Effect of a relational approach to parent education for incarcerated mothers in a work release center


Material Information

Effect of a relational approach to parent education for incarcerated mothers in a work release center
Physical Description:
ix, 171 leaves : ; 29 cm.
LeBaron, Lesley, 1942-
Publication Date:


Subjects / Keywords:
Counselor Education thesis, Ph.D   ( lcsh )
Dissertations, Academic -- Counselor Education -- UF   ( lcsh )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph.D.)--University of Florida, 2001.
Includes bibliographical references (leaves 162-168).
General Note:
General Note:
Statement of Responsibility:
by Lesley LeBaron.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 028221165
oclc - 49338378
System ID:

Table of Contents
    Title Page
        Page i
        Page ii
        Page iii
        Page iv
    Table of Contents
        Page v
        Page vi
        Page vii
        Page viii
        Page ix
    Chapter 1. Introduction
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
    Chapter 2. Review of the literature
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
    Chapter 3. Methodology
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
    Chapter 4. Results
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
    Chapter 5. Discussion
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
    Appendix A. Demographic questionnaires
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
    Appendix B. Informed consent
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
    Appendix C. Leader manual
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
    Appendix D. Approval letters
        Page 152
        Page 153
        Page 154
    Appendix E. Summary tables
        Page 155
        Page 156
        Page 157
        Page 158
        Page 159
        Page 160
        Page 161
        Page 162
        Page 163
        Page 164
        Page 165
        Page 166
        Page 167
        Page 168
    Biographical sketch
        Page 169
        Page 170
        Page 171
        Page 172
        Page 173
        Page 174
Full Text







Copyright 2001


Lesley LeBaron


I am grateful to all the individuals who provided support and assistance to me

throughout the course of this study.

First and foremost, I would like to thank the mothers who participated in this

study. They were courageous and enthusiastic as they participated in the parenting class,

and they were patient and cooperative in completing the measures. Without them, this

effort would not have been possible. I appreciate Dr. Barbara A. Greadington, director of

Shisa, Inc., for her support of my project and allowing me to come into the work-release


I would like to thank my committee: my chair, Dr. Harry Daniels, and members,

Dr. Jim Archer, Dr. Silvia Echeverria-Doan, and Dr. Dave Miller. I am grateful for their

support, patience, enthusiasm, and commitment to my project. I very much appreciated

my chair, Dr. Daniels, for his encouragement in times of near despair, his time spent

meeting with me on a regular basis despite many other demands, and his insightful and

enthusiastic guidance from beginning to end. Also, without Dr. Miller's expertise and

personal kindness, I would have felt lost in the scary world of statistics.

Finally, many thanks and blessings go to my family and friends. My mother,

Margaret, has been my greatest supporter and cheerleader. She read my dissertation as it

progressed and gave editorial assistance. Her professional and personal interest in my

project, as well as her love, really nourished me. My sister, Allison; my brother-in-law,

Greg; my niece, Jesse; and my nephew, Jonah, kept me sane and grounded with great

conversations, enjoyable meals, and play throughout the two years I worked on this

project. My friends--especially Joanna Jennie, Linda Williams, Kinsey Wingo, and Pat

Schnure--were unendingly supportive and encouraging. In the final lap, Barbara Smerage

applied her fine editing and typing skills, for which I am very grateful.



ACKNOW LEDGM ENTS........................................................................................ iii

ABSTRACT .......................................................................................................... viii


1 INTRODUCTION ............................................................................................ I

Scope of the Problem ........................................................................................ 2
Theoretical Foundations ................................................................................... 5
Statem ent of the Problem .................................................................................. 7
Purpose of the Study......................................................................................... 8
Need for the Study............................................................................................ 9
Rationale for the Study................................................................................... 10
Definition of Term s........................................................................................ 12
Organization of the Study............................................................................... 15

2 REVIEW OF THE LITERATURE ................................................................. 17

W om en in Prison............................................................................................ 18
Current Treatment Programs for Women in Prison................................. 21
M others in Prison................................................................................... 24
Parenting Program s for M others in Prison.............................................. 28
The Relational M odel ..................................................................................... 35
Jean Baker M iller................................................................................... 37
Carol Gilligan ........................................................................................ 38
Relational Health Indices....................................................................... 40
Relational Constructs of Empathy and Mutuality ................................... 42
Perception of Self.................................................................................. 43
Relational Theory in W om en's Groups.................................................. 44
Social Diversity Theory ......................................................................... 48
Application of Relational and Social Diversity Models in Programs
for Incarcerated W om en................................................................ 52
M eta-Em otion Theory .................................................................................... 56
Hiam Ginott's Influence......................................................................... 59
Em otion Coaching ................................................................................. 60

Researching M eta-Emotion.................................................................... 61
Gottman and Colleagues' Recommendations for Parenting Training...... 64
Other Theoretical Considerations.................................................................... 65
Empathy ................................................................................................ 65
Parental Acceptance, Rejection, and Control.......................................... 67

3 M ETHODOLOGY ......................................................................................... 69

Participants..................................................................................................... 69
Variables ........................................................................................................ 73
Dependent Variables.............................................................................. 73
Independent Variable: The Intervention................................................. 74
Demographic Variables.......................................................................... 75
Procedures...................................................................................................... 76
Instrumentation .............................................................................................. 78
Relational Health Indices....................................................................... 78
Balanced Emotional Empathy Scale (BEES).......................................... 81
Parental Acceptance-Rejection/Control Scale (PARQ/Control).............. 82
Assessment of Parental Style (APS)....................................................... 83
Data Analysis ................................................................................................. 84
Hypotheses..................................................................................................... 84

4 RESULTS ...................................................................................................... 86

Introduction.................................................................................................... 86
Statistical Analysis ......................................................................................... 88
Initial Group Equivalence and Bias........................................................ 88
Postintervention Analysis....................................................................... 90
Additional Comments..................................................................................... 98
Summ ary...................................................................................................... 101

5 DISCUSSION .............................................................................................. 102

Introduction.................................................................................................. 102
Overview of the Study.................................................................................. 102
Review of the Literature....................................................................... 102
M ethods............................................................................................... 103
Results................................................................................................. 104
Future Research............................................................................................ 109
Summ ary...................................................................................................... 110


A DEM OGRAPHIC QUESTIONNAIRES....................................................... 111

B INFORM ED CONSENT.............................................................................. 116

C LEADER M ANUAL.................................................................................... 123

D APPROVAL LETTERS............................................................................... 152

E SUM M ARY TABLES ................................................................................. 155

REFEREN CES ..................................................................................................... 162

BIOGRA PHICAL SKETCH ................................................................................. 169

Abstract of Dissertation Presented to the Graduate School of the
University of Florida in Partial Fulfillment of the Requirement for the
Degree of Doctor of Philosophy



Lesley LeBaron

December 2001

Chairperson: M. Harry Daniels
Major Department: Counselor Education

The purpose of this study was to examine the effects of a psycho-educational

parenting class that taught relational skills and parental emotion coaching to incarcerated

mothers in a work-release center. The researcher designed the original materials for the

parenting class, entitled Connecting and Coaching, based on the Relational Model and

Emotion Coaching from Meta-Emotion Philosophy. The basic tools for this research were

paper and pencil assessment instruments to measure relational health, empathic

tendencies, parenting behaviors, and parenting styles.

Participants were 32 mothers in two work-release centers, with 28 completing the

intervention and the posttests. The treatment group was selected to receive the

intervention based on proximity of the instructor and researcher. Participants were

equivalent in most aspects, according to the demographic questionnaire information and

the manner of selection for the work release centers. Two groups in the same location

participated in the eight-session parenting class, which was taught through explicit

instruction of the principles and through practice of skills. The control group received no

parenting instruction.

Data were collected before the intervention and at the end of the intervention. The

control group received delayed treatment after the posttests were administered. Statistical

analysis by Analyses of Covariance (ANCOVAs) of the data revealed no significant

group differences (12 < .05) on any of the measures. However, the instructor, work-release

center staff, and participants reported that they were satisfied with the class.

More research about the value of parenting classes for incarcerated mothers is

recommended. The length of the intervention and the use of paper and pencil

questionnaires alone as measures need to be examined. In addition, longitudinal studies

about parenting success after release and effects on recidivism rates utilizing a variety of

measurement and observation techniques are recommended.


LOST MOMS: The fastest-growing segment of the prison population,
women sometimes find it hard to reconnect with kids. (Cose, 2000, p. 48)

America may be winning the war on crime, but at what cost? Criminals, both

violent and nonviolent, are being incarcerated at increasing rates. A growing number of

criminals in prison are women, particularly women of color who have children. The

effort to reduce crime has led to another problem, the removal of mothers from their

families. In winning the war on crime, are we destroying the most vulnerable families in

our society and creating a new generation of criminals?

The fact that many incarcerated women have children creates a number of critical

problems for these women and their families. Recent Bureau of Justice Statistics (Cose,

2000) indicated that 65.3% of women in state prisons have children; 14.7% of the

children visit at least once a month; and 54.1% of the children never visit. Pollock (1998)

noted that young incarcerated mothers usually have more than one child, and most of

them will lose contact with their children during their incarceration. Many of these

women may have unrealistic expectations about reuniting with their children after

incarceration and may not anticipate the difficulties of day-to-day care of their children

(Pollock, 1998). Because these mothers may have been away from their children for

years at a time, often without contact, these children may become the next generation of

prisoners unless successful interventions are provided. Clearly, developing skills that

foster connection between mothers and children would be beneficial for these

incarcerated women.

Miller and Stiver (1997) contended that "an inner sense of connection to others is

the central organizing feature of women's development" (p. 16). Thus, parenting classes

that help develop this connection and increase group support may enhance mothers'

abilities to relate to their families upon release. Intended to help the mothers develop

relational skills that foster connection while incarcerated with little or no contact with

their children and at the same time learn parenting skills, the intervention was based on

two theories, the Relational Model (Garcia Coll, Cook-Nobles, & Surrey, 1997a; Liang,

Tracy, Taylor, Williams, Jordan, & Miller, 2000; Miller & Stiver, 1997) and Gottman's

meta-emotion theory (Gottman, Katz, & Gooven, 1997; Gottman, Katz, & Hooven,

1996). Parental meta-emotion is closely aligned with the Relational Model. For example,

according to meta-emotion theorists, emotion-coaching parents provide a safe

environment for their family to express feelings, thoughts, and concerns without criticism

or belittlement (Katz, Wilson, & Gottman, 1999). This study examined the effects of

teaching relational skills in an emotion-coaching parent education class for incarcerated

mothers serving the last portion of their sentence in a supervised work-release center.

Scope of the Problem

According to a recent report by the Department of Justice, more than 950,000

women in the United States, or about 1% of the female population, are in correctional

supervision (Bureau of Justice Statistics, 1999). Even though the percentage of convicted

women is small compared to convicted males, the number of women in prison continues

to grow. According to the Status Report on Female Offenders in 1999, the number of

women imprisoned in the United States has tripled in the last decade. As of 1999, more

than 80,000 females were incarcerated in our correctional system (Florida Department of

Corrections, 1999). In Florida, the number of incarcerated women rose from 1,858 in

1988 to 3,633 in 1999, and the number of supervised female offenders rose from 16,875

in 1988 to 33,024 in 1999 (Florida Department of Corrections, 1999).

Race, gender, and class issues impact the judicial system and result in more

imprisonment for women of color (Covington, 1998). According to Pollock's (1998)

summary of the problem, people of color are incarcerated about seven times more often

than Whites and in numbers that are grossly disproportional to their percentage of the

national population. National surveys indicated that about 45% of all female prisoners are

African American, and several state studies indicated that between 10% and 17% are

Hispanic (Pollock, 1998).

Some have hypothesized that minorities may have a greater involvement in

delinquent and criminal behavior, resulting in more frequent arrests and more

incarceration (Machkovitz, 1993). However, researchers contend that greater

incarceration rates of minorities cannot be explained by differences in offense behavior;

rather, minorities seem to be charged more often for serious offenses than Whites

involved in comparable criminal activity (Huizinga & Elliott, 1987). Further evidence of

the racial inequities in the prison system was found in a recent Justice Department study

that reported wide racial and geographic disparities in the federal government's demand

for the death penalty, with minorities far more prevalent than their proportion in the

population (Disparities, 2000). Regarding gender, Pollock (1998) noted some risk factors

for women that correlate with incarceration, including coming from households that

experience poverty, are led by single parents, and have other family members in prison.

Additionally, most women in prison are young, single mothers with more than one child

(Pollock, 1998).

Families are greatly affected by mothers' imprisonment in part because these

mothers often cannot maintain contact with their children while in prison nor regain

custody after release (Brownell, 1997). In addition, incarcerated mothers frequently

experience depression, anxiety, and fears about their children and their inadequacies as

parents, which require attention (Snyder-Joy & Carlo, 1998). Most incarcerated mothers

do not receive rehabilitative programs based on their needs, perhaps because of their

small numbers compared to male inmates (Brownell, 1997; Garcia Coll, Surrey, Buccio-

Notaro, & Molla, 1998). Brownell (1997) commented on the ongoing debate in the

correctional field about whether the criminal justice system should provide rehabilitation

or punishment. Brownell stated that there is increasing interest in "programs that assist

the offender [to] bridge back into the community" (p. 326). In New York, for example, an

increasing number of prisons are offering general medical services; substance abuse

programs; mental health services; violence prevention; and pregnancy, postpartum, and

parenting programs. In addition, there are community-based programs for indigent

women in transition out of correctional settings (Brownell, 1997).

There is clearly a need for parenting programs in prison and in transitional

supervised work-release centers, in part because inmates often have inadequate

knowledge and skills about child development and child behavior management

(Brownell, 1997). Inmate participants and researchers have reported benefiting from

parenting programs for mothers in prison (Harm, Thompson, & Chambers, 1998). With

few notable exceptions (e.g., Gottman et al., 1997; Showers, 1993), there has been little

theoretical discussion or empirical research in the literature about which elements of

parenting programs are helpful. Nonetheless, it is reasonable to assume that parenting

classes that help develop the "inner sense of connection to others" (Miller & Stiver, 1997,

p. 16) and increase group support may enhance mothers' abilities to relate to their

families upon release. In addition, teaching mothers about sensitive and responsive ways

to relate to their children that may help these children develop "self-regulation and

effective socioemotive functioning" (Katz et al., 1999, p. 132) is a worthwhile effort. The

theoretical basis for this assumption follows.

Theoretical Foundations

Although efforts to prepare incarcerated women for reunion with their families

may have been well intended and beneficial, few have focused on the importance of skills

that foster relationships and lead to the deep sense of connection that mothers and

children long for and that can be practiced while the women are isolated from their

families. It seems essential to teach these mothers how to relate to each other in an

empathic way so that they will be able to extend empathy and acceptance to their

children upon release. One theory that is potentially useful for conceptualizing how to

help incarcerated mothers is the Relational Model, which feminist scholars at the

Wellesley College Stone Center for Women's Studies have developed (e.g., Fedele,

1994; Garcia Coll et al., 1998; Liang et al., 2000; Miller & Stiver, 1997). This model

explores the centrality of support and relationships to women. In their extensive

observations of women's lives, Miller and Stiver (1997) found that "an inner sense of

connection to others is the central organizing feature of women's development" (p. 16).

Growth-fostering relationships are the key to the well-being of everyone involved,

according to the Relational Model. Fedele (1994) suggested that groups for women that

foster connections and promote psychological growth and healing address the four

healing factors in the Relational Model, which are "validation of one's experience,

empowerment to act in relationships, development of self-empathy, and mutuality" (p. 3).

Potential change is measured by the Relational Health Indices (Liang et al., 2000).

Several of the Stone Center researchers began applying the Relational Model to

mothers in prison using descriptive methodologies. Garcia Coll et al. (1998) reviewed

transcripts from conversations about mothering obtained in interviews from three

different groups of women (viz., incarcerated, prison-bound, and recently released

mothers), giving special attention to women of diverse racial and ethnic backgrounds.

The researchers identified sources of marginalization and issues of mothering in prison in

order to consider implications for service providers. In an earlier study, Garcia Coil,

Miller, Fields, and Mathews (1997b) used focus groups and questionnaires to obtain

women's thoughts about their most important needs to succeed outside and stay out of

prison. Of primary importance to these women were relationships in and out of prison.

In yet another report, Garcia Coll and Duff (1995) commented on the importance of a

social diversity perspective with the Relational Model for women in prison. Social

diversity is important to consider because of the increasing and disproportionate number

of women of color in prison. Diversity theory purports that "race, ethnicity, and social

climate are ... fundamental influences on human experience and developmental outcome

as well as response to treatment" (Garcia Coll et al., 1997a).

The theoretical underpinnings of the parenting intervention that focuses on

emotion coaching is based on Gottman's (1997) research about raising emotionally

intelligent children. Emotion coaching was developed from the meta-emotion philosophy

(Gottman et al., 1996; Gottman, 1997) and embodies many constructs similar to the

Relational Model, especially in its focus on emotions such as empathy and on

relationships. The research of Gottman et al. (1996) showed that "parental meta-emotion

philosophy is related to both the inhibition of parental negative affect and the facilitation

of positive parenting" and results in a positive impact on a variety of child outcomes (p.

243). A video created by the Gottman Institute (Gottman, 1996) and Gottman's (1997)

book, Raising an Emotionally Healthy Child: The Heart of Parenting, provide the guide

for the intervention discussions about parental emotion coaching that in turn will create

an opportunity for participants to practice the four curative factors of the Relational

Model: "validation of one's experience, empowerment to act in relationships,

development of self-empathy, and mutuality" (Fedele & Harrington, 1990, p. 3).

Statement of the Problem

Incarceration of mothers causes ruptures in the family system. Most mothers who

are in prison lose contact with their children and develop concern for how reunification

will be achieved after release. Parenting classes are recommended for incarcerated

mothers, but they are not offered in many states (Pollock, 1998). In prisons where parent

education is provided, the length of the programs, the number of sessions provided, and

the content of the programs vary. In addition, women also need to deal with their own

problems, which may have resulted in part from mental illness and previous physical

abuse (Henderson, 1998), in order to become effective parents.

One way to help them with their personal problems that may interfere with

parenting skills is to develop peer support using the Relational Model constructs. Women

in prison are influenced by their peers and can share experiences and practice new skills

that will help them grow as individuals and mothers (Boudin, 1998). By developing

relational skills with their peers, mothers in prison may grow to appreciate other women

and be able to share some of the anticipated family-related difficulties that they may

encounter upon release.

A major problem for society and particularly for incarcerated mothers is

recidivism, which refers to the resentencing of formerly released prisoners to prison

terms. According to Shaver (1993), preventing recidivism is a declared goal of all

modem correctional centers. The causes of recidivism are multiple and complex.

According to staff and prisoners, some of the reasons that inmates commit crimes after

release, which result in their return to prison, are their lack of work training, little

education, few life skills, lack of self-help groups or support systems, and, pertinent to

this study, lack of family bonding (Shaver, 1993). Some empirical evidence suggests that

family connections reduce future criminal activity (Hairston, 1988). By developing

relational and parenting skills, the intervention in this research study could help prevent


Purpose of the Study

The purpose of this study was to investigate the effectiveness of a psycho-

educational intervention that teaches relational skills and parental emotion coaching to

women inmates in a work-release center for the purpose of enhancing their relational

health and expanding their parenting knowledge and style. The goal of the study was to

determine whether the intervention that teaches relational skills and parental emotion

coaching would enhance women inmates' relational health, increase empathic tendencies,

increase parental acceptance, and increase their parental emotion coaching style. If this

program enhances incarcerated mothers' relational skills and parenting styles, the

benefits will extend to families, communities, and society, both because their children

will benefit from increased parental skill and because mothers who reintegrate

successfully into their families are more likely to stay out of prison. If using the

Relational Model of interaction and emotion coaching results in the expansion of

perceived parenting skill for incarcerated mothers in a work-release center, there would

be implications for designing prevention programs for incarcerated mothers with the goal

of reducing recidivism in the future. This study provides empirical evidence of the effect

of using a relational approach and emotion coaching in a parent education group with

incarcerated women.

This study employed quantitative measures to determine the effect of a parenting

education class with video and discussions guided by principles of the Relational Model

and emotion coaching on the inmates' self-reported perceptions of their peer relational

health, empathic tendencies, parental behaviors, and parenting styles.

Need for the Study

Relationships are central to women's emotional health, and researchers have

established that the Relational Model is relevant for working with women in prison

(Garcia Coll et al., 1997b). In this study, the researcher attempted to operationalize the

Relational Model with the goal of facilitating peer support in a parent education

intervention, which had not previously been attempted. This study fulfilled two of Garcia

Coil and Duffs (1995) recommendations found in their study of the needs of women in

prison, namely, to "embrace diversity and treat women within the cultural context of their

relationships, including their roles as mothers" and to "provide educational process

groups for offenders which model and enable women in prison to build mutually

empowering relationships" (p. 60). The Relational Model has not been extensively

studied empirically; thus this research added to the body of knowledge about these

constructs. As Liang et al. (2000) noted, it is important to validate relational researchers/

clinicians' recent findings across ethnic groups and other diverse populations.

Parenting programs for women in prison were described in the literature, but only

a few were studied empirically (e.g., Harm et al., 1998; Showers, 1991). Therefore, it was

beneficial to study the effectiveness for women inmates of one parenting program

teaching emotion coaching, a method which was researched extensively with other

populations of parents, with a discussion component based on the Relational Model,

which was researched with college women.

Rationale for the Study

This research was unique in its focus on relational skill development among peers

to enhance parent education for incarcerated women in a work-release center. In an effort

to operationalize the Relational Theory constructs and to test them empirically, the

researcher designed an intervention that incorporated the principles of this theory. It was

hoped that by combining relational skills with parent education based on Gottman's

parenting manual, Raising an Emotionally Intelligent Child (1997), and his video, The

Heart of Parenting (1996), the participants would be able to use their peer support group

in their residential facility to help them as they prepared to reunite with their families.

The first part of the intervention, in which relational concepts were taught and

practiced with role-plays and conversations, was intended to benefit the incarcerated

women by helping them develop empathy for themselves and others and to increase their

relational skills. The second part of the intervention involved a structured educational

group experience using Gottman's (1996) emotion-coaching video, The Heart of

Parenting. The discussions about the videos provided an opportunity to practice relational

skills. It was hypothesized that together the interventions would result in greater

relational health, empathic tendencies, greater parental acceptance, and increased

emotion-coaching parenting style for the mothers participating in the study.

By offering a parenting course on site, the study benefited the organization,

Barbara A. Graedington and Associates (BAG), which manages the Department of

Corrections contract to provide services to these inmates during the last portion of their

prison sentences in residential settings. By providing information grounded in theory and

research about programs that benefit mothers coming out of prison, BAG could apply this

information to future programs.

This same benefit applies to the Florida Department of Corrections, which

received the results of this study and may apply the knowledge to other programs. Some

empirical evidence supports the idea that strong family ties result in less future criminal

activity (Hairston, 1988). Also, children of incarcerated mothers may have a greater

chance of staying out of trouble if their mothers are empathically sensitive to them,

because research indicates that high empathy correlates to lowered aggression in children

(Mehrabian, Young, & Sato, 1988). If the intervention were successful for these

incarcerated mothers, society would benefit because recidivism may be reduced, and their

children's lives may be enhanced by more capable and empathic parenting.

Definitions of Terms

For the purposes of this study, the following terms were used in the manner


Connection refers to the Relational Model's construct that, for women, an inner

sense of being connected to others is the organizing feature of their development.

Dismissing meta-emotion philosophy refers to the thoughts, feelings, and

behaviors of parents who dismissed, ignored, or denied their child's feelings in hopes of

making the emotion go away quickly (Katz et al., 1999).

Emotion coaching refers to parents getting involved with their children's feelings

and teaching their children "strategies to deal with life's ups and downs" (Gottman, 1997,

p. 21). Parents who follow an emotion-coaching meta-emotion philosophy provide a safe

environment for their family to express feelings, thoughts, and concerns without criticism

or belittlement (Katz et al., 1999).

Emotion-dismissing parents are those who believe that their child's negative

emotions were possibly harmful to the child and thus ignore or try to suppress negative

emotions, according to meta-emotion research. They are not providing emotion coaching

to their children.

Emotion regulation describes the theory proposed by Gottman et al. (1997) of

physiological regulation in children, as measured by the opposing processes in the child's

autonomic nervous system, called the vagal "brake" and the sympathetic "accelerator" (p.


Empathy refers to "one's vicarious experience of another's emotional

experiences," according to Mehrabian (2000), and to the action of responding "with

emotions similar to those of others who are present" (Mehrabian et al., 1988, p. 221).

According to one relational theorist, Jordan (1997), empathy allows people to understand

each other's worlds, to "feel at one with" (p. 144). According to Gottman (1997),

empathetic parents use their imaginations to see the situation from the child's

perspective, use their eyes to watch for physical evidence of their children's emotions,

and use their words to reflect back what they are hearing to help their children.

Growth-fostering relationships lead to psychological development. The five

characteristics people in these relationships possess are zest; the ability to act; having an

accurate picture of herself and the other person; a greater sense of worth; and feeling

more connected to others and motivated to make further connections with others (Miller,


Incarcerated mothers are women in jail, prison, or a work-release center who have

children or are pregnant.

Incarcerated women are currently in jail, prison, or a work-release center serving

their sentences.

Marginalization refers to complications experienced by mothers in prison because

of the sense of limited recognition of their problems and needs, perhaps because of their

gender, race, or ethnic background or because of the small number of females in prison

compared to males.

Meta-emotion refers to the executive functions of emotion. Meta-emotion

philosophy can be seen as an index to the degree of intimacy in a family (Katz et al.,


Mutual empathy refers to a joining together of two people based on the authentic

thoughts and feelings of all the participants in a relationship and adds more to their

conversation, according to Jordan (1997).

Parental meta-emotion philosophy refers to an organized set of feelings and

thoughts about one's own emotions and one's children's emotions theorized by Gottman

et al. (1996). Gottman later developed emotion coaching from his extensive research

about parental meta-emotion philosophy.

Parenting classes are educational meetings that incarcerated mothers attend as part

of their rehabilitation to prepare them for reunification with their children.

Parenting styles refers to the four styles of parenting, according to Gottman's

(1997) research: the dismissing parent, the disapproving parent, the laissez-faire parent,

and the emotion coach.

Recidivism refers to ex-prisoners being incarcerated again for crimes committed

after release.

The Relational Model or relational theory explores the centrality of support and

relationships to women. Feminist clinicians and scholars from the Stone Center at

Wellesley College developed this model. It has applications for women in many settings,

including prisoners.

Relational skills refer to "a sense of self-worth, vitality, and validation, a

knowledge of self and others, and a desire for further connection," according to the

Relational Model (Liang et al., 2000, p. 3).

Scaffolding/praising refers to parents being "affectionate, enthusiastic, engaged,

responsive to the child, and positive in structuring a learning task," according to the meta-

emotion theory (Gottman et al., 1997, p. 87). Scaffolding behavior by parents helps boost

the child's confidence as she reaches the next level of competence by teaching and

remarking on small successes.

Work-release centers are the supervised residential facilities housing women in

the last phase of their incarceration. The women earned the privilege of moving to these

facilities because of good behavior in prison. They work in the community and are

supervised when not at work.

Organization of the Study

This study is presented in five chapters and appendices. The present chapter

provides background information and a rationale for the study. Chapter 2 contains a

review of the literature about women in prison, mothers in prison, current programs for

women in prison, the Relational Model, social diversity theory, women's groups, the

Relational Model as it was utilized with women in prison, meta-emotion philosophy and

its application to parent training, and finally other related theoretical considerations.

Chapter 3 presents the methodology, including a description of the participants, the

variables, the procedures, the instrumentation, the data analysis, and the hypotheses.

Chapter 4 presents the statistical analysis of the demographic variables, a postintervention

analysis of covariance of all measures, and the experimental treatment with tables and


additional comments. Chapter 5 provides a brief overview of the study, including a

summary of the literature review, a description of the participants and the procedures

utilized, a discussion of the results and the limitations, and recommendations for future

research. The appendices contain tables, the informed consent packet, required approval

letters, the introduction to the intervention, the intervention, including the goals and

objectives and a session-by-session detailed plan of action.


The purpose of this study was to construct and determine the effectiveness of an

eight-session relational and emotion-coaching parenting intervention for incarcerated

mothers in a community work-release center. The review of the related literature in this

chapter was the basis for the intervention model that was developed. This literature

review is composed of three major sections: women in prison, the Relational Model, and

the meta-emotion theory. The section Women in Prison includes characteristics of

women in prison, current treatment programs for women in prison, transition programs,

mothers in prison, marginalization of mothers in prison, effects of imprisonment on the

family, the need for programs for mothers in prison, and parenting programs for mothers

in prison. The next section, the Relational Model, concerns the theoretical basis for the

relational approach for the intervention. This section includes a discussion of the

development of the Relational Model; Carol Gilligan's work on girls' and women's

development and psychology; growth-fostering relationships; empathy and mutuality; the

perception of self; relational theory in practice; and social diversity theory and the

application of relational and social diversity models for incarcerated women. The next

section of the literature review includes an overview of the theory of meta-emotion

philosophy, the basis for the parenting intervention in this sinudy. Haim Ginott's influence

is noted; emotion coaching is reviewed; a review of the research methodology about

meta-emotion is provided; and Gottman and his associates' recommendations for

parenting training are reviewed. Finally, other theoretical considerations, including

empathy and parental acceptance and rejection, are addressed.

Women in Prison

The typical woman prison inmate is a young minority woman, about 31 years old,

who is a poor, single mother; has a history of substance abuse; and has committed a

nonviolent crime related to her drug use (Henderson, 1998; Pollock, 1998). The number

of women in prison is growing at a faster rate than the number of men in prison

(Watterson, 1996). About 950,000 women, or 1% of all women, are in prison (Bureau of

Justice Statistics, 1999). The majority are not there because they are dangerously violent

but because of arrests for prostitution, vagrancy, illegal drug possession, or property

crimes against others, such as shoplifting, fraud, or theft (Watterson, 1996). Bush-

Baskette (1998) reported that the war on drugs, which federal and state governments have

waged since the 1980s, accounts for the increasing female population in prison. The war

on drugs has also provoked an increase in the number of Black females incarcerated for

drug offenses by 828% between 1986 and 1991 (Bush-Baskette, 1998). Between these

same years, the percentage of all females incarcerated for drug-related offenses more than

doubled (Bush-Baskette, 1998). Bureau of Justice Statistics (1995, cited in Koons,

Burrow, Morash, & Bynum, 1997) indicated that "almost 50% of the women in state

facilities committed their offenses while using drugs or alcohol, and approximately 33%

of these women were serving time for drug-related offenses" (p. 521). In summary, most

of the crimes committed by women are economically driven and involve the abuse of

drugs and alcohol (Covington, 1998).

Measurable differences were noted between African American and White female

inmates. Binkley-Jackson, Carter, and Rolison (1993) enumerated five salient differences

between these two populations: African American women inmates have more children

than White women inmates; their age of first arrest and incarceration is typically earlier;

they are more likely to have used crack or cocaine prior to incarceration; White women

are more likely to have used speed or downers; they are less likely to report histories of

abuse; and they are more likely to report higher levels of self-esteem.

Incarcerated women were shown to have relatively high levels of psychiatric and

psychological disorders (Brownell, 1997; Gorsuch, 1998; Pollock, 1998). Psychological

problems often stem from "histories of physical, sexual, and emotional abuse; battering;

societal marginalization; and an overemphasis on gender-based expectations, such as

caretaking and docility" (Garcia Coll et al., 1998, p. 256). Reexposure to the traumatic

events of childhood that often occurs in prison can have serious implications for the

emotional health and behavior choices of many women in prison (Heney & Kristiansen,

1997). Life events during the 12 months preceding incarceration also were found to affect

the psychological well-being of female prisoners (Keaveny, 1999). In a psychological

survey of 62 incarcerated women, Keaveny (1999) found that nearly 60% of the

participants had "seriously high levels of anxiety and depression symptomatology," both

higher than the mean scores for working women (p. 84). Keaveny's study indicated that

these levels were coupled with limited or nonexistent coping mechanisms.

There are some similarities and some striking differences between incarcerated

women and incarcerated men. In her qualitative research study, Enos (1997) described

some of the differences and similarities between men and women in prison. She found

that men and women in prison have some similar characteristics; for example, there are

large numbers of ethnic and racial minorities. Forty-six percent of the females and 45.5%

of the males are Black; 14% of the females and 16.8% of the males are Hispanic. The

median age of female prisoners is 31 years, 30 for males. However, in terms of general

psychological profile and most other risk factors, there are striking differences. A higher

percentage of women had family members who had served time in prison (47% of the

women versus 37% of the men). Enos (1997) reported that the pain of imprisonment

revolved around family relations, especially the loss of contact with children having more

impact for women; whereas, for men, the pain of imprisonment revolved around "loss of

freedom, autonomy, personal security, heterosexual relationships, and goods and services

that can be found in the larger society" (Enos, 1997, p. 59). Garcia Coll et al. (1998)

reported that many families break ties with women in prison, perhaps because of the

shame, and other families hide the fact that a family member is in prison. This is different

from male prisoners, whose girlfriends, wives, and extended families frequently visit and

bring their children, as reported by these authors. In addition, Garcia Coil et al. (1998)

observed that mothers in prison reported that they tended to get in less trouble than their

male counterparts (e.g., lower frequency of fights and escape attempts); yet, the

restrictive treatment in prison is the same for women as men. Another difference is the

amount of treatment men and women inmates receive. Because of the large number of

males in prison compared to females, male inmates receive most of the health care and

programming (Garcia Coll et al., 1998; Koons et al., 1997). Henderson (1998) reviewed

the recent literature describing the differences between incarcerated, drug-abusing men

and women, noting that most research has focused on male inmates. Henderson noted

that the research on women inmates indicates that they are more likely to have a

coexisting psychiatric disorder, have lower self-esteem, use hard drugs (e.g., heroin and

cocaine), use drugs more frequently, have taken drugs intravenously, test HIV-positive,

and have a higher incidence of active tuberculosis. Women are also almost twice as likely

as men to receive psychotropic drug treatment while in prison, a rate consistent with

prescription drug use in the general female population (Henderson, 1998).

Current Treatment Programs for Women in Prison

Women in prison were neglected in society before their incarceration, and

afterward, by a lack of rehabilitative programming based on their special needs (Boudin,

1998; Garcia Coll et al., 1998; Martin, Cotten, Browne, Kurz, & Robertson, 1995).

Treatment programs are primarily designed to help prisoners develop the health and skills

to stay out of prison (Shaver, 1993). It is in the interest of both offenders and the public

that prison intervention programs "facilitate behavior change, improve positive mental

health, aid in the acquisition of such skills as problem solving and responsible decision

making, and improve personal effectiveness" (Pollock, 1998, p. 188). Ideally, according

to Pollock (1998), the treatment of female offenders should function to serve both types

of clients for whom corrections workers are responsible--the offenders themselves and

the public at large. "To help the offender become a better person, with more coping skills

and prosocial goals, and have the skills to succeed, also helps protect the public from

future lawbreaking" (p. 173). In some cases, the interests of both clients conflict; for

instance, the interests of an incarcerated mother who wishes to regain custody of her

children will conflict with the public interest if the offender's children had a stable home

life during her incarceration or were traumatized by her in the past (Pollock, 1998).

To study how to keep women out of prison, Shaver (1993) studied the language

culture in one women's correctional institution. In her analysis of their language culture,

Shaver attempted to understand recidivism from the point of view of correctional staff

and inmates. In her survey of staff and inmates concerning perceptions about recidivism,

corrections staff responded that among the personal reasons for recidivism were that

incarcerated women

had no social skills, were unable to be self-sufficient, had not broken ties
with old friends, kept old habits, did drugs, didn't recognize consequences
of actions, couldn't provide for self (or family), lacked networking with
recovering people, took easy way out, had no moral values, found it easier
on the "inside" than "on the street," refused to accept responsibilities,
were unable to adjust to a clean, lawful, and productive life, [and] enjoyed
the thrill of committing crimes. (p. 124)

Shaver (1993) also reported that the prison staff also recognized the existence of external

reasons for recidivism, offering factors such as the stigmatization of convicted felons,

lack of training, education, and life skills, lack of support systems, lack of social change,

and lack of family bonding. Three of these problems were addressed by this

investigation: lack of training, lack of life skills, and the need for family bonding. In

another study, Koons et al. (1997) surveyed 67 programs to find out which elements of

programs the prison program staff thought were successful. They reported that officials

and administrators linked successful outcomes to operations and structures of the


By comparison, the responses of the women inmates themselves regarding

internal reasons for recidivism were "just wanted to; power; escape from reality; low self-

esteem; depression; excitement; boredom; wrong choice; [and] ignorance" (Shaver, 1993,

p. 127). External reasons given by the inmates included "money; battering; addiction;

attorney; family problems; sexual abuse; prejudice; association with bad people;

innocent; no skills or job; police; no counseling; men; co-dependency; no education;

being away from family; being alone; the system" (Shaver, 1993, p. 128). Garcia Coll et

al. (1998) confirmed this finding in their commentary on their recent qualitative research,

stating that for some mothers the only hope for a life after prison is to be a mother.

Mothers in prison reported that a common reason for recidivism is being cut off from

their children after release (Garcia Coll et al., 1998). In the study by Koons et al. (1997),

inmates reported that staffing considerations were important, followed by inmates'

acquisition of needed skills. Inmates wanted to have a voice in planning programs,

interpersonallyy responsive and constructive" relationships between staff therapists and

offenders, and positive peer influences (Koons et al., 1997, p. 516). Recognition of the

value of interpersonal relationships and peer influence in treatment lends credence to the

decision to use the Relational Model in this intervention. Particularly relevant to this

study are Garcia Coll and Duffs (1995) recommendations to meet the needs of women in

prison based on their qualitative study: "embrace diversity and treat women within the

cultural context of their relationships, including their roles as mothers" and "provide

educational process groups for offenders which model and enable women in prison to

build mutually empowering relationships" (p. 60).

Phillips and Harm (1997) noted that the themes of support, encouragement, and

connectedness are important to women in prison and community programs after prison,

which confirm the themes of the Relational Model intervention of this study. These

authors also recommended increased community-based programs for the treatment of

drug and alcohol abuse for prevention of relapse and potential recidivism. "A

comprehensive system of care for women prisoners must include advocacy, support,

education, trauma resolution, substance abuse treatment, and family counseling," which

focus on strengths, empowerment, rather than deficits (Phillips & Harm, 1997, p. 8).

Mothers in Prison

The majority of women in prison are mothers (Brownell, 1997; Enos, 1997;

Garcia Coil et al., 1998). While 59.6% of men in prison are fathers, the proportion of

incarcerated women who are mothers is as high as 80% (Harm et al., 1998). Female

offenders were often the primary caregivers to their children prior to incarceration, and

they are likely to reunite with their children after spending less than 2 years in prison

(Martin et al., 1995). The needs of these women, their children, and the extended families

they leave behind have only recently begun to be acknowledged (Garcia Coll et al.,

1998). Some of the literature reports the need for inmates to improve parenting skills,

especially those that have neglected or abused their children (Garcia Coll et al., 1998).

Despite some programs which attempt to connect children with their mothers while in

prison, prison communities are not usually supportive of parent-child relationships

(Garcia Coil et al., 1998).

Garcia Coll and her colleagues (1998) were among the few researchers to address

the impact of incarceration on mothers. However, at least one earlier researcher, Baunach

(1985), provided the early seminal research in this field. In the qualitative study by

Garcia Coll et al. (1998), mothers in prison reported that "separation from their children

was their main concern and that they were deeply affected by it" (Garcia Coil et al., 1998,

p. 259). These incarcerated mothers defined their roles as follows: to provide emotional

care and support, teach their children, provide physical care and support, and possess

personal attributes (e.g., patience) that lead to better understanding and interaction with

their children (Garcia Coll et al., 1998). Baunach (1985) previously reported that the

majority of mothers live with their children prior to incarceration and that many of them

attempt to maintain contact with their children and plan to reunite with them after release.

Perhaps as a result of the constraints within the correctional institution to support

mothering, inmates reported "feelings of inadequacy, despondency, and fear of loss of

their children" (Baunach, 1985, p. 48). They also felt intense guilt, shame, and anger

because their own criminal behavior resulted in separation from their children (Baunach,

1985). The need for emphasis on parenting concerns for mothers in prison, including

prenatal care, the care and placement of children, visitation policies, custody rights,

parenting roles, and the developmental needs of children, was confirmed by other

researchers as well (e.g., Enos, 1997; Koons et al., 1997).

Enos (1997) studied how women managed motherhood while they were

incarcerated and the outcome of this restricted caretaking situation. For women, she

reported, the pain of imprisonment concerns family relations, especially loss of contact

with their children. Imprisonment is especially problematic for mothers who are

concerned about the whereabouts and well-being of their children. The living

arrangements of the children depend on family resources, which are related to the

ethnicity, sex, and race of the children's parents. For example, 90% of the children of

male inmates are likely to be placed with their mothers. When mothers are incarcerated,

the children's lives are more disrupted--25% of their children reside with fathers, 51%

with grandparents, 20% with other relatives, 9% in foster care, and the balance in other

placements (Enos, 1997).

In her interviews with incarcerated mothers, Enos (1997) found that White

women in prison relied more on foster care than Black women. White women had more

family-of-origin problems and were often estranged from their extended families. They

also found it easier to retrieve their children from foster care after their prison term than

from families who were taking care of children. The Black community often does not

trust child welfare agencies to care for their children because of the history of racism,

leading women in prison to usually rely on their relatives for child care (Enos, 1997).

Among the complications experienced by mothers in prison is the sense of limited

recognition of their problems and needs, which is called marginalization by Garcia Coll

et al. (1998). These authors described sources of marginalization of incarcerated mothers,

including the criminal justice system's definition of incarcerated mothers as criminals

only. "Incarcerated mothers have to prove competency; they cannot expect support for

their mothering role. It is implicitly assumed that by choosing crime, they have chosen to

abandon their children" (Garcia Coll et al., 1998, p. 263). Women prisoners are often

demonized in the media (p. 263). Women prisoners often internalize this message, and

their families may, too. For example, as mentioned above, as a difference between

women and men in prison, many families break ties with their female relatives in prison,

perhaps because of the shame, and other families hide the fact that a family member is in

prison. This is different from male prisoners, whose girlfriends, wives, and extended

families frequently visit and bring their children. Some women refuse to exercise their

visiting rights, perhaps because of the shame of being in prison or to avoid the pain of

brief and limited reunification. Marginalization leads the prison system to punish them

for their crimes, without regard for negative consequences, such as separation from their

young children, losing custody of children, and other negative consequences to the

children, according to Garcia Coil et al. (1998).

Garcia Coil et al. (1998) reported that incarcerated women often come from

deprived family systems that do not allow for healthy growth and development, which

frequently results in trauma, violation, substance abuse, poor education, poor work

history, destructive life choices. Many have posttraumatic stress disorder, with symptoms

of flashbacks, hypervigilance, exaggerated startle responses and/or dissociations,

avoidance of stimuli, and numbing. This may lead to irrational behavior on their part,

which can further "pathologize and marginalize" them (p. 264). The consequences of "the

frequently observed lack of internal resources of incarcerated mothers, resources that

could otherwise support their mothering role" also have an effect on marginalizing

mothers in prison (Garcia Coil et al., 1998, p. 263). Furthermore, trauma survivors face

traumagenic dynamics operating in prison that can lead to suicidal behaviors, self-harm,

and substance abuse (Heney & Kristiansen, 1997).

The consequences of substance abuse and addiction need to be addressed before

addicted women can be successful parents (Garcia Coil et al., 1998; Harm et al., 1998).

As Garcia Coil et al. (1998) remarked, an addiction could lead to neglect of children's

needs. When women are in recovery, they express intense shame and guilt about their

mothering during their addiction. It is only in recovery that these mothers can truly

connect with their children. Despite all these constraints, mothering remains the main

focus of incarcerated mothers (Garcia Coll et al., 1998).

Incarceration also affects the children negatively. The children's ages at the time

of incarceration have implications on both the mother and child. Garcia Coil et al.

(1997b) reported that "during infancy and the preschool years, separation from the

mother can have tremendous impact on the patterns of attachment that the child can

develop and therefore lifelong implications for the development of relationships as an

adult" (pp. 24-25). These authors reported that while less traumatic because the child may

understand more, the mother's incarceration during school age is also traumatic because

of changes in residence, school, neighborhoods, and family composition. During

adolescence, the risk of acting out with high-risk behaviors, antisocial and criminal, is

observed; this is a time when parental monitoring is crucial (Garcia Coll et al., 1997b). It

is also important to recognize that children are affected not only by separation from their

mothers but also by the stigma of having a parent in prison (Pollock, 1998). Additionally,

incarceration of mothers can result in severe financial hardship for the children. For

example, welfare reform has resulted in the denial of Title IV-A cash assistance and food

stamp benefits to individuals convicted of a felony drug charge for the possession, use, or

distribution of drugs and, as a result, to their children (Phillips & Harm, 1997).

Parenting Programs for Mothers in Prison

Relationships between incarcerated parents and their children are often difficult,

both during the separation and upon reunion (Showers, 1993). Parents released from

prison often have difficulties disciplining their children, perhaps because the "parenting

models available to them and the home environments they grew up in did not prepare

them to be skilled, effective parents" (Showers, 1993, p. 36). Currently, there is limited

availability of many programs, including parent education, for incarcerated women who

have histories of family and personal problems, which restricts the "ease with which a

prognosis for the mother's rehabilitation or reunification with her children can be

developed" (Beckerman, 1998, p. 515). Service providers face major challenges in how

to promote mothering skills, maintain the hopes of mothers, and help incarcerated

mothers face the realities of their lives (Garcia Coil et al., 1998). Mothers in prison need

help with their psychological growth by teachingig them how to reframe, resist, or cope

with insurmountable circumstances (e.g., possibility of losing custody, impossibility of

finding jobs after having a criminal record, ostracism from relatives and friends)" (Garcia

Coil et al., 1998, p. 270). If the prison is to be used not only as a strategy for punishing

criminal behavior but also as a "window of opportunity" for treatment of women's

emotional problems, the role played by past family violence must be considered in

designing treatment methods (Martin et al., 1995, p. 408).

The establishment of parenting programs that maintain strong family ties during

imprisonment is suggested by numerous authors to be positively correlated with success

of parole after an inmate has been released (Carlson & Cervera, 1991; Hale, 1988). Such

treatment strategies are supported by Baunach's (1985) research and are best instituted in

conjunction with "programs designed to improve mother-child relationships" such as Aid

to Incarcerated Mothers (AIM) and Mother Offspring Life Development (MOLD) (Hale,

1988, p. 151). According to Hairston (1988), there is little evidence to suggest that family

ties make a difference, although it is theorized that family ties can reduce the incidence of

future criminal activity. Reduction of recidivism, in fact, is an expected outcome by

directors of such programs, which typically include parent education, counseling, self-

help, and family support groups (Hairston, 1988).

The need for parenting programs in prisons can also be evaluated based on the

efficacy of such programs in stimulating the interest of participants. According to Pollock

(1998), women in prison "routinely report the need for such programs on needs

assessments and every program or class available is always full" (p. 111). Pollock found

that most women in prison do not resent counselors teaching them how to care for their

children, and factual lectures and discussions are typically attended with great interest.

Incarcerated mothers also need life skills to help them meet the demands of life

after prison, especially those that came from poverty and discrimination (Garcia Coil et

al., 1998). People who work with women prisoners recommend supporting their role as

mothers by "acknowledgment, respect, and support" and helping them maintain contact

with their children (Garcia Coll et al., 1998, p. 270). These mothers need opportunities to

practice mothering while in prison. Supporting mothering in prison helps with

prevention, treatment, and preventing recidivism of these prisoners, according to these

authors, who stated,

When women feel heard, respected as individuals, and connected with others
who care about their well-being, negative behaviors decrease. Until such
treatment approaches are institutionalized by those working with incarcerated
mothers, supporting the seeds of resistance during and after their prison
experiences is critical. (p. 272)

Garcia Coll et al. (1998) discussed one model program for mothers in prison, the

Neil J. Houston House (NJHH) in Boston. The NJHH is a minimum-security community-

based prerelease facility for pregnant women with histories of substance abuse. It was

created in 1989 with the help of Social Justice for Women (SJW), a nonprofit

organization in Boston. The NJHH, a nationally recognized and acclaimed program, is

designed to respond to three critical areas of need in the population of mothers in prison:

"pregnancy, addiction, and criminality" (Garcia Coll et al., 1998), p. 266). The women

serve their sentences in a small facility that permits them to receive prenatal and postnatal

care, substance abuse treatment, and life-skills education.

Another model program is a psychosocial group model that was facilitated by two

prisoners who were mothers (Boudin, 1998). One of the leaders, Boudin, who earned her

master's degree in adult education and adult literacy while serving her sentence, wrote

about her experience in "Lessons from a Mother's Program in Prison: A Psychosocial

Approach Supports Women and Their Children." She and another inmate designed a

program called Parenting from a Distance in Bedford Hills Correctional Facility in New

York State's maximum security prison for women. Their 3-month intensive group

therapy program, wherein participants met 5 days a week for 2.5 hours, had two goals:

"to provide a context for women to explore their experience as mothers through a review

of their lives, and simultaneously, to work on parenting from prison" (p. 105). They

created a process that encouraged women to tell their own stories, which they hoped

would help effect change in their relationships with their children, even from a distance.

Because the group was facilitated by inmates and functioned totally as an inmate group,

the leaders hoped the group would develop a "sense of self-reliance and autonomy" (p.

108). Some cognitive material about parenting and child development was presented, but

"the primary material for the group was the life stories of the women themselves" (p.

108). The approach was supported theoretically from feminist consciousness-raising

groups, group psychotherapy, and trauma recovery process. Over the years, group

members' concerns consistently focused on (a) the childhood traumatic experiences of

the mothers and their impact on their lives and the lives of their children, which tend to

result in failure to protect their children and difficulty in mothering their children; (b)

guilt and shame that most mothers in prison feel about their criminal behavior and its

impact on their children; and (c) loss of the opportunity to mother their children as they

are growing up (Boudin, 1998).

Brownell (1997) described another model program at the Rose M. Singer Center

(RMSC), housed in New York state's Rikers Island Jail, which allows incarcerated

mothers to reside with their newborns for up to 1 year, provided they do not have

communicative disease, a history of psychosis, or charges involving violent behavior.

Additionally, RMSC provides three special services for women inmates with children: (a)

Women Infant Nurturing Services (WINGS), which deals with pregnant, substance-

abusing women, providing both jail-based treatment and postrelease follow-up; (b)

Pregnant and Addicted Mothers Intervention and Rehabilitation (PAIR), which provides

residential or community-based Alternative to Incarceration (ATI) sentences for women

who show rehabilitative promise; and (c) Parent and Child Enrichment Program (PACE),

which involves comprehensive drug treatment and prenatal, postpartum, pediatric, and

parenting support services. Another prison, Bedford Hills Women's Prison, was the first

women's prison in New York State to have a nursery where mothers and their infants

could share a room on a special floor of the prison (Brownell, 1997).

Because of the high rate of drug and alcohol use prior to imprisonment by

mothers in prison, Harm et al. (1998) decided to study the effectiveness of a parenting

program with women who were drug and alcohol abusers. In their study, the researchers

found that 63% of the participants reported a high use of drugs, alcohol, or both. The

researchers used two measures for evaluation, Hudson's Index of Self-Esteem and

Bavolek's Adult-Adolescent Parenting Inventory, which were administered once before

initiation of the program and again at its conclusion, 15 weeks later. The authors

concluded that "alcohol and drug abusing women are not substantially different from

other women in their ability to benefit from parent education classes" (Harm et al., 1998,

p. 75).

Most parenting programs in America address a common set of educational

objectives, including "communication skills, development of higher self esteem,

parenting skills, knowledge of child's growth stages, and the inmate's ability to be aware

of her own feelings" (Clement, 1993, p. 95). In a survey of 36 institutional parenting

programs across the country, Clement (1993) found that Departments of Corrections

nationwide had difficulty determining the needs of female inmates. Part of the difficulty

lies in society's reluctance to recognize that "inmates themselves might be victims

needing treatment" (Clement, 1993, p. 99), an attitude that contributes to the

marginalization of incarcerated women previously outlined. Another explanation for the

inadequate success of such programs is lack of systematic organization for programs; 15

institutions claimed their programs to be modeled after Systematic Training for Effective

Parenting (STEP) or Parent Effectiveness Training (PET), while 23 used no specific

model. Volunteers taught classes in all but 12 institutions. Records in many states were

clearly inadequate, with 15 states lacking records on the number of incarcerated mothers,

32 states lacking knowledge of how many children would be living with their mothers

after incarceration, and 21 jurisdictions lacking information on how many women had

been physically or sexually abused (Clement, 1993).

Several authors have published studies that evaluated the effectiveness of

established parenting programs (Harm et al., 1998; Showers, 1993). Showers (1993) used

the Child Behavior Management Survey to measure the parenting knowledge of female

inmates at one institution, before and after administration of a parenting program based

on the nationally used STEP. It was concluded that women who completed the program

experienced very significant gains in knowledge about child development and nonviolent

behavior management. Both African American and White subjects made similar gains,

suggesting the comprehensive scope of this particular parenting skills intervention

(Showers, 1993).

The mothers who participated in this study were in community-based prerelease

work centers. The goal of community release centers is to reintroduce prisoners gradually

to society by having them live outside the prison and provide them with life skills and

experience in the community (Clement, 1993; Henderson, 1998; Koons et al., 1997;

Pollock, 1998). Most alternatives to incarceration were developed by court systems to

ameliorate overcrowding in jails and to reduce costs; however, there has been little

research concerning recidivism for participants in these programs compared to

incarcerated prisoners (Ryan, 1997). Turnbo and Murray (1997) listed treatment

programs offered by the Federal Bureau of Prisons (BOP), including transitional care

programs of about 6 months of treatment before a prisoner is discharged to prevent

recidivism or relapses in substance abuse. When offenders are released, the

Administrative Office of the U.S. Court, Probation Division, supervises them and offers

additional aftercare substance-abuse/mental health services for up to a year following

release. The BOP has also developed substance abuse treatment units in prisons for

women through which emotional issues can be addressed in group and individual

therapy. The author reported that treatment addresses core issues of incest, rape trauma,

and domestic violence, which have been experienced by many women in the program.

According to the BOP, parent education classes and specialty medical services will help

women offenders "help their own children with positive parenting techniques," thus

discouraging criminal behavior (Tumbo & Murray, 1997, p. 307).

Even though community programs may help to avoid further marginalization of

women, they are still restrictive for mothers, according to Pollock (1998). Koons et al.

(1997) reported that some promising community release programs used an empowerment

model in which skills like improving parenting and relationships were emphasized (p.

517). Their study supports the use of the Relational Model in the intervention in the

present investigation because this model addresses empowerment.

The Relational Model

The Relational Model of understanding women's development was used in this

study as a theoretical basis for implementing and evaluating programs for incarcerated

mothers. This model explores the centrality of support and relationships to women. In

their extensive observations of women's lives, Miller and Stiver (1997) found that "an

inner sense of connection to others is the central organizing feature of women's

development" (p. 16). This section of the chapter addresses the historical development

and structure of Relational Theory as it occurs in the academic literature, including a very

brief overview of the traditional western developmental emphasis as perceived by

feminist thinkers. Jean Baker Miller's (1976) seminal work in the development of the

Relational Model is then presented, with an emphasis on her construct, growth-fostering

relationships. A subsection about Carol Gilligan, who was an important influence on the

work of the relational theorists, follows. Next, Liang, Taylor, Williams, Tracy, Jordan,

and Miller's (2000) Relational Health Indices is discussed, with results from their recent

research. A discussion of Judith Jordan's and others' work on empathy and mutuality

follows. The application of the Relational Model to women's groups is then presented.

Social diversity and its relevance to the Relational Model are presented next. The final

section is devoted to the application of relational and social diversity models in programs

for incarcerated women.

Relational theory was developed by feminist scholars at the Stone Center for

Developmental Services and Studies at Wellesley College as a result of efforts since the

1970s to further understanding of women's development. Its primary contributors were

Miller (1976, 1986a), Jordan (1994, 1997), Surrey (Covington & Surrey, 2000), and

Stiver (Miller & Stiver, 1997). Preceding their work, as well as the work of other feminist

thinkers such as Gilligan (1982, 1991, 1994, 1996) and Chodorow (1989), traditional

western developmental theories emphasized only the growth of an autonomous,

individuated self. Freudian theory, for instance, asserted that relationships were

secondary to, or derived from, the fulfillment of innate instinctual forces and drives

(Jordan, 1997). Other theorists built upon Freud's theory of drives with "profoundly

limited social theories and visions of liberation, radical individualist visions that see all

sociality as constraint" (Chodorow, 1989, p. 115). The liberation that such authors refer

to involves the dissolution of all social relations, equaling complete freedom from

society. Assuming the individualist interpretations were accurate, Chodorow wrote, "We

are left with little understanding of how new human relations could be constructed or of

how people could participate in social bonds or political activity" (p. 115). As Covington

and Surrey (2000) explained, "Women's relational yearnings and the centrality of

relationship for women's psychological health has often been pathologized when viewed

through the lens of traditional, 'self-centered models" (p. 2).

Jean Baker Miller

Countering the focus on the autonomous individual of most of her predecessors,

Jean Baker Miller, a psychiatrist, began to develop a relational theory in her book,

Toward a New Psychology of Women, first published in 1976, and revised in 1986.

Miller summarized her observations of women in relational contexts with the view that

"women's sense of personhood is rooted in the motivation to make and enhance

relatedness with others" (Miller, 1986b, p. 1). She called the kinds of relationships that

lead to psychological development growth-fostering relationships, enumerating five

positive characteristics of people who have developed such connections:

1. Each person feels a greater sense of "zest" (vitality, energy).
2. Each person feels more able to act and does act.
3. Each person has a more accurate picture of her/himself and the other
4. Each person feels a greater sense of worth.
5. Each person feels more connected to the other persons) and feels a greater
motivation for connections with other people beyond those in the specific
relationship. (Miller, 1986a, p. 3)

Miller (1986b) further explained that the essence of what produces these beneficial

effects is movement and mutuality, that is, an emotional, cognitive action on the part of

everyone involved. Miller and Stiver (1997) stated that the feeling of increased vitality

and energy that results for each participant is both emotional and contextual, and without

it, the other four features cannot occur. As a result, each person feels empowered to act in

the moment of the immediate exchange, to act in realms beyond, and to take action

within the relationship (Miller & Stiver, 1997). Developing a sense of worth is predicated

upon each person conveying attention to, and recognizing, the other's experience (Miller,


According to Miller (1986a), the more growth-fostering exchanges a person has

had, the better their foundation of psychological resources. "These resources help us deal

with difficult and conflictual interactions," Miller (1986a, p. 10) wrote. "Mutually

growth-fostering interactions cannot occur if one person has an overwhelming amount of

power to determine what happens in the interactions and uses her/his power in that way."

Carol Gilligan

Carol Gilligan's (1982, 1991, 1994, 1996) theory and research about the

psychological development of girls and women at Harvard is frequently cited by the

relational researchers because her theory resonates with and cross-fertilizes their work. In

her seminal 1986 study of girls in Cleveland, Gilligan (1994) found that, when describing

their relational situations, girls would "cover their voices" with labels--"calling honesty

'stupid' and outspokenness 'unfeeling' or 'thoughtless' or 'selfish' or 'rude'" (p. 242).

Girls approaching adolescence were observed to experience "dissociation" and to devalue

honesty in coping with the realities and needs of differing relationships simultaneously.

This withdrawal of the self from relationships, Gilligan (1996) observed, was a

strategy to make and maintain relationships, and yet at the same time paradoxically

prevented connection from occurring. Such relational crises compromise three basic

human capacities: mutuality, responsive relationship, and emotional communication,

according to Gilligan (1996), concepts that parallel strongly those of the Stone Center


Gilligan's (1991) theory of development emphasizes relationships as channels of

growth, which can also act as sources of psychological pain. "If psychological health

consists," Gilligan (1991) wrote, "most simply, of staying in relationship with oneself,

with others, and with the world, then psychological problems signify relational crises" (p.

23). Included among these crises are the loss of connection with one's thoughts and

feelings, isolation from others, and personal separation from reality, according to Gilligan


Traditional developmental methodology has reflected a cultural bias toward

separation that was generally mistaken for objectivity or neutrality (Gilligan, 1996).

Social scientists who study infants and their mothers (cf., Winnicott, 1958; Benenson,

1996; Stem, 1985; Trevarthen, 1979) have since observed the relational reality that exists

between mother and child, Gilligan (1996) noted. "The psychological world of early

childhood is a relational or interpersonal world," Gilligan (1996, p. 252) wrote. "Infants

desire relationship, find pleasure in human connection, are able to make and maintain

relationship" (Gilligan, 1996, p. 252). Regardless of these findings, the idea that the self

is inextricable from the interpersonal realm within which personal development takes

place has only had a short history in psychoanalytic and behavioral theory. Emphasis

upon the separation process in the models of Freud, Erikson, Mahler, and Bowen is only

recently being questioned by feminist and relational theorists, who hold that "a girl's

development of self is dependent on mutually empathic relationships with primary

caretakers" (Nelson, 1996, p. 339).

Relational Health Indices

Liang et al. (2000) have since expanded upon Jean Baker Miller's earlier

foundations in their efforts to design a measurement device for the Relational Model,

called the Relational Health Indices. In particular, they concluded that intimate and

mutual relationships can facilitate "self-disclosure, emotional resiliency, coping

strategies, and additional social support" (p. 3). The five positive characteristics of

growth-fostering relationships introduced by Miller were expanded to seven by these


1. mutuality;
2. a sense of vitality;
3. a feeling of empowerment;
4. knowledge of self and others;
5. a sense of self-worth and validation;
6. a desire for further connection;
7. a capacity to deal constructively with conflict. (Liang et al., 2000, p. 4)

Among these, four qualities were identified as being particularly significant to relational

perspective--mutual engagement, authenticity, empowerment, and the ability to deal with

difference or conflict. According to studies on closeness and empathy, engagement, as

characterized by mutual involvement, commitment, and attunement, can help alleviate

stress and depression (Liang et al., 2000, p. 4). Engagement is also associated with "self-

esteem, self-actualization, cooperation, low interpersonal distress, and relationship

satisfaction" (Liang et al., 2000, p. 4). Authenticity was defined as the ability to be

genuine in a relationship, to be authentic regarding one's knowledge of self and other. As

a quality of growth-fostering exchanges, authenticity was recognized as being a source of

motivation in relationships. One of Miller's original constructs, empowerment/zest, was

attributed with inducing "positive affect, meaningful activity, and creativity" in a

relationship (Liang et al., 2000, p. 4). High self-esteem, in addition to being affected by

mutual engagement, is also promoted by the ability to deal with difference or conflict

(e.g., differences in background, perspective, and feeling). This leads to "more positive

attitudes toward life, less depression or anxiety, and enhanced internal locus of control"

(Liang et al., 2000, p. 5).

To understand personal development, Liang et al. (2000) identified three types of

connections that bear the most significance for adolescents and young adults: adult

mentors, peers, and community. By mentors, the researchers were referring to non-

parental adult mentors, who until recently have been analyzed only by the traditional

male models of professional development. Through a relational perspective, an adult

mentor is seen as an agent who promotes social interaction and serves as a "buffer against

social stressors" (p. 5). Regarding peer relationships, the authors noted that the vitality of

support in such connections is more important among females than among males.

Women's peer relationships have been shown to be more dyadicc, self-disclosing,

empathic, and intimate" than those of men (p. 6). The third significant type of connection,

community relationships, deals with group affiliation and contrasts with the dyadic

character of mentor and peer relationships. Community relationships contribute to a sense

of belonging that embodies two characteristics: (a) the experience of being valued by

others and (b) the experience of fitting in with others. Women with a strong sense of

belonging are believed to exhibit fewer social and psychological problems, such as

loneliness, depression, and anxiety. Liang et al. (2000) also responded to critics who have

contended that the relational model over-idealizes caring qualities and personal needs

(and thus reinforces the oppression of women) by noting that Jordan and others have

shown growth-fostering relational skills to empower individuals as well as their


Relational Constructs of Empathy and Mutuality

Jordan (1997) focused on empathy in her work. She specified that empathy allows

people to understand each other's worlds, to "feel at one with" (Jordan, 1997, p. 144). An

empathic exchange involves each person as both object and subject. Emotional influence

and communication as well as personal knowledge are mutually affected and received

among the participants of such interactions. Regarding empirical work on empathy by

Martin Hoffman, Jordan (1997) wrote,

In addition to cognitive awareness of another's inner subjective state, in empathic
attunement people resonate emotionally and physically with the other's
experience (this mirroring physiological arousal is sometimes called vicarious
affective arousal). Women typically demonstrate more emotional-physical
resonance with others' affective arousal than do men. (Jordan, 1997, pp. 15-16)

The concept of mutuality as a key component to the Relational Model was

expanded by Genero, Miller, Surrey, and Baldwin (1992). As defined in their paper, the

term mutuality refers to the "bidirectional movement of feelings, thoughts, and activity

between persons in relationships," although "its common usage is circumscribed by

notions of social exchange" (Genero et al., 1992, p. 36). According to these authors,

relationships characterized by mutuality involved both participants experiencing growth

through the relationship, an interactive developmental process. The relational perspective

suggests that both women and men are likely to value mutual relationships positively,

according to Genero et al. Mutuality in close relationships is expected to correlate

positively with the satisfaction, cohesion, and adequacy of social support offered by such

relationships (Genero et al., 1992).

In contrast, mutuality is expected to be negatively correlated with depression, an

expectation that parallels Jordan's association of low self-esteem with relational

nonresponsiveness. This theory was tested by Sperberg and Stabb (1998), who

administered questionnaires to 234 college women for the purposes of measuring

demographics, depression, anger, and mutuality. The authors found that low levels of

mutuality and high levels of suppressed anger in women's relationships were correlated

with depression. In addition, depression beyond that explained by anger could be

predicted by mutuality levels, as measured by the Mutual Psychological Development

Questionnaire (Sperberg & Stabb, 1998).

Perception of Self

Jordan (1997) noted that if self is seen as a separate entity, others may tend to be

perceived as "potential competitors, dangerous intruders, or objects to be used for the

self s enhancement" (p. 16). However, if self is understood as contextual and relational,

with the ability to connect in interaction, others will be understood as "participating in

mutual, relational growth in a particular way that contributes to the connected sense of

self' (Jordan, 1997, pp. 16-17). As Miller (1986a) proposed in her early work. Toward a

New Psychology for Women, however, a relational sense of self also enhances

vulnerabilities when there is threat of disconnection. The disruption of connections, for

many women, is seen as not simply the loss of a relationship but something closer to a

complete loss of self.

In her writings about the relational perspective on self-esteem, Jordan (1994)

suggested that the capacity to participate in growing, moving, changing relationships

should lead to "a sense of well-being, confidence, vitality, and belief in one's capacity to

participate in future relationships in a positive way" (p. 3). Situations that create poor

self-esteem, in fact, are characterized by "relational nonresponsiveness," a feeling of

being unable to create change or movement that lacks mutuality (Jordan, 1994, p. 3).

Severe disruptions in women's sense of self often occur as a result of boundary

violations, which Oakley (1996) referred to as violations of a woman's ability to define

her own physical, sexual, and psychological time and space. Boundary violations

resulting from all forms of abuse can negatively impact a woman's "self-evaluation, self-

empathy, and self-definition" (Oakley, 1996, p. 264). As a result, women often fail to

account for their needs, interests, or capacities in both their daily behavior and their self-


Relational Theory in Women's Groups

The present study was conducted in a psycho-educational group, so that an

understanding of how the shared experience of group discussion functions toward social

reintegration is particularly relevant for a study of a group intervention with incarcerated

women. In fact, Oakley (1996) identified the temporary separation from society that

occurs in group therapy as an important factor for personal and group empowerment and,

ultimately, for eventual social reintegration. Discussing her work with short-term

women's groups, Oakley (1996) concluded that a space within which women can both

develop the sense of continuity from shared life experiences as well as respect and

recognize each other's differences is crucial to women's development (Oakley, 1996).

Stone Center researchers (e.g., Fedele, 1994; Fedele & Harrington, 1990) have

taken strides in the application of relational theory to group psychotherapy. In their paper

documenting its use in women's groups, Fedele and Harrington (1990) noted that a

similar model was used in therapy as early as the 1920s, when Adler and Dreikurs used

group settings to work against the emotional isolation inherent in our culture. Sixty years

later, in 1983, Yalom conducted interpersonal group treatment that focused on "member-

to-member feedback"--the processing of immediate emotional experiences on the part of

group members for the purposes of examining defensiveness and resistance (Fedele &

Harrington, 1990, p. 2). While recognizing the historical roots of their work, Fedele and

Harrington (1990) also emphasized the differences between these early applications and

the application of modem relational theory to group work:

The relational approach concerns itself less with interpretations around
defensiveness and resistance to therapeutic change or to self-awareness. It sees
psychological growth as occurring via changes in a woman's emotional
experience, in her cognitive understanding, in her relational context, and parallel
changes in her sense of self. (p. 3)

Drawing upon the characteristics of growth-fostering relationships laid out by

Miller and other researchers, Fedele and Harrington (1990) identified four factors

important for women's therapy: validation, empowerment, self-empathy, and mutuality.

The first of these factors, validation, deals with group acknowledgment of each member's

emotional experiences. A woman whose expressed perceptions and desires are upheld by

other members of the group will experience a validation of her sense of authenticity, a

process that the authors call the first healing factor. Empowerment also functions as a

healing factor, emerging through group appreciation and encouragement of each

member's ability to effect change. This is further fostered by "repeated support of each

woman's abilities coupled with the explicit and implicit message that every woman has

the right to feel powerful" (p. 4). Healing occurs through self-empathy when women in

group therapy increasingly begin to accept self-representations, while simultaneously

decreasing criticism of their inner self-images. A woman who feels more connected with

her own feelings will set aside identifications resulting from negative past experiences in

order to make room for the healing experiences of current relationships. Lastly, mutuality

contributes to healing by placing value on the enhancement of each member's growth and

by allowing each member room for change. It involves both the open reception of

another's impact on oneself and a respect for one's impact on the other. Of particular note

is the impairment of mutuality in group settings, which occurs when women become

unable to experience and express certain feelings. When a woman becomes withdrawn

and expresses "an intense need for nurturance" from others, others in the group may

accept her neediness, associating it with her losses or other "relational precipitants" (p.

6). This acknowledgment of the woman's justification for nurturance is the first step

toward restoring lost mutuality (Fedele & Harrington, 1990).

In a later paper, Fedele (1994) applied the relational paradox outlined by Miller

and Gilligan to a group therapy setting. This paradox, characterized by a simultaneous

need to make and maintain connections and to develop strategies for disconnection,

creates special demands for therapists. Most importantly, a therapist must respect the

client's need for connection and for distancing strategies, while at the same time maintain

awareness of her own methods for disconnection.

In addition to this basic paradox, Fedele (1994) identified three other paradoxes

that occur in group settings. The first is a paradox between similarity and diversity--the

friction that results between connection of shared experiences and concerns of isolation

due to difference. The experience of growing up female, for instance, is a similarity

shared by all participants of a women's group. Differences in race, class, and sexual

orientation, on the other hand, contribute to diversity and create an important role for

empathy in holding "divergent realities within an empathic relational context" (p. 6).

According to other relational therapists, it is a misconception that when women

acknowledge their differences, they will experience a loss of solidarity (Garcia Coll et al.,

1997a). A second paradox, according to Fedele (1994), states that sharing feelings of

disconnection in group settings will lead to new connection. The process begins with one

group member discussing her disconnected or isolating emotions, which results in other

members mirroring similar feelings they had previously held back. Shared negative

experiences and emotions, then, contribute to reconnection within the group. Fedele

termed the last paradox. "conflict in connection," in reference to conflicts that sometimes

arise within the group (p. 7). By confronting divergent realities that conflict, rather than

ignoring them, the development of the group will actually be furthered. Essentially,

diversity can be more easily tolerated when group members become more connected

(Fedele, 1994).

According to Oakley (1996), the connections that women can make in a "shared

psychic space" are multiple, involving linkages "to their sense of themselves as women,

to their past experiences, to their relationships with one another as a group of women, and

to their shared experience of the culture and society in which they live" (p. 272). The

therapeutic effect of such connections leads to what Oakley (1996) called "mutual

empowerment"--a developmental stage that does not occur in traditional group

psychotherapy literature (p. 280). In this stage, women perceive their connections as

having made them participants in "something larger than oneself," a perception that helps

to increase each participant's sense of personal effectiveness in facing the problems of

daily life (Oakley, 1996, p. 280).

Social Diversity Theory

Because the incarcerated mothers in this study represented a racially and

ethnically diverse group, the social diversity theory of therapy was considered along with

relational theory as a theoretical framework for program implementation. Some Stone

Center researchers engaged in discussions that specifically dealt with the articulation of

diversity and relational theories.

Of particular interest to the researcher designing the intervention were the

obstacles that may prevent growth-fostering relationships from developing across races,

classes, cultures, and ethnic groups. Garcia Coll et al. (1997a) proposed two factors,

collective history and personal history, that function as such barriers. Collective histories

are the "prior and current history and power differentials and conflicts between the

groups that the individuals belong to" (Garcia Coll et al., 1997a, p. 179). In the case of

women with collective histories with power differentials, even between two individuals

who are striving for mutuality, one member may take the role as historical oppressor and

the other as the oppressed. An individual's personal history in dealing with others of a

culturally or racially diverse group is also important in predicting how readily connection

may occur upon future interaction. Often, it is simply a matter of being aware of one's

personal history, and the powerful feelings and prejudices it may have fostered, in order

to overcome such obstacles. Yet, acknowledging both collective and personal histories

can also cause emotional distress: "guilt for being part of the oppressor group, anger for

having been held responsible for your group's historical actions, and shame for realizing

your own contribution as a member of either the dominant or the oppressed group"

(Garcia Coll et al., 1997a, p. 181). In facing such emotions, individuals may deny the

relevance of collective and personal histories to their ongoing relationships, thus

contributing to relational disconnection.

Regarding social diversity in the psychotherapy field in general and applicable to

teaching women from culturally diverse backgrounds, De La Cancela, Jenkins, and Chin

(1993) defined diversity as "an openness to new experiences and progressive ideas

among practitioners and a more flexible and socially contextualized view of personal

histories among clients" (p. 6). A therapist who recognizes diversity in practice, then, is

expected to recognize and accept "alternative lifestyles, biculturality, human differences,

and uniqueness in individual and group life" (De La Cancela et al., 1993, p. 6). As a

result, the quality of therapy increases in both comprehensiveness and authenticity, and

the ethnically diverse client is both empowered and allowed more room for positive

identity development (Chin, De La Cancela, & Jenkins, 1993).

Similar to the Relational Model, the theory of social diversity addresses what was

considered a bias in traditional psychotherapy, in this case, a "pathological diagnosis

bias" that leads therapists to label non-Caucasian, non-western complexes of behavior

and beliefs as "character weakness, emotional disturbance, or mental illness" (De La

Cancela et al., 1993, p. 7). Traditional therapy also was criticized by diversity theorists

for concentrating solely on psychological explanations of human behavior and

overlooking structural explanations (e.g., social and economic causes). In

"decontextualizing the experiences of people of color," such therapists are essentially

supporting institutional racism and establishing obstacles to effective treatment (De La

Cancela et al., 1993). As relational theorists have noted, "psychology's over-emphasis on

personal motivation promotes a failure to develop a mature sense of self in the context of

the larger forces of systemic power" (Garcia Coil et al., 1997a, p. 192).

In addressing the concern of diversity theorists that therapists acknowledge the

pathology existing within society as well as inside the client, Turner (1997) developed six

important areas for exploration as this issue relates specifically to the Black female client:

1. How she feels about and experiences her ethnicity, along with her
perceptions about how others feel about and experience her ethnicity;
2. Her strengths and coping skills in negotiating the "two cultural worlds"
she lives in;
3. Those parts of herself which are responding resourcefully to forces
both within and without her control;
4. The parts of herself and her experiences which cause her pain, hurt,
and frustration internally and externally;
5. The interactive parts of herself, family, work, and social environment
which work beneficially for her in fostering healthy growth and change,
as well as those that interfere with this healthy process;
6. An examination of the extent to which she has acquired internalized
and external meaningful connections and bicultural support systems.
These systems include the affective and therapeutic connection with
the therapist as well as significant others, groups, organizations,
religious affiliations, and the like. (pp. 80-81)

As a solution to ethnocentric bias, the diversity theorists proposed a "conscious

violation of the neutral stance" used commonly in traditional psychotherapy (De La

Cancela et al., 1993, p. 11). This technique involves a therapist-client relationship where

political, social, and psychological connections are explicit. With the neutral stance

eliminated, the therapist and client can more readily establish a sense of solidarity that

aids in the discussion of stresses resulting from racial, ethnic, or gender diversity in an

intolerant culture (De La Cancela et al., 1993). As a positive example of fighting the

neutral stance, Garcia Coll et al. (1997a) quoted relational therapist Surrey from a

discussion regarding Surrey's work with an African American client:

I must be able to hold her profound anger at white racism, be aware of
and open to her concern about my potential for color blindness, and be
available and capable of talking about all this in the therapy without
imposing my need to do so. (p. 190)

Of particular concern for the intervention of the present study is the concept of

dissonant expectations (i.e., anticipation of academic and professional success contrasted

with anticipation of failure in relationships) and their effect upon parenting methods

(Jenkins, 1993). In examining the historical context of African Americans, Jenkins

concluded that a tradition of racism and oppression has resulted in a tendency for African

American women to develop self-defeating modes of coping. As women of color define

themselves by both gender and race (the latter being having more impact as a result of

racism), their self-definition may be one that "embodies a fundamental contradiction: to

help others adapt to, rather than to change, social oppression in order to survive"

(Jenkins, 1993, p. 120). A related expectation, placed upon African American women by

their families, is that they act as "system balancers and tension reducers" for the family

(Chin et al., 1993, p. 178). Such women develop into adulthood with a sense of

powerlessness in determining the course of their lives and experience difficulty in

building connections with others, reported Jenkins (1993). Jenkins continued,

Some gradually begin to view caretaking as a threatening process, which
may only involve total engulfment by needs of others. Consequently, by
young adulthood they are ambivalent or adverse to future possibilities for
marriage and motherhood prior to ever engaging in these experiences.
Caretaking and loving become associated with exploitation and intrusion
rather than trust, mutuality and intimacy. (p. 128)

Racial oppression also results in the unmet fulfillment of basic human needs,

which in turn leads to desperation, according to Chin et al. (1993). It is this desperation

that leads to diminished parental effectiveness, the creation of parent-child role reversals,

and obligations on the part of women to remain in the family and out of relationships

(Chin et al., 1993). Turner (1997) noted that the pressures of Black motherhood derive

from sexism, racism, and classism and, more specifically, from a distorted image of

Black womanhood portrayed through "myths, stereotypes, misconceptions, the media and

Westernized educational systems, as well as through legal, social and political

sanctioning in our society" (p. 77). Turner also pointed out that African American

mothers are faced with the challenge of raising their children in two cultural worlds and

must maintain an intricate balance between teaching and protecting. According to Turner,

to cope with the demands of a racially and culturally divergent society, Black women

who have undergone some degree of integration into White society sometimes

discriminate against other Black women. Such discrimination is founded both on physical

attributes (skin color, hair type, facial contour) and social circumstance (socioeconomic

status, regional background). Turner (1997) further asserted that this internal discord

among Black women reflects "'self-hatred,' insecurity and sometimes 'superiority,'" and

it mirrors the behavior of the majority culture in its way of treating other ethnic groups

(p. 79).

Application of Relational and Social Diversity Models in Programs for
Incarcerated Women

Few studies of incarcerated women have used both relational and social diversity

models as filters through which to understand issues affecting women in institutional

settings, with the exception of Garcia Coll and her associates (1995, 1997) and

Henderson, Schaeffer, and Brown (1998). In a survey of gender-appropriate mental

health services, Henderson et al. (1998) recognized the need to incorporate ethnic as well

as gender sensitivity in mental health treatment for female prisoners:

Women of color comprise well over half of all women incarcerated in
this country. While not rejecting the value of relational theory for
nonwhite women, experts in the mental health of African American
and Hispanic women have argued that their mental health treatment
must be considered within the context of family networks, racial
oppression, and cultural differences. (p. 47)

Most relevant to the current study is the work of Garcia Coll et al. (1997b), who

integrated both relational and social diversity theories for the purposes of evaluating

needs, establishing psycho-educational process groups, and "reframing" perceptions of

mothers in prison. In their work with women in a medium security prerelease prison in

Massachusetts, Garcia-Coll and Duff (1995) identified two sets of needs for women in,

and released from, prison: (a) education, job skills, and housing and (b) psychosocial

needs, including the "promotion of self-esteem, emotional well-being, and meaningful

connection with family, friends, and community" (p. 14). These researchers considered

psychosocial needs to be primary for incarcerated and recently released women, since it

is believed that the two sets of needs are interrelated. Without the emotional support of

others and an understanding of the cultural and relational contexts within which personal

development takes place, women at risk have increased difficulty in "accessing new job

skills, maintaining jobs, a home, economic stability, substance-free lives, and crime-free

lifestyles" ( p. 14).

It is this understanding of the interrelation between economic and psychosocial

needs that justifies the need for relational and social diversity training in the education of

incarcerated women, according to Garcia Coll and Duff (1995). These authors

emphasized that a primary objective of their study was to introduce relational theory and

social diversity concepts to both incarcerated women and co-workers and to make these

concepts understandable and useful. Their process groups were divided into six sessions,

each with a different theme: (a) relational development of women; (b) healthy

relationships; (c) connection, disconnection, and violation; (d) disconnection and

violation: posttraumatic stress syndrome and alcoholism; (e) embracing diversity; and (f)

conflict resolution.

In a later paper, Garcia Coll et al. (1998) established an additional obligation for

prisons and correctional officers to tailor women's services to meet their needs. In

particular is the divergent psychological profile of incarcerated women in comparison to

their male counterparts, resulting from histories of abuse, marginalization, and "an over-

emphasis on gender-based expectations, such as caretaking and docility" (Garcia Coil et

al., 1998, p. 256). Without a relational understanding of women's development for both

inmates and correctional officers, the specific needs that arise in conjunction with the

unique psychological profile of incarcerated women cannot be properly addressed

(Garcia Coll & Duff, 1995).

Redefining the popular perceptions and dismantling the stereotypes of mothers in

prison are equally important reasons for implementing programs of relational and social

diversity education. Most articles depict incarcerated mothers as "inadequate,

incompetent, and unable to provide adequately for the needs of their children," according

to Garcia Coll et al. (1998, p. 257). Mothers in prison are assumed to care little for the

well-being of their children, and their imprisonment is often blamed on their lack of

investment in parenting. Another bias noted in the academic literature involves an

emphasis upon the victimization of children and society that results from the

imprisonment of mothers and disregard for the suffering endured by the mothers

themselves (Garcia Coll & Duff, 1995).

In reviewing transcripts from focus groups of the 1995 study, Garcia-Coll et al.

(1998) attempted to identify evidence of a framework of resistance that incarcerated

mothers experience as a reaction to separation from their children. The authors asked the

inmates how "their actions, thoughts, and feelings" may have constituted "resistance

against the ultimate punishment (for some of them) of being separated from and not

involved in decisions regarding their children" (Garcia Coil et al., 1998, p. 261). That

mothers experience such resistance at all shows the fallacy of claims by journalists and

researchers that imprisoned mothers care little for their children. Yet, certain resistance

strategies can pose as liabilities for effective parenting upon release. As a reaction to the

trials of incarceration itself, for instance, most women shut down emotionally during their

stay in a prison environment. Therapy programs rooted in relational and social diversity

models therefore become necessary to help mothers regain their emotional presence, thus

allowing them to be emotionally available for their children before release (Garcia Coll &

Duff, 1995). Garcia-Coll et al. (1998) recognized that a focus upon resistance creates a

critical shift for those working with incarcerated mothers--a shift with important

implications for treatment and services:

It recognizes that these seemingly negative or maladaptive behavior--albeit
difficult to deal with--are not solely the manifestations of inner
psychopathology. The frustration of dealing with these behaviors can be
alleviated by reconceptualizing them as a response to the women's (present
or past) situation and an expression of their resistance. Following this
reconceptualization, treatment services that are based in solidarity, empathy,
and mutuality may lead to very different kinds of behavior change and
healing. (p. 271)

Meta-Emotion Theory

Meta-emotion theory (Gottman et al., 1997; Katz et al., 1999) is the foundation

for the parenting intervention in the present study because its concepts are closely related

to the Relational Model; both emphasize the importance of connecting with others

through empathic relationships. Based on their extensive empirical data, University of

Washington research psychologists John Gottman, Lynn Fainsilber Katz, and Carole

Hooven formulated a theory that explains the relationship of emotional communication

and the psychological well-being of family members (Gottman, 1987; Gottman et al.,

1996, 1997; Katz et al., 1999). This section of the study introduces the meta-emotion

theory and philosophy, including definitions of constructs and some historical


Katz et al. (1999) believed that meta-emotion describes how a parent feels about

and relates to emotional displays by their child, as well as how the parents feel about their

own feelings about their children in emotional situations. According to these authors, this

construct is a new addition to the field of parenting. To describe important qualities of

parent-child interactions, most developmental psychology research focused on parental

affect and discipline, with variables such as warmth, control, parenting styles, and

responsiveness, but less attention was given to the examination of "parents' feelings and

cognitions about their own affect or their feelings and cognitions about their child's

affect" (Katz et al., 1999, p. 132).

Katz et al. (1999), in their recent chapter devoted to the meta-emotion philosophy,

proposed that "successful emotional communication between family members forms the

basis for intimacy," and that this is "achieved through the awareness of one's own

emotional state and that of other family members, and through the ability to talk about

feelings" (pp. 131-132). Their new concept of parenting, called meta-emotion

philosophy, is based on their finding that "parents have an organized set of feelings and

thoughts about their own emotions and their children's emotions" (p. 132). Meta-emotion

refers to the executive functions of emotion; meta-emotion parallels the theory of

metacognition, which refers to the executive functions of cognition, which has been

formulated by other researchers. Emotion-coaching parents provide a safe environment

for their family to express feelings, thoughts, and concerns without criticism or

belittlement (Katz et al., 1999). Thus, meta-emotion philosophy can be seen as an index

of the degree of intimacy in a family. The term meta-emotion is interpreted in its broadest

sense, meaning not only emotions about emotion but also encompassing feelings and

thoughts about emotion as well as the parents' responses behaviorally to their children

around emotional expression (Katz et al., 1999).

Emotion-coaching parents practicing "a meta-emotion philosophy that values

emotion have less negative marital relations, inhibit displays of parental rejection toward

their children, and are more scaffolding and praising of children during teaching tasks"

(Katz et al., 1999, p. 132). The concept of scaffolding/praising refers to parents being

"affectionate, enthusiastic, engaged, responsive to the child, and positive in structuring a

learning task" (Gottman et al., 1997, p. 87). Scaffolding behavior by parents helps boost

the child's confidence as she reaches the next level of competence by teaching and

remarking on small successes. An example of scaffolding behavior is parents talking in a

slow, calm manner, giving the child just enough information to get started with a task and

then waiting for the child to do something right and offering her specific, not global,

praise for her action, such as, "Good! You're pushing the button at just the right time"

(Gottman et al., 1997, p. 112). Then parents might continue with more instruction.

Finally, the family would repeat the steps, with the child learning the game in increments

(Gottman et al., 1997).

Meta-emotion theory is based in part on Baumrind's research that revealed that

there were two major independent dimensions: permissive/restrictive dimension

(referring to the amount of autonomy parents permit their children) and warm/cold-

hostile dimension (referring to affect) (Katz et al., 1999). Baumnrind then identified three

parental styles based on the permissive/restrictive dimensions: authoritative (controlling,

warm, and nurturing), authoritarian (controlling and cold), and permissive (warm, not

demanding or controlling, and often less involved than the other two types) (Katz et al.,

1999). Baumrind (1967) indicated that the preschool children of authoritative parents

were both socialized and independent, self-controlled and affiliative as well as self-

reliant, explorative, and self-assertive, and generally more competent and content as

compared to the children of authoritarian or permissive parents. In more recent research,

Baumrind (1991) identified four types of parenting styles: indulgent parents who were

permissive or nondirective, responsive but not demanding, nontraditional and lenient;

authoritarian parents who were highly demanding and directive, but not responsive;

authoritative parents who were both demanding and responsive; and uninvolved parents

who were low in both responsiveness and demandingness, and could encompass both

rejecting-neglecting and neglectful parents. Baumrind's (1991) research indicated that

children's behavior related to their parent's parenting styles and that these behaviors

persisted from preschool to adolescence.

Other influences in meta-emotion theory included attachment theorists, who

added a focus of parental responsiveness (Katz et al., 1999). According to attachment

theorists, "sensitive and responsive parenting provides the child with a secure base from

which to explore his or her environment," which leads to the "development of self-

regulation and effective socioemotional functioning" (Katz et al., 1999, p. 137).

Haim Ginott's Influence

The work of Haim Ginott (1965, 1972) greatly influenced the ideas of Gottman,

Katz, and their colleagues (Gottman, 1997; Katz et al., 1999). Ginott (1965) wrote a

parenting manual, Between Parent and Child, that focused on children's emotions and

how parents can make emotional connections with their child that are not critical but

accepting. Ginott (1965) suggested that parents need to look for, understand, and respond

genuinely to the emotions of their children. Ginott believed that most parents already

have the skill of listening to their children and need to focus more on the emotion behind

the child's words. The two most important principles for parents to grasp, according to

Ginott (1965), are "(a) that messages preserve the child's as well as the parent's self-

respect" and "(b) that statements of understanding precede statements of advice or

instruction" (p. 21). Ginott taught that it is important to set limits on children's

misbehavior, but he cautioned against using contemptuous or derogatory remarks because

children may internalize a negative view of themselves. He believed that all feelings are

real and acceptable, but not all actions are, and parents need to set limits on destructive

behaviors. Ginott thought that there is a difference between expressing anger and having

contemptuous feelings toward the child, the latter being more destructive. Katz et al.

(1999) reiterated Ginott's theory in their statement, "It is contempt and defensiveness, not

anger, that is destructive in family relationships" (p. 134). Ginott (1965) believed that one

of the main goals of parenting is intimacy and that having an intimate emotional

relationship with a child results in the ability to obtain the child's compliance and

minimize power struggles. The meta-emotion philosophy attempts to operationalize and

provide research evidence to support Ginott's ideas (Katz et al., 1999).

Emotion Coaching

Emotion coaching includes talking to the child about emotions, helping the child

to verbally label emotions the child is feeling, accepting the child's emotions, discussing

what situations elicited these emotions, and intervening with strategies to cope with these

situations (Gottman et al., 1997). Resulting from research based on meta-emotion

interviews of parents, Katz et al. (1999) and Gottman et al. (1997) suggested that there

are two main parental philosophies, emotion-coaching and dismissing. Emotion-coaching

parents are aware of the role of emotions in their lives and their children's lives, can talk

about emotions, and can assist their children with negative emotions. The emotion

coaching meta-emotion philosophy has five components:

Parents were aware of low intensity emotions in themselves and in their
children, they viewed the child's negative emotion as an opportunity for
intimacy or teaching, they validated their child's emotion, they assisted the
child in verbally labeling the child's emotions, and they problem-solved
with the child, setting behavioral limits and discussing goals and strategies
for dealing with the situation that led to the emotion. (Katz et al., 1999, p,

By contrast, the dismissing meta-emotion philosophy was evidenced by parents

feeling that their child's negative emotions were possibly harmful to the child. These

parents thought that they needed to change their children's negative feelings quickly and

to convey that these feelings will not last, are not important, and that the children could

tolerate these feelings without damage. Gottman and his colleagues found that emotion-

dismissing parents could be sensitive to their children's feelings and wanted to be helpful,

but their approach was to dismiss, ignore, or deny feelings, in the hope of making the

emotion go away quickly. They did not present an insightful description of their child's

emotional experience, and they did not help the child solve problems related to the

child's negative emotions. These parents also saw their child's anger, independent of

misbehavior, as deserving of punishment (Katz et al., 1999).

Researching Meta-Emotion

Gottman and his associates hypothesized that parental meta-emotion philosophy

would have an effect through parenting practices, and they discovered that children in

families that use emotion coaching have greater physiological regulatory abilities

regarding their emotions, which impacts children in a variety of ways, including

improved peer relations, academic achievement, fewer behavior problems, and better

physical health (Gottman et al., 1997; Katz et al., 1999). Their results were based on

physiological measurements, parent reports about behavior, medical records, academic

achievement, and teacher reports to gain information. In addition, the researchers

examined whether meta-emotion philosophy would be related to scaffolding/praising

parenting or to the inhibition of parental derogation. Further, they explored whether

parental meta-emotion philosophy would directly affect emotion regulation abilities, as

assessed by parasympathetic nervous system functioning and parental report (Katz et al.,

1999). The researchers believed not only that these processes are in part biological but

also "malleable and shaped in part by parents through their emotional interactions with

the child" (Katz et al., 1999, p. 144).

To study physiological regulation, Gottman et al. (1997) built a theory called

Emotion Regulation, which refers to the opposing processes in the child's autonomic

nervous system, called the vagal "brake" and the sympathetic "accelerator" (p. 93).

Simply speaking, the vagal brake slows many physiological processes, like the heart rate,

while the sympathetic accelerator speeds them up, both regulating emotions. To study

vagal tone, the researchers used spectral analysis and modifications with high levels of

correlation with other researchers in this area (Gottman et al., 1997). In their research,

they found that "the child's ability to suppress vagal tone at age 5 was a significant

predictor of the child's emotion regulation at age 8," and "[t]he greater the child's ability

to suppress vagal tone at age 5, the less the parents had to down-regulate the child's

negative affects, inappropriate behavior, and over-excitement at age 8" (Katz et al., 1999,

p. 145). Their data suggest that parents who are aware of their child's emotions and who

utilize emotion-coaching during emotional moments have children who are better able to

self-soothe, resulting in an improved ability to calm themselves when upset. These

children also were rated by their teachers as having better peer relationships. In addition,

parents who used scaffolding/praising and were less derogatory had children who

achieved better academically. Finally, children who were coached by their parents during

emotional moments had fewer physical illnesses (Katz et al., 1999).

Research results supported many of Gottman and his colleagues' hypotheses,

including finding linkages for the major pathways they proposed, namely that "coaching

the child's emotions was negatively related to the negative parenting variable, suggesting

the hypothesis that coaching is an inhibitor of parental derogation of the child" (Katz et

al., 1999, p. 145). The researchers also found that parent meta-emotion philosophy was

positively related to the parents' use of scaffolding/praising and significantly related to

the child's physiology. As a result of their research, Gottman and his associates (1997)

suggested that parents influence a child's physiology by emotion coaching, although

directionality needs to be further researched. In addition to measuring the children's

responses physiologically, the researchers also utilized teacher ratings and academic

ratings, the children's health records, and videotapes of parent-child interactions in a

laboratory (Gottman et al., 1997).

Katz et al. (1999) observed that the parents' emotion-coaching abilities were

linked to the marriage itself Their preliminary research in a laboratory setting with

couples discussing a high-conflict subject provided evidence that meta-emotion

philosophy is related to how couples resolve marital disputes, namely, that these couples

express higher marital satisfaction and have greater marital stability, as measured by

occurrence of separation and divorce. According to Gottman and his colleagues'

extensive research on couples, "the critical dimension in understanding whether a

marriage will work or not becomes the extent to which the man can accept the influence

of the women he loves and become socialized in emotional communication" (Katz et al.,

1999, pp. 148-149). In their current research, Katz et al. (1999) hypothesized that

"spouses who have different meta-emotion philosophies will have unstable marriages and

their interaction will be characterized by disappointment, negativity, criticism,

defensiveness, and by eventual emotional withdrawal" (p. 152.).

As they examined children's adjustment from preschool through middle

childhood, Katz et al. (1999) discovered that there may be a set of "general abilities"

underlying the development of social and emotional competence with peers (p. 159).

These abilities were not simply certain behaviors like labeling and expressing one's

feelings, which might be inappropriate socially for middle schoolers when interacting

with peers, but included the ability to inhibit negative affect, being able to self-soothe,

being able to be the focus of attention, and to regulate one's own emotions. Gottman's,

Katz', and their colleagues' future research agenda includes research in more naturalistic

settings so that they can observe interactions of families, examination of gender

differences, and development of a meta-emotion philosophy measure for children, and to

learn more about how meta-emotion philosophy may impact family relationships (Katz et

al., 1999).

Gottman and Colleagues' Recommendations for Parenting Training

Numerous parent education programs have been published, yet empirical research

about which ones are most effective is limited. For example, Gordon's (1970) meta-

analysis of 26 studies of Parent Effectiveness Training indicated only a modest treatment

effect on parents' knowledge, attitudes, and behavior, and on children's self esteem, with

effects enduring up to 26 weeks.

The research study of Gottman et al. (1996) utilizing longitudinal correlational

data resulted in the development and pilot testing of a parent-training intervention in

emotion coaching, outlined in his book, Raising an Emotionally Intelligent Child, The

Heart of Parenting (Gottman, 1997). The parenting portion of the study's intervention is

based on Gottman's parenting book and video.

Gottman and associates proposed a Five-Stage Parent Training Program, which

would include the following:

(a) increasing the parent's awareness of his and her own emotions and the
labeling of these emotions with appropriate words, (b) increasing the
parents' awareness of the child's emotions and of what we might call the
child's developmental "emotion vocabulary" (for example, in preschool
this involves projection and pretend play; in infancy it might involve
Stem's notions of affective attunement (intersubjectivity across
modalities, or Tronick's notions of communicative repair), (c) using
empathy rather than dismissing, punishing or derogating the child's
negative emotions (and attributions of negative traits to the child), (d)
helping the child to label his or her feelings, and (e) problem solving about
emotions and appropriate actions (this includes communicating the
parents' values, discussing limit setting, and discussing the child's goals
and strategies. (Gottman et al., 1997, p. 285)

Other Theoretical Considerations

The theories of empathy and parental acceptance and rejection are presented

briefly in this section because two of the measures administered in this study are based on

these theories. Empathy is measured by the Balanced Emotional Empathy Scale (BEES)

(Mehrabian, 2000), and parental acceptance and rejection is measured by the PARQ/

Control (Rohner, 1999).


Empathy is an important construct of the Relational Model (Miller & Stiver,

1997), although it is not measured by the RHI. Empathy is also a key component in meta-

emotion theory, which was in part inspired by Ginott's focus on respect and empathy by

the parents for the child (Gottman et al., 1997). Given the value attributed to the construct

of empathy in both the Relational Model and emotion coaching, the researcher decided to

include a measure for empathy in the study, the Balanced Emotional Empathy Scale

(BEES) (Mehrabian, 1997).

Mehrabian (2000) studied empathy extensively. He first measured it by his

original Emotional Empathic Tendency Scale (Mehrabian & Epstein, 1972). In his

research, he found that empathic persons "tend to be more interpersonally positive,

affiliative, and adept" (Mehrabian, 2000, p. 2). Additionally, people high in emotional

empathy engaged in more altruistic behaviors, were less aggressive, and scored higher on

measures of moral judgment, according to a review of the relevant literature (Mehrabian,

1997; Mehrabian et al., 1988). He discovered that women tend to be more empathic than

men (Mehrabian et al., 1988). In another study, he reported a strong correlation between

the risk of eruptive violence scale and criminal activity of incarcerated juveniles

(Mehrabian, 1997). Another study "yielded positive intercorrelations among measure of

aggression and violence, positive intercorrelations among measure of emotional empathy,

and negative correlations (ranging from -.22 to -.50, with p < .05) of measures of

aggression and violence with measures of emotional empathy" (Mehrabian, 1997, p.


Research with children has shown that deficits in empathy and remorse are

common in children with disruptive behavior disorders (Hastings, Zahn-Waxler,

Robinson, Usher, & Bridges, 2000). These researchers reported that studies of twins have

identified inherited, genetic components for empathy and prosocial acts, with some

differences in the genders. Girls showed more concern than boys from the second year of

life through adolescence, according to numerous researchers cited in Hastings et al.

(2000). Furthermore, these researchers found that concern for others insulates aggressive

preschoolers from continuing along an aggressive path. Of importance to the current

study, the research of Hastings et al. (2000) suggested that "mothers who are overly strict

and harshly punitive, who do not tend to reason or establish reasonable and consistent

rules, and who strongly show their anger or disappointment with their children, are likely

to impede their children's prosocial development" (p. 543). This evidence validates the

need to teach empathy skills to mothers in order to benefit their children's relational


Additional evidence about the importance of teaching empathy was found in a

study about empathy in young children by Barnett, King, Howard, and Dino (1980), in

which the authors indicated that mothers were more affectionate with their children and

emphasized others' feelings in discipline and nondiscipline situations more often than

fathers. This gender-linked socialization pattern may contribute to the finding that

females tend to respond more empathically than do males during adolescence and

adulthood (Barnett et al., 1980).

Parental Acceptance. Rejection, and Control

Rohner (1999) developed an instrument, the Parental Acceptance-Rejection

Questionnaire (PARQ), "designed to measure individuals' perceptions of parental

acceptance and rejection (i.e., warmth) defining one end of the continuum and parental

rejection defining the other end" (p. 17). Parental warmth is expressed physically by

fondling, hugging, and kissing, for example, and verbally by complimenting the child or

saying nice things about one's child (Rohner, 1999). On the other hand, parental rejection

may be evidenced by the absence or significant withdrawal of warmth and affection and

is expressed in three main ways: "aggression/hostility, neglect/indifference, and in an

'undifferentiated' form where the child perceives his parents) as rejecting him, but

where the expression of rejection is neither clearly aggressive/hostile nor

neglecting/indifferent" (Rohner, 1999, p. 17). The control scale, recently added to the

PARQ, measures the individual's perceptions of parental control (permissiveness-

strictness) and the extent to which parents insist on compliance with their requests or

demands. It does not imply anything about the manner in which parents enforce their

demands. Most of the items in the PARQ refer to behavior rather than attitudes, with the

goal of avoiding the problem sometimes found in research measurement, that is, "when

one has to demonstrate that a link exists between the attitudes expressed by a person and

behavior engaged in by the person" (Rohner, 1999, p. 17).

The Parental Acceptance-Rejection Theory (PART) underlying the PARQ has

been studied worldwide in many different cultures (Rohner, 1999). Results indicated that

rejected children are significantly more hostile, aggressive, or passive aggressive than are

accepted children. Additionally, rejected children tend to exhibit more negative self-

evaluation and are more dependent than accepted children. Research with the PARQ

indicated that adults who were rejected as children are more emotionally unresponsive,

emotionally unstable, and have a more negative world view than adults who as children

were accepted (Rohner, 1999). Thus, extensive research studying PART has supported

that parental acceptance-rejection is significantly associated with the psychosocial

functioning of humans everywhere (Rohner, 1999).

Relevant to this study, PARQ has been used to measure parent education

programs as well. For example, the Compassionate Child-Rearing Parent Education

Program's (1993) success is measured by the PARQ. It was also used to evaluate the

Effective Black Parenting Program (Alvy, 2001).


The purpose of this study was to investigate the effectiveness of an eight-session

psycho-educational intervention that teaches relational skills and emotion-coaching

parenting skills to incarcerated mothers in a work-release center. The goals of the

intervention for incarcerated mothers were to enhance their peer and community

relational health, increase their empathic tendencies, increase their desire to accept and be

warm towards their children, and increase their emotion-coaching parenting style. This

chapter describes the methodology employed in this study. Included are descriptions of

the participants, the variables, the procedures, the data analysis, and the hypotheses. The

variables and instrumentation were chosen for their relationship to the important

background evidence from the literature review.


The participants in the study were incarcerated mothers serving their sentences in

two residential work-release centers in two cities in North Florida, Jacksonville and

Tallahassee. The work-release centers are managed by a nonprofit organization, Shisa

Inc., directed by Barbara A. Graedington, Ph.D., which contracts with the Florida

Department of Corrections to provide services to North Florida women for the last

segment of their incarceration. Qualified by the Florida Department of Corrections for

good behavior, these incarcerated women have earned the right to live in supervised

work-release centers and work in the community. The selected women prisoners in

North Florida can choose between the two work-release centers based on proximity to

family, employment opportunities, and bed availability. Shisa House West in Tallahassee

can house 25 women, compared to 28 in Shisa House East in Jacksonville. The number

of residents can vary in any given week. The average age is currently about 35 years,

which is about the same in both houses. About 80% are mothers, which is the same in

both houses. Most residents are there for 7 to 8 months, with some having only 30 days

remaining in their sentence. Three to five leave per month and are usually replaced within

the same week. Types of crimes committed by residents vary and may include violent

crimes, theft, embezzlement, and drug-related crimes, including possession, but

pedophiles are excluded. During their stay at the centers, they may participate in

programs that aid them in finding and keeping jobs, managing finances, and parenting.

These programs are designed to help them adjust to life after imprisonment and prevent


For this study, 32 mothers in the two centers who had at least 8 weeks remaining

in their sentence volunteered. Four participants in the experimental group discontinued,

lowering the total number of participants to 28. One of these four exercised her right not

to participate, and the other three were remanded to prison for behavior problems

unrelated to the parenting class.

The Jacksonville location, Shisa House East, was used to conduct the intervention

because it was more accessible to the researcher and to the class instructor. The instructor

was an experienced Licensed Mental Health Counselor and Rehabilitation Counselor who

had experience volunteering in prisons. The researcher provided 2 hours of training for

the instructor that included a review of the purpose of the study, the design, and a

thorough review of the leader manual. After each class, the researcher and instructor

communicated by email or telephone about what materials were covered, how the

participants responded, and suggestions for future improvement of the teaching materials.

The 15 participants in the treatment group attended a relational parenting class

entitled, Connecting and Coaching, for eight sessions, each lasting 1.5 hours. Two classes

were offered on different evenings; one was comprised of 6 members and the other of 9

members. The two experimental groups were equivalent at the p < .05 level on all

demographic variables, according to the paper and pencil demographic questionnaire (see

Appendix A), with one exception, according to the results of the t-test analysis performed

on the continuous variables and chi-square analyses on the categorical variables.

On the demographic questionnaire, participants were asked to indicate their race

or ethnic background, marital status, education level, language spoken as a child,

language spoken to children, U.S. citizenship, occupation before incarceration, alcohol

and drug use before incarceration, type of alcohol or drugs used, whether received

treatment for alcohol or drugs, whether arrested for alcohol or drug abuse, domestic

violence history, parents marital status, parents' use of alcohol and drugs, history of child

abuse, age ranges of participants' children, any adopted or foster children, whether

currently pregnant, children's current residence, frequency of visits by children while

incarcerated, was this the first arrest, was this the first incarceration, were other family

members arrested, were other family members incarcerated, and anticipated parental role

after incarceration.

The one significant difference between the two experimental groups was in the

residence of the children while the mothers were incarcerated. In the Tuesday evening

group, one child lived with the father and eight with others, whereas in the Thursday

evening group, four lived with the father and three with others. This variable was not

considered essential in the context of the study; therefore, we can conclude that the

groups were equivalent.

The control group was composed of 13 mothers in the Tallahassee work-release

center, Shisa House West. The control group was approximately equivalent to the

experimental group at Shisa House East. Some indicators of equivalency were how

residents chose the location and the number of residents in each house. Additionally, very

few significant differences between the treatment and control groups were found after

analyzing their answers on demographic questionnaire (see Appendix A). The analysis

revealed several statistically significant differences between the groups, but none that

were considered important for the assumption of equality of the groups for this study. For

example, the number of sisters and whether a boyfriend or husband was violent toward

the woman varied between groups. Additionally, the number of days until release and the

number of months until release were significantly different. These results are reported in

detail in Tables 1 and 2 in Chapter 4.

Prior to instruction, all of the participants gave their written informed consent

(Appendix B); completed the demographic questionnaire; and were pretested on

relational health, empathic tendencies, parental behaviors, and parenting style with self-

report paper and pencil questionnaires. The researcher, with input from the staff,

previously had determined that all participants had at least an eighth-grade reading level.

She was present while participants filled out the measures. She invited questions and

clarifications, including any questions about vocabulary, and was able to clarify and

answer questions.


Dependent Variables

Multiple dependent variables reflecting the relevant constructs retrieved from the

literature review were used to evaluate the outcomes of this investigation. The variables

of interest included the relational health of the participants; their empathic tendencies;

their parental behavior including acceptance, rejection, and control; and their parenting


Participants' relational health was operationalized by their scores on two of the

three scales, peer and community, of the Relational Health Indices (RHI) (Liang et al.,

2000). The empathy variable was operationalized by participants' scores on the Balanced

Emotional Empathy Scale (BEES) (Mehrabian, 2000), which measures the ability to

experience vicariously other people's emotional reactions. Self-report of parental

behaviors, including parental acceptance, rejection, and control, was measured by the

Parental Acceptance-Rejection Questionnaire/Control (PARQ/Control) (Rohner, 1999).

The PARQ/Control measure consists of four scales: perceived parental warmth/affection,

perceived parental aggression/hostility, perceived parental neglect/indifference, and

parental rejection, as well as a total score. Participants' parenting styles were assessed by

Gottman's self-test for parents, the Assessment of Parenting Style (APS) (1997), which

measures the following parenting styles: dismissing, disapproving, laissez-faire, and

emotion-coaching. A description of the psychometric properties of each of these

instruments is provided in the Instrumentation section of this chapter.

Independent Variable: The Intervention

The independent variable was the experimental eight-session parenting class,

entitled Connecting and Coaching. A detailed session-by-session description of the

complete intervention can be found in Appendix C. The two themes--Connecting, from

the Relational Model, and Coaching, from Meta-Emotion theory--were woven into the

intervention. Each session was 1.5 hours in length. The first three sessions of the 8-week

intervention were devoted to learning principles of the Relational Model and relational

skill-building activities. The researcher derived the content for the original course

materials from principles of the Relational Model of women's development (Miller &

Stiver, 1997) and the social diversity model developed by Garcia Coll and Duff (1995)

for their Women in Prison Pilot Project in Massachusetts. Relational development of

women, connection, disconnection, embracing diversity, and exploration of parenting

issues were among the topics presented. The instructor presented brief lectures with

examples, and the participants practiced the model with group discussion and group role-

play exercises in small and large groups.

The next four sessions were based on emotion-coaching parenting, a part of the

meta-emotion theory. Information was incorporated from videotaped sections from

Gottman's video, The Heart of Parenting (1996), and Gottman's parenting book, Raising

an Emotionally Intelligent Child. The Heart of Parenting (1997). The instructor presented

the five steps in Gottman's (1997) emotion-coaching model: Being Aware of the Child's

Emotions, Recognizing the Emotion as an Opportunity for Intimacy and Teaching,

Listening Empathetically and Validating the Child's Feelings, Helping the Child Verbally

Label Emotions, and Setting Limits while Helping the Child Problem-Solve. Additional

strategies from his book were presented as well, such as Avoiding Excessive Criticism,

Scaffolding and Praise to Coach Your Child, and When Emotion-Coaching Is Not

Appropriate. The instructor showed video segments, presented parenting vignettes for

role-playing, and led a discussion of the topic. Discussions were conducted with

relational principles as a foundation.

Demographic Variables

A demographic questionnaire was developed, using as a model other demographic

questionnaires found in the literature about this population. The demographic

questionnaire was administered with the initial consent package. Questions included race

or ethnic background, marital status, education level, language spoken as a child,

language spoken to children, U.S. citizenship, occupation before incarceration, alcohol

and drug use before incarceration, type of alcohol or drugs used, whether received

treatment for alcohol or drugs, whether arrested for alcohol or drug abuse, domestic

violence history, parents marital status, parents' use of alcohol and drugs, history of child

abuse, age ranges of participants' children, any adopted or foster children, whether

currently pregnant, children's current residence, frequency of visits by children while

incarcerated, was this the first arrest, was this the first incarceration, were other family

members arrested, were other family members incarcerated, and anticipated parental role

after incarceration. See the demographic questionnaire in Appendix A.


The intervention, Connecting and Coaching, was created by the researcher after

conducting an extensive review of the literature about mothers in prison. The first phase

of the intervention was inspired by the Relational Model, which had been utilized in

prisons in order to help women identify their needs and to provide a framework for

psycho-educational process groups for these women and staff by Garcia Coil and Duff

(1995). The intervention was reviewed for face validity by Cynthia Garcia Coil, Ph.D., an

expert in the Relational Model applied to women in prison (personal communication,

May 24, 2001). In this personal communication, Garcia Coil mentioned her concerns

about the short duration of eight sessions and the short length of the sessions (1.5 hours)

because it might take a long time for the women to establish the necessary trust. She also

alerted the researcher to be aware of the trauma that had been part of these women's lives

that may make connection with other adults and with their children difficult.

The researcher sought a relevant parenting program based on a theoretical model

that would be similar to the Relational Model, one that would value relational

connections and empathy. After reviewing several parenting programs, including Active

Parenting (Popkin, Garcia, & Woodward, 1986), Compassionate Child-Rearing Parent

Education Program (1993), and the Nurturing Program (Bavolek & Comstock, 1985), the

researcher chose Gottman's (1996) Emotion Coaching model based on his meta-emotion

theory as applied to parent education. The materials for the parenting portion of the

intervention were derived from Gottman's parenting video, The Heart of Parenting

(1996) and his book, Raising an Emotionally Intelligent Child (1997).

In 1999, the researcher met with Barbara A. Graedington, Ph.D., the director of

Shisa, Inc., the organization that manages the Florida Department of Corrections contract

to provide services to North Florida women in their last few months of incarceration in

North Florida, in order to gain access to a group of participants. Dr. Graedington

accepted the proposal, contingent upon Florida Department of Corrections approval,

believing that it would be beneficial to the participants. The proposal for this research

was sent first to the Florida Department of Corrections Bureau of Research and Statistics

and then to the University of Florida Institutional Review Board, both of which approved

the research proposal. See Appendix D for copies of all proposal approvals.

For the pretest, all participants in the treatment and control groups answered four

paper and pencil questionnaires to measure relational health (RHI), empathic tendencies

(BEES), parental behaviors (PARQ/Control), and parenting style (APS). The researcher

read the directions, which included the phrase, "Please answer these questions as you

intend to interact with your children after your release." The researcher also was available

to help any participant individually with vocabulary comprehension.

After the protesting was completed, the experimental group participated in an

eight-session class on relational parenting, entitled Connecting and Coaching. The

instructor was a Licensed Mental Health Counselor and Rehabilitation Counselor who

had experience volunteering in prisons. Each class was 1.5 hours in duration. Following

the intervention, the participants answered the four questionnaires again, with the same

directions. The researcher interviewed the instructor at the end of the eight sessions to

discuss how each teaching strategy seemed to work and participants' reactions to her

teaching practices. Attendance was recorded for all class sessions. The researcher

interviewed staff members at both centers at the end of the eight sessions to check their

log of all life skills classes provided so that the researcher would know of any

confounding variables or other threats to the validity of the study. After the experimental

treatment was completed, the researcher administered the four questionnaires again to the

treatment and control groups.

This study employed quantitative measures to determine the effect of teaching a

relational parenting class, Connecting and Coaching, on incarcerated mothers,

specifically, their peer and community relational health; emotional empathy tendencies;

parental acceptance, rejection, and control; and parenting styles. The study was quasi-

experimental, using a repeated measures, within subjects, pretest-posttest design, with

delayed treatment for the control group. Comparisons of the groups were conducted using

a series of Analyses of Covariance (ANCOVA) with protests as the covariates.

The design was as follows:

01 X 02

01 02 X


Four self-report paper and pencil measures (RHI, BEES, PARQ/Control, and

APS) were administered to all participants. These measures were selected based on the

literature review of the theoretical models and the concerns of incarcerated mothers.

Relational Health Indices

To assess and evaluate the Relational Model as incorporated in the experimental

treatment, the Relational Health Indices (RHI) (Liang et al., 2000) were utilized. The

Relational Model has been used to conceptualize new work in psychotherapy, inpatient

treatment, substance abuse, HIV prevention among women, and prisons (Garcia Coll &

Duff, 1995; Garcia Coll et al., 1997b; Liang et al., 2000). As Liang (2000) reported,

empirical examination of the relational model has been limited by a paucity of validated

instruments designed to measure the model's concepts. The only measure previously

available was the Mutual Psychological Development Questionnaire (MPDQ), which

measured mutuality in close relationships (Liang et al., 2000). The RHI is distinct from

the MPDQ in that it assesses three specific types of relationships (peer, mentor, and

community). Second, the MPDQ is based on impressions during verbal interactions with

a partner, whereas the RHI includes attitudinal and behavioral assessments in other

contexts. In addition, the RHI measures distinct attributes of growth-fostering

relationships, whereas the MPDQ measures the more unitary construct of mutuality.

Because the participants in this study were peers in a community (the work-

release center), the researcher theorized that it would be beneficial to evaluate

relationships in group interactions and communities by the RHI. The researcher

administered two of the three RHI scales, peer and community. Regarding peer

relationships, a review of the previous research (Liang et al., 2000) revealed that women

tend to be more socially mature when they have intense and intimate friendships. The

community scale measures community relationships or group affiliation, which

contribute to a sense of belonging, by which the authors mean the experience of being

valued and fitting into a group of women with shared characteristics. The current research

did not utilize the third measure for mentoring relationships because this was not a

feature of the study.

The Peer Relationship Scale (RHI-P) and the Community Relationship Scale

(RHI-C) are composed of 12 statements each (Liang et al., 2000). Participants are asked

to rate themselves according to these relational domains by a 5-point Likert-type scale

with responses of "never, seldom, sometimes, often, and always," numerically valued

from a 1 to 5 score. High mean scores on each of these scales correspond to a high degree

of relational health in that domain. The RHI is scored by computing a mean for each

individual across the items in the instrument. There has been no norming for this scale in

terms of cutoffs for meaningful high and low scores. The scale can be used best by

examining the covariation between it and others scales (A. Tracy, personal

communication, August 23,2001). The subscale composite Cronbach's alpha coefficients

for internal consistency are peer = .85 and community = .90 (Liang et al., 2000). To

establish validity, the RHI-P was correlated to the MPDQ, with a resultant convergent

validity ofr = .69. Additionally, the RHI-P was similar to two of the scales, mentor and

peer, of Pierce's Quality of Relationship Questionnaire (QRI). Finally, the RHI-P was

assessed for convergent validity by comparing it to the Multidimentional Scale of

Perceived Social Support (MSPSS). Liang et al. (2000) also established concurrent

validity by comparing their measure's composite and subscale scores with Rosenberg's

Self-Esteem Scale (SE); with UCLA Loneliness Scale (RULS); with a depression scale,

Center for Epidemiological Studies Depression Scale (CES-D) (2000); and finally with

the Aperceived Stress Scale (PSS). All of these measures have good construct,

discriminant validity, and good reliability. Participants in the Liang et al. (2000) study to

establish validity and reliability for the RHI were college-aged White, Asian, Hispanic,

Black, and Native American women. Their data confirmed that relational health, as

defined and operationalized in their study, was associated with mental health and

adjustment in college-aged women.

The current research project with incarcerated mothers utilizing the Relational

Model, as measured by the RHI, added to the knowledge about the usefulness of the RHI

for different populations. The scales of peer relationships and community relationships

were especially relevant for this study because of the intervention's aims to increase peer

and community support within the residential setting.

Balanced Emotional Empathy Scale (BEES)

The researcher selected a widely used reliable and valid instrument, the BEES, to

measure one of the most important constructs of the Relational Model--empathy. The

BEES measures emotional empathy, which Mehrabian (2000) defined as "one's vicarious

experience of another's emotional experiences--feeling what the other person feels" (p.

2). Mehrabian updated and improved his original empathy measure, the Emotional

Empathic Tendency Scale (EETS) (Mehrabian & Epstein, 1972) with the BEES. In

comparing the BEES with scales of violence and aggression, Mehrabian (1997) found

that BEES constituted a unitary dimension. The BEES correlates with the Risk of

Eruptive Violence Scale (REV), which has been administered to incarcerated juveniles.

In an earlier study using a modified version of the EETS for children, researchers

reported that high empathy scores correlated negatively with teacher ratings of child

aggression (Mehrabian et al., 1988). Mehrabian (2000) confirmed in his study that

emotional empathy as measured by the BEES correlated highly with relationship success,

as measured by a number of other personality scales.

The BEES is a 30-item scale, in which respondents use a 9-point agreement-

disagreement scale to report the degree of their agreement or disagreement with each

item. It generally takes the respondents about 10 minutes to answer the questions. It is

designed "to reduce 'acquiescence bias' (i.e., the tendency of some people to agree with

most statements put to them and the tendency of others to generally disagree with any

statement)" (Mehrabian, 2000, p. 3). The internal consistency coefficient alpha of the

BEES is .87 (Mehrabian, 2000). Additional evidence for validity of the BEES was

obtained indirectly through its high positive correlation of .77 with the original EETS, for

which validity has been established (Mehrabian & Epstein, 1972).

Parental Acceptance-Rejection/Control Scale (PARQ/Control)

The researcher administered a widely used self-report measure of parental

behaviors, Parental Acceptance-Rejection/Control Scale (PARQ/Control) (Rohner, 1999)

because it is a valid and reliable measure of constructs closely related to Gottman's meta-

emotion theory. This instrument is "designed to measure individuals' perceptions of

parental acceptance and rejection (i.e., warmth) defining one end of the continuum and

parental rejection defining the other end" (Rohner, 1999, p. 17). The control scale,

recently added to the PARQ, measures the individual's perceptions of parental control

(permissiveness-strictness) and the extent to which parents insist on compliance with

their requests or demands. It does not imply anything about the manner in which parents

enforce their demands. Most of the items in the PARQ questionnaire refer to behavior

rather than attitudes, with the goal of avoiding the problem sometimes found in research

measurement, that is, "when one has to demonstrate that a link exists between the

attitudes expressed by a person and behavior engaged in by the person" (Rohner, 1999, p.

17). All versions of the PARQ consist of four scales: (a) perceived parental

warmth/affection, (b) perceived parental aggression/hostility, (c) perceived parental

neglect/indifference, and (d) perceived parental rejection (undifferentiated) (Rohner,

1999). All scales, including the control scale, were utilized in this study.

Rohner (1999) achieved exterior (criterion) concurrent validity by comparing the

scales to two already validated instruments, Schaefer's (1964) Child's Report of parent

Behavior Inventory (CRPBI) and Bronfenbrenner's Parental Behavior Questionnaire

(BPB), and showed that PARQ scales were significantly related to their respective

validation scales. Cronbach's coefficient alpha, used to measure reliability, ranged from

.86 to .95. Rohner (1999) stated that "the measures of internal consistency (reliability)

and of concurrent, convergent, and discriminant validity all bear on the construct validity

of the theoretical constructs underlying the PARQ scales" (p. 29). These results were

confirmed by factor analysis as well.

The items are answered using a 4-point Likert-style scale, with Almost Always

True scored as 4, Sometimes True as 3, Rarely True as 2, and Almost Never True as 1.

Some items in the various scales are reversed scored.

Assessment of Parental Style (APS)

To measure the effect of the emotion-coaching part of the experimental treatment,

Gottman's (1997) Assessment of Parental Style (APS) was chosen. The APS is a self-

report questionnaire that measures attitudes and beliefs about parents' and children's

feelings, children's expressions of feelings and behaviors, as well as self-reported

parental behaviors. The APS is based on meta-emotion philosophy, which Gottman

extensively researched as reported in Chapter 2 of this dissertation. The APS contains

four subscale scores--dismissing style, disapproving style, laissez-faire style, and

emotion-coaching style. Although Gottman has not validated these scales yet, one of his

colleagues, L. F. Katz, gave this researcher permission to use them in the present study

(personal communication, September 27, 2000).

Data Analysis

Analysis of covariance (ANCOVA) equations assessed the effect of the

independent variable (the intervention) on the dependent variables (RHI peer and

community scales, BEES, PARQ/Control, and APS) after controlling for pretest scores.

The significance level was established at p = < .05 for all hypothesis testing.


The following hypotheses were predicted for participants after receiving the

experimental treatment, Connecting and Coaching:

H1: There will be no difference in postintervention relational health, as measured

by the Relational Health Indices, between the control group and the experimental group.

H2: There will be no difference in postintervention empathic tendencies, as

measured by the Balanced Emotional Empathy Scale, between the control group and the

experimental group.

H3: There will be no difference in postintervention parental behaviors, as

measured by the Parental Acceptance-Rejection/Control Questionnaire, between the

control group and the experimental group.


H4: There will be no difference in postintervention parenting style, as measured

by the Assessment of Parenting Style, between the control group and the experimental




The purpose of this study was to investigate the effectiveness of a psycho-

educational intervention that taught relational skills and parental emotion coaching to

incarcerated mothers in a work-release center. In this chapter, there is a brief

introduction; a statistical analysis of the demographic variables; and a postintervention

analysis of covariance of all measures and the experimental treatment, with tables,

additional comments, and a summary.

Four hypotheses about the effects of the parenting intervention, Connecting and

Coaching, on the peer and community relational health, empathic tendencies, parental

behaviors, and parenting styles of participants were developed and tested:

Hi: There will be no difference in postintervention relational health, as measured

by the Relational Health Indices, between the control group and the experimental group.

H2: There will be no difference in postintervention empathic tendencies, as

measured by the Balanced Emotional Empathy Scale, between the control group and the

experimental group.

H3: There will be no difference in postintervention parental behaviors, as

measured by the Parental Acceptance-Rejection/Control Questionnaire, between the

control group and the experimental group.

H4: There will be no difference in postintervention parenting style, as measured

by the Assessment of Parenting Style, between the control group and the experimental


The participants in this study were 32 incarcerated mothers who were serving

their sentences in residential work-release centers in North Florida, in either Jacksonville

or Tallahassee. Four participants in the experimental group discontinued, one exercising

her right not to participate and the other three being sent back to prison, lowering the total

number of participants to 28. The experimental group consisted of 15 participants, and

the control group consisted of 13 participants. The mothers at the Jacksonville location

received the intervention, Connecting and Coaching, a relational parenting class, because

of convenience of the location for the researcher and instructor. The mothers in the

Tallahassee facility were the control group. All measures were administered to both

control and treatment groups. The control group received delayed treatment after the 8-

week intervention was completed. The participants in the experimental group attended a

relational parenting class entitled, Connecting and Coaching, for eight sessions. Prior to

instruction, all of the participants were pretested on relational health, empathic

tendencies, parental behaviors, and parenting style. All of the participants were again

tested in these four areas after the intervention was completed.

Various statistical tests were performed to determine the distribution of scores,

equivalence of the groups, and the effectiveness of the intervention. Specifically, an

ANCOVA was conducted on each of the posttest scores of the two groups. More detailed

information on the tests conducted and the results of these tests along with various

follow-up tests are provided within this chapter. A thorough discussion of the conclusions

and implication that can be drawn from these results is provided in Chapter 5.

Statistical Analysis

Initial Group Equivalence and Bias

Equivalence of control and experimental groups was tested using t-tests and

Levene's Test for Equality of Variances for the continuous measures of age, number of

marriages, number of brothers, number of sisters, if parents divorced, participant's age

when parents divorced, number of girl children, number of boy children, age at birth of

first child, age at first arrest, number of previous incarcerations, total sentence length in

months, and number of days until release. The means, standard deviation, and significant

differences of continuous variables as measured by Levene's Test for Equality of

Variances for the experimental and control groups are found in Table 1. Most continuous

variables were not significantly different between groups, with the exception of number

of sisters, number of months before release, and number of days until release. The

experimental group had a mean score of 3.39 sisters, whereas the control group had a

mean score of 1.5 sisters. The experimental group had a mean 29.08 months for their total

sentence, whereas the control group had 42.33 months total. The experimental group had

a mean of 88.38 days before release, and the control group had 174.69 days before

release. These differences were not considered to be of major importance in the context

of this study; therefore, equality of the two groups as measured by continuous variables

was assumed.

Chi-square tests for the categorical measures of race or ethnic background, marital

status, education level, language spoken as a child, language spoken to children, U.S.

Table 1

Description and Significant Differences Between Treatment and Control Groups on
Continuous Variables

Treatment/Control N Mean SD Levene's Test
p value
Age Treatment 14 33.71 7.25 .98

# of Marriages

# of Brothers

# of Sisters

# of Girl Children

# of Boy Children

Age at Birth of
First Child

Age at First Arrest

# of Previous

# of Months until

# of Days until











































citizenship, occupation before incarceration, alcohol and drug use before incarceration,

type of alcohol or drugs used, treatment for alcohol or drugs, arrested for alcohol or drug

abuse, domestic violence history, parents marital status, parents' use of alcohol and

drugs, history of child abuse, age ranges of participants' children, any adopted or foster

children, whether currently pregnant, children's current residence, frequency of visits by

children while incarcerated, was this the first arrest, was this the first incarceration, other

family members arrested, other family members incarcerated, and anticipated parental

role after incarceration. There were no significant differences between the treatment

group and the control group as measured by chi-square analysis of the categorical

variables, with the exception of whether the husband or boyfriend had been the abuser in

domestic violence. The treatment group stated that five boyfriends and one husband had

been the abusers; whereas, in the control group, two boyfriends and seven husbands had

been abusive. Thus, equal variances were assumed on all categorical data of the

experimental and control groups. Hence, any differences found between groups should be

attributed to the instruction intervention. Table 2 lists numbers of respondents in the

treatment and control groups for each variable and chi-square p-value scores.

Postintervention Analysis

To determine the effectiveness of the explicit instruction intervention, a series of

ANCOVA procedures was performed. This section provides an explanation of the results

from the tests of hypotheses related to each of the four constraints. Summary tables

reporting the detailed analyses of variance for each scale of the four measures are found

in Appendix E.

Table 2

Description and Significant Differences Between Treatment and Control Groups on
Categorical Variables

Categories Treatment Control Chi-square Sig.

Race .38
White 7 9
Black 7 4
Hispanic 1 0

Marital Status .47
Single 6 3
Married 2 0
Separated 1 1
Divorced 3 6
Widowed 1 2
Never Married 2 1

Education Level .41
High School
GED 3 6
College 3 3
7th Grade 2 0
8th Grade 0 1
9th Grade 1 1
10th Grade 1 1
1 lth Grade 3 0
2 1

Language Spoken Growing Up .34
English 14 1
Non-English 13 0

Language Spoken with Children .34
English 14 1
Non-English 13 0

U.S. Citizen 1.00

Yes 15 13
No 0 0
Legal alien 0 0