e.g., PoP
compared to PC or CC; CC compared to PC) and in the presence of covariates. Additionally,
binary logistic regression analyses were used for dichotomized data (i.e., arrest/no arrest and re-
offended/did not re-offend).
Tests of intervention effects used an intent-to-treat (ITT) approach by including all
participants originally randomized to condition. This approach aimed to reduce selection bias
and is in accordance with the CONSORT guidelines for randomized trials (APA Publications
and Communications Board Working Group on Journal Article Reporting Standards, 2008).
Additional analyses including only PoP completers (n=16) were conducted to examine potential
“dosage effects” of the intervention program. All tables and appendices include versions with
full sample and completers analyses.

PRISON OF PEACE RANDOMIZED TRIAL
42

Results
Missing Data
The majority of the study sample consented to a release of information for their arrest
(92%) and incident report data (95%), and there was no missing archival data for those
participants. However, self-report data at post-intervention and follow-up was missing for a large
proportion of the sample due to the transient nature of inmates at the jail (i.e., release to the
community; transfer to prison). Seventy-one participants (95% retention) completed baseline
assessment measures, 59 (79% retention) completed T2 assessments, and 39 (52% retention)
completed T3. Attrition rates did not differ between conditions χ
2
(2, 75) = .692, p > .05.
Ethnicity, χ
2
(2, 75) = 7.761, p = .021, parenting status, χ
2
(1, 73) = 5.509, p = .019,
baseline anxiety, t(69) = 2.547, p = .013, and inmate security level, t(67) = .374, p = .064, were
associated with attrition (Appendix G) and were added as covariates in the models
7
. Participants
were more likely to miss T2 or T3 assessments if they were non-Hispanic, non-Black, parents,
low in anxiety, or were assigned lower security levels at the jail.
Baseline Characteristics and Differences
Table 1 provides information on participant demographics and baseline characteristics.
On average, inmates were approximately 32 years old (range 18-57 years) at recruitment and had
been incarcerated for 3-4 months (range 13-863 days) for non-violent felony offenses. Most
participants were single and identified as Hispanic or Black, and over 40% had not earned a high
school degree. Eighty percent lived below the poverty line prior to incarceration and most were

7
Although two additional incarceration variables were associated with missing data (days in jail pre-recruitment,
t(67) = 1.973, p = .053, and number of hostile IRTS pre-recruitment, t(69) = 1.520, p = .025), they are conceptually
linked to security level, did not differ between conditions, and thus were not added in the model as additional
covariates.
PRISON OF PEACE RANDOMIZED TRIAL
43

receiving public assistance. A large proportion were parents with a child in their care at the time
of their arrest. Nearly half of the sample indicated a prior violent arrest. Most had at least one
incident report on their jail records and 13 women reported a history of gang membership
(community- or jail/prison-based). With regard to psychological functioning, most women
reported a substance abuse history; most had been diagnosed by a health professional (informally
or formally) with a mental disorder, with depression being the most common diagnosis (36% of
the sample).
Table 1

Baseline Characteristics of Participant Sample

Characteristic N %
Ethnicity -- --
Hispanic 29 38.7
African American 29 38.7
White 8 10.7
Biracial 5 6.7
Native American 2 2.7
Asian 1 1.3
Education
Middle – High School/non-grad 31 42.5
High School graduate 20 27.4
Some College 13 17.8
AA/Vocational Degree/Bachelors 9 12.3
Single/never married 44 61.1
Parent 50 69.4
Child under care at time of arrest 32 45.1
Living below poverty line 57 80.3
Receiving public assistance 45 61.6
Arrest/offending history
Past drug offense 35 48.6
Past violent offense 34 47.2
Current arrest felony (archival) 58 84.1
Current arrest violent (archival) 25 36.2
Incident report on record (archival) 48 67.6
Family history of incarceration 46 63.9
Gang membership history 1 17.3
Substance abuse history 52 72.2
PRISON OF PEACE RANDOMIZED TRIAL
44

Note. Percentages were calculated from the number of participants endorsing a behavior divided by the number of
participants who gave reports on that behavior, not the total number of participants. Poverty line determined by 2016
federal guidelines <$20,000 for household of three. All data represent self-report.

Although the ACE measure was administered at T2 and only completed by 77% of
participants, endorsed items (Appendix H) indicated that the majority of our sample had
experienced significant childhood adversity (M=4.62; SD=2.73). Of note, the second most
commonly endorsed item, after parental separation/divorce, was being sexually assaulted as a
child by an adult (53%). Extensive research on the measure reveals that a score of 4 or more is
linked with significantly increased risk for serious negative health and social outcomes (e.g.,
Larkin, Felitti, & Anda, 2011).
The treatment conditions differed significantly along several baseline characteristics
(Appendix I). Inmate security level (1=low risk to 8=high risk) differed, F(2, 71) = 10.22, p <
.001, for Pure Controls (M=4.9) compared to PoP (M=6.8) and CC (M=6.3). The conditions also
differed by ethnicity, χ
2
(2, 75) = 10.30, p = .036, with Contamination Control including a larger
proportion of Black participants and lower proportions of Hispanic and Other participants
compared to PoP and Pure Control. CC participants were also marginally more likely to have
been arrested for a violent crime at recruitment compared to PoP and PC, χ
2
(2, 69) = 5.79, p =
Mental health history: ever diagnosed with
Any mental health problem 48 66.7
Depression 27 37.5
Bipolar Disorder 21 29.2
PTSD 21 29.2
Anxiety 20 27.8
Schizophrenia 7 9.7
M SD
Participant age, years 31.90 10.61
Age of children, years (for parents) 12.33 2.36
Number of children (for parents) 1.77 1.76
Number of days incarcerated prior to program
start date (archival)
116.03 154.07
PRISON OF PEACE RANDOMIZED TRIAL
45

.055. The conditions did not differ with regard to pre-intervention days in jail or number of prior
arrests.
When examining baseline difference for outcome variables (Appendix J), PoP
participants had marginally higher non-hostile IRTS than controls, both for incident report
counts, F(2, 71) = 2.60, p = .082, and ratios, F(2, 71) = 2.79, p = .069. Further, CC participants
reported marginally higher levels of self-efficacy compared to PoP and PC, F(2, 71) = 2.96, p =
.058. Lastly, PoP participants who completed the program had marginally higher ratings on the
hostility of solutions composite compared to controls, F(2, 71) = 3.03, p = .056.
Distribution of Archival Data
Incident report data were equally skewed for each condition across time and ranged from
0 counts (16% of sample at T1
8
; 53% at T2; 27% at T3) to 21 counts (n=1 at T1). The
distribution of IRTS data is presented in Figures 2-4 below.

8
T1 archival data refers to a 6-month period prior to PoP start date.
PRISON OF PEACE RANDOMIZED TRIAL
46

Figure 2: Distribution of T1 incident reports by condition for the full sample

Figure 3: Distribution of T2 incident reports by condition for the full sample

PRISON OF PEACE RANDOMIZED TRIAL
47

Figure 4: Distribution of T3 incident reports by condition for the full sample

Arrest data was similarly skewed across conditions. The majority of participants had
between zero and six prior arrests at T1 (not including their current arrest). Only one participant
was released and re-arrested during the 7-week program period and 19% of the sample were re-
arrested by T3. Counts of arrest and IRTS data can be found in Appendix K.
Intervention Outcomes
Means, standard deviations, and sample sizes for outcome variables at each time point are
available in Appendix L. For continuous data, GEE linear models with unstructured
covariance patterns were fit to all outcome data and predictors included time and condition. A
significant interaction can be understood as one condition exhibiting a greater “rate of change” or
a significant difference in the slope of an outcome variable compared to another condition. GEE
PRISON OF PEACE RANDOMIZED TRIAL
48

results are presented in Tables 2-4 for each of the group comparisons (PoP vs. PC; PoP vs. CC;
PC vs. CC).
Table 2

Generalized Estimating Equation (GEE) Model Estimates Comparing Prison of Peace to Pure
Control Participants on Outcome Variables at T2 and T3

Note: T1 = pre-intervention, T2 = post-intervention, T3 = 6-month follow-up. Archival outcomes: T1=6 months
prior to PoP start date. Incident Report (IRTS) ratios calculated as number of IRTS divided by number of days in jail
for the period of time indicated. Arrest ratios calculated as number of arrests divided by number of days in the
community for the period of time indicated. Arrests represent all types of arrest recorded in the jail records system.
Covariates included in model (except for anxiety outcomes): ethnicity, parenting status, security level, and baseline
GAD score. p < .10. p < .05.
Outcome Variable
Condition * T2 Condition * T3
B
95% CI
Lower Upper
p-
value
B
95% CI
Lower Upper
p-
value
Conflict Resolution
Vignettes

Perceived Hostility -.255 -.423 -.087 .003 -.064 -.308 .181 .611
Solution Hostility -.253 -.411 -.095 .002 -.114 -.366 .138 .376
Solution
Effectiveness
.074 -.194 .342 .587 -.066 -.392 .260 .692
Conflict Scale -.871 -3.825 2.083 .563 .863 -3.258 4.985 .681
Anger -4.639 -9.411 .134 .057 4.499 -1.425 10.423 .137
Depression -3.317 -7.112 .477 .087 2.331 -3.637 8.299 .444
Anxiety -2.527 -6.968 1.917 .265 .240 -5.465 5.945 .934
Parental Stress 5.890 -1.685 13.465 .128 1.487 -6.985 9.959 .731
Interpersonal
Reactivity Index

Perspective Taking -.033 -3.343 3.278 .985 -1.044 -4.958 2.871 .601
Empathic Concern -.564 -3.464 2.336 .703 -.729 -5.031 3.573 .740
Personal Distress -.115 -2.874 2.643 .935 -.083 -4.056 3.891 .967
Self-Efficacy -.401 -3.769 2.967 .815 3.938 -1.455 9.331 .152
Frequency of Contact
With Children -.148 -.740 .444 .624 .389 -.894 1.671 .553
With Family -.309 -.956 .337 .348 -.254 -1.148 .639 .577
With Friends .027 -.593 .647 .932 -.056 -1.081 .969 .915
Incident Report Ratios
Total -.006 -.021 .009 .430 .016 -.001 .033 .060
Non-hostile -.005 -.016 .006 .372 .013 -.002 .027 .083
Hostile -.001 -.010 .008 .821 .003 -.004 .010 .341
Arrests -.020 -.067 .027 .411 .007 -.006 .020 .274
PRISON OF PEACE RANDOMIZED TRIAL
49

Table 3

Generalized Estimating Equation (GEE) Model Estimates Comparing Prison of Peace to
Contamination Control Participants on Outcome Variables at T2 and T3

Note: T1 = pre-intervention, T2 = post-intervention, T3 = 6-month follow-up. Archival outcomes: T1=6
months prior to PoP start date. Incident Report (IRTS) ratios calculated as number of IRTS divided by number
of days in jail for the period of time indicated. Arrest ratios calculated as number of arrests divided by number
of days in the community for the period of time indicated. Arrests represent all types of arrest recorded in the
jail records system. Covariates included in model (except for anxiety outcomes): ethnicity, parenting status,
security level, and baseline GAD score. p < .10. p < .05.
a
=

Significance was lost when baseline self-efficacy
scores (which were significantly different between groups) were added in the model.
Outcome Variable
Condition * T2 Condition * T3
B
95% CI
Lower Upper
p-
value
B
95% CI
Lower Upper
p-
value
Conflict Resolution
Vignettes

Perceived Hostility -.100 -.307 .107 .342 -.056 -.391 .279 .743
Solution Hostility -.145 -.277 -.012 .033 -.140 -.341 .061 .173
Solution
Effectiveness
.028 -.227 .283 .829 -.228 -.569 .113 .191
Conflict Scale -2.614 -5.598 .371 .086 -3.378 -7.803 1.048 .135
Anger -9.010 -14.076 -3.944 .000 -1.054 -6.850 4.742 .721
Depression -4.732 -8.659 -.805 .018 -2.656 -9.005 3.693 .412
Anxiety -3.919 -8.598 .761 .101 -3.471 -9.076 2.135 .225
Parental Stress -1.512 -8.814 5.789 .685 -4.723 -12.519 3.072 .235
Interpersonal
Reactivity Index

Perspective Taking .195 -2.596 2.985 .891 -.551 -4.827 3.724 .800
Empathic Concern -2.025 -4.645 .595 .130 .013 -4.113 4.139 .995
Personal Distress -.518 -3.734 2.697 .752 -.778 -5.263 3.708 .734
Self-Efficacy 3.497 -.711 7.705 .103 4.558 -.486 9.603 .077
a

Frequency of Contact
With Children .519 -.293 1.330 .211 .070 -1.138 1.277 .910
With Family .349 -.420 1.118 .374 .070 -.609 .749 .840
With Friends .462 -.108 1.031 .112 .366 -.573 1.304 .445
Incident Report Ratios
Total .006 -.005 .017 .281 .019 .002 .036 .032
Non-hostile .005 -.002 .013 .152 .017 .002 .032 .030
Hostile
.001 -.007 .008 .874 .002 -.005 .009 .607
Arrests
-.002 -.011 .007 .601 .000 -.010 .009 .919
PRISON OF PEACE RANDOMIZED TRIAL
50

Table 4

Generalized Estimating Equation (GEE) Model Estimates Comparing Contamination Control to
Pure Control Participants and Outcome Variables at T2 and T3

Note: T1 = pre-intervention, T2 = post-intervention, T3 = 6-month follow-up. Archival outcomes: T1=6 months
prior to PoP start date. Incident Report (IRTS) ratios calculated as number of IRTS divided by number of days in jail
for the period of time indicated. Arrest ratios calculated as number of arrests divided by number of days in the
community for the period of time indicated. Arrests represent all types of arrest recorded in the jail records system.
Covariates included in model (except for anxiety outcomes): ethnicity, parenting status, security level, and baseline
GAD score. p < .10. p < .05.
a
=

Significance was lost when baseline self-efficacy scores (which were significantly
different between groups) were added in the model.
Outcome Variable
Condition * T2 Condition * T3
B
95% CI
Lower Upper
p-
value
B
95% CI
Lower Upper
p-
value
Conflict Resolution
Vignettes

Perceived Hostility -.154 -.368 .060 .158 -.008 -.314 .299 .962
Solution Hostility -.108 -.246 .029 .121 .026 -.187 .239 .810
Solution
Effectiveness
.046 -.225 .318 .739 .162 -.145 .468 .302
Conflict Scale 1.742 -1.407 4.892 .278 4.241 .230 8.252 .038
Anger 4.371 .447 8.296 .029 5.554 .478 10.629 .032
Depression 1.415 -2.848 5.677 .515 4.987 -1.155 11.129 .112
Anxiety 1.392 -5.671 2.888 .524 -3.771 -9.679 2.257 .223
Parental Stress 7.402 -.738 15.543 .075 6.211 .256 12.166 .041
Interpersonal
Reactivity Index

Perspective Taking -.227 -2.974 2.519 .871 -.492 -4.010 3.026 .784
Empathic Concern 1.461 -1.191 4.114 .280 -.742 -4.977 3.492 .731
Personal Distress .403 -2.564 3.370 .790 .695 -3.332 4.722 .735
Self-Efficacy -3.899 -8.111 .313 .070
a
-.620 -5.236 3.996 .792
Frequency of Contact
With Children -.666 -1.460 .127 .100 .319 -.832 1.470 .587
With Family -.658 -1.265 -.051 .340 -.324 -1.237 .589 .487
With Friends -.435 -1.004 .135 .135 -.422 -1.344 .501 .371
Incident Report Ratios
Total -.012 -.025 .001 .064 -.002 -.013 .008 .664
Non-hostile -.010 -.020 -.001 .028 -.004 -.013 .005 .411
Hostile -.002 -.010 .006 .684 .002 -.005 .008 .615
Arrests -.017 -.064 .029 .465 .008 -.006 .021 .281
PRISON OF PEACE RANDOMIZED TRIAL
51

Nine of our 19 outcomes yielded significant or marginally significant interactions at post-
treatment, follow-up, or both. Means and standard deviations for all relevant variables at each
time period are presented in Table 5. Significant and marginally significant self-report outcomes
are presented first, followed by significant and marginally significant archival outcomes, and
lastly, program completers analyses.

Table 5

Unadjusted Means and Standard Deviations of Significant GEE Interactions of Outcomes and
Conditions
Outcome Variable PoP (M) SD CC (M) SD PC (M) SD
CRM: perceived hostility
T1 1.504 .219 1.528 .336 1.450 .254
T2 1.327 .233 1.469 .269 1.503 .281
T3 1.435 .314 1.506 .378 1.417 .257
CRM: hostility of
solutions generated

T1 1.288 .277 1.167 .236 1.170 .254
T2 1.113 .201 1.169 .286 1.214 .187
T3 1.179 .250 1.289 .320 1.229 .310
Conflict Scale
T1 13.26 5.51 10.57 5.32 12.88 5.80
T2 12.65 4.95 13.74 5.53 13.00 5.01
T3 12.69 6.78 14.93 4.97 12.17 4.91
Anger
T1 19.50 7.40 16.50 9.00 20.60 8.64
T2 13.60 7.93 19.74 8.21 18.85 9.33
T3 16.00 9.07 17.36 9.35 15.25 7.35
Depression
T1 8.96 4.96 9.55 6.27 10.96 6.76
T2 5.00 4.65 10.37 6.40 10.37 7.36
T3
8.69 5.91 12.43 7.90 8.33 6.34
Parental Stress
T1 31.41 7.79 30.38 7.92 31.57 10.78
T2 29.62 8.83 31.15 11.17 26.33 6.65
T3 24.20 4.82 29.11 8.05 29.20 9.26
Incident Reports - Total
T1 .024550 .024 .014652 .016 .016650 .021
T2 .016371 .024 .005820 .022 .016527 .031
T3 .026904 .037 .008816 .015 .016309 .018
PRISON OF PEACE RANDOMIZED TRIAL
52

Note: T1=6 months pre-PoP start date; T2=post-intervention; T3=6-months post-completion; CRM=Conflict
Resolution Measure; PoP=25 participants; PoP Completers=16 participants; Depression (PHQ-9) and Anxiety
(GAD-7): minimal/mild symptoms range=5-9, moderate=10-14; CRM means range 1=non-hostile; 2=hostile

Regarding post-intervention changes in conflict resolution skills, PoP participants showed
greater decreases in perceived hostility of conflict situations compared to Pure Control
participants (B= -.255, 95% CI [-.423 to -.087]; p=.003). Further, PoP participants showed
greater decreases in hostility of solutions generated compared to Pure Controls (B= -.253, 95%
CI [-.411 to -.095]; p=.002) and Contamination Controls (B= -.145, 95% CI [-.277 to -.012];
p=.033). No significant interactions were observed at T3.

Incident Reports -
Non-hostile

T1
.017425 .021 .007548 .011 .008682 .013
T2
.014479 .022 .003576 .014 .012395 .025
T3
.021554 .032 .006040 .013 .012959 .015
Significant outcomes for
completers only
PoP
Completers
(M)
SD CC (M) SD PC (M) SD
Depression
T1 9.81 4.25 9.55 6.27 10.96 6.76
T2 5.64 5.14 10.37 6.40 10.37 7.36
T3 7.25 5.83 12.43 7.90 8.33 6.34
Parental Stress
T1 33.91 8.04 30.38 7.92 31.57 10.78
T2 30.44 10.50 31.15 11.17 26.33 6.65
T3 22.00 4.00 29.11 8.05 29.20 9.26
Incident Reports - Hostile
T1 .004731 .008 .007100 .012 .007964 .011
T2 .000000 .000 .002240 .008 .004132 .011
T3 .005113 .010 .002768 .008 .003336 .006
PRISON OF PEACE RANDOMIZED TRIAL
53

Figure 5: Perceived hostility of conflict situation T1-T3

Figure 6: Hostility of solutions generated T1-T3

PRISON OF PEACE RANDOMIZED TRIAL
54

PoP participants showed a marginal decrease in self-reported conflicts (i.e., frequency of
verbal or physical arguments) compared to Contamination Controls at T2 (B= -2.614, 95% CI [-
5.598 to .371]; p=.086). Further, Pure Controls exhibited a significant decrease in conflict
compared to Contamination Control at T3 (B= 4.241, 95% CI [.230 to 8.252]; p=.038). No
interactions were observed when comparing PoP to Pure Controls.
Figure 7: Conflict scale means T1-T3

At T2, PoP participants exhibited a significantly sharper decrease in self-reported anger
compared to Contamination Controls (B= -9.010, 95% CI [-14.076 to -3.944]; p<.001) and a
marginally significant decrease compared to Pure Controls (B= -4.639, 95% CI [-9.411 to .134];
p=.057). No PoP interactions were observed at T3. Regarding differences between the two
control groups, Pure Controls exhibited significantly greater decreases in anger compared to
PRISON OF PEACE RANDOMIZED TRIAL
55

Contamination Controls at T2 (B= 4.371, 95% CI [.447 to 8.296]; p=.029) and at T3 (B= 5.554,
95% CI [.478 to 10.629]; p=.032).
Figure 8: Anger means T1-T3

At T2, PoP participants exhibited a significantly sharper decrease in depression compared
to Contamination Control (B= -4.732, 95% CI [-8.659 to -.805]; p=.018) and a marginally
significant decrease compared to Pure Control (B= -3.317, 95% CI [-7.112 to .477]; p=.087). The
decrease observed in the PoP group indicated a clinically significant change in symptom range;
all participants were near the low end of the moderate symptom range at T1 (Ms=8.96-10.96;
mod range=10-14), but only PoP participants reported a decrease to the low end of the mild
range at T2 (M=5.00; mild range=5-9). No interactions were observed at T3.

PRISON OF PEACE RANDOMIZED TRIAL
56

Figure 9: Depression means T1-T3

No significant parenting stress interactions were observed between PoP participants and
the two control groups. However, Pure Controls exhibited a marginally greater decrease in
parenting stress at T2 (B= 7.402, 95% CI [-.738 to 15.543]; p=.075) and a significantly greater
decrease at T3 (B= 6.211, 95% CI [.256 to 12.166]; p=.041) compared to Contamination
Controls.

PRISON OF PEACE RANDOMIZED TRIAL
57

Figure 10: Parenting stress means T1-T3

Unexpectedly, an iatrogenic effect of PoP programming was observed on rates of
institutional misconduct. Specifically, PoP participants exhibited a significant increase in IRTS
compared to Contamination Controls at T3 (B= .019, 95% CI [.002 to .036]; p=.032); the
difference was marginally significant compared to Pure Controls (B= .016, 95% CI [-.001 to
.033]; p=.060). When examining non-hostile IRTS only, similar differences were observed
between the groups at T3 (Contamination Control B= .017, 95% CI [.002 to .032]; p=.030; Pure
Control B= .013, 95% CI [-.002 to .027]; p=.083). No significant PoP interactions were observed
at T2, but differences were found between the control groups.
Contamination Controls exhibited a marginal decrease in IRTS (B= -.012, 95% CI [-.025
to .001]; p=.064) and a significant decrease in non-hostile IRTS at T2 (B= -.010, 95% CI [-.020
PRISON OF PEACE RANDOMIZED TRIAL
58

to -.001]; p=.028) compared to Pure Controls. No significant differences were observed at T3 for
the two control groups.

Figure 11: Incident report ratios T1- T3

Figure 12: Non-hostile incident report ratios T1- T3

PRISON OF PEACE RANDOMIZED TRIAL
59

No significant interactions were observed for hostile IRTS, re-arrests, or other self-report
measures when comparing the PoP full sample to the control groups. It should be noted that all
significant T2 PoP interactions did not remain significant at T3. Means and standard deviations
for non-significant outcomes can be found in Appendix M.
Program completion analyses. GEE analyses were run to examine differences between
participants who completed the PoP program (N=16) and both control groups. Besides marginal
differences in parenting stress, completers did not significantly differ from PoP dropouts on
demographic characteristics and baseline outcome variables (Appendix N). Most results from the
ITT analyses were replicated, but with many marginally significant effects attaining full
significance in the “completer” models (Tables 6-8).

PRISON OF PEACE RANDOMIZED TRIAL
60

Table 6

Generalized Estimating Equation (GEE) Model Estimates Comparing Prison of Peace
Completers to Pure Control Participants and Outcome Variables at T2 and T3

Note: T1 = pre-intervention, T2 = post-intervention, T3 = 6-month follow-up. Archival outcomes: T1=6 months
prior to PoP start date. Incident Report (IRTS) ratios calculated as number of IRTS divided by number of days in jail
for the period of time indicated. Arrest ratios calculated as number of arrests divided by number of days living in the
community for the period of time indicated. Arrests represent all types of arrest recorded in the jail records system.
Covariates included in model (except for anxiety outcomes): ethnicity, parenting status, security level, and baseline
GAD score. p < .10. p < .05.

Outcome Variable
Condition * T2 Condition * T3
B
95% CI
Lower Upper
p-
value
B
95% CI
Lower Upper
p-
value
Conflict Resolution
Vignettes

Perceived Hostility -.278 -.439 -.117 .001 -.143 -.366 .080 .209
Solution Hostility -.204 -.359 -.048 .010 -.124 -.367 .119 .316
Solution
Effectiveness
.023 -.273 .320 .877 .007 -.337 .351 .969
Conflict Scale -.883 -3.904 2.137 .566 1.187 -3.841 6.214 .644
Anger -3.811 -9.280 1.657 .172 2.980 -3.355 9.316 .357
Depression -4.099 -8.298 .100 .056 -1.174 -7.056 4.705 .695
Anxiety -2.469 -7.574 2.636 .343 -.682 -6.719 5.355 .825
Parental Stress 4.539 -4.498 13.577 .325 -2.287 -10.512 5.937 .586
Interpersonal
Reactivity Index

Perspective Taking 1.458 -1.999 4.914 .408 .088 -4.419 4.596 .969
Empathic Concern .893 -2.095 3.880 .558 -.371 -5.194 4.451 .880
Personal Distress .058 -2.313 2.429 .962 -.031 -4.635 4.573 .989
Self-Efficacy -.129 -3.866 3.607 .946 3.755 -2.471 9.981 .237
Frequency of Contact
With Children -.276 -.863 .311 .356 .636 -.625 1.896 .323
With Family -.224 -.665 .217 .320 -.089 -1.014 .835 .849
With Friends -.079 -.753 .595 .817 -.222 -1.376 .932 .707
Incident Report Ratios
Total -.004 -.020 .012 .603 .022 .004 .040 .019
Non-hostile -.004 -.017 .009 .518 .016 .003 .008 .000
Hostile .000 .0039 -.007 .980 .006 -.001 .012 .079
Arrests
-.019 -.069 .031 .449 .007 -.008 .021 .345
PRISON OF PEACE RANDOMIZED TRIAL
61

Table 7

Generalized Estimating Equation (GEE) Model Estimates Comparing Prison of Peace
Completers to Contamination Control Participants and Outcome Variables at T2 and T3

Note: T1 = pre-intervention, T2 = post-intervention, T3 = 6-month follow-up. Archival outcomes: T1=6 months
prior to PoP start date. Incident Report (IRTS) ratios calculated as number of IRTS divided by number of days in jail
for the period of time indicated. Arrest ratios calculated as number of arrests divided by number of days living in the
community for the period of time indicated. Arrests represent all types of arrest recorded in the jail records system.
Covariates included in model (except for anxiety outcomes): ethnicity, parenting status, security level, and baseline
GAD score. p < .10. p < .05.

Outcome Variable
Condition * T2 Condition * T3
B
95% CI
Lower Upper
p-
value
B
95% CI
Lower Upper
p-
value
Conflict Resolution
Vignettes

Perceived Hostility -.121 -.321 .078 .233 -.142 -.462 .179 .385
Solution Hostility -.198 -.353 -.042 .013 -.153 -.399 .093 .224
Solution
Effectiveness
-.037 -.324 .250 .798 -.162 -.524 .199 .379
Conflict Scale -2.613 -5.671 .446 .094 -2.907 -8.187 2.373 .281
Anger -8.042 -13.669 -2.416 .005 -2.347 -8.616 3.922 .463
Depression -5.459 -9.832 -1.086 .014 -5.932 -12.265 .400 .066
Anxiety -3.756 -9.104 1.593 .169 -4.450 -10.311 1.410 .137
Parental Stress -2.818 -11.594 5.958 .529 -8.632 -16.376 -.888 .029
Interpersonal
Reactivity Index

Perspective Taking 1.754 -1.139 4.648 .235 .456 -4.381 5.293 .853
Empathic Concern -.572 -3.310 2.167 .682 .280 -4.377 4.937 .906
Personal Distress -.300 -3.192 2.592 .839 -.800 -5.833 4.234 .756
Self-Efficacy 3.670 -.874 8.214 .113 4.146 -1.817 10.108 .173
Frequency of Contact
With Children .348 -.442 1.138 .388 .189 -.967 1.344 .749
With Family .435 -.173 1.043 .161 .184 -.522 .890 .609
With Friends .349 -.279 .977 .276 .218 -.836 1.272 .685
Incident Report Ratios
Total .008 -.005 .020 .214 .024 .006 .043 .011
Non-hostile .006 -.004 .016 .251 .020 .002 .038 .029
Hostile .002 -.004 .008 .560 .004 -.002 .011 .203
Arrests
-.001 -.014 .011 .834 .000 -.012 .012 .971
PRISON OF PEACE RANDOMIZED TRIAL
62

Table 8

Generalized Estimating Equation (GEE) Model Estimates Comparing Contamination Control to
Pure Control Participants and Outcome Variables at T2 and T3

Note: T1 = pre-intervention, T2 = post-intervention, T3 = 6-month follow-up. Archival outcomes: T1=6 months
prior to PoP start date. Incident Report (IRTS) ratios calculated as number of IRTS divided by number of days in jail
for the period of time indicated. Arrest ratios calculated as number of arrests divided by number of days living in the
community for the period of time indicated. Arrests represent all types of arrest recorded in the jail records system.
Covariates included in model (except for anxiety outcomes): ethnicity, parenting status, security level, and baseline
GAD score. p < .10. p < .05.

Three additional significant interactions were observed in the completers analyses. First,
PoP completers reported a marginally greater decrease in depression symptoms at T3 compared
Outcome Variable
Condition * T2 Condition * T3
B
95% CI
Lower Upper
p-
value
B
95% CI
Lower Upper
p-
value
Conflict Resolution
Vignettes

Perceived Hostility -.157 -.370 .057 .150 -.001 -.306 .304 .994
Solution Hostility -.102 -.239 .035 .144 .025 -.189 .238 .821
Solution
Effectiveness
.061 -.211 .333 .661 .169 -.138 .476 .281
Conflict Scale 1.729 -1.403 4.861 .279 4.094 .043 8.145 .048
Anger 4.231 .292 8.170 .035 5.327 .237 10.418 .040
Depression 1.360 -2.892 5.611 .531 4.785 -1.422 10.939 .131
Anxiety 1.287 -3.019 5.593 .558 3.769 -2.182 9.719 .214
Parental Stress 7.358 -.793 15.508 .077 6.345 .278 12.411 .040
Interpersonal
Reactivity Index

Perspective Taking -.297 -3.008 2.415 .830 -.367 -3.867 3.132 .837
Empathic Concern 1.464 -1.182 4.110 .278 -.651 -4.925 2.180 .765
Personal Distress .358 -2.613 3.330 .813 .768 -.257 4.794 .708
Self-Efficacy -3.799 -7.991 .392 .076 -.390 -4.907 4.126 .865
Frequency of Contact
With Children -.624 -1.427 .179 .128 .447 -.705 1.599 .447
With Family -.659 -1.268 -.049 .340 -.273 -1.183 .636 .556
With Friends -.428 -.998 .142 .141 -.440 -1.363 .484 .351
Incident Report Ratios
Total
-.012 -.025 .001 .064 -.002 -.013 .009 .701
Non-hostile -.010 -.020 -.001 .028 -.004 -.013 .006 .444
Hostile -.002 -.010 .006 .681 .002 -.005 .008 .615
Arrests -.018 -.065 .029 .453 .007 -.006 .021 .295
PRISON OF PEACE RANDOMIZED TRIAL
63

to Contamination Controls (B= .006, 95% CI [-.001 to .012]; p=.079). Specifically, program
completers remained in the mild symptom range while CC participants increased from the mild
to the moderate range. Although the full PoP sample and PoP completers both exhibited a
significant decrease in depression at T2 compared to controls, completers alone maintained a
significant decrease at T3.

Figure 13: Depression means with PoP completers T1- T3

Second, PoP completers exhibited a significant decrease in parenting stress at T3
compared to Contamination Controls (B= -8.632, 95% CI [-16.376 to -.888]; p=.029). Compared
to full PoP sample means, completers reported greater parenting stress at baseline and a sharper
decrease in parenting stress by T3. No significant interactions were observed at T2 between the
three conditions.
PRISON OF PEACE RANDOMIZED TRIAL
64

Figure 14: Parenting stress with PoP completers T1- T3

Lastly, PoP completers exhibited a marginally greater increase in hostile IRTS compared
to Pure Controls at T3 (B= .006, 95% CI [-.001 to .012]; p=.079). It should be noted that post-
intervention GEE comparisons were not possible because none of the PoP completers had any
Hostile IRTS on record during this assessment period. No differences were observed in relation
to Contamination Controls.

PRISON OF PEACE RANDOMIZED TRIAL
65

Figure 15: Hostile incident report ratios with PoP completers T1-T3

Re-arrest and re-offending. There were no differences for incarceration outcomes at
T3; across conditions, participants were transferred to prison or released from the jail at an equal
rate (χ
2
(2, 75) = 1.833, p = .400; χ
2
(2, 75) = .108, p = .948, respectively). With regard to
dichotomous variables, we evaluated the impact of PoP on several indices of misconduct. Binary
Logistic Regression analyses (ITT and completers) yielded no significant results for re-arrest and
the re-offending composite (Table 9).

PRISON OF PEACE RANDOMIZED TRIAL
66

Table 9

Binary Logistic Regression of Dichotomized Data with Prison of Peace Completers,
Contamination Control, and Pure Control Participants
Outcome Variable
PoP Full Sample (N=25),
Contamination Control,
Pure Control
PoP Completers (N=16),
Contamination Control,
Pure Control
X
2
df p-value X
2
p-value df
Re-arrested (yes/no) 3.158
2 .269 2.626
.206
2
Re-offended (yes/no) .463 2 .558 1.168 .793 2
Note: Re-offended=participants with one or more incident report and/or one or more re-arrest. Re-arrests and
Re-offending from post-intervention completion to 6-month follow-up. Covariates included in model:
ethnicity, parenting status, security level, and baseline GAD score.

Discussion
The physical, financial, and psychological cost of institutional misconduct has become a
pervasive problem in correctional institutions (Welsh et al., 2007) and rates of infractions have
been linked to future recidivism (Trulson et al., 2011). Women are increasingly incarcerated in
the US (Minton & Zeng, 2015) and exhibit similar rates of misconduct as their male counterparts
(Steiner & Wooldredge, 2014a), yet they remain absent in most of the violence prevention
literature. Existing program evaluations have failed to assess misconduct (Kubiak et al., 2015) or
have been limited to small juvenile samples (Goldstein et al., 2007) or non-randomized designs
(Eamon et al., 2001; Messina et al., 2016). The current randomized trial examined the impact of
a conflict resolution program (Prison of Peace; PoP; Kaufer & Noll, 2014) for female inmates at
a large county jail; the intervention group was compared to two control groups with varying
levels of contact with program participants. Results indicated that the 7-week program was
effective at enhancing conflict resolutions skills directly after program completion, but
ineffective at reducing institutional misconduct six months later. Moreover, and contrary to our
primary hypothesis, PoP participants actually exhibited an increase in misconduct at T3.
PRISON OF PEACE RANDOMIZED TRIAL
67

Interestingly, inmates housed in the same module as PoP participants exhibited a reduction in
post-intervention incident reports despite a marked increase in self-reported conflict and anger.
Several hypotheses are presented to explain this potentially iatrogenic effect of the program on
participant misconduct.
First, the restrictions of the jail setting likely affected program fidelity. A key element of
PoP, and indeed to most behavioral interventions (e.g., Barlow, 2008; Beck, 2011), is the
consistent practice of skills – namely, leading Peace Circles. As part of the program, inmates
were required to engage in discussion groups when conflict situations naturally arose in their
housing units. This enabled students to practice the skills learned in the classroom and gain
deeper understanding of PoP concepts, such as perspective-taking, communication,
collaboration, and strengthening community bonds. Outside the one allotted hour of program
time, the women in Mod A (PoP and Contamination Control participants) were placed on
constant lockdown and provided with little opportunity to interact outside of their two- to four-
person group of cellmates. The program instructor therefore deviated from the manualized
intervention, visited the jail on non-class days, and pulled women out of their cells in order to
create opportunities for practice. Unfortunately, these exercises did not accurately match
naturally-occurring situations inmates would typically address during Peace Circles.
Second, our study, which only measured the effects of the first PoP workshop (i.e.,
Peacemaker-Student level), changed the conditions of the program because students did not
continue on to subsequent PoP levels (i.e., Mentor, Rising Trainer, Trainer, and Mediator). After
the 7-week course, the instructor left and inmates returned to their 23-hour-lockdown routines.
Because inmates knew that further opportunities to advance through the PoP levels would not be
available to them, they may have felt less motivated to continue practicing and refining their
PRISON OF PEACE RANDOMIZED TRIAL
68

newly acquired skills. Moreover, they may have attempted to engage in conflict resolution in a
non-effective way which may have led to more incident reports. Also, the PoP instructor
condensed her eight-lesson program from 12 weeks to 7 weeks. This was done to limit
participant attrition (i.e., due to prison transfers and community releases) and to reduce the
amount of time movement was frozen within the jail. However, shortening the time frame of the
program also limited the number of opportunities students had to complete Peace Circles.
Third, deputy attitudes may have contributed to increased incident reports for PoP
participants, and thus IRTS may not have been valid representations of institutional misconduct.
For example, during the course of the program, researchers were told that a Mod A deputy
announced the following to inmates housed there: “No one gets to shower on Tuesdays or
Thursdays because the Prison of Peace students are in class and we can’t let everyone out at the
same time. So, you can thank them for not getting your showers today” (Laurel Kaufer personal
communication, August 25, 2016). If this report is accurate, two study confounds were
potentially introduced: hostile deputy attitudes and the prompting of hostile attitudes for
Contamination Controls toward PoP participants.
The creation of a hostile climate may account for the increase of IRTS for PoP
participants and the rise in self-reported conflicts and anger in the Contamination Control group
at T2. It is possible that deputies, knowing which women were involved in PoP, highly
scrutinized those inmates’ behaviors and thereby inflated the number of incidents reported for
those women. This may explain why Contamination Controls showed a significant decrease in
IRTS in the same time period that PoP showed an increase. However, when participants were
asked to indicate the frequency of various conflicts they were involved with over time (i.e., from
general disagreements to situations involving violence), a counterintuitive pattern emerged in
PRISON OF PEACE RANDOMIZED TRIAL
69

relation to official reports of misconduct (Figures 7 and 12). Within Mod A, PoP IRTS increased
while conflicts remained stable, but Contamination Control IRTS decreased and conflicts
increased.
One interpretation of these findings may be that some confrontational situations were
viewed as “conflict” by Contamination Controls but not by PoP participants, who were then
disproportionately written up for misconduct arising from those interactions. This is consistent
with PoP outcomes on the conflict resolution measure (i.e., decreases in perceived hostility in
conflict situations). Further, deputy attitudes regarding PoP may have led to biased sanctions.
Indeed, several sources indicated that some deputies felt that allowing the students out of their
cells was undeserved “special treatment” (Laurel Kaufer personal communication, August 25,
2016; two inmates, personal communication, October 7, 2016). Some research shows a link
between officer attitudes and rates of incident reporting (Freeman, 2003; Griffin, 2002; Lerman
& Page, 2012). However, attitudes among the jail deputies were not measured in the present
study and warrant attention in future evaluations. Lastly, it should be noted that the Conflict
Scale was the least validated measure in our assessment battery but was the only available
inmate-based conflict questionnaire in the literature; therefore, results should be interpreted with
caution.
Lastly, the Peacemaker course may have been iatrogenic in terms of inmate misconduct.
For one, women who completed PoP exhibited an interesting pattern of hostile behavior over
time. During the program, there were zero hostile incidents recorded for those 16 inmates, yet
once the course ended, a significant increase in hostile IRTS was observed. PoP participants may
have been incentivized to behave more cooperatively during the intervention phase, as they
would have been placed on lockdown and missed classes if they behaved otherwise. There were
PRISON OF PEACE RANDOMIZED TRIAL
70

also fewer opportunities for hostile conflict as the women spent more hours engaged in the
workshop. Further, PoP students were guided and supervised by the program instructor, which
may have been effective at curtailing their hostile impulses. However, once the intervention
ended, PoP completers may have felt a sense of confidence and agency in their conflict
resolution skills, thereby choosing to engage in conflict situations more often. Yet, without
consistent guidance from the program or sufficient skill practice, these approaches to conflict
may have been perceived as abrasive and led to more hostile situations.
Regarding the positive effects of the program, women randomized to PoP appeared to be
less angry and depressed and were also less hostile when approaching hypothetical conflict
situations after program completion, although these effects were not maintained six months later.
The decrease in depression symptoms may have resulted from PoP participation or alternatively
from a change in the weekly routines of PoP inmates. As for the latter explanation, PoP inmates
may have emotionally benefitted from a lift in the 23-hour lockdown procedures, whereas
Contamination Controls had to remain in their cells and observe their neighbors’ newfound
freedom of movement two days a week. This issue is particularly salient for incarcerated women
given that nearly 30% of female inmates report significant depressive symptoms (Fazel &
Danesh, 2002). Similar to national averages, 21% to 31% of our sample reported symptoms of
severe depression on the PHQ-9 across the three time points and PoP participants exhibited the
most consistent decline over time. Lastly, PoP completers reported a decrease in parenting stress
at follow-up. This effect may have stemmed from the program’s emphasis on active listening and
communication, which are skills that are linked to parenting satisfaction for incarcerated women
(Loper & Tuerk, 2011).
PRISON OF PEACE RANDOMIZED TRIAL
71

In summary, we found significant treatment effects for nearly half of our outcome
domains at either post-treatment or follow-up. However, we did not find group differences in
relation to re-arrest or the third vignette composite, solution effectiveness. Further, no treatment
effects were found for self-efficacy, perspective taking, empathic concern, personal distress, and
frequency of contact with children, family, and friends. As noted earlier, PoP effects may have
been attenuated due to low dosage of the intervention, with students only receiving the
Peacemaker-Student workshop.
Jails versus Prisons: Future Directions
Prison of Peace was originally requested by and developed for women serving life in
prison who wanted to develop stronger community bonds and communication skills. However,
our randomized trial was conducted in a jail setting due to the favorable institutional allowances
made to accommodate randomization procedures, and to specifically test the effects of PoP in
this transient setting. Our results raise an important question pertaining to generalizability: was
this program evaluation representative of Prison of Peace effectiveness in prisons? First, we
consider differences between jail and prison correctional populations.
Women serving long sentences in prison may seek out and benefit from programs like
Prison of Peace, which depend on the development of trust and compassion to facilitate
communication skills. However, women serving shorter sentences in jails may not invest
themselves in peer relationships that are perceived as more transient. Past research indicates that
inmate relationships (non-romantic; non-gang related) are valued more by female prisoners than
male inmates and can positively influence the way in which women serve their time in prisons
(Clone & DeHart, 2014; Hart, 1995; Severance, 2005). However, inmates uninterested in
PRISON OF PEACE RANDOMIZED TRIAL
72

developing stronger bonds may artificially “use” these relationships without benefitting from the
potential impact these connections have on their behavior (Wright et al., 2013). Although PoP
completers may have worked on strengthening family relationships, which may be illustrated by
their parenting stress decreases, jail-based relationships may not have been valued as much.
Women in our sample may not have been engaged in the program as they would have in
a prison setting due to the uncertainty of their sentences. In general, jails house a variety of
offenders – individuals awaiting trial or prison transfers, and inmates with crimes that typically
warrant less than one year of incarceration. However, a shift occurred in California in 2011 when
the Criminal Justice Realignment law (AB-109) transferred low-level felony offenders from state
prisons to county facilities, which likely altered jail climates. Indeed, some research shows a link
between AB-109 and increases in jail-based violence (Caudill, Trulson, Marquart, Patten,
Thomas, & Anderson, 2014). During study recruitment, jail officials indicated that our two
housing units included AB-109 inmates (i.e., long sentences) and most participants assured
researchers that they would be at our jail facility for at least one year. Unfortunately,
incarceration status varied greatly across participants (e.g., pre-sentence/awaiting trial, short-
term or long-term sentences, prison transfer) and 45% of our sample was released before the 6-
month follow-up. This fact highlights the highly transient, heterogeneous, and unstable nature of
our sample compared to prison populations. Our evaluation outcomes, therefore, may not be
entirely generalizable to prisoners.
Further, our jail participants may have been more hostile, impulsive, and psychologically
unstable than inmates housed in prison. While this may seem counterintuitive, we posit that the
sudden transition to a carceral setting, separation from children and family, and restricted
freedom of movement may lead to more volatile behavior for jailed individuals. Regarding
PRISON OF PEACE RANDOMIZED TRIAL
73

mental health differences, a higher proportion of female inmates in jails (32%) report serious
psychological distress than do women in prisons (20%; Bronson & Berzofsky, Bureau of Justice
Statistics, 2017). Moreover, jail inmates with psychological distress are more likely to have
institutional assaults on their records than inmates who do not report mental health problems
(Bronson & Berzofsky, 2017). This is consistent with prior research, which considers
misconduct and distress to be the primary indicators of maladjustment in correctional settings
(Salisbury, Van Voorhis, & Spiropoulos, 2009; Steiner & Wooldredge, 2014b; Wright et al.,
2007). Thus, our sample may have been less responsive to the intervention compared to a prison
sample due to greater maladjustment and associated differences (i.e., more distressed and
hostile).
Second, setting disparities should be considered, with greater stability of deputies being
more common in prisons compared to jails. Whereas Mod A was overseen by several different
deputies every week, prisons have been reported to have the same correctional officers
supervising a housing unit for several months (Laurel Kaufer, personal communication, June
2018). High deputy turnover in Mod A may have led to low trust and weak bonds between staff
and inmates, which may have fostered hostile deputy attitudes. Indeed, some research indicates
that minimal officer-inmate contact leads to harsher officer attitudes and punishment of inmates
(Kelly, 2014). Additionally, the severe living conditions in our experimental module (i.e., 23-
hour lockdown) are generally not found in California women’s prisons where PoP students have
reported greater freedom of movement (Laurel Kaufer, personal communication, June 2018). As
previously mentioned, lockdown conditions and restricted peer interactions likely influenced
program fidelity.
PRISON OF PEACE RANDOMIZED TRIAL
74

Furthermore, jail deputies and prison officers may differ in their responses to misconduct
and attitudes toward inmates. A study conducted by Santos, Lane, and Gover (2012) showed that
officers in prison settings used more methods of informal control (e.g., removal of privileges;
reduced phone time) as opposed to formal methods (e.g., official sanctions and write-ups;
solitary confinement) compared to officers in jails. This difference in use of authority suggests
that our results might differ if the study had been carried out in a prison setting. Thus, future PoP
research should attempt to measure official incident reports and informal sanctions to
comprehensively capture misconduct. Additionally, given the aforementioned potential biases in
IRTS reporting in our sample, officer attitudes of inmates and the program itself should be
assessed. For example, the Klofas-Toch Measure of Professional Orientation (1982), which
categorizes officers as either believing or not believing in rehabilitation practices, has been found
to predict rule violation reporting for prison inmates (Freeman, 2003). This effect was also found
in a jail setting predicting readiness to use force (Griffin, 2002). Other research has shown that
officer attitudes regarding the purpose of imprisonment (i.e., punishment versus rehabilitation)
may impact their behavior toward inmates (Lerman & Page, 2012). Considering the major
influence that officers have over their wards, their attitudes, beliefs, and methods of discipline
warrant further attention.
Although the jail-based violence prevention literature is scarce with regard to female
inmates, several programs have been examined in prison settings, with peer educators receiving
much positive attention in recent years (Ashcraft & Anthony, 2011; Clone & DeHart, 2014). For
example, Beyond Violence (BV; Covington, 2013), a gender-responsive manualized violence-
prevention program, yielded positive outcomes for women who had been incarcerated for at least
a decade at the time of enrollment. Unfortunately, published program evaluations rarely include
PRISON OF PEACE RANDOMIZED TRIAL
75

archival records (e.g., incident reports; Kubiak et al., 2015) and even fewer include control
groups (Messina et al., 2016). A randomized trial of PoP in a prison setting would certainly add
to the literature on peer-led, gender-responsive programs for female inmates.
PoP Modifications and Future Evaluations
Following our evaluation, Laurel Kaufer, the PoP co-founder, modified the Peacemaker
portion of the program to address challenges she encountered during the implementation process.
To allow more opportunities for advancement, PoP’s first level was broken into two separate 16-
hour workshops – Circle Keeper and Peacemaker. Only participants who have graduated from
the Circle Keeper course, which introduces inmates to restorative justice, active listening, and
Peace Circles, are eligible to continue on to the Peacemaker workshop, which delves deeper into
the management of strong emotions. Since implementing this format in numerous prison settings,
Ms. Kaufer has found that voluntary dropout rarely occurs beyond the Circle Keeper workshop.
Moreover, most participants asked for letters of support from PoP to verify their participation
and request that they not be transferred until completion of the program. In the event that they
must be transferred, inmates request facilities where they can continue with the program.
However, the fact that some inmates are unwilling or unable to continue with the
program during the Circle Keeper workshop indicates that PoP may not be engaging and/or
effective for everyone. Our attrition analyses indicated that PoP dropouts had marginally lower
parenting stress compared to completers. No other significant differences were found, but, low
power may have impeded our ability to identify disparities. When examining raw means at
baseline, non-completers were older and had higher rates of “current arrest: violent” (44% v.
PRISON OF PEACE RANDOMIZED TRIAL
76

completers at 25%) and hostile IRTS (1.25 vs. completers at .44). Therefore, the program may
have been less engaging for more established and violent offenders.
Future PoP evaluations should include an examination of the entire program, from Circle
Keeper to Trainer levels, and should also measure the differential effectiveness of peer versus
non-peer educators. One of the unique and financially desirable elements of PoP is its self-
sustainability within a correctional setting once enough inmates have been certified trainers and
can organize their own courses.
Limitations
Our study had several strengths, including a quasi-experimental design (with
randomization to two of the three conditions), the freezing of inmate movement to minimize
crossover effects, and a mix of self-report and archival outcome data. Nonetheless, several
limitations should be considered when interpreting results. First, the restrictions of program
classes limited the size of our total sample (N=75). Therefore, when a significant portion of our
participants were released during the study and could not be reached to complete assessments,
power to detect significant group differences was negatively affected. Moreover, participants
with missing data were more likely to be non-Black, non-Hispanic parents with low anxiety and
a less serious incarceration history, thereby introducing potential bias in our results.
Additionally, the $10 incentives for completing each assessment may not have been a sufficient
reward for completing lengthy and personal questionnaires. Future evaluators should aim to
provide adequate compensation, particularly for released prisoners who may have more
opportunities to earn income and may be less incentivized to complete follow-up measures.
PRISON OF PEACE RANDOMIZED TRIAL
77

A great strength of our study was the inclusion of archival records for at least 92% of our
sample. IRTS data included detailed deputy observations of inmate behavior and arrest data
included various aliases and booking numbers for each inmate in our study. Moreover, we
controlled for the number of days inmates were in the community or at the jail by creating re-
offending ratios, as opposed to raw number of offenses. However, participant arrest data was
limited to the jail’s database and did not include incarceration outside of California. Further,
once inmates were transferred to prison and crossed agency types (e.g., from county jail to
California Department of Corrections and Rehabilitation), we did not have access to their
institutional records and were not informed when they were released. Therefore, potentially
salient information was not captured pertaining to misconduct and re-arrest for transferred
participants.
Further, living arrangements for participants housed in the experimental module impacted
program fidelity and potentially affected study outcomes. Although jail officials gave researchers
access to the two high-risk modules at their facility, these differed significantly with regard to
inmate risk level (i.e., lower risk inmates were housed in the Pure Control module) and living
situation (i.e., 23-hour lockdown only in PoP/Contamination Control module). Lockdown
regulations prevented PoP participants from completing their required Peace Circles, which may
have hindered their progress. Further, an iatrogenic effect of PoP may have occurred partially
due the lift in lockdown for program participants. Contamination Controls, as well as deputies
who may have disagreed with PoP’s “special treatment” of the inmates, may have fostered
hostility toward PoP students. A more direct comparison would have been to release
Contamination Controls from their cells for the same amount of time as PoP participants. Also,
this would have controlled for the potential confound that freedom of movement had on
PRISON OF PEACE RANDOMIZED TRIAL
78

participant outcomes (e.g., anger, depression). However, it is unlikely that this method would
have been accepted by jail officials.
Additional limitations included the skewed distribution of certain outcomes (e.g., hostile
IRTS), program effectiveness being measured for one class and one PoP instructor, ethnicity
differences across conditions (i.e., Contamination Control including a larger proportion of Black
inmates), and lack of generalizability of study findings to other prison-based PoP programs.
Indeed, PoP students in California prisons are able to graduate from the Peacemaker course and
subsequently complete the next levels and the Mediator workshop.
Conclusion
This was the first randomized trial of a conflict resolution program for female inmates in
a jail setting. Our study adds to the current and limited literature on violence prevention
initiatives for the growing population of female offenders (Eamon et al., 2001; Goldstein et al.,
2007; Kubiak et al., 2015; Messina et al., 2016) and provides some support for the teaching of
problem-solving and communication skills. Although program participants appeared to have
exhibited an increase in misconduct compared to controls, this effect may have been driven by
living situations (i.e., 23-hour lockdown) and potentially hostile deputy attitudes. PoP
programming was effective at enhancing conflict resolution skills and decreasing self-reported
depression and anger at post-completion compared to both control groups. Moreover, program
completers exhibited a sharp decrease in parenting stress at 6-month follow-up. However, it is
unlikely that these results are fully generalizable to most PoP programs currently being
implemented in prison settings, as population and facility differences are likely to impact
outcomes. Future evaluations should measure the effects of the entire PoP program (i.e., from
PRISON OF PEACE RANDOMIZED TRIAL
79

Circle Keeper-student to Mediator-trainer) in prison settings, examine the impact of peer
educators, and include measures of correctional officer attitudes.

PRISON OF PEACE RANDOMIZED TRIAL
80

References
Andrews, D. & Bonta, J. (1994). The psychology of criminal conduct. Cincinnati, OH: Anderson
Pub. Co.
Andrews, D. A., & Bonta, J. (2003). The psychology of criminal conduct (3rd ed.) Anderson
Publishing Co, Cincinnati, OH. Retrieved from
http://search.proquest.com.libproxy1.usc.edu/docview/620953565?accountid=14749
Andrews, D. A., Zinger, I., Hoge, R. D., Bonta, J., Gendreau, P., & Cullen, F. T. (1990). Does
correctional treatment work? A clinically relevant and psychologically informed meta-
analysis. Criminology, 28, 369-397.
APA Publications and Communications Board Working Group on Journal Article Reporting
Standards (2008). The American Psychologist, 63, 839-851.
Ashcraft, L., & Anthony, W. (2011, November 15). Inmates thrive with peer support training.
Behavioral Healthcare. Retrieved from www.behavioral.net
Austin, J., Naro-Ware, W., Ocker, R., Harris, R., & Allen, R. (2012). Evaluation of the Current
and Future Los Angeles County Jail Population. Retrieved from http://www.jfa-
associates.com/publications/
Auty, K. M., Cope, A., & Liebling, A. (2017). Psychoeducational programs for reducing prison
violence: A systematic review. Aggression and Violent Behavior, 33, 126.
Barlow, D. H. (2008). Clinical Handbook of Psychological Disorders: A Step-By-Step Treatment
Manual (4th edition). New York: Guilford Press.
Bartlett, A., Jhanji, E., White, S., Anne Harty, M., Scammell, J., & Allen, S. (2015).
Interventions with women offenders: A systematic review and meta-analysis of mental
health gain. The Journal of Forensic Psychiatry & Psychology, 26, 133-165.
PRISON OF PEACE RANDOMIZED TRIAL
81

Beck, J. S. (2011). Cognitive behavior therapy: Basics and beyond (2nd ed. ed.) Guilford Press,
New York, NY.
Beck, R., & Fernandez, E. (1998). Cognitive-behavioral therapy in the treatment of anger: A
meta-analysis. Cognitive Therapy and Research, 22, 63-74.
Berry, J.O., & Jones, W.H. (1995). The parental stress scale: Initial psychometric evidence.
Journal of Social and Personal Relationships, 12, 463-472.
Bolton, R., 1929. (1986). People skills: How to assert yourself, listen to others, and resolve
conflicts (1st Touchstone ed.). New York: Simon & Schuster.
Bureau of Justice Statistics. (2004). Survey of Inmates in State Correctional Facilities. [Data
file]. Retrieved from http://www.bjs.gov/index.cfm?ty=dcdetail&iid=275#Questionnaires
Bronson J. & Berzofsky, M. (2017). Indicators of Mental Health Problems Reported by Prisoners
and Jail Inmates, 2011-2012. Bureau of Justice Statistics. Retrieved from
http://www.bjs.gov/index.cfm?ty=pbdetail&iid=5946
Buss, A. H., & Perry, M. P. (1992). The aggression questionnaire. Journal of Personality and
Social Psychology, 63, 452-459.
California Sentencing Institute (2016). Data downloaded from the Center on Juvenile and
Criminal Justice: 2016 Data Report on Incarcerated Adults. Retrieved from:
http://casi.cjcj.org/Adult/Los-Angeles
Cao, L., Zhao, J. and Van Dine, S. (1997) Prison disciplinary tickets: A test of the deprivation
and importation models. Journal of Criminal Justice, 25, 103–113.
Casey-Acevedo, K., & Bakken, T. (2001). The effect of time on the disciplinary adjustment of
women in prison. International Journal of Offender Therapy and Comparative
Criminology, 45, 489-492.
PRISON OF PEACE RANDOMIZED TRIAL
82

Caudill, J. W., Trulson, C. R., Marquart, J. W., Patten, R., Thomas, M. O., & Anderson, S.
(2014). Correctional destabilization and jail violence: The consequences of prison
depopulation legislation. Journal of Criminal Justice, 42, 500-506.
Chamberlain, A. (2012). Offender rehabilitation: Examining changes in inmate treatment
characteristics, program participation, and institutional behavior. Justice Quarterly, 29,
183-228.
Clone, S., & DeHart, D. (2014). Social support networks of incarcerated women: Types of
support, sources of support, and implications for reentry. Journal of Offender
Rehabilitation, 53, 503-521.
Covey, M. K., & Dengerink, H. A. (1984). Development and validation of a measure of
heterosocial conflict resolution ability (relational behaviors survey). Behavioral
Assessment, 6, 323-332.
Covington S. (2013). Beyond Violence: A Prevention Program for Criminal Justice-Involved
Women. (John Wiley & Sons, Inc., Hoboken, NJ).
Craddock, A. (1996). A comparative study of male and female prison misconduct careers. The
Prison Journal, 76, 60-82.
D'Zurilla, T. J., & Goldfried, M. R. (1971). Problem solving and behavior modification. Journal
of Abnormal Psychology, 78, 107-126. doi:10.1037/h0031360
Davis, M. H. (1980). A multidimensional approach to individual differences in empathy. JSAS
Catalog of Selected Documents in Psychology, 10, 85.
Duhart, D. T. (2001). Violence in the workplace, 1993–99 (Bureau of Justice Statistics Special
Report NCJ 190076). Retrieved from the U.S. Department of Justice, Office of Justice
Programs, Bureau of Justice Statistics website:
PRISON OF PEACE RANDOMIZED TRIAL
83

Eamon, K. C., Munchua, M. M., & Reddon, J. R. (2001). Effectiveness of an anger management
program for women inmates. Journal of Offender Rehabilitation, 34, 45-60.
Faul, F., Erdfelder, E., Lang, A., & Buchner, A. (2007). G*Power 3: A flexible statistical power
analysis program for the social, behavioral, and biomedical sciences. Behavior Research
Methods, 39, 175-191.
Fazel, S., & Danesh, J. (2002). Mental disorders in prisoners. The Lancet, 360, 573.
Feinstein, J. (2011). Reforming Adult Felony Probation to Ease Prison Overcrowding: An
Overview of California S.B. 678, 14 Chap. L. Rev. 375.
Felitti, V. J., Anda, R. F., Nordenberg, D., Williamson, D. F., Spitz, A. M., Edwards, V., Marks,
J. S. (1998). Relationship of childhood abuse and household dysfunction to many of the
leading causes of death in adults: The adverse childhood experiences (ACE) study.
American Journal of Preventive Medicine, 14, 245-258.
Foglia, W. D. (1996). Exploring the interaction of cognitive and social learning influences to
enhance understanding of self-reported delinquency among inner city youths (Order No.
9627920). Available from ProQuest Dissertations & Theses Full Text; ProQuest
Dissertations & Theses Global. (304312164). Retrieved from
http://search.proquest.com.libproxy1.usc.edu/docview/304312164?accountid=14749
Freeman, R. M. (2003). Social distance and discretionary rule enforcement in a women's prison.
The Prison Journal, 83, 191-205.
French, S. A., & Gendreau, P. (2006). Reducing prison misconducts: What works! Criminal
Justice and Behavior, 33, 185-218.
Gendreau, P., Goggin, C., & Cullen, F. (1999). The effects of prison sentences on recidivism.
Report to the Corrections Research and Development and Aboriginal Policy Branch (Cat.
PRISON OF PEACE RANDOMIZED TRIAL
84

No. J42-87/1999E), Solicitor General of Canada. Ottawa: Public Works and Government
Services Canada.
Gendreau, P., & Keyes, D. (2001). Making prisons safer and more humane environments.
Canadian Journal of Criminology, 43, 123-130.
Glaze, L. & Maruschak, L. (2010). Parents in prison and their minor children. Washington, DC:
U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics.
Retrieved from www.bjs.gov/content/pub/pdf/pptmc.pdf
Goldstein, N. E. S., Dovidio, A., Kalbeitzer, R., Weil, J., & Strachan, M. (2007). Anger
management for female juvenile offenders: Results of a pilot study. Journal of Forensic
Psychology Practice, 7, 1-28.
Gonzalez, P., Romero, T., & Cerbana, C. B. (2007). Parent education program for incarcerated
mothers in Colorado. Journal of Correctional Education, 58, 357-373.
Gover, A. R., Perez, D. M., & Jennings, W. G. (2008). Gender differences in factors contributing
to institutional misconduct. The Prison Journal, 88, 378-403.
Griffin, M. L. (2002). The influence of professional orientation on detention officers' attitudes
toward the use of force. Criminal Justice and Behavior, 29, 250-277.
Hall, E., Prendergast, M. L, Wellisch, J., Patten, M., & Cao, Y. (2004). Treating drug-abusing
women prisoners: An outcome evaluation of the Forever Free program. The Prison
Journal, 84, 81-105
Hart, C. B. (1995). Gender differences in social support among inmates. Women and Criminal
Justice, 6, 67-88.
PRISON OF PEACE RANDOMIZED TRIAL
85

Hastings, M. E., Krishnan, S., Tangney, J. P., & Stuewig, J. (2011). Predictive and incremental
validity of the violence risk appraisal guide scores with male and female jail inmates.
Psychological Assessment, 23, 174-183.
Houck, K. F., & Loper, A. B. (2002). The relationship of parenting stress to adjustment among
mothers in prison. American Journal of Orthopsychiatry, 72, 548–558.
James, D. & Glaze, L. (2006). Mental health problems of prison and jail inmates. Washington,
DC: U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics.
Retrieved from www.bjs.gov/content/pub/pdf/mhppji.pdf
Kaufer, L. & Noll, D. (2014). Peacemaker Trainer Workshop Manual. Prison of Peace: From
serving life to a life of service. Unpublished Manual.
Kaufer, L., Noll, D. E., and Mayer, J. (2014). Prisoner facilitated mediation: Bringing peace to
prisons and communities. Cardozo Journal of Conflict Resolution, 16, 187-219.
Kaya, C., Blake, J., & Chan, F. (2015). Peer-mediated interventions with elementary and
secondary school students with emotional and behavioural disorders: A literature review.
Journal of Research in Special Educational Needs, 15, 120-129.
Kelly, D. (2014). Punish or reform? predicting prison staff punitiveness. Howard Journal of
Criminal Justice, 53, 49-68.
Kennon, S., Mackintosh, V., & Myers, B. (2009). Parenting education for incarcerated mothers.
Journal of Correctional Education, 60, 10–30.
Klofas, J. M., & Toch, H. (1982). The guard subculture myth. Journal of Research in Crime and
Delinquency, 19, 23.
Kroenke, K, Spitzer, R.L., Williams, J.B.W. (2001). The PHQ-9: Validity of a brief depression
severity measure. Journal of General Internal Medicine. 16, 606-613.
PRISON OF PEACE RANDOMIZED TRIAL
86

Kubiak, S., Kim, W., Fedock, G., & Bybee, D. (2015). Testing a violence-prevention
intervention for incarcerated women using a randomized control trial. Research on Social
Work Practice., 25, 334.
Landenberger, N. A., & Lipsey, M. W. (2005). The positive effects of cognitive-behavioral
programs for offenders: A meta-analysis of factors associated with effective treatment.
Journal of Experimental Criminology, 1, 451-476.
Larkin H., Felitti VJ., & Anda R.F. (2014) Social work and Adverse Childhood Experiences
(ACE) research: implications for practice and health policy. Social Work in Public
Health, 29, 1-16
Lerman, A.E., & Page, J. (2012). The state of the job: An embedded work role perspective on
prison officer attitudes. Punishment & Society, 14, 503-529.
Lieberman, M. D., Eisenberger, N. I., Crockett, M. J., Tom, S. M., Pfeifer, J. H., & Way, B. M.
(2007). Putting feelings into words: Affect labeling disrupts amygdala activity in
response to affective stimuli. Psychological Science, 18, 421-428.
Lipsey, M. W., & Cullen, F. T. (2007). The effectiveness of correctional rehabilitation: A review
of systematic reviews. Annual Review of Law and Social Science, 3, 297-320.
Loper, A. B., & Tuerk, E. H. (2011). Improving the emotional adjustment and communication
patterns of incarcerated mothers: Effectiveness of a prison parenting intervention.
Journal of Child and Family Studies, 20, 89-101.
Loper, A. B., Carlson, W., Levitt, L., & Scheffel, K. (2009). Parenting stress, alliance, child
contact and adjustment of imprisoned mothers and fathers. Journal of Offender
Rehabilitation, 48, 483–503.
PRISON OF PEACE RANDOMIZED TRIAL
87

Loper, A. B., Thompson, C., Tuerk, E., Daoust, S., Kramer, M. E., & Komarovskaya, I., et al.
(2007). Parenting from the inside: Making the mother-child connection inmate manual.
University of Virginia.
Loper, A. B., & Tuerk, E. (2006). Parenting programs for incarcerated parents: Current research
and future directions. Criminal Justice Policy Review, 17, 407–427.
Löwe, B., Kroenke, K., Spitzer, R. L., & Williams, J. B. W. (2006). A brief measure for
assessing generalized anxiety disorder: The GAD-7. Archives of Internal Medicine, 166,
1092-1097.
Macpherson, S. L. (1986). An investigation of the components of anger management as applied
to an incarcerated population Available from PsycINFO. (617333680; 1987-54249-001).
Retrieved from http://libproxy.usc.edu/login?url=https://search-proquest-
com.libproxy1.usc.edu/docview/617333680?accountid=14749
Marcum, C. D., Hilinski-rosick, C., & Freiburger, T. L. (2014). Examining the correlates of male
and female inmate misconduct. Security Journal, 27, 284-303.
Marsh, D.T. (1982). The development of interpersonal problem solving among elementary
school children. The Journal of Genetic Psychology, 140, 107.
Messina, N., Burdon, W., & Prendergast, M. (2006). Prison-based treatment for drug-dependent
women offenders: Treatment versus no treatment. Journal of Psychoactive Drugs, 3, 333-
343.
Messina, N. P., Braithwaite, J., Calhoun, S., & Kubiak, S. (2016). Examination of a violence
prevention program for female offenders. Violence and Gender, 3, 143-149.
PRISON OF PEACE RANDOMIZED TRIAL
88

Minton, T., & Zeng, Z. (2015). Jail inmates at midyear 2014. Washington, DC: U.S. Department
of Justice, Office of Justice Programs, Bureau of Justice Statistics. Retrieved from
http://www.bjs.gov/index.cfm?ty=pbdetail&iid=5299
Pearson, F. S., Lipton, D. S., Cleland, C. M., & Yee, D. S. (2002). The effects of
behavioral/cognitive-behavioral programs on recidivism. Crime and Delinquency, 48,
476-496. Retrieved from
http://search.proquest.com.libproxy1.usc.edu/docview/221426319?accountid=14749
Petersila, J. (2006). Understanding California corrections: A policy research program report.
California Policy Research Center, University of California.
Phillips, S., & Cooney, M. (2005). Aiding peace, abetting violence: Third parties and the
management of conflict. American Sociological Review, 70, 334-354.
Pierce, M. B., Freiburger, T. L., Chapin, J. R., Epling, B., & Madden, T. J. (2018). Assessing the
impact of visitation on inmate misconduct within a county jail. Security Journal, 31, 1-
20.
Prendergast, M. L., Hall, E. A., Wexler, H. K., Melnick, G., & Yao, Y. (2004). Amity prison-
based therapeutic community: 5-year outcomes. The Prison Journal, 84, 36-60.
Prince, J. R. (1994). Aggression management training: An evaluation (Order No. 9502834).
Available from ProQuest Dissertations & Theses Full Text; ProQuest Dissertations &
Theses Global. (304104403). Retrieved from
http://search.proquest.com.libproxy1.usc.edu/docview/304104403?accountid=14749
Ruiz, M. A., Zamorano, E., García-Campayo, J., Pardo, A., Freire, O., & Rejas, J. (2011).
Validity of the GAD-7 scale as an outcome measure of disability in patients with
PRISON OF PEACE RANDOMIZED TRIAL
89

generalized anxiety disorders in primary care. Journal of Affective Disorders, 128, 277-
286.
Salisbury, E. J., Van Voorhis, P., & Spiropoulos, G. V. (2009). The predictive validity of a
gender-responsive needs assessment. Crime & Delinquency, 55, 550-585.
Santos, S. D., Lane, J., & Gover, A. R. (2012). Former prison and jail inmates' perceptions of
informal methods of control utilized by correctional officers. American Journal of
Criminal Justice: AJCJ, 37, 485-504.
Schenk, A. M., & Fremouw, W. J. (2012). Individual characteristics related to prison violence: A
critical review of the literature. Aggression and Violent Behavior, 17, 430-442.
Schwarzer, R. (1993). Measurement of perceived self-efficacy. Psychometric scales for cross-
cultural research. Berlin, Germany: Freie Universität Berlin.
Schwarzer, R., & Jerusalem, M. (1995). Generalized Self-Efficacy scale. In J. Weinman, S.
Wright, & M. Johnston, Measures in health psychology: A user’s portfolio. Causal and
control beliefs (pp. 35-37). Windsor, UK: NFER-NELSON.
Slaby, R. G., & Guerra, N. G. (1988). Cognitive mediators of aggression in adolescent offenders:
I. assessment. Developmental Psychology, 24, 580-588.
Sorbello, L., Eccleston, L., Ward, T., & Jones, R. (2002). Treatment needs of female offenders:
A review. Australian Psychologist, 37, 198-205.
Steiner, B., Butler, H. D., & Ellison, J. M. (2014). Causes and correlates of prison inmate
misconduct: A systematic review of the evidence. Journal of Criminal Justice, 42, 462-
470.
PRISON OF PEACE RANDOMIZED TRIAL
90

Steiner, B., & Wooldredge, J. (2014a). Comparing self-report to official measures of inmate
misconduct. Justice Quarterly, 31, 1074. Retrieved from
http://search.proquest.com.libproxy1.usc.edu/docview/1562483393?accountid=14749
Steiner, B., & Wooldredge, J. (2014b). Sex differences in the predictors of prisoner misconduct.
Criminal Justice and Behavior, 41, 433-452.
Stevahn, L., Johnson, R., Johnson, D., & Schultz, R. (2002). Effects of conflict resolution
training integrated into a high school social studies curriculum. The Journal of Social
Psychology, 142, 305-331.
Swavola, E., Riley, K., & Subramanian, R. (2016). Overlooked: Women and jails in an era of
reform. Vera Institute of Justice. Retrieved from Policy File Index Retrieved from
http://libproxy.usc.edu/login?url=https://search-proquest-
com.libproxy1.usc.edu/docview/1918305153?accountid=14749
Thompson, C., & Loper, A. B. (2005). An analysis of differences based on sentence length.
Criminal Justice and Behavior, 32, 714-732.
Tripodi, S. J., Bledsoe, S. E., Kim, J. S., & Bender, K. (2011). Effects of correctional-based
programs for female inmates: A systematic review. Research on Social Work Practice,
21, 15-31.
Twisk, J.W.R. (2013) Applied longitudinal data analysis for epidemiology; a practical guide, 2d
ed. Reference and Research Book News, 28.
Van Voorhis, P., Wright, E. M., Salisbury, E., & Bauman, A. (2010). Women's risk factors and
their contributions to existing Risk/Needs assessment: The current status of a gender-
responsive supplement. Criminal Justice and Behavior, 37, 261-288.
PRISON OF PEACE RANDOMIZED TRIAL
91

Ward, T., & Stewart, C. (2003). Criminogenic needs and human needs: A theoretical model.
Psychology, Crime, and Law, 9, 125-143.
Welsh, W. N., McGrain, P., Salamatin, N., & Zajac, G. (2007). Effects of prison drug treatment
on inmate misconduct: A repeated measures analysis. Criminal Justice and Behavior, 34,
600-615.
Wilson, D., Gallagher, C. A., & MacKenzie, D. L. (2000). A meta-analysis of corrections-based
education, vocation, and work programs for adult offenders. Journal of Research in
Crime and Delinquency, 37, 347-368.
Wolff, N., Blitz, C. L., Shi, J., Siegel, J., & Bachman, R. (2007). Physical violence inside
prisons: Rates of victimization. Criminal Justice and Behavior, 34, 588-599.
Wright, E. M., Salisbury, E. J., & Van Voorhis, P. (2007). Predicting the prison misconducts of
women offenders: The importance of gender-responsive needs. Journal of Contemporary
Criminal Justice, 23, 310-340.
Zeng, Z. (2016). Jail inmates in 2016. Washington, DC: U.S. Department of Justice, Office of
Justice Programs, Bureau of Justice Statistics. Retrieved from
https://www.bjs.gov/content/pub/pdf/ji16.pdf

PRISON OF PEACE RANDOMIZED TRIAL 92

APPENDIX A: Detailed CONSORT Flow Diagram

PRISON OF PEACE RANDOMIZED TRIAL 93

APPENDIX B: Assessment Battery Questionnaires
Demographics (T1)

1. Age: ______________

2. Ethnicity/Race:
a) White
b) Hispanic or Latino
c) Black or African American
d) Native American or American Indian
e) Asian / Pacific Islander
f) Multiracial
g) Other: _____________________

3. Primary Language:
a) English
b) Spanish
c) Other: ____________

4. Were you born in the United States or some other country?
a) US
b) Other country: ____________

5. If you were born in another country, how long have you been living in the United States? ________

6. Relationship Status:
a) Single, never married
b) Married or domestic partnership
c) Committed relationship
d) Widowed
e) Divorced
f) Separated

7. Number of children: ______ and their age(s): ___________________________

8. Number of children under your care or living in your household prior to arrest: ______ and their age(s):
___________________________

9. What is your total household income?
a) Less than $10,000
b) $10,000 to $19,999
c) $20,000 to $29,999
d) $30,000 to $39,999
e) $40,000 to $49,999
f) $50,000 to $59,999
g) $60,000 to $69,999
h) $70,000 to $79,999
PRISON OF PEACE RANDOMIZED TRIAL 94

i) $80,000 to $89,999
j) $90,000 to $99,999
k) $100,000 to $149,999
l) $150,000 or more

10. Were you or anyone permanently living with you receiving public assistance or welfare, for example,
Temporary Assistance for Needy Families (TANF), food stamps, Medicaid, Women, Infants, and Children
Program (WIC), or housing assistance, before you were arrested?
a) Yes
b) No
c) Don't know

11. Most recent employment status before arrest:
a) Employed for wages
b) Self-employed
c) Out of work and looking for work
d) Out of work but not currently looking for work
e) A homemaker
f) A student
g) Military
h) Retired
i) Unable to work

12. Highest level of education completed:
a) Middle School
b) High school, but did not graduate
c) High school diploma or GED
d) Some college, no degree
e) AA or other two-year degree
f) Vocational school
g) Bachelor’s degree
h) Graduate degree

13. Religious affiliation/belief system: __________________________

14. When you were growing up, which caretaker did you live with most of the time? (If more than one relative,
mark closest blood relative)
a) Both parents (including one step-parent)
b) Mother
c) Father
d) Grandparent(s)
e) Other relatives (Include step-relations)
f) Friends
g) Foster homes
h) Homeless
i) Agency or institution (including correctional or religious institution)
j) Someone else - Specify __________________________

PRISON OF PEACE RANDOMIZED TRIAL 95

Arrest history
15. How many times have you been arrested? ___________
16. How many times have you been incarcerated at a jail? _____________
17. How many times have you been incarcerated at a prison? _____________
18. What is the longest time you have spent incarcerated? ____________
19. What is your current offense/charges against you? __________________________________
20. What other offenses have you been charged with in the past?
_______________________________________________________________________________________
_______________________________________________________________________________________
___________________________________________________

21. Have you ever been convicted for a (circle all that apply):
a) Drug offense (for example, possession of drugs)
b) Violent offense (for example, battery)
c) Property offense (for example, robbery)
d) None of the above

22. Please check all that apply about your involvement with gangs:
a) I am currently a member of a gang here at the jail
b) I used to be a member of a gang here at the jail, but I am not anymore
c) I have never been a member of a gang here at the jail
d) I have been a member of a gang at least once in my life (in prison or in the community)
e) I have never been a member of a gang (in prison or in the community)

23. Has anyone in your immediate family ever been incarcerated (went to jail or prison)?
a) Yes. Please indicate which member: ____________________________
b) No
c) Not sure

24. Have you ever taken part in a program while incarcerated? For example, a GED class or substance abuse
education.
a) Yes
b) No
c) Not sure

25. If you selected YES or NOT SURE to the question above, please list or describe the class(es) you have taken
in the past:
_______________________________________________________________________________________
_______________________________________________________________________________________
___________________________________________________

26. Have you ever had a substance abuse problem? This can include being addicted to or commonly using
alcohol, heroin, methamphetamines, cocaine, marijuana, or other illegal drugs?
a) Yes
b) No
c) Not sure

PRISON OF PEACE RANDOMIZED TRIAL 96

27. Were you using alcohol or other drugs at the time of your current or any past arrest?
a) Yes
b) No
c) Not sure

28. Briefly describe your substance use history and treatments you may have received:
_______________________________________________________________________________________
_______________________________________________________________________________________
___________________________________________________

29. Have you ever been told by a mental health professional, such as a psychiatrist or psychologist, that you had
(circle all that apply):
a) A depressive disorder
b) Manic-depression, bipolar disorder, or mania
c) Schizophrenia or another psychotic disorder
d) Post-traumatic stress disorder
e) Another anxiety disorder, such as a panic disorder
f) A personality disorder (such as an antisocial or borderline personality disorder)
g) Any other mental or emotional condition - Specify__________________________

30. Were you experiencing symptoms of mental illness at the time of your current or any past arrest?
a) Yes
b) No
c) Not sure

PRISON OF PEACE RANDOMIZED TRIAL 97

Conflict Resolution Measure #1 (T1)

Please read the scene and answer the 4 questions (use the back of the page if you need more space):

Scenario 1: An inmate has been on the phone for half an hour. You’re next in line and you need to call your
family. The cop announced “your 15 minutes are up” and you ask nicely for your turn. The inmate ignores
you the first time you ask. The second time you ask, she says “I’m not getting off this phone.”

1. What is the problem in this situation? Describe what you think is wrong:

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
2. What is your goal in this situation? What do you want/need?

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
3. Come up with at least 3 solutions to this problem:

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
PRISON OF PEACE RANDOMIZED TRIAL 98

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
4. Choose the best solution to achieve your goal:

________________________________________________________________________________________________
________________________________________________________________________________________________
________________________________________________________________________________________________
________________________________________________________________________________________________
________________________________________________________________________________________________
Scenario 2: It’s shower time and you were able to put your stuff on the 4
th
chair in line to hold your spot. You
run back to your bunk to grab something and come back. Your stuff’s been moved three chairs back and
another girl is standing by the 4
th
chair with her stuff.

1. What is the problem in this situation? Describe what you think is wrong:

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________

2. What is your goal in this situation? What do you want/need?

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
PRISON OF PEACE RANDOMIZED TRIAL 99

3. Come up with at least 3 solutions to this problem:

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________

4. Choose the best solution to achieve your goal:

________________________________________________________________________________________________
________________________________________________________________________________________________
________________________________________________________________________________________________
________________________________________________________________________________________________
________________________________________________________________________________________________
________________________________________________________________________________________________
PRISON OF PEACE RANDOMIZED TRIAL 100

Conflict Scale (T1)
Response choices:
1. Never
2. Once or twice
3. About once a month
4. About once a week
5. Several times a week
6. Daily

Conflicts with inmates/staff in the last 2 months. I have had/I have been in:

A discussion with some disagreement(s) _____
A discussion with some tension _____
A discussion where I or someone else got angry _____
A situation where some physical force was used on someone _____
A situation where some physical harm was done to someone _____
A situation where weapons were used _____

PRISON OF PEACE RANDOMIZED TRIAL 101

Perspective Taking (T1-T3)

The following statements ask about your thoughts and feelings in different situations. For each item, indicate
how well it describes you by choosing the appropriate number (1-5) on the scale at the top of the page. When
you have decided on your answer, fill in the number next to the item number. READ EACH ITEM
CAREFULLY BEFORE RESPONDING. Answer as honestly as you can.

Answer Scale:
1 2 3 4 5
Does NOT Describes me
describe me well very well

1. I daydream and fantasize often about things that might happen to me. _____
2. I often have tender, concerned feelings for people less fortunate than me. _____
3. I sometimes find it difficult to see things from the "other guy's" point of view. _____
4. Sometimes I don't feel very sorry for other people when they are having problems. _____
5. I really get involved with the feelings of the characters in a book or movie. _____
6. In emergency situations, I feel anxious and uneasy. _____
7. I am usually uninvolved when I watch a movie, and I don't often get completely caught up in it. _____
8. I try to look at everybody's side of a disagreement before I make a decision. _____
9. When I see someone being taken advantage of, I feel kind of protective towards them. _____
10. I sometimes feel helpless when I am in the middle of a very emotional situation. _____
11. I sometimes try to understand my friends better by imagining how things look from their perspective. _____
12. Becoming extremely involved in a good book or movie is somewhat rare for me. _____
13. When I see someone get hurt, I tend to remain calm. _____
14. Other people's misfortunes do not usually disturb me a great deal. _____
15. If I'm sure I'm right about something, I don't waste much time listening to other people's arguments. _____
16. After seeing a movie, I have felt as though I were one of the characters. _____
17. Being in a tense emotional situation scares me. _____
18. When I see someone being treated unfairly, I sometimes don't feel very much pity for them. ____
19. I am usually pretty helpful in dealing with emergencies. _____
20. I am often quite touched by things that I see happen. _____
21. I believe that there are two sides to every question and try to look at them both. _____
22. I would describe myself as a pretty soft-hearted person. _____
23. When I watch a good movie, I can very easily put myself in the place of a leading character. ___
24. I tend to lose control during emergencies. _____
25. When I'm upset at someone, I usually try to "put myself in his shoes" for a while. _____
26. When I am reading an interesting story or book, I imagine how I would feel if the events in the story were
happening to me. _____
27. When I see someone who badly needs help in an emergency, I fall apart. _____
28. Before criticizing somebody, I try to imagine how I would feel if I were in their place. _____

PRISON OF PEACE RANDOMIZED TRIAL 102

Anger (T1-T3)

How true of you is this statement?

Answer Scale:

1 2 3 4 5 6 7
Extremely UNTRUE Extremely TRUE of me

1. If I am provoked enough, I may hit another person. _____
2. I get into fights a little more than the average person. _____
3. There are people who pushed me so far that we came to blows. _____
4. I have threatened people I know. _____
5. I have become so mad that I have broken things. _____

PRISON OF PEACE RANDOMIZED TRIAL 103

Depression (T1-T3)

PRISON OF PEACE RANDOMIZED TRIAL 104

Anxiety (T1-T3)

PRISON OF PEACE RANDOMIZED TRIAL 105

Self-efficacy (T1-T3)

Please read the sentence and decide how true it is of you in general.

Answer Options
1=Never
2=Almost Never
3=Sometimes
4=Fairly Often
5=Very Often

1. I can manage to solve difficult problems if I try hard enough. _____
2. If someone opposes me, I can find the means and ways to get what I want. _____
3. It is easy for me to stick to my aims and accomplish my goals. _____
4. I am confident that I could deal efficiently with unexpected events. _____
5. Thanks to my talents and skills, I know how to handle unexpected situations. _____
6. I can solve most problems if I try hard enough. _____
7. I stay calm when facing difficulties because I can handle them. _____
8. When I have a problem, I can find several ways to solve it. _____
9. If I am in trouble, I can think of a solution. _____
10. I can handle whatever comes my way. _____

PRISON OF PEACE RANDOMIZED TRIAL 106

Friend and Family Relationships (T1-T3)

When applicable, answer the following questions about the last 2 months:

1. About how often have you made or received calls to/from your child(ren)?
a) Daily or almost daily
b) At least once a week
c) At least once a month
d) Less than once a month
e) Never
f) Not applicable

2. How often have you sent or received mail from your child(ren)?
a) Daily or almost daily
b) At least once a week
c) At least once a month
d) Less than once a month
e) Never
f) Not applicable

3. How often have you been personally visited by your child(ren)?
a) Daily or almost daily
b) At least once a week
c) At least once a month
d) Less than once a month
e) Never
f) Not applicable

4. How would you describe most interactions/contacts between you and your child(ren)?
a) Negative
b) Mostly negative
c) Neutral
d) Mostly positive
e) Positive
f) Not applicable

5. About how often have you made or received calls to/from family members?
a) Daily or almost daily
b) At least once a week
c) At least once a month
d) Less than once a month
e) Never
f) Not applicable

6. How often have you sent or received mail from your family?
a) Daily or almost daily
b) At least once a week
c) At least once a month
d) Less than once a month
e) Never
f) Not applicable

PRISON OF PEACE RANDOMIZED TRIAL 107

7. How often have you been personally visited by your family?
a) Daily or almost daily
b) At least once a week
c) At least once a month
d) Less than once a month
e) Never
f) Not applicable

8. How would you describe most interactions/contacts between you and your family?
a) Negative
b) Mostly negative
c) Neutral
d) Mostly positive
e) Positive
f) Not applicable

9. About how often have you made or received calls to/from friends?
a) Daily or almost daily
b) At least once a week
c) At least once a month
d) Less than once a month
e) Never
f) Not applicable

10. How often have you sent or received mail from your friends?
a) Daily or almost daily
b) At least once a week
c) At least once a month
d) Less than once a month
e) Never
f) Not applicable

11. How often have you been personally visited by your friends?
b) Daily or almost daily
c) At least once a week
d) At least once a month
e) Less than once a month
f) Never
g) Not applicable

12. How would you describe most interactions/contacts between you and your friends?
a) Negative
b) Mostly negative
c) Neutral
d) Mostly positive
e) Positive
f) Not applicable

PRISON OF PEACE RANDOMIZED TRIAL 108

13. If you have had minimal contact and/or negative interactions with friends, family, or children in the last
2 months, please explain why that is or describe some things that may be influencing your
relationships/interactions:

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
14. If you have had positive interactions with friends, family, or children in the last 2 months, please explain
why that is or describe some things that may be influencing your relationships/interactions:

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________

PRISON OF PEACE RANDOMIZED TRIAL 109

Parental Stress (T1-T3)

The following statements describe feelings and perceptions about the experience of being a parent. Think of
each of the items in terms of how your relationship with your child or children typically is. Please indicate the
degree to which you agree or disagree with the following items by placing the appropriate number in the space
provided.

1 = Strongly disagree 2 = Disagree 3 = Undecided 4 = Agree 5 = Strongly agree

I am happy in my role as a parent
There is little or nothing I wouldn't do for my child(ren) if it was necessary.
Caring for my child(ren) sometimes takes more time and energy than I have to give.
I sometimes worry whether I am doing enough for my child(ren).
I feel close to my child(ren).
I enjoy spending time with my child(ren).
My child(ren) is an important source of affection for me.
Having child(ren) gives me a more certain and optimistic view for the future.
The major source of stress in my life is my child(ren).
Having child(ren) leaves little time and flexibility in my life.
Having child(ren) has been a financial burden.
It is difficult to balance different responsibilities because of my child(ren).
The behaviour of my child(ren) is often embarrassing or stressful to me.
If I had it to do over again, I might decide not to have child(ren).
I feel overwhelmed by the responsibility of being a parent.
Having child(ren) has meant having too few choices and too little control over my life.
I am satisfied as a parent
I find my child(ren) enjoyable

PRISON OF PEACE RANDOMIZED TRIAL 110

Program Questionnaire (T2-T3)
1. Have you taken part in any programs (like substance abuse or high school classes) since your
last set of questionnaires (on August 10
th
or 11
th
about 8 weeks ago)?

a) Yes: Please list the program(s):
___________________________________________________________________________
___________________________________________________________________________
___________________________________________________________________________
___________________________________________________________________________
b) No
c) Unsure

2. To your knowledge, have you had any of the following contact with the program Prison of
Peace since your last set of questionnaires (August 10
th
-11
th
)? Circle all that apply:

a) Yes, I was enrolled in the program and went to classes
b) Yes, I was housed in a module with women who are in Prison of Peace
c) Yes, I had conversations with someone enrolled in Prison of Peace (talked to my bunkee or
someone else)
d) Yes, I had conversations with someone enrolled in Prison of Peace (talked to my bunkee or
someone else) about their time in the program
e) Yes, I was part of a Peace Circle (a conversation with someone in Prison of Peace about a
conflict that was happening in the module)
f) No, I have not had any contact with Prison of Peace that I know of
g) I’m not sure if I’ve had any contact. Explain :
___________________________________________________________________________
___________________________________________________________________________
___________________________________________________________________________
___________________________________________________________________________

PRISON OF PEACE RANDOMIZED TRIAL 111

3. Answer this ONLY if you had a conversation with someone enrolled in Prison of Peace (circled C,
D, or E in the last question). What was it like to have this conversation? Circle all that apply:

a) It was different than other conversations I usually have
- How?
_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________

b) It was better than other conversations I usually have
- How?
_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________

c) It was worse than other conversations I usually have
- How?
_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________

d) It was the same as other conversations I usually have
e) Other type of experience having this conversation
_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________

PRISON OF PEACE RANDOMIZED TRIAL 112

Conflict Resolution Measure #2 (T2)

Please read the scene and answer the 4 questions:

Scenario 1: You’re in line to order your commissary and your program time is about to run out. The
inmate in front of you is taking forever ordering her stuff. You listen closely to her order and realize
she is placing orders for other inmates as well.

1. What is the problem in this situation? Describe what you think is wrong:

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
______________________________________________________________________________________
2. What is your goal in this situation? What do you want/need?

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
3. Come up with at least 3 solutions to this problem:

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
PRISON OF PEACE RANDOMIZED TRIAL 113

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
______________________________________________________________________________________

4. Choose the best solution to achieve your goal and explain why you chose it:

________________________________________________________________________________________________
________________________________________________________________________________________________
________________________________________________________________________________________________
________________________________________________________________________________________________
________________________________________________________________________________________________
Scenario 2: It’s program time and you get in line to get hot water. An inmate cuts in a few people in
front of you. Right before it’s your turn, the hot water runs out. They put the hot pot away because the
person in front of you got the last cup of hot water.

5. What is the problem in this situation? Describe what you think is wrong:

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
6. What is your goal in this situation? What do you want/need?

_______________________________________________________________________________________
_______________________________________________________________________________________
PRISON OF PEACE RANDOMIZED TRIAL 114

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________

7. Come up with at least 3 solutions to this problem:

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________

8. Choose the best solution to achieve your goal and explain why you chose it:

________________________________________________________________________________________________
________________________________________________________________________________________________
________________________________________________________________________________________________
________________________________________________________________________________________________
________________________________________________________________________________________________
________________________________________________________________________________________________
PRISON OF PEACE RANDOMIZED TRIAL 115

Scenario 3: You just got your bag of commissary and you’re looking forward to making a meal. You
go to your bag during program time and find that half of your bag of your coffee is gone. You look
across the room and see your bunkee sipping on a cup of coffee.

1. What is the problem in this situation? Describe what you think is wrong:

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________

2. What is your goal in this situation? What do you want/need?

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________

3. Come up with at least 3 solutions to this problem:

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
PRISON OF PEACE RANDOMIZED TRIAL 116

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________

4. Choose the best solution to achieve your goal and explain why you chose it:

________________________________________________________________________________________________
________________________________________________________________________________________________
________________________________________________________________________________________________
________________________________________________________________________________________________
________________________________________________________________________________________________
________________________________________________________________________________________________
PRISON OF PEACE RANDOMIZED TRIAL 117

Conflict Scale (T2-T3)

Think about conflicts with inmates or staff in the last 2 months.

I have had/I have been in (circle the best answer for you):

A discussion with some disagreement(s)
1 2 3 4 5 6
Never Once Once Once Several times Daily
or twice a month a week a week

A discussion with some tension
1 2 3 4 5 6
Never Once Once Once Several times Daily
or twice a month a week a week

A discussion where I or someone else got angry
1 2 3 4 5 6
Never Once Once Once Several times Daily
or twice a month a week a week

A situation where some physical force was used on someone
1 2 3 4 5 6
Never Once Once Once Several times Daily
or twice a month a week a week

A situation where some physical harm was done to someone
1 2 3 4 5 6
Never Once Once Once Several times Daily
or twice a month a week a week

A situation where weapons were used
1 2 3 4 5 6
Never Once Once Once Several times Daily
or twice a month a week a week

PRISON OF PEACE RANDOMIZED TRIAL 118

Adverse Childhood Experiences (ACE) Questionnaire (T2)

While you were growing up, during the first 18 years of your life (check one choice):

1. Did a parent or other adult in the household often …
Swear at you, insult you, put you down, or humiliate you?
or
Act in a way that made you afraid that you might be physically hurt?
No___ Yes __

2. Did a parent or other adult in the household often …
Push, grab, slap, or throw something at you?
or
Ever hit you so hard that you had marks or were injured?
No___ Yes __

3. Did an adult or person at least 5 years older than you ever…
Touch or fondle you or have you touch their body in a sexual way?
or
Try to or actually have oral, anal, or vaginal intercourse with you?
No___ Yes __

4. Did you often feel that …
No one in your family loved you or thought you were important or special?
or
Your family didn’t look out for each other, feel close to each other, or support each other?
No___ Yes __

5. Did you often feel that …
You didn’t have enough to eat, had to wear dirty clothes, and had no one to protect you?
or
Your parents were too drunk or high to take care of you or take you to the doctor if you needed it?
No___ Yes __

6. Were your parents ever separated or divorced?
No___ Yes __

7. Was your mother or stepmother:
Often pushed, grabbed, slapped, or had something thrown at her?
or
Sometimes or often kicked, bitten, hit with a fist, or hit with something hard?
or
Ever repeatedly hit over at least a few minutes or threatened with a gun or knife?
No___ Yes __

8. Did you live with anyone who was a problem drinker or alcoholic, or who used street drugs?
No___ Yes __

9. Was a household member depressed or mentally ill, or did a household member attempt suicide?
No___ Yes __

10. Did a household member go to prison?
No ___ Yes __
PRISON OF PEACE RANDOMIZED TRIAL 119

Location questionnaire (T3)
INCARCERATION STATUS AND HOUSING MOVEMENT since the last set of questionnaires
(on October 7
th
/ about 6 months ago):
Where have you spent the last 6 months? Circle all that apply and provide details if you can:

a) At the jail
a. List all modules you were housed in over the last 6 months (list length of time/date(s) if
possible): ________________________________________________________________
________________________________________________________________________
________________________________________________________________________
b. If you were placed in SHU, please list the length of time/date(s) if possible: ___________
________________________________________________________________________
b) Outside the jail (circle all that apply below):
a. Transferred to another facility/prison. Location and Date(s): ________________________
________________________________________________________________________
b. In the community/Released. Date(s): __________________________________________
i. Were you ever re-arrested once you were released? Date(s)/details: ____________
__________________________________________________________________
__________________________________________________________________

PRISON OF PEACE RANDOMIZED TRIAL 120

Conflict Resolution Measure #3 (T3)

Please read the scene and answer the 4 questions:

Scenario 1: An inmate has been on the phone for half an hour. You’re next in line and you need to call
your family. The cop announced “your 15 minutes are up” and you ask nicely for your turn. The
inmate ignores you the first time you ask. The second time you ask, she says “I’m not getting off this
phone.”

5. What is the problem in this situation? Describe what you think is wrong:

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
6. What is your goal in this situation? What do you want/need?

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
7. Come up with at least 3 solutions to this problem:

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
______________________________________________________________________________________
PRISON OF PEACE RANDOMIZED TRIAL 121

_______________________________________________________________________________________
_______________________________________________________________________________________

8. Choose the best solution to achieve your goal and explain why you chose it:

________________________________________________________________________________________________
________________________________________________________________________________________________
________________________________________________________________________________________________
________________________________________________________________________________________________
________________________________________________________________________________________________
Scenario 2: It’s shower time and you were able to put your stuff on the 4
th
chair in line to hold your
spot. You run back to your bunk to grab something and come back. Your stuff’s been moved three
chairs back and another girl is standing by the 4
th
chair with her stuff.

1. What is the problem in this situation? Describe what you think is wrong:

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
2. What is your goal in this situation? What do you want/need?

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________

PRISON OF PEACE RANDOMIZED TRIAL 122

3. Come up with at least 3 solutions to this problem:

_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________
_______________________________________________________________________________________

4. Choose the best solution to achieve your goal and explain why you chose it:

________________________________________________________________________________________________
________________________________________________________________________________________________
________________________________________________________________________________________________
________________________________________________________________________________________________
_______________________________________________________________________________________________

PRISON OF PEACE RANDOMIZED TRIAL 123

APPENDIX C: Coding Manual for Conflict Resolution Measure Vignettes

Welcome to the Huey Lab! You have agreed to spend 8-10 hours a week on this project (which includes our
two hour meeting) and to meet on Wednesdays from 12-2pm for coding meetings. Please complete the
following steps when coding vignettes for the Prison of Peace (PoP) Project. Given the different scenarios
and to lessen confusion during our discussions, we will be coding items categorically – in other words, we
will be coding all “Phone” vignettes starting with question 1 first and then moving on to question 2 for that
vignette. But, you will not know if the answers are from waves 1 or 3 (which have the same vignettes) in
order to reduce bias. Since some participants refer back to answers from a previous question (writing “same”
on an item), answers will stay grouped per participant to see the entire train of thought. When you see “888”
it means that the participant left that questions blank.
Wave 1 2 vignettes N=71
Wave 3 2 vignettes N=39________________________
Scenario 1: An inmate has been on the phone for half an hour. You’re next in line and you need to call your
family. The cop announced “your 15 minutes are up” and you ask nicely for your turn. The inmate ignores
you the first time you ask. The second time you ask, she says “I’m not getting off this phone.”
Scenario 2: It’s shower time and you were able to put your stuff on the 4th chair in line to hold your spot.
You run back to your bunk to grab something and come back. Your stuff’s been moved three chairs back and
another girl is standing by the 4th chair with her stuff.
Wave 2 3 vignettes N=59
Scenario 1: You’re in line to order your commissary and your program time is about to run out. The inmate
in front of you is taking forever ordering her stuff. You listen closely to her order and realize she is placing
orders for other inmates as well. (similar to phone vignette)
Scenario 2: It’s program time and you get in line to get hot water. An inmate cuts in a few people in front of
you. Right before it’s your turn, the hot water runs out. They put the hot pot away because the person in front
of you got the last cup of hot water. (similar to shower vignette)
Scenario 3: You just got your bag of commissary and you’re looking forward to making a meal. You go to
your bag during program time and find that half of your bag of your coffee is gone. You look across the
room and see your bunkee sipping on a cup of coffee.

PRISON OF PEACE RANDOMIZED TRIAL 124

We will be coding for each of the 4 questions in each vignette, here are the instructions for the PHONE
CONFLICT scenario #1 from waves 1 and 3 (with associated examples)
An inmate has been on the phone for half an hour. You’re next in line and you need to call your family. The
cop announced “your 15 minutes are up” and you ask nicely for your turn. The inmate ignores you the first
time you ask. The second time you ask, she says “I’m not getting off this phone.”
1) What is the problem in this situation? Describe what you think is wrong:
• Conflict definition: perceived hostility versus perceived object need
o Hostility (2): the girl on the phone is disrespectful/unfair
o Need (1): I need to call my family/I won’t have time to call my family
o IF the participant mentions “Need” anywhere in the answer (even if Hostility is mentioned),
code as “Need” – example: the girl on the phone is being a bitch and I need to call my moms -
-- Simply saying “The girl needs to get off the phone” is not considered Need unless it is
specified that it’s so the participant can call her family/use the phone. Using language that is
“other-centric” is appropriate for non-hostile if there is an object need. For example, “The
inmate ignoring me, not having consideration for others that need to use the phone” implies
the Participant’s need to use the phone and would be coded as 1.
o Ask yourself, “is the participant focusing all energy on the person in the way of their goal
(inmate on the phone) or on the goal itself (calling family)?
o 999: If the participant doesn’t answer the questions and instead offers a solution or an opinion
(“I don’t really care…. I would call the cop”). Code 999 if the answer doesn’t clearly fit
hostility or need.
• Empathy/Positivity: Code 1=yes or 2=no → to indicate if the participant presented an empathetic
answer, looked at things from another’s point of view, or tried to view things from a
positive/optimistic stance (“maybe this was her first time using the phone and she needed it more”).
2) What is your goal in this situation? What do you want/need?
• Goal Orientation: perceived hostility versus perceived object need
o Hostility (2): to show that girl some respect/ I need to get her off the phone (without any
mention of why/needing to call family)
o Need (1): I to call my family
o IF the participant mentions “Hostility” anywhere in the answer (even if Need is mentioned),
code as “Hostility” following these exceptions/caveats. The hostility needs to be clear enough
to overshadow the need. There needs to be a direct link between the hostile portion and what
will help them get their need met – examples:
PRISON OF PEACE RANDOMIZED TRIAL 125

- Hostile = I need to call my family and kick her ass to show her respect Need = I need
to get her off the phone to call my family
- Hostile = I need to get her off the phone
o Ask yourself, “is the participant focusing most of her energy on the person (inmate on the
phone) or on the goal itself (calling family)?”
o 999: If the participant doesn’t answer the question “what is the GOAL” and instead offers a
solution or an opinion (“I don’t really care…. I’d tell the deputy”), or if the answer doesn’t
clearly fit hostility or need. Only code 999 if there is a solution included here that CANNOT
be inferred as a goal. Something like “talk to the cop” or “go to a different phone” are JUST
solutions without a larger goal “I need to be the bigger person here and go to a different
phone.”
• Empathy/Positivity: Code 1=yes or 2=no → to indicate if the participant presented an empathetic
answer, looked at things from another’s point of view, or tried to view things from a
positive/optimistic stance (“maybe this was her first time using the phone and she needed it more”).
Being polite by itself or using kind language does not mean they should receive a code of 1. Avoiding
a fight does not automatically mean the participant is being positive or empathic.
o What does NOT count here are comments like “she needs to understand I have a family to call
too… she needs to be considerate” because it isn’t other-centric
• Conflict Avoidance: Code 1=yes or 2=no → indicate if the participant expressed a desire to avoid
conflict (“I need to get her off the phone without causing a problem with the officers”). Being
empathic/optimistic does not automatically mean the participant is avoiding conflict. They have to
explicitly indicate they want to avoid hostility; do not infer.
o The responsibility to avoid the conflict must be on the participant, not wishing it on someone
else. So what does NOT count here are other-centric inferences – for example, “I need her to
be considerate and to walk away” would not be coded as 1 because the action/decision is not
the participant’s.
o Code 1 if the Participant indicates wanting to stay calm (I don't want to get mad; stay calm;
don't get angry). Simply writing "talk to her with a calm voice" is not the same thing as the
entire goal is not to avoid conflict, but rather indicates a general approach to the situation.
3) Come up with at least 3 solutions to this problem:
• Alternative solution thinking; hostile versus non-hostile solutions generated; Code for the method
used most often in the answers and use this hierarchy. For example, if the Participant gave 2 non-
hostile answers and 1 avoidant answer, code as non-hostile. If they gave one of each answer, code as
hostile. Any mention of violence within an answer choice makes that solution hostile (e.g., “tell her I
PRISON OF PEACE RANDOMIZED TRIAL 126

need to use the phone or take the phone from her”). If a participant provides several 999 solutions but
one or more legitimate solution, code for the legitimate solution(s).
o Hostile (2): solutions can include “beating her ass” and “disconnect her call” and “tell her
she’s going to regret this later” (clear provoking or verbal threats)
o Non-hostile (1): solutions can include “explain to her why I need to call my family” or “talk to
the cop in charge” or “look for another phone” (anything that would solve the problem of
being able to have access to a phone)
o Avoidant (0): solutions can include “I’ll try again tomorrow” or “just walk away” or “be
patient.”
o 999: If the participant did not generate any solutions, wrote something unrelated to the
question, or was wishing others to do something differently/wishing themselves to have done
something differently in the past/future
• Three solutions provided: This is a manipulation check/adherence code to see if the participant
followed directions, as opposed to providing 3 legitimate solutions. Code 1 if the participant
generated AT LEAST 3 solutions (even if the solutions don’t seem effective/possible) and Code 2 if
they generated less than 3 solutions, did not generate any solutions, or only generated solutions that
were unrelated to the question (e.g., ALL solutions were wishing other people did different things).
4) Choose the best solution to achieve your goal and explain why you chose it:
• Effectiveness of solution chosen by the subject as the best one. Refer back to the previous item if the
participant indicates “#1” from their choices listed. Effective refers to being able to use the phone (or
making a concrete/time-specific plan to) and not engaging in violence. Remember that PoP is about
mediation, not about avoiding a situation.
o Ineffective (2): Did the participant indicate that their best choice was a hostile one (“beat her
up”), an ineffective one that never enabled them to use a phone (“walk away”), was too
vague (“use the phone”), or a solution that wishes others would change their behavior/wishes
they could change what they did in the past/future. If the solution is effective in getting to use
the phone AND also includes violence, code as ineffective.
o Effective (1): Did the participant indicate that their best choice was a non-hostile one
(“explain to her why I need to use the phone”), that was effective in getting to use a phone or
making a plan to. Although an effective solution is generally an active one, “Be patient” is
another example of effective because it is avoidant but indicates there is still the goal of
getting to the phone. When in doubt, code 1.
o 999: If the participant did not choose any solutions (“I don’t know”), left this blank, wrote
down several solutions, or wrote something unrelated to the question.
PRISON OF PEACE RANDOMIZED TRIAL 127

o Note about several solutions: employ a “you’ll know it when you see it” approach when an
answer provides advanced problem-solving solutions providing multiple non-hostile solutions
in a sequential manner (e.g., “I would ask her nicely and tell her she can use it after me. If
that doesn’t work, then I guess I’ll go to the cop” for the phone vignette would be coded a 1
for EF, ST, and BSP)
o Note about one plan of action: If the solutions chosen are clearly connected, code as the same
solution. For example, “Tell her why she shouldn’t take such a long time and just grab the
phone because she was in the wrong.” Should be codded as Ineffective (2) since both
solutions/this one approach is hostile and one plan of action, ST would be (2), and BSP would
be (1). The only exception to this would be if the participant used the word “or” between the
sentences or if the several solutions are not considered as part of one approach to the problem.
• Solution Type: Use the same criteria as AST. Was the best solution provided Hostile (2), Non-hostile
(1), or avoidant (0). Code 999 if this was left blank/stated “I don’t know,” if the participant selected
more than one type of answer (so you can’t code which to type)- if the answer does not fit hostile,
non-hostile, or avoidant, or indicates wishful/past/future change-based thinking.
• Best Solution provided: Code 1 if the participant generated a best solutions and Code 2 if they left
the item blank/wrote I don’t know, gave multiple solutions, or gave an answer that had nothing to do
with the question

PRISON OF PEACE RANDOMIZED TRIAL 128

We will be coding for each of the 4 questions in each vignette, here are the instructions for the
SHOWER CONFLICT scenario #2 from waves 1 and 3 (with associated examples)
It’s shower time and you were able to put your stuff on the 4th chair in line to hold your spot. You run back
to your bunk to grab something and come back. Your stuff’s been moved three chairs back and another girl is
standing by the 4th chair with her stuff.
1) What is the problem in this situation? Describe what you think is wrong:
• Conflict definition: perceived hostility versus perceived object need – Here, it is appropriate (need-
based) if the participant indicates the problem as “my stuff has been moved” because that goes
directly to the goal/problem of not being able to shower in time. BUT something like “someone
touched my property” or “someone disrespected me” is hostility based because it overshadows the
need to shower.
o Hostility (2): “that girl’s not respecting the order of the line” (focuses on SHE is not
respecting versus “my stuff is no longer in the right order”); “she touched my shit.” Here the
main theme is disrespect.
o Need (1): “I need to take a shower and now I’ll have to wait longer.” Other answers that
would be coded as need are “someone moved my stuff…. Someone cut in line/took my spot”
because the focus in inherently taking the shower here.
o IF the participant explicitly mentions “Need” in the answer (even if Hostility is mentioned),
code as “Need” – example: No one should disrespect my stuff. Everyone needs to shower and
I was first. However, an answer like “my stuff has been moved. Someone wanted to go ahead
of me but didn't ask” would be considered hostility because “my stuff has been moved” in this
context is not explicitly enough of a need (because it doesn’t mention taking a shower at all)
to trump hostility. The focus here is also mostly on the other person. Using language that is
“other-centric” is appropriate for non-hostile if there is an object need. For example, “The
inmate ignoring me, not having consideration for others that need to shower” implies the
Participant’s need to shower and would be coded as 1.
o Ask yourself, “is the participant focusing all energy on the person in the way of their goal
(inmate who moved the stuff) or on the goal itself (having her stuff in the same spot in line so
she can shower in time)?
o 999: IF the participant doesn’t answer the questions and instead offers a solution or an opinion
or takes on the blame (“I don’t really care…. I would call the cop… it’s my fault, I left”).
Code 999 if the answer doesn’t clearly fit hostility or need.
• Empathy/Positivity: Code 1=yes or 2=no → to indicate if the participant presented an empathetic
answer, looked at things from another’s point of view, or tried to view things from a
positive/optimistic stance (“maybe no one ever taught her that it’s not okay to move other people’s
stuff… tomorrow will be a better day”)
PRISON OF PEACE RANDOMIZED TRIAL 129

2) What is your goal in this situation? What do you want/need?
• Goal Orientation: perceived hostility versus perceived object need
o Hostility (2): To show that girl some respect… to tell that girl I was first
o Need (1): I need to get my spot back…. I need to shower
o IF the participant mentions “Hostility” anywhere in the answer (even if Need is mentioned),
code as “Hostility” following these exceptions/caveats. The hostility needs to be clear enough
to overshadow the need. There needs to be a direct link between the hostile portion and what
will help them get their need met – examples:
- Hostile = I need to shower and kick her ass to show her respect
- Need = I need to get my stuff back from that bitch because I need to get my shower in
time
- Hostile = I need to get my stuff back from that bitch
o Ask yourself, “is the participant focusing most of her energy on the person (inmate that
moved her stuff) or on the goal itself (showering/getting back in line)?”
o 999: If the participant doesn’t answer the question “what is the GOAL” and instead offers a
solution or an opinion (“I don’t really care…. I’d tell the deputy”), or if the answer doesn’t
clearly fit hostility or need. Only code 999 if there is a solution included here that CANNOT
be inferred as a goal. Something like “talk to the cop” or “go to a different shower” are JUST
solutions without a larger goal “I need to be the bigger person here and use another shower.”
• Empathy/Positivity: Code 1=yes or 2=no → to indicate if the participant presented an empathetic
answer, looked at things from another’s point of view, or tried to view things from a
positive/optimistic stance (“if she was new on the unit I’d let her go first”). Being polite by itself or
using kind language does not mean they should receive a code of 1. Avoiding a fight does not
automatically mean the participant is being positive or empathic.
o What does NOT count here are comments like “she needs to understand I have to shower
too… she needs to be considerate” because it isn’t other-centric
• Conflict Avoidance: Code 1=yes or 2=no → indicate if the participant expressed a desire to avoid
conflict (“I need to get my stuff back without causing a problem with the officers”). Being
empathic/optimistic does not automatically mean the participant is avoiding conflict. They have to
explicitly indicate they want to avoid hostility; do not infer.
o The responsibility to avoid the conflict must be on the participant, not wishing it on someone
else. So what does NOT count here are other-centric inferences – for example, “I need her to
be considerate and respect the rules of the line” would not be coded as 1 because the
action/decision is not the participant’s.
PRISON OF PEACE RANDOMIZED TRIAL 130

o Code 1 if the Participant indicates wanting to stay calm (I don't want to get mad; stay calm;
don't get angry). Simply writing "talk to her with a calm voice" is not the same thing as the
entire goal is not to avoid conflict, but rather indicates a general approach to the situation.
3) Come up with at least 3 solutions to this problem:
• Alternative solution thinking; hostile versus non-hostile solutions generated; Code for the method
used most often in the answers and use this hierarchy. For example, if the Participant gave 2 non-
hostile answers and 1 avoidant answer, code as non-hostile. If they gave one of each answer, code as
hostile. Any mention of violence within an answer choice makes that solution hostile (e.g., “tell her to
move or move her stuff for her”). If a participant provides several 999 solutions but one or more
legitimate solution, code for the legitimate solution(s).
o Hostile (2): solutions can include “beating her ass” and “move her shit back…. Jump in front
of her” and “tell her she’s going to regret this later… confront the girl” (clear provoking or
verbal threats)
o Non-hostile (1): solutions can include “ask her to please move nicely” or “talk to the cop in
charge”
o Avoidant (0): solutions can include “just let it go/be patient” or “just take the spot and
whatever” – only use this code if there is no other “active” solution like talking to the cop in
charge
o 999: If the participant did not generate any solutions, wrote something unrelated to the
question, or was wishing others to do something differently/wishing themselves to have done
something differently in the past/future, or took blame for the situation
• Three solutions provided: This is a manipulation check/adherence code to see if the participant
followed directions, as opposed to providing 3 legitimate solutions. Code 1 if the participant
generated AT LEAST 3 solutions (even if the solutions don’t seem effective/possible) and Code 2 if
they generated less than 3 solutions, did not generate any solutions, or only generated solutions that
were unrelated to the question (e.g., ALL solutions were wishing other people did different things).
4) Choose the best solution to achieve your goal and explain why you chose it:
• Effectiveness of solution chosen by the subject as the best one. Refer back to the previous item if the
participant indicates “#1” from their choices listed. Effective refers to being able to use the shower
(or making a concrete/time-specific plan to), getting their spot back in line and not engaging in
violence. Remember that PoP is about mediation, not about avoiding a situation.
o Ineffective (2): Did the participant indicate that their best choice was a hostile one (“take my
stuff back”), an ineffective one that never enabled them to use the shower in time/is vague
(“take a shower”)/does not share intent to shower (“take a breath/talk calmly”), or a solution
that wishes others would change their behavior/wishes they could change what they did in
PRISON OF PEACE RANDOMIZED TRIAL 131

the past/future. If the solution is effective in getting to use the shower AND also includes
hostility, code as ineffective.
o Effective (1): Did the participant indicate that their best choice was a non-hostile one
(“explain to the girl I need my spot back”), that was effective in getting to use the shower or
making a concrete plan to do so. Although an effective solution is generally an active one,
“Just stay where they moved me…. Just take a shower… be patient” is avoidant but another
example of effective because it is non-hostile and indicates there is still the goal of getting a
shower. There is no assumption that “just wait in that spot” means they won’t have time to
take a shower. “Take a breath” (in the examples above) is ineffective compared to “just wait”
because there is no intention to take a shower When in doubt, code (1).
o 999: If the participant did not choose any solutions (“I don’t know”), left this blank, wrote
down several solutions, or wrote something unrelated to the question.
o Note about several solutions: employ a “you’ll know it when you see it” approach when an
answer provides advanced problem-solving solutions providing multiple non-hostile solutions
in a sequential manner (e.g., “I would ask her nicely and tell her she can use it after me. If
that doesn’t work, then I guess I’ll go to the cop” for the phone vignette would be coded a 1
for EF, ST, and BSP)
o Note about one plan of action: If the solutions chosen are clearly connected, code as the same
solution. For example, “Tell her why she shouldn’t touch my stuff and just get in because she
was in the wrong.” Should be codded as Ineffective (2) since both solutions/this one approach
is hostile and one plan of action, ST would be (2), and BSP would be (1). The only exception
to this would be if the participant used the word “or” between the sentences or if the several
solutions are not considered as part of one approach to the problem.
• Solution Type: Use the same criteria as AST. Was the best solution provided Hostile (2), Non-hostile
(1), or avoidant (0). Code 999 if this was left blank/stated “I don’t know,” if the participant selected
more than one type of answer (so you can’t code which to type)- if the answer does not fit hostile,
non-hostile, or avoidant, or indicates wishful/past/future change-based thinking.
• Best Solution provided: Code 1 if the participant generated a best solutions and Code 2 if they left
the item blank/wrote I don’t know, gave multiple solutions, or gave an answer that had nothing to do
with the question
PRISON OF PEACE RANDOMIZED TRIAL 132

We will be coding for each of the 4 questions in each vignette, here are the instructions for the
COMMISSARY CONFLICT scenario #1 from wave 2 (with associated examples)
You’re in line to order your commissary and your program time is about to run out. The inmate in front of
you is taking forever ordering her stuff. You listen closely to her order and realize she is placing orders for
other inmates as well. (similar to phone vignette)
1) What is the problem in this situation? Describe what you think is wrong:
• Conflict definition: perceived hostility versus perceived object need
o Hostility (2): the girl in front of me is disrespectful/unfair/breaking the rules
o Need (1): I need to order my commissary
o IF the participant mentions “Need” anywhere in the answer (even if Hostility is mentioned),
code as “Need” – example: that girl needs to stop ordering for other people so I can order my
stuff --- Simply saying “that girl needs to stop ordering for other people” is not considered
Need unless it is specified that it’s so the participant can order her commissary. Saying “time
is running out” is an indication that the need is there to order commissary in time. Using
language that is “other-centric” is appropriate for non-hostile if there is an object need. For
example, “The inmate is not having consideration for others that need to order their stuff”
implies the Participant’s need to order commissary and would be coded as 1.
o Ask yourself, “is the participant focusing all energy on the person in the way of their goal
(inmate ahead in line) or on the goal itself (getting her commissary)?
o 999: If the participant doesn’t answer the questions and instead offers a solution or an opinion
(“I don’t really care…. I would call the cop”). Code 999 if the answer doesn’t clearly fit
hostility or need.
• Empathy/Positivity: Code 1=yes or 2=no → to indicate if the participant presented an empathetic
answer, looked at things from another’s point of view, or tried to view things from a
positive/optimistic stance (“I think it’s nice she’s trying to help someone out”).
2) What is your goal in this situation? What do you want/need?
• Goal Orientation: perceived hostility versus perceived object need
o Hostility (2): to show that girl some respect/ I need her to stop placing other orders (without
any mention of why/needing to get her stuff)
o Need (1): to put my order in before they close
o IF the participant mentions “Hostility” anywhere in the answer (even if Need is mentioned),
code as “Hostility” following these exceptions/caveats. The hostility has to be clear enough to
PRISON OF PEACE RANDOMIZED TRIAL 133

overshadow the need. There needs to be a direct link between the hostile portion and what will
help them get their need met – examples:
- Hostile = I need to get my stuff and let her know she’s being a bitch
- Need = I need her to get her ass off the machine so I can get my stuff
- Hostile = I need her to get her ass off the machine
o Ask yourself, “is the participant focusing most of her energy on the person (inmate ahead in
line) or on the goal itself (getting her commissary)?”
o 999: If the participant doesn’t answer the questions and instead offers a solution (“I’d tell the
staff”) or an opinion (“she needs to be more considerate”), or if the answer doesn’t clearly fit
hostility or need. Only code 999 if there is a solution included here that CANNOT be inferred
as a goal. Something like “talk to the cop” or “go to a different kiosk” are JUST solutions
without a larger goal “I need to be the bigger person here and order my stuff some other way.”
• Empathy/Positivity: Code 1=yes or 2=no → to indicate if the participant presented an empathetic
answer, looked at things from another’s point of view, or tried to view things from a
positive/optimistic stance (“maybe she doesn’t know the rules that you can’t order for others”).
Being polite by itself or using kind language does not mean they should receive a code of 1. Avoiding
a fight does not automatically mean the participant is being positive or empathic.
o What does NOT count here are comments like “she needs to be considerate and realize I have
stuff to order too” because it isn’t other-centric
• Conflict Avoidance: Code 1=yes or 2=no → indicate if the participant expressed a desire to avoid
conflict (“I need to get her off the machine without causing a problem with the officers”). Being
empathic/optimistic does not automatically mean the participant is avoiding conflict. They have to
explicitly indicate they want to avoid hostility; do not infer.
o The responsibility to avoid the conflict must be on the participant, not wishing it on someone
else. So what does NOT count here are other-centric inferences – for example, “I need her to
be considerate about my needs” would not be coded as 1 because the action/decision is not the
participant’s.
o Code 1 if the Participant indicates wanting to stay calm (I don't want to get mad; stay calm;
don't get angry). Simply writing "talk to her with a calm voice" is not the same thing as the
entire goal is not to avoid conflict, but rather indicates a general approach to the situation.
3) Come up with at least 3 solutions to this problem:
• Alternative solution thinking; hostile versus non-hostile solutions generated; Code for the method
used most often in the answers and use this hierarchy. For example, if the Participant gave 2 non-
hostile answers and 1 avoidant answer, code as non-hostile. If they gave one of each answer, code as
PRISON OF PEACE RANDOMIZED TRIAL 134

hostile. Any mention of violence within an answer choice makes that solution hostile (e.g., “tell her to
please hurry or move her to the side to make my order”). If a participant provides several 999
solutions but one or more legitimate solution, code for the legitimate solution(s).
o Hostile (2): solutions can include “tell that bitch to hurry up or else” and “count to ten out
loud to let her know I’m about to step in” (clear provoking or verbal threats) or “have other
inmates get in line to make her feel the pressure”
o Non-hostile (1): solutions can include “talk to her” or “go tell the deputy” or “use a bubble
sheet instead” or “wait patiently” (anything that would solve the problem of being able to
order the commissary)
o Avoidant (0): solutions can include “I’ll try again tomorrow”
o 999: If the participant did not generate any solutions, wrote something unrelated to the
question, or was wishing others to do something differently/wishing themselves to have done
something differently in the past/future
• Three solutions provided: This is a manipulation check/adherence code to see if the participant
followed directions, as opposed to providing 3 legitimate solutions. Code 1 if the participant
generated AT LEAST 3 solutions (even if the solutions don’t seem effective/possible) and Code 2 if
they generated less than 3 solutions, did not generate any solutions, or only generated solutions that
were unrelated to the question (e.g., ALL solutions were wishing other people did different things).
4) Choose the best solution to achieve your goal and explain why you chose it:
• Effectiveness of solution chosen by the subject as the best one. Refer back to the previous item if the
participant indicates “#1” from their choices listed. Effective refers to being able to order the
commissary (or making a concrete/time-specific plan to; here only “order the next time” is sufficient)
and not engaging in violence. Remember that PoP is about mediation, not about avoiding a situation.

o Ineffective (2): Did the participant indicate that their best choice was a hostile one (“keep
telling her to hurry the hell up”), an ineffective one that never enabled them to order
commissary (“walk away”), a vague one (“order commissary”), or a solution that wishes
others would change their behavior/wishes they could change what they did in the
past/future. If the solution is effective in getting to order commissary AND also includes
hostility, code as ineffective.
o Effective (1): Did the participant indicate that their best choice was a non-hostile one (“have a
trustee order my stuff for me if time runs out… explain to her why she can’t do that”), that
was effective in getting to order commissary or making a concrete plan to. Although an
effective solution is generally an active one, “Be patient” is another example of effective
because it is avoidant but indicates there is still the goal of getting to order her stuff
eventually. When in doubt, code (1).
PRISON OF PEACE RANDOMIZED TRIAL 135

o 999: If the participant did not choose any solutions (“I don’t know”), left this blank, wrote
down several solutions, or wrote something unrelated to the question.
o Note about several solutions: employ a “you’ll know it when you see it” approach when an
answer provides advanced problem-solving solutions providing multiple non-hostile solutions
in a sequential manner (e.g., “I would ask her nicely and tell her she can use it after me. If
that doesn’t work, then I guess I’ll go to the cop” for the phone vignette would be coded a 1
for EF, ST, and BSP)
o Note about one plan of action: If the solutions chosen are clearly connected, code as the same
solution. For example, “Tell her why she shouldn’t order for others and just get in front of her
because she was in the wrong.” Should be codded as Ineffective (2) since both solutions/this
one approach is hostile and one plan of action, ST would be (2), and BSP would be (1). The
only exception to this would be if the participant used the word “or” between the sentences or
if the several solutions are not considered as part of one approach to the problem.
• Solution Type: Use the same criteria as AST. Was the best solution provided Hostile (2), Non-hostile
(1), or avoidant (0). Code 999 if this was left blank/stated “I don’t know,” if the participant selected
more than one type of answer (so you can’t code which to type)- if the answer does not fit hostile,
non-hostile, or avoidant, or indicates wishful/past/future change-based thinking.
• Best Solution provided: Code 1 if the participant generated a best solutions and Code 2 if they left
the item blank/wrote I don’t know, gave multiple solutions, or gave an answer that had nothing to do
with the question

PRISON OF PEACE RANDOMIZED TRIAL 136

We will be coding for each of the 4 questions in each vignette, here are the instructions for the HOT
WATER CONFLICT scenario #2 from wave 2 (with associated examples)
It’s program time and you get in line to get hot water. An inmate cuts in a few people in front of you. Right
before it’s your turn, the hot water runs out. They put the hot pot away because the person in front of you got
the last cup of hot water. (similar to shower vignette)
1) What is the problem in this situation? Describe what you think is wrong:
• Conflict definition: perceived hostility versus perceived object need
o Hostility (2): “that girl’s not respecting the order of the line… I got cut”
o Need (1): “the pot ran out of water and I didn’t get any”
o IF the participant mentions “Need” anywhere in the answer (even if Hostility is mentioned),
code as “Need” – example: That girl was disrespectful and cut in line and now I don’t have
water
o Ask yourself, “is the participant focusing all energy on the person in the way of their goal
(inmate who cut in line) or on the goal itself (needing hot water)?
o 999: IF the participant doesn’t answer the questions and instead offers a solution or an opinion
(“I don’t really care…. I would call the cop… I’ll leave it alone”). Code 999 if the answer
doesn’t clearly fit hostility or need.
• Empathy/Positivity: Code 1=yes or 2=no → to indicate if the participant presented an empathetic
answer, looked at things from another’s point of view, or tried to view things from a
positive/optimistic stance (“maybe she needed the water more than me”).
2) What is your goal in this situation? What do you want/need?
• Goal Orientation: perceived hostility versus perceived object need
o Hostility (2): To try to control my anger cuz that bitch cut in line … to let her know she can’t
do that
o Need (1): To get hot water
o IF the participant mentions “Hostility” anywhere in the answer (even if Need is mentioned),
code as “Hostility” following these exceptions/caveats. The hostility needs to be clear enough
to overshadow the need. There needs to be a direct link between the hostile portion and what
will help them get their need met. Using language that is “other-centric” is appropriate for
non-hostile if there is an object need. For example, “The inmate is not having consideration
for others that need to get the water” implies the Participant’s need to get hot water and
would be coded as 1.
PRISON OF PEACE RANDOMIZED TRIAL 137

o Ask yourself, “is the participant focusing most of her energy on the person (inmate that’s
cutting in line) or on the goal itself (getting hot water)?”
o 999: If the participant doesn’t answer the question “what is the GOAL” and instead offers a
solution or an opinion (“I don’t really care…. I’d tell the deputy”), or if the answer doesn’t
clearly fit hostility or need. Only code 999 if there is a solution included here that CANNOT
be inferred as a goal. Something like “talk to the cop” or “get water from my room” are JUST
solutions without a larger goal “I need to be the bigger person here and get hot water some
other way.”
• Empathy/Positivity: Code 1=yes or 2=no → to indicate if the participant presented an empathetic
answer, looked at things from another’s point of view, or tried to view things from a
positive/optimistic stance (“if she was new on the unit I’d let her go first”). Being polite by itself or
using kind language does not mean they should receive a code of 1. Avoiding a fight does not
automatically mean the participant is being positive or empathic.
o What does NOT count here are comments like “she needs to understand I have to get water
too… she needs to be considerate” because it isn’t other-centric
• Conflict Avoidance: Code 1=yes or 2=no → indicate if the participant expressed a desire to avoid
conflict (“to not get in a fight and control my temper”). Being empathic/optimistic does not
automatically mean the participant is avoiding conflict. They have to explicitly indicate they want to
avoid hostility; do not infer.
o The responsibility to avoid the conflict must be on the participant, not wishing it on someone
else. So what does NOT count here are other-centric inferences – for example, “I need her to
be considerate and respect the rules of the line” would not be coded as 1 because the
action/decision is not the participant’s.
o Code 1 if the Participant indicates wanting to stay calm (I don't want to get mad; stay calm;
don't get angry). Simply writing "talk to her with a calm voice" is not the same thing as the
entire goal is not to avoid conflict, but rather indicates a general approach to the situation.
3) Come up with at least 3 solutions to this problem:
• Alternative solution thinking; hostile versus non-hostile solutions generated; Code for the method
used most often in the answers and use this hierarchy. For example, if the Participant gave 2 non-
hostile answers and 1 avoidant answer, code as non-hostile. If they gave one of each answer, code as
hostile. Any mention of violence within an answer choice makes that solution hostile. If a participant
provides several 999 solutions but one or more legitimate solution, code for the legitimate solution(s).
o Hostile (2): solutions can include “take the last cup from that chick” and “cut her off” and
“tell her she better share or she’s going to regret this later… confront the girl” (clear
provoking or verbal threats)
PRISON OF PEACE RANDOMIZED TRIAL 138

o Non-hostile (1): solutions can include “talk it out with the inmate” or “ask if I can get more
water to the cop” (anything that would solve the problem of being able to get/seek out hot
water)
o Avoidant (0): solutions can include “just let it go” or “go to my room to avoid fighting” – only
use this code if there is no other “active” solution like talking to the cop in charge
o 999: If the participant did not generate any solutions, wrote something unrelated to the
question, or was wishing others to do something differently/wishing themselves to have done
something differently in the past/future
• Three solutions provided: This is a manipulation check/adherence code to see if the participant
followed directions, as opposed to providing 3 legitimate solutions. Code 1 if the participant
generated AT LEAST 3 solutions (even if the solutions don’t seem effective/possible) and Code 2 if
they generated less than 3 solutions, did not generate any solutions, or only generated solutions that
were unrelated to the question (e.g., ALL solutions were wishing other people did different things).
4) Choose the best solution to achieve your goal and explain why you chose it:
• Effectiveness of solution chosen by the subject as the best one. Refer back to the previous item if the
participant indicates “#1” from their choices listed. Effective refers to being able to get hot water or
making a concrete plan to seek out hot water and not engaging in violence. Remember that PoP is
about mediation, not about avoiding a situation.
o Ineffective (2): Did the participant indicate that their best choice was a hostile one (“take her
hot water”), an ineffective one that never enabled them to get water (“walk away”), a vague
one (“get hot water”), or a solution that wishes others would change their behavior/wishes
they could change what they did in the past/future. If the solution is effective in getting water
AND also includes violence, code as ineffective.
o Effective (1): Did the participant indicate that their best choice was a non-hostile one (“ask
her to share with me”) that was effective in getting water. Although an effective solution is
generally an active one, “be patient … get hot water tomorrow” is avoidant but another
example of effective because it is non-hostile and indicates there is still the goal of getting
water. When in doubt, code (1).
o 999: If the participant did not choose any solutions (“I don’t know”), left this blank, wrote
down several solutions, or wrote something unrelated to the question.
o Note about several solutions: employ a “you’ll know it when you see it” approach when an
answer provides advanced problem-solving solutions providing multiple non-hostile solutions
in a sequential manner (e.g., “I would ask her nicely and tell her she can use it after me. If
that doesn’t work, then I guess I’ll go to the cop” for the phone vignette would be coded a 1
for EF, ST, and BSP)
PRISON OF PEACE RANDOMIZED TRIAL 139

o Note about one plan of action: If the solutions chosen are clearly connected, code as the same
solution. For example, “Tell her why she shouldn’t cut and just get in front of her because she
was in the wrong.” Should be codded as Ineffective (2) since both solutions/this one approach
is hostile and one plan of action, ST would be (2), and BSP would be (1). The only exception
to this would be if the participant used the word “or” between the sentences or if the several
solutions are not considered as part of one approach to the problem.
• Solution Type: Use the same criteria as AST. Was the best solution provided Hostile (2), Non-hostile
(1), or avoidant (0). Code 999 if this was left blank/stated “I don’t know,” if the participant selected
more than one type of answer (so you can’t code which to type)- if the answer does not fit hostile,
non-hostile, or avoidant, or indicates wishful/past/future change-based thinking.
• Best Solution provided: Code 1 if the participant generated a best solutions and Code 2 if they left
the item blank/wrote I don’t know, gave multiple solutions, or gave an answer that had nothing to do
with the question
PRISON OF PEACE RANDOMIZED TRIAL 140

We will be coding for each of the 4 questions in each vignette, here are the instructions for the
COFFEE CONFLICT scenario #3 from wave 2 (with associated examples)
You just got your bag of commissary and you’re looking forward to making a meal. You go to your bag
during program time and find that half of your bag of your coffee is gone. You look across the room and see
your bunkee sipping on a cup of coffee. (ambiguous situation)
1) What is the problem in this situation? Describe what you think is wrong:
• Conflict definition: perceived hostility versus perceived object need
o Hostility (2): “She stole my coffee” … “people who steal shit/stealing”
o Need (1): “Half of my coffee is missing”
o IF the participant mentions “Need” anywhere in the answer (even if Hostility is mentioned),
code as “Need” – example: She stole my coffee and I was really looking forward to having a
cup, now I can’t enjoy my coffee OR that there’s missing coffee and that she took it.
o Ask yourself, “is the participant focusing all energy on the person in the way of their goal (the
someone that stole the coffee) or on the goal itself (having the coffee)?
o 999: IF the participant doesn’t answer the questions and instead offers a solution or an opinion
(“I don’t really care…. I would call the cop… I think it’s wrong when people steal things”).
Code 999 if the answer doesn’t clearly fit hostility or need.
• Empathy/Positivity: Code 1=yes or 2=no → to indicate if the participant presented an empathetic
answer, looked at things from another’s point of view, or tried to view things from a
positive/optimistic stance (“maybe it’s just a misunderstanding and I should not jump to
conclusions”). Do not code yes simply if the participant is considering other “suspects.”
• Attribution of Blame: Code 1= other (“someone” took my coffee) or the possibility that it may not
be the bunkee (“it may be a coincidence… if she took it”), Code 2= assumption that the bunkee took
the coffee (“she stole my coffee”). Code 999 if the answer is unrelated to the question of “what is the
problem” or there is no attribution of blame at all (“stealing is wrong”).
2) What is your goal in this situation? What do you want/need?
• Goal Orientation: perceived hostility versus perceived object need
o Hostility (2): To show my bunkee some respect and tell her not to steal my stuff
o Need (1): To get my coffee back
o IF the participant mentions “Hostility” anywhere in the answer (even if Need is mentioned),
code as “Hostility” following these exceptions/caveats. The hostility needs to be clear enough
PRISON OF PEACE RANDOMIZED TRIAL 141

to overshadow the need. There needs to be a direct link between the hostile portion and what
will help them get their need met – examples:
- Hostile = To find out who took my coffee so I can beat their ass
- Need = To find out who took my coffee so I can get it back
- Need = To find out who took my coffee or To get my coffee back from my bunkee (this
is acceptable as an answer because hostility isn’t clear enough here)
o Ask yourself, “is the participant focusing most of her energy on the person (inmate that took
her coffee) or on the goal itself (getting the coffee back)?”
o 999: If the participant doesn’t answer the question “what is the GOAL” and instead offers a
solution or an opinion (“I don’t really care…. I’d tell the deputy… I can’t even think about
this stuff”), or if the answer doesn’t clearly fit hostility or need. Only code 999 if there is a
solution included here that CANNOT be inferred as a goal. Something like “have her moved”
or “get more coffee” are JUST solutions without a larger goal “I need to figure out what’s
going on here to find my coffee.”
• Empathy/Positivity: Code 1=yes or 2=no → to indicate if the participant presented an empathetic
answer, looked at things from another’s point of view, or tried to view things from a
positive/optimistic stance (“maybe it’s just a misunderstanding and I should not jump to
conclusions…I can learn from my mistakes and keep my coffee next time”). Do not code yes simply
if the participant is considering other “suspects.” Being polite by itself or using kind language does
not mean they should receive a code of 1. Avoiding a fight does not automatically mean the
participant is being positive or empathic.
o What does NOT count here are comments like “she needs to be considerate and understanding
that other people want coffee too” because it isn’t other-centric
• Conflict Avoidance: Code 1=yes or 2=no → indicate if the participant expressed a desire to avoid
conflict (“To not get mad… move cells”). Being empathic/optimistic does not automatically mean the
participant is avoiding conflict. They have to explicitly indicate they want to avoid hostility; do not
infer.
o The responsibility to avoid the conflict must be on the participant, not wishing it on someone
else. So what does NOT count here are other-centric inferences – for example, “I need her to
be considerate and respect my stuff” would not be coded as 1 because the action/decision is
not the participant’s.
o Code 1 if the Participant indicates wanting to stay calm (I don't want to get mad; stay calm;
don't get angry). Simply writing "talk to her with a calm voice" is not the same thing as the
entire goal is not to avoid conflict, but rather indicates a general approach to the situation.

PRISON OF PEACE RANDOMIZED TRIAL 142

3) Come up with at least 3 solutions to this problem:
• Alternative solution thinking; hostile versus non-hostile solutions generated; Code for the method
used most often in the answers and use this hierarchy. For example, if the Participant gave 2 non-
hostile answers and 1 avoidant answer, code as non-hostile. If they gave one of each answer, code as
hostile. Any mention of violence within an answer choice makes that solution hostile (e.g., “ask her
nicely if she took it and get in her face”). If a participant provides several 999 solutions but one or
more legitimate solution, code for the legitimate solution(s).
o Hostile (2): solutions can include “make her give me my coffee back” (as opposed to just “get
my coffee back”) and “tell her she’s going to regret this later… confront my bunkee” (clear
provoking or verbal threats) or “have her moved” (different than “move rooms” because
forcefully making someone else move).
o Non-hostile (1): solutions can include “talk it out with my bunkeee… find out why she did
that” or “get my coffee back” (anything that would solve the problem of being able to get the
coffee back)
o Avoidant (0): solutions can include “just let it go” or “move rooms”
o 999: If the participant did not generate any solutions, wrote something unrelated to the
question, or was wishing others to do something differently/wishing themselves to have done
something differently in the past/future (“don’t let this happen”).
• Three solutions provided: This is a manipulation check/adherence code to see if the participant
followed directions, as opposed to providing 3 legitimate solutions. Code 1 if the participant
generated AT LEAST 3 solutions (even if the solutions don’t seem effective/possible) and Code 2 if
they generated less than 3 solutions, did not generate any solutions, or only generated solutions that
were unrelated to the question (e.g., ALL solutions were wishing other people did different things).
4) Choose the best solution to achieve your goal and explain why you chose it:
• Effectiveness of solution chosen by the subject as the best one. Refer back to the previous item if the
participant indicates “#1” from their choices listed. Effective refers to being able to get the coffee
back, discuss the situation with the bunkee/seek out where the coffee went and not engaging in
violence. Remember that PoP is about mediation, not about avoiding a situation.
o Ineffective (2): Did the participant indicate that their best choice was a hostile one (“make her
give me my coffee back”), an ineffective one that never enabled them to get their coffee
back or figure out who took the coffee (“move rooms”), a vague one (“get my coffee”), or a
solution that wishes others would change their behavior/wishes they could change what they
did in the past/future (“next time hide my stuff”). If the solution is effective in getting the
coffee AND also includes violence, code as ineffective.
PRISON OF PEACE RANDOMIZED TRIAL 143

o Effective (1): Did the participant indicate that their best choice was a non-hostile one (“ask
her why and see if she can give me back the coffee”), that was effective in getting the coffee
back/seeking out where it went. In this conflict, an effective solution will almost always be
an active one. When in doubt, code (1).
o 999: If the participant did not choose any solutions (“I don’t know”), left this blank, wrote
down several solutions, or wrote something unrelated to the question.
o Note about several solutions: employ a “you’ll know it when you see it” approach when an
answer provides advanced problem-solving solutions providing multiple non-hostile solutions
in a sequential manner (e.g., “I would ask her nicely and tell her she can use it after me. If
that doesn’t work, then I guess I’ll go to the cop” for the phone vignette would be coded a 1
for EF, ST, and BSP)
o Note about one plan of action: If the solutions chosen are clearly connected, code as the same
solution. For example, “Tell her why she shouldn’t touch my stuff and beat her up because she
was in the wrong.” Should be codded as Ineffective (2) since both solutions/this one approach
is hostile and one plan of action, ST would be (2), and BSP would be (1). The only exception
to this would be if the participant used the word “or” between the sentences or if the several
solutions are not considered as part of one approach to the problem.
• Solution Type: Use the same criteria as AST. Was the best solution provided Hostile (2), Non-hostile
(1), or avoidant (0). Code 999 if this was left blank/stated “I don’t know,” if the participant selected
more than one type of answer (so you can’t code which to type)- if the answer does not fit hostile,
non-hostile, or avoidant, or indicates wishful/past/future change-based thinking. (e.g., “next time I’ll
hide my coffee better… to have my bunkee ask me for some cause I would shared if she asked”).
• Best Solution provided: Code 1 if the participant generated a best solutions and Code 2 if they left
the item blank/wrote I don’t know, gave multiple solutions, or gave an answer that had nothing to do
with the question.
PRISON OF PEACE RANDOMIZED TRIAL 144

GENERAL GUIDELINES
• Timeline: STOP coding on Monday nights at 7pm, as I will be compiling the reliability ratings at that
time. I will send you a “discrepancies” worksheet on Tuesday early afternoon. Please be prepared to
discuss your ratings/make supporting or conceding arguments by noon on Wednesday.
• Only code for a consecutive 1.5 hour at a time; take a break, come back with fresh eyes to avoid
coding fatigue. Only code the set amount of vignettes you are given for that week.
• Use the “vignettes” SPSS file in your name-folder on Dropbox to record your codes and use the
“vignettes” Excel file in your folder as well.
• Do not make assumptions and simply code what the participant has written.
• Language: Interchange the word “shit” with “stuff” (e.g., ask her if she took my shit) to avoid a
hostility bias. However, if a sentence is clearly added in to indicate hostility (e.g., “ask her where the
fuck she put my coffee) assume hostility.
• If the participant provides many disjointed sentences for one question item, break down each “part”
to see if any of them fits the question and code that one. However, you may use the entire sentence to
code for the rest of the variables.
o Example below for the coffee vignette Goal Orientation. The three “parts” would be “I would
be pissed” (not a goal, in itself 999) then “want to confront my bunkee” (a goal coded as
2/hostile), and “I didn’t see her steal it I can’t just go off on her” (not a goal/opinion, 999).
Although you are coding one of the parts to be “hostile” you can use the last part to code 2 for
conflict avoidance.
Well first off I would be pissed and want to confront my bunkie beacuse I didnt see her steal it I cant
just go off on her
a. Goal Orientation: _______2________
b. Empathy/Positivity: _____2________
c. Conflict Avoidance: ______1_______
• Remember that the goal of PoP is mediation tactics and conflict resolution, not avoidance without
planning to resolve the issue
• When you’re stuck, take a step back and ask “what am I trying to capture with this code? Does this
participant represent ___ (conflict avoidance/hostility) with this answer?”
• Take your time and re-read the vignette answers as reading them quickly can lead to mis-reading
sentences.
PRISON OF PEACE RANDOMIZED TRIAL 145

• You will be coding these in random order and will not know when the participants answered these
(waves 1-3) and you will not know which group the participant was assigned to in the study to reduce
bias.
• You will both (Hailey and Mady) be coding all the vignettes. Once a week, we will all (Gabby and
Marie) come together to discuss discrepancies in the codes. Do NOT access the other RA’s folder on
Dropbox at any time.
• Please email me if you have any doubts/questions about the coding/code book.
LEGEND OF DESCRIPTORS
• To keep coders blind to the time period of the answers, Waves 1-3 are indicated by random letters (R,
E, S). The following identifiers will be used to indicate which vignettes/questions/variables you are
coding:
• Vignette Type
o Ph = Phone
o Sh = Shower
o Hw = Hot Water
o Cm = Commissary
o Cof = Coffee
• Question Number
o 1 = What is the problem
o 2 = What is the goal
o 3 = Generate 3 solutions
o 4 = Choose the best solution
• Variable being coded (in order of the code book)
o CD = Conflict Definition
o EP = Empathy/Positivity
▪ EP con = EP for conflict definition; EP go = EP for goal orientation
o CA = Conflict Avoidance
o GO = Goal Orientation
o AST = Alternate Solution Thinking
o TSP = Three Solutions Provided
o EF = Effectiveness
o ST = Solution Type
o BSP = Best Solution Provided
o AoB = Attribution of Blame (only for Coffee vignette)
• Identifiers will be “Wave #_Vignette Type_Question Number_Variable being coded”
• Ex: The variable “R_Ph_1_CD” will indicate Wave (unknown), phone vignette, question 1 (what is
the problem) for the conflict definition variable.

PRISON OF PEACE RANDOMIZED TRIAL 146

APPENDIX D: Conflict Resolution Measure Vignette Codes by Vignette Type and Time Point

Note: Bolded numbers indicate the majority proportion.

Phone Vignette
Hostility/Hostile Need/Non-hostile Invalid Response
n % n % n %
Codes (T1 N=72)
Conflict Definition 50 69.5% 9 12.5% 13 18%
Goal Orientation 5 7% 46 64% 21 29%
Alternate Solution Type 8 11% 57 79% 7 10%
Best Solution Type 1 1% 51 71% 20 28%
Yes No Invalid Response
n % n % n %
Empathy/Positivity (problem) 6 16% 65 90% 1 1%
Empathy/Positivity (goal) 4 6% 67 93% 1 1%
Conflict Avoidance 19 26% 52 72% 1 1%
Effective Solution 43 60% 9 12% 20 28%
Adherence Check
Three Solutions Provided 57 79% 15 21% --
Best Solution Provided 49 68% 23 32% --
Phone Vignette
Hostility/Hostile Need/Non-hostile Invalid Response
n % n % n %
Codes (T3 N=38)
Conflict Definition 26 68% 9 24% 3 8%
Goal Orientation 5 13% 25 66% 8 21%
Alternate Solution Type 6 16% 32 84% 0 0%
Best Solution Type 3 8% 29 76% 6 16%
Yes No Invalid Response
n % n % n %
Empathy/Positivity (problem) 1 3% 37 97% 0 0%
Empathy/Positivity (goal) 0 0% 38 100% 0 0%
Conflict Avoidance 9 24% 29 76% 0 0%
Effective Solution 24 63% 8 21% 6 16%
Adherence Check
Three Solutions Provided 32 84% 6 16% --
Best Solution Provided 32 84% 6 16% --
PRISON OF PEACE RANDOMIZED TRIAL 147

Shower Vignette
Hostility/Hostile Need/Non-hostile Invalid Response
n % n % n %
Codes (T1 N=72)
Conflict Definition 40 56% 16 22% 16 22%
Goal Orientation 7 10% 45 62.5% 20 27.5%
Alternate Solution Type 31 43% 32 44.5% 9 12.5%
Best Solution Type 9 12.5% 43 60% 20 28.5%
Yes No Invalid Response
n % n % n %
Empathy/Positivity (problem) 2 3% 69 96% 1 1%
Empathy/Positivity (goal) 1 1% 69 96% 2 3%
Conflict Avoidance 7 10% 64 89% 1 1%
Effective Solution 40 56% 16 22% 16 22%
Adherence Check
Three Solutions Provided 42 58% 30 42% --
Best Solution Provided 54 75% 18 25% --
Shower Vignette
Hostility/Hostile Need/Non-hostile Invalid Response
n % n % n %
Codes (T3 N=38)
Conflict Definition 20 53% 13 34% 5 13%
Goal Orientation 5 13% 26 68% 7 18%
Alternate Solution Type 13 34% 25 66% 0 0%
Best Solution Type 11 29% 19 50% 8 21%
Yes No Invalid Response
n % n % n %
Empathy/Positivity (problem) 1 3% 37 97% 0 0%
Empathy/Positivity (goal) 0 0% 38 100% 0 0%
Conflict Avoidance 2 5% 36 95% 0 0%
Effective Solution 17 45% 13 34% 8 21%
Adherence Check
Three Solutions Provided 23 61% 15 39% --
Best Solution Provided 31 82% 7 18% --
PRISON OF PEACE RANDOMIZED TRIAL 148

Commissary Vignette
Hostility/Hostile Need/Non-hostile Invalid Response
n % n % n %
Codes (T2 N=59)
Conflict Definition 39 66% 13 22% 7 12%
Goal Orientation 4 7% 45 76% 10 17%
Alternate Solution Type 9 15% 45 76% 5 8%
Best Solution Type 1 2% 47 79.5% 11 18.5%
Yes No Invalid Response
n % n % n %
Empathy/Positivity (problem) 0 0% 59 100% 0 0%
Empathy/Positivity (goal) 0 0% 59 100% 0 0%
Conflict Avoidance 9 15% 50 85% 0 0%
Effective Solution 45 76% 7 12% 7 12%
Adherence Check
Three Solutions Provided 51 86% 8 14% --
Best Solution Provided 50 85% 9 15% --
Hot Water Vignette
Hostility/Hostile Need/Non-hostile Invalid Response
n % n % n %
Codes (T2 N=59)
Conflict Definition 32 54% 18 31% 9 15%
Goal Orientation 5 8% 45 76% 9 15%
Alternate Solution Type 13 22% 40 68% 6 10%
Best Solution Type 5 8% 42 71% 12 21%
Yes No Invalid Response
n % n % n %
Empathy/Positivity (problem) 2 3% 57 97% 0 0%
Empathy/Positivity (goal) 0 0% 59 100% 0 0%
Conflict Avoidance 3 5% 56 95% 0 0%
Effective Solution 35 59.5% 18 30.5% 6 10%
Adherence Check
Three Solutions Provided 41 69.5% 18 30.5% --
Best Solution Provided 53 90% 6 10% --
PRISON OF PEACE RANDOMIZED TRIAL 149

Coffee Vignette
Hostility/Hostile Need/Non-hostile Invalid Response
n % n % n %
Codes (T2 N=53)
Conflict Definition 28 53% 16 30% 9 17%
Goal Orientation 14 27% 23 43% 16 30%
Alternate Solution Type 17 32% 31 58.5% 5 9.5%
Best Solution Type 5 9.5% 36 68% 12 21%
Bunkie Other Invalid Response
n % n % n %
Attribution of Blame
31 58.5% 17 32% 5 9.5%
Yes No Invalid Response
n % n % n %
Empathy/Positivity (problem) 2 4% 51 96% 0 0%
Empathy/Positivity (goal) 2 4% 51 96% 0 0%
Conflict Avoidance 8 15% 45 85% 0 0%
Effective Solution 19 36% 29 55% 5 9.5%
Adherence Check
Three Solutions Provided 39 73.5% 14 26.5% --
Best Solution Provided 46 87% 7 13% --
ALL Vignettes (N=391)
Hostility/Hostile Need/Non-hostile Invalid Response
n % n % n %
Conflict Definition 235 19% 94 76% 62 5%
Goal Orientation 45 12% 255 65% 91 23%
Alternate Solution Type 97 25% 262 67% 32 8%
Best Solution Type 35 9% 267 68% 89 23%
Yes No Invalid Response
n % n % n %
Empathy/Positivity (problem) 14 3.5% 375 96% 2 0.5%
Empathy/Positivity (goal) 7 2% 381 97% 3 1%
Conflict Avoidance 57 14.5% 332 85% 2 0.5%
Effective Solution 223 57% 100 25.5% 68 17.5%
Three Solutions Provided 285 73% 106 27% --
Best Solution Provided 315 80.5% 76 19.5% --
PRISON OF PEACE RANDOMIZED TRIAL
150

APPENDIX E: Intraclass Correlation Coefficients (ICCs) for Conflict Resolution Measure
Composite Scores
Note: Coefficients based on 377 vignettes across all time points. All codes double-coded by two trained research
assistants; Intraclass correlation coefficients using a consistency definition; Single measures are reported; Two-way
mixed model; * fair reliability, ** good reliability, *** excellent reliability.

Variable
T1
(n=2 vignettes)
T2
(n=3 vignettes)
T3
(n=2 vignettes)

Mean Perceived Hostility

.735**
.878*** .881***
Mean Hostility of Solutions
Generated
.774*** .809*** .579*
Mean Effectiveness of Solution .738** .753*** .672**
PRISON OF PEACE RANDOMIZED TRIAL
151

APPENDIX F: Arrest Descriptions at T1 for Participant Sample

Note: Data collected from archival jail records from consenting participants (N=69).
ADW=Assault with a deadly weapon; GBI=Great bodily injury

Description N %
Violent 27 39%
Descriptions: ADW Not Firearm w/GBI;
Corporal Injury on Spouse/Cohabitant; Murder;
Attempted Murder; Stalking/Terrorize Causing
Fear; Attempt to Prevent Victim from
Testifying/Induce False Testimony by Force;
Assault w/ Firearm on Police; Corporal Injury on
Child/ Cruel to Child Likely to Produce
GBI/Death; Battery w/GBI

Thefts 19 28%
Descriptions: Robbery/Burglary; Grand Theft
Auto/Carjacking; Petty Theft; Credit Card Theft

Drugs/Weapons 9 13%
Descriptions: Possession of Drugs (all types,
intent to sell); Under Influence of
Drugs/Alcohol; DUI Causing Injury; Carry
Loaded Firearm: Gang Member; Illegal
Possession of Ammunition

Other 7 10%
Descriptions: Court Ordered Returnee;
Accessory After the Fact/Witness; Disobey
Domestic Relations Court Order; False
Impersonation; Obstruct/Resist Police;
Possession of False Birth Certificate

Missing Description 7 10%
PRISON OF PEACE RANDOMIZED TRIAL
152

APPENDIX G: Significant Interactions Between Missing Data and Baseline Variables

Note: Missing Data=participant missing T2 or T3 assessment data. Incident Reports=number of hostile incidents
recorded at the jail six months prior to recruitment into the study. *= incarceration outcomes occurring post-
recruitment; Re-offended=Participants with one or more IRTS and/or one or more re-arrest. Ethnicity is self-
reported; Other=White, Asian American, Native American, Biracial; ACE=Adverse Childhood Experiences;
*=added in GEE models as covariates.

Categorical Variables
Pearsons Chi Square and Crosstabs
Missing
(%)
Not
Missing
X
2
df
Sig.
(2-sided)
Ethnicity* 7.761 2 .021
Black 10 (34%) 19
Hispanic 16 (55%) 13
Other 13 (76%) 4
Parent Status* 5.509 1 .019
Parent 30 (60%) 20
Not a parent 7 (30%) 16
Condition .692 2 .618
Prison of Peace 14 (56%) 11
Contamination Control 11 (44%) 14
Pure Control 14 (56%) 11

Independent Samples T-Test
Continuous Variables Missing
Not
Missing
t df
Sig.
(2-sided)

GAD Total Score (Mean)* 8.00 11.50 2.547 69 .013
Security Level (Mean)* 5.94 6.09 .374 67 .064
Days in jail pre-recruitment 81.72 153.45 1.973 67 .053
Hostile Incident Reports .51 .91 1.520 69 .025
PRISON OF PEACE RANDOMIZED TRIAL
153

APPENDIX H: Adverse Childhood Experiences (ACE) Questionnaire Item Endorsement

Note: Percentages based on 58 participants who completed ACE at T2.

ACE Item
N %
1. Did a parent or other adult in the household often swear at you,
insult you, put you down, or humiliate you or act in a way that
made you afraid that you might be physically hurt?
30 51%
2. Did a parent or other adult in the household often push, grab,
slap, or throw something at you or ever hit you so hard that you
had marks or were injured?
28 47%
3. Did a parent or other adult in the household often touch or fondle
you or have you touch their body in a sexual way or try to or
actually have oral, anal, or vaginal intercourse with you?
31 53%
4. Did you often feel that no one in your family loved you or
thought you were important or special or your family didn’t look
out for each other, feel close to each other, or support each other?
31 53%
5. Did you often feel that you didn’t have enough to eat, had to
wear dirty clothes, and had no one to protect you or your
parents were too drunk or high to take care of you or take you to
the doctor if you needed it?
12 20%
6. Were your parents ever separated or divorced
47 80%
7. Was your mother or stepmother often pushed, grabbed,
slapped, or had something thrown at her or sometimes or often
kicked, bitten, hit with a fist, or hit with something hard or ever
repeatedly hit over at least a few minutes or threatened with a gun
or knife?
24 41%
8. Did you live with anyone who was a problem drinker or
alcoholic, or who used street drugs
31 53%
9. Was a household member depressed or mentally ill, or did a
household member attempt suicide?
16 27%
10. Did a household member go to prison?
24 41%
PRISON OF PEACE RANDOMIZED TRIAL
154

APPENDIX I: Tests of Differences on Demographic and Correctional Data
Note: PoP=Prison of Peace; CC=Contamination Control; PC=Pure Control. Percentages were calculated from the
number of participants endorsing a behavior divided by the number of participants who gave reports on that
behavior, not the total number of participants.
a
= Incarceration outcomes occurring anytime post-recruitment. Re-
offended=Participants with one or more IRTS and/or one or more re-arrest. Ethnicity is self-reported; Other=White,
Asian American, Native American, Biracial; ACE=Adverse Childhood Experiences; ACE Total Score collected at
T2 and does not represent full sample (N=58).

Prison of Peace, Contamination Control, and Pure Control Participants
Categorical Variables
Pearsons Chi Square and Crosstabs
PoP
n (%)
CC
n (%)
PC
n (%)
X
2
df
Sig.
(2-sided)
Incarceration Outcomes
a

Transferred to Prison 7 (28.0) 8 (32.0) 4 (16.0) 1.833 2 .400
Released 11 (44.0) 11 (44.0) 12 (48.0) .108 2 .948
Re-arrested 5 (22.7) 3 (12.0) 2 (9.0) 1.847 2 .397
Re-offended 15 (68.2) 12 (48.0) 14 (63.6) 2.215 2 .330
Ethnicity 10.296 4 .036
Black 6 (24.0) 16 (64.0 7 (28.0)
Hispanic 12 (48.0) 6 (24.0) 11 (44.0)
Other 7 (28.0) 3 (12.0) 7 (28.0)
Other Variables
Self-reported as a parent 16 (64.0) 17 (73.9) 17 (68.0) .550 2 .760
Current arrest: felony 17 (81.0) 23 (92.0) 19 (90.5) 2.473 4 .649
Current arrest: violent 8 (36.4) 13 (52.0) 4 (18.2) 5.793 2 .055
One-way ANOVA
Continuous Variables PoP (M) CC (M) PC (M) F df
Sig.
(2-sided)

Age (years) 31.56 30.64 33.52 .474 2 .625
Security Level 6.77 6.32 4.91 10.224 2 .000
Days in jail pre-recruitment 130.09 101.16 118.86 .207 2 .814
ACE Total Score 4.45 5.21 4.21 .693 2 .505
PRISON OF PEACE RANDOMIZED TRIAL
155

Note: PoP=Prison of Peace, CC=Contamination Control, PC=Pure Control. Percentages were calculated from the
number of participants endorsing a behavior divided by the number of participants who gave reports on that
behavior, not the total number of participants. Prior IRTS=number of incident reports recorded at the jail six months
prior to recruitment into the study.
a
= incarceration outcomes occurring post-recruitment. Re-offended=Participants
with one or more IRTS and/or one or more re-arrest. Ethnicity is self-reported. Other=White, Asian American,
Native American, Biracial. ACE=Adverse Childhood Experiences.

Prison of Peace Completers (N=16), Contamination Control, and Pure Control Participants
Categorical Variables
Pearsons Chi Square and Crosstabs
PoP
n (%)
CC
n (%)
PC
n (%)
X
2
df Sig. (2-sided)
Incarceration Outcomes
a

Transferred to Prison 6 (37.5) 8 (32.0) 4 (16.0) 2.727 2 .256
Released 7 (43.8) 11 (44.0) 12 (48.0) .105 2 .949
Re-arrested 3 (21.4) 3 (12.0) 2 (9.0) 1.189 2 .552
Re-offended 10 (71.4) 12 (48.0) 14 (63.6) 2.340 2 .310
Ethnicity 9.057 4 .060
Black 4 (25.0) 16 (64.0 7 (28.0)
Hispanic 8 (50.0) 6 (24.0) 11 (44.0)
Other 4 (25.0) 3 (12.0) 7 (28.0)
Other Variables
Self-reported as a parent 10 (62.5) 17 (73.9) 17 (68.0) .583 2 .747
Current arrest: felony 12 (85.7) 23 (92.0) 19 (90.5) 1.912 4 .752
Current arrest: violent 4 (28.6) 13 (52.0) 4 (18.2) 6.204 2 .045
One-way ANOVA
Continuous Variables PoP (M) CC (M) PC (M) F df Sig. (2-sided)

Age (years) 29.63 30.64 33.52 .789 2 .459
Security Level 6.79 6.32 4.91 9.438 2 .000
Days in jail pre-recruitment 133.36 101.16 118.86 .211 2 .811
ACE Total Score 4.50 5.21 4.21 .670 2 .516
PRISON OF PEACE RANDOMIZED TRIAL
156

APPENDIX J: Tests of Differences on Outcome Variables at Baseline
Note: PoP=Prison of Peace, CC=Contamination Control, PC=Pure Control. Incident reports and arrests= number 6
months pre-PoP start date.

Prison of Peace, Contamination Control, and Pure Control Participants
Outcome Variable
One-way ANOVA
PoP (M) CC (M) PC (M) F df Sig. (2-sided)
Anger 19.50 16.50 20.60 1.485 2 .234
Anxiety 10.08 9.77 9.48 .060 2 .942
Conflict Resolution Vignettes
Perceived Hostility 1.504 1.528 1.450 .503 2 .607
Solution Hostility 1.288 1.167 1.170 1.708 2 .189
Solution Effectiveness 1.370 1.177 1.217 1.455 2 .241
Conflict Scale 13.26 10.57 12.88 1.488 2 .234
Depression 8.96 9.55 10.96 .710 2 .495
Friend and Family Frequency of
Contact

With Children 1.353 1.375 1.311 .014 2 .986
With Family 2.083 2.030 2.107 .027 2 .973
With Friends 1.403 1.492 1.406 .038 2 .962
Parental Stress Scale 31.41 30.38 31.57 .085 2 .919
Interpersonal Reactivity Index
Perspective Taking 17.22 19.45 16.80 1.621 2 .205
Empathic Concern 20.48 20.91 19.95 .166 2 .847
Personal Distress 11.17 12.95 11.24 1.372 2 .261
Self-Efficacy 36.58 40.73 36.92 2.964 2 .058
Incident Reports
Total 2.54 1.32 1.64 1.630 2 .204
Non-hostile 1.83 .60 .95 2.596 2 .082
Hostile 0.71 0.72 0.68 .007 2 .993
Arrests 5.73 6.52 7.00 .180 2 .836
PRISON OF PEACE RANDOMIZED TRIAL
157

Tests of Differences on Outcome Variables at Baseline
Note: PoP=Prison of Peace, CC=Contamination Control, PC=Pure Control. Incident reports and arrests= number 6
months pre-PoP start date.

Prison of Peace Completers (N=16), Contamination Control, and Pure Control Participants
Outcome Variable
One-way ANOVA
PoP (M) CC (M) PC (M) F df Sig. (2-sided)
Anger 20.44 16.50 20.60 1.736 2 .185
Anxiety 10.25 9.77 9.48 .079 2 .924
Conflict Resolution Vignettes
Perceived Hostility 1.511 1.528 1.450 .498 2 .610
Solution Hostility 1.354 1.167 1.170 3.031 2 .056
Solution Effectiveness 1.406 1.177 1.217 1.777 2 .179
Conflict Scale 14.06 10.57 12.88 1.872 2 .163
Depression 9.81 9.55 10.96 .359 2 .700
Friend and Family Frequency of
Contact

With Children 1.273 1.375 1.311 .029 2 .971
With Family 2.000 2.030 2.107 .048 2 .953
With Friends 1.542 1.492 1.406 .062 2 .940
Parental Stress Scale 33.91 30.38 31.57 .502 2 .609
Interpersonal Reactivity Index
Perspective Taking 16.07 19.45 16.80 2.190 2 .121
Empathic Concern 19.67 20.91 19.95 .256 2 .775
Personal Distress 11.06 12.95 11.24 1.505 2 .230
Self-Efficacy 36.50 40.73 36.92 2.608 2 .082
Incident Reports
Total 2.56 1.32 1.64 1.280 2 .285
Non-hostile 2.13 .60 .95 3.085 2 .053
Hostile .44 0.72 .68 .476 2 .624
Arrests 6.21 6.52 7.00 .051 2 .950
PRISON OF PEACE RANDOMIZED TRIAL
158

APPENDIX K: Arrest and Incident Report (IRTS) Counts

N=71 available IRTS Records
T1 T2 T3
Counts of IRTS (all) N % N % N %
0 25 33% 53 71% 33 44%
1 14 19% 7 9% 8 11%
2 10 13% 5 7% 6 8%
3 5 7% 5 7% 11 15%
4 6 8% -- -- 3 4%
5 3 4% 1 1.3% 2 3%
6 2 3% -- -- 3 4%
7 1 1.3% -- -- 2 3%
8 -- -- -- -- 1 1.3%
9 -- -- -- -- 1 1.3%
10 1 1.3% -- -- -- --
12 1 1.3% -- -- 1 1.3%
13 1 1.3% -- -- -- --
18 1 1.3% -- -- -- --
21 1 1.3% -- -- -- --
N=69 available arrest Records
Count of Arrests N % N % N %
0 11 15% -- -- 59 79%
1 7 9% 1 1.3% 5 7%
2 6 8% -- -- 4 5%
3 6 8% -- -- 1 1.3%
4 7 9% -- -- -- --
5 4 5% -- -- -- --
6 6 8% -- -- -- --
7 1 1.3% -- -- -- --
8 2 3% -- -- -- --
9 2 3% -- -- -- --
10 2 3% -- -- -- --
11 1 1.3% -- -- -- --
12 2 3% -- -- -- --
13 3 3% -- -- -- --
15 2 3% -- -- -- --
17 2 3% -- -- -- --
19 2 3% -- -- -- --
22 1 1.3% -- -- -- --
29 1 1.3% -- -- -- --
34 1 1.3% -- -- -- --
PRISON OF PEACE RANDOMIZED TRIAL 159

APPENDIX L: Means, Standard Deviations, and Sample Sizes for Outcome Variables

Outcome Variable
T1 T2 T3
Mean SD n
%
miss

Mean SD n
%
miss

Mean SD n
%
miss
SELF-REPORT

Anger 18.96 8.41 71 5% 17.36 8.80 59 21% 16.26 8.50 39 48%
Anxiety 9.77 6.01 71 5% 8.02 6.03 58 23% 9.56 6.42 39 48%
Conflict Resolution
Vignettes

Perceived Hostility 1.49 0.27 69 8% 1.44 0.27 53 29% 1.45 0.32 38 49%
Solution Hostility 1.21 0.26 68 9% 1.17 0.23 58 23% 1.23 0.29 38 49%
Solution Effectiveness 1.26 0.39 63 16% 1.35 0.29 58
23%
1.34 0.40 35
53%
Conflict Scale 12.29 5.60 68 9% 13.12 5.10 59 21% 13.33 5.61 39 48%
Depression 9.85 6.02 71 5% 8.52 6.62 58 23% 9.92 6.90 39 48%
Friend and Family
Frequency of Contact

With Children 1.35 1.05 48 36% 1.37 1.12 37 51% 1.62 1.35 22 71%
With Family 2.08 1.12 71 5% 1.98 1.13 57 24% 2.12 1.19 38 49%
With Friends 1.43 1.21 69 8% 1.46 1.17 56 25% 1.76 1.28 37 51%

PRISON OF PEACE RANDOMIZED TRIAL 160

Outcome Variable
T1 T2 T3
Mean
SD n
%
miss

Mean SD n
%
miss

Mean SD n
%
miss
Parental Stress Scale 31.11 8.64 47 37% 29.34 9.25 35 53% 27.84 7.62 19 75%
Interpersonal Reactivity
Index

Perspective Taking 17.77 5.40 70 7% 17.78 5.25 59 21% 17.57 4.71 37 51%
Empathic Concern 20.45 5.38 66 12% 19.92 5.10 59 21% 18.82 5.16 38 49%
Personal Distress 11.76 4.12 70 7% 10.19 4.52 59 21% 10.97 4.60 39 48%
Self-Efficacy 38.00 6.54 70 7% 39.10 6.57 59 21% 37.05 6.80 37 51%
ARCHIVAL (Ratios)

Arrests .01460 .021 69 8% .00161 .013 69 8% .00278 .008 69 8%
IRTS .01862 .021 71 5% .01270 .026 71 5% .01725 .026 71 5%
IRTS – Non-hostile .01124 .016 71 5% .00100 .021 71 5% .01343 .023 71 5%
IRTS – Hostile .00738 .011 71 5% .00271 .009 71 5% .00381 .008 71 5%
Note: IRTS=Jail Incident Report; % miss = percentage of missing data based off N=75. T1 = pre-intervention, T2 = post-intervention, T3 = 6-month follow-up. Arrests
represent all types of arrest (drug and violent) recorded in the jail records system. Arrests at T1 represent past arrests 6 months pre-recruitment.
c
Conflict Resolution Measure subscales: Perceived Hostility=average of Problem Definition and Goal Orientation; Solution Hostility=average of Alternate Solutions Type
and Best Solution Type. Friend and Family Questionnaire: Frequency of Contact coded as 1=daily to 6=never. Arrests and IRTS values represent ratios of number of arrests
per days in the community and number of IRTS per days in the jail. T1 data=number of offenses in the 6 months pre-PoP start date.

PRISON OF PEACE RANDOMIZED TRIAL 161

APPENDIX M: Unadjusted Means and Standard Deviations of Non-Significant GEE Interactions of
Outcomes and Conditions
Outcome Variable PoP (M) SD
PoP Comp
(M)
SD CC (M) SD PC (M) SD
CRM: Mean
effectiveness of solution

T1 1.3696 .43 1.4063 .42 1.1765 .35 1.2174 .36
T2 1.4417 .32 1.4167 .35 1.3241 .26 1.2750 .26
T3 1.2500 .40 1.3750 .44 1.4231 .34 1.3500 .47
Anxiety
T1 10.08 5.68 10.25 5.20 9.77 6.17 9.48 6.40
T2 6.15 6.43 7.36 7.20 9.89 5.90 8.11 5.39
T3 8.69 6.16 8.38 6.44 11.93 6.21 7.75 6.64
Friend and Family
Frequency of Contact

With children
T1 1.3527 1.03 1.2727 1.18 1.3750 1.22 1.3111 .94
T2 1.5641 1.39 1.2222 1.29 1.1282 1.00 1.4242 .93
T3 1.9444 1.39 2.0000 1.55 1.7000 1.48 1.1667 1.17
With family
T1 2.0833 1.06 2.0000 1.08 2.0303 1.24 2.1067 1.12
T2 1.9500 1.24 2.0000 1.12 1.6667 1.49 2.2982 .94
T3 2.0833 1.02 2.2083 1.01 2.0476 1.25 2.2500 1.35
With friends
T1 1.4028 1.18 1.5417 1.15 1.4924 1.30 1.4058 1.21
T2 1.6140 1.05 1.5714 1.02 1.1852 1.29 1.5789 1.18
T3 1.8056 1.40 2.0000 1.44 1.5952 1.38 1.9091 1.08
Interpersonal Reactivity
Index

Perspective Taking
T1 17.22 4.97 16.07 4.86 19.45 5.34 16.80 5.65
T2 17.20 4.23 17.07 4.81 19.05 5.83 17.15 5.63
T3
16.58 5.25 16.13 5.54 19.07 5.20 16.73 3.07
Empathic Concern
T1 20.48 4.79 19.67 5.39 20.91 5.87 19.95 5.66
T2 19.10 4.62 19.71 5.30 21.00 5.45 19.70 5.29
T3 18.62 5.30 18.88 6.15 18.69 6.20 19.17 4.11
Personal Distress
T1
11.17 4.19 11.06 3.42 12.95 4.70 11.24 3.40
T2 9.70 4.00 9.50 4.60 11.53 4.64 9.40 4.82
T3 10.38 5.04 10.38 6.46 12.29 4.80 10.08 3.85
Self-Efficacy
T1 36.58 4.75 36.50 4.94 40.73 5.61 36.92 8.16
T2
38.40 6.02 38.79 6.30 39.21 6.75 39.70 7.16
T3
36.50 8.94 38.13 7.72 38.57 5.30 35.73 6.04

PRISON OF PEACE RANDOMIZED TRIAL 162

Note: T1=6 months pre-intervention, T2=post-intervention, T3=6 months post-completion. IRTS=jail incident reports.
CRM=Conflict Resolution Measure. PoP=25 participants, PoP Completers=16 participants CRM effectiveness means range
1=effective, 2=ineffective. Frequency of contact coded as 1=daily to 6=never. Arrests and IRTS values represent ratios of number of
arrests per days in the community and number of IRTS per days in the jail. T1 data=number of offenses in the 6 months pre-PoP
start date.

Hostile IRTS
T1 .007125 .012 .004731 .008 .007100 .012 .007964 .011
T2
.001892 .006 .000000 .000 .002240 .008 .004132 .011
T3 .005342 .009 .005113 .010 .002768 .008 .003336 .006
Arrests
T1
.014091 .016 .015293 .019 .011212 .009 .018964 .031
T2
.000000 .000 .000000 .000 .000000 .000 .027775 .024
T3
.006500 .009 .008063 .010 .003807 .008 .003791 .007
PRISON OF PEACE RANDOMIZED TRIAL 163

APPENDIX N: Tests of Differences Between Prison of Peace Completers (N=16) and Non-completers (N=9)

Note: Percentages were calculated from the number of participants endorsing a behavior divided by the number of participants who
gave reports on that behavior, not the total number of participants. M=Mean. Ethnicity is self-reported. Other=White, Asian
American, Native American, Biracial.
a
= incarceration outcomes occurring post-recruitment. Re-offended=Participants with one or
more incident report and/or one or more re-arrest. ACE=Adverse Childhood Experiences.
b
= T2 ACE scores only provided for 14
completers and 6 non-completers. Chi square analyses with cells with low expected counts (<5) may not have enough power to
detect significant differences.

Categorical Variables
Pearsons Chi Square and Crosstabs
Completers
n (%)
Non-
Completers
n (%)
X
2
df Sig. (2-sided)
Ethnicity .198 2 .906
Black 4 (25.0) 2 (22.2)
Hispanic 8 (50.0) 4 (44.4)
Other 4 (25.0) 3 (33.3)
Other Variables
Self-reported as a parent 10 (62.5) 6 (66.7) .043 1 .835
Current arrest: felony 12 (75.0) 5 (55.6) 2.118 2 .347
Current arrest: violent 4 (25.0) 4 (44.4) 1.010 1 .315
Missing data 8 (50.0) 6 (66.7) .649 1 .420
Incarceration Outcomes
a

Transferred to Prison 6 (37.5) 1 (11.1) 1.999 1 .158
Released 7 (43.8) 4 (44.4) .001 1 .973
Re-arrested 3 (18.8) 2 (22.2) .037 1 .848
Re-offended 10 (62.5) 5 (55.6) .187 1 .665

Independent Samples T-Test
Continuous Variables
Completers
(M)
Non-
Completers
(M)
t df Sig. (2-sided)
Age 29.63 35.00 -1.311 23 .203
Security Level 6.79 6.75 .060 20 .953
Days in jail pre-recruitment 133.36 124.38 .097 20 .923
ACE Total Score
b
4.50 4.33 .134 18 .895
PRISON OF PEACE RANDOMIZED TRIAL 164

Note: PoP=Prison of Peace, CC=Contamination Control, PC=Pure Control. Incident reports and arrests=6 months pre-PoP start
date.

Outcome Variable
One-way ANOVA
Completers (M)
Non-completers
(M)
F df
Sig. (2-
sided)
Conflict Resolution Vignettes
Perceived Hostility 1.511 1.489 .047 1 .830
Solution Hostility 1.354 1.156 2.959 1 .100
Solution Effectiveness 1.406 1.286 .368 1 .551
Conflict Scale 14.06 11.43 1.118 1 .302
Anger 20.44 17.63 .763 1 .392
Depression 9.81 7.25 1.454 1 .241
Anxiety 10.25 9.75 .040 1 .844
Parental Stress Scale 33.91 26.83 3.759 1 .072
Interpersonal Reactivity Index
Perspective Taking 16.07 19.38 2.463 1 .131
Empathic Concern 19.67 22.00 1.255 1 .275
Personal Distress 11.06 11.43 .036 1 .852
Self-Efficacy 36.50 36.75 .014 1 .907
Friend and Family Frequency of
Contact

With Children 1.273 1.499 .178 1 .679
With Family 2.000 2.250 .288 1 .597
With Friends 1.542 1.125 .651 1 .428
Incident Reports
Total 2.56 2.50 .002 1 .968
Non-hostile 2.125 1.250 2.186 1 .153
Hostile .44 1.25 .476 1 .624
Arrests 6.21 4.88 .527 1 .604

Abstract (if available)

Abstract Prisoner misconduct (e.g., rule violations, fighting) among incarcerated women leads to significant physical, psychological, and economic costs (Steiner & Wooldredge, 2014a). Several behavioral interventions targeting mental health problems (e.g., trauma, substance abuse) are available for the growing population of female offenders, but very few studies use strong designs and there are no published randomized trials of programs targeting institutional misconduct. Our evaluation tested the efficacy of a 7-week conflict resolution program, Prison of Peace (PoP

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Revision of the multisystemic therapy (MST) adherence coding protocol: assessing the reliability and predictive validity of adherence to the nine MST principles

PDF

A "second chance" program for vulnerable young adults: a program evaluation and a randomized controlled trial of adjunctive motivational interviewing to improve retention

PDF

Predictors and outcomes across the transition to fatherhood

PDF

Targeting sexism with cognitive dissonance

PDF

Investigating discrimination and depression within a couple context

PDF

The role of primary care in the use of specialty mental health services: an investigation of utilization patterns among African Americans and non-Hispanic Whites

PDF

Cumulative risk as a moderator of multisystemic therapy effects for juvenile offenders

PDF

Effects of childhood adversity on physiology and health in emerging adulthood

PDF

The successful African American Parenting (SAAP) Program: an evaluation of a culturally sensitive parenting intervention

PDF

An investigation of the factors associated with electronic aggression among college students

PDF

Is affluence a developmental risk for Hong Kong Chinese adolescents?

PDF

Evaluating a cultural process model of depression and suicidality in Latino adolescents

PDF

Using integrative data analysis to evaluate gender differences in effects of multisystemic therapy for justice-involved youth

PDF

Young adult dating couple interactions in daily life: links to family aggression and physiological processes

PDF

Incarceration trajectories of mothers in state and federal prisons and their relation to the mother’s mental health problems and child’s risk of incarceration

PDF

Reducing gang involvement through employment: a pilot intervention

PDF

Peer Coach Training for disruptive youth

PDF

Cultural influences on mental health stigma in Asian and European American college students

PDF

Psychological distance in the public’s response to terrorism: An investigation of the 2016 Orlando nightclub shooting

PDF

Computational modeling of behavioral attributes in conversational dyadic interactions

Asset Metadata

Creator Gillespie, Marie L. (author)

Core Title Prison of Peace in a jail setting: a randomized trial of a conflict resolution program for incarcerated women

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Psychology

Publication Date 10/18/2018

Defense Date 07/10/2018

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag conflict resolution,incarcerated women,misconduct,OAI-PMH Harvest,program evaluation

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Huey, Stanley J., Jr. (committee chair), John, Richard (committee member), Margolin, Gayla (committee member), Valdez, Avelardo (committee member)

Creator Email marie.gillespie@usc.edu,mls.gillespie@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-83539

Unique identifier UC11675575

Identifier etd-GillespieM-6877.pdf (filename),usctheses-c89-83539 (legacy record id)

Legacy Identifier etd-GillespieM-6877.pdf

Dmrecord 83539

Document Type Dissertation

Format application/pdf (imt)

Rights Gillespie, Marie L.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA