 

 

THEmC
Ll

ﬂra’Q
xiv
V

This is to certify that the

dissertation entitled

Coordinating Councils as Vehicles for Achieving

a Coordinated Community Response to Domestic

Violence: An Examination of the Correlates of
Council Effectiveness

presented by
Nicole Elizabeth Allen

has been accepted towards fulﬁllment
of the requirements for

Ph.D. Psychology
degree in

 

 

/ /l:’lagr professfr

May 10, 2001
I)ate

 

MSU Lt an Afﬁrmative Action/Equal Opportunity Institution 0-12771

 

 

LIBRARY
Michlgan State
University

 

 

 

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAﬁE RECALLED with earlier due date if requested.

DATE DATE DUE DATE DUE

MAiZloO‘EZU’W

i V '
I
ll .t K U ; in;

 

6/01 cJCIRC/DateDuolpst'rp. 15

COORDINATING COUNCILS As VEHICLES FOR ACHIEVING A COORDINATED
COMMUNITY RESPONSE To DOMESTIC VIOLENCE: AN EXAMINATION OF
THE CORRELATES OF COUNCIL EFFECTIVENESS
By

Nicole Elizabeth Allen

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Psychology

2001

ABSTRACT
COORDINATING COUNCILS AS VEHICLES FOR ACHIEVING A COORDINATED
COMMUNITY RESPONSE To DOMESTIC VIOLENCE: AN EXAMINATION OF
THE CORRELATES OF COUNCIL EFFECTIVENESS
By

Nicole Elizabeth Allen

It is widely recognized that intimate partner violence against women is a pervasive
social issue. Approximately two million women are physically assaulted annually in the
United States (Tjaden & Thoennes, 1998). Given the need to involve multiple service
systems (e. g., criminal justice, human service, educational) in the response to domestic
violence (Clark, Burt, Schulte, & Maguire, 1996), communities are increasingly focused
on creating a ‘coordinated community response’ (CCR). This is critical because these
systems Often have diverse goals and sometimes competing priorities. The goal of a CCR
is to encourage collaboration across systems, and to implement new policies and practices
that ultimately increase women’s safety and batterer accountability. While efforts to
develop a CCR are underway in many communities across the United States, little is
known empirically about what contributes to the effectiveness of these efforts. One
common vehicle for encouraging a CCR is the use of coordinating councils. These
councils are collaborative bodies that include key stakeholders in the community response
to domestic violence, and that meet regularly to address the unique issues of responding to
domestic violence in a given community. Coordinating councils in general face barriers to

achieving collaboration across multiple stakeholders; the collaborative process of domestic

violence coordinating councils is likely to include such challenges. Furthermore, domestic
violence coordinating councils must deal effectively with a history of conﬂict and inequity
that has characterized the ﬁght for social justice within the domestic violence movement.
Given this context, it is critical to attend to the factors that facilitate or impede the
effectiveness of such councils. The purpose of this study was to: I) examine the
effectiveness of domestic violence coordinating councils in one state and 2) investigate
correlates that positively contribute to council effectiveness. Correlates examined
included aspects of the council climate (e. g., Shared power in decision-making; quality
leadership) and council characteristics (e.g., degree of formality of council Structure).
Survey data were collected from 511 coordinating council members across 43 councils.
Given that this study included multiple councils as well as numerous individuals within
each council, special attention was paid to multiple levels of analysis. This study found
that both council climate and council characteristics were important correlates of council
effectiveness. Speciﬁcally, a council climate characterized by effective conﬂict resolution,
shared power in decision-making, a shared mission and effective leadership was associated
with greater council effectiveness. However, elements of council climate were
differentially related to indicators of council effectiveness, indicating that one element may
be more salient than another regarding different aspects of the councils’ collective work.
In addition, the degree to which councils received support from their communities, ﬁInders
and local policy makers and the degree to which they fostered active participation from a
diverse set of stakeholders also inﬂuenced their effectiveness. The implications of these

ﬁndings are discussed.

Copyright by
NICOLE ELIZABETH ALLEN

2001

In loving memory of
Georgia Belosic
Woman Warrior
February 3, 1922

February 11, 2000

ACKNOWLEDGMENTS

First and foremost, I thank all of those individuals working diligently to respond to
Violence against women in their communities and all of those who took time out of their
busy schedules to participate in this study. I thank Leslie Hagen for working
collaboratively with me to design and implement this study and for her continued
willingness to work in partnership with me. I also thank Debi Cain, Janet Strahan and
Veronica Zoltowski for their thoughtﬁIl comments on the instruments used in this study. I
feel deep appreciation for all of my committee members. I thank Pennie F oster-Fishman,
my committee chair, for her thoughtﬁIl comments, and commitment of time and energy to
help me reﬁne my work. I cannot adequately summarize how I have grown as a scholar
with Pennie’s guidance. I could not have completed this project without the support and
insight of Cris Sullivan who introduced me to the movement to end violence against
women. Cris’ passion and clarity has provided me with a model for affecting social
change via research and action and I aspire to contribute as she has. I thank Deborah
Salem for her conceptual clarity, steadfast support, honest assessment of my work and for
being a model for how to create balance in academe. Finally, I thank Deborah Bybee, the
statistical goddess, for increasing my understanding of the nuances Of statistical
phenomenon and giving of her time so freely. I also thank Deb for her gentle spirit,
always making me feel conﬁdent in my abilities even when correcting something I was

doing incorrectly! I thank all of the Ecological Community graduate students for their

vi

support and inspiration. In particular, I thank my cohort — traveling through graduate
school with each of them has enriched my experience. I thank Angie Wolf for her
friendship, support, and many welcome breaks, Pamela Martin for her support and
prayers, especially as we jumped hurdles to reach the ﬁnish line, David Loveland for our
statistical exchanges and his humor and Christina Ayala-Alcantar for supporting me all the
way from California with her love, friendship and laughter. I thank Ruth Fleury for many
welcomed breaks during my data collection and Elaine Shpungin and Mikhail Lyubansky
for a versatile friendship including wonderful conversations and thoughtful feedback on
my written work. I thank the committed undergraduate Students who assisted me with
data collection and entry — Clary Goedert-Gasper, Shelley Leiphart, Shannon Mullally,
Paul Perez, Cheryl ReﬁIerzo and Katie Gregory . I also thank Sheila Balmer for always
having kind words. These acknowledgments would not be complete without Special
thanks to my family and friends for their limitless love and support, especially Nicholas
and Margo Allen, Jon Allen, Jaimee Allen, Elizabeth Allen, John Belosic, Faith Menard,
and Arthur Larrabee. Their constant belief in me translates into my belief in myself I
thank my close ﬁ'iends Christine and Marc Marin and Nicole Middleton Keller for
maintaining our friendship across distance and time. I also express my appreciation for my
friends, teachers and classmates from the Great Lakes Pathwork community who have
provided me with much love, guidance and support on the journey. Above all, I thank my
partner, Christopher Menard who anchors me to the larger truth, picks up the pieces as I

go, provides me with many needed diversions from my work and loves me as I am..

vii

TABLE OF CONTENTS

List Of Tables ........................................................ xv
Overview ............................................................ I
A Coordinated Community Response ................................. 2
Examining the Role of Coordinating Councils .......................... 3
Investigating Correlates of Council Effectiveness ........................ 6
Council Climate ........................................... 7

Council Characteristics ..................................... 10

Current Study ................................................. ll
Introduction ........................................................ I 3

History of the Domestic Violence Movement: Encouraging Community

Accountability ................................................. 13
Coordinated Community Response Efforts ............................ 15
Interagency Coordinating Councils .................................. l8
Examining Domestic Violence Coordinating Council Effectiveness .......... 20
Council Goals ............................................ 20
Council Members’ Perceptions of Council Effectiveness ............ 23
Investigating Correlates of Council Effectiveness ....................... 25
Council Climate .......................................... 26

Elements of the Council Climate as Correlates Of Council
Effectiveness ....................................... 26

viii

Conﬂict Transformation ......................... 30

Power/Inﬂuence in Decision-Making ............... 35

Shared Mission ............................... 41

Council Leadership ............................ 42

Council Characteristics and Council Effectiveness ................. 44
Breadth of ‘Ofﬁcial’ and ‘Active’ Membership ............. 45

Presence of External Support .......................... 47

Council Structure ................................... 48

Council ‘Age’ ..................................... 49

Exploring the Multi-Level Nature Of Perceptual Data .................... 50
Current Study ................................................. 54
Method ............................................................ 59
Sample ....................................................... 59
Domestic Violence Coordinating Councils ...................... 59
Council Members ......................................... 61
Council Leaders and/or Conveners ............................ 62
Procedures .................................................... 63
Recruitment of Councils and Key Inforrnants .................... 63
Recruitment of Council Members ............................. 65

Survey Collection ......................................... 65

Key Informant Interviews ................................... 65

ix

Follow-up ............................................... 65
Measures ..................................................... 67
Council Member Survey .................................... 67

Council Member Demographic and Descriptive Information . . . 67

Council Effectiveness. ................................ 68

Conﬂict Transformation .............................. 69
Power\Inﬂuence in Decision-Making ..................... 70

Council Leadership .................................. 7]

Presence of a Shared Mission .......................... 72

Aggregating Perceptions at the Council Level .............. 73

Key Informant Interview .................................... 75
Council Demographic and Descriptive Information .......... 75

Assessing Council Goals and Accomplishments ............. 75

Breadth of Council Membership ........................ 76

Presence of External Support .......................... 77

Council Structure ................................... 77

Data Analysis Strategy ........................................... 78
Results ............................................................ 80
Univariate Analysis ............................................. 8O
Descriptives ............................................. 80
Conﬂict Resolution .................................. 80

Shared Power in Decision-Making ...................... 84

Leadership ........................................ 84

Shared Mission ..................................... 85

Breadth of ‘Oﬂicial’ and ‘Active’ Membership ............. 85

Support for Council Activities .......................... 87

Forrnality of Council Structure ......................... 87

Council Goals ...................................... 88
Accomplishment of Goals ............................. 89

Perceived Effectiveness ............................... 89

Validating Indicators of Council Effectiveness ......................... 90
Intercorrelations Among Independent and Dependent Variables ............ 93
Modeling Perceived Council Effectiveness ............................ 97
Estimating the Unconditional Model ........................... 98
Estimating Models for Each Independent Variable ................ 99
Conﬂict Resolution ................................. 101

Shared Power in Decision-Making ..................... 101

Shared Mission .................................... 103

Leadership ....................................... 103

Estimating Combined Climate Model ......................... 105
Adding Council Characteristics to the Final Climate Model ......... 108
Fonnality of Council Structure ........................ 108

xi

Breadth of ‘Ofﬁcial’ and ‘Active’ Membership ............ 108

External Support for Council Activities .................. 111

Estimating a Final Climate/Characteristics Model ................ 1 11
Modeling the Breadth of Council Goals ............................. 1 14
Estimating Models for Each Independent Variable ............... 114
Conﬂict Resolution ................................. 1 14

Shared Power in Decision-Making ..................... 1 16

Shared Mission .................................... 1 17

Leadership ....................................... 1 19

Estimating a Combined Climate Model ........................ 121
Adding Council Characteristics to the Model ................... 126
Breadth of ‘Ofﬁcial’ and ‘Active’ Membership ............ 126

External Support for Council Activities .................. 130

Formality of Council Structure ........................ 132

Estimating a Combined Climate/Characteristics Model ............ 133
Modeling the Accomplishment of Goals ............................. 134
Estimating Models for Each Independent Variable ............... 134
Conﬂict Resolution ................................. 137

Shared Power in Decision-Making ..................... 137

Shared Mission .................................... 137

Leadership ....................................... I 4 1

xii

Estimating a Combined Climate Model ........................ 141

Adding Council Characteristics to the Model ................... 154
Breadth of ‘Ofﬁcial’ and ‘Active’ Membership ............ 154
External Support for Council Activities .................. 154
F orrnality of Council Structure ........................ 157

Estimating a Combined Climate/Characteristics Model ............ 157

Examining Variability in Perceptual Variables as a Function of

Organizational Role ............................................ 1 71
Power Analysis ............................................... 175
Discussion ......................................................... 1 77
Council Effectiveness ........................................... 17 8
Elements of the Council Climate as Correlates of Council Effectiveness ..... 182
Conﬂict Resolution ....................................... 183

Shared Power in Decision-Making ........................... I90

Shared Mission .......................................... 195
Leadership ............................................. 1 96
Council Characteristics as Correlates of Council Effectiveness ............ 201
Breadth of ‘Ofﬁcial’ and ‘Active’ Membership .................. 201
Presence of External Support ........ ' ....................... 203
Formality of Council Structure .............................. 205
Summary of the Correlates of Indicators of Council Effectiveness ......... 206
Revisiting Levels of Analysis and Organizational Role .................. 209

xiii

Implications for Future Research .................................. 214

Limitations ................................................... 2 l 6
Conclusions ........................................................ 219
Appendices

Appendix A .................................................. 224

Appendix B .................................................. 237

Appendix C .................................................. 261

Appendix D .................................................. 264

Appendix E .................................................. 267

Appendix F .................................................. 269

Appendix G .................................................. 271

Appendix H .................................................. 273
References ......................................................... 275

xiv

Table 1:

Table 2:

Table 3:

Table 4:

Table 5:

Table 6:

Table 7:

Table 8:

Table 9:

Table 10:

Table 1 1:

Table 12:

Table 13:

Table 14:

Table 15:

Table 16:

LIST OF TABLES

Intraclass Correlations for Dependent and Independent Variables ........ 74
Descriptive Statistics for Dependent and Independent Variables at Level I . 81
Descriptive Statistics for Independent Variables at Level II ............. 82
Descriptive Statistics for Dependent Variables at Level II .............. 83

Percent of Councils with Ofﬁcial and Active Membership from Each

Stakeholder Group ........................................... 86
Intercorrelation Matrix Of Indicators of Council Effectiveness .......... 91
Intercorrelation Matrix of Independent Variables at Level I ............ 94
Intercorrelation Matrix of Independent Variables at Level II ............ 95
Correlations of Independent and Dependent Variables at Level II ........ 96
Correlations of Independent and Dependent Variables at Level I ........ 97

Hierarchical Linear Modeling of Perceived Effectiveness Regressed onto
Council Climate Variables Individually ........................... 102

Hierarchical Linear Modeling of Perceived Effectiveness Regressed onto
Council Climate Variables Individually (continued) .................. 104

Hierarchical Linear Modeling of Perceived Effectiveness Regressed onto
Council Climate Variables Combined ............................ 107

Hierarchical Linear Modeling of Perceived Effectiveness Regressed onto
Council Climate and Characteristic Variables ...................... 109

Hierarchical Linear Modeling of Perceived Effectiveness Regressed onto
Council Climate and Characteristic Variables (continued) ............. 112

Hierarchical Linear Modeling of Perceived Effectiveness Regressed onto
Council Climate and Characteristic Variables: Final Models ........... 113

XV

Table 17:

Table 18:

Table 19:

Table 20:

Table 21:

Table 22:

Table 23:

Table 24:

Table 25:

Table 26:

Table 27:

Table 28:

Table 29:

Table 30:

Breadth of Criminal Justice System Goals Regressed onto Effective Conﬂict
Resolution Strategies (N = 40) ................................. 115

Breadth of Goals Beyond the Criminal Justice System Regressed onto
Effective Conﬂict Resolution Strategies (N = 40) ................... 115

Breadth of Criminal Justice System Goals Regressed onto Shared Power in
Decision-Making (N = 40) .................................... 116

Breadth of Goals Beyond the Criminal Justice System Regressed onto Shared
Power in Decision Making (N = 40) ............................. 117

Breadth of Criminal Justice System Goals Regressed onto Shared Mission
(N = 40) .................................................. 118

Breadth of Goals Beyond the Criminal Justice System Regressed onto Shared
Mission (N = 40) ........................................... 118

Breadth of Criminal Justice System Goals Regressed onto Inclusive Leadership
(N = 40) .................................................. 119

Breadth of Goals Beyond the Criminal Justice System Regressed onto
Inclusive Leadership (N = 40) ................................. 120

Breadth of Criminal Justice System Goals Regressed onto the Instrumental
Leadership (N = 40) ......................................... 120

Breadth of Goals Beyond the Criminal Justice System Regressed onto
Instrumental Leadership (N = 40) ............................... 121

Breadth of Criminal Justice System Goals Regressed onto Council Climate
Variables (N = 40) .......................................... 123

Breadth of Criminal Justice System Goals Regressed onto Council Climate
Variables: Reduced Model (N = 40) ............................. 124

Breadth of Criminal Justice System Goals Regressed onto Council Climate
Variables: Second Reduced Model (N = 40) ....................... 125

Breadth of Goals Beyond the Criminal Justice System Regressed onto Council
Climate Variables (N = 40) .................................... 127

xvi

Table 31:

Table 32:

Table 33:

Table 34:

Table 35:

Table 36:

Table 37:

Table 38:

Table 39:

Table 40:

Table 41:

Table 42:

Table 43:

Table 44:

Breadth of Goals Beyond the Criminal Justice System Regressed onto Council
Climate Variables: Reduced Model Including Leadership Variables Only
(N = 40) .................................................. 128

Breadth of Criminal Justice Goals Regressed onto the Breadth of Official
Membership (N = 40) ........................................ 128

Breadth of Goals Beyond the Criminal Justice System Regressed onto the
Breadth of Official Membership (N = 40) ......................... 129

Breadth of Criminal Justice System Goals Regressed onto the Breadth Of
Active Membership (N = 40) ......................... . ......... 129

Breadth of Goals Beyond the Criminal Justice System Regressed onto Breadth
of Active Membership (N = 40) ................................ 130

Breadth Of Criminal Justice System Goals Regressed onto External Support for
Council Activities (N = 40) .................................... 131

Breadth of Goals Beyond the Criminal Justice System Regressed onto External
Support for Council Activities (N = 40) .......................... 131

Breadth of Criminal Justice Goals Regressed onto Formality of Council
Structure (N = 40) .......................................... 132

Breadth of Goals Beyond the Criminal Justice System Regressed onto
F orrnality Of Council Structure (N = 40) .......................... 133

Breadth of Criminal Justice Goals Regressed onto Council Climate and
Characteristic Variables (N = 40) ............................... 135

Breadth of Goals Beyond the Criminal Justice System Regressed onto Council
Climate and Characteristic Variables (N = 40) ..................... 136

Degree to Which Criminal Justice Goals Accomplished Regressed onto
Effective Conﬂict Resolution Strategies (N = 40) ................... 138

Degree to Which Goals Beyond the Criminal Justice System Accomplished
Regressed onto Effective Conﬂict Resolution Strategies (N = 40) ....... 138

Degree to Which Criminal Justice Goals Accomplished Regressed onto Shared
Power in Decision-Making (N = 40) ............................. 139

xvii

Table 45:

Table 46:

Table 47:

Table 48:

Table 49:

Table 50:

Table 51:

Tables
52a - (1:

Tables
53a - d:

Table 54:

Table 55:

Table 56:

Table 57:

Degree to Which Goals Beyond the Criminal Justice System Accomplished
Regressed onto Shared Power in Decision Making (N = 40) ........... 139

Degree to Which Criminal Justice Goals Accomplished Regressed onto Shared
Mission (N = 40) ........................................... 140

Degree to Which Goals Beyond the Criminal Justice System Accomplished
Regressed onto Shared Mission (N = 40) ......................... 140

Degree to Which Criminal Justice Goals Accomplished Regressed onto
Inclusive Leadership (N = 40) ................................. 142

Degree to Which Goals Beyond the Criminal Justice System Accomplished
Regressed onto Inclusive Leadership (N = 40) ..................... 142

Degree to Which Criminal Justice Goals Accomplished Regressed onto the
Instrumental Leadership (N = 40) ............................... 143

Degree to Which Goals Beyond the Criminal Justice System Accomplished

Regressed onto Instrumental Leadership (N = 40) .................. 143

Hierarchical Regression: Degree to Which Criminal Justice Goals
Accomplished Regressed onto Council Climate Variables (N = 40) ...... 145

Hierarchical Regression: Degree to Which Goals Beyond the Criminal Justice
System Accomplished Regressed onto Council Climate Variables (N = 40) 150

Degree to Which Criminal Justice Goals Accomplished Regressed onto the
Breadth of Official Membership (N = 40) ......................... 155

Degree to Which Goals Beyond the Criminal Justice System Accomplished
Regressed onto the Breadth of Oﬂicial Membership (N = 40) .......... 155

Degree to Which Criminal Justice Goals Accomplished Regressed onto the
Breadth of Active Membership (N = 40) .......................... 156

Degree to Which Goals Beyond the Criminal Justice System Accomplished
Regressed onto Breadth of Active Membership (N = 40) ............. 156

xviii

Table 58:

Table 59:

Table 60:

Table 61:

Tables
62a-e:

Tables
63a - (I:

Table 64:

Table 65:

Degree to Which Criminal Justice Goals Accomplished Regressed onto
External Support for Council Activities (N = 40) ................... 158

Degree to Which Goals Beyond the Criminal Justice System Accomplished
Regressed onto External Support for Council Activities .............. 158

Criminal Justice Goals Accomplished Regressed onto F ormality of Council
Structure (N = 40) .......................................... 159

Goals Beyond the Criminal Justice System Accomplished Regressed onto
Formality of Council Structure (N = 40) .......................... 159

Hierarchical Regression: Degree to Which Criminal Justice Goals
Accomplished Regressed onto Council Climate and Characteristic Variables
(N = 40) .................................................. 161

Hierarchical Regression: Degree to Which Goals Beyond the Criminal Justice
System Accomplished Regressed onto Council Climate and Characteristic
Variables (N = 40) .......................................... 167

Perceived Effectiveness Regressed and Climate Variables Regressed onto
Organizational Role ......................................... 172

Perceived Effectiveness Regressed and Climate Variables Regressed onto
Organizational Role (Continued) ............................... 174

xix

OVERVIEW

Intimate violence against women1 is now commonly recognized as a pervasive
social problem. While this may seem self-evident, it represents a substantial shift from the
widespread perception of domestic violence as a ‘private matter’ common only two
decades ago and still lingering today. A recent research brief produced by the National
Institute of Justice and the Centers for Disease Control and Prevention estimates that 1.9
million women are physically assaulted annually in the United States (Tjaden & Thoennes,
1998). The prevalence of intimate violence against women coupled with evidence
documenting the detrimental psychological (e. g., depression and anxiety) and physical
health consequences of such abuse (e. g, Cascardi & O’Leary, 1992; Campbell, Sullivan, &
Davidson, 1995; Eby, Campbell, Sullivan, Davidson, 1995; Goodman, Koss, & Russo,
1993; Haber & Roos, 1985; Koss etal., 1991; Sutherland, Bybee, & Sullivan, 1998)
emphasizes the need to respond to this phenomenon. While it is critical that we
understand the experiences of women with abusive partners and the deleterious
psychological and physical consequences of abuse, we must also broaden the scope Of our
inquiry to focus on the current efforts of communities to develop effective responses to

end intimate violence against women. This focus is a critical component of understanding

 

1

The phrases ‘intimate violence against women,’ ‘domestic violence,’ and ‘intimate partner
violence’ will be used interchangeably to describe intimate partners’ systematic use of
psychological and physical abuse to exert power and control.

1

not only why intimate violence against women persists, but what actions can be taken to
end this violence.
A Coordinated Community Response

Historically, the community response to intimate violence against women has been
characterized by inadequate services (Gondolf, 1988; Sullivan, 1991), and a lack of
coordination across the systems involved in responding to domestic violence cases (Hart,
1995; McEvoy, Brookings, & Brown, 1983). In an effort to address these shortcomings,
current community response efforts have centered on what is commonly termed a
‘coordinated community response’ (CCR) to intimate violence against women. A
‘coordinated community response’ refers to community-wide efforts to bring together
relevant stakeholders to respond to intimate violence against women in a comprehensive
manner by instituting policies and procedures that increase women’s safety and batterer
accountability (Pence & Shepard, 1999). While many CCRS initially focused on reforming
only the criminal justice system (e. g., police, prosecutors, judges, probation officers), there
has been an increasing recognition that stakeholders in arenas outside of the criminal
justice system must be included in efforts to end violence against women (Clark et al.,
1996). This is important given that responding to intimate violence against women
requires the mobilization of a variety of community systems including the health care,
criminal justice, educational, and social services systems, all levels of government,
religious organizations, and businesses (Clark et al., 1996). To date, there have been few

examinations of the efﬁcacy of ‘coordinated community responses’ of this nature, but

early evaluations and anecdotal evidence suggests that where these efforts exist, greater
Strides are being made toward addressing the complex issues that arise when responding
to violence against women (Buel, 1997; Clark et al., 1996; Edleson, 1991; Hart, 1995).

Given the promise of CCR’s to improve the community response to violence
against women (Pence & Shepard, 1999), gaining a better understanding of what vehicles
facilitate such a response is critical. One such vehicle is the domestic violence
coordinating council. Although coordinating councils have become increasingly
widespread, we know little about their effectiveness in facilitating a CCR. While there is
preliminary evidence to suggest coordinating councils can play a positive role in the
development of a CCR (Buel et al., 1997; Clark et al., 1996), there is also competing
anecdotal evidence suggesting that coordinating councils are not the most effective way to
achieve a CCR (Gamache & Asmus, 1999). This study is one of few systematic
examinations of the effectiveness of domestic violence coordinating councils.

Examining the Role of Coordinating Councils

Broadly speaking, coordinating councils2 are collaborative bodies that include key
stakeholders (i.e., individuals/organizations that are affected by or inﬂuence a given issue)
who meet regularly to respond to a given issue in a collaborative manner. While there is

relatively little research on the application of coordinating councils to domestic violence

 

2

The term coordinating council will be used in this Study to refer to any interdisciplinary
collaborative body responding to domestic violence (e. g., domestic violence task force or
coordinating committee).

intervention, there is some evidence that this vehicle may lead to increased coordination
among agencies responding to domestic violence (Clark et al., 1996). In a study done in
San Francisco, the presence of a coordinating council was found to facilitate interactions
between agencies, promote broader institutional change and increase the responsiveness of
the service system to the needs of battered women (Clark et al., 1996). These ﬁndings
provide early evidence of the usefulness of these councils. However, there is still a great
deal to be uncovered about the degree to which councils are effective in encouraging a
CCR to domestic violence in their communities. A goal of this study was to examine the
capacity of coordinating councils to engage in collective work and affect change in their
communities.

While investigating direct indicators of community change (e. g., increased
community awareness or access to resources) was beyond the scope of this study, this
study takes a ﬁrst step in understanding the effectiveness of domestic violence
coordinating councils. Council effectiveness as operationalized in this study refers to each
council’s capacity to engage in collective work to address their goals and the goals of a
CCR. This iS important given that councils engaging in collective work and addressing
issues relevant to multiple systems are more likely to achieve broad systems change.

Thus, the ﬁrst purpose of this study was to examine multiple indicators of council
effectiveness: a) the breadth of goals that councils focused on to stimulate a CCR (i.e., the
number of areas of reform they have addressed), b) the degree to which councils have

accomplished their goals in these areas according to council leaders, and c) the degree to

which council members perceive their council as effective in meeting its goals and
stimulating needed reforms of policy and practice that encourage batterer accountability
and women’s safety.

This study utilized two sources of information to gauge councils’ effectiveness in
encouraging a CCR to domestic violence. First, information was gathered from domestic
violence coordinating council leaders and/or coordinators in one state regarding a) the
goals they were setting and b) the degree to which they were accomplishing these goals.
These data provide information regarding the breadth of councils’ collective work (i.e.,
the total number of goals councils addressed3 in the criminal justice system and in the
human service delivery system) as well as their effectiveness in achieving these goals,
according to a single key informant.

Second, council members’ perceptions of the effectiveness of councils (i.e., the
degree to members view their councils as effective in meeting intermediate goals such as
improving relationships among stakeholders and the goals of a CCR, such as reforming
policies and practices) were assessed. This second proxy for council effectiveness is
important for a number of reasons. First, there is a long tradition of utilizing the
perceptions of multiple setting members to understand the character of a setting (e. g.,

Forehand & Gilmer, 1962). In fact, other studies of collaborative processes and outcomes

 

3

For the purposes of this study ‘addressing’ an area of reform is equivalent to simply
identifying it as a formal or informal goal of the council; ‘accomplishing’ or ‘achieving’ a
goal or an area of reform refers to the degree to which it was successfiilly completed.

5

have used perceived effectiveness as a proxy for the effectiveness of collaborative settings
(e. g., Gottlieb, Brink, & Gingiss, 1993; Kumpfer, Turner, Hopkins & Librett, 1993).
Second, examining the breadth of council goals and the degree to which these goals are
accomplished provides only a partial understanding of council effectiveness. Council
members’ perceptions provide a valuable source of insight into council functioning that
extends beyond the number of goals they are addressing. For example, councils could
address a broad range of goals, but may be relatively ineffective in achieving change in any
area. Simply attending to the breadth of goals would obscure councils’ relative
effectiveness in meeting their goals. In addition, gathering data from multiple informants
creates the Opportunity to validate key informants’ impressions of the degree to which
councils are accomplishing their goals. This is important because there is some evidence
that leaders may evaluate a setting differently than other setting members (Foster-Fishman
& Keys, 1997; Martin, 1992). Finally, the degree to which council members perceive
councils as effective is related to their participation and commitment to the collaborative
endeavor (Skaff, 1988). Understanding those factors that contribute to council members’
perceptions of the effectiveness of their councils provides valuable information about how
to encourage continued participation in council activities.
Investigating Correlates of Council Effectiveness

Examining collaborative bodies requires a focus not only on the degree to which

they are effective, but also on what contributes to their effectiveness. While the

collaboration literature has identiﬁed numerous facilitators and barriers to collaboration

(e.g., Butterfoss, Goodman, & Wandersman, 1993; Foster-Fishman, et al., 2001; Roussos
& Fawcett, 2000), efforts to build models of effective collaboration and to examine trends
across collaborative settings have been limited by the relatively small number of studies
that examine more than a single council. Thus, the second purpose of this study was to
examine multiple settings in an effort to understand more about what aspects of the
collaborative setting inﬂuence council effectiveness. This study focused on: a) the council
climate for collaboration (e. g., the degree to which power is Shared in decision-making)
and b) characteristics of the council setting (e. g., the degree of formality in the council
stnIcture) as correlates of council effectiveness.

Council Clirna_te_

Social climate refers to the personality or character of a setting along a number Of
dimensions (Moos, 1973, 1979). Such dimensions refer to a variety of elements of the
social setting, including, for example, the degree to which the setting fosters cooperative
versus competitive relationships and/or promotes the personal-development of setting
members (e. g., achievement; Moos, 1973, 1979). Moos suggested that social climate
could be characterized by employing the same methods researchers use to characterize
individual personality traits. A focus on the climate of a setting provides a useful
framework for the investigation of councils because it attends not only to those
characteristics of a setting that can be described ‘objectively’ (e. g, the council size or
breadth of membership), but to the character of various ‘intangible’ aspects of the setting

that can be understood only by assessing the experiences of individuals within the setting

(Rousseau, 1988).

In addition, an assessment of council climate is an important part of examining
council effectiveness given substantial evidence that the climate of a setting has a powerful
effect on organizational outcomes. These include individual level outcomes such as the
degree to which setting members implement desired innovations (Klein & Sorra, 1996), as
well as group level outcomes such as workgroup innovation (Anderson & West, 1998).
Organizational scientists also stress the importance of assessing climate with a particular
referent such as identifying a climate for innovation (Schneider & Reichers, 1983). The
focus of this study was to begin to uncover essential elements of a climate for
collaboration.

Speciﬁcally, this study examined four elements of the council climate hypothesized
to facilitate the collective work of domestic violence coordinating councils. While these
elements of council climate are not unique to the domestic violence context, there is some
evidence that certain issues may be more salient for collaborative efforts in one domain
(e. g., early intervention with families) than they are for those in another domain, in this
case, responding to domestic violence (Gray, 1989; Mattessich & Monsey, 1992; Reilly,
1998). Considering the unique aspects of the domestic violence context, this study
focused on assessing the degree to which the council climate encouraged common ground
and the inclusion of all stakeholders within a context of diversity and power disparity.

While power issues - speciﬁcally how councils deal with conﬂict and power

disparities among stakeholders - are common in many collaborative efforts (Winer & Ray,

1994), they are likely to be particularly salient in the domestic violence context. Feminist
scholars have long asserted that the sexism of a patriarchal structure is the root cause of
violence against women and its persistence across cultures (e. g., Radford & Stanko, 1996;
Schechter, 1982). Structural inequalities and imbalances of power lie at the center of the
perpetuation of violence against women (Schechter, 1982) and have been reproduced in
efforts to respond to domestic violence, both within the feminist movement and within
larger social institutions (e. g., the criminal justice system; Radford & Stanko, 1996). For
example, while domestic violence advocates have had to ﬁght for battered women’s basic
rights to safety to be upheld by courts, batterers have been treated leniently and have not
been held accountable by the courts.

These patterns reﬂect a history of structural inequity and necessitate a focus on
power dynamics in an examination of the collaborative process of domestic violence
coordinating councils. This is essential given that many of the stakeholders who were
engaged in these struggles are now required to collaborate with each other. Yet, these
varied stakeholders have disparate power to inﬂuence the reforms in policy and practice
that are central to the creation of a CCR to domestic violence. For example, high ranking
criminal justice officials, such as judges and prosecuting attorneys, have much greater
inﬂuence over whether cases are prosecuted and sentencing practices than domestic
violence advocates and survivors. However, it is often the latter that have greater
knowledge of domestic violence and what is needed to enhance the safety of women. To

be successﬁIl, these settings need to foster a climate that encourages all voices to be heard

and does not suppress the expertise of any stakeholder, especially those who have
traditionally been the least powerful. Given differences in the relative power of
stakeholders and the need to overcome divergent viewpoints, this study focused on the
relationship of four elements of the council climate to council effectiveness: a) the degree
to which councils handle conﬂict effectively, b) the degree to which power/inﬂuence in
decision making is shared across all stakeholders, c) the degree to which a shared/unifying
mission has been achieved and d) the degree to which leaders are efﬁcient and organized
as well as skilled at encouraging the active and equitable participation of all stakeholders.
Council Characteristics

Council characteristics, such as the council structure, were also investigated as
important correlates of council effectiveness. Such an investigation is critical given that
there is some evidence that characteristics of the council setting are related to council
effectiveness and that structural changes (e. g., incorporating the use Of an agenda) may be
more easily implemented than changes in council climate (e. g., fostering shared power in
decision-making). Thus, the relationship of four council characteristics to council
effectiveness were examined in this study: a) the breadth of ofﬁcial and active council
membership (i.e., the number of stakeholder groups who are official and/or active
members), b) the perceived presence of external support for council activities (e.g., the
support of the community and policy makers), and c) the degree to which the council has
formal structures in place (e. g., established processes for decision-making). In addition,

given that councils go through different developmental stages and that older councils will

10

likely have accomplished more than younger councils (Butterfoss, Goodman, &
Wandersman, 1993) the number of years the council has been in existence was included as
a covariate in all analyses.
Current Study

The current study had two primary purposes. The ﬁrst purpose was to examine
council effectiveness. To address this purpose, the following research questions were
investigated: 1) To what degree do councils address a broad range of goals relevant to the
development of a CCR (within and beyond the criminal justice system)? 2) To what
degree are councils accomplishing their goals? and 3) To what degree do council members
perceive councils as effective in meeting intermediate goals and in facilitating reforms in
policy and practice that encourage women’s safety and batterer accountability? The
second purpose of this study was to examine the correlates of council effectiveness. To
address this purpose, two ﬁnal research questions were explored: 4) To what degree is
council climate (e. g., shared power in decision-making) related to council effectiveness?
and 5) To what degree are council characteristics (e. g., formality of council stmcture)
related to council effectiveness? In order to investigate the issues that might emerge in the
process of developing a coordinated community response to intimate violence against
women, this study employed both survey research methods and semi-structured interviews
with key informants.

The following literature review begins by providing a historical context for

understanding the need for a coordinated community response to violence against women.

11

This is followed by an overview of current research on coordinated community response
efforts and the potential role of coordinating councils. Finally, a review is presented of the

multiple aspects of collaborative settings that are the focus of this study.

12

INTRODUCTION

History of the Domestic Violence Movement: Encouraging Community Accountability

A brief history of the battered women’s movement provides an appropriate starting
place for understanding current efforts to mobilize communities to respond to violence
against women in a coordinated fashion. This historical perspective sets the stage for
exploring the collaborative efforts of key stakeholders from multiple service systems. The
direction of current community interventions addressing domestic violence rests on the
historical developments of the last 25 years (Pence & Shepard, 1999). The battered
women’s movement affected major social change, successfully arguing that the state bears
the responsibility for protecting women and for holding their abusive partners accountable
(Dobash & Dobash, 1992; Schechter, 1982). While there is still much work to be done to
achieve these goals in all communities, many communities across the United States have
begun to transform their response to domestic violence.

While many events can be viewed as precursors to the domestic violence
movement that began in the early 1970's (Schechter, 1982), one marker for the ‘beginning’
of this movement was the opening of the ﬁrst women’s refuge in Great Britain (Dobash &
Dobash, 1979). This marker is particularly signiﬁcant considering that shelter programs
formed the cornerstone of the domestic violence movement and still play a critical role in
advocating for women with abusive partners. The domestic violence movement quickly

spread throughout the United States and battered women’s shelter programs were opened

13

across the country. It is important to note that the battered women’s movement was not
simply a service delivery reform movement, but a feminist movement that sought not only
greater access to services, but social change in societies that subordinate women and
perpetuate their abuse (Dobash & Dobash, 1992; Schechter, 1982). Understanding the
original goal of this movement is important because it underscores the tensions that
emerge as the response to violence against women becomes increasingly professionalized
(Radford & Stanko, 1996). Schechter (1982) discusses this struggle, noting that being
“caught up in daily survival and the need to build institutions, activists have sometimes lost
sight of the political visions that inspired the formation of a movement” (p. 5). Still, the
work of activists in the battered women’s movement started an ongoing transformation Of
the community response to violence against women, beginning the slow process of shifting
the responsibility for ending violence from survivors to the community.

To date, considerable progress has been made in increasing women’s safety and
batterer accountability. For example, there have been multiple reforms to the criminal
justice system, expanded shelter and advocacy services, and an increase in batterers’
interventions (Pence & Shepard, 1999). Regarding the criminal justice system, in every
state there is now some version of a personal protection order to remove abusive partners
from women’s homes (Pence & Shepard, 1999). This represents a signiﬁcant policy
change from the previous stance that required women to leave their homes if they wanted
to escape their abusers. Similar advances have occurred outside of the criminal justice

system. For example, both the American Medical Association and the American Bar

14

Association, two powerﬁIl lobbying groups, have taken stands against domestic violence
(F litcraft, 1992 as cited by Pence & Shepard, 1999).

It is also important to note the advances made in federal legislation that support
ending violence against women. Most notable was the passage of the Violence Against
Women Act of 1994 (VAWA; Title IV of the Violent Crime and Law Enforcement Act of
1994) and its subsequent renewal in 2000. VAWA was responsible for appropriating over
$130 million to states and US territories to support their efforts to effectively address
violence against women (Burt, Newmark, Jacobs & Harrell, 1998) and the renewal of this
legislation promises more ﬁJnds will follow. Legislative support was also evident in the
special provisions made in the sweeping welfare reform of 1996 to mitigate the potential
negative effects of welfare reform on women with abusive partners (i.e., Violence Against
Women Option of the Personal Responsibility and Work Opportunity Reconciliation Act
of 1996). While not all states are taking advantage of this option, and early evidence
indicates it is not widely used by women when it is available, its presence in the federal
legislation speaks to effective institutional advocacy and an increasing awareness regarding
the dangers survivors of domestic violence face.

Coordinated Community Response Efforts

This history Of efforts to stimulate a response to domestic violence illustrates the
necessary involvement Of varied stakeholders, including the human service, criminal
justice, and educational systems as well as religious organizations, all levels of government

and community businesses (Clark et al., 1996). The central goal of a coordinated

15

community response is to encourage cooperation and communication across these systems
in order to increase batterer accountability and women’s safety (Clark et al., 1996; Pence
& Shepard, 1999). These goals are achieved by the slow and arduous process of altering
policies and practices that have a direct impact on the safety of women. For example,
numerous communities across the United States have adopted pro or mandatory arrest
policies that shift the burden of deciding to arrest the batterer from the survivor of the
assault to the state (Gwinn & O’Dell, 1993; O’Dell, 1996).

The history of the battered women’s movement has demonstrated a core
component of systems change - that changes must occur on the institutional level to
inﬂuence changes among individual practitioners (e. g., law enforcement personnel; Pence,
1999). In order to affect change on the institutional level, dialogue and cooperation must
ensue on the part of all key organizations and systems. Hart (1995) emphasized that
parallel reforms of individual components of the systems responding to domestic violence
may decrease women’s safety and increase fragmentation across systems. However, it is
important to underscore that the creation of a CCR does not only involve coordinating
existing systems; changes must also be made to these systems SO that their policies and
practices protect women rather than place them in greater danger. That is, coordination
should not ensue for coordination’s sake, but for the purpose of stimulating reform in
policies and practices while simultaneously improving communication across systems
(Pence & McDonnell, 1999). For example, it is not simply that law enforcement needs to

better coordinate their efforts with domestic violence advocates, but that they also need to

16

utilize advocates’ understanding of the dynamics of domestic violence to inform arrest
policies and practices that increase the safety of survivors rather than inadvertently place
survivors in greater danger. In fact, Pence and McDonnell (1999) describe coordination
without reform to existing systems as potentially harmful to battered women, actually
resulting in decreased safety. For example, partnerships between domestic violence
advocates and child protective workers are essential (Carter & Schechter, 1997; Schechter
& Edleson, 1994), but in the absence of reform to the existing child protection system that
holds women exclusively accountable for the safety of their children, coordination efforts
could actually result in more women becoming visible to the child protection system,
placing them at greater risk for facing neglect charges or losing their children. Taken
together, Hart (1995) and Pence and McDonnell (1999) are suggesting that reform Should
not occur in the absence of coordination and coordination should not occur in the absence
of reform.

To date, the development of a coordinated community response to domestic
violence has taken three forms: a) free standing organizations responsible for encouraging
cooperation and institutional change (e. g., the Domestic Abuse Intervention Project
(DAIP) in Duluth), b) programs within existing organizations that are responsible for
encouraging coordination and institutional change (e. g., within a Shelter program or
prosecutor’s Office), and c) domestic violence coordinating councils (i.e., ﬁ'ee-standing
committees, task forces or boards formed to lead the coordinating effort; Gamache &

Asmus, 1999; Hart, 1995).

17

While there is no empirical evidence determining which of these three strategies is
most effective in facilitating a CCR, domestic violence coordinating councils have become
very popular and are increasingly formed as a way to meet the collaboration requirement
to receive federal and state ﬁinding (Burt, Newmark, Jacobs, & Harrell, 1998; personal
communication, Gamache, 1999). In fact, Gamache and Asmus (1999) report that
numerous national organizations have Speciﬁcally called for the development of
coordinating councils including: the National Council Of Juvenile and Family Court
Judges, the American Prosecutors Research Institute, the American Bar Association, and
the American Medical Association. The increasing presence of domestic violence
coordinating councils warrants fiIrther investigation regarding their effectiveness and the
factors that impede or facilitate their effectiveness.

Interagency Coordinating Councils

Broadly Speaking, coordinating councils are collaborative bodies that include key
stakeholders (i.e., individuals/organizations that are affected by or inﬂuence a given issue)
who meet regularly to respond to a given issue in a collaborative manner. In human
services, coordinating councils have gained some recognition for their potential role in
facilitating collaborative efforts by encouraging relationships among relevant stakeholders
(e. g., Butterfoss, Goodman & Wandersman, 1996; Foster-Fishman, Salem, Allen &
Fahrbach, in press; Penner, 1995). It is not surprising, then, that one of the vehicles
commonly employed in the process of creating a coordinated community response to

violence against women is a coordinating council (Clark et al., 1996). These councils

18

typically consist of representatives from many service systems and organizations that are
deemed central to intervening in domestic violence cases. Hart (1995) described the
purpose of domestic violence coordinating councils as bodies that seek to “coordinate all
the components of the criminal justice system to improve justice system practice and to
better communicate and collaborate in work to end violence against women” (p. 3).
However, Clark et a1. (1996) noted that often partners from multiple systems (e. g.,
criminal justice, human services, child protection) are included in these coalitions .

While there is little research on domestic violence coordinating councils,
coordinating councils and coalitions‘ have been commonly employed in the human service
delivery system to reduce duplication of services, encourage cooperation rather than
competition, and increase communication and coordination across service delivery
organizations (e.g., Abott, 1995; Chavis, 1995; Fawcett, 1997; Foster-Fishman et al., in
press; Penner, 1995; Skaff, 1988). Research in the human services arena suggests that
coordinating councils play a role in encouraging interorganizational exchanges and
developing more highly integrated service delivery systems (Foster-Fishman, et al., in
press) and enhancing communication among the agencies involved (Abott, 1995). For
example, Penner (1995) found that the presence of HIV coalitions increased

organizational interdependence, which is thought to be a critical component of successﬁll

 

4

The collaboration literature includes studies of collaborative bodies that are referred to by
a variety of names. For the purposes of this study, literature on collaborative bodies
including both coalitions and coordinating councils is included.

19

collaboration (Bond & Keys, 1993; Gray, 1989) and the efﬁciency of the service system.
Examining Domestic Violence Coordinating Council Effectiveness
Given that the implementation of domestic violence coordinating councils to
encourage a CCR has not been systematically studied, there is little information available
describing council effectiveness. The ﬁrst step in examining domestic violence
coordinating councils is to examine their effectiveness in facilitating a CCR in their
community. This is critical given there is some concern that coordinating councils may not
be the most effective vehicle for implementing a CCR (Gamache & Asmus, 1999). This
study utilized three sources of information to examine council effectiveness in facilitating a
CCR: a) a detailed description of the goals councils are addressing, b) the degree to which
council leaders report these goals as accomplished and c) the degree to which council
members perceive their councils as effective in meeting intermediate goals (i.e., increasing
stakeholders’ knowledge of one another) and the goals of a CCR (i.e., reforming policies
and practices to encourage women’s safety and batterer accountability). Utilizing multiple
sources of information provides a more complete examination of council effectiveness.
CM
Examining the goals of councils is important for three primary reasons. First,
domestic violence advocates have maintained that CCR efforts should focus on increasing
women’s safety and batterer accountability. This requires extensive reform to the criminal
justice system (Pence & Shepard, 1999). Thus, it is of interest to examine the degree to

which councils are addressing the broad range of criminal justice reforms necessary to

20

achieve women’s safety and batterer accountability. For example, there are a number of
issues that councils might address, including: a) increasing the accessibility of personal
protection orders, b) enforcing the adoption of pro or mandatory arrest policies, and c)
encouraging evidenced based prosecution (i.e., prosecution that is not dependent on
survivor’s testimony). Thus, this study examined the degree to which councils are
addressing a broad range of issues in the criminal justice system to increase women’s
safety and batterer accountability according to council leaders.

Second, while reforms to the criminal justice system are particularly essential with
regard to batterer accountability, there is a concern that councils must attempt to stimulate
reforms beyond the criminal justice system to encourage women’s safety (e. g., in the
social service or medical systems; Schechter, 1999; Hart, 1995). This is particularly
important given that many battered women are not involved with the criminal justice
system (Schechter, 1999; Sullivan & Keefe, 1999). In addition, community resources are
often not commensurate with women’s needs: there is a dramatic shortage of temporary
shelter for battered women (Hart, 1995), and battered women need a host of community
supports to protect them from their violent partners and ex-partners (Sullivan, 1991;
Sullivan & Bybee, 1999). In short, addressing issues outside of the criminal justice system
is also central to ensuring women’s safety. Issues councils might address that extend
beyond the criminal justice system include: a) developing early identiﬁcation systems in
emergency room settings, b) increasing women’s access to essential community resources

(e. g., transportation, housing, food, clothing, counseling, child care, and employment),

21

and c) fostering partnerships between child protective service workers and domestic
violence advocates. Thus, similar to examining the breadth of areas addressed in the
criminal justice system, this study examined the degree to which councils have addressed
needed reforms beyond the criminal justice system according to council leaders.

Third, this study examined the degree to which councils were accomplishing these
goals. The breadth of goals a council addresses is important given that the creation of a
CCR involves multiple reforms and a focus in one area exclusively would likely fall short
Of a successful CCR to domestic violence. However, simply addressing multiple areas on
a cursory level would also likely fall short of an effective CCR. Given that collective
efforts are difﬁcult to implement, there may be considerable disparity between the stated
goals of collaborative settings and what they are actually able to achieve. They may be
adept at determining where reform is required, but may be unable to generate needed
changes. Thus, examining the degree to which councils are accomplishing their goals
provides another dimension of understanding council effectiveness.

In summary, this study examined council goals as a proxy for council effectiveness
by investigating the breadth of goals councils addressed both within the criminal justice
system (e. g., enforcing the adoption of pro or mandatory arrest policies) and in other
systems (e. g., developing early identiﬁcation systems in emergency room settings). This
examination included providing descriptive information regarding the goals of councils,
speciﬁcally attending to the degree to which: a) they reﬂect a broad array of reforms to the

criminal justice system, b) they address issues beyond the criminal justice system and c)

22

they are accomplished.
Council Members’ Perceptions of Council Effectiveness

While an examination of the breadth of council goals and the degree to which they
are accomplished provides us with one source of information regarding the collective
work of councils, it is not viable to use only this information to measure council
effectiveness. To date, the knowledge base concerning what constitutes an effective
council is very limited and comparisons between councils on objective criteria such as the
number of goals they have addressed must be validated with other sources of information
on effectiveness. For example, one council may focus exclusively on issues in the health
care system as a vehicle for increasing women’s safety while another may focus only on
the criminal justice system response. These councils may be equally effective at creating
change in their communities depending on the unique need for reform in their respective
settings. While the information gathered from council leaders provides one source of
information regarding the degree to which council goals were being accomplished, a
second starting point for an examination of council effectiveness is to measure council
members’ perceptions of the degree to which their council is achieving its goals, the stated
goals of a CCR (i.e., women’s safety and batterer accountability) and intermediate goals
(i.e., actions that lead to increased coordination such as increasing communication across
stakeholders).

Examination of council members’ perceptions of council effectiveness is useful for

three reasons. First, there is a long tradition of assessing the features of a setting via the

23

perceptions of those who are a part of the setting (e. g., Florin, Giamartino, Kenny &
Wandersman, 1990; Forehand & Gilmer, 1964; James, 1982; James & Jones, 1974;
Koslowski & Hults, 1987). This practice in the social sciences emerges from the belief
that members of a setting are well positioned to evaluate the character of the setting
(Forehand & Gilmer, 1964; Moos, 1973, 1979). Given that the perceptions of the
members of a setting are often utilized to evaluate the character of the setting, council
members’ perceptions provide a useful proxy for council effectiveness given that a clearly
objective criterion is not currently available.

Second, the perceptions of council leaders sometimes do not reﬂect the
experiences of the broader membership of an organization (Foster-Fishman & Keys, 1997;
Martin, 1992). In addition, asking multiple members of a setting to rate its effectiveness
can provide a broader perspective and a more reliable assessment of the setting (Bliese,
2000). Thus, council members’ perceptions provide another source of information to
examine council effectiveness and an opportunity to validate the varied indicators of
council effectiveness employed in this study.

Finally, the degree to which council members perceive their council as effective
inﬂuences their commitment to the collaborative process and their continued participation
in council activities (Skaff, 1988). Across a variety of collaborative settings several
researchers have found that the more satisﬁed council members are with the success and
accomplishments of their councils the more likely they are to actively participate in and be

committed to council activities (Butterfoss, Goodman, & Wandersman, 1996; Foster-

24

Fishman, Shpungin, Bergeron, & Allen, 2001). Given that retaining members and
ensuring their active participation is a critical component of successful collaboration,
examining the degree to which council members view their council as effective is
important.
Investigating Correlates Of Council Effectiveness

Examining collaborative bodies requires a focus not only on the degree to which
they are effective, but what contributes to their effectiveness. While the collaboration
literature has identiﬁed numerous facilitators and barriers to collaboration (e. g.,
Butterfoss, Goodman, & Wandersman, 1993; Foster-Fishman, et al., 2001; Mattesich &
Monsey, 1992), efforts to build models of effective collaboration and examine trends
across collaborative settings have been limited by the relatively small number of studies
that examine more than a single council (see Butterfoss, Goodman, & Wandersman, 1996;
Gottlieb, Brink & Gingiss, 1993; Kegler et al., 1993;Yin & Kaﬁarian, 1997 for examples).
Thus, the second purpose of this study was to examine multiple settings in an effort to
understand more about what aspects of the collaborative setting inﬂuence council
effectiveness. This study focused on the: a) council climate for collaboration (e. g., the
degree to which power is shared in decision-making; the degree to which leadership is
effective) and b) characteristics of the council setting (e. g., the degree of formality in the

council structure) as correlates of council effectiveness.

25

Council Climate

Climate refers to the personality or character of a setting (Moos, 1973, 1979) or
the internal environment that distinguishes one organization from another (Pritchard &
Karasick, 1973). A focus on the climate of domestic violence coordinating councils
provides a useful framework for this study given that it captures the character of various
‘intangible’ aspects of the setting that can be understood only by assessing the experiences
Of individuals within the setting (Rousseau, 1988). The collaboration literature does
provide some evidence that the internal climate councils foster can be integral to their
success (Foster-Fishman et al., in press; Gray, 1985; Harrison et al., 1990; McNulty,
1990). Studies of coordinating councils and coalitions have focused on numerous
facilitators and barriers that could be conceived as aspects of the climate of councils
including the presence of compatible stakeholder ideologies (Winer & Ray, 1994), mutual
respect (Mattesich & Monsey, 1992), and sense of community (Kumpfer et al., 1993). In
an effort to build a model of domestic violence coordinating council effectiveness, this
study focused on those aspects of the council climate that seemed particularly germane to
collaborative efforts in the domestic Violence context: a) the ability to handle conﬂict
productively, b) the need to share power/inﬂuence in decision-making, c) the presence of a
Shared mission and d) the presence of quality leadership.
Elements of the Council Climate as Correlates of Council Effectiveness

Studies of collaborative bodies rarely attend to the speciﬁc context of collaborative

efforts (i.e., what issue the collaborative effort is trying to address) and the dynamics

26

associated with the nature of addressing a particular social issue. Attending to such
context-speciﬁc issues is an essential component of building a knowledge base to support
collaborative efforts in a particular domain (Gray, 1996; Reilly, 1998). When considering
the domestic violence context it seems some of the most important issues in current
collaborative efforts are related to the history of the domestic violence movement and
power differences among key stakeholders.

While the history of the battered women’s movement is inspirational, it also
highlights a history of tensions and hard fought battles between domestic violence
advocates and other critical stakeholders, most notably those from the criminal justice
system and multiple levels of government. In the words of Pence and Shepard (1999), “in
every state, in every courthouse, and in every squad car, there has been resistance to the
full measure of what this social movement seeks to gain for women” (p. 5). Laws were
passed and policies were reformed, but these changes are not always upheld and are
sometimes used against women with abusive partners (Pence & Shepard, 1999). The gap
between the spirit of the law and enforcement of the law is well exempliﬁed by the
consequences battered women have faced when they have chosen not to take an active
role in the prosecution of their abuser. While domestic violence advocates focus on the
safety of survivors and see this as a viable choice for women, many prosecuting attorneys
view this as “failure to cooperate.” In fact, many women have been criminally charged
with failing to comply with a subpoena, ﬁling a false report and/or neglecting their children

(Pence & Ritmeester, 1992). This backlash is also evident in the increased number of

27

women who are arrested and charged with assault of their batterers when acting in self-
defense (Pence & Shepard, 1999). Further, domestic violence advocates themselves have
been charged with practicing law without a license or obstructing justice (Davies, 1995 as
cited by Pence & Shepard, 1999).

The purpose of highlighting these tensions is that the history of relations among
multiple stakeholders often impacts the current relations of those stakeholders and their
ability to form collaborative alliances (F oster-Fishman, Perkins, & Davidson, 1997).
Domestic violence advocates have played the historical role of applying needed pressure
to create consistency in written law and legal practices. It is unrealistic to expect that
when domestic violence advocates and representatives from the criminal justice system sit
at the same table to create a CCR that these historical patterns will not continue to
emerge. For example, in a comprehensive study of the coordinated community response
in Baltimore, Maryland, Clark et al. (1996) found that the presence of the local shelter
program as a member of the Domestic Violence Coordinating Council had caused some
tension. They describe that in some instances the Shelter program “had served as a willing
collaborator and at other times as an advocacy group pushing criminal justice agencies to
do more” (p. 29). Clark et al. (1996) report that in one instance, the shelter released
criminal justice data shared at a Council meeting to the press without the permission of the
Council. This resulted in a future reluctance to share this type of information which
impeded openness among Council members. This tension exempliﬁes how the historical

relationships among stakeholders and the historical roles these stakeholders have played

28

have an impact on their current collaborative relationships.

The history of the domestic violence movement provides insight into how to frame
a study of factors that may contribute to council effectiveness. First, examining this
historical context illuminates the need for councils to expect conﬂict and address it as it
arises. Historically, stakeholders responding to domestic violence have not shared a
common philosophical base regarding why domestic violence occurs (Pence, 1999) and
sometimes have had different priorities in responding to domestic violence cases
(Gamache & Asmus, 1999). These differences have often led to conﬂict regarding how
domestic violence should be addressed. A failure to effectively manage conﬂict is
associated with stagnation (Chavis, 1999), mistrust (Wichnowski & McCollum, 1995),
and sometimes withdrawal of key stakeholders from the collaborative effort (Bitter, 1977;
Wichnowski & McCollum, 1995). Therefore, the degree to which councils can effectively
handle their conﬂict may impact council effectiveness.

Second, history reveals a consistent pattern of disparities in power among critical
stakeholders in the domestic violence arena that is likely to be reproduced in the structure
of councils (e. g., Gamache & Asmus, 1999). These power differences exist both in
relation to structural inequalities in the social context and in relationships among
stakeholders with different organizational roles. Addressing these power disparities is
critical for domestic violence coordinating councils, given that encouraging Shared power
in the decision-making process of the council has been found to be related to the success

of collaborative efforts (Bartunek, Foster-Fishman, & Keys, 1996; Bond & Keys, 1993;

29

Gray, 1989).

Third, the presence of a shared mission is a critical counterpart to overcoming
conﬂict effectively. That is, the ability of councils to develop a Shared mission has
important implications for their collective work because it is within this unifying
framework that they overcome their differences and meet their shared goals (Bond &
Keys, 1994). An inability to foster such a shared direction can result in stagnation and the
dissolution of collaborative settings (Bitter, 1977).

Finally, effective council leadership has repeatedly been linked to the success of
collaborative endeavors (Butterfoss, Goodman, & Wandersman, 1996; F oster-Fishman et
al., in press; Mattesich & Monsey, 1992; Roussos & Fawcett, 2000). Leaders can play a
critical role in keeping a group focused and active (Bitter, 1977) and fostering an inclusive
climate (Kumpfer et al., 1993). In a context of power disparity and conﬂict, the role of an
effective leader seems to be paramount.

In summary, consideration of the domestic violence context suggests that an
investigation into the collaborative efforts of domestic violence coordinating councils must
attend to the degree to which these collaborative bodies foster a climate characterized by
effective conﬂict resolution, shared power across stakeholders, a shared mission and
effective leadership.

Conﬂict trangsformation. Conﬂict is viewed as endemic to the collaborative process
(Bitter, 1977; Byles, 1985; Fargason et al., 1994; Frederickson, 1996; Gray, 1989; Gray,

1996; Jones & Bodtker, 1998; Smith & Berg, 1997; Reilly, 1998; Tjosvold & De Dreu,

3O

1997; Wischnowski & McCollum, 1995). In fact, Smith and Berg (1997) describe
“opposition, polarities and conﬂict [as] a part of the DNA of collective life” (p. xxvii).
Conﬂict, while challenging for the members of a collaborative body, is a potential catalyst
for generating creative solutions (Gray, 1989; Jones & Bodtker, 1998), encouraging
needed change (Bitter, 1977), avoiding “groupthink” (Burnett, 1993) and ensuring trust in
ﬁrture exchanges (Reilly, 1998). Therefore, it is not the presence of conﬂict that is a
problem, but how the collaborative body addresses this conﬂict (Chavis, 1999).

While there is little research regarding how collaborative bodies transform their
conﬂict, there is a recognition that conﬂict resolution skills are important for the
membership of interagency coordinating councils (Foster-Fishman et al., in press;
Wischnowski & McCollum, 1995). The literature on conﬂict resolution provides some
insight regarding what constitutes an effective response to conﬂict and what does not.
Bitter (1977) draws on this research and suggests that conﬂict may be handled by
coalitions in one of the following ways: a) withdrawal of one or more participants, b)
“smoothing over” or ignoring conﬂict, c) compromising, d) ending the conﬂict by
introducing a mediator external to the conﬂict or e) encouraging a confrontation that leads
to an effort to solve the underlying problems that fuel the conﬂict. Bitter (1977) suggests
that the latter is the most effective, but does not explicate how this confrontation would
proceed productively. Similarly, Fargason et al. (1994) suggest that oﬁen conﬂict is not
handled constructively because the members Of a collaborative body focus on the needs of

individuals within organizations rather than the conﬂicts between organizations. Thus, in

31

an effort to meet the needs of an individual member, the underlying reason for the conﬂict
remains unaddressed. They identify three dysfunctional ways that multidisciplinary teams
handle conﬂict: avoidance, diffusion, and power. The avoidance technique denies the
existence of the conﬂict; the diffusion technique acknowledges the conﬂict, but attempts
to minimize it; and the power technique meets the needs of the most powerful member.
Similar to Bitter (1977), Fargason et al. (1994) suggest that as an alternative to these
dysfunctional methods Of handling conﬂict, members of collaborative bodies look for
“systematic dysﬁmction” in the interaction of organizations.

Addressing the relationship between how councils handle their conﬂict and council
effectiveness is particularly important within the domestic violence arena for two primary
reasons. The ﬁrst of these is the history of tension and adversarial relations in the
women’s movement to end violence against women. History is not easily overcome and
leads to current manifestations of conﬂict (Gray, 1989; Jones & Bodtker, 1998). In
describing the early years of developing the Domestic Assault Intervention Project
(DAIP), one of the most well respected CCR’S in the United States, Pence & McDonnell
(1999) describe “an atmosphere of distrust, defensiveness, and ﬁnger pointing” (p. 45). In
the beginning of their development it seemed that conﬂict fueled further conﬂict. They
explained that “Shelter advocates challenged agencies and institutions, which oﬂen
responded with hostility. Battered women’s advocates were usually seen as ‘pushy, single

3”

issue, and inherently biased outsiders (p. 45). Thus, in reﬂecting on their experience,

Pence & McDonnell (1999) note that organizations must work to overcome differences in

32

order to be effective. They suggest this requires relinquishing polarizing misconceptions
that plague the collaborative process. For example, they suggest that advocates must
recognize that they are not the only individuals concerned with the welfare of battered
women and that legal practitioners must discontinue viewing advocates as biased and
recognize the biases inherent in their own training and organizational role.

The second reason that conﬂict must be handled effectively in the context of
domestic violence coordinating councils is that Often key stakeholders have different
understandings of why domestic violence occurs (e. g., mental health perspectives
emphasizing pathology and poor conﬂict resolution skills versus feminist perspective
emphasizing power and control). Very oﬂen individuals working in different systems do
not have a shared philosophy regarding the etiology of domestic violence (Jenkins &
Davidson, 1999). Differences in beliefs about why domestic violence occurs are
polarizing because they are not only “personal opinions” but have an impact on the actions
organizations choose to take. For example, Pence (1999) notes that one of the greatest
differences between practitioners in the criminal justice system and domestic violence
advocates is that criminal justice practitioners often argue that women perpetuate the
violence against themselves while advocates argue that the state’s response to violence
against women not only perpetuates violence, but causes it. Thus, advocates may focus
on requiring the state to take responsibility and action for protecting women (e. g.,
encouraging evidence-based or ‘victimless’ prosecution), while the state requires women

to take a more active role in their safety (e.g., participating as witnesses for the

33

prosecution).

Of course, conﬂicts in philosophy do not only emerge around how to work with
and protect women. Differences in philosophy also emerge with regard to batterer
interventions. While advocates often recognize the need to intervene with batterers, there
is also a concern that the presence of these groups would “decriminalize” the behavior of
batterers, making it more difficult to hold them accountable in the criminal justice system
(Pence & Shepard, 1999).

Addressing conﬂict regarding why domestic violence occurs and how to intervene
is of critical importance for domestic violence coordinating councils. In fact, Gamache &
Asmus (1999) suggest that the absence of a shared understanding of the etiology of
domestic violence will result in the lack of “a clear commitment to the priority of victim
safety, [which could result in a] council [that] will propose changes in policy that cause
additional problems for victims or actually endanger them” (p. 77). While they stress the
importance of addressing differences in philosophical base regarding why domestic
violence occurs, they also suggest that this is one of the greatest challenges councils face.
Thus, addressing conﬂicts that emerge from differences in philosophy is critical to councils
achieving the central goals of a CCR - women’s safety and batterer accountability.

There is ample evidence suggesting that domestic violence coordinating councils
may face conﬂict as a result of grappling with historical tensions or differences across
stakeholders in their understanding of why domestic violence occurs and how to best

respond. This study focused on the degree to which councils were viewed by members as

34

engaging in conﬂict resolution behaviors that are constructive (e. g., stimulating needed
changes) rather than destructive (e. g., avoiding conﬂict) and explored the relationship of
the use of constructive conﬂict resolution tactics to council effectiveness. This is essential
because a failure to deal with conﬂict productively can seriously impede the ability of a
collaborative effort to be effective. Research suggests that a failure to deal with conﬂict
results in mistrust (Wichnowski & McCollum, 1995), withdrawal of key participants
(Bitter, 1977; Wichnowski & McCollum, 1995), stagnation (Chavis, 1999) and an
avoidance of addressing any issues that could be viewed as controversial (Kegler,
Steckler, McLeroy, & Malek, 1998). Closely related to the need to transform conﬂict
effectively is the need for Shared power across stakeholders. Both of these processes
involve facilitating equitable solutions and hearing the voice of all council members.

Power\inﬂuence in decis_ion- _a_kigg._ Issues of shared power in decision making
are intimately related to issues of effective conﬂict transformation. In his review of
political models of organizations, Pfeffer (1999) links the resolution of conﬂict to that Of
power, noting that in the presence of conﬂict “the power of the various actors determines
the outcome of the decision process” (p. 369). Thus, the examination of the relationship
of Shared power and inﬂuence in decision-making is an important companion to the study
Of effective conﬂict transformation.

For the purposes Of this study, power will be deﬁned as summarized by Shafritz
and Ott (I987) - “the ability to get things done the way one wants them done...the latent

ability to inﬂuence people” (p. 399). Considering power and inﬂuence in an examination

35

of collaboration is critical given that shared power in decision-making is an important
component of success in the collaborative process (e. g., Bartunek, F oster-Fishman, &
Keys, 1996; Bitter, 1977; Bond & Keys, 1993; Butterfoss, Goodman, & Wandersman,
1993; Chrislip & Larson, 1994; Gray, 1985; Gray, 1989; Jones & Bodtker, 1998; Lipman,
1997). Collaboration cannot ensue in situations where one stakeholder has unchallenged
power or in situations where power is not shared “roughly” equally (Gray, 1985).
However, this ‘requirement’ is more easily proposed than achieved. While the
collaboration literature does address the existence of disparities in power, there is little
attention paid to how these disparities are overcome, what happens if critical stakeholders
are being excluded from collaborative bodies or included only as token members, and
whether the presence or absence of Shared power in decision making is related to council
effectiveness.

In her review of the conditions facilitating interorganizational collaboration, Gray
(1985) and McCann and Gray (1986) assert that stakeholders who are involved in
collaboration must be perceived as having a legitimate stake in the problem. A legitimate
stake is deﬁned as having the “perceived right and capacity to participate in the
developmental process” (p. 921). While the right to participate is shared by all who are
affected by a given problem, the capacity to participate is not as widely applicable. Gray
(1985) notes the direct implications of power disparity here and suggests that the less
powerful stakeholder must “build their power base before they can gain legitimate Status

as a stakeholder” (p. 922) (Gricar & Brown, 1981 as cited by Gray, 1985). The problem

36

with this solution is that it places responsibility on those stakeholders who are the most
disenfranchised to acquire power. This presents a double bind for disenfranchised
Stakeholders given that they often possess inadequate resources yet are expected to claim
power from those who control the resources. This places less powerful stakeholders at
risk for having even less inﬂuence because some of the tactics they typically use to
encourage accountability and assert their agenda may be deemed inappropriate by the
collaborative body. In fact, those who have power will often label those with less power
as ‘radical’ or ‘biased,’ making their equitable participation very difﬁcult. This has direct
implications for domestic violence coordinating councils given the substantial power
differences among those who are represented on these councils. These power differences
stem from two primary sources: a) structural inequities with regard to social location (e.g.,
race, gender, class, sexual orientation) and b) inequities in organizational role across and
within systems (Gamache & Asmus, 1999).

With regard to stnIctural inequities, key stakeholders on domestic violence
coordinating councils have varied social locations (e. g., race, class, gender, sexual
orientation), each of which has implications for the amount of power they may be able to
exert in the setting (Pfeffer, 1999; Wood, 1989). For example, from a structural
standpoint, men have more power than women, individuals from upper socioeconomic
classes have more power than those in lower socioeconomic classes, and whites have
more power than people of color (Frye, 1983; Anderson & Hill Collins, 1998). It is not

uncommon for individuals with less power in the social strata to exert less power in

37

collaborative contexts (Gruber & Trickett, 1987; Krogh, 1998; Lipman, 1997). Gruber
and Trickett (1987) assert that it is impossible to share power within a context of inequity.
This is well illustrated in Lipman’s (1997) study of a collaborative effort attempting to
create a more equitable learning environment for low-income, minority students. She
found that the voices and inﬂuence of marginalized stakeholders (i.e., teachers Of color,
Aﬁican-American community activists) were suppressed by the white, upper middle class
stakeholders. Given the power inequities embedded in our social fabric it is likely that
encouraging the voices of all stakeholders is a difﬁcult task for councils which have a
diverse membership with regard to power and requires a concerted effort.

While structural inequities reﬂect our social context, differences in power among
organizational roles represented on the council presents an equally challenging issue with
regard to shared power in decision making. In this case, two individuals with the same
social location may still have vastly different inﬂuence. Wood (1989) found that
differences in status (i.e., organizational role) negatively affected the ability of all group
members to contribute equally to a volunteer group. While the intention of this group was
to create equal voice across participants, the absence of a formal structure ensuring such
equity resulted in less powerﬁIl members (i.e., teachers versus principals) having less
inﬂuence. This same dynamic is particularly challenging with regard to the existing power
structure of the criminal justice system. For example, prosecuting attorneys or domestic
violence legal advocates may not feel comfortable speaking openly in front of judges

whom they may subsequently face in court (Gamache & Asmus, 1999). Similarly, a staff

38

person who is seated next to the leader of their organization may not be forthcoming with
information about the Shortcomings of a particular policy or procedure or other
intraorganizational issues. It is possible that on councils where issues of power disparities
have not been addressed, certain individuals will have a limited voice.

Finally, it is not unusual for stakeholders who are the target of a council or
coalition to be excluded or to have less voice in the setting. For example, Armbruster et
al. (1999) found that elders in an Eldercare coalition had the least amount of inﬂuence.
Gamache and Asmus (1999) note that many councils fail to include or create a structure
that supports input from survivors or advocates. Even if representatives from these
groups are included, they often face situations where they are heavily outnumbered by
those from government agencies. In fact, given that domestic violence service providers
or women’s advocates have traditionally represented the interests of women with abusive
partners they may ﬁnd themselves particularly challenged in their new role as collaborative
partners. Abernathy, Edwards, & Gamache (1999) speciﬁcally warned domestic violence
advocates about the possibility of being cO-Opted by other stakeholders in the process of
developing a CCR. They suggested that advocates avoid this by systematically examining
every decision in which they participate in collaboration with others in order to insure that
the policy or practice with which they are agreeing beneﬁts women with abusive partners,
does not place them at greater risk, and holds batterers accountable for their actions.

While multiple studies discuss the need to have shared power in order to

successﬁllly collaborate (Gray, 1989; Mattesich & Monsey, 1992; Winer & Ray, 1994)

39

few studies have closely examined issues related to shared power in multiple Stakeholder
groups. Two exceptions, Bond and Keys (1993) and Bartunek, Foster-Fishman, & Keys
(1996), focused on the co-empowerment of members of multiple stakeholder groups
addressing developmental disabilities. Bond & Keys (1993) deﬁned co-empowerrnent as
occurring when each stakeholder group was able to inﬂuence the decision-making process.
In a sense, co-empowerment can be thought of as antithetical to a lack of shared power,
given that no one group will dominate or suppress the unique aims of another (Bond &
Keys, 1993). Bond and Keys (1993) and Bartunek, F oster-F ishman and Keys (1996)
found that it was possible to achieve desired collaborative outcomes while at the same
time encouraging maintenance of within group identity and goals. This is critical because
as Bond and Keys (1993) point out, in the absence of each group retaining its unique
goals, “collaboration can easily deteriorate into cooptation by the stronger group” (p. 54).
Given the potential for power inequity among stakeholders on a domestic violence
coordinating council, it is critical to examine a) the degree to which councils are viewed by
their members as sharing power in decision-making and b) the relationship of shared
power in decision-making to council effectiveness. Shared power is a crucial component
Of council effectiveness for two reasons. First, to be a successful collaborative body all
voices represented must be heard - particularly the voices of those who have traditionally
been excluded. Often, the purpose of the collaborative process is to create a forum where
previously adversarial parties can work toward achieving a common interest or goal

(Gray, 1989). The exclusion of key stakeholders in the decision making process seems

40

counter to this purpose. Encouraging shared power is especially important when
considering that each stakeholder brings unique pieces of useful information to the council
(Larson, Foster-Fishman & Keys, 1994). In fact, groups where more information is
shared make better decisions (Larson, F oster-Fishman, & Franz, 1998). Second, in the
domestic violence context, those members who have traditionally had the least power
from a structural standpoint (domestic violence advocates and survivors) have a great deal
Of expertise to offer. In order for councils to be effective at achieving the goals of a CCR,
they must allow for all members to have a voice in the decision-making process of
councils.

Shared mission. The presence of a shared mission has been repeatedly linked to
the effectiveness of collaborative bodies (e. g., Gray, 1985). The development and
maintenance of a Shared mission serves as the connective tissue of a collaborative body.
Collaboration is characterized by multiple stakeholders coming together to accomplish a
goal that is greater than what any stakeholder could accomplish alone (Gray, 1985).
However, to do this, stakeholders must often overcome disparate views about how a goal
should be accomplished or about what should constitute the focus of the collaborative
effort (Allen et al., 1994; Kagan, 1991). While deﬁning a common goal can be quite
difﬁcult, it is a necessary part of the collaborative process (Butterfoss, Goodman &
Wandersman, 1993; Gray, 1989; Kagan, 1991). In fact, in their comprehensive review of
the literature on community coalitions, Butterfoss, Goodman and Wandersman (1993)

conclude that the “most important element in coalition formation appears to be the

41

articulation of a clear mission or guiding purpose” (p. 11).

The presence of a shared mission is related to the effectiveness of a collaborative
body ultimately meeting its shared goals (Bond & Keys, 1994). Gray (1985) describes this
as the ﬁrst stage of coalition development - direction setting. During this stage,
stakeholders develop a common or shared purpose. Developing such a shared purpose is
essential given that the degree to which stakeholders are operating within a unifying
framework inﬂuences their desire to work together (e. g., Hord, 1986).

Although recognized as a critical factor, the role a shared mission plays in
inﬂuencing council effectiveness has not been examined within the context of domestic
violence coordinating councils. Given the divergent perspectives and priorities Of
stakeholders on such councils, achieving a shared mission may be a particularly important
component inﬂuencing council effectiveness within the domestic violence arena.
Therefore, this study examined the degree to which councils are viewed by their members
as having a shared mission and the relationship between the presence of a shared mission
and council effectiveness.

Council leadership. Numerous studies have established that effective leadership of
a collaborative body is integral to its success (Bitter, 1977; Butterfoss, Goodman, &
Wandersman, 1993; Butterfoss, Goodman, & Wandersman, 1996; Feighery & Rogers,
1989; Kegler, et al., 1998; Kumpfer et al., 1993; Mattessich & Monsey, 1992; Reilly,
1998; Winer & Ray, 1994). Leaders are thought to keep collaborative bodies focused and

active (Bitter, 1977), to contribute to the perceived legitimacy of a group (Chrislip &

42

Larson, 1994), to inﬂuence team effectiveness (Kumpfer et al., 1993 ), to support the
membership of the coalition (Brown, 1984) and to enhance member satisfaction (Kegler et
al., 1998).

In addition, there is some evidence that particular styles of leadership have
differential effects. For example, there is evidence that participative leadership styles -
those that encourage the active participation and voice of all members -— are more effective
at promoting member discussion of all available information than autocratic leaders
(Larson, Foster-Fishman, & Franz, 1998). Similarly, Kumpfer et al., (1993) argued that
an empowering style of leadership (e. g., one in which the leader takes time to encourage
member participation) is positively related to team effectiveness. Participative,
empowering leaders who encourage the active participation of all members are important
when considering the domestic violence context where stakeholders bring a diverse array
of expertise and vantage points.

Given that leadership is consistently viewed as an essential component of
collaboration and, to date, has not been investigated with regard to domestic violence
coordinating councils, this study examined the degree to which council leadership is
viewed as effective by council members and the relationship between the quality of council
leadership and council effectiveness. Considering the importance of participative and
empowering leadership styles, this study hypothesized that councils will be more likely to
be effective if their leaders are perceived to be: a) sensitive to member concerns, b)

competent in negotiations, garnering resources, problem solving and conﬂict resolution, c)

43

able to promote equity and collaboration among members and value all members’ input, d)
able to plan council activities efficiently and e) knowledgeable in the area the council is
addressing (Butterfoss, Goodman, & Wandersman, 1993).

In summary, this study proposes that in order for domestic violence coordinating
councils to be effective vehicles for encouraging collaboration and facilitating a
coordinated community response, they must foster a climate where they transform their
conﬂict constructively, share power/inﬂuence with all stakeholders, develop a shared
mission and foster effective leadership. The need for such a climate is illustrated by Gwinn
& O’Dell’s (1993) description of the CCR efforts in San Diego, a model for the country.
They recall that “meetings were characterized by humility, transparency, and at times,
heated exchanges. The advocates understood little about the requirements of successful
criminal prosecution. And the prosecutors and police officers understood even less about
the dynamics of domestic violence” (p. 1504). The ability of councils to overcome
conﬂict, ﬁnd common ground and incorporate the input of all stakeholders to foster shared
understanding were examined as critical predictors of council effectiveness.

Council (Mcteristics and Council Effectiveness

While features of the council climate (e. g., effective conﬂict resolution and shared
power/inﬂuence in decision-making) are important predictors of perceived council
effectiveness, there are also aspects of council structure (e. g., the degree to which the
council has a broad membership) that may explain differences in council effectiveness.

Research on coalitions suggest a number of council characteristics that may inﬂuence their

44

effective ﬁInctioning. For the purposes of this study, a few characteristics hypothesized to
affect council effectiveness have been targeted: a) the breadth of official and active council
membership (Butterfoss, Goodman, & Wandersman, 1993; Gray, 1985; Mattesich &
Monsey, 1992), b) the perceived presence of external support for council activities (e. g.,
Reilly, 1998), and c) the formality of council structure (e. g., Foster-Fishman & Mauricio,
1999; Kegler et al., 1998). Given the relatively small number of studies focusing on the
relationship of council effectiveness to the breadth of council membership and degree of
formality in the council structure, these characteristics were examined in an exploratory
fashion, while the presence of external support was hypothesized to be positively related
to council effectiveness. In addition, the number of years the council has been in existence
was examined as a covariate in all analyses.
Bﬂadth of ‘Ofﬁciwd Active Council Membership

A critical council level predictor Of council effectiveness is the degree to which the
council’s membership is inclusive (Florin, Mitchell, & Stevenson, 1993; Gray, 1985;
Mattesich & Monsey, 1992). Gray (1985) suggests that a critical component of forming a
collaborative body is that all relevant stakeholders are included. Butterfoss et al. (1993)
describes a coalition’s membership as a primary asset given that each stakeholder brings a
unique set of skills and resources to the collaborative body. Further, they suggest that
diversity on a coalition allows the coalition to reach a broader constituency.

Gray (1985) suggests that the “stakeholder set needs to reﬂect the complexity of

the problem under consideration if collaboration is to occur” (p. 919). With regard to

45

domestic violence coordinating councils, these stakeholder groups include survivors,
domestic violence advocates from all local domestic violence programs, and members of
the criminal justice community (Gamache & Asmus, 1999). Clark et a1. (1995) expanded
this set, suggesting that human service providers, school administrators, clergy and local
business representatives must be included. Gamache and Asmus (1999) note that it is
imperative for all advocacy programs in a community to be involved. This not only
increases the presence and voice of survivors and/or advocates on the council, but it also
increases input from Otherwise disenfranchised groups such as ethnic or racial minorities,
gays, lesbians and bisexuals, immigrant women and refugees.

In this study, two aspects of council inclusiveness, or the degree to which the key
stakeholder set is fully represented were examined: a) the breadth of ‘official’ members of
the council (i.e., the number of stakeholder groups, for example, domestic violence service
providers and judges, that have been invited to participate), and b) the breadth of members
who are active participants (i.e., the number of official members who are actively engaged
in council activities). Examining both aspects of membership is critical given that many
coalition members may be included in ‘name only’ and are not actively engaged or
incorporated into the work Of the council. While it is clear that a diverse membership is
central to the success of collaborative efforts, it is also possible such diversity may
generate more conﬂict and power struggles. Therefore, in the proposed study, the
relationship of the breadth of ofﬁcial and active council membership were examined in an

exploratory manner.

46

P_resence of External Support

In addition to examining council inclusiveness, it is critical to explore the presence
of support for council activities (Reilly, 1998). There is evidence to suggest that in the
development of a new organization it is critical for the organization to have support from
the broader community - also termed the ‘external environment’ (Bartunek & Betters-
Reed, 1987). Assessing the supportiveness of the external environment is critical given
that collaborative bodies are impacted by the environment in which they operate (Skaff,
1988; Staggenborg, 1986). In her study of child maltreatment coordinating committees,
Skaff (1988) found that professional and agency support and community support were
two major facilitators of committee effectiveness.

With regard to domestic violence coordinating councils, the degree to which the
‘external environment’ supports council activities and goals may impact the degree to
which a council is effective. F oster-Fishman et al., (in press) found that service providers’
perceptions of the ‘external environment’ (e. g., organizational leaders, policy makers) as
supportive of service delivery reforms was the most important predictor of whether or not
these reforms were adopted by human service providers. Such external support is likely to
be critical in the domestic violence context, given that the goals of the CCR require
signiﬁcant reforms in both the criminal justice and human service delivery systems. If
leaders from key organizations involved in developing a CCR are not willing to provide
time for staff to participate in meetings and are not willing to make needed changes in the

policies and practices of their own organization, council progress is likely to be stiﬂed

47

(F oster-Fishman, Salem, Allen & Fahrbach, in press). In a Similar fashion, the presence of
support from public officials and policy makers can bolster the efforts of a coordinating
council and increase their legitimacy in the community (Gray, 1985).

For the purposes of this study, the ‘external environment’ includes leaders from
organizations critical to CCR efforts, local policy makers, ﬁmders and community
members. Each can provide a particular type of support for councils. For example,
organizational leaders can alter policies and practices with their organizations, and can
provide ﬁnancial and human resources (e.g., allowing staff to attend meetings) to support
councils accomplishing their goals. Funders can provide ﬁnancial support to fund council
initiatives. Community members can demonstrate support by attending council activities
and organizing grassroots campaigns to support council efforts (e. g., community
education efforts). This study examined the perceived presence of support in the ‘external
environment’ as an important predictor of council effectiveness.

Council Structure

Issues related to the relative beneﬁts and costs of fonnalizing organizations have
been long debated (e. g., Rothschild-Whitt, 1979). Those who advocate for alternative
settings that challenge the typical bureaucratic structure associated with organizations
suggest that the elimination of rules, procedures, and agendas encourages shared power
and creativity (Holleb & Abrams, 1975 as cited by Heller et al., 1984). While it seems that
collaborative bodies could beneﬁt from such a non-bureaucratic structure, the

preponderance of research supports the presence of more formalized structures on

48

councils.

There is some evidence that council structures are positively related to multiple
indicators of council effectiveness (Butterfoss, Goodman, & Wandersman, 1993; Foster-
Fishman & Mauricio, 1999; Gottlieb, Brink, & Gingiss, 1993; Kegler et al., 1998; Kuhn,
Ducett & Edwards, 1999; Rogers et al., 1993). For example, Kegler et al., (1998) found
that councils which had more complex structures (i.e., bylaws, written agenda & written
minutes) mobilized greater resources and higher levels of implementation. However, they
caution that there is some concern that while formalized structures may facilitate
collaboration between professionals and organizations it may be counter productive with
regard to grassroots participation. Given that domestic violence coordinating councils
include organizations and also survivors of domestic violence, it is unclear whether
increased council structure will have a positive or negative inﬂuence on effectiveness. In
addition to the importance of formalized written procedures, others have found that the
presence of work groups or subcommittees is a critical component of effectiveness
(Foster-Fishman & Mauricio, 1999). Given these equivocal ﬁndings, for the purposes of
this study, the relationship of council structures (including the presence of bylaws,
subcommittees, formal decision-making processes and conﬂict resolution processes) and
council effectiveness were examined in an exploratory manner.

Council ‘Age’
It is important to examine the amount of time a council has been in existence as a

covariate of effectiveness. Council members’ perceptions of council effectiveness may

49

vary over time given that councils develop over time (e. g., Butterfoss, Goodman, &
Wandersman, 1993). This is evidenced by the numerous stage theories that have emerged
to explain council activities (e. g., Butterfoss et a1, 1993; Chavis, 1999; Gray, 1985).

Councils that have formed very recently are more likely to be focusing on creating
their goals and mission. As a result, recently formed councils may be perceived as less
effective when compared to councils who have had time to implement their goals. In
addition, councils who have been together for longer periods of time may have resolved
many key issues and created a balance between dealing with internal issues (e. g., conﬂict)
and external issues (e. g., accomplishing goals; Chavis, 1999). This is illustrated by
Gamache & Asmus (1999), who in describing the interagency efforts of the Domestic
Assault Intervention Project (DAIP), note that “despite the conﬂicts and tensions that are
inherent in these types of negotiations, a high level of trust has been established among the
cooperating agencies as they build a history of mutually satisfactory outcomes and
successful reforms” (emphasis added; p. 70). This suggests that the degree to which
councils are effective may be a ﬁinction of the amount of time they have been working
together. Given the potential inﬂuence of council ‘age’, the proposed study included the
number of months a council has been operating as a covariate of council effectiveness.

Exploring the Multi-Level Nature of Perceptual Data

There is a long history and common practice of assessing settings based on the

perceptions of the setting members (e. g., Florin, et al., 1990; Forehand & Gilmer, 1964;

James & Jones, 1974; Koslowski & Hults, 1987). However, organizational scientists have

50

been careful to note that organization members may view the same environment differently
as a result of their unique vantage points (e. g., Foster-Fishman & Keys, 1997; Forehand &
Gilmer, 1964; James & Jones, 1974; Martin, 1992; Powell & Butterﬁeld, 1978; Schneider
& Reichers, 1983). For example, according to Forehand and Gilmer (1964), such
differences in perception could be due to differences in ability, personality traits, social
positions, prior experiences, or differences in organizational role.

This suggests that there are multiple sources of variation in the perceptions of
setting members. One source of variation may be the ‘reality’ of the setting itself, while
the other may be the characteristics or position of the setting member. That is, councils
are likely to vary with regard to their effectiveness and climate; however, the perceptions
of their members may also vary based on their individual experiences of the setting.
Because of the potential for this individual level variation, discussion regarding the
assessment of context based on the aggregated perceptions of setting members has been
characterized by debate and disagreement with regard to both methodological and
conceptual issues (e.g., Glick, 1985; Glick, 1988; James, Joyce, & Slocum, 1988).
Conceptually, organizational scientists have debated whether perceptions of organizational
climate constitute psychological constructs or organizational attributes (e. g., Drexler,
1977; Glick, 1985; Glick, 1988; James, Joyce, & Slocum, 1988; James & Joyce, 1974;
James, 1982; Powell & Butterﬁeld, I978). Methodologically, there is a great deal of
concern about when perceptual measures can be aggregated to the setting level and when

disagreement between setting members is too great (e.g., James, 1982) for taking the

51

average of member perceptions to characterize a setting.

One approach is to employ James, Demaree and Wolf‘s (1984) strategy to
determine within group agreement (rum) to justify aggregating members’ perceptions of
organizational phenomenon. Given that the elements of council climate examined in this
study were measured using perceptual data, interrater agreement was calculated using this
method (rwgm) to determine whether it was appropriate to aggregate council members’
scores to the council level to model the breadth of council goals address and the degree to
which those goals were accomplished.

Another approach attempts to ease methodological and conceptual concerns
regarding levels of analysis by separating individual and group level effects (Bryk &
Raudenbush, 1992; Florin, et al., 1990; Kenny & La Voie, 1985). The individual level
refers to the individual perceptions of setting members; the group level refers to the
aggregated perceptions of setting members. Using this method, individual and group level
relationships can be examined Simultaneously, and the covariation accounted for at one
level can be separated from the other. This approach is useful as it avoids a common
problem of aggregating to the setting level without considering individual level variation
or simply viewing individual level variation as measurement error. Further, it allows for a
systematic examination of those factors that may explain individual variation in
perceptions of the same setting while simultaneously examining differences across settings.
Given this ﬂexibility, this approach (e.g., using multi-level modeling to decompose

individual and group level effects) was utilized in this study to examine the correlates Of

52

members’ perceptions of council effectiveness.

With regard to domestic violence coordinating councils, organizational role (e. g.,
domestic violence advocate, probation officer, prosecuting attorney, judge, mental health
professional) may inﬂuence the perceptions of council members. Therefore, it may be a
useful predictor of variation in perceptions of council members in the same setting if such
variation exists. Research has repeatedly demonstrated that individuals who hold more
powerful organizational roles (i.e., leaders) view organizational settings differently than
those who hold less powerful organizational roles (Foster-Fishman & Keys, 1997; Martin,
1992). Similarly, in our societal context, those in power are Often ‘blind’ to power
inequities while those with less power are acutely aware of power disparities (e. g.,
McIntosh, 1989). Taken together, these ﬁndings suggest that in the context Of domestic
violence coordinating councils, council members with more powerful organizational roles
may not perceive power inequities in the same manner as those with less power. For
example, judges may perceive power/inﬂuence in decision-making as more equitably
distributed than domestic violence advocates by virtue of the power differences endemic
to their roles.

In summary, in this Study, the impact of organizational role was introduced to
assist in explaining variability in member perceptions within councils. TO capture
variability in individual perceptions based on organizational role, this factor was examined
as a predictor of perceived council effectiveness and in relationship to each aspect of

council climate (i.e., having a shared mission, having effective leadership, transforming

53

conﬂict effectively and sharing power/inﬂuence in decision-making).5
Current Study

The current study has two primary purposes. The ﬁrst was to examine council
effectiveness. To address this purpose, the following research questions were
investigated: 1) To what degree do councils address a broad range of goals relevant to the
development of a CCR (within and beyond the criminal justice system)? 2) To what
degree do council leaders perceive their councils as accomplishing their goals? and 3) TO
what degree do council members perceive councils as effective in meeting intermediate
goals and facilitating reforms in policy and practice that encourage women’s safety and
batterer accountability? The second purpose of this study was to examine the correlates
of council effectiveness. To address this purpose two ﬁnal research questions were
explored: 4) To what degree is council climate (e.g., shared power in decision-making)
related to council effectiveness? and 5) To what degree are council characteristics (e. g.,
external support for council activities, formality of council stnIcture) related to council
effectiveness? In relationship to Research Question 4, the following hypotheses were

examined.

 

5

Given that domestic violence advocates have the most expert knowledge regarding
domestic violence and are generally less powerﬁll stakeholders, they were used as the
baseline (comparison) group in the analyses of organizational role.

54

Hypothesis 1a:

Hypothesis 1b:

Hypothesis 1c:

Hypothesis 2a:

Hypothesis 2b:

Hypothesis 2c:

Effective Cpnﬂig Resglutipn

The degree to which a council is perceived as employing
effective conﬂict transformation tactics will be positively
related to perceived council effectiveness.

The degree to which a council is perceived as

employing effective conﬂict transformation tactics will be
positively related to the breadth of Council goals.

The degree to which a council is perceived as

employing effective conﬂict transformation tactics will be
positively related to the degree to which council goals have

been accomplished.

S_hared Power in Decision-Making

The degree to which a council is perceived as having shared
power in decision-making will be positively related to
perceived council effectiveness.

The degree to which a council is perceived as having shared
power in decision-making will be positively related to the
breadth of council goals.

The degree to which a council is perceived as having shared
power in decision-making will be positively related to the

degree to which council goals have been accomplished.

55

Hypothesis 3a:

Hypothesis 3b:

Hypothesis 3c:

Hypothesis 4a:

Hypothesis 4b:

Hypothesis 4c:

Presence of a Shared Mission

The degree to which a council is perceived as having a
Shared mission will be positively related to perceived
council effectiveness.

The degree to which a council is perceived as having a
shared mission will be positively related to breadth of \
council goals.

The degree to which a council is perceived as having a
shared mission will be positively related to the degree to
which council goals have been accomplished.

Effective Council Leadership

The degree to which council leadership is perceived as
effective will be positively related to perceived council
effectiveness.

The degree to which council leadership is perceived as
effective will be positively related to breadth of council
goals.

The degree to which council leadership is perceived as
effective will be positively related to the degree to which

council goals have been accomplished.

56

Only one council characteristic was examined with a directional hypothesis, the
presence of external support for council activities. The breadth of ofﬁcial and active
membership and the formality of council structure were examined in an exploratory

manner. Thus, for Research Question 5, the following hypotheses were examined:

Presence of External Support for Council Activities

 

Hypothesis 5a: The presence of external support for council activities will

be positively related to perceived council effectiveness.

Hypothesis 5b: The presence of external support for council activities will

be positively related to breadth of council goals.

Hypothesis 5c: The presence of external support for council activities will

be positively related to the degree to which council goals
have been accomplished.

To account for variability in perceived council effectiveness, the hypotheses for
research question four (Hypotheses 1a, 2a, 3a, and 4a) separated individual level effects
(i.e., the relationship of individual council members’ perceptions of council processes and
their perceptions of council effectiveness) from group level effects (i.e., the relationship of
aggregated council member perceptions of council processes as predictive of aggregated
council member perceptions of council effectiveness). The latter is of greater interest
given that this study is concerned with explaining variability in effectiveness across
councils. The hypotheses examining correlates of the breadth of council goals

(Hypotheses 1b, 2b, 3b, and 4b) and the degree to which these goals have been met

57

(Hypotheses 1c, 20, 3c, and 4c) were examined only at the council level (i.e., group level)
given that these data were gathered only from key informant interviews and do not reﬂect
multiple perceptions. Finally, council characteristics were examined in an exploratory
fashion except for the analyses concerning the presence of external support for council
activities (Hypotheses 5a, 5b and 5c).

Finally, to explain variability in the perceptions of individuals within the same
setting the following question was addressed: 6) To what degree does organizational role
explain variation in council members’ perceptions of council effectiveness and council
climate (conﬂict resolution tactics, shared power in decision-making, leadership and
shared mission)?

In order to investigate the issues that might emerge in the process of developing a
coordinated community response to violence against women, this study employed both
survey research methods and semi-structured interviews with key informants. Research
questions one and two were addressed utilizing descriptive data from both closed-ended
and open-ended questions regarding the goals and activities of councils. Research
questions four through six were explored utilizing both hierarchical linear modeling

(HLM) and standard regression analyses.

58

METHOD

This study employed quantitative methods to address the research questions posed.
To maximize the degree to which this study would be relevant and useful to domestic
violence coordinating councils two steps were taken. First, the instruments in this study
were developed in conjunction with the primary domestic violence trainer from the
Violence Against Women Project of the Prosecuting Attorneys Association of Michigan
(PAAM) and the director of the Domestic Violence Prevention and Treatment Board
(DVPTB), who were interested in learning more about how councils can be supported in
facilitating a coordinated community response to violence against women. These key
informants contributed to both the content of the instruments and the phrasing of speciﬁc
items. Second, two current council members from different counties completed the survey
prior to distribution and edits were made based on their feedback.

Sample
Domestic Violence Coordinating Councils

This study included council members from domestic violence coordinating councils
across the state of Michigan. Of the 44 councils which were approached all but one chose
to participate. The participating councils varied in size from 8 to 116 members and council
‘age’ ranged from 7 months to 192 months (16 years) with an average of 64 months (5
years, 4 months). Councils were most likely to be chaired by an employee of the domestic

violence Shelter program (3 6%), followed closely by prosecuting attorneys (25%). Other

59

council chairs represented a range of organizations including law enforcement (8%),
victims’ advocates from the prosecuting attorney’s ofﬁce (6%), community based human
service agencies (6%), the judiciary (3%), local government including the county or
mayor’s ofﬁce (3%), probation (3%), someone hired specifically to direct the council (3%)
and survivors (3%). In some cases, councils were co-chaired; most often including a
combination of judges or prosecuting attorneys with shelter program staff.

The organizations included on each council varied; however, some organizations
were represented on the majority of councils. Of the 41 councils from whom this
information was gathered, 100% included representatives from the local domestic violence
programs, 100% from law enforcement (e. g., police, probation ofﬁcers), 88% from the
prosecuting attorney’s ofﬁce, 88% from batterer’s intervention programs, 80% from the
district court (e.g., a judge, magistrate or clerk), and 78% from health care organizations.
In addition, some included legal aid (60%), mental health services (59%), social services
agencies (58%), child protective services (51%), religious organizations (46%), the
circuit court (42%), educational institutions (39%), and local businesses (15%). Very few
councils included a domestic violence survivor representative (29%).

Of these councils, 38 of 43 met the criterion (i.e., representation across at least
two sectors, e. g., criminal justice and human service, surveys returned from at least three
council members with representation from at least three roles, e.g., prosecuting attorney,
probation officer and domestic violence service provider) and had adequate data for

inclusion in the HLM analyses proposed in this study. In addition, 40 of 43 councils had

60

adequate data to be included in analyses utilizing key informant data and aggregated
perceptual data to examine differences across councils.‘S
Council Members

Council members from each council were surveyed via mail (see Appendix A for
the council member survey).7 Survey data was collected from 511 council members.
Individual council return rates ranged from 14% to 75% with an overall return rate of
35%. Across councils, the return rate for members who participated at least sporadically
was 42% and 51% for members who attended meetings regularly. These ﬁgures are
important because many councils membership lists included individuals who had only
peripheral involvement with council activities or no involvement at all, which deﬂates the
overall return rate. Still, the surveys of all council members who responded were included
given that this study was concerned with capturing a broad range of viewpoints to assess
the collaborative processes and effectiveness of each council.

Thirty-eight percent of council members were between the ages of 40 and 49, 26%
were between 50 and 60, 22% were between 30 and 39, 9% were 20 to 29 and 5% were

over 60. The majority of participants were female (62%) and white (90%). Minority

 

6

Two councils with an N of 2 were included in these analyses. The response rates for these
councils (14% and 20%) were not appreciably smaller than those for other councils with a
larger N which were included in the study. Thus, these organizations were retained in the
standard regression analyses, but not the HLM given the small Level I N. Analyses
including these organizations did not differ appreciably when these councils were
removed.

7Please note that this survey includes items that were not utilized in the current study.

61

groups were not well represented in this sample; ﬁve percent of participants were African-
American, 2% Latino, 1% Native American and less than 1% Asian-American. Over half
of this sample had over six years of experience in the domestic violence arena (52%),
while 7% had less than a year of experience. Eighty-four percent had received training in
the area on domestic violence.

Participants represented a variety of organizational roles and sectors. The largest
portion of respondents were from the human service delivery system (including physical,
mental health and social services; 23%), followed by law enforcement (19%), domestic
violence service providers (18%), prosecuting attorneys (7%), probation ofﬁcers (6%),
judges (4%), community members (4%), court staff (4%), school administrators and
educators (3%), legal aid (3%), religious Ofﬁcials (2%), victims’ advocates from the
prosecuting attorney’s or sheriffs office (2%), batterers’ intervention (1%) and others
(5%).

Council Leaders and/or Conveners

The council member responsible for convening or chairing each council (e. g.,
managing meetings, creating schedules) was interviewed at greater length regarding
council characteristics, activities, and history (e. g., the structure of the council, the history
of membership; see Appendix B for the key informant interview). A key informant
interview was completed for 41 of 43 councils. Usually, the key informant was a present
or former chair of the council (56%) or the individual primarily responsible for council

coordination (27%). In ﬁve cases ( 12%) the key informant was a long time member

62

because there was no formal chair or he/she was not available. In most cases, the key
informant was a council member who played a primary role in the operation of the council
(e.g., convening meetings) and had a historical perspective regarding the council’s
functioning (he/she must have been a council member for at least one year or since the
council’s inception if it was formed less than one year ago). However, in two cases (5%)
the key informants had been council members for less than one year, but were the current
coordinators and the only members available to be interviewed.

Key informants represented a variety of organizations: 56% were employees of the
local domestic violence shelter program; 20% were prosecuting attorneys or assistant
prosecuting attorneys; 7% were victim’s advocates from the sheriffs or prosecuting
attorney’s office; 7% were council staff or directors; 5% were law enforcement; 2% were
probation Officers; 2% were from a community-based organization.

Procedures

The procedures in this study included ﬁve phases of data collection: a) recruitment
of councils and key informants, b) recruitment of council members within each council and
survey distribution, c) survey collection, (1) key informant interviews and d) follow-up.

Recruitment of Councilsand KeyInformgits

To begin the data collection process a letter was sent to council leaders and/or the
council convener (i.e., the person who manages the council), informing them about the
purpose of the study, the proposed survey, and requesting the participation of his/her

council and a complete member list so that surveys could be sent to council members (see

63

Appendix C for the letter sent to council leaders). To encourage participation, the
recruitment letter was sent jointly by this author and the community agency who requested
the development of the survey - the Prosecuting Attomey’s Association of Michigan
(PAAM). In the letter, the purpose of the study was detailed, including the desire to learn
more about: a) what councils are currently accomplishing in the State of Michigan, b) the
factors that contribute to the effectiveness of coordinating councils/task forces in
facilitating a coordinated community response, and c) what technical support coordinating
councils/task forces need to be effective. The letter explained that the survey would ask
council/task force members to reﬂect on their experiences working on the council,
including their perceptions of the practices, accomplishments and needs of the councinask
force. The council leader/staff person had the option of discussing the councils’
participation with the council. While this was not required due to potential logistical
difﬁculties, it was suggested in the recruitment letter.

If the leader or staff person did not respond to the letter by providing a staff list, a
phone call was made requesting participation, asking for a membership list and answering
any questions the council staff person or leader may have had (see Appendix D). If the
leader/staff person did respond to the letter by providing a staff list, a phone call was made
to answer any questions the council leader/staff person may have had (see Appendix E).

In both cases, if the council leader/staff person was willing to participate, he/she was
asked to indicate who the most appropriate person for the key informant interview would

be based on the criteria detailed earlier (i.e., the individual manages council activities and

64

has been with the council for at least a year or since its inception if less than a year). If the
leader/staff person was the most important individual an interview time was set for his/her
key informant interview. If not, the person identiﬁed by the leader/staff person was called
to request his/her participation and to set up an interview time if they were willing to
participate (see Appendix F).
Recruitment of Council Memﬂs

After the council staff or leader agreed to participate and provided a membership

list, council members were mailed a survey and invited to participate in the study.
Survey Collection

Surveys were mailed to participants with self-addressed, postage paid envelopes

included. They were asked to return the survey via mail within two weeks Of receipt.
Kev Infonmant Interviews

Each key informant was called at the time agreed upon during recruitment. Key
informants were also surveyed and key informant interviews were not conducted until
after the key informant had ample time to return his/her survey if he/she planned to
complete it. If they had not completed their survey at the scheduled interview time and
planned to complete it, a new time and date for the interview was scheduled. Each key
informant interview took approximately one hour.

Follow-up
On the survey due date, a postcard was mailed including a telephone number

participants would call if they required a second survey (see Appendix G). One to two

65

weeks after this postcard had been mailed follow-up calls began for councils who included
telephone numbers on their membership lists (see Appendix H). Unfortunately, almost
half of the councils did not include telephone numbers. While some councils could easily
send this information or access it when it was requested, others said it was not information
they had recorded anywhere and could not be easily accessed. Efforts were made to ﬁnd
telephone numbers using the intemet, but this was not very productive. Ultimately, calls
were made to 29 of 43 councils. During the call, council members were informed that the
due date had been extended, and that they could still return the survey by the new deadline
(see Appendix F). They were also asked if they needed another survey mailed or faxed to
them. The new deadlines were set to allow two weeks for the return of the survey. Given
that not all individuals could be called, a second postcard with a ﬁnal due date was mailed
to council members identiﬁed as at least sporadic attendees at meetings (limited resources
did not permit mailihg the second card to all members).

Given the focus of this study on potential differences among council members who
have different organizational roles or who represent different sectors (i.e., criminal justice,
domestic violence advocacy, health care), follow-up calls were continued until a
representative sample was drawn from the majority of councils (95%) and it was clear that
such a sample could not be drawn from other councils (e. g., everyone from a given sector
chose not to participate even after follow-up efforts were made). For the purposes of this
study, a representative sample consisted of at least two sectors (e. g., criminal justice and

domestic violence advocacy) and at least three organizational roles (e.g., domestic

66

violence advocate, judge, prosecuting attorney or health care provider, educator,
probation officer)" Thus, ﬁnal follow-up efforts focused on achieving such a
representative sample. Accordingly, a second set of follow-up telephone calls was made
that focused Speciﬁcally on particular groups of council members (e. g., domestic violence
advocates and prosecuting attorneys). In addition, a second set of surveys was mailed to
domestic violence advocates who were not represented across a number of councils.
Measures
Council Member Survey

Council Member Demographic and Descriptive Information

Demographic data were collected on the council members’ age, gender, race,
organizational role, number of years/months they have been working on domestic violence
issues, the number of years/months they have been a member of the council, their type and
degree of participation, and the service sector and organization they represent. Type and
degree of participation and the service sector represented by cOuncil members was
measured using items from Florin’s (1996) Task Force Member survey. These data were
collected for descriptive purposes. However, a series of dummy variables was created to
capture organizational role to model differences in perceptions of council effectiveness and

climate across council members.

 

8This was the minimum criterion to be included in the HLM analyses.

67

Cguncil Effectiveness

A ten-item scale constructed Speciﬁcally for this Study was utilized to assess
council effectiveness in meeting their goals and the goals of a CCR. Utilizing a six-point
Likert type scale (1 = not at all true, 6 = very true) council members and key informants
were asked to indicate their agreement with a series of statements regarding the degree to
which the council has met its goals and the goals of a CCR (e.g., “The council has
addressed shortcomings in practices in community organizations regarding their response
to domestic violence”), the degree to which the council has met intermediary goals (e. g.,
“The council has increased members’ knowledge about each other’s organizations”) and
the degree to which the council has met its goals in general (e. g., “The council has
consistently met its goals”). Cronbach’s alpha for the overall effectiveness scale was .96,
with item-total correlations ranging from .57 to .86.

To explore the underlying structure of this scale, an exploratory principle
components analysis was performed. This analysis yielded two factors; one factor
included nine items regarding the degree to which the council has met its goals (alpha =
.95; item-total correlations ranged from .58 to .89) and the second factor included six
items regarding the degree to which the council was meeting intermediary goals (alpha =
.93; item-total correlations ranged from .60 to .86). The overall scale score9 was utilized

for the purposes of this study given its high internal consistency and the goal of examining

 

9

Unless otherwise speciﬁed, all scale and subscale scores were constructed by calculating
the mean.

68

overall effectiveness. However, subscales were included in some analyses to detemrine
the degree to which these aspects of perceived council effectiveness were differentially
related to the independent and dependent variables explored in this study.
Conﬂict Transformation

An eight-item scale constructed speciﬁcally for this study was utilized to assess the
degree to which the council handles conﬂict effectively. Utilizing a six-point Likert type .
scale (1 = not at all true, 6 = very true), council members and key informants were asked
to indicate their agreement with a series of statements regarding the tactics the council
uses to handle emergent conﬂict (e.g., “Conﬂicting Opinions among council members have
led to needed changes in the council”). These items were constructed to reﬂect tactics
employed by collaborative bodies to handle conﬂict. This set of items included: a) two
tactics that are somewhat neutral mechanisms for handling conﬂict including
compromising, (e. g.,“Disagreements among council members are Often resolved by
compromise” ) and ‘smoothing over’ the conﬂict (e.g., “When conﬂict arises, council
members agree to disagree”), b) three tactics that are considered ineffective at handling
conﬂict, including ignoring or avoiding the conﬂict (e. g., “When conﬂict arises the council
ignores it”), and c) three tactics that are considered effective in handling conﬂict (e. g.,
“The council has handled conﬂict by attempting to get to the root of the disagreement”)
(based on Bitter, 1990; Faragason, 1994).

To explore the underlying structure of this scale, an exploratory principle

components analysis was performed. This analysis yielded two factors; one factor

69

included six items regarding the degree to which councils have handled conﬂict with
tactics that are considered neutral or constructive (e. g., conﬂict has led to effective
problem solving; alpha = .78; item total correlations ranged from .35 to .67) and the
second factor included three items regarding the degree to which councils have handled
conﬂict with tactics that are considered ineffective (e.g., when conﬂict arises the council
ignores it; alpha = .73; item-total correlations ranged ﬁ'om .25 to .33).

To reﬁne the climate measures and reduce redundancy in each of the scales an
exploratory principle components analysis including all items from all scales (i.e., shared
power in decision-making, leadership, shared mission) was performed. This resulted in a
change to the ‘ineffective’ conﬂict resolution scale; one item from the shared power in
decision-making scale was added given its conceptual relevance (“council decisions are
dominated by a few members”) and two items from the shared mission scale were added .
(e.g., council members disagree about what direction the council should take”).
Cronbach’s alpha for this six-item subscale was .73 with item-total correlations ranging
from .49 to .59. Given the focus of this study on the use of effective conﬂict resolution,
the constructive conﬂict tactics subscale was utilized to model variation in council
effectiveness and the subscale focused on ineffective strategies was used for descriptive
purposes.

Power\Inﬂuence in Decision-Mm
A six-item scale constructed speciﬁcally for this study was utilized to measure the

degree to which council members perceived the council as Sharing power in decision-

70

making across its membership. Utilizing a six-point Likert type scale (1 = not at all true, 6
= very true), council members and key informants were asked to indicate their agreement
with four statements regarding the degree to which the council as a whole included all
members in the decision-making process of the council (e. g., “The input of all council
members inﬂuences the decisions the council makes”).

To explore the underlying structure of this scale, an exploratory principle
components analysis was performed. This analysis yielded only one factor. However,
given that the exploratory principle components analysis described above (i.e., including
all climate items) indicated that one shared power in decision making item (i.e., “Council
decision are dominated by a few members”) be included in the ‘ineffective’ conﬂict
resolution subscale it was not included in the ﬁnal shared power in decision-making scale.
Cronbach’s alpha for the ﬁnal ﬁve-item scale was .88, with item-total correlations ranging
from .62 to .77.

Council Leadership

To assess the effectiveness of council leadership a portion of Butterfoss’ (1993)
Coalition Effectiveness Inventory was utilized. This leadership index measures multiple
aspects of effective leadership including, for example, effective management of meetings,
commitment to the council’s mission, knowledge of domestic violence, and competence in
problem solving and resolving conﬂict. Utilizing a six-point Likert type scale (1 = not at
all true, 6 = very true), all council members (including the key informant even if they are

the council leader) were asked to indicate their agreement with each item.

71

To explore the underlying structure of this scale, an exploratory principle
components analysis was performed. This analysis yielded two factors; one factor
included ten items regarding the degree to which leaders had inclusive leadership skills
(e. g., including and engaging all stakeholders; alpha = .96; item total correlations ranged
from .73 to .88) and the second factor included 4 items regarding the degree to which
leaders had instrumental leadership Skills (e.g., organizing meetings; alpha = .87; item-total
correlations ranged from .65 to .79). Given the conceptual differences between these
factors, these subscale scores were utilized separately to explain variation in council
effectiveness. Cronbach’s alpha for the overall scale was .96 with item-total correlations
ranging from .64 to .86.

Presence of a Shared Mission

A six-item scale constructed speciﬁcally for this study was utilized to assess the
degree to which a shared mission was present on the council. Utilizing a six-point Likert
type scale (1 = strongly disagree, 6 = strongly agree) council members and key informants
were asked to indicate their agreement or disagreement with a series of statements
regarding the degree to which the council had a shared mission. These statements
included four general items (e. g., “The council has developed a shared mission.” and
“Council members are working together to achieve a common goal”) and two items that
were speciﬁcally related to domestic violence issues (e. g., “Council members have a
shared understanding of domestic violence.” and “Council members have a Shared vision

regarding what changes are needed in the community response to domestic violence”).

72

To explore the underlying structure of this scale, an exploratory principle components
analysis was performed. This analysis yielded only one factor. However, given that the
exploratory principle components analysis including all items ﬁ'om all climate scales
(described above) indicated that two items be included in the ‘ineffective’ conﬂict
resolution subscale (e. g., “on the council special interests get in the way of having a Shared
mission”) these items were removed from the ﬁnal shared mission scale. Cronbach’s alpha
for the ﬁnal four-item scale was .89, with item-total correlations ranging from .69 to .81.
Aggregating Perceptions at the Council Level

The predictors described thus far (e. g., perceptions of council leadership) have
been operationalized at the individual level as council members’ perceptions of various
aspects of the climate of the council. However, in order to compare perceived
effectiveness across councils (i.e., as a proxy for council effectiveness) and to examine the
factors that explain differences across councils, for each of the indices described above a
Level II aggregate score was created for each council. For example, council members’
perceptions of the degree to which the council has a shared vision was averaged for each
council. Creating these aggregate scores was justiﬁed because a signiﬁcant portion of the
variance existed between councils (i.e., see Table l for a summary of the intraclass
correlations for each variable). In addition, James, Demaree and Wolf’s (1984) rwgm was.
calculated to assess within-group agreement. These analyses demonstrated adequate
agreement across settings (i.e., > .70) for shared power in decision-making (.82), conﬂict

resolution (.73), shared mission (.77), instrumental leadership (.82), and inclusive

73

leadership (.83) justifying taking the mean to represent the setting climate.lo These
aggregate scores were included in subsequent hierarchical linear modeling analyses to
model variability in the average perceived effectiveness of councils as well as standard
regression analyses explaining variability in the breadth Of goals addressed and the degree
to which they were accomplished.

Table 1

Intraclass Correlations for Dependent and Independent Variables

 

 

 

Variable Name Variance Variance Within ICC p-value
Between (Too) (0')

Perceived Effectiveness .19 .75 .20 .000
Instrumental Leadership .21 .84 .20 .000
Shared Mission .15 .65 .19 .000
Inclusive Leadership .14 .73 .16 .000
Conﬂict Resolution .11 .76 .13 .000
Shared Power in .08 .78 .09 .000
Decision-Making

 

 

10

Two organizations were excluded from the analyses of within group agreement given their
small N of 2. In addition, some organizations did not have agreement above .70 on all
elements of the council climate, but were retained because on some they had adequate
agreement

74

Key Informpnt Interview

Qppnpil Demographic and Descriptive Information

Information was collected regarding when the council was formed, council
membership and participation, whether the council had a paid staff person and the amount
of ﬁnancial resources the council had.
Assessing Council Goals and Accomplishments

The breadth of goals addressed and the degree to which goals were accomplished
were assessed via the key informant interview. More speciﬁcally, key informants were
asked about a series of issues (12 total; 6 criminal justice goals and 6 goals beyond the
criminal justice system) that are Often associated with reforms needed to achieve a CCR
(Clarke et al. 1996; Hart, 1995; Pence and Shepard, 1999). Sample issues include: a)
improving access to personal protection orders, b) implementing early identiﬁcation
policies in emergency rooms, c) providing community education for local human service
providers who serve battered women, and d) reforming arrest policies. For each item, the
key informant was asked whether or not addressing this issue was a goal of the council
and the degree to which each goal was accomplished (1 = not at all accomplished; 6 =
completely accomplished). If they had not addressed the issue, they were asked if
addressing the issue was a need in their community.

This information was used to describe the degree to which councils are addressing
issues consistent with the aims of a CCR. To accomplish this, two scores were created to

represent the breadth of council goals addressed: a) within the criminal justice system and

75

b) beyond the criminal justice system. Each score was created to reﬂect the proportion of
needs that a council addressed relative to the needs in their community. Speciﬁcally, the
total number of areas a council had addressed was divided by the total number of areas
that needed to be addressed in their communities. Calculating a proportion takes into
account that needed reforms vary across communities; therefore councils were not
penalized for failing to address an issue that was not a need in their community. Finally,
two means were calculated to assess the degree to which councils accomplished their
goals to create reform a) within the criminal justice system and b) beyond the criminal
justice system.
fwdth of Council Membership

Breadth of council membership was assessed by asking key informants to indicate
whether or not a series of potential stakeholder groups were current council members and
whether or not they were active council members. This list included all stakeholder
groups thought to be central to the development of a CCR: domestic violence service
providers, judges, prosecuting attorneys, police ofﬁcers, probation officers, batterers’
intervention facilitators, social service providers, school administrators, local business
representatives, and clergy. Two sum scores were calculated: 1) the total number of
official members (i.e., stakeholder groups) and 2) the total number of members who were

actively participating in council activities (i.e., whether the stakeholder group/organization

76

sends someone to meetings regularly and participates in council activities)" Calculating
both scores was important because some organizations are members “on paper,” but do
not participate in council activities. Both of these scores were utilized to explain variation
in effectiveness across councils. In addition, a dichotomous score was created to indicate
whether or not survivor representatives were included in the council for descriptive
purposes.
Presence of External Support

To assess the degree to which each council receives support from the external
environment, key informants were asked to indicate the degree to which they felt a number
of key stakeholders (e. g., organizational leaders, funders, local policy makers, and
community members) were supportive of council activities. Key informants responded
utilizing a six-point Likert type scale (1 = not at all supportive, 6 = very supportive). The
mean of these items was calculated for each council as a measure of the external support
received by councils. Cronbach’s alpha for this six-item scale was .73 with item-total

correlations ranging from .37 to .63.

 

11

These scores reﬂect the number of members at the stakeholder group level rather than the
individual level. For example, domestic violence service providers were considered a
stakeholder group and counted once even when there were multiple domestic violence
providers who were oﬂicial members of the council. Thus, this provides a measure of the
breadth of official and active membership across stakeholder groups rather than the depth
of membership within any given stakeholder group.

77

Cguncil Structure

To assess the degree to which councils have formalized Structures in place, two
segments of Butterfoss et a1. (1993) Coalition Effectiveness Inventory (CEI) were utilized
(coalition structures and coalition processes). Key informants were asked a series of 18
yes/no items to determine the degree to which the coalition has structures (e. g., bylaws,
mission statement in writing) and processes (e. g., mechanisms for decision-making and
problem-solving) in place. A mean score was created reﬂecting the degree to which the
council has formalized structures and processes in place. Cronbach’s alpha was .88 with
item-total correlations ranging from .00 to .82.‘2

Data Analysis Strategy

To examine the research questions posed in this study quantitative analyses were
employed. Given that council members were nested within councils, hierarchical linear
modeling (HLM) was utilized to examine relationships among independent variables
measured via council member and key informant perceptions and perceived council
effectiveness. More speciﬁcally, HLM was utilized to examine the relationship of
perceived council effectiveness and 1) effective conﬂict resolution, 2) shared power in

decision-making, 3) leadership quality, 4) the presence of a shared vision and 5) council

 

12

Three items with .00 item-total correlations had low variability, but given the purpose of
the scale was to assess degree of formality, these items were retained even if they were
endorsed by many organizations (e. g., Do you have regular meetings?) or only a few
organizations (e. g., DO you have mechanisms in place to encourage accountability of
member organizations?)

78

characteristics (e.g., age, breadth of membership). In addition, HLM was utilized to
separate individual level effects from group level effects and to account for variation in
perceptions across members in a single council by examining the relationship of their
perceptions to their organizational roles. Finally, standard multiple and hierarchical
regression was employed to examine the relationship of these same predictors to the
indicators of council effectiveness collected from key informants (i.e., the breadth of
council goals and the degree to which these goals were accomplished).

Council age was included as a covariate in all analyses given that it would be likely
to account for some of the variation in council effectiveness. Speciﬁcally, it was expected
that council age would be positively related to the indicators of council effectiveness.

One-tailed tests were used to examine all directional hypotheses while two-tailed
test were used to examine all exploratory research questions (i.e., the relationships of the
breadth of Oﬂicial and active council membership and the formality Of council structure to
council effectiveness).

Finally, it is important to note that in some cases (11 = 23) key informants also
chose to complete the survey. Their scores were retained in subsequent analyses given
that their perspective was considered an important part of the assessment Of the setting.
However, their scores were removed from the perceived effectiveness outcome measure
for the purposes of validating this indicator of council effectiveness with the others

collected only via the perceptions of key informants.

79

RESULTS

Univariate Analysis

Univariate analyses were performed (e.g., descriptive statistics, histograms,
ﬁequency distributions) to examine independent and dependent variables for skewness and
outliers at the council level and individual level (i.e., across all council members where
applicable). See Tables 2, 3, and 4 for a summary of descriptive statistics for independent
and dependent variables at both the individual and council levels. For the most part these
analyses did not reveal any extreme outliers or skewness in the data. However, the
breadth of goals addressed within the criminal was negatively skewed (skewness = -1.58).

Descriptives

Conﬂict Resolution

Overall, councils were viewed by their members as employing a moderate degree
of effective conﬂict resolution strategies. On average, various statements reﬂecting
effective conﬂict resolution tactics (e.g., “conﬂict has led to effective problem solving”)
were viewed as somewhat true of councils (Mean = 3.46, SD = .42). While effective
conﬂict resolution tactics were rated as somewhat true (51.2%) or mostly true (46.5%) for
the majority of councils, it is important to note that only a small portion (2.3%) rated the
statements as being true and none were rated as very true. In addition, members did not
view their councils as having appreciable amounts of conﬂict (in response to items such as

“conﬂict has emerged in the council’s attempts to work collaboratively”; Level 11 Mean =

80

 

 

81

8.0 cod ooh 8.0 00.0 cod 83::me
00.. 2: co; mN._ ow; cog Ens—£2
_:. mS. _Z. 0:. o:. o:. mmogoxmwostmﬁum
v2..- ozr wmor 03.. www.. Q19- 306585
2.2: memo. memo. Noon: Ema. Taco.— cosﬂgﬂ .Bm
200v mmmvd ovwmé owmmé whmhé ooomé :32
9%. >3» 5v oav 3v oov m
comm: cows—80M .838 3:32:84 938384 mmocezuootm
@225 86:00 332m 18:25:85 2522: 33080;

 

 

0?: cam “52599 .8 8:3anle 26 temlloa

 

N 033.

 

 

 

8.: 8.3 8.: mun «swam mhé Tram 3% New Esp—ca:
cod oo.m ooh mw._ 3M VON Sam 00m _w.m 82552
men. Elm. 31m. 06m. Gm. Em. Sm. Sm. Em. Hesse—m mo cotm .Em
Em. oovr moor _o_ .. 03.- mem. em _. _m _. 30.- 32.386
QOVM 8va hmmvd wmmo; 3:. 33. Scam. mam. mmmv. =2§>on .Bm
2:6 ommhw om”: owmo...V mmtlu mmmv.m mummé moové swine :82
2‘ CV ov :V mv mv 3 Q Q Z
saaslﬁe
23025 0384‘ 95.3802 tongsm coma: coca—08M 526A— quconmoA qiﬁeuﬂﬁ
zen—50 mo 532m we 532m Enowxm 3.8% 65:00 33.8 35833.:— 95205

 

 

= ~0>01~ an mO—n—wtdxl «ﬁe—vac 0mg: ho..— momuwmuaum o>mu COmQD

m 2an

82

 

 

 

 

83

8.. 8.... 8.. 8.. 8.. 8.. 8.. case...
8.. SN 8.. 8.. t. 8.. 2.. 52......2
8.. 8.. 8.. 8.. 8.. 8.. 8.. .858... .0 .2... .2.
2...- N. . .- .8- 8.. E. .- 8...... S..- .85....
.8... 5... .8. as... N... 8:... m. .n 8.88.. a.
a... .. 8.... .8. 8:... 2... 8.... 2...... :82
2. K .N .N .. .4 a. m
8...... 8...... .885 8...... 8...... 8.....5
8...... .885 5......» 8%... 8.88.... 8:55 a. 8.8.8...

.885 888.... 8......Eo8< .888. 88.2 8...... are... 8.82 280 8...... 882.85

88:38.. .80 .5 .o 8882.. .5 n .o 8882.. 8.....5 .. 8.88..

 

 

= .054 85515.3 €09.33. 8.. 8.38% «3.9.8on

v 03....

2.48, SD = .56), and few viewed their councils as employing ineffective conﬂict
resolution strategies (e. g., “when conﬂict arises the council ignores it”; Level 11 Mean =
1.85, SD = .47). This suggests that most councils were perceived by their members as
facing relatively small amounts of conﬂict and as employing effective conﬂict resolution
strategies to some degree.
Sigred Power in Decision-Making

For the most part, councils were rated overall as sharing power in decision-making
to a moderate degree. On average, various statements reﬂecting shared power in
decision-making (e.g., “the input of all active council members inﬂuences the decisions the
council makes”) were viewed as mostly true of councils (Mean = 4.34, SD = .39). In
general, shared power in decision-making was rated as mostly true (62.8%) or true
(37.2%) for the majority of councils while none were rated as very true. This indicates
that the majority of councils were viewed by their members as having shared power in
decision-making to some extent, but there was room for growth in this area.
Leadership

On average, various statements reﬂecting quality of leadership (e. g., “council
leaders plan meetings eﬂ‘ectively and efﬁciently”) were viewed as mostly true of councils
(Mean = 4.67, SD = .46). For the most part quality of leadership was rated as mostly true
(32.6%) or true (62.7%) of councils with a few rated as very true (4.7%). This indicates

that for the most part councils were viewed as having quality leadership.

84

Shared Mission

On average, various statements reﬂecting the presence of a shared mission (e. g.,
“my council has developed a mission that is shared and supported by all council
members”) were viewed as mostly true of councils (Mean = 4.48, SD = .48). Speciﬁcally,
the presence of a shared mission was rated as mostly true (39.5%) or true (55.8%) of
councils with a few rated as only somewhat true (4.7%). Councils were viewed as having
a shared mission to a large extent. However, a signiﬁcant proportion of councils were
rated on average as only slightly above the mid-point of the scale indicating room for
growth in this area for many councils.

Breadth of ‘Oﬂicial’ and ‘Active’ Membership

 

Councils had a broad range of ‘official’ members (i.e., they were on the
membership list), but not all participated. On average, councils had an average of 11
official members ranging from 5 to 17, while the average number of active members was
slightly lower (9 members/council, ranging form 3 to 14). Not surprisingly, domestic
violence service providers were the most likely to participate in council activities (100%
of councils reported active participation), followed by law enforcement ofﬁcials (84%),
batterers’ intervention service providers (74%), prosecuting attorneys or assistant
prosecuting attorneys (70%), health care providers (61%), and victim’s rights advocates
of the prosecuting attorney’s office (61%). See Table 5 for a complete summary of

ofﬁcial and active membership.

85

Table 5

Percent of Councils with ‘Oﬂicial’ and ‘Active’ Membership from Each Stakeholder

QOUQ

 

 

Member Organization ‘Ofﬁcial’ ‘Active’
Membership Membership
Domestic Violence Service Provider 100% 100%
Law Enforcement Ofﬁcial 100% 84%
Prosecuting Attorney or Assistant 98% 70%
Prosecuting Attorney
Probation Ofﬁcer 88% 63%
Victim’s Rights Advocate in the 87% 61%
Prosecuting Attomey’s Ofﬁce
Batterers’ Intervention 85% 74%
District Court Representative 82% 54%
Health Care Provider 78% 61%
Legal Aid Attorney 62% 37%
Mental Health Provider 60% 44%
Social Service Provider 57% 37%
Child Protective Service Worker 53% 35%
Clergy 45% 28%
Circuit Court Representative 58% 42%
School Administrators or Educators 38% 21%
Survivor Representative 29% 0
Local Businesses Representative 15% 9%

 

*This item was not asked regarding survivor representatives.

Support for Council Activities

On average, key informants rated various key players, such as local policy makers
and funders, as being somewhat supportive of council activities (Mean = 4.10, SD = 1.03).
It is interesting that while 68% experienced these key players as at least somewhat
supportive of council activities, 32% of councils described key players as at least
somewhat unsupportive. This suggests that overall councils receive a moderate amount of
support from leaders, policy makers and ﬁinders, but even this level of support is not
uniform across councils.
Formality of Council Structure

On average, councils implemented 7.34 of 18 formal structures and/or processes.
Some structures and processes were employed by nearly all councils. For example, 100%
had regular meetings, 97% used an agenda, 92% kept minutes, 85% had subcommittees,
76% had a mission statement in writing and 63% had objective and goals in writing.
Other structures and processes were employed by a smaller number of councils. For
example, 34% councils had a formalized decision-making process, 32% of councils had an
organizational chart, 29% had written bylaws, 24% had a process in place to orient new
members, 23% had job descriptions in writing, 22% had a core planning group, and 21%
had a formalized problem-solving or conﬂict resolution process and 20% had a process to
train new and old members. Very few councils had any mechanisms for holding members
accountable for completing assignments in a timely manner (15%) or holding member
organizations accountable (5%). No council had an advisory board made up of domestic

violence survivors.

87

Council Goals

To address Research Question 1 (To what degree do councils address a broad
range of goals reﬂecting issues relevant to the development of a CCR?), descriptive
statistics were employed (see Table 4 for a summary of these ﬁndings). Councils
addressed a somewhat broad range of goals both within and beyond the criminal justice
system. On average, councils addressed 7.95 (SD = 2.34) of 12 possible goals. However,
councils were addressing a greater number of reforms regarding policy and practice in the
criminal justice system (Mean = 4.54 of six possible goals, SD = 1.42; 84% of reforms
needed in a given community on average) than reforms beyond the criminal justice system
(Mean = 3.41 of six possible goals, SD = 1.45; t = 3.87, d.f. = 40, p < .001; 68% of
reforms needed in a given community on average). Frequencies were performed to
describe the areas in which councils are setting goals. Within the criminal justice system,
93% identiﬁed weaknesses or holes in the criminal justice system, 83% of councils
addressed access to and enforcement of personal protection orders (PPO), 78% addressed
reforming arrest practices, 74% addressed altering sentencing practices, 68% addressed
prosecution practices, and 61% addressed the processing of court cases. Beyond the
criminal justice system, 90% conducted training for the community or key stakeholder
groups such as health care professionals, clergy or law enforcement, 76% supported or
developed batterers’ intervention programs, 49% addressed the need for early
identiﬁcation in the health care system, 48% worked on improving relationships between
child protective services and domestic violence advocates, 39% worked on making it

easier for battered women to access needed community resources, and 39% worked on

88

developing new services for battered women or their children.
Accomplishment of Goals

To address Research Question 2 (To what degree do council leaders perceive their
councils as achieving their goals?), descriptive statistics and frequencies were performed
(see Table 4 for a summary of the descriptive statistics). Key informants rated their
councils’ accomplishment of goals as moderate overall (Mean = 3.86, SD = .82 on a scale
of one to six). Key informants reported council goals to reform the criminal justice system
as slightly less accomplished (Mean = 3.84, SD = .95) than those to reform areas beyond
the criminal justice system (Mean = 4.13, SD = 1.06), however, these means were not
signiﬁcantly different ( t = -1.63, d.f. = 40, p = .1 l).
Perceived Effectiveness

To address Research Question 3 (To what degree do council members perceive
their councils as effective in facilitating reforms in policy and practice (i.e., CCR goals)
and meeting intermediate goals?) a series of descriptive analyses were performed. On
average, various statements reﬂecting council effectiveness (e. g., “inﬂuenced changes in
practice in community organizations that have increased women’s safety” and “increased
member’s knowledge about each other’s organizations”) were rated by members as mostly
true of councils (Mean = 4.35, SD = .53). Councils in general were viewed as less
effective at meeting CCR goals (e. g., changing policies and practice that encourage
women’s safety; Mean = 4.27, SD = .58) than intermediate goals (e. g., getting people
talking about domestic violence; Mean = 4.50, SD = .52; t = -4.46, d.f. = 40, p < .001).

While statements describing council effectiveness were rated by members as mostly true

89

(48.8%) or true (41.9%) for the majority of councils, it is important to note that for a
small portion of councils (7%) these statements were rated on average as being only
somewhat true, and for very few (2.3%) as very true. This indicates that the majority of
councils are viewed as mostly effective in meeting intermediate goals (getting people
talking) and the goals of a CCR (inﬂuencing policy regarding the organizational response
to domestic violence), but not very effective.
Validating Indicators of Council Effectiveness

To validate and examine the relationships among indicators of council
effectiveness, a series of Pearson correlations were performed (see Table 6 for the
intercorrelation matrix of dependent variables). Examination of the intercorrelation matrix
indicates the indicators of council effectiveness have robust relationships with one another;
however, these relationships were not uniform. The breadth of criminal justice goals (i.e.,
the proportion of needs addressed in each community) addressed by councils was
positively related to their average perceived effectiveness with regard to meeting the goals
of a CCR such as reforming policies and practices in the criminal justice system (a subset
of 7 items including, “The council has inﬂuenced the policy of organizations regarding
their response to domestic violence;” r = .30, p = .06). However, the breadth of criminal
justice goals addressed by councils was not signiﬁcantly related to their average perceived
effectiveness in meeting intermediary goals such as improving relationships among
stakeholders (a subset of 8 items including, “The council has increased members’
knowledge about each other’s organizations;” r = .18, p = .27) or with their overall

perceived effectiveness (r = .26, p = .1 1). This indicates that councils who addressed a

90

 

3. 3. 3. 3. 2. 3. 3. Z

8.... .8. .8. .8. .8. 8... 8.8-..
605...... 8.8.8.8....
8.... ml... 8...- 8.. a a. alum. .85 8...... 8.85 80.8.. .
......................... .. 8.8.8.8....82
8.. .8. 8... 8... 8... 8.9-..
63.88.8828...
8.... a... a. 8.8 8% mm. .85 8...... 8.....5
.................................................. .. ...........:.z
:8. 8... N... .2. 8.9-..
.888... 8...... 8.8....
8.... RN. .8. .8. e... .88.. 8.8.8... .8... .o 58....
............................................................................. ............z
8.. .8. 8.. 8.8-..
3.5.0.8 8.8.83.
8.... E. 8% .5. .85 8...... .88.... .8 £88..
8 ................. m .82 ......................................................................
8... 8... 2.8.-..
Ezra... 858.5
8.... a.“ 8% 888.58.... a. 8883.85 8288..
......................................................................................................................... m. .82
8... 2.8.-..
359......
8.... mg 888.5 50 a. 8.83.85 8288..
................................................................................................................................................. m .2
88.... ESE... .885 88828.... 8.8....

 

00<~0m 003.0 4<00n0m 4.8.00.0 Emu—m 1200me gob—mm

 

.mo:o>..oo.mm .6500 «o £386... .0 5.82 :o....._otoo.o.£

 

m. 8.89

greater proportion of the needed criminal justice reforms in their communities were more
likely to be viewed as effective with regard to meeting changes in policies and practice
consistent with increasing women’s safety and batterer accountability. However, the
breadth of goals addressed beyond the criminal justice system was not signiﬁcantly related
to overall perceived effectiveness (r = .12, p = .47) or either of its subscales (CCR goals, r
= .09, p = 57; Intermediate Goals, r = .08, p = .61). This suggests that councils who
addressed a broader range of goals outside of the criminal justice system were no more or
less likely to be viewed by their members as meeting their goals and the goals of a CCR.

The degree to which criminal justice goals were accomplished was signiﬁcantly
positively related to the overall perceived effectiveness of councils (r = .58, p < .001) as
well as the CCR goals subscale (r = .62, p < .001) and the intermediate goals subscale (r =
.45, p < .01). The degree to which councils accomplished goals beyond the criminal
justice system was also positively signiﬁcantly related to the council members’ rating of
the perceived effectiveness of councils overall (r = .57, p < .001) in addition to each
subscale (CCR goals, r = .57, p < .001; Intermediate Goals, r = .49, p < .01). This
suggests the degree to which goals were accomplished was consistently related to the
perceived effectiveness of councils.

While the breadth of goals addressed in the criminal justice system was
signiﬁcantly related to the degree to which criminal justice goals were accomplished (r =
.52, p < .01), it was not related to the degree to which goals were accomplished outside of
the criminal justice system (r = .14, p = .39). This suggests that councils who addressed a

greater number of areas within the criminal justice system were also accomplishing these

92

goals to a greater degree. Further, the breadth of goals addressed beyond the criminal
justice system was not related to either the degree to which goals were accomplished
beyond the criminal justice system (r = -.08, p = .63) or within the criminal justice system
(r = .17, p = .30). Finally, the degree to which criminal justice goals were accomplished
was signiﬁcantly positively related to the degree to which goals were accomplished
beyond the criminal justice system (r = .38, p < .05). This indicates that councils who
were successful in achieving their goals in one domain were successful in other domains as
well. Considered together, this pattern of relationships suggests that these indicators of
effectiveness are related to one another. The breadth of goals addressed had the weakest
link with the other indicators of effectiveness. This is not surprising, however, given that
simply identifying an area of need does not insure councils can effectively accomplish what
they set out to do.
Intercorrelations Among Independent and Dependent Variables

The intercorrelation matrix of independent variables indicated that all measures of
council climate were signiﬁcantly related to one another (see Tables 7 and 8 for the Level
I and Level II independent variables intercorrelation matrices) at both Level I and Level II.
In addition, at Level 11 external support was strongly related to the perceived climate
variables. While the proportion of active participation was signiﬁcantly related to external
support it was not related to any of the perceived climate variables. Similarly, the breadth
of ‘oﬂicial’ membership, and the formality of council structure were not correlated with
other independent variables with one exception; these variables were positively

signiﬁcantly related to one another. See Tables 9 and 10 for a summary of the zero-order

93

 

03. w...» 2% vwv m wv Z

8... 8... 8... 8.. 02?-..
8.. 3w m...l a. a 9.88.8.2 88.5
.83... ........... a... ............. a... .............. 7... .......................................................................................
8... 8... 8... 8....-.
8.. $8 8". ﬂ... .8. 8...... 8...... 5.8.6.. 8......
.33....37. ......................................................................................
8... 8... 8.9.-..

8.. Nola. a. 9.... 8.3.0.8. .288 2.8....
8.. ............ .. mm ............... 7... ......................................................................................
coo. 02.3-..

8.. 8% 3.82.. 8.8.88. 3.883....
...................................................................... .87.
8.. 352.. .2288. 3.8.2..

 

2m mm M0 H57: .302.

 

 

 

_ 654 a 83am; Eoqulco 2.... we; x2... 82 5.3.8.88...—

h 2an

94

 

Z
2.37..

.8. 232:5 =82... .0 boss...

I0.0IOOIIOIIIOCIOOIIOIIIO' IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

33>-..
.2... 95.3502 .o 532..

0'00...II0notIout00OOIUIO'OIOIIOIIOOIOIOOIOIOIOIIIIIIIOIDIOIIIVIIIII-I'll!IIOIIDOOIIIIIIDIOIOIIIUIIIIICOIIIIIllIIIIUIIIIDCIIOIOIOIOIII. IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

.v ov ov
. go. o:.
ooo. a an.
ov ov
vmo.
ooo. n2..-
ov
ooo.

 

33>-..
..<. 8.3.5.5.. 2.3.

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

33>-..
.mm. cooaom 3:3...

GUIDIIOUOOOOICOOIIOOOOOOIIOOIOIOOOI00...IOIIIIIIIt.IIOOOOIOIIOIOIODDOII0.0.000OI.IIIOOIOUDIIIOOOOODIOIIIIIIOOOIIOIIOOIIUI'OIIIUIOOIOIII' IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

33>-..
.35. 8.3.2 3.95

ODIIO0.ODCOIOOOIOOII'IOOOOIOIOIOIOCOIIII0000-000OOIIOOIOOIIIOIDOIIOOIUIIIIIIOIIUOIOIIIIIIIIUIIII-IIOIIlllllIIIOIOOIIIIIOOIIIIIOIIIIDIII' IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

33>-..
E... was“: 8.28.. o. .36.. 3.2..

IIOIOOOOOIOIDOOOOOOIIIUOI'OIIUUIOIOIOIIIIDIOOUOIIIIIIDOOIOOIIOIIOOIIOOIIOOI'DOOOnoonoOIlnOOIIIIOUIIOICIIIOODIIIIIIIIIIIIIIUIIIIIOICOII' IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

33>-..
9.0V cog—83. 85.50 03.8.”...

IOOIIIOOIOOllititIOOIIOOI'IIIUOIOOOIIIIoCOCO...IIIIIIIUIOOCIOIIIOOIIIIInn-Io.-OOOIIIOI-OOIIIIICOOIOIIIOIIOUIIIIUIOllIIllllltootlIIIIIO' IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

23>-..
352.. .2333 3583.2.

OOOIIOOOIOIOOIIIOIOOOO‘OOIIOIIIOOOIIOOOOOOOOOIOIIOOIDUOOIIOOIOOCIUIOIOIOICOOIIOI.OOIOIOIIIIIOICIOIIDDOIIOI-tlIll...OUIIIUIOIIOCIIOIIDOO' IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

own mmo. o.. oo.
moo: “no: mo.. "no.
oo ow oo oo
.oo Now. mm. ooo.
m~.: boo. ooo. ooo.
oo ov ov oo
ovo q... o.o omm
o.o. o.o. .n.. .o..
......... .os-e-s.m-s-sa.w-s=szs..s:z-s
.oo Noo. .oo ooo
a... mom" «mm. wmm"
mo m. m. mo
ooo moo ooo ooo
w... mmm" NW.“ om."
no mo m. M.
o.o. ooo. ooo.
ooo.“ mom" mm.“ wmmu
mg no no
”No. .oo
ooo. mwm" mama
mg no
ooo.
ooo.. m...
mo
ooo.

3.62.. 953.3. n50.22..

 

mu 2m m<

mm

2m mm MD 952. 4402.

 

 

.. .33 a owing; 8353?. we 55.2 zoom—2.8.8:.

w 2an

95

 

.v ov ov .v .v .v .v .v 3 Z

 

8.. SN. ooo. So. So. omo. 8.. .oo. ooo. 029,...
3.5.3883.
8...... 12.55
.ﬁ. 2.. % wﬂ am am .3. a a 28.0.. 280
.oov ..................... o o. ....................... w... .................. z... ................ ....... ................ a... .................. a... ..................... :Vz .............................
ooo. «on. :o. ooo. .8. .oo. 2... .oo. .oo. 02?-..
85.3808...
.300
So. So. a. a. a». .3. ooo. % % 8...... 12.8....
.vov ..................... o o. ....................... o... .................. . ..... ................ a... ................ .. .m .................. .. ..... ..................... .vz ..............................
2... 2o. 3... on... ooo. 2... Rm. .3. So. .29....
333.
255.... 253..
So. So. .~o.- .8. 2o. .oo. ooo: ooo. ooo. 28.. .6 5.32m
.vov ..................... o .m ....................... 2.. .................. . ..... ................ . .m ................ .. ..... .................. . ..... ..................... .vz .............................
on... E. om... EN. 2.. 3m. 2N. w. .. o8. 02$-..
£80
8...... .86....
«no: .3. So. 2... 8.. ooo. o... oz. .5. .o .68...
.vov ..................... ... ..... ....................... 2.. .................. m ..... ................ m o. ................ m ..... .................. m ..... ..................... m «z .............................
.3. So. «8. ooo. ooo. ooo. ooo. ooo. ooo. 02?-..
mmo=o>_.oo.mm
ooo. ooo: odd 3%. ﬂ oolo. molt .Nloﬂ Manon. 3385..
832:5 3:95.382 5.3.533. toaasm 20.3.2 .026. sous—08M 35.09.01. Giacomo.
=2.on .o 5.38m o>so< 3:38... 3.2m 3.2% 8:300 3.582%... 02.33:.

 

 

.. .051. 3 833.8.» .595 on 23 .53.... 2.... .o macaw—atom.

o 2.3...

correlations of the independent and dependent variables.
Table 10

Qgrrelations of Independent and Dependent Variables at Level I

Inclusive Instrumental Conﬂict Shared Shared

 

 

Leadership Leadership Resolution Power Mission

Perceived .730 .660 .41 1 .531 .721
Effectiveness

p-value .000 .000 .000 .000 .000

N 484 483 44 l 477 482

 

Modeling Perceived Council Effectiveness

Hierarchical linear modeling (HLM) is a multi-level modeling technique that allows
for both individual level (e. g., organizational role; individual perceptions of council
climate) and council level (e. g., aggregate perceptions of council climate and council
characteristics) predictors to be examined simultaneously. Such multi-level analysis is
required when data is nested and there is a likelihood of non-independence of errors. That
is, one would expect council members from the same setting to be more similar to one
another than to members of other settings. Hierarchical linear modeling is appropriate
because it accounts for the non-independence of errors and controls for the problem of
downwardly biased standard errors that would occur if regular multiple regression was
utilized to analyze this nested data (Bryk & Raudenbush, 1992). Modeling Level II
variation in the intercept, or differences in average perceived effectiveness scores across
councils was the focus of this study. Given the relatively low power for Level II analyses

(i.e., those analyses predicting differences across councils), these ﬁndings must be

97

considered preliminary and exploratory in nature.
Estimating the Unconditional Model

When employing HLM a necessary ﬁrst step is the estimation of the unconditional
model (i.e., a model with no independent variables) to determine what proportion of
variance in the outcome variable exists within councils and what proportion exists across
councils (i.e., calculating the Intraclass Correlation (ICC)). Estimating the unconditional
model for perceived council effectiveness and calculating the ICC indicated that 20% of
the variance in perceived effectiveness was at the council level. This indicates that
councils varied signiﬁcantly with regard to their average perceived effectiveness and
demonstrates non-independence or errors in the dependent variable, justifying the use of
multi-level modeling. That is, council members in the same setting were more similar to
one another with regard to their perceptions of council effectiveness than they were to
members of other councils.

The unconditional model was also performed and the ICC calculated for each
independent variable to determine whether signiﬁcant variation existed between councils
and whether it was appropriate to aggregate members’ scores to the council level (see
Table 1 on p. 74 for a summary of these ﬁndings). The intraclass correlation (ICC) for
each independent variable indicated a signiﬁcant portion of the variance in shared power in
decision-making, conﬂict resolution, shared mission, instrumental leadership and inclusive
leadership was at Level II, or between councils (9%, 12%, 19%, 20% and 16%,
respectively). This provided partial justiﬁcation for taking the average score for each of

these independent variables to model average perceived effectiveness across councils (i.e.,

98

modeling the intercept in Level II HLM analyses) and the other indicators of effectiveness
(i.e., breadth and accomplishment of goals across councils; Bliese, 2000; Bryk &
Raudenbush, 1992; Florin, Giamartino, Kenny, & Wandersman, 1990).'3
Estimating Models for Each Independent Varjajg

To address Research Questions 4 and 5 as they relate to explaining perceived
effectiveness (To what degree is council climate related to council effectiveness? and To
what degree are council characteristics related to council effectiveness?) the council
average for each element of the council climate (shared power in decision-making,
effective conﬂict resolution, effective leadership and shared mission) was entered into the
Level II model of the intercept, or the average perceived effectiveness score for each
council. In addition, to account for within group variation on each independent variable,
each council member’s score was entered into the Level I model of perceived effectiveness
group-mean centered. Group-mean centering allows one to separate individual level
effects (i.e., Level I) from council level effects (Bryk & Raudenbush, 1992). For the
purposes of this study, HLM allowed for the examination of the degree to which each
element of the council climate explains variation in average perceived effectiveness across
councils (i.e., the council level effect) while accounting for the effect of each member’s
perception of the elements of the council climate relative to others in their group (i.e., the
individual level effect). For example, council members who viewed their councils as

having a shared mission, relative to the mean of their group, were more likely to view their

 

13

Please note that James, Demaree and Wolf’s (1984) r
aggregating this data to the council level.

ng) was also calculated to justify

99

councils as effective (the individual level effect). In addition, councils who were viewed
on average as having a shared mission were more likely to be viewed on average as
effective (the council level effect). Given the focus of the questions posed in this study
concerned the council level effect (i.e., what explained differences in perceived
effectiveness across councils), only Level II ﬁndings are described below. Please note that
Level I ﬁndings are summarized in italics in Tables 11 to 16.

Analyses were performed for each independent variable separately followed by
analyses of combined models which included multiple climate variables to determine the
most parsimonious and most informative HLM model relating council climate to council
effectiveness." Finally, to begin to explore the relationship of council characteristics
(measured via key informant report) to perceived council effectiveness, each council
characteristic was added to the Level II estimation of the intercept of the ﬁnal climate
model separately." This allowed an examination of the degree to which council
characteristics explained differences across councils in their average perceived

effectiveness.

 

14

Again, all climate variables were tested using one-tailed tests. All council characteristics
were tested using two-tailed tests except for ‘external support’ which was tested with a
one-tailed test given its directional hypothesis.

15

While a combined model was estimated, characteristics were entered separately to
examine their unique effects. This was an important step given the relatively small number
of Level 11 units (N = 38).

100

Conﬂict Resolution

Effective conﬂict resolution was a signiﬁcant predictor of council effectiveness at
both the individual and council levels (see Table 11). These ﬁndings indicate that the
degree to which councils were rated by their members as handling conﬂict effectively on
average was positively related to the degree to which they were viewed as effective in
meeting their goals and the goals of a CCR (Level II Coefﬁcient = .48, p < .001, one-
tailed). That is, councils with more productive conﬂict resolution strategies were more
likely to be perceived as effective in their collective work.

S_hared Power in Decision-Making

Similarly, shared power in decision-making was a signiﬁcant predictor of council
effectiveness at both the individual and council levels (see Table 11). These ﬁndings
suggest that the degree to which councils were viewed as sharing power in decision-
making on average was positively related to the degree to which they were viewed as
effective in meeting their goals and the goals of a CCR (Level II Coefficient = .86, p <

.001, one-tailed). That is, councils where power is shared in decision-making were more

likely to be perceived as effective in their collective work.

101

Table 1 1

Hierarchigg Linear Modeling of Perceived Effegiveness Regressed onto Council Climm

V ' 1 In ividu ll evel IN = 484' Level II N = 38

 

 

 

 

Variables Unconditional Climate Climate
Model Model 1 Model 2
Unstandardized Coefficient (SE)
Intercept 437*" (.08) 230*" (.52) .40 (.59)
Conﬂict Resolution (CR) --- .29 "' * * (.05) ----
---- 48*" (.15) —---
Shared Power (SP) ---- --- .43 “* (.05)
«~- --- 86*" (.14)
Shared Mission (SM) ---- ---- -...

Instrumental Leadership (IT) ---- ---- ----

Inclusive Leadership (1C) ---- ---- ----

Council Age ---- 01"" (.00) 01"" (.00)

 

Variance Components

02 .75 .67 .58

too . l9AM .1 1M" .OS’V‘A
tn (CR) .03M -..-

122 (SP) "" ""' '02”
Deviance ((11) 1287.62 (2) 1242.74 (4) 1164.37 (4)

 

Note: one-tailed T'p-value < .10, *p-value < .05, "p—value < .01, M"‘p - value < .001; two4ailed 17p—

value < .10, "p - value < .05, Mp-value < .01. ’W‘p-value < .001; Level 1 Coefﬁcients in italics.

102

sum

Similar to the other climate variables, the presence of a shared mission was a
signiﬁcant predictor of council effectiveness at both the individual and council levels (see
Table 12). These ﬁndings indicate that the degree to which councils were viewed as
having a shared mission on average was positively related to the degree to which they
were viewed as effective in meeting their goals and the goals of a CCR (Level II
Coefﬁcient = .17, p < .05, one-tailed). That is, councils with a shared mission were more
likely to be perceived as effective in their collective work.
mm

Consistent with the ﬁndings for other climate variables, inclusive leadership and
instrumental leadership were signiﬁcant predictors of council effectiveness at both the
individual and council levels (see Table 12). These ﬁndings indicate that councils whose
leaders are viewed as effective on average were more likely to be viewed as effective in
meeting their goals and the goals of a CCR (Level II Coefficient for Inclusive Leadership
= .97, p < .001, one-tailed; Level II Coefﬁcient for Instrumental Leadership = .73, p <
.001, one-tailed). That is, councils with more effective leaders were more likely to be
perceived as effective in their collective work.

Given the shared variance in instrumental and inclusive leadership (r = .88, p <
.001), these factors were also examined simultaneously. In this model, only inclusive
leadership was signiﬁcantly related to perceived effectiveness at the council level (Level II
Coefﬁcient for Inclusive Leadership = .92, p < .001, one-tailed). This indicates that the

unique variance associated with inclusive leadership was signiﬁcantly related to the

103

Table 12

Hieregehiﬂ Linea; Medeling ef Pereeived Effﬁtiveness Regressed onte Qeeneil Climate
Vm'ables Individually (continued)

 

 

 

Variables Unconditional Climate Climate Climate
Model Model 3 Model 4 Model 5
Unstandardized Coefﬁcient (SE)
Intercept 437*“ (.08) .35 (.34) .85* (.35) -.37 (.33)
Conﬂict Resolution (CR) --—- ---- ---- ----
Shared Power (SP) ---- ---- ---- ----
Shared Mission (SM) ---- .64 *** (. 04) ---- ----
--- .17* (.08)
Instrumental Leadership (IT) ---- ---- .5 2*" (. 04) ----
--- ---- 73*" (.08) ---
Inclusive Leadership (IC) ---- ---- --- 04*" (. 04)
..-- ..-- --- 97"" (.07)
Council Age ---- 01*” (.00) .00" (.00) .00" (.00)

 

Variance Components

o2 .75 .43 .51 .43

10., .19AM .131‘2 .029 .00

In (SM) .00

144 (IT) .00

155 (IC) .02“
Deviance (df) 1287.62 (2) 999.01 (4) 1087.27 (4) 1001.71 (4)

 

Note: one-tailed Tlp-value < .10, *p-value < .05, "p-value < .01, *“p - value < .001; two-tailed ”p—

value < .10, "p - value < .05, Mp-value < .01, “Mp-value < .001; Level 1 Coefﬁcients in italics.

104

average perceived council effectiveness across councils while the unique variance
associated with instrumental leadership was not. This suggests that when examining these
multiple indices of leadership, what is most important in predicting effectiveness across
councils is the extent to which council leaders use an inclusive leadership style. However,
it is important to note that given the collinearity in these independent variables, the relative
importance of inclusive leadership could also be a statistical artifact.
Estimating a Combined Climete Model

As a ﬁrst step in estimating the ﬁnal model relating council climate to perceived
council effectiveness all ﬁve climate variables were entered simultaneously into the Level I
model, group-mean centered, with each corresponding council mean entered into the
Level 11 model of the intercept, or average perceived effectiveness (see Table 13). Given
that conﬂict resolution, shared power in decision-making and inclusive leadership
demonstrated signiﬁcance or a trend toward signiﬁcance in the variance of their slopes
across councils (1’s) the random effects associated with these slopes were not constrained
(i.e., set to zero). However, the random effects of the slopes for shared mission and
instrumental leadership were constrained given they did not vary signiﬁcantly across
councils. Constraining the random effects is an important part of increasing model
parsimony.

In this full model, the only signiﬁcant predictor of variation across councils in their
average perceived council effectiveness (i.e., the intercept) was inclusive leadership (Level

II Coefﬁcient = .60, p < .05, one-tailed). This suggests that inclusive leadership subsumes

105

the effects of the other independent variables. Still, the presence of a shared mission and
shared power in decision-making demonstrated trends toward signiﬁcance at Level 11
indicating they might also explain variation across councils (see Table 13).

To increase the parsimony of the ﬁnal model, conﬂict resolution was removed
from the model given it was not a signiﬁcant Level I or Level II predictor, nor did it
demonstrate a trend toward signiﬁcance. In addition, the random effects for all of the
slopes were constrained given that none varied signiﬁcantly in the combined model. In the
ﬁnal model (i.e., including all climate variables except for conﬂict resolution), the pattern
of signiﬁcance was similar; inclusive leadership was a signiﬁcant Level II predictor (Level
II Coefficient = .45, p < .05, one-tailed; see Table 13). However, shared mission also
emerged as a signiﬁcant Level II predictor (Level II Coefficient = .26, p < .05, one-tailed).
Instrumental leadership was not signiﬁcant (Level II Coefﬁcient = .14, p = .16, one-
tailed). Shared power in decision-making approached a trend toward signiﬁcance at the
council level (Level II Coefﬁcient = .15, p = .12, one-tailed)“ One possible explanation
for shared power in decision-making and instrumental leadership not emerging as
signiﬁcant Level II predictors was the multicollinearity among the predictor variations.
Inclusive leadership was strongly positively related to perceived effectiveness as well as
the other independent variables, thus possibly masking other relationships. However, it is

also important to note that the magnitude of the relationship of inclusive leadership to

 

16

While shared power in decision-making did not achieve signiﬁcance it was retained
because it appeared to contribute to the model. Thus, this model was treated as the ﬁnal
climate model.

106

Table 13

Hiergchical Linear Modeling of Perceived Effectiveness Regressed onto Council Climate
Variables Combined (Level I N = 484; Level II N = 38)

 

 

 

 

 

Variables Unconditional Combined Final
Model Climate Climate
Model Model
Unstandardized Coeﬂicient (SE)
Intercept 437*" (.08) -.41 (.36) -.39 (.35)
Conﬂict Resolution (CR) --- .03 (.04) ----
--- -.09 (.09) ----
Shared Power (SP) ---- .01 (.05) .03 (.04)
--- .18Tl (.12) .14 (. 12)
Shared Mission (SM) ---- 37*" (.04) 38*" (.04)
---- .21TI (.13) .26" (.13)
Instrumental Leadership (IT) ---- .12" (.05) .13" (.05)
~--- .08 (. 15) .14 (.14)
Inclusive Leadership (IC) ---- .26 "* (.07) 24*" (.07)
...- .60* (.28) .45“ (.26)
Council Age ---- .00" (.00) .00” (.00)
Variance Components
02 .75 .36 .37
to, .19” .00 .oo
1,, (CR) --..- .00 ---
122 (SP) .03
1,, (IC) -- .02
Deviance ((11) 1287.62 (2) 937.85 (1 1) 934.50 (2)

 

Note: one-tailed T'p-value < .10, *p-value < .05, "p—value < .01, *“p - value < .001; two-tailed ”p-

value < .10, "p - value < .05, Mpovalue < .01. MAp-value < .001; Level 1 Coefficients in italics.

107

perceived council effectiveness was reduced in the presence of the other predictor
variables, suggesting partial mediation of this relationship by the other climate variables.

In summary, at the council level, these ﬁndings suggest that councils who were
viewed on average as having strong inclusive leadership, a shared mission and perhaps to
some extent shared power in decision-making were more likely to be viewed as effective
in their collective work.

Adding Council Characteristics to the Final Chm Model

To create a model that included not only council climate variables, but also council
characteristics, each characteristic was added individually to the Level 11 model of the
intercept in the ﬁnal climate model (i.e., the model including inclusive leadership,
instrumental leadership, shared mission and shared power).
Formality of Council Structure

The degree of formality in the council structure was not a signiﬁcant predictor in
the Level II model of the intercept (Level II Coefficient = -.01, p = .94, two-tailed; see
Table 14) nor did its addition to the ﬁnal climate model result in any appreciable changes.
This indicates that councils with a higher degree of formality were no more or less likely
than councils with less formality to be viewed as effective by members on average.
Bgtdth of ‘Oificia_l’_and ‘Active’ Membership

The breadth of oﬂicial council membership was not a signiﬁcant Level II predictor
(Level II Coefficient = .05, p = .82, two-tailed) and did not make an appreciable change in
the ﬁnal climate model as a whole (see Table 14).

However, the breadth of active membership was a signiﬁcant predictor of variation

108

Table 14
Hierarchical Linear Modeling of Perceived Effectiveness Regressed onto Council Clim_ate

and Characteristic Variables (Level I N = 484;.evel II N = 38)

 

 

 

Variables Unconditional Combined Combined
Model Climate! Climate!
Characteristics Characteristics
Model 1 Model 2
Intercept 4.37”" (.08) -.38 (.35) -.42 (.39)
Shared Power (SP) -- .03 (.04) .03 (.04)
~— .l4(.12) .l4(.ll)
Shared Mission (SM) ~— .33“‘ (.04) .38‘“ (.04)
-- .26" (.13) .26‘ (.14)
Instrumental Leadership (IT) -— .13” (.05) .13‘ (.05)
-— .14 (.14) .13(.l4)
Inclusive Leadership (IC) -— .24‘“ (.07) .24‘” (.07)
--- .45‘ (.26) .46“ (.27)
Council Age .00“I (.00) .00” (.00)
Council Structure -—- -.0l (.16) -—-—
Breadth of ‘Ofﬁcial’ Membership -- -- .05 (.21)
Breadth of ‘Active’ Membership -- __ _
External Support .... .. ..
Variance Components
0’ .75 .37 .37
10° .19""" .00 .00
Deviance (dl) 1287.62 (2) 934.44 (2) 933.94 (2)

 

 

Note: one-tailed Tlp-value < .10, ‘p-value < .05, "p-value < .01, "*p - value < .001; two-tailed ”p«

value < .10, "p - value < .05. Mp-value < .01. “Mp-value < .001; Level 1 Coefficients in italics.

109

in the intercept (i.e., variation in average perceived effectiveness across councils; Level II
Coefficient = .54, p < .05, two-tailed; see Table 15). The addition of this predictor also
created some changes in the ﬁnal climate model: 1) shared power in decision-making
emerged as a signiﬁcant Level II predictor (Level II Coefficient: .20, p < .05, one-tailed),
and 2) inclusive leadership was no longer a signiﬁcant Level II predictor of average
perceived effectiveness across councils (Level II Coefficient = .35, p < .10, one-tailed).
Speciﬁcally, the Level II Coefﬁcient for shared power in decision-making changed from
.14 to .20 and the Level II Coefﬁcient for inclusive leadership changed from .49 to .35.
This indicates that councils with a greater proportion of active members were more
effective on average than councils with a smaller breadth of active members. However,
these ﬁndings also suggested that the breadth of active membership: I) acted as classical
suppressor of the relationship between shared power in decision-making and perceived
council effectiveness at the council level and 2) reduced the inﬂuence of the unique
variance associated with inclusive leadership. This suggests the unique association
between shared power in decision-making and council effectiveness was stronger when the
effect of the number of active members on the council was controlled. That is, shared
power in decision-making is more important when you account for the number of people
actively participating in council activities. The reduced contribution of inclusive leadership
might also be due in part to the suppression effect bolstering the inﬂuence of shared power

in decision-making, given they are signiﬁcantly related to one another.‘7

 

17

Interestingly, the breadth of active membership is not signiﬁcantly related to shared power
in decision-making or to inclusive leadership. Thus, this is not collinearity suppression.

llO

Extgmﬂ Support for Council Activities

While external support for council activities had a strong correlation with averaged
perceived council effectiveness (r = .55, p < .001), it only approached a trend toward
signiﬁcance in the Level 11 model of the intercept (Level II Coefficient = .04, p = .13, one-
tailed; see Table 15). In addition, with external support in the equation, shared mission
demonstrated only a trend toward signiﬁcance (Level II Coefficient = .21, p < .10, one-
tailed) when previously it had been a signiﬁcant predictor (Level II = .26, p < .05, one-
tailed). These ﬁndings are probably due to multicollinearity considering the strong
intercorrelation of external support with the climate indicators including shared mission.
This ﬁnding suggests that the unique variance associated with external support for council
activities did not predict the degree to which councils were viewed as effective on
average.

Estimating a Final Combined ClimatdCimcteﬁstics Model

Given that only the breadth of active membership emerged as a signiﬁcant
predictor of average perceived effectiveness, the ﬁnal climate model which included the
breadth of active participation variable was the best combined climate/characteristics
model (see Table 16). The combined climate/characteristics model demonstrated a
reduction in the deviance statistic with the same degrees of freedom, indicating the model

provides a better ﬁt to the data than the ﬁnal climate model.

111

Table 15

Hierarchical Linear Modeling of Perceived Effectiveness Regressed onto Council Climate
and Characteristic Variables (continued)

 

 

 

 

 

Variables Unconditional Combined Combined
Model Climate! Climate/
Characteristics Characteristics
Model 3 Model 4
Unstandardized Coefficient (SE)
Intercept 4.37‘” (.08) .-73’ (.37) -.29 (.36)
Shared Power (SP) --- .03 (.04) .03 (.04)
--- .20" (.12) .13(.12)
Shared Mission (SM) --- .38‘“ (.04) .38‘” (.04)
--- .28“ (.13) .217I (.14)
Instrumental Leadership (IT) -- .13” (.05) .13” (.05)
...- .18(.l4) .12(.14)
Inclusive Leadership (IC) -- .24‘" (.07) .24‘" (.07)
--- .33 (.26) .48’ (.26)
Council Age .00“I (.00) .00“ (.00)
Council Structure ---- .... ....
Breadth of ‘Ofiicial’ Membership -- - —-
Breadth of ‘Active’ Membership .. .54" (.21) --
External Support .... -- 04 (.04)
Variance Components
(32 .75 .37 .37
too . 19M" .00 .00
Deviance ((11) 1287.62 (2) 927.77 (2) 936.02 (2)

 

Note: one-tailed T'pcvalue < .10, *p-value < .05, "p-value < .01, ”*p - value < .001; two-tailed 1"p-

value < .10, "p - value < .05, Mp-value < .01, MAp-value < .001; Level 1 Coefficients in italics.

Table 16
Hierarchical Linear Modeling of Perceived Effectiveness Rggressed onto Council Climate

and Chgacteristic Variablgs: Final ModelgLevel I N = 484; Level II N = 38)

 

Variables Unconditional Model Final Final
Climate Climate/
Model Characteristics

 

Unstandardized Coefﬁcient (SE)

 

Intercept 437"" (.08) -.39 (.35) .-73‘ (.37)
Shared Power (SP) ---- .03 (.04) .03 (.04)
-- .14 (.12) .20‘ (.12)
Shared Mission (SM) «u .38‘” (.04) .38‘“ (.04)
-- .26‘ (.13) .28‘ (.13)
Instrumental Leadership (1T) ---- .13" (.05) .13“ (.05)
-- .14 (.14) .18(.l4)
Inclusive Leadership (IC) ---- .24 ‘" (.07) .24 ‘“ (.07)
- .45’ (.26) .33 (.26)
Council Age -—- .00" (.00) .00" (.00)

Council Structure ~— .... --..
Breadth of ‘Ofﬁcial’ Membership --- _. ..
Breadth of ‘Active’ Membership ---- -- .54" (.21)

External Support --— --- --

 

Variance Components

 

(,2 .75 .37 .37
too .19W .00 .00
Deviance (do 1287.62 (2) 934.50 (2) 927.77 (2)

 

Note: one-tailed Tlp-value < .10, *p-value < .05, "p-value < .01, *"p — value < .001; two-tailed 11p-

 

value < .10, "p - value < .05. Mp-value < .01, AMp-value < .001; Level 1 Coefficients in italics.

113

Modeling the Breadth of Council Goals
Estimating Models for Each Independent Vamla

To further address Research Questions 4 and 5 (To what degree is council climate
related to council effectiveness? and To what degree are council characteristics related to
council effectiveness?) a series of standard regression analyses were performed to relate
council climate and characteristics to another indicator of council effectiveness -- the
breadth of goals addressed within and beyond the criminal justice system. These analyses
included each of the average climate variables (i.e., shared power in decision-making,
effective conﬂict resolution, effective leadership and shared mission) and the council
characteristics (i.e., breadth of council membership, breadth of active membership,
formality of structure, presence of external support). Given the relatively small number of
councils (N = 40) that had adequate data to be included in these analyses, the breadth of
goals addressed within the criminal justice system and the breadth of goals addressed
beyond the criminal justice system were regressed onto each independent variable
separately. Following this, combined models were performed to identify the best model of
the breadth of areas councils addressed.
Conﬂict Resolution

Effective conﬂict resolution was not signiﬁcantly related to the breadth of criminal
justice goals addressed by councils (B = .18, p = .12, one-tailed) or the number of goals
addressed beyond the criminal justice system (B = -. 10, p = .74, one-tailed; see Tables 17
and 18). This indicates that councils where conﬂict was resolved effectively were no more

or less likely to address a broad range of goals with or beyond the criminal justice system.

114

Table 17

Breadth of Criminal Justice Svstem Goals Regressed onto Effective Conﬂict Resolution

Strategies (N = 40)

 

 

Variables B _S_E B Q p-value
(one-tailed)
Intercept .3 88 .303 . 104
Effective Conﬂict Resolution .102 .086 .183 .120
Council Age .001 .001 .245 .060

 

N04233=JO

Table 18

Breadth of Goals Bevond the Criminal Justice Svstem Iggressed onto Effective Conﬂict

Resolution Strategies (111 = 40)

 

 

 

Variables _B_ S_E _B_ Q p-value
(one-tailed)
Intercept .777 .290 .001
Effective Conﬂict Resolution -.054 .082 -.103 .741
Council Age .001 .001 .230 .077

 

Note. E = .06

115

Shagd Power in Decision-Mam

Shared power in decision-making was not signiﬁcantly related to the breadth of
criminal justice goals councils addressed (13 = .01, p = .48, one-tailed) or the number of
goals addressed beyond the criminal justice system (B = .01, p = .47, one-tailed; see
Tables 19 and 20). This indicates that councils where power was shared in decision-
making were no more or less likely to address a broad range of goals within or beyond the

criminal justice system.

Table 19

Breadth of Criminal Justice System Goals Regressed onto Shared Power in Decision-

 

 

 

Makin = 40

Variables B a: B Q p-value
(one-tailed;

Intercept .716 .401 .041

Shared Power in Decision-Making .006 .096 .010 .477

Council Age .001 .001 .253 .067

Note. E = .07

116

Table 20

Breadth of Goals Bevond the Criminal Justice System Regressed onto Shared Power in

 

 

 

D i ion Makin = 40

Variables _B _S_E_E B B a-value
(one-tailed)

Intercept .564 .379 .073

Shared Power in Decision-Making .007 .091 .013 .470

Council Age .001 .001 .220 .097

Note. _R_2 = .05

Shared Mission

The presence of a shared mission was not signiﬁcantly related to the breadth of

criminal justice goals councils addressed (B = .06, p = .37, one-tailed) nor the number of

goals addressed beyond the criminal justice system (B = -.02, p = .54, one-tailed; see

Tables 21 and 22). This indicates that councils with a shared mission were no more or less

likely to address a greater breadth of goals within or beyond the criminal justice system.

117

Table 21

 

 

 

 

Breadth f riminal ti e S stem Mission (N = 40)
Variables B S_ B B p-valua
(one-tailed)
Intercept .625 .337 .036
Shared Mission .026 .076 .055 .367
Council Age .015 .001 .245 .066
Note. B2: .07
Table 22

Braadth of Goals Bayond the Criminal Justice System Regressed onto Shared Mission

 

 

 

= 40
Variables B _S_E_ B B p-value
(one-tailed)
Intercept .62 1 .3 19 .030
Shared Mission .007 .072 -.015 .536
Council Age .001 .001 .227 .084

 

Note. 33 = .05

118

Mime

Leadership being inclusive (i.e., encouraging the active participation of all
members) was not signiﬁcantly related to the breadth of criminal justice goals councils
addressed (B = .10, p = .28, one-tailed) nor the number of goals address beyond the
criminal justice system (B = -.08, p = .69, one-tailed). Similarly, instrumental leadership
(e. g., planning meetings eﬂiciently) was not signiﬁcantly related to the breadth of criminal
justice goals councils addressed (B = .19, p = .12, one-tailed) nor the number of goals
addressed beyond the criminal justice system (B = .03, p = .42, one-tailed). This indicates
that, considered separately, councils which had strong inclusive and instrumental
leadership were no more or less likely to address a broad range of goals within or beyond

the criminal justice system (see Tables 23 through 26).

Table 23

Breadth of Cﬁminﬂustice System Goaﬁ Regressed onto Inclusive Leadershig (N = 40)

 

 

 

Variables B BB B B p-value

(one-tailed)
Intercept .5 17 .3 86 .095
Inclusive Leadership .050 .085 .098 .282
Council Age .001 .001 .220 .099
Note. B2 = .07

119

Table 24

Brgdth of Goals Beyond the Criminal Justice System Regressed onto Inclusive
Leadership (B = 40)

 

 

 

Variables B S_B B B -value
(one-tailed)

Intercept .77 l .3 66 .021

Inclusive Leadership -.O40 .081 -.084 .688

Council Age .001 .001 .254 .071

Note. B2 = .06

Table 25

Braadth of Criminal Justice System Goals Regressed onto the Instrumental Leadership

 

 

 

= 40
Variables B S_E B B p-value
(one-tailed)
Intercept .3 81 .294 . l 10
Instrumental Leadership .084 .070 .193 . 1 18
Council Age .001 .001 .204 . 105
Note. Bf = .10

120

Table 26

Breadth of Goals Beyond the Criminal Justice System Regressed onto Instrumental

 

 

 

Qaderahip (E = 40)

Variables B _B B B p-value
(one-tailed)

Intercept .537 .294 .038

Instrumental Leadership .0 I 3 .067 .03 2 .424

Council Age .001 .001 .21 5 .099

Note. E = .05

Estimating a Combined Climate Model

While no climate variables emerged as signiﬁcant predictors of the breadth of goals
addressed, a combined model including all of the predictors together was still performed.
This was important given that sometimes suppression effects emerge when variables are
simultaneously entered into a model, revealing signiﬁcant relationships. This was the case
with regard to the breadth of goals addressed within the criminal justice system.
Speciﬁcally, in the model where the breadth of goals addressed within the criminal justice
system was regressed onto all of the climate variables simultaneously, one predictor
emerged as signiﬁcant and another emerged with a trend toward signiﬁcance. In the
presence of the other independent variables, instrumental leadership was signiﬁcantly

positively related to the breadth of criminal justice goals addressed (B = .65, p < .05, one-

tailed) and there was a trend toward signiﬁcance for effective conﬂict resolution (B = .26,

121

p < .10, one-tailed; see Table 27). This indicates that councils with a stronger
instrumental leader and to some extent those who handled conﬂict more effectively
addressed a broader range of goals within the criminal justice system. To create a more
parsimonious model, the breadth of criminal justice goals was regressed onto instrumental
leadership and conﬂict resolution without the other climate variables in the model; neither
instrumental leadership (B = .14, p = .20, one-tailed) or conﬂict resolution (B = .14, p =
.21, one-tailed) emerged as signiﬁcant predictors of the breadth of criminal justice goals
addressed (see Table 28). While it is not a signiﬁcant predictor of the breadth of goals
addressed in the criminal justice system (B = -.63, p = .94, one-tailed”), adding inclusive
leadership into this model affects the contributions of other predictors, suggesting that it
acts as a suppressor, allowing the relationships of instrumental leadership (B = .65, p <
.05, one-tailed) and, at the level of a trend, conﬂict resolution (B = .25, p < .10, one-
tailed) to emerge (see Table 29). This indicated that when the variance associated with
inclusive leadership in the independent variable was accounted for, the unique variance
associated with instrumental leadership was signiﬁcant. The variance in instrumental
leadership that is unshared with inclusive leadership may reﬂect leaders having a greater
task orientation (e. g., focusing on efficiency). This suggests that councils with a stronger
instrumental leader who is more oriented toward tasks than an inclusive process are more

likely to address a greater breadth of goals within the criminal justice system. Similarly,

 

18

Note that given the directional hypothesis, this one-tailed test was not signiﬁcant nor did it
demonstrate a trend.

122

Table 27

Braadth of Criminal Justice Svstem Goals Regressed onto Council Clim_ate Variables

 

 

 

= 40
Variables B BB B B p-value
(one-tailed)

Intercept .592 .441 .094
Inclusive Leadership -.265 .233 -.521 .867
Instrumental Leadership .284 . 162 .650 .044
Conﬂict Resolution .147 .102 .264 .080
Shared Power in Decision-Making -.083 .138 -.143 .724
Shared Mission -.010 .113 -.021 .535
Council Age .018 .001 .303 .044
Note. B2 = .18

123

Table 28

Braadth pf Qriminal Justiga Systam ﬁpala Ragpessed ontg Qpanpil Climata Variablea:
Reduced Model (B = 40)

 

 

 

Variables B S_E B B p-value

(one-tailed)
Intercept .214 .369 .283
Instrumental Leadership .063 .075 .144 .204
Conﬂict Resolution .075 .092 .135 .210
Council Age .001 .001 .209 . 101
Note. B2 = .12

the unique variance associated with effective conﬂict resolution demonstrated a trend
toward signiﬁcance with inclusive leadership in the model. The variance in conﬂict
resolution that is unshared with inclusive leadership may reﬂect councils’ practices of
“agreeing to disagree,” which is oriented toward efficiency as opposed to working to “get
to root of the problem” which implies a greater need to hear all voices. '9

For the breadth of goals addressed beyond the criminal justice system, no

signiﬁcant predictors emerge in the combined climate model. However, there was a trend

 

19

An item level analysis provided some conﬁrmation for this assertion indicating that the
only item in the positive conﬂict resolution subscale uncorrelated with the items from the
inclusive leadership subscale was “When faced with conﬂict council members ‘agree to
disagree.”

124

Table 29
Breadth of Criminal Justice System Goals Reggssed onto Council Climate Variables:

Second Reduced Model m = 4m

 

 

 

 

Variables B S_E B B p-value
(one-tailed)

Intercept .488 .401 . 1 16
Instrumental Leadership .285 . 158 .651 .040
Conﬂict Resolution .139 .099 .249 .085
Inclusive Leadership -.322 .203 -.633 .939
Council Age .002 .001 .296 .042
Note. Bf = .17

toward signiﬁcance for instrumental leadership (B = .51, p < .10, one-tailed; see Table 30).
This indicates that for the most part this set of climate predictors is not explaining
differences in the breadth of goals addressed beyond the criminal justice system, but that
councils with more effective instrumental leadership addressed goals in a greater number
of areas outside of the criminal justice system. However, the model described above
including only instrumental leadership indicated it was not a signiﬁcant predictor of the
breadth of goals addressed beyond the criminal justice system. Again, inclusive leadership
appears to suppress extraneous variance in instrumental leadership resulting in a trend

toward signiﬁcance in the relationship of instrumental leadership and the breadth of goals

125

addressed beyond the criminal justice system (B = .53, p < .10, one-tailed; see Table 31).
This suggests that the unique variance associated with instrumental leadership (i.e., the
variance unshared with instrumental leadership) was positively related to the degree to
which councils addressed a broad range of goals outside of the criminal justice system.
Again, this suggests councils with leaders who have a greater task orientation are more
likely to address a greater breadth of goals in beyond the criminal justice system. Taken
together, these ﬁndings suggest that the most important factor regarding the breadth of
goals addressed is task-oriented leadership that is competent in garnering resources and
planning meetings efﬁciently.
Addirlg Council Characteristics to the Model

Braadth pf ‘Oﬁicial’ and ‘Active’ Membership

The breadth of council membership was not signiﬁcantly related to the number of
goals addressed beyond the criminal justice system (B = .09, p = .58, two—tailed), but
demonstrated a trend toward signiﬁcance regarding the number of criminal justice goals
addressed by councils (B = .24, p = .06, two-tailed).20 See Tables 32 and 33. This
indicates that councils with a broader official membership were somewhat more likely to
address a broad range of goals within the criminal justice system, but were no more or less
likely to address a broad range of goals in areas outside of the criminal justice system.
However, the breadth of active participation was not signiﬁcantly related to the breadth of

criminal justice goals addressed by councils

 

20

Examining the breadth of official and active membership in relationship to the breadth of
goals addressed was exploratory, requiring a two-tailed test of signiﬁcance.

126

(B = -.01, p = .93) nor to the breadth of goals addressed beyond the criminal justice

system (B = -.07, p = .67). See Tables 34 and 35. This indicates that councils with

broader or more active membership were no more or less likely to be addressing a broader

range of goals within or beyond the criminal justice system.

Table 30

Breadth of Goals Beyond the Criminal Justice Satstem Ragessed onto Council Clima_ta

Variables (N = 40)

 

 

Variables B BB B B ma
(one-tailed)
Intercept .796 .429 .036
Instrumental Leadership .210 . 1 57 .5 13 .095
Inclusive Leadership -.330 .227 -.691 .922
Conﬂict Resolution -.026 .099 -.050 .604
Shared Power in Decision-Making .079 .135 .145 .281
Shared Mission .032 .109 .281 .387
Council Age .002 .001 .28 I .063

 

Note. B2212

Table 31

Breadth of Goals Beyond the Criminal Justice System Regpessed onto Council Climate

Vag'ables: Reduced Model Including Leadership VaL'ables Only (N = 40)

 

 

 

Variables ﬂ S_E_ E B. p-value
(one-tailed)
Intercept .888 .370 .01 1
Instrumental Leadership .2 16 . 149 .526 .078
Inclusive Leadership -.272 .178 -.57 I .932
Council Age .002 .001 .289 .047
Note. 31 = .1 I
Table 32

Breadth of Criminal Justice Goals Regressed onto the Breadth of Official Membership

 

 

 

= 40
Variables B BB B B p-value
(two-taileQ
Intercept .470 .186 .016
Breadth of Council Membership .390 .248 .243 .125
Council Age .002 .001 .277 .082
Note. B3 = .12

128

Table 33

Br h f als B ond the Criminal Justice S stem Re essed onto the Breadth of

Qﬂicial Memﬁrship (N = 40)

 

 

 

Variables B BB B B p-value
(two-tailed)

Intercept .492 . 179 .009

Breadth of Council Membership .134 .239 .089 .579

Council Age .001 .001 .243 . l3 7

Note. B2 = .06

Table 34

Braadth of Criminal Justice System Goals Regressed onto the Breadth of Active

Members_hip (N = 4CD

 

 

 

Variables B S_E B B p-value

(two-tailed)
Intercept .679 . 187 .000
Breadth of Active Membership -.020 .230 -.014 .931
Council Age .002 .001 .257 .121
Note. B2 = .07

129

Table 35

Breadth of Goals Beyond the Criminal Justice System Regressed onto Breaﬁth of Active

Membership (5 = 40)

 

 

 

Variables B BB B B p-value

(two-tailed)
Intercept .492 . 175 .001
Breadth of Active Membership -.093 .215 -.O7I .666
Council Age .001 .001 .249 .135
Note. B2 = .06

External Support for Council Activities

 

The presence of support for council activities was not signiﬁcantly related to the

breadth of criminal justice goals councils addressed (B = .13, p = .21,one-tailed) nor the

number of goals address beyond the criminal justice system (B = .01, p = .48, one-tailed;

see Tables 36 and 37). This indicates that the degree to which councils are receiving

external support for council activities does not explain the breadth of goals councils are

addressing within or outside of the criminal justice system.

130

Table 36

Breadth of Criminal Justice System Goals Regessed onto Exterml Support for Council

 

 

 

Aptivitiaa (E = 40)

Variables B _S_E B B p-value
(one-tailed)

Intercept .633 . 150 .000

External Support .030 .037 .131 .213

Council Age .001 .001 .216 .096

Note. _R_2 = .08

Table 37

Breadth of Goals Beyond the Criminal Justice System Reggessed onto External Support

for Council Activities (N = 401

 

 

Variables B S_ B B p-value
(one-tailed)
Intercept .5 85 . l 43 .000
External Support .002 .03 5 .010 .477
Council Age .001 .001 .22 I .096

 

Note. E = .05

131

Bimhty of Council Structure

The degree of formality in the council structure was not related to breadth of
criminal justice goals councils addressed (B = .03, p = .86, two-tailed) nor the number of
goals address beyond the criminal justice system (B = .13, p = .42, two-tailed; see Tables
38 and 39). This indicates that councils who have a more formal structure are not more or

less likely to address a broader range of goals within or beyond the criminal justice system.

Table 3 8

Breadth of Criminal Justice Goals Regressed onto Formality of Council Structure

 

 

 

= 40
Variables B S_ B B p-value
Qwo-tailed)
Intercept .721 . I 22 .000
Formality in Council Structure .003 .197 .028 .861
Council Age .001 .001 .268 . 104
Note. B2 = .07

132

Table 39

Brgagth of Goals Beyond the Criminal Justice System Reggesseg onto Formality of

 

 

 

ncil Structure = 40
Variables B _SB B B p-value
(two-taileQ
Intercept .517 .114 .000
Formality in Council Structure .151 .185 .132 .418
Council Age .001 .001 .255 .122
Note. Bf = .07

Estimating a Combined CIirnate/Chﬂeristics Model

No council characteristics emerged as signiﬁcant predictors of the breadth of goals
addressed either within or outside of the criminal justice system. However, a ﬁnal model
was performed for each dependent variable. For the breadth of goals addressed in the
criminal justice system a model including instrumental leadership, conﬂict resolution,
inclusive leadership and the council characteristics was performed. This model indicated a
trend toward signiﬁcance for instrumental leadership (B = .58, p > .10, one-tailed; see
Table 40), but for no other variables. Consistent with the individual models for each
independent variable, this suggests that the council characteristics investigated in this
study do not explain variation in the breadth of goals councils address within the criminal
justice system.

Similarly, for the breadth of goals addressed beyond the criminal justice system a

133

model including inclusive leadership, instrumental leadership and the council
characteristics was performed. No predictors emerged as signiﬁcant (see Table 41).
Again, consistent with the individual models, these ﬁndings suggest that the council
characteristics investigated in this study do not explain variation in the breadth of goals
councils address beyond the criminal justice system.
Modeling the Accomplishment of Council Goals
Estimating Models for Each Independent Variabla:

To complete the analyses required to address Research Questions 4 and 5 (To
what degree is council climate related to council effectiveness? and To what degree are
council characteristics related to council effectiveness?) a series of standard regression
analyses were performed to relate council climate and characteristics to the ﬁnal indicator
of council effectiveness -- the degree to which council goals were accomplished both
within and beyond the criminal justice system. These analyses included each of the climate
variables (i.e., shared power in decision-making, effective conﬂict resolution, effective
leadership and shared mission) and the council characteristics (i.e., breadth of council
membership, breadth of active membership, formality of structure, presence of external
support). Given the relatively small number of councils (N = 40) that had adequate data
to be included in these analyses, the degree to which goals were accomplished within the
criminal justice system and the degree to which goals were accomplished beyond the
criminal justice system were regressed onto each predictor separately. Following this,
combined models were performed to identify the best model of the degree to which

councils’ goals have been accomplished.

134

Table 40

Breadth of Criminal Justice Goals Regressed onto Council Climate and Characteristic

 

V 'bles =40

 

 

Variables B _S_E B B m
Jone-tailed)

Intercept .3 94 .5 16 .226
Instrumental Leadership .254 . 177 .582 .08 1
Conﬂict Resolution .109 .109 .192 .164
Inclusive Leadership -.300 .235 -.581 .894
Breadth of Membership .295 .292 .184 320*
Breadth of Active Participation .002 .254 .002 993"
Formality of Council Structure -.094 .238 -.077 697*
External Support .020 .049 .088 .344
Council Age ’ .002 .001 .268 .085

 

Note. B3 = .21; * = two-tailed p-value.

135

Table 41

Braadth pf goals Bayond the Criminal Justica System Ragpessed onto Council Climapa

Ed Characteristic Variables (N = 4m

 

 

 

Variables B _S_ B B -value
(one-tailed)

Intercept .872 .483 .041
Instrumental Leadership . 177 . I67 .434 . 149
Inclusive Leadership -.233 .21 1 -.485 .860
Breadth of Membership -.008 .277 -.005 .977“
Breadth of Active Participation -.123 .245 -.093 .617 *
Formality of Council Structure .177 .224 .155 436*
External Support .003 .047 .013 ' .476
Council Age .002 .001 .342 .042

 

Note. 33 = .12; * = two-tailed p-value.

136

Cpnﬂict Resolution

Effective conﬂict resolution was not signiﬁcantly related to the degree to which
criminal justice council goals were accomplished (B = .02, p = .45, one-tailed). However,
there was a trend toward signiﬁcance for the degree to which goals were accomplished
beyond the criminal justice system (B = .24, p < .10, one-tailed; see Tables 42 and 43).
This indicates that councils which handle conﬂict effectively are no more or less likely to
accomplish their goals in the criminal justice arena, but may be more likely to accomplish
their goals in areas outside of the criminal justice system.
BBared Power in Decision-Magipg

Shared power in decision-making was signiﬁcantly related to the degree to which
councils’ goals were accomplished both within (B = .40, p < .01, one-tailed) and beyond
the criminal justice system (B = .29, p < .05, one-tailed; see Tables 44 and 45). This
indicates that councils who shared power in decision-making were more likely to
accomplish their goals both within and outside of the criminal justice system.
Shared Mission

The presence of a shared mission was signiﬁcantly related to the degree to which
goals beyond the criminal justice system were accomplished (B = .44, p < .01, one-tailed).
However, there was only a trend toward signiﬁcance regarding the degree to which
criminal justice goals were accomplished (B = .24, p < .10, one-tailed; see Tables 46 and
47). This indicates that councils with a shared mission were more likely to achieve their

goals beyond the criminal justice system as well as within the criminal justice system to

501116 extent.

137

Table 42

Degree to Which Criminal Justice Goals Accomplished Regressed onto Effective Conﬂict

 

Resolution Strategies (N = 40)

 

 

 

Variables B _S_B B B p-value
ﬁne-tailed)

Intercept 3.214 1.247 .007

Effective Conﬂict Resolution .059 .354 .021 .446

Council Age .007 .004 .281 .040

Nate. 3: = .08

Table 43

Degree to Which Goala Beyond the Criminal Justice System Accomplished Regpessed

onto Effective Conﬂict Resolution Stra_t§gies (N = 40)

 

 

 

Variables B BB B B p-value

(one-tailed)
Intercept 1 .822 1.406 . 102
Effective Conﬂict Resolution .604 .399 .237 .070
Council Age .003 .004 .l 16 .232
Note. B} = .07

138

Table 44

Daggee to Which Criminal Justice Goals Accomplished Regressed onto Shared Power in

 

 

 

 

Decision-Makin = 40

Variables B S_E B B p-vaIue
(one-tailed)

Intercept -.483 1.489 .3 74

Shared Power of Decision-Making .939 .356 .397 .006

Council Age .004 .004 .159 . 149

Note. Bf = .22

Table 45

Bagree to Which Goals Beyond the Criminilustice System Accomplished Rpgressed

 

 

 

 

onto Shared Power in Decision Making (B = 40)

Variables B _B B B p-value
(one-tailed)

Intercept .742 1 .807 .342

Shared Power in Decision-Making .765 .432 .288 .043

Council Age .001 .004 .040 .403

Note. Bf = .09

139

Table 46

Deggee to Which Criminal Justice Goals Accomplished Reggessed onto Shared Mission

 

 

 

 

= 40

Variables B S_E B B p-value
(one-tailed;

Intercept 1.416 1.325 .146
Shared Mission .454 .299 .234 .069
Council Age .006 .004 .238 .066
Note. B: = .13
Table 47

_D_egree to Which Goals Beyond the Criminal Justice System Accomplished Regressed

 

onto Shmd Mission (N = 40)

 

 

 

Variables B S_E B B p-value

(one-tailed)
Intercept -.306 1.422 .416
Shared Mission .971 .321 .444 .002
Council Age .001 .004 .045 .382
Note. B2 = .21

140

Lgdership

Leadership being inclusive (i.e., encouraging the active participation of all
members) was signiﬁcantly related to the degree to which goals were accomplished both
within the criminal justice system (B = .38, p < .05, one-tailed) and beyond the criminal
justice system (B = .57, p < .001, one-tailed; see Tables 48 and 49). In addition,
instrumental leadership was signiﬁcantly related to the degree to which goals were
accomplished within the criminal justice system (B = .46, p < .01, one-tailed) as well as
beyond the criminal justice system (B = .50, p < .01, one-tailed; see Tables 50 and 51).
This indicates councils which have strong inclusive and instrumental leadership are more
likely to accomplish their goals both within and beyond the criminal justice system.

Estimating a Combined Climate Model

Given that a number of climate variables were signiﬁcantly related to the degree to
which goals were accomplished both within and beyond the criminal justice system, to
estimate ﬁnal climate models hierarchical regression was employed. Given that the
purpose of this analysis was to identify the most parsimonious model as well as the most
informative model, independent variables were entered into the model in stages starting
with the strongest predictor.“

To model the degree to which council goals were accomplished with the criminal
justice system, independent variables were entered as follows: instrumental leadership,

shared power in decision-making, inclusive leadership, shared mission and effective

 

21The covariate council age was entered into the ﬁrst block.

141

Table 48

Daggee to Which Criminﬂ Justice Goala Acpomplished Reggassed onto Inclusive

L a ershi = 40

 

 

 

Variables B BB B B p-valua
(one-tailed)

Intercept -. 124 l .458 .467

Inclusive Leadership .786 .321 .380 .010

Council Age .003 .004 . 144 . 181

Note. B_2 = .21

Table 49

Degr_‘ee to Which Goals Beyond the Criminal Justice System Accomplished Regressed

onto Inclusive Leadership (N = 40)

 

 

 

Variables B _S_E B B p-_va_lu_a

(one-tailed)
Intercept -2.036 1.535 .097
Inclusive Leadership 1.330 .338 .572 .000
Council Age -.002 .004 -.O79 .704
m B2 = .30

142

Table 50

Begree to Which Criminal Justice Goals Accomplished Regressed onto the Inatmmenjg

 

 

 

Leadershi = 40

Variables B BB B B p-value
(one-tailed)

Intercept -.O97 1. l 15 .466

Instrumental Leadership .818 .255 .460 .002

Council Age .004 .003 . 158 . 140

Note. B2 = .28

Table 51

Degree to Which (ﬁls Beyond the Criminal Justice System Accomplished Regressed

onto Instrumental Leadership (111 = 40)

 

 

 

 

Variables B B_ B B p-value

(one-tailed)
Intercept -.33 1 1.277 .399
Instrumental Leadership .994 .292 .497 .001
Council Age .000 .004 -.005 .5 14
Note. Bf = .25

143

conﬂict resolution (see Tables 52a - 52d). The ﬁrst block indicated that instrumental
leadership was a signiﬁcant predictor of the degree to which goals were accomplished in
the criminal justice system (B = .46, p < .01, one-tailed). While instrumental leadership
was still signiﬁcant in the second block (B = .35, p < .05, one-tailed), shared power in
decision-making was not (B = .36, p = .18, one-tailed). However, in the third block, while
inclusive leadership was not a signiﬁcant predictor of the degree to which criminal justice
goals were accomplished (B = -.44, p = .88, one-tailed), instrumental leadership was still a
signiﬁcant predictor (B = .67, p < .05, one-tailed) and a trend toward signiﬁcance emerged
for shared power in decision-making (B = .28, p < .10, one-tailed). This indicated that
when the variance associated with inclusive leadership was accounted for in the degree to
which council goals were accomplished within the criminal justice system and in shared
power in decision-making, the unique variance associated with shared power in decision-
making emerged as signiﬁcant. Similar to the ﬁndings regarding the breadth of goals
addressed, councils who have leaders with a greater task orientation are more likely to
accomplish their goals to reform the criminal justice system. In addition, the unique
variance associated with shared power in decision-making (i.e., the variance unshared with
inclusive leadership) may reﬂect not only members’ input being encouraged by the leader,
but their input being applied to the decision-making processes of the councils. This
suggests that councils where decision-making power is shared broadly were more likely to
accomplish their goals to reform areas of the criminal justice system. In the fourth and
ﬁnal blocks shared mission and conﬂict resolution did not contribute signiﬁcantly to the

models.

144

Table 52a

Hiararphigal Regression: Degree to Which Criminal Justice Goals Accomplished

 

Regressed onto Council Climate Va_riables (N = 40)

 

 

Variables B _S_ B B p-value A_Rf
(one-tailed) (p-value)

Block 1 .28
(.00)

Intercept -.097 l. l 15 .466

Council Age .004 .003 . 158 . 140

Instrumental Leadership .818 .255 .460 .002

Block-2' ........................................................................................................................................ .02. .......
(.36)

Intercept -.974 1 .462 .255

Council Age .003 .004 . 133 . 185

Instrumental Leadership .628 .328 .353 .032

Shared Power in Decision- .410 .441 .173 .180

Making

 

145

Table 52b

Hierarchical Regression: Degree to Which Criminal Justice Goals Accomplished

R_e_gressed onto Council Climate Vamibles (Continued) (N = 40)

 

 

Variables B _B B B p-value
(one-tailed) (p-value)
Block 3 .03
Intercept -.327 1.548 .417
Council Age .004 .004 . 175 . 124
Instrumental Leadership 1 . 194 .571 .671 .022
Shared Power in Decision- .662 .486 .280 .091
Making
Inclusive Leadership -.9 I 8 .759 -.444 .880

AB}.

(.23)

 

146

Table 52c

Hiararchipal Regression: Deggea to Which Criminal Justipa Baals Accomplished

Begressed onto Council Climate Variables (Continued) (B = 40)

 

 

 

Variables B BB B B p-value A_R2
(one-tailed) (24131116)

Black 4 .00
(.58)

Intercept -.246 1.570 .43 8

Council Age .004 .004 .164 . 145

Instrumental Leadership 1.192 .576 .670 .023

Shared Power in .722 .502 .305 .080

Decision-Making

Inclusive Leadership -.766 .812 -.370 .824

Shared Mission -.231 .409 -.1 19 .711

 

147

Table 52d

Hiararchical Regression: Degree to Which Criminal Justice Goals Accomplished

 

Ragressed onto Council Climate Variables (Continued) (N = 40)

 

 

Variables B S_E B B p-value ﬂ
(one-tailed) (p-value)

Block 5 .02
(.35)

Intercept .079 1 .609 .481

Council Age .003 .004 . 139 . 188

Instnimental Leadership 1 .075 .590 .604 .039

Shared Power in .772 .505 .326 .068

Decision-Making

Inclusive Leadership -.525 .852 -.254 .729
Shared Mission -.206 .41 l -.106 .690
Conﬂict Resolution -.355 .373 -.156 .826

 

No model beyond the ﬁrst explained a signiﬁcant proportion of the variance in the
degree to which goals have been accomplished within the criminal justice system

indicating the ﬁrst block of this hierarchical regression is the most parsimonious model.

148

Taken together, these models suggest that instrumental leadership is the strongest
predictor of the degree to which councils have accomplished their goals to reform areas of
the criminal justice system. However, in the interest of examining all contributing
variables and building the most informative model, shared power in decision-making and
inclusive leadership (because of its suppression effect) were retained in the ﬁnal model
utilized later to examine models including council characteristics.

For the degree to which goals have been accomplished in areas beyond the criminal
justice system a different picture emerged from a hierarchical regression including all
climate variables. Independent variables were entered into the model as follows: inclusive
leadership, instrumental leadership, shared mission, shared power in decision-making, and
conﬂict resolution (see Tables 53a - 53d). In the ﬁrst block, inclusive leadership was a
signiﬁcant predictor of the degree to which goals were accomplished beyond the criminal
justice system (B = .57, p < .001, one-tailed). With the introduction of instrumental
leadership in the second block, inclusive leadership was still signiﬁcant (B = .57, p < .05,
one-tailed); however, instrumental leadership was not (B = .01, p = .49, one-tailed). The
introduction of shared mission in the third block, shared power in decision-making in the
fourth block and conﬂict resolution in the ﬁnal block did not contribute signiﬁcantly to the
model. However, it is important to acknowledge that with each of these climate
predictors added to the model and in the ﬁnal block model, inclusive leadership was

reduced to a trend (B = .65, p < .10, one-tailed).

149

Table 53a

Hiergchical Regression: Degree to Which Goals Beyond the Briminal Justice Svstem

Agpmplished Regressed onto Council Climate Variables (N = 40)

 

 

 

 

Variables B BB B B m

BELL

Intercept -2. 036 1 .53 5 .097

Council Age -.002 .004 -.079 .704

Inclusive Leadership 1.330 338 .572 .000

Block“.2. ...................................................................................................................................................
Intercept -2.028 I .594 . 106

Council Age -.002 .004 -.078 .699

Inclusive Leadership 1.314 .770 .565 .048

Instrumental Leadership 1 .495”2 .641 .007 .491

A132.

(one-tailed) (R'Valllﬂ
.30

(.00)

 

150

Table 53b

Hierarchical Regpession: Degree to Which Goals Beyond the Criminal Justice System

Accomplished Regressed onto Council Climate Variables (ContinuedMN = 4m

 

 

Variables B S_E B B p-value _A_R_2
(one-tailed) (p-value)

Block 3 .00
(.61)

Intercept -2. 187 1.639 .096

Council Age -.002 .004 -.069 .674

Inclusive Leadership 1.120 .863 .482 .102

Instrumental Leadership .0 1 9 .647 .01 0 .488

Shared Mission .232 .449 .106 .304

 

151

Table 53c

Hiararchical Ragpeaaipn: Deggee to Which Goals Beyond tha Criminal Justice System

 

Accomplished Regressed onto Council Climate Varatbles (Continued) (N = ﬂ)

 

 

Variables B BB B B p-value A_R2
(one-tailed) (p-value)

Block 4 .03
(.23)

Intercept -1.404 1.753 .215

Council Age -.001 .004 -.053 .636

Inclusive Leadership 1.476 .907 .635 .057

Instrumental Leadership -.002 .643 -.001 .50 1

Shared Mission .3 50 .457 .160 .225

Shared Power in Decision -.678 .560 -.255 .883

Making

 

152

Table 53d

Hierarchical Reggession: Degree to Which Goals Beyond the Criminal Justice Svstem
Appomplished Regressed onto Council Climate Variables (Bontinued) (ﬂ = 40)

 

 

Variables B S_E B B p-value B133
(one-tailed) (n-value)

Block 5 .00
(.93)

Intercept -l .368 1.819 .229

Council Age -.002 .004 -.056 .638

Inclusive Leadership 1.502 .963 .646 .064

Instrumental Leadership -.0 14 .667 -. 007 .508

Shared Mission .353 .465 .162 .227

Shared Power in Decision -.673 .571 -.253 .876

Making

Conﬂict Resolution -.039 .422 -.015 .536

 

Taken together these models suggest that in the presence of the other climate variables
only the unique variance associated with having inclusive leadership was an important

predictor of accomplishing goals in areas beyond the criminal justice system.

153

Again, no model beyond the ﬁrst accounted for a signiﬁcant proportion of the
variance in the degree to which goals were accomplished beyond the criminal justice
system nor did any other independent variables emerge as signiﬁcant or with a trend
toward signiﬁcance. In this case, the most parsimonious and most informative climate
model were the same. This ﬁnding indicates that councils with inclusive leadership were
more likely to achieve their goals to reform areas outside of the criminal justice system.

AddinLCouncil Characteristics to the Model

 

Breadth of Council Membership and Active Membership

The breadth of council membership was not signiﬁcantly related to the degree to
which councils achieved their goals within the criminal justice system (B = .11, p = .47,
two-tailed) nor beyond the criminal justice system (B = .20, p = .22, two-tailed; see Tables
54 and 55). However, the breadth of active membership was signiﬁcantly related to the
degree to which councils achieved their goals within the criminal justice system (B = .35, p
< .02, two-tailed) and beyond the criminal justice system (B = .40, p < .01, two-tailed; see
Tables 56 and 57). This suggests that it is not simply having ‘ofﬁcial’ membership that
inﬂuences the degree to which councils are accomplishing their goals, but the degree to
which membership is active. Councils with a greater proportion of active members were
more likely to achieve their goals both within and beyond the criminal justice system.
Bxﬂpal Support for Council Activities

The presence of support for council activities was signiﬁcantly related to the
degree to which councils accomplished their criminal justice goals (B = .49, p < .01, one-

tailed) and to the degree to which they accomplished their goals beyond the criminal

154

Table 54

Degree to Which Criminal Justice Goals Accomplished Regressed onto the Breadth of

Qf’ﬁcig Membership (B = 40)

 

 

 

Variables B _S__ B B p-value
(two-tailed)

Intercept 2.843 .765 .001

Breadth of Council Membership .744 1.024 .1 I4 .472

Council Age .007 .004 .299 .065

Note. R_2 = .10

Table 55

Degree to Which Goals Beyond the Criminal Justice System Accomplished Regressed

 

onto the Breadth of Ofﬁcijal Membership (N = 40)

 

 

 

Variables B B_ B B p-value

(two-tailed)
Intercept 2.871 .880 .002
Breadth of Council Membership 1.470 1.178 .200 .220
Council Age .004 .004 . 149 .369
Note. B2 = .06

155

Table 56

Begpee to Which Criminﬂustice Goals Accomplished Regressed onto the Breadth of

Aptive Membership (ﬂ = 40)

 

 

 

Variables B BB B B p-value
(two-tailed)

Intercept 1.836 .701 .013

Breadth of Active Membership 2.036 .862 .353 .024

Council Age .005 .004 .218 .154

Note. _B_2 = .20

Table 57

Degree to Which Goals Beyond the Criminal Justice System Accomplished Regressed

onto Breadth of Active Members_hip (N = 40)

 

 

 

Variables B S_ B B p-value

Qwo-tailed)
Intercept l .954 .806 .020
Breadth of Active Membership 2.584 .991 .399 .013
Council Age .001 .004 .051 .741
Note. B2 = .17

156

justice system (B = .44, p < .01, one-tailed; see Tables 58 and 59). This indicates that
councils with more support ﬁ'om their external environment were more likely to
accomplish their goals both within and beyond the criminal justice system.
Formalig of Council Structure

The degree of formality in the council structure was not related to the degree to
which councils accomplished their criminal justice goals (B = .09, p = .57, two-tailed).
However, there was a trend toward signiﬁcance regarding the degree to which councils
addressed their goals beyond the criminal justice system (B = .27, p = .10, two-tailed; see
Tables 60 and 61). This suggests that councils with a more formal structure are not
necessarily more likely to accomplish their goals in the criminal justice system, but may be
more likely to accomplish their goals in areas outside of the criminal justice system.

Estimating a Combined Climate/Chﬂcteristics Model

To estimate ﬁnal models including both council climate and characteristic
variables, a second set of hierarchical regressions was performed. To model the degree to
which criminal justice goals were accomplished, the ﬁnal climate model, including
instrumental leadership, shared power in decision-making and inclusive leadership
(because of suppression effect) was entered into the ﬁrst block. Each council
characteristic from the strongest to weakest predictor (i.e., external support, breadth of
active membership, breadth of official membership and formality of council structure)
were entered into the in the second, third, fourth and ﬁfth blocks. See Tables 62a to 62c.

In the second block, external support for council activities was signiﬁcantly related

to the degree to which criminal justice goals were accomplished (B = .38, p < .05, one-

157

Table 58

mgree to Which Criminal Justice Goals Accomplished Regressed onto External Support

for Bouncil Activities (B = 40)

 

 

 

Variables B BB B B p-value
(one-tailed)

Intercept 1 .775 .534 .001

External Support .450 .132 .488 .001

Council Age .003 .003 . 136 . 174

Note. B: = .30

Table 59

Bagree to Which Ggls Beyond the Criminal Justice Svstem Accomplished Regressed

onto External Support for Council Activities

 

 

 

Variables B BB B B p-value

(one-tailed)
Intercept 2.254 .64 1 .000
External Support .458 . 158 .443 .003
Council Age -.009 .004 -.003 .491
Note. B2 = .20

158

Table 60

Criminal Justice Goals Accomplished Regpessed onto Forrnalig of Council Structure

 

 

 

= 40

Variables B S_E B B p-value
(two-tailed)

Intercept 3. 1 56 .491 .000
Formality in Council Structure .449 .793 .090 .574
Council Age .007 .004 .303 .065
Note. Rf = .09
Table 61

_ijals Beyond the Criminal Justice Svstem Accomplished Regressed onto Formality of

Council Structure Q1 = 40)

 

 

 

Variables B BB B B p-value

(two-tailed)
Intercept 3.148 .553 .000
Formality in Council Structure 1.485 .893 .266 .104
Council Age .005 .004 . 192 .237
Note. Bf = .08

159

 

tailed) as well as instrumental leadership (B = .64, p < .05, one-tailed) and a trend toward
signiﬁcance for shared power in decision-making (B = .26, p < .10, one-tailed).

In the third block the breadth of active participation emerged as a signiﬁcant
predictor (B = .32, p < .05, two-tailed) of the degree to which councils’ criminal justice
goals were accomplished along with instrumental leadership (B = .70, p < .05, one-tailed)
and shared power in decision-making (B = .36, p < .05, one-tailed). Given that shared
power was previously a trend, the strengthening of this relationship suggests that the
breadth of active participation acts as a suppressor. That is, when the variance associated
with the breadth of active participation was accounted for in the dependent variable and
shared power in decision-making, the relationship between shared power in decision-
making and the degree to which goals were achieved in the criminal justice system was
stronger. This suggests that, accounting for the variation associated with having broader
participation, shared power in decision-making is more strongly related to the
accomplishment of criminal justice goals. However, in the third block, external support is
reduced to a trend toward signiﬁcance (B = .26, p < .10, one-tailed). This is likely due to
the shared variance between total participation and external support. Those councils with
greater participation also have greater external support (r = .37, p < .05). Thus, the
unique variance associated with external support was not as strong in the presence of the
breadth of active participation.

In the fourth and ﬁfth blocks the breadth of official membership and the degree of

formality in the council structure did not signiﬁcantly contribute to the model. Thus, the

160

ﬁnal model (the third block) includes instrumental leadership, shared power in decision-
making, external support and the breadth of active participation. This suggests that
councils with stronger instrumental leaders, greater shared power in decision-making,
more external support and a greater breadth of active participants are more likely to

accomplish their goals to create reform within the criminal justice system.

Table 62a
Hierﬂhicaﬂlegression: Degree to Which Criminal Justice Goals Accomplished

Bagressed onto Council Climate and Chtﬂcteristic Vambles (N = 40)

 

 

Variables B BB B B p-value _A_R_2
(one-tailed) (p-value)

Block 1 .33
(.01)

Intercept -. 579 1.605 .361

Council Age .004 .004 . 173 . 130

Instrumental Leadership 1.098 .593 .618 .036

Shared Power in .682 .490 .285 .087

Decision-Making

Inclusive Leadership -.795 .787 -.379 .840

 

161

 

Table 62b

Hierarchical Regression: Degree to Which Briminal Justice (gala Accomplished

Regressed onto Council Climate and Characteristics Vgiables (Continued) (N = 40)

 

 

 

Variables B BB B B p-value _AE
(one-tailed) (p-value)

Block 2 .10
(.03)

Intercept .138 1.541 .465

Council Age .003 .003 .127 . 191

Instrumental Leadership 1.138 .558 .640 .025

Shared Power in Decision- .619 .462 .258 .095

Making

Inclusive Leadership -1.21 1 .761 -.578 .939

External Support .351 .149 .378 .013

 

I62

Table 62c

Hiararchical Regression: Deggee to Whiph Criminal Juatipa Baals Accomplished
Ragpassag onto Coancil Blimata ang Bhagactariatics Vap'ables (@ntinued) (N = 4m

 

 

 

Variables B S_ B B p-value AR:
(one-tailed) (n-value)

Block 3 .08
(.03)

Intercept -1.281 1.574 .211

Council Age .002 .003 .093 .248

Instrumental Leadership 1 .23 7 .527 .696 .013

Shared Power in Decision- .865 .448 .361 .031

Making

Inclusive Leadership -1.432 .723 -.684 .972

External Support .238 .149 .256 .059

Breadth of Active 1.836 .791 .319 .027*

Participation

 

Note. * = two-tailed p-value.

163

 

 

Table 62d

Hieragchical Regression: Degree to Whiph Criminal Justice Goala Accomplished

Regressed onto Council Climate and Characteristics Variables (Continued) (N = 40)

 

 

 

Variables B BB B B p-value ﬂ:-
(one-tailed) (p-value)

Black 4 .00
(.73)

Intercept -1.588 1.818 .195

Council Age .002 .003 .097 .243

Instrumental Leadership 1 . 194 .548 .672 .01 9

Shared Power in Decision- .900 .464 .375 .031

Making

Inclusive Leadership -1.397 .739 -.667 .966

External Support .225 .155 .243 .078

Breadth of Active 1.863 .806 .324 027*

Participation

Breadth of Membership .308 .876 .047 .728*

 

Note. * = two-tailed p-value.

164

Table 62e

Hierarchical Regpassion: Degree to Which Criminal Justipe @als Accomplished

 

 

 

Re esse ont o ncil lim te har eristics V ' 1 Continued = 40
Variables B 5a B B p-value 4133
(one-tailed) (p-value)

Block 5 .014
(.35)

Intercept -l.550 1.821 .201

Council Age .001 .003 .054 .355

Instrumental Leadership 1 .232 .550 .693 .016

Shared Power in Decision- .848 .468 .354 .040

Making

Inclusive Leadership -1 .405 .741 -.671 .966

External Support .253 .158 .273 .060

Breadth of Active 2.022 .824 .351 020*

Participation

Breadth of Membership .577 .922 .088 536*

Formality in Council -.703 .742 -. I42 .350“

Structure

 

Note. * = two-tailed p-value.

165

This process was repeated to model the degree to which council goals for reform
were achieved beyond the criminal justice system. A hierarchical regression beginning
with inclusive leadership in the ﬁrst block (inclusive leadership was the only signiﬁcant
climate predictor), followed by external support in the second block, the breadth of active
participation in the third block, the degree of formality in the council structure in the
fourth block and the breadth of council membership added in the ﬁfth block was
performed. See Table 63a - 63d for a summary of these ﬁndings.

In the second block, while inclusive leadership remained signiﬁcant (B = .47, p <
.01, one-tailed) external support did not emerge as signiﬁcant (B = .21, p = .11, one-
tailed) although it approached a trend. Similarly, in the third block, the breadth of active
participation B = .31, p < .05, two-tailed) and inclusive leadership (B = .49, p < .01, one-
tailed) were signiﬁcant, while external support was not (B = .10, p = .28, one-tailed). This
is likely due to the shared variance of external support with both inclusive leadership (r =
.53, p < .001) and the breadth of active participation (r = .37, p < .05). As the sole
predictor in the model, external support is signiﬁcantly positively related to the degree to
which goals were accomplished outside of the criminal justice system, but the unique
variance associated with external support in the presence of the other predictors was not.
In the fourth and ﬁfth blocks, the degree of formality in the council structure and the
breadth of official membership did not signiﬁcantly contribute to the overall model. Taken

together these ﬁndings suggest that councils with leaders who encouraged collaboration

I66

and input from all members and a broad range of active members were more likely to
accomplish their goals to create reforms beyond the criminal justice system.

Table 63a

Hiargchical Regression: Degree to Which @als Beyond tha Briminal Justice System

 

Accomplished Regpessed onto Council Climate and Characteristic Variables (N = 40)

 

 

 

Variables B BB B B p-value AB}
(one-tailed) (2-value)

Block 1 .31
(.00)

Intercept -2.247 1 .568 .008

Council Age -.002 .004 -.077 .700

Inclusive Leadership 1.370 .344 .581 .000

Blck203 ...........
(.21)

Intercept -1.868 1.584 .123

Council Age -.003 .004 -. 104 .759

Inclusive Leadership 1.112 .398 .472 .004

External Support .220 . 174 .210 . 107

 

167

Table 63b

Hierarchical Regression: Degrpe to Which Goals Beyond the Criminal Justice System

Apﬂmpliahed Regressed onto Council Climate and Characteristic Variables (Continued)

 

 

= 40
Variables B BB B B p-value A_R2
(one-tailed) (g-value)

Block 3 .08
(.03)

Intercept -3. 1 15 1.608 .03 1

Council Age -.004 .004 -. 136 .831

Inclusive Leadership 1.152 .378 .489 .002

External Support . 101 . 173 .097 .281

Breadth of Active 1.989 .902 .307 034*

Participation

 

Note. * = two-tailed p-value.

168

ix .iiill. l

..'1|l.. III .lllil' 1.11.1 iii

 

Table 63c
Hierarchical Regression: Dagree to Which Goals Beyond the Criminal Justice System

Accomplished Regressed onto Council Climate and Chaaacteristic Variables (Continued)

 

 

= 40
Variables B ﬂ: B B p-value BE
(one-tailed) (p-value)

Block 4 .01
(.52)

Intercept -3 .21 I 1.628 .029

Council Age -.003 .004 -.103 .750

Inclusive Leadership 1.155 .382 .490 .003

External Support .075 . 179 .072 .339

Breadth of Active 1.870 .928 .288 052*

Participation

Formality of Council .525 .804 .094 .518*

Structure

 

Note. * = two-tailed p-value.

169

Table 63d

Hierarchical Regression: Degree to Which Goals Beyond the Criminal Justice System

 

Accomplished Regressed onto Cpuncil Climate and Charapteristic Variables (N = 40)

 

 

 

Variables B BB B B p-value BIL2
(one-tailed) ( g-value)

Block 5 .02
(.26)

Intercept -4.020 1 .767 .015

Council Age -.003 .004 -.104 .754

Inclusive Leadership 1.187 .381 .503 .002

External Support .059 .180 .047 .394

Breadth of Active 2.019 .933 .311 038*

Participation

Formality of Council .176 .855 .032 838*

Structure

Breadth of 1.195 1.040 .163 259*

Membership

 

Note. * = two-tailed p-value.

170

Examining Variability in Perceptual Variables as a Function of Organizational Role

To address research question 6 (To what degree does organizational role explain
variation in council members’ perceptions of council effectiveness and council climate?),
organizational role was included as a sole Level I predictor of each of the perceptual
variables included in this study (see Tables 64 and 65). These analyses suggest that
organizational role explains variation in council members’ perceptions of council
effectiveness, shared power in decision-making, the presence of a shared mission,
instrumental leadership and inclusive leadership to some extent. However, for each model
the random effects associated with the intercept were signiﬁcant, indicating that there is
considerable variability across settings in each of these variables that is unexplained by
accounting for variation due to organizational role.

To conduct these analyses, domestic violence advocates were used as the baseline
or comparison group. Therefore, all unstandardized coefﬁcients reﬂect the degree to
which each group differed from domestic violence providers in their ratings of council
effectiveness and climate; for the most part, few differences emerged.

For perceived council effectiveness, judges were more likely than domestic
violence providers to rate councils as effective (Level I Coefficient = .53, p < .05), as were
prosecuting attorneys (Level I Coefficient = .45, p < .05). There was a trend toward
signiﬁcance for clergy (Level I Coefficient = .49, p < .10). This indicated that judges,
prosecuting attorneys and to some extent clergy viewed councils as more effective than

did domestic violence advocates.

171

 

Table 64

Parceived Effectiveness Regressed and Climate Variables Reggessed onto Organizational

 

 

 

 

Bola
Organizational Role Perceived Effective Conﬂict Shared Power in
Effectiveness Resolution Decision-
Making
Unstandardized Coefficient (Standard Error)
Intercept 4.20M“ (. 12) 3.52”“ (.11) 4.17““ (.11)
Human Service Provider .16 (.13) .06 (.13) .20 (. 13)
Clergy .491”2 (.31) .22 (.31) .02 (.31)
Judge .53" (.21) .19 (.21) .55" (.21)
Law Enforcement .16 (.13) -.18 (.13) .03 (.13)
Probation .11 (.19) -. 12 (.19) .06 (.19)
Prosecuting Attorney .45" (. 18) -.00 (.18) .34T2 (.18)
Educator .07 (.24) -.24 (.24) .31 (.24)
Community Member -.05 (.28) -.28 (.28) -.11 (.28)
Court .27 (.21) -.28 (.22) .22 (.22)
Attorney .19 (.24) -.22 (.24) .09 (.24)
Batterers’ Intervention -.59 (.45) .30 (.46) -.51 (.46)
Victim’s Advocate in .28 (.35) .10 (.35) .40 (.36)

Criminal Justice System

 

Variance Components

 

oz .75 .77 .78

Intercept (too) 18““ .1 IM" .08"""

Note: two-tailed T2p-value < .10, Ap - value < .05, Mp-value < .01, ’W‘p-value < .001.

 

172

 

For effective conﬂict resolution, there were no signiﬁcant differences between
domestic violence providers and any of the other stakeholder groups. However, with
regard to shared power in decision-making, judges were more likely than domestic
violence service providers to view their councils as having shared power in decision-
making (Level I Coefficient = .55, p < .05), while prosecuting attorneys demonstrated a
trend toward signiﬁcance (Level I Coefﬁcient = .34, p < .10). Similar to perceived
effectiveness overall, judges and to some extent prosecuting attorneys rated their councils
as sharing power in decision-making to a greater degree than did domestic violence
providers.

Human service providers were more likely to view their council as having a shared
mission (Level I Coefficient = .40, p < .01), while judges (Level I Coefﬁcient = .41, p <
.10) and educators demonstrated a trend toward signiﬁcance (Level I Coefﬁcient = .40, p
< .10). This indicated that human service providers and to some extent judges and
educators viewed their councils as having a shared mission to a greater degree than did
domestic violence advocates.

While no signiﬁcant differences emerged with regard to council members’ rating of
instrumental leadership, clergy (Level I Coefficient = .53, p < .10) and educators (Level I
Coefficient = .45, p < .10) emerged with trends toward signiﬁcance. This indicated that
clergy and educators may be somewhat more likely to view their councils as having strong

instrumental leadership than did domestic violence advocates.

173

 

Table 65

Perceived Effectiveness Regpessed and Climate Variables Reggessed onto Organizational
Role (Continued )

 

Organizational Role Shared Instrumental Inclusive
Mission Leadership Leadership

 

Unstandardized Coefficient (Standard Error)

 

 

 

Intercept 4.28 (.12)~\A 4.32m (.12) 4.61““ (.11)
Human Service Provider .40M (.13) 5 .36M (.13) E .32" (.12)
Clergy .48 (.31) .53T2 (.32) .20 (.30)
Judge .41T2 (.21) .29 (.22) .37“2 (.21)
LE .13 (.13) .13 (.14) .03 (.13)
Probation .23 (. 19) .04 (.20) .05 (.19)
Prosecuting Attorney .44" (.18) .18 (.19) .33T2 (.18)
Educator .40T2 (.24) .45T2 (.25) .20 (.24)
Community Member .28 (.28) -. 17 (.30) —.03 (.27)
Court .13 (.22) .33 (.23) .37T2 (.21)
Attorney -.16 (.24) .05 (.25) -.04 (.24)
Batterer’s Intervention .14 (.46) -.35 (.48) -.84T2 (.45)
Victim’s Advocate in .37 (.36) .06 (.37) .23 (.35)
Criminal Justice System
Variance Components
0’2 .76 .84 .73
Intercept (too) 17““ .20""" .13"""

 

Note: two-tailed T2p-value < .10, "p - value < .05, Mp-value < .01, AAAp-value < .001.

 

 

Human service providers were more likely than domestic violence providers to
view their councils as having inclusive leaders (Level I Coefficient = .32, p < .05). Judges
(Level I Coeﬂicient = .53, p < .05) along with prosecuting attorneys (Level I Coefficient =
.53, p < .05), court personnel (Level I Coefficient = .53, p < .05) , and batterers’
intervention providers (Level I Coefficient = .53, p < .05) demonstrated trends toward
signiﬁcance. These ﬁndings indicate that human service providers and to some extent
judges, prosecuting attorneys and court personnel viewed their councils as having
inclusive leadership to a greater extent than domestic violence service providers.
However, these ﬁndings suggest that batterers’ intervention providers are less likely than
domestic violence service providers to view their councils as having inclusive leadership.
Taken together, the ﬁndings regarding organizational role indicate that for the most part,
different stakeholder groups (i.e., organized by role) do not differ appreciably from
domestic violence service providers regarding their assessment of the setting. However,
judges, prosecuting attorneys, and human service providers do appear to view councils
more favorably when compared to domestic violence providers.

Power Analysis

This study had varied power to detect signiﬁcant relationships. The number of
independent variables included in the analyses ranged from two to eight and the effect
sizes ranged from R2 = .05 to R2 = .33. Using Cohen’s (1988) conservative estimates,
power was calculated for a variety of circumstances based on three criterion: a) effect size,

b) number of independent variables, and c) sample size. More speciﬁcally, power estimate

175

 

were calculated for small (R2 = .05), medium (R2 = .13) and large (R2 = .30) effect sizes
with two, four and eight independent variables with a sample size of 40.22 These
calculations indicate that with a sample size of 40 and a small effect size statistical power
to detect signiﬁcant relationships was quite low (.21 with 2 independent variables, .16
with four independent variables, and .11 with eight independent variables). This indicates
that weak relationships were not likely to emerge as signiﬁcant in this study. While such
power was slightly improved with a medium effect size, it was still somewhat low (.54
with 2 independent variables, .41 with four independent variables, and .30 with eight
independent variables). However, with a sample size of 40 and a large effect size, power
to detect signiﬁcant relationships was adequate (.96 with 2 independent variables, .90 with
four independent variables, and .75 with eight independent variables).

In addition, when utilizing multi-level modeling techniques such as HLM,
determining the power is somewhat more complicated. While there are no clear
guidelines, Kreﬁ and DeLeeuw (1998) suggest that at least 20 groups are required to
examine Level 11 level relationships. By these standards this study had an adequate
number of Level 11 units (N = 38), but the speciﬁc power to detect small, medium and
large effects is difficult to estimate accurately. Overall, it is important to note that this
study had limited power to detect small and perhaps even moderate council level

relationships. Thus, ﬁndings of non-signiﬁcance should be interpreted with due caution.

 

22

All power estimates are approximate and reﬂect the mid-point of the range provided in
Cohen’s (1988) power table for an N = 20 and an N = 60.

176

 

DISCUSSION

Creating a Coordinated Community Response (CCR) to domestic violence
requires fundamental reforms in both the human service and criminal justice systems (Hart,
1995; Pence & Shepard, 1999). Scholars, funders and community members agree that
collaboration across the many stakeholder groups that play a role in responding to
domestic violence is central to these reform efforts (Hart, 1995; Pence & Shepard, 1999).
This study provides preliminary evidence that domestic violence coordinating councils can
facilitate such collaboration and that these councils are more likely to be effective when
they attend to their internal climate, as well as the degree to which they foster diverse
active membership and garner support from the community, funders, and policy makers.

Speciﬁcally, this study found that councils perceived as effective by council leaders
and members were more likely to have an internal climate characterized by effective-
conﬂict resolution, shared power in decision-making, a shared mission and effective '
leadership that embodied both an inclusive and task-oriented leadership style. Further,
there is evidence that these features of the council climate are differentially related to
different indicators of council effectiveness. For example, the degree to which councils
had a shared mission was related to the degree to which they were perceived as effective
by council members, but was not related to the degree to which goals were accomplished
within the criminal justice system according to council leaders. The ﬁndings from this

study also suggest that it is not only the internal climate of councils that contributes to

177

 

their effectiveness, but also the degree to which councils have fostered active member
participation and the degree to which they have garnered support from their environment
(i.e., ﬁinders and policy makers). Finally, in addition to the substantive ﬁndings regarding
the correlates of council effectiveness, this study raises important methodological issues
regarding the use of perceptual measures in the study of councils and the need to attend to
the multi-level nature of studying collaborative settings.
Council Effectiveness

Overall, there was consensus among council members and leaders that councils are
stimulating needed reforms in their communities. While there was variability across
councils with regard to their average perceived effectiveness, overall, council members
rated their councils as at least somewhat effective at meeting their intermediate goals and
stimulating changes in policy and practice that increase women’s safety and batterer
accountability. Further, council leaders’ ratings of the degree to which councils have
accomplished their goals also suggest that councils are succeeding in the creation of
reform in their communities. These ﬁndings support a small but growing body of
literature that suggests that domestic violence coordinating councils provide an important
vehicle for the creation of a CCR to domestic violence by improving relationships across
key stakeholders, providing a forum to create institutionalized change (i.e., reforms in
policy and practice) and engaging in prevention activities, including the education of
professionals and community members (Clark et al. 1995; Edwards, 1992; Short, &

DeBruyn, 2000).

178

 

Councils were viewed as especially effective in meeting intermediate goals such as
increasing council members’ knowledge of different stakeholder groups (including both
their roles and limitations), and improving stakeholders’ ability to coordinate their efforts.
That is, members rated councils’ effectiveness in meeting intermediate goals more
favorably than goals to reform practice and policies that increase women’s safety and
batterer accountability. Accomplishing these intermediate goals is a critical ﬁrst step in
building the relationships among stakeholders that provide the foundation upon which
their collective work ensues. However, in some ways, fostering such relationships may be
easier to achieve than creating actual reforms within or beyond the criminal justice system.
For example, some councils began their collaborative efforts by using a series of meetings
to have each organization share information about its unique role in the response to
domestic violence, its general operations, and the areas in which the organization hoped to
see change. Councils may function more effectively as a forum for the exchange of
information than for the implementation of changes in policy and practice.

Interestingly, councils were more likely to have goals to stimulate changes in the
criminal justice system than to create reforms in other systems, such as healthcare.
Speciﬁcally, council leaders reported addressing a greater number of needed reforms
within the criminal justice system than they did in areas beyond the criminal justice system.
The greater attention to criminal justice issues is not surprising given that to date CCR
efforts have focused largely on the need to protect women and hold batterers accountable

by advocating for reforms within the criminal justice system. This focus reﬂects the

179

 

emphasis of the movement to end domestic violence on recognizing domestic violence as a
crime rather than a private matter. Still, the almost exclusive focus of some councils on
the criminal justice system is concerning considering the important role different systems
play in encouraging the safety of women (Hart, 1995; Schechter, 1999; Sullivan &
O’Keefe, 1999) and the need for coordinated efforts to increasingly address areas in other
human service systems (Clark et al., 1995; Schechter, 1999). Schechter (1999)
underscores this concern, reﬂecting on the large number of battered women who never
become involved with shelter programs or the criminal justice system, but who may access
services in other areas of the human service system including neighborhood based clinics,
hospitals, and welfare and child protection offices. For some councils, attending to
needed reforms outside of the criminal justice system would require a refocusing of their
efforts. For example, they would need to set goals to address issues in other systems such
as health care and human service and incorporate collaborative partners from these
systems who play a central role in the accomplishment of these goals. The membership of
many of the councils in this study did not include stakeholder groups that would be central
to the implementation of these efforts. Still, setting new goals and inviting new members
would have to be balanced with the need to keep the work of the domestic violence
coordinating council relevant to the areas of expertise and desires of participating
members. For example, criminal justice ofﬁcials may not feel they are centrally involved in
plans to implement a domestic violence education program in schools, while educators

may ﬁnd their involvement peripheral to discussions of the enforcement of Personal

180

 

 

 

Protection Orders. Organizing the councils’ work in such a way that all stakeholders
understand their roles in the setting is critical given that stakeholders may remain more
engaged when the collective work is relevant to their area of expertise and the beneﬁts of
their involvement outweigh the costs (Butterfoss, Goodman, & Wandersman, 1996;
Chinman, Anderson, 1mm, Wandersman, & Goodman, 1996; Gray, 1985).

While this study provides preliminary evidence supporting the role of coordinating
councils as a vehicle for stimulating community change, ﬁiture research must link
perceptions of council effectiveness to community outcomes and indicators such as
community members’ knowledge of domestic violence, women’s safety and access to
resources, and batterers’ recidivism. In their review of the work of coalitions in the
community health arena, Roussos and Fawcett (2000) found that linking the work of
collaborative bodies to more ‘distant’ health outcomes within the community poses a
signiﬁcant challenge. However, they concluded that “at least under some conditions,
implementation of collaborative partnerships is associated with improvements in
population-level outcomes” (p. 375). In the domestic violence context, ultimately, it is the
increased safety of survivors that will indicate the degree to which coordinating councils
are making a difference in their communities (Hart, 1995). Yet, very few of the councils
in this study received any formal feedback from survivors or incorporated survivors into
their efforts. Collaborative efforts to respond to complex social issues must be
accountable to the constituency for which they advocate reform. Future research must

include feedback from survivors about how the reforms for which domestic violence

181

 

coordinating councils advocate inﬂuence their lives and about the degree to which they
experience the criminal justice, human service and health care systems as more responsive
to their needs.
Elements of the Council Climate as Correlates of Council Effectiveness

This study suggests that the internal climate councils foster is integral to their
effectiveness. While some collaboration researchers have suggested that the internal
climate of councils is important (Foster-Fishman et al., in press; Larson & LaFasto, 1989),
few studies have systematically examined the relationship of council climate to
effectiveness across multiple settings (see Butterfoss, Goodman, & Wandersman, 1996;
Gottlieb, Brink & Gingiss, 1993; Kegler et al., 1993 for examples). The ﬁndings from this
study indicate that domestic violence coordinating councils were more likely to be
successﬁil in their collaborative endeavors when they fostered an internal climate where
conﬂict was handled effectively, power in decision-making was shared, a shared mission
was in place and leadership was both inclusive and task-focused. However, while there is
evidence that each element of the council climate examined in this study was related to
council effectiveness, these elements were often differentially related to each indicator of
council effectiveness. In addition, the effect of each element of the council climate shifted
in the presence of the other elements of council climate. Given the relatively small sample
size in this study and the strength of the interrelationships of the elements of council
climate (and the resulting collinearity), each element and its relationship to the indicators

of council effectiveness is presented in two ways: a) as the sole predictor of the indicators

182

 

 

of council effectiveness, and b) in the presence of the other elements of council climate.23
Milt—65m

The ﬁndings from this study suggest that the relationship between conﬂict
resolution and council effectiveness is complex. When conﬂict resolution was considered
alone, it was related to two indicators of council effectiveness. However, when conﬂict
resolution was considered in the presence of other variables, its relationship to council
effectiveness changes. These shifts appear to be the result of the relationships of effective
conﬂict resolution with the other elements of council climate. Taken together, the ﬁndings
from this study suggest that effective conﬂict resolution may not be as strongly related to
council effectiveness as other aspects of the council climate. This may be due to relatively
low levels of conﬂict on councils or insensitive measurement of conﬂict and conﬂict
resolution within the council setting.

When effective conﬂict resolution was considered individually as a predictor of the
indicators of council effectiveness two signiﬁcant relationships emerged. Speciﬁcally,
councils whose members rated them as handling conﬂict effectively were more likely to be
viewed as effective by council members. These councils were also somewhat more likely
to be accomplishing their goals beyond the criminal justice system, according to council

leaders (at the trend level). This provides some support for the assertion that councils’

 

23

Please note that the ﬁndings discussed refer to the council level relationships (e. g., the
relationship between average perceived effectiveness and average shared power in
decision-making) unless otherwise noted.

183

 

"air .iWM‘i’illllﬁlllll'll’liiltlllt"

 

ability to handle conﬂict is integral to their success (Fargason et al., 1994; Wichnowski &
McCollum, 1995). However, when considered in the presence of the other elements of
council climate the relationship of effective conﬂict resolution and the indicators of council
effectiveness shifted somewhat. First, as a predictor of the breadth of goals addressed
within the criminal justice system a suppression effect emerged. Second, as a predictor of
perceived council effectiveness, it appears that the relationship of conﬂict resolution and
perceived council effectiveness is mediated by the other climate variables.

While this ﬁnding emerged only at the trend level and must be considered
preliminary, councils whose members viewed them as handling conﬂict effectively
addressed a broader range of reforms within the criminal justice system. Interestingly, this
trend emerged when inclusive leadership was added to the model including effective
conﬂict resolution, instrumental leadership and council age (i.e., conﬂict resolution was
not a signiﬁcant predictor in the absence of inclusive leadership)?" This ﬁnding suggests
that inclusive leadership (and perhaps instrumental leadership) suppresses the relationship
between conﬂict resolution and the breadth of goals addressed. The challenge in
interpreting such a ﬁnding is to identify the unique variance in conﬂict resolution (i.e.,
variance unshared with the other predictors) that is related to the remaining variance in the

breadth of criminal justice goals addressed (i.e., when the variance related to the other

 

24

It is important to note that this suppression effect only emerged when instrumental
leadership and council age were also included in the model. This suggests that it is some
combination of these effects that bolsters the relationship of conﬂict resolution and the
breadth of goals addressed within the criminal justice system.

184

 

 

WWW

m a"!
it"? it:
i 11

is: lit

it's/rt

ll

.- l

predictors is accounted for). In this study, the unique variance associated with conﬂict
resolution (i.e., not overlapping with inclusive leadership and instrumental leadership)
appears to be the practice of ‘agreeing to disagree’. Conceptually, ‘agreeing to disagree’
is the only conﬂict resolution strategy that does not emphasize the need to foster input
from all stakeholders to achieve a resolution. In contrast, the other conﬂict resolution
strategies, such as ‘getting to the root of the disagreement’ or ‘compromise,’ may be more
related to the ability of council leadership to be “adept in organizational and
communication skills” (instrumental leadership) or be “competent in negotiating, solving
problems and resolving conﬂict” (inclusive leadership). Providing some support for this
assertion, ‘agreeing to disagree’ was the only item in the conﬂict resolution subscale
unrelated or only weakly related to most of the inclusive and instrumental leadership
items. While this ﬁnding is tentative, it highlights an important direction for ﬁiture
research, suggesting that domestic violence coordinating councils using this conﬂict
resolution strategy (‘agreeing to disagree’) attempt reform in a greater number of areas
within the criminal justice system. It is not surprising that ‘agreeing to disagree’ may play
a more prominent role in this realm given there are many contentious issues surrounding
reform of the criminal justice system, ranging from addressing dual arrest practices (i.e.,
arrests of both the assailant and survivor versus identiﬁcation and arrest of only the
‘primary aggressor’) to survivors’ participation in the prosecution of domestic violence
cases (i.e., survivors as key witnesses versus an emphasis on evidenced-based

prosecution). Focusing on areas of agreement may be essential given that underlying

185

differences are unlikely to be easily resolved. For example, in some communities the
practice of dual arrest may reﬂect an underlying attitude among law enforcement that
women are just as aggressive as men. Domestic violence advocates would counter this
argument emphasizing that women are assailants in only ﬁve percent of domestic violence
cases. While these stakeholder groups may not come to agreement about the underlying
issue, they may both agree that the system can not handle the volume of cases and
subsequent dilemmas (e. g., determining who to press charges against) that result from
dual arrest.

The preliminary ﬁnding that ‘agreeing to disagree’ may be a useﬁil conﬂict
resolution strategy is consistent with research examining coordinated efforts in the sexual
assault arena where this tactic was also employed as a way for collaborative bodies to
engage in joint action (Campbell, 1999). However, this ﬁnding contradicts the
preponderance of research on conﬂict resolution which stresses the importance of
transforming conﬂict by addressing complex underlying issues (Bitter, 1977; Fargason et
al., 1994). Yet, ‘agreeing to disagree’ may reﬂect an aspect of the collaborative process
where stakeholders gain a more clear understanding of their differences (Gray, 1989), and
choose to focus on areas of agreement as a way to begin or continue their work together.

Interestingly, some studies suggest that younger councils may be more likely to
focus on areas of agreement, while over time councils may focus on more contentious
issues (Bitter, 1977; Rosenkoetter et al., 1995). For example, Rosenkoetter et al., (1995)

suggest that in the beginning of the work of collaborative settings, they tend to focus on

186

“non-controversial issues that build broad-based community ownership” (p. 276). As
councils mature, they suggest that there is a greater need to resolve the more sensitive
issues that were avoided initially (Rosenkoetter et al., 1996). However, in this study,
younger councils were no more or less likely to ‘agree to disagree’.25 In the domestic
violence context, it may be that this tactic is useﬁil not only in the beginning of
collaborative work, but at varied stages of the collaborative process when councils are
identifying areas to address within the criminal justice system. Given that collaborative
groups are not going to achieve complete consensus before they engage in cooperative
activities (Bitter, 1977), ‘agreeing to disagree’ may provide a starting point with the
potential that, as stakeholders work together over time on a given issue, they will foster
greater consensus (Wood & Gray, 1991). Still, it is also important to note that ‘agreeing
to disagree’ may prove to be problematic if it becomes a way to avoid the development of
a truly shared mission and understanding of domestic violence and results in a prolonged
avoidance of the more complex issues domestic violence coordinating councils face.
Future research is needed to identify when and how ‘agreeing to disagree’ is an effective
conﬂict resolution strategy.

In addition to the suppression effect, the ﬁndings from this study suggest that

effective conﬂict resolution has an indirect effect on council effectiveness. Conﬂict

 

25

Post hoc analyses reveal that the item, “when faced with conﬂict council members ‘agree
to disagree’” was not signiﬁcantly related to council age (r = .17, p = .30, two-tailed).
However, the item, “conﬂict has created opportunities for open discussion among council
members” was signiﬁcantly related to council age (r = .43, p < .01).

187

 

 

resolution was not retained in the ﬁnal model of the degree to which councils were viewed
by members as effective, but was a signiﬁcant predictor when considered individually.
This suggests that the other climate variables may mediate the relationship between
conﬂict resolution and council effectiveness. In particular, effective conﬂict resolution
might inﬂuence the degree to which councils have developed a shared mission, which
explains its relationship with perceived council effectiveness. Future research is needed to
examine such mediating relationships in an effort to understand how conﬂict and conﬂict
resolution positively contribute to the work of domestic violence coordinating councils.

Finally, while there is some preliminary evidence that aspects of conﬂict resolution
are related to council effectiveness, the ﬁndings from this study suggest that conﬂict
resolution may simply not be as strongly related to council effectiveness as the other
elements of the council climate. Even considered separately, conﬂict resolution was not
signiﬁcantly related to most indicators of council effectiveness (e. g., it was not related to
the breadth of goals addressed beyond the criminal justice system nor the degree to which
goals were accomplished within the criminal justice system).

In addition, overall, this study suggests that councils did not experience high levels
of conﬂict or employ ineffective conﬂict resolution strategies with great frequency. The
generally low level of conﬂict experienced by the majority of councils might explain the
more limited role of effective conﬂict resolution in explaining council effectiveness. In a
paper written for the National Resource Center on Domestic Violence, Schechter (1999)

suggests that in contrast to the past, there are professionals in every ﬁeld who take

188

domestic violence seriously. While this is not likely to be true for all communities, this
may explain why for many councils the ability to handle conﬂict constructively did not
covary with the degree to which they were effective in accomplishing their goals. It is
probable that the degree to which conﬂict resolution is related to council effectiveness
depends on the level of conﬂict in the setting. For example, the level of conﬂict may
moderate the relationship between effective conﬂict resolution and council effectiveness.
Future research is necessary to more completely examine what mediates and/or moderates
the relationship of conﬂict resolution to council effectiveness.

The ﬁnding that conﬂict resolution played a more limited role in explaining council
effectiveness may also reﬂect a weakness in the conﬂict resolution measure and the
approach to examining conﬂict in this study. During interviews with key informants, when
asked about the speciﬁc areas in which conﬂict emerged, many indicated they did not
experience conﬂict on their councils. Some went on to explain they had disagreements
and diverse viewpoints, but not conﬂict. Others indicated that the conﬂict was always
below the surface, but did not characterize this as ignoring conﬂict because the conﬂict did
not actually emerge during council meetings. These comments raise important issues
regarding the measurement of this construct and ﬁiture research on conﬂict in the work of
domestic violence coordinating councils. Smith and Berg (1997) bemoan the paucity of
research on conﬂict within groups. They suggest that while scholars frequently mention
that conﬂict is endemic to group work, it is not ﬁilly investigated. Further, they suggest

that the critical task in examining conﬂict is not only to understand the role of conﬂict

189

resolution in the work of collaborative groups, but the nature of the conﬂict itself and how
it is ﬁtnctional and/or dysfunctional for groups. In future research on conﬂict, and conﬂict
resolution, in the domestic violence context, researchers must operationalize and examine
both conﬂict and conﬂict resolution processes with more precision and contextual
information.
Shared Power in Decision-Making

This study suggests that shared power and inﬂuence in decision-making is
positively related to the degree to which councils achieve their goals. Considered as the
sole predictor, councils whose members rated them as having shared power in decision-
making were more likely to be viewed as effective by council members. They were also
more likely to be accomplishing their goals both within and beyond the criminal justice
system, according to council leaders. This provides support for the assertion that the
success of collaborative efforts depends upon the ability of collaborative bodies to share
power and authority (Kagan, 1991) and create opportunities for all stakeholders to
inﬂuence direction-setting (Butterfoss, Goodman, & Wandersman, 1996; Gray, 1985).
Fostering a climate where each stakeholder can share their unique vantage point and
experience is critical to council effectiveness given that greater information sharing leads
to better decision-making (e.g., Larson, Foster-Fishman, & Keys, 1994) and greater
satisfaction and participation among council members (e.g., Butterfoss, Goodman, &
Wandersman, 1996). In the domestic violence context, such an equitable climate is

essential; individuals with the most expert knowledge on domestic violence are sometimes

190

those with the least power to inﬂuence outcomes in the collaborative setting (e. g.,
domestic violence advocates and survivors). However, this study also suggests that
shared power in decision-making has complex interrelationships with both inclusive
leadership and the breadth of active membership that inﬂuence its relationship with council
effectiveness. In addition, similar to conﬂict resolution, shared power in decision-making
was not uniformly related to all indicators of council effectiveness and was frequently
reduced to a trend toward signiﬁcance or non-signiﬁcance in the presence of the other
climate variables. These ﬁndings highlight two important issues regarding the relationship
of shared power in decision-making to council effectiveness.

First, similar to effective conﬂict resolution, inclusive leadership acted to suppress
the relationship between shared power in decision-making and the degree to which goals
were accomplished within the criminal justice system (at the trend level). While these
ﬁndings must be interpreted with caution, the unshared variance between shared power in
decision-making and inclusive leadership may reﬂect the decision-making focus of shared
power in decision-making. That is, both constructs emphasize council member input, but
only shared power in decision-making assesses the degree to which this input inﬂuences
the decisions the council makes. This implies that it is not only important that leaders
gather member input, but that this input informs decisions made in the collaborative
context. Again, it is important to interpret this ﬁnding with caution. It is possible that this
suppression effect is due only to multicollinearity. These variables are highly correlated

and this suppression effect may reﬂect only a statistical artifact. That is, when numerous

191

highly correlated variables are examined simultaneously, they may act to mask or suppress
their relationships with the outcome variable.

Interestingly, the relationship between shared power in decision-making and
council effectiveness is also augmented when the breadth of active membership is included
in models of council effectiveness. Speciﬁcally, a second set of suppression effects
emerged with regard to the relationships of shared power in decision-making to both
perceived council effectiveness and the degree to which goals are accomplished within the
criminal justice system. This suggests that when the variance due to the breadth of active
participants was accounted for in these indicators of council effectiveness, shared power in
decision-making had a stronger relationship with each.26 Thus, the ﬁndings from this
study indicate that shared power in decision-making is not only related to council
effectiveness above and beyond the breadth of active membership, but that taking this
covariate into account strengthens this relationship. This suggests that in order for
councils to be effective they must not only ‘get stakeholders to the table,’ but create a
climate where the input of all members inﬂuences the decision-making process of the
council. This ﬁnding also raises the possibility that fostering shared power in decision-
making becomes more essential when a diverse array of stakeholders are actively engaged

in the collaborative process. To conﬁrm such an assertion, ﬁiture research should examine

 

26

It is important to recall that shared power in decision-making is not signiﬁcantly related to
the breadth of active membership. Thus, this is a case of classical suppression (Cohen &
Cohen, 1983).

192

the breadth of active membership as a moderator of the relationship between shared power
in decision-making and council effectiveness.

Second, the ﬁndings from this study suggest that shared power in decision-making
was not uniformly related to all indicators of council effectiveness. Speciﬁcally,
considered individually, shared power in decision-making was not signiﬁcantly related to
the breadth of goals addressed both within and beyond the criminal justice system. This
suggests that the ability of councils to share power in decision—making is not essential to
all aspects of the collective work of domestic violence coordinating councils. Further,
shared power in decision-making did not remain a signiﬁcant predictor in the ﬁnal climate
models of perceived council effectiveness nor the degree to which goals were
accomplished beyond the criminal justice system. Substantively, these ﬁndings may
indicate that shared power in decision-making was not as centrally related to council
effectiveness as the quality of council leadership. However, these ﬁndings may also be
explained by the multicollinearity among the independent variables. Shared power in
decision-making was strongly related to both instrumental and inclusive leadership, but
especially the latter. In the ﬁnal models, where inclusive leadership was the strongest
predictor, shared power in decision-making was not a signiﬁcant predictor. In the ﬁnal
models where instrumental leadership was the stronger predictor, inclusive leadership
acted to suppress the relationship of shared power in decision-making and council
effectiveness. The complexity of these interrelationships suggests that ﬁiture research is

needed to examine non-additive models, including an examination of mediating structures.

193

While the ﬁndings from this study seem to converge on the need to incorporate the
voices of all stakeholders in the direction the council takes, this research does not reveal
how power relationships inﬂuence the collaborative process, how power disparities are
successﬁilly transformed to encourage equity with a setting or how power relationships
outside of the collaborative setting, and within the community context, inﬂuence
collaborative work. Regarding the latter issue, councils may foster shared power among
stakeholders within the collaborative context, but may be unable to foster such equity with
key stakeholders in their community who are not currently council members or are
‘oﬂicial’ but not active members. Those with greater power in the community context
may participate of their own good will, but there are no mechanisms in place to encourage
their participation. This is problematic because outside of the collaborative context,
stakeholders may not be able to inﬂuence powerful players whose actions affect the degree
to which councils effectively stimulate changes in their communities. Thus, the ﬁndings
from this study highlight two questions for future research on shared power and
collaboration: 1) How do councils foster equitable power relationships where all voices are
heard within their collaborative settings? and 2) How can councils transform power
relationships with individuals outside of their collaborative settings? These are both
critical questions when the ultimate goal of collaboration is to address complex social
issues, such as domestic violence. Such efforts require a more fundamental transformation

of power relationships (Himmelman, 1996; Jones & Bodtker, 1995; Lipman, 1997).

194

W29

Consistent with previous research, this study indicates that the presence of a
shared mission was related to the degree to which councils were meeting intermediate
goals and the goals of a CCR. Considered individually, councils whose members viewed
them as having a shared mission were more likely to be rated as effective by council
members and more likely to accomplish goals beyond the criminal justice system and to
some extent within the criminal justice system (at the trend level) according to council
leaders. In addition, considered in the presence of the other elements of council
effectiveness, the degree to which councils were viewed by their members as having a
shared mission positively contributed to the ﬁnal model of the degree to which councils
were viewed by members as effective. Thus, in part, the ﬁndings from this study are
consistent with those from previous research suggesting that forging a shared mission in a
context of diversity is a critical part of the collaborative process and is linked to the
effectiveness of collaborative endeavors (e. g., Butterfoss, Goodman & Wandersman,
1993; Bond & Keys, 1994; Foster-Fishman et al., in press; Roussos & Fawcett, 2000).

Similar to both conﬂict resolution and shared power in decision-making, these
ﬁndings were not uniform across indicators of council effectiveness. Speciﬁcally, the
presence of a shared mission was not signiﬁcantly related to the breadth of goals
addressed within or beyond the criminal justice system. This suggests that domestic
violence coordinating councils may be able to work collectively in some areas, such as

identifying goals, even when they have not developed an entirely shared mission. The

195

presence of a shared mission may become more important as councils attempt to
accomplish the goals they have identiﬁed. This is evidenced by the relationship between
the presence of a shared mission and the degree to which goals have been accomplished.
Still, shared mission remained a signiﬁcant predictor in only one of the ﬁnal models of the
indicators of council effectiveness. Again, this ﬁnding may be explained in part by shared
variance with other predictors and a lack of power to detect signiﬁcant relationships.
However, this ﬁnding may also suggest that the presence of shared mission, like conﬂict
resolution and to some degree shared power in decision-making, was not as strongly
related to council effectiveness as other elements of the council climate, especially the
quality of council leadership. Future research is need to examine circumstances when the
the presence of a shared mission is most critical and those when other factors are more
paramount.
Lam

Consistent with previous literature (e.g., Butterfoss, Goodman, & Wandersman,
1996; Kegler et al., 1998), this study ﬁnds that leadership is strongly related to council
effectiveness. Further, this study suggests that effective leaders must be skilled at both
facilitating the collaborative process (e.g., attending to the participation of all members),
and being adept at focusing the group on completing tasks (e. g., planning meetings
efﬁciently). While leadership was consistently the strongest predictor of council
effectiveness, different aspects of leadership were related to different indicators of

domestic violence coordinating council effectiveness.

196

Instrumental leadership, a leadership style focused on strong organizational skills,
commitment to the council’s mission and garnering resources, was the most important
predictor of the breadth of goals councils addressed. That is, the more councils were
viewed as having instrumental leadership by their members, the more likely they were to
be addressing areas of reform within and, to some extent, beyond the criminal justice
system according to council leaders. However, this relationship only emerged when
inclusive leadership was also included in the model. As with the other elements of the
council climate, the interrelationships of the leadership variables require special attention
in the interpretation of these ﬁndings. There are two explanations for this pattern of
relationships. First, this ﬁnding may reﬂect a statistical artifact of the collinearity of
inclusive and instrumental leadership. These variables were highly correlated, indicating
that, for the most part, leaders were viewed as having both strong instrumental and
inclusive leadership skills.27 Second, substantively, this ﬁnding may indicate that it was the
unique variance associated with instrumental leadership that was related to the breadth of
goals councils addressed. This unique variance may reﬂect the degree to which the
council leadership has the organizational skills and resources to effectively lead (e. g.,
planning meetings eﬂiciently, providing guidance in maintaining the council). While it is

critical to interpret these ﬁndings with caution, councils with leaders who have such skills

 

27

Further, in some cases, council members may have been responding about multiple leaders
within the setting, rather than only a single leader. It is not uncommon for coalitions to
have a core group of leaders (Roussos & Fawcett, 2000).

197

may be able to manage a broader array of desired reforms with the human service, health
care and criminal justice systems.

Interestingly, such a suppression effect did not emerge in the relationship of
instrumental leadership and other indicators of council effectiveness. Instrumental
leadership appears to be critical to the ability of councils to successﬁilly create reforms
with criminal justice system. Councils whose members viewed them as having strong
instrumental leaders were also more likely to accomplish their goals within the criminal
justice system. These ﬁndings suggest that operating successfully within the criminal
justice context may require a greater emphasis on the ability of the leader to plan meetings
effectively and efficiently and garner resources to support the councils efforts. Future
research is needed to ascertain which leaders are the most likely to employ such a
leadership style to explore a potential confound of this ﬁnding. If leaders from the
criminal justice system are also more likely to embody instrumental leadership traits this
may explain why this leadership style is more strongly related to goals being accomplished
within the criminal justice system.

With regard to inclusive leadership, a different pattern of relationships emerges.
While instrumental leadership is somewhat related to the breadth of goals addressed
beyond the criminal justice system (at the trend level and only with inclusive leadership in
the model), inclusive leadership is the strongest predictor of the degree to which these
goals were accomplished. That is, councils whose members viewed them as having strong

inclusive leaders were more likely to be accomplishing their goals beyond the criminal

198

justice system according to council leaders. This indicates that to create needed reforms in
the human and social services councils require a leader who incorporates the input of all
members and promotes equality among members. This may be particularly important
given that domestic violence coordinating councils are made up largely of criminal justice
officials. In the absence of a leader skilled at encouraging and valuing the input of all
members, stakeholders who are not afﬁliated with the criminal justice system could have
peripheral involvement in council activities resulting in goals not being accomplished in the
areas they represent.

Councils with inclusive leadership were also more likely to be rated as effective by
their membership. The degree to which councils were viewed by members as having
inclusive leadership was the strongest predictor of perceived council effectiveness. This
ﬁnding may suggest that the presence of inclusive leadership plays a more primary role in
the degree to which councils are effective overall. Perceived effectiveness assesses a
broader scope of council effectiveness (i.e., it includes an assessment of the degree to
which councils have addressed both the goals of a CCR and intermediate goals). Thus,
inclusive leadership may play a critical role in the degree to which councils are
accomplishing their intermediate goals, such as learning about each’s other roles and
limitations.

In their review of collaborative partnerships, Roussos and Fawcett (2000) found
that less is known about how different sets of leadership skills may be USCﬁJl during

different stages of partnership development. While this study can not provide information

199

regarding the role of leadership at different stages of council development, these ﬁndings
provide evidence that inclusive and instrumental leadership are related to different aspects
of the collaborative work of domestic violence coordinating councils. Thus, it is not
leaders’ exclusive use of one style or another that is related to council effectiveness, but
rather that different aspects of leadership were related to different areas of success in the
collaborative work of domestic violence coordinating councils. This is consistent with
previous research which suggests that different aspects of leadership inﬂuence group
ﬁanctioning in different ways and these relationships depend in part on other situational
characteristics (e. g., Barrow, 1976; Larson, Foster-Fishman & Franz, 1998). For
example, Larson, F oster-Fishman and Franz (1998) found that while participative leaders
elicited a greater amount of information from group members, directive leaders managed
the information more effectively and ultimately made better decisions. However, the
quality of decisions made in groups with a directive leader was in part dependent on the
degree to which the leader held information, prior to the group discussion, that was
congruent with the best decision. Future research on the role of leadership in the
collective work of domestic violence coordinating councils should measure different
aspects of leadership with more precision and less shared variance. Further, this research
should attend to situational factors such as the leaders’ knowledge base regarding
domestic violence, their viewpoints regarding the direction of the council, and their
organizational affiliation which may also affect how and when instrumental versus inclusive

leadership are important.

200

Finally, with regard to the relationship of leadership and council effectiveness, this
study provides some preliminary evidence that inclusive leadership is partially mediated by
other elements of an inclusive climate, speciﬁcally shared power in decision-making and
presence of a shared mission. When these variables were introduced into the ﬁnal model
of perceived council effectiveness, the strength of the association between inclusive
leadership and council effectiveness was reduced. Future research is necessary to examine
those factors that mediate the relationship of leadership and council effectiveness. Such an
examination would provide insight regarding the mechanisms by which leaders foster
council effectiveness.

Council Characteristics as Correlates of Council Effectiveness

In addition to council climate, council characteristics play an important role in
explaining council effectiveness. Attending to such characteristics is critical given that
there are many aspects of the council structure (e. g., Kegler et al., 1998), membership
(e. g., Chrislip & Larson, 1994) and resources (e. g., Reilly, 1998) that have been linked to
council ﬁmctioning in previous research. This study provides some conﬁrmation of
previous research suggesting that active membership and support from the external
environment are important correlates of the effectiveness of domestic violence
coordinating councils.

Breﬂh of ‘Oﬂicﬂnd ‘Active’ Membership
While the breadth of ‘oﬂicial’ membership was not related to council effectiveness,

the breadth of ‘active’ membership was. Councils with a greater number of their official

201

membership actively participating in council activities were more likely to be rated by their
members as effective, and more likely to accomplish their goals both within and beyond
the criminal justice system according to council leaders. These ﬁndings are consistent
with previous research that suggests that councils with broader membership have broader
agendas for change (Rosenkoetter et al., 1995) and are more effective (Butterfoss,
Goodman, & Wandersman, 1996; Chrislip & Larson, 1994; ISA Associates, 1993 as cited
by Chinman et al., 1996). In the domestic violence context, the presence of a broad range
of stakeholders who are active members creates greater opportunity for reform in a broad
range of systems. In addition, the active participation of stakeholders suggests that they
are committed to the work of the council, which is a critical precursor to successﬁil
collaboration (Gray, 1985).

Huxham and Vangen (2000) suggest it is critical to investigate council membership
more careﬁilly, taking into account the complexity and ambiguity inherent in determining
who is actually a member of a collaborative setting. This study demonstrates that when
examining the inﬂuence of council membership, it is critical to distinguish between
‘oﬂicial’ and ‘active’ membership. There were a considerable number of councils where
‘official’ membership included the entire set of relevant stakeholders, but this diversity of
membership in ‘name only’ was not related to council effectiveness. Stakeholders may
agree to become members of a council, but may not necessarily be engaged or committed
to the work of the council. This membership in ‘name-only’ does not appear to foster

council effectiveness.

202

Interestingly, the breadth of membership was not related to the breadth of goals
addressed either within or beyond the criminal justice system. It is possible that councils
identify goals not based on their membership, but based on their desire for reform. For
example, 74% of councils were addressing sentencing practices, but only 54% had active
participation from a member of the judiciary, the stakeholder group most central to the
actual implementation of this reform. The degree to which they can actually implement
these reforms depends on the degree to which they encourage ‘buy in’ from the
stakeholder groups than can facilitate the implementation of these reforms. The challenge
in the domestic violence context is to encourage involvement from the most powerful
stakeholder groups. Future research is needed to address how the participation of
stakeholders in relative positions of power can be fostered.

Presence of External Support

Consistent with previous research, the support councils receive from the external
environment (e.g., ﬁinders and policy makers) is an important component of their success.
Councils who received support from ﬁinders, local policy makers, and community
members according to key informants were more likely to accomplish their goals both
within and beyond the criminal justice system. This echoes the ﬁndings of other studies
which suggest that the environment in which collaborative settings engage in collective
work is critical to their success (Gray, 1996) and that the degree of support provided in
the external environment is critical to the adoption of reform (F oster-Fishman, Salem,

Allen, & F ahrbach, 2000). External support is critical to the efforts of domestic violence

203

coordinating councils given that the purpose of their collective work is to create change in
complex systems. Such change efforts require support from many external stakeholders
(i.e., those stakeholders external to the collaborative group, but critical to the
collaborative work; Sink, 1996) including, for example, ﬁinders, organizational leaders
and community members. For example, in this state, there has been a great deal of
controversy regarding which batterers intervention programs judges should refer assailants
to as a part of their sentences. Some council leaders complained that many judges would
routinely refer to weekend programs that were viewed as ineffective. Then a Govemor’s
Task Force was put together to develop batterers intervention standards. This state level
initiative supported the local efforts of councils who were advocating for similar reforms.
Such support from the external environment may be most critical when the desired
reforms involve powerful stakeholders such as judges.

While this study provides evidence for the importance of support from the external
environment, similar to effective conﬂict resolution, the presence of such support did not
remain a signiﬁcant predictor in any of the ﬁnal models of council effectiveness in this
study. Perceived support from the external environment was a signiﬁcant predictor of the
degree to which goals were accomplished, both within and beyond the criminal justice
system, yet these relationships were weakened in the presence of the breadth of active
council membership. It is possible that the relationship between the presence of support
from the external environment and council effectiveness is mediated by the breadth of

active council members. Such a mediating relationship might suggest that support from

204

the external environment is important because it results in greater participation from key
stakeholders. For example, in one community, organizational leaders were appointed to
the council by the County Commissioner. Thus, from the start, this council had some ‘buy
in’ from key players who could enact change within their respective organizations and
allocate staff for regular council involvement. The ﬁndings from this study suggest that
future research is needed to examine how the external environment supports council
activities (e. g., by providing funds and staff; by increasing legitimacy; by changing policies
that encourage reform; by broadly engaging community members in prevention activities).
Formality of Council Structure

The degree of formality in the council structure was not related to the breadth of
goals councils addressed, the degree to which they accomplished their goals nor the
degree to which their members viewed them as effective. These ﬁndings contradict other
research, suggesting that coalitions with more complex structures were more successful
(Gottlieb, Brink, & Gingiss, 1993; Kegler et al., 1998). However, it is critical to note that
in this study the vast majority of councils had at least basic organizational structures in
place. For example, 97% of councils used an agenda and 92% kept minutes. In addition,
fewer councils had more formal structures, such as decision-making processes, in place
(34%) or written bylaws (24%). The lack of variability limits the conclusions that can be
drawn from these ﬁndings. In short, the ﬁndings from this study do not conﬁrm or reﬁite
the ﬁndings of previous research which suggests that basic organizational tools such as the

use of an agenda are important to council effectiveness (e. g., Foster-Fishman et al., 2001).

205

Summary of the Correlates of Indicators of Council Effectiveness

It is interesting that each indicator of council effectiveness was explained by
diﬁ'erent constellations of the elements of the council climate and council characteristics.
This ﬁnding suggests that those factors that facilitate collaboration may be dependent on
the nature of the collective work and has important implications for researchers and
practitioners interested in supporting domestic violence coordinating councils. This
section provides a brief summary of those factors that were most salient in explaining each
indicator of council effectiveness.

The breadth of goals reﬂects the degree to which councils have attempted a broad
array of reforms within their communities both within the criminal justice system and in
other systems including human service and healthcare. The most important factors in
explaining the breadth of goals councils addressed were the degree to which the council
had strong instrumental leadership and perhaps the degree to which they focus on ﬁnding
common ground in the midst of diverse viewpoints (i.e., ‘agreeing to disagree’). This
suggests that, above all, attempting broad based reform requires highly organized
leadership. However, it is also important to note that for the most part the set of
predictors examined in this study was not strongly related to the breadth of goals
addressed. There are ﬁve plausible explanations for this. First, other factors may play a
more prominent role regarding the number of reforms a council undertakes. For example,
identifying council goals may be prescribed by the resources made available to councils.

There is a great deal of written work deﬁning the reforms required in a CCR (e.g., Berry,

206

1995; Pence, 1996). The breadth of goals addressed may be a ﬁinction of the knowledge
base of council members regarding what the development of a CCR entails. Second, this
outcome measure may not be a reliable proxy for council effectiveness. Councils who are
addressing a broad range of goals without success are equivalent to those who have
addressed the same number of goals with great success. Thus, this measure may be less
sensitive to differences in effectiveness across councils. Third, this measure reﬂected the
proportion of needed reforms councils addressed in their community. However, only the
key informant deﬁned which reforms were needed. Thus, the breadth of goals addressed
as measured in this study may not accurately reﬂect the degree to which needed reforms
were being addressed in a given community. Fourth, the breadth of goals addressed
within the criminal justice system was negatively skewed (54% of councils reported
addressing 100% of the needed criminal justice reforms in their communities). Such
skewness may limit the ability to detect signiﬁcant relationships. Finally, it is important to
note that council age remained a signiﬁcant predictor of the breadth of goals councils
addressed. Thus, this indicator of council effectiveness may be strongly related to the
amount of time councils have been engaging in collective work; clearly, councils that have
been working together for a number of years will have addressed a broader range of goals
than those working together for only a number of months.

The degree to which goals are accomplished reﬂects each council’s ability to
follow through on their goals and stimulate desired reforms. To accomplish their goals

within the criminal justice system councils require attention to both their internal climate

208

and external environment. Councils with shared power in decision-making, strong
instrumental leaders, a broad array of active members and an external environment
perceived to be supportive of council activities (e.g., ﬁinders, policy makers) were more
likely to accomplish their goals to reform the criminal justice system. This suggests that
achieving these reforms requires a high degree of efficiency, a forum for a broad range of
voices to be heard, and adequate resources and contact with key external stakeholders
(i.e., those who are not active council members, but are key players) to support their
efforts. A somewhat different picture emerges when considering the accomplishment of
goals in areas outside of the criminal justice system (e.g., reforms in healthcare and/or
child protective services). To accomplish these reforms, what is most essential is inclusive
leadership and having a diverse set of active stakeholders. This indicates that it is essential
for councils addressing reforms in healthcare and human service to foster the participation
of key stakeholders from the systems in which reform is desired and to create an internal
climate that emphasizes their inclusion.

Finally, perceived effectiveness reﬂects the degree to which setting members view
the council as achieving intermediate goals as well as goals to stimulate changes in policy
and practice that increase women’s safety and batterer accountability. What appears to be
most central in explaining variation across councils with regard to perceived effectiveness
is: a) the degree to which members’ viewed their councils’ leaders as fostering an
inclusive process and their councils’ as having a shared mission and to some extent shared

power in decision-making, and b) the degree to which the council has a diverse set of

208

active participants. These ﬁndings may reﬂect, in part, the focus of this measure on both
intermediate and CCR goals. An inclusive process may be a critical component of
councils meetings intermediate goals given that such goals reﬂect the degree to which
council members are learning about one another and fostering mutual respect. These
ﬁndings also suggest that the degree to which councils have developed a shared mission is
related to their members’ perceptions of council effectiveness. While the presence of a
shared mission was not as strongly related to the other indicators of council effectiveness,
its role in fostering members’ perceptions of effectiveness is critical. Member participation
and council effectiveness may have a reciprocal relationship; member participation is
critical to council success and members who view their councils as successful are more
likely to participate (Butterfoss, Goodman, & Wandersman, 1996).

In summary, each indicator of council effectiveness examined in this study requires
attention to different elements of the council climate and setting. While these ﬁndings
must be considered preliminary, they suggest that practitioners who are working to assist
domestic violence coordinating councils must attend to different aspects of the council
climate, council characteristics, and external environment depending on the focus of the
collective work in which councils are engaging. These ﬁndings also suggest that future
research must pay attention to how facilitators and barriers to the collaborative process
are linked to speciﬁc components of collaborative work.

Revisiting Levels of Analysis and Organizational Role

In this study, the examination of council members’ perceptions of council

209

effectiveness required attention to multiple levels of analysis. Given that council members
were nested within councils, I-ILM was used to regress this indicator of council
effectiveness onto the elements of the council climate while separating individual and
group level effects (i.e., group-mean centering was employed to examine the effect of each
element of the council climate). Interestingly, those factors that explained variation in
council members’ perceptions of council effectiveness at the individual level were
somewhat different than those which explained differences in average perceived
effectiveness across settings. For example, at the individual level, the degree to which
members’ viewed their councils as having a shared mission, relative to others in their
council, was the strongest predictor of the degree to which they viewed their council as
effective. However, at the council level, the strongest predictor of average perceived
effectiveness across settings was the degree to which councils were rated as having
inclusive leadership. Thus, when explaining differences in perceived effectiveness across
council members the degree to which they view their councils as having a shared mission
is most important; when explaining differences across councils the degree to which they
are rated as having strong inclusive leadership is most important. This provides one
example of how the pattern of relationships at the individual and council levels is different.
Highlighting these subtle differences is critical given that failure to decompose
individual and group level effects can result in potentially inaccurate ﬁndings (Florin et al.,
1995) and failure to attend to multiple levels of analysis is somewhat common in the

collaboration literature. Some studies of collaborative settings examined relationships

210

only at the individual level without attention to the nested nature of the data and without
attention to the appropriate level of analysis (e. g., Butterfoss, Goodman & Wandersman,
1996; Kumpfer et al., 1993). Other studies have examined relationships at the council
level without attention to within-group agreement or variation (e. g., Butterfoss,
Goodman & Wandersman, 1996; Gottlieb, Brink, & Gingiss, 1993; Kegler et al., 1998;
Mansergh, Rohrbach, Montgomery, Pentz, & Johnson, 1996).

Given the potential for within-group differences, another goal of this study was to
explain variability in perceptions across members within settings; organizational role was
examined as one such predictor. The ﬁndings ﬁom this study suggest that there are some
differences between the varied stakeholder groups (as deﬁned by organizational role) and
domestic violence advocates in their perceptions of council effectiveness, and other
elements of the council climate. With regard to perceived council effectiveness, judges
and prosecuting attorneys were more likely to view their councils as effective (as well as
clergy at the trend level). This could be due in part to differing standards for effectiveness
espoused by judges, prosecuting attorneys and clergy as opposed to those of domestic
violence advocates. That is, lower expectations regarding reforms required to create a
CCR may result in the view that councils are more effective in meeting the goals of a
CCR. However, this difference may also reﬂect the sense of personal reform they have
experienced by these stakeholders as a result of their council involvement. For example,
they may experience signiﬁcant shifts in their practice, making their “yard stick” for rating

effectiveness somewhat different than those of domestic violence advocates. It is also

211

possible that councils who have active participation from judges and prosecuting attorneys
are, in fact, more effective at meeting their goals and the goals of a CCR. These elected
officials are central to a CCR given that the prosecution of batterers, and appropriate
sentencing, form the cornerstone of batterer accountability. A ﬁnal plausible explanation
is that judges and prosecuting attorneys are also politicians who may have an investment in
the council being viewed as an effective entity.

With regard to the council climate variables, there were no differences in the
perceptions of conﬂict resolution between the different stakeholder groups (deﬁned by
organizational role) when compared to those of domestic violence advocates. However,
judges, and to some extent prosecuting attorneys (at the trend level), were more likely to
view the councils as sharing power in decision-making. This may be due to their relative
positions of power within the settings. It is common for those in privileged positions to be
less aware of power inequities when they do exist (Bond & Pyle, 1998; McIntosh, 1989).

Human service providers, prosecuting attorneys, and to some extent educators (at
the trend level), were more likely to view councils as having a shared mission than
domestic violence advocates. Again, this may reﬂect different expectations regarding the
direction of the councils’ work and different assumptions regarding the degree to which
this direction is shared. Human service providers, and to some extent clergy and
educators, were also more likely to view councils as having instrumental leadership than
were domestic violence advocates. Domestic violence advocates may have higher

expectations regarding the degree to which leaders should provide guidance and garner

212

resources to support council activities, skills that describe aspects of the instrumental
leadership style. Finally, human service providers, and to some extent judges, prosecuting
attorneys and court personnel (at the trend level), were more likely to view councils as
having inclusive leadership when compared to domestic violence advocates. It is
interesting that batterers intervention service providers were less likely than domestic
violence advocates to view councils as having inclusive leadership. This may reﬂect the
marginalization of the interests of batterers intervention providers within the council
setting.

While this study can not provide sufﬁcient contextual information to ﬁilly explain
these differences across groups, these ﬁndings suggest that perceptions of the council
setting do vary systematically by organizational role, even though there were a relatively
small number of differences. These ﬁndings are consistent with research that suggests that
perceptions of climate may vary by subgroup or workgroup or deﬁning identity
characteristics such as organizational role (e. g., Forehand & Von Gilmer, 1964; Schneider
& Richers, 1983). This systematic variation raises important issues with regard to when it
is appropriate to aggregate such data and how accurately the within-group agreement
statistics reﬂect the actual variation within groups. It may also suggest that multi-level
examination of data avoids making faulty assumptions about the degree to which
individual level data can simply be aggregated to the setting level. Finally, it is important
to emphasize that these analyses were performed with only domestic violence providers as

the baseline group. Future research must include more exhaustive analyses to better

213

understand differences between groups, not only in comparison to domestic violence
advocates.
Implications For Future Research

While implications for ﬁrture research have been identiﬁed throughout, there are
also some overarching issues that should be taken into consideration in ﬁxture research on
collaborative settings. First, in the study of collaborative settings researchers must
continually examine the way they approach research design with regard to three issues: 1)
balancing contextual information with the examination of trends across settings, 2)
employing longitudinal designs, and 3) attending to multiple levels of analysis. While there
is utility in examining trends across settings, such an examination limits a researcher’s
capacity to gain a contextually grounded understanding of the collaborative process. This
study provides information about how settings differed from one another and factors
which explained these differences; however, these ﬁndings are more diﬂicult to ﬁilly
understand given the relative lack of information regarding the unique dynamics of the
collaborative process within each setting. Conversely, case study research examining only
one or two settings makes it difﬁcult to understand which elements of the collaborative
setting are most linked to effectiveness. Thus, there is a need to thoughtﬁilly balance
examining issues across council settings and collecting ample information to understand
those trends.

Relatedly, it is becoming more paramount that longitudinal designs are employed

to study collaborative processes and outcomes. Collaboration is a dynamic process (Smith

214

& Berg, 1997) and should be examined and captured as such. This study highlights some
of the correlates of council effectiveness; however, ﬁiture research must also begin to
understand the causal relationships of the collaborative process. Clearly, the independent
variables in this study were highly interrelated. Longitudinal research could play an
important role in examining the mediating relationships that may exist among these climate
variables.

Finally, as in all community-based research, the design of collaborative research
should attend more careﬁilly to levels of analysis (Shinn, 1990). The ﬁndings from this
study suggest that different patterns of relationships emerge based on the level of analysis
and that it is important to address within-group variation when using perceptual data.
Attention to such multiple levels of analysis will ultimately increase the accuracy of
research on collaborative processes and outcomes.

There is also a need to begin building more parsimonious conceptual models of the
collaborative process that explore the interrelationships of different aspects of the council
climate and setting. This study focused on four elements of the council climate; however,
these constructs were highly interrelated suggesting that they may reﬂect one overarching
construct. Future research must focus on the identiﬁcation of meta-constructs that
capture the multiple factors that have been identiﬁed as critical to the collaborative
process (see Foster-Fishman et al., in press, for one such framework).

In addition, it is imperative that our examinations of collaborative efforts delve

more fully into the unique realities of collaborative endeavors in a particular context (e. g.,

215

the domestic violence context) and, more broadly, into the issues which emerge when
collaboration is being utilized to respond to complex social issues. Himmelman (1996)
refers to collaboration that aims to create social change as transformational collaboration.
Such efforts might have different implications regarding the ﬁill participation of
community members and the power issues that ensue when the goal is to address social
issues that are inherently power laden, such as domestic violence.

Finally, future research must link the work of collaborative settings to outcomes
within their communities. Relatively little research has been conducted to examine how
the work of collaborative bodies is inﬂuencing the community at large. In the work of
domestic violence coordinating councils examining such outcomes is imperative. To date,
we know very little about how the changes in policy and practice, both within and beyond
the criminal justice system, are affecting women’s safety.

Limitations

Although this study is one of few that investigates multiple collaborative settings
permitting an examination of trends across settings, it is important to highlight some of its
limitations. First, given the challenge of studying multiple coalitions, there was a relatively
small N to examine council level relationships which are the focus of this study. This leads
to greater challenges in modeling multiple independent variables simultaneously. That is,
there is reduced power to detect signiﬁcant relationships when they do exist, increasing
the probability of Type II errors. Thus, ﬁnal models should be interpreted with caution as

some variables that do contribute to the effectiveness of councils may not have achieved

216

signiﬁcance due to limited power to detect such relationships.

Second, the independent variables, particularly the climate variables, were highly
interrelated. While this may be due in part to shared method variance, the
interdependence of these variables is not surprising given that they all reﬂect elements of
an inclusive internal climate. While this study begins to examine the relationships among
independent variables, it is important to remember that collinearity can result in statistical
artifacts such as relationships being masked or suppressed due to shared variance.

Third, single method variance is a concern with regard to the analysis of perceived
council effectiveness (i.e., as perceived by members). However, the degree to which goals
were accomplished was assessed via key informant interviews, providing some evidence
that the relationship of the council climate to council effectiveness was not only the result
of shared method variance.

Fourth, response bias is always a concern when utilizing perceptual measures of
climate. That is, council members may have a tendency to rate their councils favorably
and may be reluctant to rate their councils as ineffective or as having high levels of
conﬂict. Thus, for example, the low levels of conﬂict reported by council members may
reﬂect response bias rather than an accurate assessment of the actual level of conﬂict
within the council setting. However, it is important to note that, overall, there was
sufﬁcient variability in each of the independent and dependent variables with no extreme
skewness.

Fifth, sampling bias may also be a concern. While surveys were returned from

217

almost half of those members who were active participants, those who participated may be
more satisﬁed with the council that those who did not participate. Thus, council members
perceptions of the setting may be biased and not reﬂect the reality of the setting. Still, it is
important to note that for the majority of councils, participants included individuals from a
variety of sectors and organizational roles increasing the probability that a diverse set of
perspectives is represented.

Sixth, this data is cross-sectional. Thus, while this study provides important
information regarding the correlates of council effectiveness, any causal assumptions or
statements are based only in theory, not in the data. The cross sectional nature of the data
makes understanding the interrelationships of the predictor variables more difﬁcult to
interpret.

Finally, there are numerous barriers and facilitators of collaborative work (F oster-
Fishman et al., in press; Gray, 1989; Mattesich & Monsey, 1992; Sink, 1996) that are not
examined in the current study. For example, this study does not attend to any of the
individual representatives’ personalities or personal agendas (Sink, 1996) or the degree to
which members’ have an adequate knowledge base regarding domestic violence. It is
likely that such member characteristics also inﬂuence the degree to which domestic

violence coordinating councils are effective.

218

CONCLUSIONS

This study provides preliminary evidence that domestic violence coordinating
councils can serve as effective vehicles for the creation of a coordinated community
response (CCR) to domestic violence. While councils were more likely to address areas
of needed reform within the criminal justice system (e. g., the accessibility and enforcement
of personal protection orders), they also addressed areas of needed reform in systems such
as health care and child protective services. In addition, domestic violence coordinating
councils were particularly effective at meeting their intermediate goals such as learning
more about each others’ organizations, and working together more efficiently. In this
way, councils may provide a backdrop for needed community change by facilitating
relationships across the many stakeholder groups that are critical to the formation of an
effective community response to domestic violence.

While this study provides preliminary evidence that domestic violence coordinating
councils can play an important role in stimulating needed reforms in the community, they
were not uniformly effective. The ﬁndings from this study suggest that the internal
climate councils foster is central to their effectiveness in accomplishing their goals and the
goals of a CCR. First, such a climate must be characterized by quality leadership that is
organized and efficient, as well as skilled at encouraging the voices of all stakeholders to
be heard. Second, the council climate must be characterized by shared power in decision-

making where the input of all stakeholders inﬂuences councils’ collective work. Third, the

219

internal climate must be characterized by the presence of a shared mission, and, ﬁnally, by
effective conﬂict resolution. Taken together, these ﬁndings provide preliminary evidence
that fostering effective leadership that is highly organized, efficient, and inclusive,
engaging in shared decision-making, developing a shared mission, and productively
addressing areas of conﬂict are all critical to the success of domestic violence coordinating
councils.

Interestingly, while the elements of the council climate were all related to council
effectiveness, each was differentially related to varied aspects of councils’ collective work.
This ﬁnding suggests that the relationship of climate and council effectiveness depends, in
part, on the nature of the collective work itself (e.g., identifying goals versus
accomplishing goals). Future research must attend to those factors that facilitate and pose
barriers to speciﬁc aspects of collaborative work.

Further, the ﬁndings from this study suggest that council effectiveness is not only
affected by the council climate, but also by other features of the council setting. First,
councils are more successful when they are able to secure the active participation of a
broad range of stakeholders. The emphasis on active membership is critical given that the
degree to which councils had a broad range of ofﬁcial members was not related to their
effectiveness. Second, councils who were able to garner support from the external
environment, including policy makers, funders, organizational leaders and community
members were more successful in their collective work. Taken together, these ﬁndings

suggest that it is not only the internal climate of councils that is important, but also the

220

degree to which councils operate within a supportive context and the degree to which a
broad range of stakeholders actively engage in councils’ activities.

Finally, this study suggests that the investigation of councils requires attention to
multiple levels of analysis. First, it is essential to examine relationships at the council level
while simultaneously accounting for individual level variability. If the analyses performed
for this study had ignored the nested nature of these data, examining only individual level
relationships, different ﬁndings would have emerged. Second, it is important to examine
the sources of variability in members’ perceptions within a given council. This study
provides preliminary evidence that individual characteristics such as organizational role
may account for differences in perceptions across council members within a single setting.

In summary, domestic violence coordinating councils provide promising venues for
the creation of a CCR to domestic violence. While such councils are likely to face
challenges in their collective work, the creation of a climate that is both inclusive and task-
oriented, the engagement of a broad array of critical stakeholders and the acquisition of
the external resources and supports necessary to meet their goals can all positively

contribute to the success of their collective work.

221

APPENDICES

222

APPENDIX A

223

APPENDIX A

Council Member Survey

ID Number:

 

Please respond to this survey regarding your perceptions and experiences with
your council as a whole rather than only one subcommittee or work group. In
this survey, the use of the term council refers to Domestic Violence Coordinating
Councils, Task Forces, and Coordinating Boards.

1. Are you currently a member of a domestic violence coordinating council?
Yes No
x
If NO:

1a. How many months ago did you stop being a member?—
1b. For what reason(s) did you stop being a member?

 

 

 

Please continue to next question even if you answered NO.
2. When did you join this council? Month Year

3. For what reason(s) did you join the council?

 

 

4. How many council meetings did you attend Qvar (ha past 12
mpnths .......... _

224

5. What kinds of activities do you or did you engage in with/for the

coordinating council?
Do (did) you EVER: (please circle)

Talk at meetings? ................................. YES
Do work for the council outside of meetings? ............ YES
Help organize activities other than meetings? ............ YES
Chair the entire council? ............................ YES
Direct the implementation of a particular program or policy? YES
Serve as a member of a subcommittee? ................ YES
Chair a subcommittee? ............................. YES

(N/A = Not Applicable)

5a. Are you currently a subcommittee member? ........ YES

7

If YES, of which subcommittee?

NO
NO
NO
NO
NO
NO . . N/A
NO . . N/A

NO . .N/A

 

In an average month, about how many hours of your time have you given

to the council in the following activities (WW3

MW):
hours for general council meetings
hours for subcommittee meetings

hours for council work outside of regular meetings

(e. g., preparing for meetings, activities, doing paperwork)

Other
Total Hours

Is (or was) your participation on the council: (Please check one)

(1) Voluntary (not compensated by any source)

(2) Part of your job for an agency or organization (which agency?)
(3) As a direct employee of the council (e. g., paid coordinator)

225

8. Domestic Violence Councils often have members who come from many different
community sectors, such as law enforcement, education, etc. What community sector
do you come from or represent at council meetings? (Please check one)

(1)
(2)
(3)
(4)
(5)
(6)
(7)
(3)
(9)
(10)
(11) __
(12) __
(13)____
(14)—
(15) __
(16) __
(17) _
(18) _
(19) _

Business
Child Welfare Agencies

Civic/Volunteer Organizations
Community/Neighborhood Group

Cooperative Extension

Cultural/Ethnic Groups

Domestic Violence Service Providers/Advocates
Domestic Violence Survivors

Health Services

Higher Education

Justice Systems/Courts

Law Enforcement

Local Government

Mental Health

Preschool/K42 Education

Religious Organizations

Media

Youth Members
Other (please specify)

 

9. Do you have authority to make decisions on behalf of your organization or
group at council meetings? Please check one:

(1) Yes

(2) Only with approval from others in my organization
(3) Only with approval of my board or membership
(4)__ No

226

 

 

 

 

 

 

 

 

 

 

 

 

l I 5
582m
38M 38M
CEO 02 8.x C80 02 8> o m e m N _ “mam .o
288$
390— 33
ream oz 8> :89 oz 8». c w v m N _ “mam .m
:5on
390. 399
reen— oZ mo> r80 02 8> e w v m N _ “mam .v
.535
399 305—
rcon 02 8» C89 02 8> w m e m N _ “mam .m
“ennui
390— 399
reoQ 02 BS 589 02 ma; 0 w v m N _ Oman .N
588m
329 399
race 02 mo> CEO 62 8> o w v m N _ «mam .—
Aoee 29:8 A23 385
«View A28 29:3 A25 29:8 “585 $30 .6550
$5 9:33 5 S9: 3:on .5 couaNanCo 308.an0 N c an a 8: u 3 .5 “me.—
025a 36333 59A 8 “Stag 28m $5 3:39:83
3:303 0:883 2Q Rom £5 £8? _moesoo of we: 2.on :23 oh
n A: N A: _

.bEEeEoo 2: 5 825—95 3 we?“ a 3:38 2: mow—8:9 328% 2: 2 .32 £80 .mA mess—8 E £83823
warning 3 Row £5 Sosa 29: 3:813» :23 .5.— 98 3585 98 ~33 mﬁom Eatong 29: can 2: 35 .o—

227

 

 

 

 

 

 

 

 

 

 

 

 

11. What would you say best characterizes the portion of time your council spends on
internal issues, or those issues that keep the council functioning (e.g., dealing with
internal conﬂicts, the organization of the council, recruiting members) versus
external issues, or those issues that lead to the council accomplishing its goals
(e. g., changing protocols and policies)? (Please check one)

(1) 0% Internal 100% External (4) 7 5% Internal 25% External
(2) 25% Internal 75% External (5) 100% Internal 0% External
(3) 50% Internal 50% External (6) Don’t Know

12. Below are several statements about how decisions are made within councils.
Circle the number to the right of each statement that shows how accurately you
think each statement describes decision-making within your council.

Not at A Little Somewhat Mostly True Very Don't
All True True True True Know
True (V)
a. The input of all active council 1 2 3 4 5 6
members inﬂuences the decisions the
council makes.
b. When making decisions the council is 1 2 3 4 5 6
responsive to all of the viewpoints
represented on the council.
c. Council decisions are dominated by a 1 2 3 4 5 6
few members.
d. The council does not move forward on 1 2 3 4 5 6
a decision until all input is heard
e. If a council member shares a dissenting l 2 3 4 5 6
opinion at council meetings his/her
perspective is considered important.
f. 'l‘hegeneralmembershiphasreal l 2 3 4 5 - 6
decision making control over the policies
and actions of the council.
g My input inﬂuences the decisions the 1 2 3 4 5 6
council makes.
h. My stance for or against an issue can 1 2 3 4 5 6
sway the council in that direction.
i. 1 have an active role in the decision- 1 2 3 4 5 6
making process of the council.
j. The council will not move forward on 1 2 3 4 5 6
a decision if 1 am not in agreement.
It. The council takes actions that my 1 2 3 4 5 6
Eganization does not support

 

 

 

228

 

13. Below are several statements about how conﬂict is addressed on your council.
Circle the number to the right of each statement that shows how accurately each

statement represents your council. [N/A = not applicable]

 

Not at
All

True

A Little
True

Somewhat
True

Mostly
True

True

Very
True

N/A
(V)

Don’t
Know
(V)

 

1. There are differences in
opinion among council
members

1

 

b. Conﬂict has emerged in
the council’s attempts to
work collaboratively.

 

c. Council members disagree
about the definition of
domestic violence.

 

d. Disagreements among
council members are often
resolved by compromise.

 

e. Conﬂict has led to
effective problem-solving

 

 

1’. When conﬂict arisesthe
council ignoresit

 

g. Conﬂict has created
opportunities for open
discussion among council
members.

 

h. Disageements typically
stiﬂe the progress of the
council.

 

1. When faced with conﬂict
council members ‘agree to

disagree.’

 

j. The council has handled
conflict by attempting to get
to the root of the
disagreement.

 

k. Conﬂicting opinions
among council members
have led to needed changes
in the council.

 

l. The council has avoided
addressing diverse
viewpoints represented on
the council.

 

 

 

 

229

 

 

 

14. Below are several statements about leadership. Circle the number of each statement
that shows how accurately each statement represents the leadership of your council.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Not at A Somewhat Mostly True Very Don’t
All Little Tme True True Know
True True (V)
a. Council leadership is committed to the 1 2 3 4 5 6
council's mission.
b. Council leader(s) provide leadership and 1 2 3 4 5 6
guidance in maintaining the council.
c. Council leader(s) have appropriate time 1 2 3 4 5 6
to devote to the council.
(1 Council leader(s) plan meetings 1 2 3 4 5 6
eﬂ’ectively and efficiently.
e. Council leader(s) have knowledge in the l 2 3 4 5 6
area of domestic violence.
f. Council leader(s) are ﬂexible in 1 2 3 4 5 6
accepting different viewpoints.
g Council leader(s) promote equality and l 2 3 4 5 6
collaboration among members.
h. Council leader(s) is/are adept in l 2 3 4 5 6
organizational and communication skills.
i. Council leader(s) are competent in 1 2 3 4 5 6
negotiating, solving problems and resolving
conﬂict.
j. Council Ieader(s) are attentive to 1 2 3 4 5 6
individual member concerns.
k. Council leader(s) are adept in obtaining 1 2 3 4 5 6
resources.
1. Council leader-(s) value members’ input 1 2 3 4 5 6
m. Council Ieader(s) recognize members 1 2 3 4 5 6
t for their contributions.
n. Overall, to what extent does your 1 2 3 4 5 6 ll
leader’s style work for your council.

 

230

15. Below are several statements about your council’s mission. Circle the number to
the right of each statement that shows how accurately each statement represents

your council.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Not at A Little Somewhat Mostly True Very Don't
All True True True True Know
True (II)
a. My council has developed a mission that 1 2 3 4 5 6
is shared and supported by all council
members.
b. Council members have a shared 1 2 3 4 5 6
understanding of domestic violence.
c. Council members disagree about what 1 2 3 4 5 6
direction the council should take.
d. On the council. special interests get in l 2 3 4 5 6
the way of having a shared mission.
e. Council members have a shared vision 1 2 3 4 5 6
regarding what changes are needed in the
community response to domestic violence.
f. Council members are working together 1 2 3 4 5 6
to achieve a common goal.
g. Council members have a history of l 5 6
mutual support.
h. Council members act in the best interest 1 3 4 5 6
of the survivors of domestic violence.
i. The members of my council act in ways 1 3 4 5 6
that beneﬁt the community.
j. Councils members act in ways that l 3 4 5 6
benefit themselves or their organization.
k. Members of my council are likely to 1 3 4 5 6
reciprocate with each other.
I. Council members trust each other. 1 3 4 5 6

 

 

231

 

16. Below are statements describing the eﬂ’ects a council may have on the
community. Please circle the number to the right of each statement that shows
how accurately each statement represents the effect your council has had.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

The council has:
Not at A Somewhat Mostly True Very Don’t
All Little True True True Know
True True (V)
a. increased the ability of organizations to l 2 3 4 5 6
coordinate their efforts.
b. increased members’ knowledge about each 1 2 3 4 5 6
other’s organizations.
c. increasedmembers’respectforeachother"s 1 2 3 4 5 6
work.
d. increased members’ knowledge of other 1 2 3 4 5 6
members’ roles and limitations.
e. resulted in agencies working together more 1 2 3 4 5 6
efﬁciently
f. addressed shortcomings in practices in l 2 3 4 5 6
community organizations regarding their response to
domestic violence (cg, police, probation,
prosecution, domestic violence shelters).
g. inﬂuenced the policy of agencies regarding their 1 2 3 4 5 6
response to domestic violence.
h. inﬂuenced changes in practice in community 1 2 3 4 5 6
organizations that have increased batterer
accountability.
i. inﬂuenced changes in practice in community 1 2 3 4 5 6
organizations that have increased women’s safety.
j. been effective at facilitating needed changes in l 2 3 4 5 6
our community regarding the response to domestic
violence.
k stimulated policy changes within my 1 2 3 4 5 6
organization regarding our response to domestic
violence.
I. got people talking about domestic violence. 1 2 3 4 5 6
m. inﬂuenced changes in practice in community 1 2 3 4 5 6
organizations that have increased both women and
their children's safety.
n. consistently moved toward meeting its goals. 1 2 3 4 5 6
0. been productive in accomplishing what it set out 1 2 3 4 5 6
to do.

 

 

 

232

l7.

 

To what degree are you satisﬁed with the councils’ eﬁ’orts to:

 

Not at all Unsatisﬁed Somewhat Somewhat Satisﬁed Very

 

 

 

Satisﬁed Unsatisfied Satisﬁed Satisﬁed
a. consistently move 1 2 3 4 5 6
toward meeting its goals
b. create needed changes 1 2 3 4 5 6
in the community
response to domestic
violence
c. get people talking 1 2 3 4 5 6
about domestic violence

 

 

 

18. What factors have facilitated Council members working together to achieve the
Council’s goals?

 

 

 

 

19. What factors have impeded Council members working together to achieve the
Council’s goals?

 

 

 

 

 

233

20. What types of support (i.e., technical support, training, etc) would be
most helpful to your council?

 

 

 

 

 

21. This question is about other council members with whom you currently discuss
domestic violence issues or council activities. Please list these council members
by name (including both their ﬁrst and last name) and indicate how often you
discuss such issues or activities with each of them.

 

 

 

 

 

 

 

 

Name Every Day Once a Week Once a Once
Month a Year
1 2 3 4
1 2 3 4
l 2 3 4
l 2 3 4
1 2 3 4
1 2 3 4
l 2 3 4

 

 

 

 

234

BACKGROUND INFORMATION

Please respond to the items below. Your responses will be held in the strictest
conﬁdence. For each question please circle the appropriate response.

Bl.

B2.

B3.

AGE B4.

(1) under 20
(2) 20 to 29
(3) 30 to 39
(4) 40 to 49
(5) 50 to 59
(6) 60 or over

GENDER B5.

(1) female
(2) male

ORGANIZATIONAL
ROLE
(Circle one)

(1) FIA Case Worker
(2) CPS Case Worker

(3) Clergy
(4) Domestic Violence

Service Provider B6.

(5) Judge
(6) Nurse
(7) Police Ofﬁcer

(8) Probation Ofﬁcer

(9) Prosecuting Attorney
(10) Psychologist

(11) Social Worker

(12) School Administrator
(13) Teacher

(14) Other

 

RACE/ETHNICITY

(1) Aﬁican—Amen'can/Black
(2) Asian/Paciﬁc Islander
(3) Hispanic/Latino

(4) Native American

(5) White/Caucasian

(6) Other:

 

YEARS WORKING ON
DOMESTIC VIOLENCE
ISSUES

(l) 6 months or less

(2) 7-12 months

(3) 13 months to 2 years
(4) 2.1 years to 4 years
(5) 4.1 years to 6 years
(6) 6.1 years to 10 years
(7) 10.1 years to 15 years
(8) more than 15 years

Have you attended any training
regarding domestic violence issues?
(1) YES (0) NO

If YES, approximately how many
hours of training have you had?

Thank you so much for your participation.

235

APPENDIX B

236

3a.

APPENDIX B

Key Informant Interview
COUNCIL DEMOGRAPHICS

What is the ofﬁcial name of your council or task force?

 

When was the council formed? (Months)

 

What precipitated the council forming (VAWA/STOP grant funding, one agency
initiated collaboration)?

 

 

 

3b. What is the purpose of your council?

 

 

 

 

 

 

[IF NOT CLEAR FROM #3]

Which organization took the ‘lead’ in forming the ?

How long have you been the [SPECIFY ROLE] of the council?
(Months)

5b. What is your organizational role?

 

If NOT council chair: Who is chair?
His/her org role ?

 

 

237

9a.

How often does your council meet? Please circle one.
1 Bi-Monthly (every other month)

 

2 Monthly

3 Weekly

4 Other (describe)

Does your council have a paid staff person? (circle one) YES NO
7a. If YES, which organization or funding source supports this person’s

position? (please specify)

 

Does the council have any funding or ﬁnancial support?
(circle one) YES NO

8a. If YES, what types of ﬁnancial support? (please specify with amount)

 

 

 

 

 

 

 

 

 

 

( 1) Federal Grant 3
(2) State Grant $
(3) County Grant $
(4) City Grant $
(5) Public Donations $
(6) Fundraisers $
(7) Other: 5
8b. How much do you receive yearly?
9. To what degree does your council have adequate staff/volunteers to
complete desired objectives? Would you say your council has
staff/volunteers to complete desired objectives?
[CIRCLE ONE]

1 Not at all adequate
2 Somewhat adequate
3 Adequate

4 Very Adequate

What type(s) of geographic communities does your council serve?

Rural Suburban Urban

238

10.

11.

12.

13.

Has your Council/Task Force or another agency in your community conducted a
needs assessment of your community with regard to domestic violence?
[CIRCLE ONE]

YES NO

If yes, what did you/they do?

 

 

 

Does your Council/Task Force address issues related to sexual assault?
(Please circle) YES NO

If YES, what has the Council done with regard to sexual assault?

 

 

 

Has your Council/Task Force facilitated a Safety and Accountability Audit or
something similar? (Please circle) YES NO

If YES, what type of audit did the Council do (e. g., shelter system or criminal
justice system response)?

 

 

 

Has your council worked to address the needs of speciﬁc groups (e. g., ethnic or
racial minorities, gay men or lesbians)? YES NO

If YES, what did your council do?

 

 

 

239

14. Council Goals/Activities/Accomplishments: I want to ask you a series of
questions about the types of issues your council/task force has addressed in your
community. For each issue, I will ask you if it was a goal of your council to
address the issue, what activities the council engaged in to address the issue, the
degree which your council/task force accomplished what it set out to do, and what
evidence you have that your actions were effective. If you have not addressed the
issue, I will ask you if this issue needs to be addressed in your community and
whether your council/task force plans to address it.

 

Is/was
council?

a goal for your

 

Gl. Improving access to or
enforcement of personal protection
orders

YES NO

 

IF YES

IF NO

 

G1Y2. What activities did your council
engage in to address this goal?

G1N2. Is this an unmet need in your
community?

 

council accomplished this goal:

YES NO
G1Y3. On a scale of 1-6 (1=not at all IF NO IF YES
accomplished, 6=completely GO T 0 NEXT
accomplished) to what degree has the GOAL

 

G1Y4. What evidence, if any, do you
have that this goal was accomplished (e.g.,
indicators of its success)?

 

G1N3. Does your
council/task force
plan to do this in
the future?

YES NO

 

 

IF YES
GO TO
NEXT
GOAL.

IF NO,
why

not?

 

 

 

240

 

 

Is/was a goal for your
councﬂ?

 

G2. reforming arrest practices (e.g.,
ad0pting mandatory or pro arrest
policies)

YES NO

 

IF YES

IF NO

 

G2Y2. What activities did your council

G2N2. Is this an unmet need in your

 

engage in to address this goal? community?
YES NO
G2Y3. On a scale of 1-6 (1=not at all IF NO IF YES
accomplished, 6=completely GO T 0 NEXT
accomplished) to what degree has the GOAL

council accomplished this goal:

 

 

G2Y4. What evidence, if any, do you
have that this goal was accomplished (e. g.,
indicators of its success)?

G2N3. Does your
council/task force
plan to do this in
the future?

YES NO

 

 

 

IF NO, IF YES
why GO T 0
not? NEXT

 

 

241

. GOAL... .

ﬂ

 

 

Is/was a goal for your
council?

 

G3. reforming prosecution practices
(e.g., encouraging evidenced based
prosecution)

YES NO

 

IF YES

IF NO

 

G3Y2. What activities did your council
engage in to address this goal?

G3N2. Is this an unmet need in your
community?

 

 

 

 

YES NO
G3Y3. On a scale of 1-6 (1=not at all IF NO IF YES
accomplished, 6=completely GO T 0 NEXT
accomplished) to what degree has the GOAL
council accomplished this goal:
G3Y4. What evidence, if any, do you G3N3. Does your
have that this goal was accomplished (e.g., council/task force
indicators of its success)? plan to do this in
the future?
YES NO
IF NO, IF YES
why GO T0
not? NEXT
GOAL...

 

 

 

 

242

 

 

Is/was a goal for your
councﬂ?

 

G4. reforming the processing of court
cases (i.e., speed of processing, notifying
women about court cases; providing
advocacy and support to women during

 

 

 

 

 

trial)
YES NO
IF YES IF NO
G4Y2. What activities did your council G4N2. Is this an unmet need in your
engage in to address this goal? community?
YES NO
G4Y3. On a scale of 1-6 (1=not at all IF NO IF YES
accomplished, 6=completely GO T 0 NEXT
accomplished) to what degree has the GOAL
council accomplished this goal:
G4Y4. What evidence, if any, do you G4N3. Does your
have that this goal was accomplished (e. g., council/task force
indicators of its success)? plan to do this in
the ﬁiture?
YES NO

 

IF NO, IF YES

why GO T 0
not? NEXT
GOAL...

 

 

 

 

243

 

 

Is/was a goal for your
councﬂ?

 

G5. altering sentencing practices (e.g.,
extending minimum sentence;
extending probation period; mandating
batterer’s intervention)

YES NO

 

IF YES

IF NO

 

G5Y2. What activities did your council

G5N 2. Is this an unmet need in your

 

engage in to address this goal? community?
YES NO
GSY 3. On a scale of 1-6 (1=not at all IF NO IF YES
accomplished, 6=completely GO T 0 NEXT
accomplished) to what degree has the GOAL

council accomplished this goal:

 

 

GSY 4. What evidence, if any, do you
have that this goal was accomplished (e. g.,
indicators of its success)?

G5N3. Does your
council/task force
plan to do this in
the ﬁiture?

YES NO

 

 

IF NO.
why not

 

IF YES
GO T 0
NEXT

GOAL...

 

 

 

b

 

244

 

Is/was a goal for your
councﬂ?

 

G6. identifying weaknesses or ‘holes’
in the criminal justice or human service
delivery system (e.g., conducting a
death review)

YES NO

 

IF YES

IF NO

 

G6Y2. What activities did your council

G6N2. Is this an unmet need in your

 

engage in to address this goal? community?
YES NO
G6Y3. On a scale of 1-6 (1=not at all IF NO IF YES
accomplished, 6=completely GO T 0 NEXT
accomplished) to what degree has the GOAL

council accomplished this goal:

 

 

G6Y4. What evidence, if any, do you
have that this goal was accomplished (e. g.,
indicators of its success)?

YES NO

G6N 3. Does your
council/task force
plan to do this in
the future?

 

 

IF NO,
why
not?

 

 

IF YES
GO T 0
NEXT

GOAL...

 

245

 

 

Is/was a goal for your
councﬂ?

 

G7. developing or supporting
batterer’s intervention program(s)

 

 

 

 

 

 

 

 

YES NO
IF YES IF NO
G7Y 2. What activities did your council G7N 2. Is this an unmet need in your
engage in to address this goal? community?
YES NO
G7Y3. On a scale of 1-6 (1=not at all IF NO IF YES
accomplished, 6=completely GO T 0 NEXT
accomplished) to what degree has the GOAL
council accomplished this goal:
G7Y4. What evidence, if any, do you G7N3. Does your
have that this goal was accomplished (e.g., council/task force
indicators of its success)? plan to do this in
the ﬁxture?
YES NO
IF NO, IF YES
why GO T 0
not? NEXT
GOAL...

 

 

 

 

 

246

 

Is/was
councﬂ?

a goal for your

 

G8. implementing early identiﬁcation
policies in healthcare settings (e.g.,
emergency rooms; doctor’s ofﬁces)

YES NO

 

IF YES

IF NO

 

G8Y2. What activities did your council
engage in to address this goal?

G8N2. Is this an unmet need in your
community?

 

YES NO
G8Y3. On a scale of 1-6 (1=not at all IF NO IF YES
accomplished, 6=completely GO T 0 NEXT
accomplished) to what degree has the GOAL

council accomplished this goal:

 

 

G8Y4. What evidence, if any, do you
have that this goal was accomplished (e. g.,
indicators of its success)?

G8N3. Does your
council/task force
plan to do this in
the future?

YES NO

 

IF NO, IF YES
why GO T 0
not? NEXT
GOAL...

 

 

 

 

247

 

 

Is/was
councﬂ?

a goal for your

 

G9. providing training or community
education regarding domestic violence
(e.g., to health care providers, human
service workers, youth in schools, the
community at large)

YES NO

 

IF YES

IF NO

 

G9Y2. What activities did your council
engage in to address this goal?

G9N 2. Is this an unmet need in your
community?

 

YES NO
G9Y3. On a scale of 1-6 (1=not at all IF NO IF YES
accomplished, 6=completely GO T 0 NEXT
accomplished) to what degree has the GOAL

council accomplished this goal:

 

 

G9Y4. What evidence, if any, do you

have that this goal was accomplished (e.g.,

indicators of its success)?

council/task force
plan to do this in
the future?

YES NO

G9N 3. Does your

 

 

IF NO.
why
not?

IF YES
GO T 0
NEXT

 

 

 

248

ll

 

 

Is/was a goal for your
councﬂ?

 

G10. making it easier for women’s to
access needed community resources
(e..g., housing, ﬁnancial support,
clothing, transportation, child care)

YES NO

 

IF YES

IF NO

 

G10Y2. What activities did your council
engage in to address this goal?

G10N2. Is this an unmet need in your
community?

 

 

 

YES NO
G10Y3. On a scale of 1-6 (1=not at all IF NO IF YES
accomplished, 6=completely GO T 0 NEXT
accomplished) to what degree has the GOAL
council accomplished this goal:
G10Y4. What evidence, if any, do you G10N3. Does
have that this goal was accomplished (e.g., your council/task
indicators of its success)? force plan to do this
in the ﬁJture?
YES NO
IF NO, IF YES
why GO T 0
not? NEXT
GOAL...

 

 

 

 

 

 

249

 

Is/was a goal for your
councﬂ?

 

G11. developing new services for
battered women and their children

YES NO

 

IF YES

IF NO

 

Gl 1Y2. What activities did your council
engage in to address this goal?

G11N2. Is this an unmet need in your
community?

 

 

 

YES NO
G11Y3. On a scale of 1-6 (1=not at all IF NO IF YES
accomplished, 6=completely GO T 0 NEXT
accomplished) to what degree has the GOAL
council accomplished this goal:
G11Y4. What evidence, if any, do you G11N3. Does
have that this goal was accomplished (e.g., your council/task

indicators of its success)?

force plan to do this
in the ﬁxture?

YES NO

 

IF NO, IF YES
why GO T 0
not? NEXT

GOAL...

 

 

 

 

250

 

 

Is/was a goal for your
councﬂ?

 

G12. encouraging partnerships
between child protective services and
domestic violence advocates

YES NO

 

IF YES

IF NO

 

G12Y2. What activities did your council
engage in to address this goal?

G12N2. Is this an unmet need in your
community?

 

 

 

YES NO

G12Y3. On a scale of 1-6 (1=not at all IF NO IF YES
accomplished, 6=completely GO T 0 NEXT
accomplished) to what degree has the GOAL
council accomplished this goal:
G12Y4. What evidence, if any, do you G12N3. Does
have that this goal was accomplished (e. g., your council/task
indicators of its success)? force plan to do this

in the future?

YES NO

 

IF NO, IF YES fl
why GO T 0
not? NEXT

GOAL...

 

 

 

 

251

 

 

Is/was
councﬂ?

a goal for your

 

 

G13. evaluating outcomes related to the
council’s work

YES NO

 

IF YES

IF NO

 

G13Y2. What activities did your council
engage in to address this goal?

G13N2. Is this an unmet need in your
community?

 

YES NO
G13Y3. On a scale of 1-6 (1=not at all IF NO IF YES
accomplished, 6=completely GO T 0 NEXT
accomplished) to what degree has the GOAL

council accomplished this goal:

 

 

G13Y4. What evidence, if any, do you
have that this goal was accomplished (e. g.,
indicators of its success)?

G13N3. Does
your council/task
force plan to do this
in the ﬁJture?

YES NO

 

 

IF NO.
why
not?

IF YES
GO T 0
NEXT

 

 

 

252

‘1

 

. GOAL... .

 

OTHER

 

G14. Were there any other areas your
council wished to address or did
address?

YES NO

 

IF YES

IF NO

 

G14Y2. What activities did your council

G14N2. Is this an unmet need in your

 

engage in to address this goal? community?
YES ' NO
G14Y3. On a scale of 1-6 (1=not at all IF NO IF YES
accomplished, 6=completely GO T 0 NEXT
accomplished) to what degree has the GOAL

council accomplished this goal:

 

 

G14Y4. What evidence, if any, do you

have that this goal was accomplished (e. g.,

indicators of its success)?

G14N3. Does
your council/task
force plan to do this
in the future?

 

 

 

 

YES NO

IF NO, IF YES

why G0 T 0

not? NEXT
GOAL...

 

253

 

15. Has your Council/Task Force or another agency in your community gotten
feedback or input ﬁ'om domestic violence survivors regarding your work?

YES NO

If yes, how did you/they get this feedback from survivors?

 

 

 

15a. Thinking back to before the council/task force began, on a scale from 1 to 6, 1
being not at all effective and 6 being very effective, how effective was your
community’s response to domestic violence?

 

Now, thinking about the time since your council’s inception, how effective is your
community’s response to domestic violence?

 

16. The next set of questions is about the structure and processes of your

Council/Task Force. Does your Council or Task Force:

0 Have a written agenda YES NO
- Record and distribute minutes YES NO
- Have bylaws/rules of operation? YES NO
0 Have a mission statement in writing? YES NO
0 Have goals and objectives in writing? YES NO
0 Have regular meetings? YES NO
0 Have organization chart? YES NO
- Have written job/role descriptions? YES NO
0 Have a core planning group? YES NO
0 Have subcommittees or workgroups? YES NO

If YES, how often do they meet? Bimonthly, monthly, weekly or other
If YES, what are they?

 

 

- Have established processes for decision making? YES NO

[If YES, specify type ]
0 Have established processes for problem solving and

conﬂict resolution? YES NO
0 Have established processes for resource allocation? YES NO
0 Have established mechanisms for process and

impact evaluation? YES NO
0 Have a mechanism established for accountability

of members completing assignments in a timely manner? YES NO
0 Have a mechanism in place to encourage

accountability among member organizations? YES NO

254

17.

0 Have a mechanism in place to encourage
accountability among non-member organizations
in the community?

0 Have a mechanism for new member orientation?
0 Have a mechanism for training new
and old members?
0 Have a domestic violence survivor member?
If YES, how many ?
0 Have an advisory group made up of domestic survivors?
- Is your council a 501c3 nonproﬁt organization?

Does the Council allow for different representatives from member organizations to

be sent to your council meetings? (circle one)

If YES, does this disrupt the work of the council?

255

YES NO
YES NO
YES NO
YES NO
YES NO
YES NO
YES NO
YES NO

18.

The next set of questions have to do with member participation on your council. While I
have your membership list I would like to collect more speciﬁc information about what
types of organizations attend and whether or not they are active participants. First, I
wanted to check with you about which type of organizations are currently council
members and the degree to which they are actively participating on your council. What I
mean by active participation is whether the organization sends someone to meetings
regularly and participates in council activities.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Organization Current Active
Member Participant
Batterer’s Intervention Yes No Yes No
Child Protective Services Yes No Yes No
Circuit Court (Judge) Yes No Yes No
District Court (Judge) Yes No Yes No
Domestic Violence Yes No Yes No
Shelters/ Service Providers
Health Care Organizations Yes No Yes No
Legal Aid Yes No Yes No
Local Businesses Yes No Yes No
Local Police Department Yes No Yes No
Mental Health Organizations Yes No Yes No
Religious Organizations Yes No Yes No
Probation Yes No Yes No
Prosecuting Attomey’s Ofﬁce - Yes No Yes No
Prosecuting Attorney
Prosecuting Attomey’s Ofﬁce - Yes No Yes No
Victim’s Rights Advocate
School Administrators/ Yes No Yes No
Educators
Social Services (e.g., FIA) Yes No Yes No
Others: Yes No Yes No

 

 

 

256

 

19.

20.

21.

22.

Second, I would like to go through the membership list your organization provided
and have you indicate how often each member attends your council meetings. For
each person, please let me know if over the past year they have: 1) attended
meetings regularly, 2) attended meetings once in a while, 3) not attended meetings
at all, or 4) are on the membership list only to receive mailings or minutes. I know
this information will not be exact, just give me your general impressions.

GO THROUGH LIST
19a. Is there anyone else who attends meetings who was not mentioned here?

Overall, what percent of the members of your council/task force are in attendance
at the average council or subcommittee meeting? %

Are there any organizations or other community groups that you wish were
represented on the council? These could be organizations that have refused to join
or have not yet been asked to join. (Please circle)

YES NO

If yes, who? (please list)

 

 

Why do you wish they would join?

 

 

 

Since the councils inception, are there any organizations or other community
groups that were council members, but have since leﬁ or discontinued their
participation? (Please circle)

 

 

 

 

YES NO
If yes, who? (please list)
Why did they leave?

 

 

 

257

23. Council Support: Now, I want to ask you about how much support your council
has received from the local community, organizational leaders and policy makers.

 

 

 

 

 

 

 

 

 

 

 

To what degree have (has) Not at all Unsupportive Somewhat Somewhat Supportive Very

been supportive of the work Supportive Unsupportive Supportive Supportive

of the council?

the leaders of council 1 2 3 4 5 6

member organizations

the leaders of non -council 1 2 3 4 5 6

member organizatiom who

are integral to the work of

the council

local policy makers 1 2 3 4 5 6

state policy makers 1 2 3 4 5 6

members of the community 1 2 3 4 5 6

ﬁmders 1 2 3 4 5 6
24. Your probably remember from the survey that I asked a series of questions about

conﬂict on the council and how conﬂict is handled. I did this because it is very
common when multiple groups are working together that they encounter conﬂict.
On your council/task force where does conﬂict emerge? [Probez Around which
issues?]

25. When conﬂict does arise, how does your council typically handle it?

258

26. How does dealing with conﬂict affect your council/task force?

27. Is there anything else you would like to add about your experience working with
the ?

 

259

APPENDIX C

260

APPENDIX C

Recruitment Letter

COUNCIL CHAIR
ADDRESS HERE

Dear COUNCIL CHAIR,

As you know, efforts to facilitate a coordinated community response (CCR) to domestic
violence are occurring across the state of Michigan. We know that you are currently the
chair of a council, committee or task force undertaking this task and wanted to inform you
about a series of regional training conferences which will be held in May and June of 2000.
The purpose of these conferences is to support communities across the state who are
developing a coordinated response to domestic violence. Materials regarding the training
and registration will arrive in February and March of this year.

The Prosecuting Attorney’s Association of Michigan is working in conjunction with
researchers from Michigan State University to conduct a survey of the members of
domestic violence coordinating councils and task forces across the state. This survey will
be distributed in February of 2000. The purpose of the survey is to provide useﬁll
information to coordinating councils/task forces across the state. We have four primary
goals: First, we are interested in collecting information about what councils/task forces
are currently accomplishing in the State of Michigan. Second, we hope to learn more
about the factors that contribute to the effectiveness of councils/task forces in facilitating a
CCR. Third, we want to ﬁnd out more about what technical support councils/task forces
need to be effective. Finally, this survey will provide us with the information needed to
develop a follow-up survey to help us evaluate how useﬁil the training was to
councils/task forces and to develop a follow-up training. The survey will ask council/task
force members to reﬂect on their experiences working on the council, including their
perceptions of the practices, accomplishments and needs of the council/task force.

We hope that you and members of your coordinating council/task force will be willing to
participate. As a ﬁrst step, we ask that you please send us a copy of your member list so
that we can mail surveys to members of your council beginning in February to request
their participation. If you could, we would appreciate you mentioning the survey to your
council at an upcoming meeting. In addition, we also hope to conduct a phone interview
with you or the person on your council/task force who is responsible for managing the
council/task force and who has been with the council for at least a year or since its
inception. The survey of council members should take thirty to forty minutes to complete

261

and the interview should take approximately one hour. Your participation and the
participation of other council/task force members is completely voluntary.

 

We also want to assure you that only members of the Michigan State University research
team will have access to completed surveys. All results will be conﬁdential and
participants will remain anonymous in any report of research ﬁndings from these surveys.
The information collected in this survey will be used to provide feedback to each
council/task force in the state regarding what councils/task forces are currently
accomplishing, what their needs are, and what factors facilitate or create barriers to
councils/task forces achieving their goals. Upon project completion, a ﬁnal report will be
available to you. This feedback will not be connected to a speciﬁc council/task force or
council/task force member. The information in any report produced will be summarized
across people and across councils/task forces.

The participation of you and your council is very valuable to us. We really appreciate
your time and hope you will participate. Please send us your membership list by using the
self-addressed stamped envelope included with this letter or by faxing Nicole Allen at 517-
432-2945. Please call Nicole Allen at 517-353-6449 with any questions or comments. In
addition, if you have any questions about your rights as a participant in research you can
call David Wright at (517) 355-2180.

Sincerely,
Leslie A. Hagen Nicole E. Allen, M.A.
Violence Against Women Training Attorney Michigan State University

Prosecuting Attomey’s Association of Michigan

262

APPENDIX D

263

‘_

 

APPENDIX D

Follow-up Phone Call for Council Leaders Who Have Not Yet Agreed to Participate

Hello [Council Chair]. My name is Nicole Allen. I am a graduate student at Michigan
State University. I was calling to follow-up on the letter I mailed to you regarding the
survey of domestic violence councils/task forces. Have you received it [IF YES GO TO
1, IF NO GO TO 2]?

1. If YES: Okay, I was wondering if I could answer any questions for you.
If YES: [RESPOND TO QUESTIONS]
If NO: GO TO 2

2. If NO: I wanted to follow-up with you to see if you were willing to
participate in the study and to provide a membership list so other members
of your council/task force can be surveyed?

If YES: Great, would you prefer to send it via mail or fax?

If mail: Do you still have the return envelope or should I
provide my address again? [IF NEEDED PROVIDE
ADDRESS]

If fax: [IF NEEDED PROVIDE FAX NUMBER]

If NO: Okay, is there an reason in particular that your council has
declined participation? [RECORD IF THERE IS A REASON]. If your
council/task force decides to participate, please feel free to call me at
(517) 353-6449. Thank you for your time [CLOSE
CONVERSATION].

3. [FOR THOSE WHO ARE WILLING TO PARTICIPATE] I would also like
to conduct a phone interview with someone from your council who is
responsible for managing the council and who has been a council member for at
least one year. Would you be the best person for this interview?

[If YES] I would like to set up an appointment with you for a phone interview
that should take about one hour. What day and time works best for you?

264

 

[SET DATE AND TIlny FOR INTERVIEW]

[If NO] Who do you think would be the best person for this interview?
[IF THEY PROVIDE A NAME] Could I have his/her telephone number to see
if they would be willing to be interviewed?

. Thank you so much for your time and participation. If you have any questions
or need to change our interview time please feel free to call me. Do you still
have my number? [PROVIDE NUMBER IF THEY NEED IT] I look forward

to speaking with you.

265

 

APPENDIX E

266

APPENDIX E

F ollow-up Phone Call for Council Leaders Who Have Agreed to Participate

Hello [Council Chair]. My name is Nicole Allen. I sent you the letter regarding the
survey of domestic violence coordinating councils and task forces. I wanted to thank you
for sending us your membership list. I am calling to see if you have any questions that I
can answer and to set up a time to do a phone interview with you or someone from your
task force.

1. Do you have any questions about the survey or interview?

If YES: [RESPOND TO QUESTIONS]
IfNO: [GO TO 2]

2. I would also like to conduct a phone interview with someone from your council
who is responsible for managing the council and who has been a council
member for at least one year. Would you be the best person for this interview?

[If YES] I would like to set up an appointment with you for a phone interview
that should take about one hour. What day and time works best for you?

[SET DATE AND TIME FOR INTERVIEW]

[If NO] Who do you think would be the best person for this interview?
[IF THEY PROVIDE A NAME] Could I have his/her telephone number to see

if they would be willing to be interviewed?

3. Thank you so much for your time and participation. If you have any questions
or need to change our interview time please feel free to call me. Do you still
have my number? [PROVIDE NUMBER IF THEY NEED IT] I look forward
to speaking with you.

267

APPENDIX F

268

APPENDIX F

Protocol to Contact Alternate Key Inforrnants

Hello [Person Named by Council Chair]. My name is Nicole Allen. I spoke to the leader
of the Council/Task Force. He/she suggested you would be a good
person to conduct a phone interview with given that you are responsible for managing the
council/task force. I would like to interview you about the council - how the council
operates, what the councils goals are, and what the council is currently accomplishing.

1. Do you have any questions about the interview?

If YES: [RESPOND TO QUESTIONS]
IfNO: [GO TO 2]

2. Would you be willing to be interviewed about the council?

[If YES] I would like to set up an appointment with you for a phone interview
that should take about one hour. What day and time works best for you?

[SET DATE AND TIME FOR INTERVIEW]

[If NO] Okay, do you have any suggestions for other council members
have been long-time members or have a leadership role on the council?

3. Thank you so much for your time and participation. If you have any questions
or need to change our interview time please feel free to call me. Do you still
have my number? [PROVIDE NUMBER IF THEY NEED IT] I look forward
to speaking with you.

269

APPENDIX G

270

APPENDIX G

Follow-up Postcard
Dear Council/Task Force Member:

We just wanted to remind you that the council/task force member survey due date is
quickly approaching. Please mail your survey to us by . We really
appreciate the time and effort required for your participation. Please call Nicole Allen
with any questions or if you need another copy of the survey (517) 353-6449. Thank you
for your participation.

 

271

APPENDIX H

272

APPENDIX H

Follow-up Telephone Calls

Hello [Council/T ask Force Member]. My name is . I am a (graduate)
student at Michigan State University. I was calling to follow-up on the survey I mailed to
you regarding the [TASK FORCE NAME] domestic violence councils/task force. Have
you received it [IF YES GO TO 1, IF NO GO TO 2]?

 

1. If YES: Okay, I was wondering if I could answer any questions for you.
If YES: [RESPOND TO QUESTIONS]
If NO: GO TO 3
2. If NO: I would be happy to mail you another survey.
If YES: Would you prefer me to send the survey via mail or fax?
If mail: [CONFIRM ADDRESS]
If fax: [IF NEEDED PROVIDE FAX NUMBER]
3. Okay, I just wanted to let you know that we have extended the due date for
surveys. The new due date is [SET DATE FOR TWO
WEEKS FROM CURRENT DATE]. I hope you will consider ﬁlling out the

survey and sending it to us by then. We really appreciate your time and
participation.

 

4. Thank you so much for your time and participation. If you have any
questions please feel free to call Nicole Allen at 517-353-6449.

273

REFERENCES

274

REFERENCES

Abbot, B. (1995). Interagency collaboration for children’s mental health services:
The San Mateo County model for managed care. Special Issues: Public sector planning
for managed mental health care. Administratign and Policy in Mental Health, 22, 301-
3 13.

Abernathy, 1., Edwards, D., & Gamache, D. (1999). Navigating a tough course:
Advocacy and collaboration. Presentation at the Next Millennium Conference: Ending
Domestic Violence, August, 1999.

Allen, D.G., Gilchrist, L.D., Brown, L., Cox, G.B., Semke, J., Thomas, M.D., &
Perry, RD. (1994). One system, many perspectives: Stakeholders and mental health
system evaluation. Evaluation and Program Planning, 17, 47-51.

 

Anderson, M.L. & Collins, RH. (1998). Race, class and gender: an anthology.
Belmont, CA: Wadsworth Publishing Company.

Anderson, N. R., & West, M. A. (1998). Measuring climate for work group
innovation: Development and validation of the team climate inventory. Journal of
Organizational Behavior. 19. 23 5-258.

Armbruster, C., Gale, B., Brady, J ., & Thompson, N. (1999). Perceived
ownership in a community coalition. Public Health Nursing, 16 (1), 17-22.

Barrow, J. C. (1976). Worker performance and task complexity as causal
determinants of leader behavior style and ﬂexibility. Journal of Applied Psychology, 61
(4), 433-440.

Bartunek, J. M., & Betters-Reed, B. L. (1987). The stages of organizational
creation. American Journal of Community Psychology, 15 (3), 287-303.

Bartunek, J. M., Foster-Fishman, P. G., & Keys, C. B., (1996). Using
collaborative advocacy to foster intergroup cooperation: A joint insider-outsider
investigation. Human Rﬁtions. 49 (6), 701-731.

Bidwell, C. E., Frank, K., & Quiroz, P. A. (1997). Teach types, workplace
controls and the organization of schools. Sociology of Education, 70, 285-307.

275

 

lull it! .alwlvfllIll b I

Bitter, G. B. (1977). From conﬂict to coalition: Disability groups, parents,
professionals. Volta Review. 79(5), 317-326.

Benard, B. (1989). Working together: principles of effective collaboration.
Prevention Forum. October. 4-9.

Berry, D. B. (1995). The Domestic Violence Sourcebook. Los Angeles, CA:
Lowell House.

Bliese, P. D. (2000). Within-group agreement, non-independence, and reliability:
Implications for data aggregation and analysis. In K. J. Klein & S. W. Kozlowski (Eds)
Multilevel Theog, Research and Methods in Organizations: Foundations. Extensions an_d
New Directions. San Francisco, CA: Jossey-Bass, 349 - 381.

Bond, M.A. & Keys, CB. (1993). Empowerment, diversity, and collaboration:

promoting synergy on community boards. American Journal of Community Psychology,
2_1, 37-57.

Bond, M. & Pyle, J. (1998). The ecology of diversity in organizational settings:
Lessons from a case study. Human Relations 51 (5), 589 - 623.

 

Buel, S. (1997). Family violence councils to the rescue to coordinate your
resources. The Texas Prosecutor, July/August.

Burt, M. R., Newmark, L. C., Jacobs, L. K., & Harrell, A. V. (1998). Evaluation
of the STOP Formula Grants Under the Violence Against Women Act of 1994. Report
produced by the Urban Institute, Washington, D. C.

Butterfoss, F. D., Goodman, R. M., & Wandersman, A. (1993). Community
coalitions for prevention and health promotion. Health Education Resrmch. 8 (3), 315-
330.

Butterfoss, F.D., Goodman, R.M., & Wandersman, A. (1996). Community
coalitions for prevention and health promotion: Factors predicting satisfaction,

participation, and planning. Health Education marterlv. 2_3, 65-79.

Brown, C. (1984). The art of coalition building: a guide for communityleaders.
The American Jewish Committee, New York.

Bryk, A. 8., & Raudenbush, S. W. (1992). Hierﬂhigl Linea; Models. Newbury
Park, CA: Sage Publications.

276

Burnett, R. E., (1993). Conﬂict in collaborative decision-making. In N. R. Blyler,

& C. Thralls (Eds) Professional Communication: The Social Perspective. Thousand
Oaks, CA: Sage Publications, 144 - 162.

Byles, J. A. (1985). Problems in interagency collaboration: Lessons from a project
that failed. Child Abuse_and Neglect, 9, 549-554.

Campbell, R. (1999). Exploring diverse appligatipna of multiple stakeholder
collaboratign: trials and triumphs across intervention domaips. Symposium at the 7th
Biennial Conference of the Society for Community Research and Action. New Haven,
CT.

Campbell, R., Sullivan, C., & Davidson, W. S. (1995). Women who use domestic
violence shelters: Changes in depression over time. Psychology of Women Quarterly, 19
237-255.

Carter, J. & Schechter, S. (1997). Child abuse and domestic violence: creating
community partnerships for safe families: suggested components of an effective child
welfare response to domestic violence. Family Violence Prevention Fund.

Cascardi, M. & O’Leary, K. D. (1992). Depressive symptomatology, and self-
esteem and self-blame in battered women. Journal of Lamilv Violenceg7(4), 249-259.

Chavis, D. M. (1995). Building community capacity to prevent violence through
coalitions and partnerships. Journal of health Care for the Pomind Underserved, 6 (2),
234-245.

Chavis, D. M. (1999). Evaluation of community collaborations and building their
capacity. Workshop presented by the Evaluators’ Institute, Washington, DC.

Chrislip, D. & Larson, C. (1994). Collaiaogrative leagship: How citizens and civic
leaders can make a difference. San Francisco: Jossey-Bass.

Chinman, M. J ., Anderson, C. M., 1mm, P. S., Wandersman, A., & Goodman, R.
M. (1996). The perception of costs and beneﬁts of high active versus low active groups in
community coalitions at different stages in coalition development. Journal of Community

Psychology, 24 (3), 263-274.

Clark, S. J., Burt, M. R., Schulte, M. M., & Maguire, K. (1996). Coordinated
wunity responses to domestic violence in six communities: Beyond the Justice
System. Final Report to the U. S. Department of Health and Human Services by the
Urban Institute. Washington, D. C.

277

Cohen, J ., & Cohen, P. (1983). Applied Multiple Regression/Correlation Analysis
fpr 1h; Bahaviorg Spiances. (2"‘I Edition). Mahwah, New Jersey: Lawrence Erlbaum
Associates, Inc., Publishers.

Cohen, J. (1988) Statistical Power Analysis for the Behavioral Sciences (2“1
Edition). Hillsdale, New Jersey: Lawrence Erlbaum Associates, Publishers.

Dobash, R. E., & Dobash, R. (1979). Violence Agagtst Wives. New York: Free
Press.

Dobash, R. E. & Dobash, R. (1992). Women. Violence and Social Change.
London: Routledge.

Eby, K. K., Campbell, J. O, Sullivan, C. M., & Davidson, W. 8., II (1995).
Health effects of experiences of sexual violence for women with abusive partners.” Health
_(_Z_are for Women International. 16. 563-576.

 

Edleson, J. L. (1991). Coordinated community responses. In M. Steinman (Ed),
Woman battering: Policy responses. Cincinnati, OH: Anderson Press, pp. 203-220.

Edwards, L. P. (1992). Reducing family violence: The role of the family violence
council. Juvenile & Family Court Journal. 43 (3), l - 18.

Fargason, Jr., C. A., Barnes, D., Schneider, D., & Galloway, B. W. (1994).
Enhancing multi-agency collaboration in the management of child sexual abuse. Child

Abuse and Neglect, 18 (10), 859-869.

Fawcett, S.B., Lewis, R.K., Andrews-Paine, A., Francisco, V.T., Richter, K.P.,
Williams, E.L., & Copple, B. Evaluating community coalitions for prevention of
substance abuse: the case of project freedom. ﬁalth Educationﬁand Behavior. 24. 812-
828.

Feighery, E., Rodgers, T. (1992). Building and maintaining effective coalitions:
How to guides in community health promotion no. 12. Palo Alto, CA: Standford Health
Promotional Resource Center.

Florin, P. (1996). 1996 Task Force Member Survey. Community Research and

Services Team. Center for Alcohol and Addiction Studies. Brown University, Rhode
Island.

278

Florin, P., Giamartino, G., Kenny, D. & Wandersman, A. (1990). Levels of
analysis and effects: Clarifying group inﬂuence and climate by separating individual and

group effects. Journal of Applied Social Psychology, 20, 881-900.

Florin, R, Mitchell, K, & Stevenson, J. (1993). Identifying training and technical
assistance needs in community coalitions: a developmental approach. Health Education
Research 8 417-432.

 

Forehand, G. A., & Gilmer, B. (1964). Environmental variation in studies of
organizational behavior. Psychological Bulletin, 62 (6), 361-382.

Foster-Fishman, P.G., Berkowitz, S., Lounsbury, D., Jacobsen, S., & Allen, N. E.
(in press). Creating collaborative capacity. American Journal of Community Psychology.

Foster-Fishman, P.G. Perkins, D.F., & Davidson, W.S. (1997). Developing
effective evaluation partnerships: paradigmatic and contextual barriers. Analise

Psicologig, 3, 389-403.

Foster-Fishman, P. G., & Keys, C. B. (1997). The person/environment dynamics
of employee empowerment: An organizational culture analysis. American Journal of

Community Psychology.

Foster-Fishman, P., Salem, D., Allen, N. E., & Fahrbach, K. (in press). Ecological
factors impacting provider attitudes towards human service delivery reform. American
Journal of Community Psychology, 27.

F oster-Fishman, P.G., Shpungin, E., Bergeron, J ., & Allen, NE. (2000). Unity
within diversity: The unique factors facilitating the participation of multiple stakeholders
in a community-based coalition. Unpublished manuscript, Michigan State University, E.
Lansing, MI.

Foster-Fishman, P. G., Salem, D., Allen, N. E., & Fahrbach, K. (in press).
Facilitating interorganizational exchanges: The contributions of interagency alliances.
American Journal of Community Psychology.

 

Frederickson, P. J. (1996). Community collaboration and public policy making:
Examining the long term utility of training in conﬂict management. American Behavioral
Science 39 (5), 552-569.

 

Frye, M. (1983). The Politics of Realm; Essays in feminist theory. Freedom, CA:
The Crossing Press.

279

Gamache, D. & Asmus, M. (1999). Enhancing networking among service
providers. In M. F. Shepard & E. Pence (Eds) Coordinating Community Resppnses 10
meestic Violence: Lessons from Duluth and Beyond. Thousand Oaks, CA: Sage
Publications, 65 - 88.

Glick, W. H. (1985). Conceptualizing and measuring organizational and
psychological climate: Pitfalls in multilevel research. Apademy of Management Reviaw,
19 (3), 601-616.

Glick, W. H. (1988). Response: Organizations are not central tendencies:
Shadowboxing in the dark, round 2. Academy of Management Review. 13 (1), 133-137.

Goodman, L. A, Koss, M. P., & Russo, N. F. (1993). Violence against women:
Physical and mental health effects. Part I: Research Findings. Applied and Preventive

Psychology, 2, 79-89.

Goodman, R.M., Wandersman, A., Chinman, M., Imm, P., & Morrissey, E.
(1996). An ecological assessment of community-based interventions for prevention and
Health promotion: approaches to measuring community coalitions. American Journal of
Community Psychology, 24, 33-61.

Gondolf, E. (1988). The survivor theory. In Battered womew survivors. CA:
Lexington, pp. 11-25.

Gottlieb, N.H., Brink, S., & Levenson Gingiss, P. (1993). Correlates of coalition
effectiveness: the smoke free class of 2000 program. Health Education Research 8 375-
384.

 

Gray, B. (1985). Conditions facilitating interorganizational collaboration. Human
Relations 38 (10), 911-936.

 

Gray, B. (1989). Collaboration: FindingCommon Ground for Multioarty
Problems. San Francisco, CA: Jossey-Bass, Inc.

Gray, B. (1996). Cross-sectoral partners: Collaborative alliances among business,

government and communities. In C. Huxman (Ed) Creating Collaborative Advantage.
Thousand Oaks, CA: Sage Publications.

280

Gruber, J. & Trickett, El (1987). Can we empower others? The paradox of
empowerment in the governing of an alternative school. American J oum of Community

Psychglogy, 15, 353-371.

Gwinn, C. G., & O’Dell, A. (1993). Stopping the violence: The role of the police
ofﬁcer and the prosecutor. Western State University Law Review 20. 1501-1521.

Haber, J. D. & Roos,’ C. (1985). Effects of spouse abuse and/or sexual abuse in
the development and maintenance of chronic pain in women. Advances in Pain Research

and Therapy, 9, 889-895.

Harrison, P.J., Lynch, E.W., Rosander, K.,. & Borton, W. (1990). Determining
success in interagency collaboration: an evaluation of process and behaviors. Infants and

Young Children, 31, 69-78.

Hart, B]. (1995). Coordinated community approaches to domestic violence.
Paper presented at the Strategic Planning Workshop on Violence Against Women,

National Institute of Justice, Washington DC. www.mincava.umn.edu.

Heller, K., Price, R. H., Shulamit, R., Riger, S., Wandersman, A. (1984).
Psychology and Community Change: Challenges for the future. Paciﬁc Grove, CA:
Brooks/Cole Publishing.

Himmelman, A. T. (1996). On the theory and practice of transformational
collaboration: From social service to social justice. In C. Huxham’s (Ed) Creating
Collaborative Advantage. Thousand Oaks, CA: Sage Publications.

Hord, S. (1986). A synthesis of research on organizational collaboration._
Educational Leadership, February, 22-26.

Huxham, C., & Vangen, S. (2000). Ambiguity, complexity and dynamics in the
membership of collaboration. Human Relations 53 (6), 771-793.

James, L. R. (1982). Aggregation bias in estimates of perceptual agreement.
Journal of Applied Psychology, 67 (2), 219-229.

James, L. R., & Jones, A. P. (1974). Organizational climate: A review of theory
and research. Psychological Bulletin. 81 (12), 196-1112.

James, L. R., Joyce, W. F., & Slocum, J. W. (1988). Comment: Organizational do
not cognize. Academy of Management Review. 13 (1), 129-132.

281

Jenkins, P. & Davidson, B. (1999). Consensus and contradictions in
understanding domestic violence: implications for policy and model programs. In T.
Gullotta and S. McElhand (Eds) Violence in Homes and Communities: Prevention
I_n_tervention and Tﬁtﬁment I§_sues in Children’s and Families’ Lives. Thousand Oaks, CA:
Sage Publications, 245 - 267.

 

Jones, T. S., & Bodtker. (1998). A dialectical analysis of a social justice process:
International collaboration in South Africa. Journal of Applied Communication Research,
g, 357-373.

Kagan, S.L. (1991). United we stand: Collaboration for child care and early
education amices. New York: Teachers College Press.

 

Kegler, M.C., Steckler, A., McLeroy, K. & Malek, SH. (1998). Factors that
contribute to effective community health promotion coalitions: a study of 10 Project
ASSIST Coalitions in North Carolina. Health EducatiorLand Behavior. 25. 338-353.

Klein, K. J ., & Sorra, J. S. (1996). The challenge of innovative implementation.
Academy of Management Review. 21 (4), 1055-1080.

Koslowski, S. W. J ., & Hults, B. M. (1987). An exploration of climates for
technical updating and performance. Personnel Psychology, 40, 539-563.

Kreft, I., & DeLeeuw, J. (1998). Introducing Multi-Level Modelirg, London:
Sage Publications.

Kumpfer, K.L., Turner, C., Hopkins, R. & Librett, J. (1993). Leadership and team
effectiveness in community coalitions for the prevention of alcohol and other drug abuse._
ﬂe_a,lth Education Research 8 359-374.

 

Larson, J. R., Jr., Foster-Fishman, P. G., Franz, T. M. (1998). Leadership style
and the discussion of shared and unshared information in decision-making groups. PSPB,
2_4 (5), 482-495.

Larson, J. R., F oster-Fishman, P. G., & Keys, C. B. (1994). Discussion of shared
and unshared information in decision-making groups. Journal of PersomJity and Socﬂ

Psychology, 67 (4), 446-461.

Larson, C. E., & LaFasto, F. M. J. (1989). Mm Work: What must go @ht/Wha;
can go wrong. Newbury Park, CA: Sage Publications.

282

Lipman, P. (1997). Restructuring in context: A case study of teacher participation
and the dynamics of ideology, race, and power. American Educational Research Joumjﬂ,
3_4_ (1), 3-37.

Mansergh, G., Rohrbach, L. A.., Montgomery, S. B., Pentz, M. A., & Johnson, C.
A. (1996). Process evaluation of community coalitions for alcohol and other drug abuse
prevention: A case study comparison of researcher- and community-initiated models.
Journal of Communig Psychology, 24, 118-135.

Martin, J. (1992). Cultures in Organizations: Three perspectives. New York:
Oxford University Press.

Mattessich, P.W., & Monsey, BR. (1992). Collaboration: What makes it work: a
review of resezp'ch literature on factors inﬂuencing successﬁil collaboration. Wilder
Research Center: St.Paul, Minnesota.

McCann, J. E. (1983) Design guidelines for social problem-solving interventions.
Journal of Applied Behavioral Science. 19. 177-189.

McCann, J E & Gray, B. (1986). Power and collaboration in human service
domains. International Journal of SociologjLand Social Policy. 6. 58-67.

McIntosh, P. (1989). White privilege: Unpacking the invisible knapsack. Peace
and Freedom. July/August, 10-12.

McEvoy, A., Brookings, J .B., & Brown, CE. (1983). Responses to battered
women: problems and strategies. Social Casework: The Journal of Contemporag Social
Work, February, 92-96.

McNulty, B. A. (1986). Interagency cooperation: The state of the art.
Unpublished manuscript, Colorado Department of Education, Denver.

Moos, R. H. (1973). Conceptualizations of human environments. American
Psychologist, 28, 652-665.

Moos, R. H. (1979). Social climate measurement and feedback. In R. Munoz, L.
Snowden, & J. Kelly (Eds), Social and Psychological Reseaﬁrch in Communjty Setting;
San Francisco: J ossey-Bass.

 

O’Dell, A. (1996). Domestic violence homicides. The Police Chief. February,
1996.

283

Pence, E. L. (1996). Coordinated Communng Response to Domestic Assault

Cases: A Qpide fpr Policy Development. Available from the Domestic Abuse
Intervention Project, 206 W. Fourth Street, Duluth, MN 55806.

Pence, E. L. (1999). Some thoughts on philosophy. In M. F. Shepard & E. Pence

(Eds) Cpordinating Community Responses to Domestic Violence: Lessons from Duluth
and Beyond. Thousand Oaks, CA: Sage Publications, 25 - 40.

 

Pence, E. L., & McDonnell (1999). Developing policies and protocols. In M. F.
Shepard & E. Pence (Eds) Coordinating Community Responses to Domestic Violence:
Lessons from Duluth and Beyond. Thousand Oaks, CA: Sage Publications.

Pence, E. L., & Ritmeester, T. (1992). A cynical twist of fate: How process of
ruling in the criminal justice system and the social sciences impede justice for battered
women. University of Southern CalifomiaLReview of Law and Women’s Studie§,_2_, 25 5-
292.

Pence, E. L., & Shepard, M. F. (1999). An introduction: Developing a
coordinated community response. In M. F. Shepard & E. Pence (Eds) Coordinating

Community Responses to Domestic Violence: Lesson_§ from Duluth and Beyond.
Thousand Oaks, CA: Sage Publications, 3 - 24.

Penner, S. (1995). A study of coalitions among HIV/AIDS service organizations.
Sociological Perspectives, 38, 217-239.

Pfeffer, J. (1996). Understanding the role of power in decision-making. In J. M.
Shaﬁitz & J. S. Ott (Eds) Classics of Organization Theog. (4th Edition). Philadelphia:
Harcourt Brace College Publishers, 359-3 74.

Powell, G. N., & Butterﬁeld, D. A. (1978). The case for subsystem climates in
organizations. Academy of Management Review. 3. 151-157.

Pritchard, R. D. & Karasick, B. W. (1973). The effect of organizational climate
on managerial job performance and job satisfaction. Organizational Behavior aprd Human
Performance 9 126-146.

 

Radford, J. & Stanko, EA. (1996). Violence against women and children: the
contradictions of crime control under patriarchy. In Hester, M., Kelly, L., & Radford, J.

(Eds), Women. violence and male power: feminist activism, research and practice (pp.65-
89). Philadelphia, PA: Open University Press.

 

284

Reilly, T. (1998). Communities in conﬂict: Resolving differences through
collaborative efforts in environmental planning and human service delivery. Journal of
Sociology and Social Welfare. XXV (3), 115-141.

Rogers, T., Pitney-Howard, B., Feighery, E.C., Altman, D.G., Endres, J .M., &
Roeseler, AG. (1993). Characteristics and participant perceptions of tobacco control
coalitions in California. Health Education Research 8 345-357.

 

Rothschild-Whitt, J. (1979). The collectivist organization: An alternative to
rational-bureaucratic models. American Sociological Review. 44. 509-527.

Roussos, S. T., & Fawcett, S. B. (2000). A review of collaborative partnerships
as a strategy for improving community health. Annual Review Public Health 21 369-
402.

 

Rousseau, D. M. (2000). Multilevel competencies and missing linkages. In K. J.
Klein & S. W. Kozlowski (Eds) Multilevel Theogy, Research and Methods in
Organizations: Foundations, Extensions and New Direction_s, San Francisco, CA: Jossey-
Bass, 572 - 582.

Schechter, S. (1982). Women & Male Violence: The visions and struggles of the
_b_aLttered women’s movement. Boston, MA: South End Press.

Schechter, S. (1999). New Challenges for the Battered Women’s Movement:
Building Collaborations and Improving Public Policy for Poor Women. Published by the
National Resource Center on Domestic Violence, Harrisburg, PA.

Schechter, S. & Edleson, J. (1994). In the best interest of women and children: a
call for collaboration between child welfa1e_and domestic violence constituencies. Brieﬁng
paper prepared for the conference Domestic Violence and Child Welfare: Integrating
Policy and Practice for Families, Racine, Wisconsin.

Schneider, B. & Reichers, A. E. (1983). On the etiology of climates. Personnel
Psychology, 36, 19-39.

Shafritz, J. M., & Ott, J. S. (1996). Classics of Organization Theory. (4th
Edition). Philadelphia: Harcourt Brace College Publishers.

Shinn, M. (1990). Mixing and matching: Levels of conceptualization,
measurement, and statistical analysis in community research. In P. Tolan, C. Keys, F.
Chertok, & L. Jason (Eds), ReseLchiryg Community Psychology: Issues of theory and
methods. Washington, D. C., American Psychological Association, 111 - 126.

285

Short, L. M., & DeBruyn, L. M. (2000). Coordinated Community Responses To
Prevent Intimate Partner Violence: Application of the Public Health Model. Unpublished
Manuscript. Analytic System Associates, Inc.

Sink, D. (1996). Five obstacles to community-based collaboration and some
thoughts on overcoming them. In C. Huxham’s (Ed) Creating Collaborative Advantage.
Thousand Oaks, CA: Sage Publications, 101 - 109.

Skaff, L. (1988). Child maltreatment coordinating committees for effective service
delivery. Child Walfge Laagpe of Ameriaa, 3, 217-230.

Smith, K. K., & Berg, D. N. (1997). Paradoxes of Group Life: Understanding
conﬂict, paralysis, and movement in group dmamics. San Francisco: Jossey-Bass
Publishers.

Staggenborg, S. (1986). Coalition work in the pro—choice movement:
organizational and environmental opportunities and obstacles. Social Problems, 33, 374-
390.

Sullivan, CM. (1991). Battered women as active helpseekers. Violence UndaLe,
August, pp. 1,8.

Sullivan, C.M. & Bybee, D.I. (1999). Reducing violence using a community-based
advocacy project for women with abusive partners: Two-year follow-up. Journal of

Consulting and Clinical Psychology, 67 (1), 43-53.

Sullivan, C.M. & Keefe, M. (1999, Fall). Evaluations of advocacy efforts to end
intimate male violence against women. Michigan Coalition Against Domestic and Sexual
Violence (MCADSV) Review. 2. 3 and 11-14.

Sutherland, C., Bybee, D., & Sullivan, C. (1998). The long-term effects of
battering on women’s health. Women’s Heamh: Research on ﬁnder. behavior. and policy,
4 (1), 41-70.

Tjaden, P. & Thoennes, N. (1998). Prevalence, incidence and consequences of
violence against women: Finding from the National Violence Against Women Survey.
Research Brief. National Institute of Justice and the Centers for Disease Control and
Prevention.

Tjosvold, D., & De Dreu, C. (1997). Managing conﬂict in Dutch Organizations: A
testing of the relevance of Deutsch’s cooperation theory. Joumg of Applied Social
Psychology, 27 (24), 2213-2227.

286

Winer, M., & Ray, K. (1994). Collaboration handbook: Creating, sustaining, and
enjpya'ng the joumey. Saint Paul, MN: Amherst H. Wilder Foundation.

Wischnowski, M. W., & McCollum, J. A. (1995). Managing conﬂict on local
interagency coordinating councils. Topics in Early Childhood Education, 15 (3), 281-295.

Wood, C]. (1989). Challenging the assumptions underlying the use of
participatory decision-making strategies. Small Group Behavior, 20, 139-162.

Wood, D. J ., & Gray, B. (1991). Toward a comprehensive theory of
collaboration. Journal of Applied Behavioral Science. 27 (2), 139-162.

Yin, R. K., & Kaﬁarian, S. J. (1997). Introduction: Challenges of Community-
Based Program Outcome Evaluations. Evaluation and Program Planning, 20 (3), 293-
297.

 

Zapka, J .G., Marrocco, G.R., Lewis, B., McCusker, J ., Sullivan, J ., McCarthy, 1.,
& Birch, X. (1992). Inter-educational responses to AIDS: a case study of the Worcester
AIDS Consortium. 331th Education Ougterly. 7. 31-46.

287