«a:
a .
[1.14:2

fix

v .

 

 

 

 

. v
:nifiqntil I
j... 5?

\
. m.
in .
e a... .
x E»...

r.» ,2 a ..

L . A
J .v a .
..r.. Junta“ ..a .11.... £5; a x
aw. a .41 .mW a.

Elm: _ «Eu
.fizmwuhaw. .

u. in.

$3“ .

..s

{'1

i .. .
t:
.5
\Q ..

”mm

X: 11$.
. 5h", .1
(.2... at;
.12 a?
l ,

"l .
s: x ¢
Lt. "-991

1 ,
.....
.

. h. 3‘ -
uxfl. , .
1 ‘ $‘v.e\?iatt.lta
x .. Itilit i...
‘22? .1310...
i 3.2}!

 

'l ICQ‘|.

100)

This is to certify that the
dissertation entitled

SOCIAL CAPITAL IN A
COMMUNITY COLLABORATIVE SETTING

presented by

Mary B. Mc Donald

has been accepted towards fulfillment
of the requirements for

PhD. Child and Family Ecology

degree in

 

 

 

Majo professor

Date December 7, 2002

 

MSU is an yfirmatiw Action/Equal Opportunity Institution 0-12771

 

 

 

.UBRARY i
Michigan State
University

 

 

 

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

 

DATE DUE DATE DUE DATE DUE

 

NW :1 .8 arm

 

SEP 1 7 200!r

 

$3er 8 mar

 

   

NOV 14’ 2504

 

SIEP “135” i009?

 

 

 

 

 

 

 

 

 

 

6/01 c:/CIRC/DateDue.p65-p.15

SOCIAL CAPITAL IN A COMMUNITY COLLABORATIVE NETWORK
By

Mary B. MC Donald

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Family and Child Ecology

2002

ABSTRACT

SOCIAL CAPITAL IN A COMMUNITY COLLABORATIVE NETWORK
By

Mary B. Mc Donald

Community collaboration has grown significantly in the last two decades.
With those efforts have come many opportunities for community organizations to
form both formal and informal networks to facilitate the work (Foster-Fishman,
Berkowitz, Lounsbury, J acobsen & Allen, 2001). These collaborative networks
operate on both organizational and personal ties (Bruner & Chavez, 1996; Chavis,
1994; Wolff, 2001) and can be analyzed using social techniques. In addition,
given the multiple levels of the networks, the collaboratives can be conceptualized
using dimensions of social capital.

This study examined the relationships between members of a community
collaborative network. It had two purposes. The first purpose was to learn how
members build, maintain, and access resources in a collaborative network and the
relevance of applying social capital as a concept to explain the relationships
among members of the network. The second purpose was to pilot the use of
quantitative network analysis techniques and qualitative interviews and
observations to inform participants and encourage action.

The results of the quantitative sociograms, key informant interviews, and
participant observation indicated that social capital was evident in stakeholders’

relationships. Resources, such as money, information, and access, were

distributed in the network through social ties. In addition, mandates, proximity,
trust, reciprocity, selection, influence, and work were distinguished by social
capital level: individual, dyad, and network.

The method encouraged collaborative dialogue and group learning. Given
the personal vulnerability involved with a network analysis, however, this social
capital study stopped short of participatory action. Stakeholders preferred that the
information be interpreted and reported, rather than participating in action
planning.

The results of the study implore policy and grant makers to encourage the
development and support of collaborative networks. In addition, it provides a
method to observe other collaborative structures to further study the presence and

influence of social capital in access to and distribution of resources.

To Dale Mc Donald,
the first of Grace and George Mc Donald’s children
to go to Michigan State,
from the last one.

ACKNOWLEDGMENTS

There is an old adage that advises, “A smart man knows everything; a wise man
everybody.” This dissertation is the result of that adage as it is based on the collective
wisdom that came from kindred souls. I am in awe at the number of people who helped
me. I am grateful, of course, that you did. I will not forget any of you for your interest,
patience, and gentle encouragement. I also am grateful for those moments when some
were less than gentle.

I want to thank Joanne Keith, Chico Villarruel, and Annette Abrams. I also thank
Ken Frank, David Imig, and Esther Onaga for their willingness to help.

I want to thank the FCC, who gave me flexibility in my job and demonstrated a
willingness to learn with me in this effort. I recognize my personal board: Janeane, who
talks us off the ledge; Linda, who loves us because we live and breathe; and Susan, our
ever-steady counsel. I thank Pam Leppek for emotional engagement through the Ph.D.
process, not once, but twice. I thank Bruce and Craig and Tom, whom I love like
brothers. Ithank Judy and Char, my first to call; Connie Navarro, my first amiga; and my
sister Jennie, my first best friend. I am forever in debt to Lynn Mack and to Kern Tanner.

I respect and thank my parents for teaching me tenacity, my brothers and sisters
for teaching me independence, and my nieces and nephews for teaching me hope. We

have come full circle, the Mc Donalds and Michigan State. We have done our job.

TABLE OF CONTENTS

LIST OF TABLES ....................................................................................................... viii
LIST OF FIGURES ....................................................................................................... ix
CHAPTER I
INTRODUCTION TO THE STUDY .............................................................................. 1
Background ................................................................................................................ 1
Statement of the Problem ........................................................................................... 2
Purposes of the Study ................................................................................................. 3
Significance of the Study ............................................................................................ 3
Conceptual Framework .............................................................................................. 5
Research Questions .................................................................................................. 11
Assumptions ............................................................................................................ 12
CHAPTER 11
REVIEW OF THE LITERATURE ............................................................................... 13
Introduction ............................................................................................................. 13
Collaborative Organizations ..................................................................................... 13
Philosophical Foundation of Collaboratives ...................................................... 13
Practical Aspects of Collaboratives ................................................................... 15
Social Networks ....................................................................................................... 17
Social Capital ........................................................................................................... 21
Participatory Action Research .................................................................................. 25
CHAPTER [[1
RESEARCH METHODOLOGY .................................................................................. 27
Introduction ............................................................................................................. 27
Research Objectives and Questions .......................................................................... 27
Participatory Framework .......................................................................................... 28
Role of the Researcher ............................................................................................. 35
Research Design ....................................................................................................... 39
Research Context ..................................................................................................... 42
Background of the Family Coordinating Council .............................................. 42
Background on the Social Network Study ......................................................... 45
The Research Population .......................................................................................... 46
Data Sources and Research Tools ............................................................................. 47
Archival Data ................................................................................................... 48
Stakeholder Meetings ....................................................................................... 48
Community Delivery Survey ............................................................................ 48
Interviews With Key Informants ....................................................................... 50
Conceptual and Operational Definitions ................................................................... 51
Influence .......................................................................................................... 51
Mandates .......................................................................................................... 51

Proximity ......................................................................................................... 52

Selection .......................................................................................................... 52

Social Capital Dimensions ................................................................................ 53
Social Network ................................................................................................. 53
Strong/Weak Ties ............................................................................................. 54
Research Validity ..................................................................................................... 54
Limitations ............................................................................................................... 55
CHAPTER IV
RESULTS ..................................................................................................................... 56
Introduction ............................................................................................................. 56
Data Gathering and Generation of Results ................................................................ 56
Point 1: Stakeholder Meetings ......................................................................... 58
Point 2: Community Delivery Survey .............................................................. 64
Point 3: Stakeholder Meeting: Interpret Sociograms and Inform About
Key Stakeholder Interviews ......................................................................... 67
Point 4: Snowball Sample ................................................................................ 74
Point 5: Data Analysis of Key Informant Interviews ........................................ 78
Point 6: Sociogram/Interview Triangulation .................................................... 86
Point 7: Stakeholder Meeting: Reactions, Reflections, and More Questions 92
CHAPTER V
FINDINGS AND IMPLICATIONS .............................................................................. 94
Restatement of Goals and Objectives ....................................................................... 94
Major Findings: Question One ................................................................................ 95
Major Finding: Question Two ................................................................................ 105
How Research Advances Scholarship and Collaborative Practice ............................ 106
Influence on Personal Practice ................................................................................. 107
Research Implications ............................................................................................. 109
Policy Implications ................................................................................................. 109
Questions for Future Research ................................................................................. 111
Benefits and Limitations of Insider Research ........................................................... 111
APPENDICES
A. Community Delivery Survey ............................................................................. 114
B. Informant Interview Protocol ............................................................................. 115
C. Letter to Members of the Stakeholder Group ..................................................... 116
D. UCRIHS Approval Letter .................................................................................. 120
BIBLIOGRAPHY ........................................................................................................ 121

vii

LIST OF TABLES

3.1 Models of Action-Oriented Research ................................................................... 29
4.1 Snowball Sample Subgroups ............................................................................... 76
5.1 Collaborative Dimensions .................................................................................. 101

viii

1.1

1.2

3.1

3.2

3.3

3.4

4.1

4.2

5.1

B.l

B2

B3

B4

B5

LIST OF FIGURES

Social Capital Among Actors in a Network ........................................................... 6
Conceptual Map .................................................................................................... 8
Social Capital in a Community Collaborative Network Research Design ............. 40
Organizational Evolution of the Family Coordinating Council ............................. 43
Focus Groups Involved in the Community Assessment Process, 1995 ................. 44
Relationship Among Research Tools and Data Sources ....................................... 47
Social Capital in a Community Collaborative Network: Generation of

Results ................................................................................................................ 57
Illustration of Importance Ties in the Collaborative ............................................. 68
Position of Key Stakeholders Within the Importance Sociogram ........................ 104
Sociogram: Importance Ties ..................................................................................
Sociogram: Substance Abuse .................................................................................
Sociogram: Violence ..............................................................................................
Parenting ................................................................................................................
Teen Pregnancy ......................................................................................................

CHAPTER I

INTRODUCTION TO THE STUDY

Background

During the 19908, a strong focus on coalition building at the community level
emerged, spurred by funding requirements and an interest in maximizing organizational
capacity and assets (Gittel & Vidal, 1998, p.14). In Michigan, for example, a 1994
Executive Order established a state central committee to oversee the coalition process.
“Putting It Together for Michigan Families” or the PIT crew, comprising the state
directors of the departments of mental health, public health, education, social services,
and aging, was formally identified to help communities “bust barriers” to agency
collaboration and service integration (Executive Order, 1994). The governor’s office
earmarked specific finding that passed from the state to the local, county level. To
access the money, county public and private agencies were required to form a
collaborative group and to construct a plan that articulated both collective capacity and
shared resources.

The collaborative-building action was designed to formally open the community
planning and service delivery table. PIT executive actions required that the collaborative
group at the community level include a “threshold” membership roster. The traditional
“big public agencies” (e.g., Family Independence Agency, Family Court, the schools, and
law enforcement) were required by their state departments to organize formally with
private agencies and with each other (Executive Order, 1994; Putting It Together for

Michigan Families, 2002).

The state-level emphasis on community collaboration brought a new dimension to
community agencies and their organizational structure. It is not that agencies did not
cooperate before the PIT process was established. In one western Michigan county, for
example, two standing committees existed with the sole purpose of coordinating and
integrating services to families. After 1994, however, a collaborative structure was
required by state offices and by funding restrictions. Before an agency could access
certain funds, it was required to document collaborative agreements and processes, so
agencies forged alliances and partnerships. Community collaboratives became the
infrastructure for organizations to formally promote mutual goals (Chaskin, 1999;
Chavis, 1995) and met the threshold requirements of mandates (Putting it Together for
Michigan Families, 2002). Thus, a community-directed and collaborative way of doing
business was transformed into a mandated structure (Foster-Fishman, Salem, Allen, &
Fahrbach, 2001). “The Collaborative” became a proper noun.

Community collaboratives have been the subject of numerous studies designed to
assist in both the creation (Chavis, 1995; Gulati & Garsinlo, 1999; Wolfe, 2001) and the
evaluation (Goodman, Wandersman, Chinman, Irnrn, & Morrisey, 1996) of such efforts.
One of the overarching themes of this literature is that collaboratives are an emerging
network structure in communities and that, to understand them, one must consider both

the individual and the network as units of analysis (F oster-Fishman et a1., 2001).

Statement of the Problem
Faced with legislative mandates and funding requirements, community members
have formed networks to formalize a collective process to plan and deliver services to
families. These networks are largely defined by the social, relational ties between

members; therefore, it would seem that the underlying social structure would be the

2

actual infrastructure for collaborative work. Yet, without an understanding of the nature
of the relational ties or a method to analyze them, collaborative members have no

systematic way to inform their work.

Purposes of the Study

In this research, I examined the relationships among members of a community
collaborative network. I had two purposes for conducting this study. The first was to
learn how members build, maintain, and access resources in a collaborative network and
the relevance of applying social capital as a concept to explain the relationships among
members of the network. The questions underlying this purpose were: Do relationships
influence access to and distribution of resources? If so, in what way?

The second purpose was to pilot the use of quantitative network analysis
techniques and qualitative interviews and observations to inform participants and
encourage their action. Questions underlying this purpose were: How do members of the
collaborative respond to information about relational patterns? Do members

(participants) apply the results as a development strategy in their community work?

Significance of the Study

Work in a community is a social process. It involves particular people coming
together for a specific purpose. Therefore, it is based on ties between and among those
pe0ple, their positions within the community, and the potential that lies in the interaction
of those ties and positions.

In this research I assessed collaborative members’ ties and positions by applying
the framework of social capital and testing whether the resulting information would

inform collaborative action. Theoretically (and practically), research of this nature is

important because work in a community is a social process that requires social analysis to
be understood.

Community members come to collaborative work representing their own
particular system’s constituency and responsibilities. The Family Court, for example has
a legislated responsibility to enforce the policies of the juvenile justice code. Schools
have a legislated responsibility to provide equal Opportunity for quality education. A
private, mission-driven delinquency prevention program shares responsibilities with both
the court and the schools by reducing delinquency as enforced by the court among the
adolescents who attend the school. Collaborative infrastructure was established to
formalize the system integration for people such as those adolescents.

Yet researchers have found that interagency, coalition work is accomplished
through the relationships that exist, not necessarily through the creation of special groups
(Chavis, 1995; Foster-Fishman, Berkowitz, Lounsbury, Jacobsen, & Allen, 2001).
Research has not, however, provided insight into how the underlying social structure is or
is not influenced by the formal infrastructure. Collaboration takes effort, both in time
(Cohen & Prusak, 2001) and resources (Taylor, 2002; Wolfe, 2001). Are community
relational ties a motivating reason for an organization to participate in such an effort, or is
it just those organizations required to participate? Do people come together because they
have to and then access resources, or is there, in fact, an underlying social structure that
informally gives access to the formal collaborative process? Is the presence of social
capital in relationships or the wealth of social capital in a network the result of the
collaborative infrastructure, or is the opposite the case? This research was based on the

assumption that it is important to answer such questions.

Conceptual F Lameworlg

The overarching theoretical framework for this research was based on a human
ecological view of individuals, families, and organizations as units that interact and
transform larger systems within the natural, physical, human-constructed, and
sociocultural/behavioral environments (Andrews, Bubolz, & Paolucci, 1980; Buboltz &
Sontag, 1994). Within a community, individuals and their relational ties form such units,
and from an ecological perspective they are studied as units and as part of a larger
ecosystem where the whole has equal consideration to the unit. In addition, a human
ecological model of research calls for multiple applicability and generalization that can
be translated into practical application (deGroot, 1988; Westney, Brabble & Edwards,
1988). True to the model, the present study was designed as participatory action research
using participants to define the study parameters, research questions, and design, and to
interpret the results. In determining the utility of the results in informing collaborative
action, I could test the applicability of the work.

The emerging concept of social capital was the primary source of theory. Social
capital is a reliable conceptual framework for community collaborative analysis for two
reasons. First, social capital accommodates the individual, the dyad, and the network
units of analysis (Coleman, 1990; Wasserman & Faust, 1994). Second, the dimensions of
social capital define social ties in the network and distinguish those that hold the potential
for access to resources and mutual action.

According to Coleman (1990), social capital exists in an actor’s individual
(micro) position related to other actors in the network. That social capital, in turn, has
implications for the network (macro) in that context. The social capital framework

allows a researcher to observe individuals and their context, to make micro/macro

transitions (Frank, 1995; Paxton, 2002). For illustration, consider three actors who are
part of a network of agency directors. Actor A has a tie with actor B, who in turn has
such a tie with actor C. The relationship tie between actors A and B is a social-capital tie,
one that holds actual or potential access to resources (Bourdieu, 1985) where the actors
'share values, have reciprocal regard, and act to mutual benefit (Coleman, 1990). The tie
between actors B and C possesses social capital as well. Access to resources within the
network, the potential between AB and B/C is implied by the position of the three actors
within the network. Actors A and B share a level of potential and actors B and C share a
level of potential, but to access or distribute resources through the network, actor A
would want to go to actor B and actor C would want to go to actor B. Actor B, on the
other hand, could go to either actor A or B and could, theoretically, access resources for
one from the other. The relationships between actors have implications for resource

based on reciprocity and mutual action. Figure 1.1 illustrates how social capital flows.

0 Individual agency directors,
actors in the network

Social ties that possess

¢ social capital providing
the potential for accessto
resources

Figure 1.1: Social capital among actors in a network.

Within the context of the network, the potential to access resources or social
capital is greater than just the two separate relationships. Basically, the whole is greater
than the sum of the parts. Coleman (1988) asserted that social capital creates a form of
organizational structure in which the structure of relationships among people facilitates
action and produces results that would not be possible without such organization.
Applying this notion to actors A, B, and C, then, the position of actor B influences the
flow of resources within this network as the resources flow from actor C to actor A.

After reviewing a number of collaborative studies, Blackman and Smith (2000)
concluded that the quality of social ties or linkages within a community directly affects
the capacity of social networks to mobilize for collective action and address social
problems. This quality of ties can be represented in the presence of social capital as those
particular social ties, the “stock of active connections among people” (Cohen & Prusak,
2001, p. 14). The presence of social capital qualifies the social ties of a network and, as
such, is a fundamental dimension of a community collaborative network and a valuable
asset to communities (Woolcock & Narayan, 1995).

The relationship between social capital and a community collaborative network is
shown in Figure 1.2. When an individual becomes involved in a network, one or more of
the three individual-social-ties situations may exist for that person. The person may or
may not have social ties to other members of the network before the network association.
Oliver and Ebers (1998) defined a social network as a pattern of interactions that connect
individual actors with one another. Applied to this study, those actors would be the
members of the collaborative, and the social ties for those individuals could be any
combination of the three illustrated in Figure 1.2. One actor may have some ties with

other collaborative network actors that possess social capital and ties that do not. There

may also be network actors with whom the individual has no ties. All of the potentially
varied social ties are brought into the collaborative network, where, through the formal

association of the collaborative, the individual ties may be influenced.

Social Capital
in a Community Collaborative Network

Network Social Ties 2‘5 (V >

 

 

 

 

 

with Social Capital Network No Network
oShared values Social Ties Social Ties
oRecrprocrty
oMutual Action

 

 

 

 

 

 

 

 

 

 

l

 

 

 

   
 
 
 
    
   
  

( Legend
Network Structure \ ‘4 .
-Mandates and Proximity [1 Social
-Trust ties
-Reciprocity
-Selection N k
-Influence etwor
A -Work D structure
Increased

 

 

J Connection
m between

ties

social capital
Increased social

collaboration
. Access to

  
   

resources
‘ Outcome 1 Direction
of ties
Decreased social capital
Decreased collaboration B 0
Decision

 

 

 

Figure 1.2: Conceptual map.

The structure of a community collaborative network is defined formally and
informally. The formal structure includes traditional organizational items such as
governing agreements, job descriptions, and organizational charts. It is the formal
structure that is most often the focus of collaborative research. (Chavis, 2001; Wolfe,
1999). Those organizational items do connect the individual members by providing
structure and context. However, in a review of 80 articles, Foster-Fishman and her
colleagues (2001) concluded that collaboration requires broader relational networks (as
opposed to traditional delivery) and new ways of interacting within established networks.
The work of interagency coalitions within communities is accomplished through the
relationships that exist (Chavis, 1995), which can be viewed successfully through an
analysis of the network’s social ties.

In Figure 1.2, the social ties converge into the process labeled Network Structure,
which includes all ties, both formal (organizational structure) and informal (social
ties/relational structure). Formal ties include the working agreements that define the
members’ prescribed access or proximity both to one another and to the formally
mandated work. Into that formal process flows the relational structure, represented by
the varied social ties of the individual members. Interagency collaboration is a pattern of
association and includes exchange, sharing of resources, and joint efforts (Foster-

F ishman et al., 2001; Gulati & Gargiulo, 1999).

When an interagency collaborative is studied as a social network, the associations
between and among members can be analyzed more precisely. A social network is a
specific type of relationship linking a defined set of persons, objects, or events (Mitchell,
1969), one that includes both a framework for identifying how a set of actors is linked

together and the consequences of those linkages for the actors and the system (Knoke &

Kuklinski, 1982). Within the collaborative network, individual actors may form
subgroups and hold a primary affiliation with members of their subgroups while they also
identify through ties with others (Frank & Yasumoto, 1998). There is a selection that
occurs when one actor chooses to interact with others based on their attributes
(Wasserman & Faust, 1994, part VI). Theoretically, the selection process occurs a
network for one of two reasons. People choose others either because they are similar to
themselves (balance seeking) or to obtain some type of information (information
seeking). The information can help a person achieve a goal or gain power; it can be used
to gain understanding, reduce uncertainty, or inoculate an individual against risk (Frank
& F ahrbach, 1999). Influence can affect an actor’s pattern of social ties in a network.
Interpersonal influence occurs when one influences what another thinks and subsequently
does (Frank & Fahrbach, 1999).

At the junction illustrated in Figure 1.2 as the Network Structure process, the
linkages of individuals to each other and to the system are combined with the
organizational structures of collaboration. At some point in that process, decisions are
made about the individual and system linkages, and the consequences of those linkages
become clear.

The consequences are represented in Figure 1.2 as arrow A and arrow B. If the
process results in increased actual or potential access to resources (social capital) or
increased interagency collaboration (shared resources, joint efforts), then the
consequences would follow a feedback loop. In that case, the collaborative network
process would have built social capital. If the process results in a decrease, then the

collaborative network would have “burned” social capital, or at least failed to build any.

10

The point where individual social ties and the collaborative structure combine to
form a network process, the decisions that are made, and the resulting consequences (as
illustrated in Figure 1.2) were the nexus of this study. The relationship among social ties,
social capital, and the collaborative structure could be observed constructing that
process/decision point through the lens of individual network members. It was the
synthesis of concepts represented in that process/decision/consequences point that was

the basis for the research questions.

Research Questions

This study was conducted to answer two primary questions. The first was: How
does the concept of social capital relate to members’ ability to build, maintain, and access
resources in a collaborative network? To answer this primary question, the following
subquestions were formulated to guide the study:

1. Is there an underlying social structure that determines social ties?

2. Are individual members positioned within the network in predictable
locations?

3. Does the network show differences in access and distribution when the
collaborative work differs?

4. Are there patterns of ties that emerge from an analysis of the network?

5. Does the selection process differ for individuals in the network?

6. Do individuals describe an influence process in the network?

The second primary research question was: How do network members respond to
social capital, network concepts, and the research process? Further, does action result?
In answering this question, I explored the level of knowledge generated by the research—

that is, whether network members increased their representational knowledge

11

(explanation through identifying the relationship between discrete variables and their
meaning), reflective knowledge (creation of a vision of what ought to be through
consciousness raising), and relational knowledge, in which the social interaction among

members generates new relations (Park, 1999; Reason & Bradbury, 1998; Small, 1995).

Assumptions

Before this research study, the phenomenon of the process/decision point in the
collaborative network was a mystery, at worst, or an assumption, at best. I assumed that
if actors in the collaborative network would describe their personal linkages to others and
the network using a social capital framework, those descriptions could be used to explain
the process, the decisions, and the consequences that occur.

Using an overarching human ecological model, I further assumed that community
collaborative network members would welcome the opportunity to participate in such a
discussion, would engage in the study as partners, and would embrace the results as
strategic tools for action. The accuracy of these initial assumptions was verified by the

results of the study, which are presented in Chapter IV.

12

CHAPTER II

REVIEW OF THE LITERATURE

Introduction
This chapter contains a review of the literature that was the foundation for this
research. Four main topics were deemed relevant to the study: collaborative
organizations, social networks, social capital, and participatory research. They are
discussed in this chapter to build a rationale for the research design, which is the focus of

the last section of this chapter.

Cofiorative Organizations

The discussion of collaborative organizations is divided into two parts. First is
the philosophical foundation for collaborative organizations. In this part, I explore why
leaders believe that collaboratives are a necessary vehicle for community change. The
other part of the discussion of collaborative organizations has a practical focus,
presenting information gleaned from the literature about how the vehicle works. In the
first part, it can be seen that collaborative efforts in communities are philosophically
based in an ecological perspective and, as such, can be viewed through beliefs about
survival, adaptation, and interdependency. The second part highlights the dimensions of

collaborative organizations and the elements that produce success.

Philosophical Foundation of Collaboratives
The plight of youths and their families in America resonates from both human
(Benson, 1997; Garbarino, 1995; Catalano & Hawkins, 1996) and community

development leaders (Bruner, 1995; Chaskin, 1992). Both have argued that

13

circumstances for adolescents are complex, are dangerous, and require collective and
innovative change. Leaders call for supports to the inner circles around youths in
communities (Brooks-Gunn, Duncan, Klebanov, Kato & Sealand (1993), a social
structure that includes a network of community supports that Bronfenbrenner (198 6)
called “Microsystems.” Researchers have promoted integrated and collaborative
interventions and interagency work as strategies to achieve successful community
outcomes, including resiliency among adolescents (Bruner & Chavez, 1996; Chavis,
1995; Gittell & Vidal, 1998; Catalano & Hawkins, 1996).

Conditions for youths and families, when measured by indicators of well-being,
are attention getting. Indicators such as people’s having children too soon, leaving
school illiterate and unemployable, and committing violent crimes have been at alarming
levels for decades. Schorr (1988) called indicators like these “rotten outcomes.” She
urged that more attention he paid to American adolescents and that public policy be
changed to support local and innovative interventions.

Collaboration is a manifestation of the human ecological perspective. Survival is
a fundamental ecological principle, considered a core value found through adaptation and
interdependency (Bubolz & Sontag, 1993) between humans and their environment.
Collaboration is a means to apply multiple strategies synergistically across systems to
maximize the community’s adaptive capacity (Goodman, Wandersman, Chinman, Irnrn
& Morrisey, 1993). When leaders promote comprehensive change, they are responding
to the risk of failure for youths, families, and communities and the urgent need for public
policies and programs that adapt to the complexity of the times.

This study of social capital in a community collaborative network was designed

within a human ecological philosophical framework. I questioned the belief system of

14

collaborative members by asking members how they viewed their interdependency. Did
they believe that a collaborative organization is a vehicle for adaptation and survival?
Why build and participate in community collaborative networks? According to
the current literature, one needs to build and participate collaboratively because
community issues are not resolved when one works alone (Brunet, 1998; Gittell & Vidal,

1998; Rubin & Rubin, 2001).

Practical Aspects of Collaboratives

Chaskin (1992) defined a community collaborative as “a group of individuals,
usually representing different backgrounds and fields, who come together and organize to
address a concern regarding child, family or neighborhood well-being that transcends any
one organizational response” (p. 9). These groups vary widely, ranging from groups of
community members who organize around community assets (Mc Knight & Krctzman,
McKnight, 1989; 1993; McKnight, 1997) to those that gather to respond to pubic policy
mandates or funding opportunities (Butterfoss, Goodman & Wandersman, 1996;
Goodman, Wandersman, Chuman, Irnm, & Morrissey, 1996; Mandel], 2001). Yet, in
either case, a collaborative organization is one in which individuals come together to
achieve shared outcomes (Cigler, 1999; Florin, Mitchell, & Mitchell, 1995; Rubin &
Rubin, 2001). In this section, the dimensions of that collaborative’s “coming together”
are reviewed. The aspects of collaboration that defined this study are highlighted, and the
rationale for choosing social capital as the focus of the work is set forth.

Collaborative organizations range along a continuum from loosely formed
partnerships with a narrow focus to more structured and interdependent collaboratives

encompassing broad systems change to accomplish a common policy goal (Mandell,

15

2001, p. 298). Although the formal structure may differ based on the organization’s
position on this continuum, the relational structure has some common elements.

In a review of literature on collaboratives, F oster—Fishman, Berkowitz,
Lounsbury, J acobsen, and Allen (2001) found that four critical levels of capacity exist to
some extent in all collaboratives: member capacity, relational capacity, organizational
capacity, and programmatic capacity. This collaborative capacity refers to the conditions
necessary for coalitions to be effective and build sustainable community change.
Member capacity refers to individual and collective knowledge and attitudes.
Programmatic capacity is the extent to which individual members and the collaborative
can deliver programs. Organizational capacity is the level of identification with the
collaborative and the ability of the group to develop and to deliver a group vision. In
collaboration, an effective organization has broad relational capacity based on members’
having new ways of interacting with current contacts (relationships across participating
members and organizations) and external organizations (Foster-Fishman, et al., 2001).
Change is required in how individuals behave in their work, in the broad socioeconomic
structure, and in the networks in which people function (Chaskin, Brown, Venkatesh, &
Vidal, 2001). In an effective collaborative, members identify with a shared vision that
influences the work they do individually as well as the work of the collective (Bruner &
Chavez, 1996; Cox, 2000; Foster-Fishman et al., 2001).

Collaborative relationships bring with them a redistribution of power. Given that
the group has come together to accomplish something beyond its individual members’
missions, power redistribution is a function of the coming-together process (Wolfe,
2001). Redistribution of power takes the form of sharing information and referrals

(Foster-Fishman et al., 2001), funding (Oliver & Ebers, 1998), and including diverse

16

voices (Bartunek, F oster-F ishman, & Keyes, 1996) and can result in increased
identification with a collaborative and with a collective vision (Chavis, 1995; Gulati &
Gargiulo, 1999; Catalano & Hawkins, 1996).

Michigan’s “Putting It Together for Michigan Families” initiative (Executive
Order, 1994) was promoted to increase collaborative capacity at the local level and to
facilitate the redistribution of power among community service providers. To assess the
success of such an effort, one would need to consider the relationship between an
individual and the collaborative social structure. For that reason, social network analysis

is an appropriate method for assessing the success of such relationships.

Social Networks

A network analysis documents how patterns of ties are used to allocate resources
in a social structure. It is a way of “taking social structure seriously” (Wellman, 1988, p.
20). A network comprises patterns of relationships linking a defined set of persons,
objects, or events. The network is a function of the nature of those ties, in terms of when
actors come together (frequency), why they do so (motivation), and what outcomes or
consequences result (Knoke & Kuklinski, 1982; Mitchell, 1969).

Network ties are the stable, relational patterns that represent an individual’s
position within a network (Oliver & Ebers, 1998; Stevenson & Greenberg, 2000) and the
emergence of subgroups based on the cross-linkages of ties (Wellman, 1988). Subgroups
in a network emerge when an actor is more likely to interact with a particular subgroup of
actors than with any other actors. In a study of an interagency collaborative, Frank and
Yasumoto (1998) found that individuals within the collaborative formed cross-linked
subgroups and were more likely than those not involved in the collaborative to exchange

resources. The researchers also discovered that the ties between members of a subgroup

17

could be distinguished from ties within a subgroup. The within-group ties bonded actors
together with norms and reciprocity, whereas ties between groups bridged actors to
potential resources.

Relational ties can be classified as strong or weak. Strong relational ties are
emotional bonds of friendship, intimacy, and reciprocity; they tend to endure over time,
whereas weak ties are less frequent and less intimate (Granovetter, 1973; Krackhard,
1992). Weak ties, most common among those who are unequal or heterogeneous in their
social identities (Blau, 1994), tend to be more “instrumental” and are more likely to be
threatened by conflict (Ashman, Brown, & Zwick, 1998). Weak ties have been found to
increase the capacity of an individual or a subgroup to elicit new information and
resources that would otherwise not be accessible. The cohesive subgroups of a network
define an actor’s primary affiliation, but he or she can still pursue resources through ties
with other groups (Frank & Yasumoto, 1998; Granovetter, 1973).

In collaborative networks, there are strong and weak ties between subgroups.
Operating in that context, people form associations most easily with those who are most
like themselves (Blau, 1994; Putnam, 1993), whereas those with differing values,
interests or degrees of power often find it difficult to establish the bonds necessary to
build trust (Ashman et al., 1998). In such cases, the use of fiiendships (Uphoff, 1992) or
intermediaries who perform bridging roles (Brown, 1993) can shore up associations or
sustain relationships. The nature of individual relationships between and among
members of a social network determines how those individuals behave within the
network structure. Strong ties provide the stable structure of shared values, trust, and
reciprocity, and weak ties furnish access to unique resources (Frank & Yasumoto, 1998;

Granovetter, 1973).

18

An actor’s behavior within a network might also be influenced by his or her
position in the social structure, akin to the similarity among actors in a bounded network.
Wasserman and Faust (1994) defined structural equivalence as a mathematical property
of subsets of actors in a network, where “two actors are structurally equivalent if they
have identical ties to and from all other actors in the network” (p. 356). In his study of
social exchange, Bearman (1997) applied the mathematical property of subsets to
partition the population into blocks of structurally equivalent actors. This “block
modeling” process was used to partition the network into blocks of actors based on
similar patterns of relations. This means that actors have homogeneous internal relations
and homogeneous external relations with others, all relations being considered
simultaneously (Wasserman & Faust, 1994). Thus, exchange with others is based on
structural equivalence, whereby actors are interchangeable in terms of social position,
roles, or obligations. In community collaborative networks, this might be seen in the
similarity of organizational positions; i.e., all directors or all middle managers or those in
similar functions, such as all funding agencies or all advocacy groups, which could
influence one actor to have a structural predisposition toward another.

Another determinant of an actor’s behavior within a network is the position the
actor occupies. Stevenson and Greenberg (2000) found that actors can be enabled by
their network position or constrained by it. Those individuals who are central to the
network are powerful in their capacity to influence a number of relational ties directly
(and to allocate corresponding resources), whereas those on the periphery need brokers
within the network to exert influence. There is, however, power in the periphery. Burt
(2001) found that connections carry obligations and that there are strategic advantages to

not incurring those obligations. Although a central position likely brings an actor access

19

to information and resources, actors on the outside tend to be specialists, or to possess
unique advantages in specific areas and can leave those in a central position reacting to
innovations on the periphery (Stevenson & Greenberg, 2000). Burt called gaps between
actors or subgroups in a network “structural holes” and concluded that peripheral actors
find an advantage to filling such holes in a network. According to Gittell and Vidal
(1998), occupying a peripheral position gives an entrepreneurial advantage to those on
the outside.

Actor position is one factor in how a network functions to access and distribute
resources. Two types of exchanges are direct or generalized (indirect). These types of
exchanges operate in the exchange of goods or in a social exchange in which the
exchange results in acquisition of social rewards, as in quid quo pro exchanges (Bearman,
1997). In direct exchange, there is an exchange of some equivalence in perceived values.
A gift is given that induces direct reciprocation. This could be an exchange of identical
items, as in “You speak to me and I will speak to you,” or items of equivalent value, as in
“You speak tome and I will open the door for you.” Another example would be an
individual’s acting in a certain way because doing so gives him or her a sense of warmth,
or regard.

The second type of exchange, generalized exchange, rests in the norm of
reciprocity in that “the takers are obliged to be givers” (Bearman, 1997, p. 1390). In this
type of exchange, for example, an individual will give preferential access or attention to
another based on the norm that such would be afforded to him in some form at some
time. Whereas generalized exchange is a force in moving individual actions to be
reallocated into a network of actors, it breaks down if the chain of exchange is broken

(Bearman, 1997). In a community network, for example, a generalized exchange

20

between actors A and C would be supported by direct ties between A and C, a direct
chain of exchange, and not through indirect ties between actors A and C through a third
actor, B. Generalized exchange, then, is most successful in social systems with few
external opportunities for the exchange to be broken, where the system can support the
integrity of the direct ties (Frank & Zhao, 2001).

The ties between and among individuals in a community collaborative network
define position in a network, the presence of cohesive groups, and access to potential
resources through a method of exchange. Thus, ties influence both the individual actors
and the network of actors within a social structure. To successfully study these ties, one
must view them at both the micro (individual) and the macro (network) levels. In this

research, the concept of social capital was used to accommodate such a view.

Social Capital

Social capital is the value that is inherent in the social ties of a network (Cohen &
Prusak, 2001), “the aggregate of the actual or potential resources which are linked to
possession of a durable network” (Bourdieu, 1985, p. 248). Social capital facilitates
coordination and cooperation for mutual benefit (Putnam, 1993, p. 36).

An individual draws on social capital to access necessary resources; researchers
have linked this concept to successful, collective outcomes for adolescents. In a study of
teen pregnancy rates among Latina youths, it was found that a high percentage of Latino
residents, strong social networks, cultural norms, and social capital was correlated to low
rates of births to teens (Denner, Kirby, Coyle, & Brindis, 2001).

Sampson, Morenoff, and Felton (1999) highlighted three dimensions of
neighborhood social organization that affect the lives of children. They contended that

many efforts to operationalize social capital miss Coleman’s view that social capital is

21

lodged not in individuals but in the structure of social organizations. Their results
showed that residential stability and concentrated affluence, more so than poverty and
racial/ethnic composition, predict intergenerational closure and reciprocal exchange. In
other words, a stable neighborhood and a focused point of resources resulted in increased
links between adults and children (intergenerational closure) and interaction between
families and adults with respect to childrearing. Similarly, in a longitudinal study of at-
risk youths, Furstenburg and Hughes (1995) found that extended family, church
connections, and quality of relationships were significant in the lives of youths who had
successful outcomes. Social capital was found in the “aspects of social structure used to
access the resources to achieve their interests” (Furstenberg & Hughes, 1995, p. 581).

At the network level, social capital has been promoted as the relational
infrastructure that represents a community’s capacity to act on common concerns
(Temkin & Rohe, 1996) and its capacity to mobilize stakeholders (Bruner, 1998).
Working with more than three dozen community foundations, Putnam (2002)
administered the Social Capital Community Benchmark Survey to assess the strength of a
community network, based on the quality of civic ties. The survey measured dimensions
of civic life, including social trust, political participation, civic leadership, and
associational involvement. The survey also included a measure of giving and
volunteering, a measure of faith-based engagement, a measure of the diversity of
friendships, and a measure of the equality of civic engagement at the community level.
Aggregate results indicated that ethnic diversity in a community was a strong predictor
for a number of these dimensions. Ethnically diverse communities reported a higher
density of neighborhood, ethnic, or self-help groups, and in these diverse communities

residents were more likely to report friendships with people of color and gays. These

22

residents also reported more tolerance; the greater the community’s ethnic diversity, the
less likely its residents were to say that books should be banned from a public library
because most people do not approve of them. The study further indicated that the more
diverse a community, the less likely its residents were to trust each other, including those
in their own ethnic group, and the less likely they were to connect with each other.
Residents of more diverse communities were more likely to be personally isolated, to
claim fewer friends, and to have less of a sense of community. Community foundations
involved in the study committed to funding programs that would build levels of
connectedness and strengthen social ties within their community by giving residents the
opportunity and the space to connect.

Putnam and the community foundations participating in the Social Capital
Community Benchmark Survey were not alone in their focus on proximity and
opportunity as critical conditions for social capital construction. Paxton (2002) referred
to them as the association between individuals who are tied within a social space. After
conducting a series of organizational studies, Cohen and Prusak (2001) concluded that
regular patterns of association are critical for the development of the trusting, reciprocal
ties that hold social capital. Such conditions also have been found to exist in housing
coalitions (Gittell & Vidal, 1998) and human service agencies (Foster-Fishman et al.,
2001).

In a study of computer innovation in schools, Frank and Zhao (2000) found
evidence that social capital can be built within a social space such as a school. Whereas
traditional ties tended to fall within grade or curriculum groups, new collegial ties
between teachers emerged informally through “technology talk” across school subgroups.

This supported the informal movement of resources between groups and the potential to

23

form shared values about the technology, and subsequently built new social capital within
the school network.

Understanding social capital, how it is built, and its relationship to collaborative
networks is somewhat problematic because it is relevant on both the individual and the
network levels (Bourdieu, 1985; Coleman, 1990). In a community network, for example,
an individual actor’s capacity to access actual or potential resources through social ties
(i.e., information about funding, assistance with a client) is social capital. The resources,
as in the allocation of funding or the coordination of services, rest in the network and are
based on collective norms, trust, and resulting action. The question is, Under what
conditions does the individual level make the transition to the network level?

Frank (2002) used the metaphor of a rotating system to represent aspects of social
capital in a dynamic model. He conceptualized a system with a set of actors and ties
among them, where the system has mass based on the number of actors and density based
on the number of ties among actors. The ties represent that the actors are confined in a
system with each other (as in teachers in a school or members of a collaborative) and
must exchange with one another (see Frank, 2001). The more equal the interests of each
actor in the exchange of social and nonsocial rewards, the greater the capacity for
exchange. When an exchange of social or nonsocial resources occurs across the ties of
two actors, the system rotates and gains momentum, positioning actors for further
exchange. The force of exchanges fuels rotation of the system and moves the social tie
between two individuals together, merging into collective interests and a spiraling
identification with a collective. When applied to a collaborative network, this theory
suggests that, when members of such a network define themselves as confined to a

system and exchange resources, two things occur. First, the distance between the

24

interests of individual social ties will compress toward collective interests as exchange
across social ties gains momentum and force, and second, individual exchanges will be
transformed into network resources because of the dynamic nature of social capital.

Coleman (as cited in Frank & Zhao, 2001) saw norms as the form of social capital
that transforms individual, direct exchange of resources to the network level.

Individuals’ actions “cause norms, which in turn cause sanctions and conformity” (pp. 4-
5), providing momentum to a system of exchange. Actors allocate resources based on a
level of certainty that their interests will compensated, based on the strength of the norms
as evidenced by the accumulated experience of resource allocation.

The dynamic nature of social capital lends itself to a research model that is
equally dynamic, where the actual and potential resources for the work are accessed
through social ties, and resources that result from the work are distributed and owned by
the network. Participatory action research was selected for this study because it is such a

dynamic method.

Participatory Action Research

Participatory action research has two goals: to make both a scholarly and a
practical contribution by including research, education, and action (Bartunek, 1993;
Small, 1995). It includes the following characteristics of research articulated by Reason
and Bradbury (1998):

l. Produces practical knowledge that is useful.

2. Increases well-being.

3. Involves all stakeholders in both the questioning and the sense-making that
inform the research and in the action that is its focus.

4. Allows the research to evolve and develop over time.

25

5. Creates new abilities and new knowledge over time.
The research design described in Chapter HI integrated qualitative and
quantitative measures into the participatory approach. Results were produced that can

inform both theory and practice.

26

CHAPTER III

RESEARCH METHODOLOGY

Introduction
This chapter relates the objectives and questions of this study overall research
framework that was used to structure the research design. The design is then applied to
the role of the researcher and the research context. Then the design components are
reviewed, including the research tools, the sample and population, and the data-
collection and analysis methods. The conceptual and operational definitions are related
to the study tools. Finally, issues of research validity and limitations of the study are

addressed.

Research Objectives and Ouestions

In this research I examined the relationships between and among members of a
community collaborative network. The study had two purposes. The first was to learn
how members build, maintain, and access resources in a collaborative network and the
relevance of applying social capital as a concept to explain the relationships. The
questions were: Do relationships influence access to and distribution of resources? If so,
in what way?

The second purpose was to pilot the use of quantitative network analysis
techniques and qualitative interviews and observations to inform participants and to
encourage their action. Questions underlying this purpose were: How do members of the
collaborative respond to information about relational patterns? Do members

(participants) apply the results as a development strategy in their community work?

27

_P_articipatory Framework

The research goals dictated the selection of an action-oriented framework for the
study of social capital in a community collaborative network. This was a study of a
community social process and, as such, focused on how network members build,
maintain, and access resources within the collaborative network. Thus, the work needed
to be done within the community context and involve real network members’ resource
needs. To capture the social process, the research process needed to emerge dynamically
from the relationships and the community ties through reflection. Action models
accommodate all of these requirements as they are carried out in a field setting of the
phenomenon being studied (Creswell, 1994; Miles & Huberrnan, 1994), with a focus on
the felt needs of the participants (Park, 1999). Action-oriented research is concerned
primarily with process rather than outcome (Creswell, 1994) in mind. It is a dynamic
framework that allows definitions to flow from the relationships among the researcher,
the participants, and the emerging knowledge that is generated by the inquiry process
itself (Marshall & Rossman, 1995; Reason & Bradbury, 1998; Tesch, 1990). In addition,
an action orientation accommodated the overarching ecological dimensions that
supported this research, the interrelationship between humans and their environment, and
the need to conduct research that has both practical and scholarly implications (Bubolz &
Sontag, 1993). Action-oriented research is contextually defined and is conducted to solve
practical problems as well as to generate new knowledge (Bartunek, 1996; Miles &
Huberrnan, 1994; Small, 1995).

There are, however, some variations in the use of the terms action oriented and
participatory action research. Miles and Huberrnan (1994) used the term collaborative

social research, Reason and Bradbury (1998) used the term action research, and Park

28

(1999) called this participatory research. In a review of action-oriented research, Small

(1995) suggested that three aspects determine a research model: the agenda, the

epistemology, and the method. He then used those aspects to distinguish between action-

oriented models. Small’s models of action-oriented research across those aspects are

summarized in Table 3.1

Table 3.1: Models of action-oriented research.

 

 

 

 

 

 

 

 

of research process

 

Agenda Epistemology Method

Action Contribute to Generating critical Methods vary but
Research practical concerns knowledge are collaborative

and social science and focused on

practical problems

Participatory Combination of Both researcher and Participants are the
Research research, participant source of primary

education and critical knowledge implementers-

action Methods user

fi'iendly

Empowerment Empowerment Generating critical Ethics influence
Research (where people knowledge by a methods-consistent

gain mastery 0f focus of with

thEiT affairs) participants empowerment

“voice”

Feminist Advocates social, Focus on Methods that allow
Research political and reflexivity in the for a reflective

economic equality critical process process between

for women and Emphasis on researcher and

men emotional aspects participants

 

Source: Adapted from Small (1995).

29

 

The primary goal of action-oriented research, according to Small, is to “[address]
who are the beneficiaries of the wor ”; this is called the research agenda. The
epistemology aspect concerns what forms of knowledge are considered scientific and
method, the most commonly used data-collection strategies (Small, 1995). When
compared across the four models, the agendas move progressively from a development
agenda (action research) to a political-change agenda (feminist research).

The epistemology aspect also moves progressively in the Small model as the
knowledge generated increases the reflective capacity of the participants and the
researcher in the “knowing process.” In the action model, the inquiry informs immediate
concerns, helping participants to change their circumstances based on the outcomes. The
action model is based in representational knowledge, which separates the knower and the
known (Park, 1999) and provides explanation through discrete variables and
understanding through interpretive meaning (Reason & Bradbury, 1998). As the research
models smnmarized in Table 3.1 increase the reflective role of the participants and the
researcher, they take on a consciousness-raising component. The relationship between
the researcher and the participants can be summarized as one that produces a worldview
that is “co-authored” (Reason & Bradbury, 1998, p. 7).

Knowledge includes community ties, critical awareness, and objective
understandings of reality. The inquiry process is more than collaborative in nature
(action research); it is transforming and forms a basis for mobilization around action
(Park, 1999; Tesch, 1990). This “relational knowledge” (Park, 1999, p.146) is generated
by the social interaction and makes the relationships developed in the research process an
outcome of the work. The scope of this type of inquiry (participatory, empowerment, and

feminist research) creates a “reflection-action-reflection” process in which the research

30

cycle is mediated by the production of needed knowledge (Bartunek, 1993; Freire, 1970;
Park, 1999; Small, 1995). Knowledge unfolds as participants integrate a new worldview
through reflection, act on the knowledge, and then reflect again.

The four models of action-oriented research use methods that are consistent with
the ethics (or politics) of the research model. A feminist model of research, for example,
would employ methods consistent with feminist ethics and include techniques that
measure intuitive as well as attitudinal data. All methods are at least collaboratively
designed, in that the relationships between variables are not predetermined or developed
outside of the research participants or the research context. Action-oriented research is,
therefore, a dynamic process. In Small’s models, the roles of participant and researcher
articulate the method. The collaborative roles of researcher and participants in action
research allow the research method to be designed by the researcher with participants
responding to the proposed design based on their context.

Case studies are often used in action research (Small, 1995). In the participative
model, participants conduct the research, so the researcher works in partnership with
them to design the relevant method. As such, both qualitative and quantitative methods
can be used to generate knowledge for critical reflection. Quantitative methods, in this
case, cannot increase the distance between the participant-researcher partners and thus
need to be defined (i.e., identification of variables, parameters, sample selection) by all
partners. Participatory research can also be conducted using a participant-observer
method to facilitate the involvement of participants in design and implementation. An
empowerment method must expand participants’ capacity to control their world; thus, it
requires that the participants’ voice supersedes that of the researcher. In this model of

action-oriented research, there is a need to ask research questions and employ methods

31

that help participants recognize their strengths and resources and gain mastery over the
forces that affect their lives (Small, 1995). In feminist research, the roles of participants
and researcher converge in a dual reflective process. The present study lends itself to
qualitative, interpretive research methods because it is personal and interactive and tends
to be consistent with a feminist approach (Small, 1995).

The study of social capital in a community collaborative network fit the
participatory model. The study had a two-point agenda: (a) understanding whether (and
how) the concept of social capital relates to a member’s ability to build, maintain, and
access resources in a collaborative network and (b) discerning how network members
respond to social capital, network concepts, and the research process and whether their
work results in action. This was a participatory model as the agenda included research,
education, and action.

The epistemological aspect of this study was also participatory. Both the
participants and I as the researcher were needed as sources of critical knowledge if the
research questions were to be answered adequately. The study context was a community
collaborative network and needed to be defined by the collaborative members. A social
network analysis, which I provided, was then needed to display the network, as defined,
to the participants for further study. Knowledge emerged from the relationship between
the researcher and the participants; both participants and researcher generated knowledge
from the inquiry process.

Both qualitative and quantitative methods were used in this network study.
Kliquefinder, a network analytic software tool, was used to assess the network structure.
This tool was developed to measure whether and what types of subgroups emerge from

exchanges between members (Frank, 1995; Frank & Yasumoto, 1998; Frank & Zhao,

32

2001). Subgroups emerge when actors are more likely to interact with a particular subset
of other actors than with any other actors in a network (this method was used by Foster-
F ishman et al., 2001). The sociograms that are developed using Kliquefinder show the
ties between participants as well as subgroups.

A number of qualitative methods, namely, interviews with key informants,
participant observation, and archival collection was also used. All of these methods are
used in social network analysis (Miles & Huberman, 1994; Wasserman & Faust, 1994).

A team, with stakeholder and researchers as members, made the design decisions
for this study. The group reaffirmed the community needs (questions to be answered),
and the variables to be included (parameters of the network, nature of relationships) in an
understandable design that the stakeholders could implement. This went beyond a joint
insider/outsider methodology in which community members work collaboratively with a
research team on the study design (Bartunek & Louis, 1996). In this study, the
stakeholders were members of the research team and decided process, interpreted
meaning, and determined results. The stakeholders’ reactions to the network displays
determined the protocol for key informant interviews that were open-ended and used to
document the informants’ insights into the network display. The interviews both verified
the network display and provided explanations for various dimensions of the network.
The interviews and the various stakeholder meetings served to facilitate the development
of relational ties and knowledge. I presented the results of the network displays and the
interviews to the stakeholder group, and then incorporated their reactions and reflections
into the study.

A participatory research model is strengthened by the use of participant

observation because a participant observer has access to events and activities that an

33

outsider would likely not be able to observe. Participant observation is a data-collection
technique that requires the researcher to be present at, to be involved in, and to record the
routine daily activities with people in the field setting. It assists in forming an intuition
about how people are organized, how they relate to one another, and how social and
physical boundaries might relate (Schensul, Schensul, & LeCompte, 1999). That insight
informs the dynamic nature of participatory research, in that the participant observer’s
insight can frame questions and uncover potential themes from a group reflective process
(stakeholder meetings, for example) for further exploration and study by the group.
Because a participant observer is present in the field over time, she or he can discern
patterns that connect various elements in their complexity based on the actions, feelings,
and reflections of the participants. The participant observer comes to understand the
pe0p1e with whom she or he works through engagement with them (LeCompte, Schensul,
Weeks, & Singer, 1999) and thus is in a perfect position for co-authoring action-oriented
research.

If the participant observer is, in fact, a member of the participant group, the
individual is an “insider” to the phenomenon under study. In participatory research, this
means that the insider is not only an actor in the study process but also a player in any
actions that result from the process. A traditional participant observer engages with
participants in their context to understand relational ties in their daily lives. Insider
engagement, in contrast, results in a view of not just the lives of others but of his or her
own life as well. The insider is not observing the lives of others with whom she or he has
developed relationships within the context of the research. Rather, the insider is one of
these people and has relationships with at least some of them (required by one’s inclusion

“inside” the group).

34

The context of the research could be seen as the context of the insider’s life and
the lives of those with whom she or he may or may not engage in normal day-to-day life
(see LeCompte et al., 1999). An inside researcher can be a member of the group under
study and is uniquely charged with the responsibility of facilitating the research and
acting within it. The role of researcher as insider, participant observer brings with it risks
of bias and other ethical considerations that must be accommodated in the research

design.

Role of the Researcher
Three primary risks are involved in the role of an insider, participant observer.
They are personal risks to the insider and can be reduced by a clear, participant-driven
research model.

The first risk is what can be called influence bias. As a member of the group
under study, the insider has a particular political, social, and economic position in that
group. That position may be apparent to group members (as in formal relationships or
mandated positions) or not apparent (embedded in affiliations or personal experiences).
Apparent or not, the power or relational ties of the insider could influence the research in
terms of who is willing to be involved and to what degree. Politics might influence some
actors to “need to be involved” or might have the opposite effect. A strong participatory
model can reduce this risk because it requires that all research team members (researchers
and participants) design the process and that all participants implement the research
methods. Given the process focus of action-oriented research, dealing with the risk of
influence bias due to the insider’s position itself can be a source of data. That process

could yield insight into the group’s relational ties and the dynamic nature of the work.

35

The insider also might risk her or his position in the group. That is the second
risk of an insider role. To insure that all participants have access to the process, the
insider might need to use her or his existing relationships to facilitate access. The person
might need to use her or his connections and credibility with people to get them involved.
The risk is that the insider would be held responsible at some level for the credibility of
the research, either eroding or shoring up one’s personal credibility. The recruitment of
participants or a study site can be successful when inside researchers use a group in
which they are members (Bartunek et al., 1996). The personal risk can be reduced by
adding an outside researcher to the research team who can assist with the research design,
data collection, and data analysis (Louis & Bartunek, 1992). The risk would also be
reduced with an early group affirmation of the tenets of participatory research; the
participants define the work and own the outcome, and the process is an outcome of the
effort.

Participatory research is a political process (Park, 1999; Small, 1995). Because of
its inclusionary nature, all participants have a voice in definitions, in design, in data
collection and analysis, and in the actions that result. The third risk is that either the
process or the results of the process change the power structure of which the insider is a
part and that such change could adversely affect the insider’s situation. This risk is
inherent in participatory research; in fact, it is really the point of the work. It is designed
to reframe the conditions in people’s lives through their individual and collective
reflections on the research, on the learning that occurs, and on the action that results. By
its nature, participatory research redistributes power; thus, for the insider as well as for all

participants, the risk of change cannot be reduced through attempts to control outcomes.

36

Such efforts would only serve to abort the process; hence, they would be not only self-
defeating but unethical.

Manipulating research outcomes is an ethical violation that could occur in an
insider design. It could be done, as described above, in a deliberate attempt to control the
outcomes and protect someone or something. It also could be done through sloppy
methodology. An insider has to have a clear process to gather her or his own
knowledgeurepresentational, relational, and reflective (Park, 1999). The process has to
be systematic so as to ensure that the insider’s opinion is not given more weight than that
of other participants. This can be accomplished through the use of personal journals and
reflective writings (Schensul et al., 1999). In addition, the participatory process should
include a clear stakeholder-driven decision process so that participants can respond to
interpretations of the insider and the group can collectively determine actions.

The researcher risks identified above are all possible concerns for members of the
group under study with an insider, participant observer. Because there can be influence,
positional, or power implications for participants, it is critical in participatory research
that the process be structured to respect the equality and the potential vulnerability of all
group members. Inidesigning the research, all participants (including researchers) need
to speak frankly about the methods, the data collection, and the results, as well the issue
of confidentiality. In addition, all participants need access to the skills necessary to
engage in the work. If a survey is constructed, for example, all people involved need to
be comfortable with the method and the data-analysis process. In an insider/outsider
design, the outside researcher can serve to facilitate safety for participants by, for
example, being responsible for coding of instruments or being available to conduct

interviews with participants who would prefer it.

37

To successfully conduct insider participant observation, the study must be
grounded in two threshold stands. One is that the insider research role is different from
the individual’s role as a member of the group. The researcher must hold an identity that
is not fully coterrninous with that of the individuals who are members of the group under
study (Schensul et al., 1999). The perspective of a participant observer is that of the
members of the group being studied (Small, 1995). The process “demands that the
researcher learn how it is to ‘not be me’ in a physical setting that is ‘not mine’”
(Schensul, et al, 1999, p. 71). For the insider this requires continuous framing and
reframing. This threshold position needs to be clearly articulated to the group under
study. The insider’s boundaries for research must be expressed and agreed to by the
participatory research team. Without a definition of the insider role, the participants
might be unaware of when, in living and doing business every day, the insider is a group
member and when she or he is a researcher.

The second threshold strand is that the purpose of the research is genuinely a need
of the group to be studied, determined by a decision process that is locally defined. In
other words, research participation needs to be decided by whatever method the group
normally uses to make decisions. Participatory research is implemented through its
stakeholders (Park, 1999). It is action oriented (Small, 1995) and relevant to community
needs (Creswell, 1994; Reason & Bradbury, 1998). It can be successful only if the
questions being asked resonate with the group. For an insider, this is ethically
problematic. From the beginning of the process, there needs to be clear stakeholder
involvement with the value of the research to the group being studied.

Participatory research is similar to most grounded-theory models and has as its

goal the generation of new understanding in a local context (Geertz, 2000; Park, 1999).

38

This is an important point to include in a stakeholder discussion. It can be time
consuming for participants to define the research, to implement the components, and to
be the source of much of the data (LeCompte et al., 1999; Marshall & Rossman, 1995).
The work is defined in the local context, and the group’s perceptions of the world, their
ties to each other, and the interactions involved in their everyday lives is the work.
Action-oriented research takes time, but if the work answers real questions about
real life, it can be worth it (Reason & Bradbury, 1998). To study social capital in a
community collaborative network, the first step is to build a structure that will

successfillly frame the work.

Research Design

The design of this study of social capital in a community collaborative network is
illustrated in Figure 3.1. There were nine points of the research design, and the
illustration shows how each stakeholder meeting informed the research tools, which in
turn informed the next stakeholder meeting in the process. The design included meetings
in which all stakeholders, participants, and both the inside and outside researchers came
together. In the first series of meetings (Point 1), the group (a) defined the research
parameters (those present would be the network of study), the overarching concepts
(networks, social capital, and participatory research) and (b) learned about potential tools.
They decided on a sociogram-producing survey process that would be grounded, verified,
and explained through a key informant process. At that meeting, the stakeholders also
defined the networks to be included in the study, which were the basis for the survey tool.
The survey tool, the Community Delivery Survey (Point 2), was administered to the

stakeholders and then processed using the Kliquefinder technique to create five

39

sociograms, which were then taken to the stakeholders (Point 3) for interpretation,

reflection, and sharing of knowledge.

 

 

 

Social Capital
in a Community Collaborative Network
Research Design
Stakeholder Meetings 3318913019“
Parameters] 9. ”“898
DefixfifionsJQuesfions ‘ Informed Actions
1. a.

 

 

 

 

 

 

  
   
     

 
       
   
   
 

“Survey

  
   

Analysis

 

 

of .
C ollab oration” 01"
Legend _ . .
Sociogram Postion
Analysis Use
Research 2. of ties
Tools .
Stakeholder {1
Stakeholder M°°538
E] Process Interpret
Sociogram
—5 Flow of Questions 3
Information ‘ 6.

 

 

 

 

 

 
     
   
   
   

 

  

Key
Stakeholder

 

 

 

Intani ewal
Snowball Stakeholder
4_ Meeting
ReactionstRefle ction
Questions
5.

 

 

 

Figure 3.1: Social capital in a community collaborative network research design.

The stakeholder meeting process (Figure 3.1, Points 1, 3, and 5) was designed to
generate three types of knowledge: representational, relational, and reflective. This is
where the inside and outside researcher and the key stakeholders came together and

discussed the relationship between variables and the meaning of their work and structure.

40

This discussion created representational knowledge, an explanation of what relationships
mean, and how stakeholders’ relationships to each other and their work formed the
foundation of their collaborative effort. Stakeholders talked about their own personal and
organizational ties and, in so doing, generated new relational knowledge about those ties.
Together they also considered how such ties and variables existed in their real lives to
collectively come to a new understanding about how such relationships “ought to be”
(Park, 1999; Reason & Bradbury, 1998). At each of these points, the design required
stakeholders to reflect on the themes emerging from the study. The stakeholder meetings
were the platform for participants’ implementation of this research model.

The questions that surfaced in the stakeholder meeting (Point 1) were the basis for
constructing the Community Delivery Survey. The questions that emerged in the second
meeting (Point 3) were the basis for questions included in the key stakeholder interviews
(Point 4). In addition, the stakeholders decided to expand the survey to include a
snowball sample of those identified in the first survey. Results of the interviews and
snowball sample (in the form of new sociograms) were reported to the stakeholders for
critical reflection in the third meeting (Point 6). This reflection led to new questions and
brought about the need to analyze the interview results in the context of the sociogram,
network position of the respondents, as a method to compare and contrast (Point 7).
Results of the interview-sociogram comparison were presented to the stakeholders (Point
8) for interpretation and potential action. The research design provided for the sustained
reflection-action-learning process as the actions of the stakeholders were used to inform
the network parameters and could, therefore, restart the process.

The participatory nature of this research required the flexibility illustrated in

Figure 3.1. This was a dynamic process and required a dynamic design as well as a

41

dynamic setting. The context for this study needed to be a group that would engage not
only in the research, but also in the education required to implement it and the action that
would result. The group needed to be able to adjust to the emerging demands of inquiry,
such as having the time to reflect on findings, to analyze relationships, and to develop a
shared understanding. In addition, it needed to be a group that would see itself in the
knowledge-generating business and would take time to learn together. It also would need
to be a group that considered process to be an outcome and one that could be assessed for
maximum learning about relationships from a participant observer. Such a context for
the study was found in a collaborative group, organized at the onset of the Michigan
“Putting It Together for Michigan Families” initiatives and now, after five years of
operations, wondering how relationships between and among members influenced

resource allocation.

Research Context
The research had two components. One was the development of the community
collaborative itself, the Family Coordinating Council (FCC), which was a local response
to state and local policy and funding decisions. The other was the collaborative’s
decision to study its relationship structure. Both dimensions are presented here in
historical sequence to document the participatory, dynamic history of the council and the

research.

Background of the Family Coordinating Council
In 1995, the FCC was organized for two reasons. One, it was a response to the
Michigan “PIT” initiative, which was introduced with a funding opportunity. Up to

$30,000 was made available to community collaborative groups to be used for planning

42

across public and private service agencies. The money was to be coordinated by the
Family Independence Agency (F IA) in most of Michigan’s 83 counties. In a particular
county in Michigan, there was a second reason for organizing the Council. A private
donor to the local community foundation had approached the foundation president about
a designated gift. The donor was interested in assessing conditions for children and
families in the county. He wanted to know whether the money being spent to help
families was actually making a difference (Children’s Agenda Report, 1998).

Through a partnership between the foundation president and the F IA director, the
two planning initiatives were combined. The state funding required that certain key
stakeholders sign as collaborators for the planning grant. Those organizations that were
involved in the collaboration at its inception have remained involved, and 22 additional

members have joined in the last seven years (Figure 3.2).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Organizational Evolution of the Family Coordinating Council
1994@ginal Members 2002 Membership
Community Foundation "‘ Three Neighborhood Groups
F amibl Independence Agency Public Health
Juvenile Court Two Public Health Clinics
Intermediate School District Two Hospitals
Mental Health Five Human Service Agencies
Prosecutor University Extension
Sheriff United Way
Domestic Violence Agency Six youth development agencies
Foster Care/Adoption Agency Senior Agency
Juvenile Day Treatment Age—ricy
10 Oflanizations + 22 Orgnizations
2002 Total Council Members 32
* The Foundation was not a state required
signature in the planning grant

 

 

 

 

 

Figure 3.2: Organizational evolution of the Family Coordinating Council.

43

The original stakeholders included the major public agencies that served children
in the community, the FIA, the Juvenile Court, the Intermediate School District, and
Mental Health. It also included the community foundation that allocated approximately
half a million dollars for services to families and the highest elected law enforcement and
criminal justice officers--the sheriff and the prosecutor. In addition, the original members
included the directors of three private agencies that served children and families. This
was the first collaborative effort for that particular group of agencies or individuals (FCC
archives, 1995). The original members were the directors or presidents of eight large
community institutions.

The original collaborative received the community foundation donation and state
planning money to conduct a community assessment of conditions for children and
families in the county. Using demographic and descriptive data indicators, the group
identified those topic areas in which significant issues seemed apparent (organization
archives, FCC). Those issues were grounded through a series of 13 focus groups
conducted with a wide range of community stakeholders (Figure 3.3). The original
collaborative members conducted or attended all of the focus groups and subsequently
summarized the results to determine the planning focus for the FCC. In addition, the
group invited the focus group participants to join the community efforts, took names, and

later recruited members of critical-isSue groups from the lists.

Chambers of Commerce Agency directors

F oster-care parents Teachers and administrators
Police and judges Public-housing residents

F aith—based leaders Healthcare providers
Adolescents Town meetings (3)

Grandparents Raising Grandchildren

Figure 3.3: Focus groups involved in the community assessment process, 1995.

44

With information gathered in the focus group process, the FCC was organized
into collaborative groups that were designed to work together and influence critical
issues. Overtime, four collaboratives emerged: teen sexuality, violence, effective
parenting, and substance abuse.

Organization of the FCC took most of 1995. The original members of the Council
agreed on a two-part mission. They committed to countywide, comprehensive planning
around children and families. They also agreed to include in that planning a goal to build
the community-based infrastructure in the county. From 1996 to date, the Council has
allocated resources to community organizations and interagency initiatives.

As shown in Figure 3.2, the Council expanded significantly, with the core,
original leadership staying active and engaged. Of the original 10 members, eight were
still actively involved with the Council. They attended meetings, took on responsibilities,
and tended to be the chairs of the collaboratives or ad hoc groups (FCC minutes, 1996-
2001). During 2000/2001, however, two of the eight active members changed jobs and
two retired. At the same time, new leadership was emerging and, with the changes, came

an interest in assessing the Council’s relationships.

Background on the Social Network Study:

In the spring of 2001, the Council decided to hold a two-day retreat to discuss and
assess the structure of the Council and to recommit resources to the future. A
professional with expertise in organizational development facilitated the retreat. During
that retreat, a number of issues surfaced concerning the way the organization made
decisions and allocated resources. Some of the newer Council members expressed
confusion about how the funding decisions were made. Some also said that they

sometimes felt disconnected from the meeting agendas and the work that was being done.

45

Other members strongly believed that the organizational culture was defined in the
original mission. They generally agreed that the planning work and commitment to
building community-based infrastructure were still relevant for the Council. The group
decided that, given the differing opinions among members, the Council needed to assess
the relationships between members and the members’ involvement with the current
collaboratives and critical issues (Retreat minutes and breakout results, 2001). Staff were
assigned responsibility for recommending an assessment process.

The FCC director had an association with social capital, social network
researchers at Michigan State University. The methods used there had recently been
applied to an assessment of cross-organizational relationships in a community
collaborative structure. The initial design was developed and accepted by the Council.
An existing standing committee of the Council agreed to serve as part of the research
team as the stakeholder group for the study. The team selected the research model,

including the sample, the research tool variables, and the data sources.

The Research Population

A social network consists of a finite set or sets of actors and the relationship or
relationships defined by them. The collection of ties among the members of a group is
called a relationship and it is the relational information that defines the boundary of the
network (Frank, 1995; LeCompte et al., 1999; Wasserman & Faust, 1994). In this study
of social capital in a community collaborative network, the potential network was the
FCC membership; as such, it could have served as the research population.

In a participatory research design, however, the study group must be actively

engaged in the work. The whole FCC is a large group (see Figure 3.1), which might have

46

been problematic. The research team (including the stakeholder group) decided to use
the stakeholder group as the network group of study. They defined the variables and the

direction of the research tools.

Data Sources and Research Tools
Data were collected in this study using a variety of methods: (a) an archival
review to provide context, (b) insider participant observation of stakeholder meetings, (c)
the Community Delivery Survey, and (d) key informant interviews. The relationship
among these sources and tools is illustrated in Figure 3.4. Each source flowed from the
one that preceded it because data were recorded, coded, and analyzed along the way. The

following is a description of each data source.

Archival Data

Stakeholder
Meetings

Stakeholder

, Research Tools
Intervrews

and

Data Sources

 

Community
Survey

Figure 3.4: Relationship among research tools and data sources.

47

Archival Data

Archival data were obtained through a systematic review of existing FCC
documents, including minutes of all meetings, publications, and reports. With rare
exceptions, the Council has met monthly since its inception. Attendance and
participation information for individual members and organizations were tracked to
provide a context for the relationships that formed the network. The various initiatives of

the Council were also noted to describe the context in greater detail.

Stakeholder Meetings

The participant observer took detailed field notes during all stakeholder meetings,
noting both key comments and actions. In addition, a tape recorder was used at those
meetings at which the inside researcher had facilitative responsibilities. The
transcriptions were coded, as were the field notes, to reveal emerging themes and further

research direction.

Communi Delive Surve A endix A

Based on previous social network research, the inside and outside researchers
hypothesized that enduring social ties could be captured through a survey in which
members would be asked to list the individuals who were important to them in doing
their jobs. This technique was used by Frank (1996) to study relationships between
French bankers, by Frank and Zhao (2001) to study relationships in schools, and by
Foster-Fishman et al. (2001) to study the patterns of relationships between community
organizations. As in those studies, the goal in using this data-collection strategy in the

present research was to construct a graphic representation of the reported ties between

48

and among members of the network. Respondents were asked to measure the ties in two
ways: the strength of the ties based on frequency (Granovetter, 1993) and type (Chavis,
1995). Results of the surveys were processed using Kliquefinder (Frank, 1996), and five
network sociograms were created based on importance ties and the allocation of
resources across those importance ties. During the initial design meeting (Point 1 in
Figure 3.1), the research team decided that the four collaborative critical issues defined
the FCC relationships. In addition to the network based on importance ties, networks
also were created for the information and service delivery networks of parenting,
substance abuse, violence, and teen sexual behavior.

Other variables included in the survey came from a review of literature on social
networks, social capital, and collaboratives. When he studied the network of the French
banking elite, Frank (1998) found that strong ties (and corresponding subgroup
membership) correlated with the school a participant had attended. Considering that such
association might be related to a potential cohort affect, descriptive variables were
included in the survey to accommodate for associations such as school attendance, length
of time in the community, and length of time in the current job.

The literature on community collaboration suggests that the strength of
identification an individual has with an organization could influence that person’s
identification with other organizations or other individuals (Foster-Fishman et al., 2001).
To accommodate this possibility, three indexes were created. The following items
constituted the three identification measures:

Identify with _(Place)__
IdenOl “My values fit in__(Place)__.”
Iden02 “I belong in _(Place)___.”

Iden04 “I identify with other people in (Place) .”
Iden07 “People in (Place) like what I like.”

49

Identify with the FCC
IdenOl “My values fit in the FCC.”
Iden02 “I belong in FCC.”
Iden04 “I identify with other people in FCC.”
Iden08 “People in FCC like what I like.”
Identify with My Organization
Iden01 “My values fit in my organization.”
Iden02 “I belong in my organization.”
Iden04 “I identify with other peOple in my organization.”
Iden08 “People in my organization like what I like.”

The sociograms were used in three ways in this study. The first was to inform
the research team about importance ties between them and, in that way, to give direction
to ongoing work. The second was to create a picture of the FCC network based on the
reported ties of the members. The sociograms were used as a template to organize the
data from field notes and key informant interviews. The graphs served as a method to
triangulate the quantitative survey information and the qualitative interview and field-
note data. Interview responses, for example, were placed next to the respondent’s spot on
the sociogram, yielding information to verify the location in the network and to offer
some explanation for it. The sociograms also gave members the ability to visually
compare and contrast the different ways resources generally flow through the network by

creating separate pictures for each of the four critical issues without having to know the

identity or location of particular individuals.

Interviews With Key Informants

The key informant interview protocol (Appendix B) was based on the reflection
of the stakeholder group on the survey work that preceded it and previous social capital,
social network, and collaborative research. Interviews are often used in social network
analysis (Schensul et al., 1999; Wasserman & Faust, 1994). Originally, the interviews

were planned for just a few of the network members; given the participative nature of the

50

work, no specific number was proposed. But after reflecting on the sociograms, the
research team decided that “more rather than fewer” interviews would help bring clarity
to the distribution of the network (Field notes, July 2002).

The interview protocol included questions that were Open-ended and allowed for
flexibility in ongoing questioning. They were exploratory in nature, a questioning
process consistent with qualitative (Creswell, 1994; Miles & Huberrnan, 1994), social
network analysis (W asserman & Faust, 1994), and participatory action (Park, 1999;
Reason & Bradbury, 1994) designs. In the interviews, group members were asked to
give details, examples, and illustrations. These data were the basis for what Geertz
(1973) called “thick description,” weaving the fabric of community life by connecting the
threads. The data-collection and research tools were used to operationalize the concepts

and theories associated with this work, as defined in the following section.

Conceptual_and Operatimial Definitions
Influence
Conceptual definition. Interpersonal influence occurs when one manipulates what
another thinks and subsequently does (Frank & F ahrbach, 1999).
Operational definition. Influence was measured in this study by responses to
question 8 in the interview protocol. Evidence of influence was when a respondent

indicated that he would change his behavior based on the opinion of others, or vice versa.

Mandates
Conceptual definition. Mandated relations are inescapable requirements that one
individual or system imposes on another. Formal relations require traditional

organizational structures such as Operating agreements, job descriptions, and protocols

51

(Chavis, 1995; Stevenson & Greenberg, 2000; Wolff, 2001). Collaborative relations
exist between individuals who come together and organize to address a concern regarding
child, family, or neighborhood well-being that transcends any one organizational
response (Bruner, 1998; Chaskin, 1992).

Operational definition. Mandates were qualified in the network section of the
Community Delivery Survey when respondents were asked to indicate whether the
relationship to each person they listed as important or a source of resources was formal,

mandated, or collaborative.

mm

Conceptual definition. Proximity refers to an individual actors relative distance
from other actors or the network (Cohen & Prusak, 2001) translated into time and space
to make connections (Gittel & Videl, 1998; Frank & Yasumoto, 1998; F oster-Fishman,
Salem, Allen & Fahlbach, 2001).

Operational definition. Time and space proximity was measured in this study
through responses to interview questions in which respondents explained the distribution

of their time or their assessment of the distribution of others’ time.

Selection

Conceptueflefinition. Selection is the process that occurs when one actor
chooses to interact with others based on their attributes (Frank & Fahrbach, 1999; Frank
& Yasumoto, 1998; Wasserman & Faust, 1994).

Operational definition. Selection was measured in this study by network

members’ responses to question 5 in the interview protocol. In a network, selection

52

normally occurs because a people seek balance (people who are like themselves) or

information.

Social Capital Dimensions

Conceptual definition. Social capital refers to an investment in social relations
that provides access to actual or potential resources (Bourdieu, 1985; Lin, 1999),
advantage (Burt, 1998; Frank, 1996), shared values, reciprocity, and action between and
among actors for benefit (Coleman, 1990; Gittell & Vidal, 1998).

Operational definition. Social capital was represented in the use of social ties
between and among members of the group and measured by the members’ explanations
of how such ties were used (Key informant interview protocol, questions 4 and 6).
Reciprocity was represented in interview comments that indicated action based on a sense
of regard beyond quid pro quo where a respondent indicated that she had somehow

internalized the well-being of another actor.

Social Network

Conceptual definition. A social network consists of a finite set or sets of actors
and the relationship or relationships defined by them (Frank, 1996; LeCompte et al.,
1999; Wasserman & Faust, 1994).

QJperational definition. The FCC network was defined and assessed using the
results of the nine network sections in the Community Delivery Survey. Respondents
were asked to identify individuals with whom they interacted or from whom they
accessed resources. The sociograms were verified through comments made by key

stakeholders in the interviews (questions 2, 4, and 5).

53

Strong/Wealglles

Conceptual definition. The strength of the ties between members of a social
network determines how those individuals behave within the network structure. Strong
ties provide the stable structure of shared values, trust, and reciprocity. Weak ties
provide access to unique resources (Frank & Yasumoto, 1998; Granovetter, 1973).

Operafimal definition. Strength of ties was measured in two ways. Respondents
were asked to qualify the frequency of their contact with those they identified as being
important to their work or from whom they received resources. Data from the interviews
also were considered as representing strength of ties, with weak ties being described as “a

way to get resources” rather than conveying any sense of mutual benefit or identification.

Research Validity

To establish validity, I triangulated the data. This was accomplished by
grounding key stakeholders’ interview responses within the context of the sociogram
pictures. Because the survey network selection process resulted in the sociograms, the
sociograms provided a collective picture of all members’ relational ties to one another.
Those pictures were then verified and explained through the triangulation of the
quantitative data (the pictures) with the interview data.

The other strategy I used to validate the work was grounded in the participatory
method itself. In this work, all parts of the design, implementation, and analysis passed a
peer review process (Park, 1999; Small, 1995), which served to validate the work in the

day-to-day community lives of participants.

54

Limitations

One limitation of this study was that a participatory design takes a long time to be
completed. The reflection-action-reflection framework (Freire, 1970) requires time for
participants to gather information, learn together on a number of knowledge levels, and
then to act. Given the complexities of community collaboration, the design requires a
longitudinal effort to observe with completeness the process being suggested.

The second limitation was that the research was an attempt to converge complex
(social networks) and emerging (social capital/community collaborative) concepts.
Despite a relative lack of clarity about these concepts at the community level (Foster-
Fishman et al., 2001), I attempted to use a participatory model to explore the concepts.
The limitation was that the intellectual sophistication of the study limited the individuals

who could engage in the analytical aspects of the work.

55

CHAPTER IV
RESULTS

Introduction

This research study was conducted to answer two primary questions. The first
was: How does the concept of social capital relate to members’ ability to build, maintain,
and access resources in a collaborative network? To answer this primary question, the
following subquestions were formulated to guide the study:

1. Is there an underlying social structure that determines the social ties?

2. Are individual members positioned within the network in predictable
locations?

3. Does the network show differences in access and distribution when the
collaborative work differs?

4. Are there patterns of ties that emerge from an analysis of the network?

5. Does the selection process differ for individuals in the network?

6. Do individuals describe an influence process in the network?

The second primary research question was: How do network members respond to
social capital, network concepts, and the research process? Further, does action result?
In the following pages, the data that were gathered to answer these questions are

presented and discussed.

Data Gathering and Generation of Results
The research questions were answered using the seven-point data-gathering and
result-generating process illustrated in Figure 4.1. The survey data are presented here in

a sequential manner that operationalizes the "reflection-action-reflection” process of

56

participatory action research (F reire, 1970; Miles & Huberrnan, 1994; Park, 1999).
Reflection at stakeholder meetings was the basis for “next step” action. Answers to the
research questions can be confirmed only through a thorough consideration of the process
from which that knowledge emerged. Therefore, the seven points in Figure 4.1 are
discussed in detail in the following pages to report the convergence of research,

education, and action in this participatory research (Bradbury & Reason, 1994; Small,

 

      
 

 

 

 

 

 

 

 

 

 

 

 

1995).
Sltak 611.01% Social Capital
eetmgs .
D . Data Generation Process
esrgn
l.
S Ongoin feedback
urveys
l6
Distributed
15 returned
2.
Stakeholder
Meeting
Sigigggs - Stakeholder
3 Meeting
' 7.
Snowball sample Data Sociogram
Key informant Analysis linterview
interviews 5. triangulation
4. 6.

Figure 4.1: Social capital in a community collaborative network: Generation of results.

In this study, field notes and interview data in raw form Often included the

participants’ names. When the data were analyzed, I replaced participants’ names with

57

the codes assigned in the survey process. Comments, when quoted, are referenced with
the code number. The numbers can be located on the sociograms to consider

stakeholders’ remarks within the network as constructed.

Point 1: Stakeholder Meetings

Three formal meetings were conducted to initiate the study.

1. The researchers met with the Executive Committee of the collaborative (the
FCC) to gain their approval.

2. The researchers, with the recommendation of the Executive Committee, met
with a standing committee of the collaborative and recruited them to serve as the study’s
stakeholder group.

3. The researchers and stakeholders met to design the study both conceptually
and practically.

The relevant events that took place at those three meetings are discussed in the
following paragraphs.

Executive Committee. The insider introduced the outside researcher, who had
been accompanied to the meeting by a representative from the university who had
previous association with the FCC. It was thought that her presence would provide
credibility to the work. In the introductions, she talked about the relationship between
her work and the proposed study. Stakeholders commented that good experience with
her office in the past, encouraged them to pursue the study. The proposed work was
“realistic,” one remarked (1101). Another commented that “they,” referring to the
university, “will get this done” (1901). Five Executive Committee members joined in a

dialogue with the researchers about the relevance of the work to the needs of the FCC.

58

Two questions emerged in the discussion: What would the research tell
stakeholders? What would they do with the information? After an hour-long discussion,
the stakeholders agreed that the study would give the FCC information about how
members related to one another and whether the collaborative as a whole seemed to be
“connected” (3001). When asked, they said “connections” could be measured by “who
was talking to whom about what” (1101, 1901; Field notes, July 18, 2001, meeting).

This group expressed some concern about studying the “whole” FCC, given the
size of the group. There was consensus that one particular standing committee of the
FCC would be asked to serve as the stakeholder group for the study. At Point 1 of the
work, this was considered a pilot of the method. The leadership thought that, alter the
first wave (completed study of this standing committee), decisions could be made about
involving the entire FCC.

At this meeting, stakeholders and the two researchers discussed their various
roles. The role of the insider was clarified. The study was to serve as the basis for the
insider’s doctoral dissertation. There were favorable comments about that fact. One
stakeholder said, “We will do whatever to get her done” (1101). A subsequent question,
however, spurred 10 minutes of conversation: How is she (the insider) going to use the
information? (1301). The research team assured the stakeholders that strict
confidentiality would apply. The group reviewed the Human Subjects Review form from
the insider’s university.

In the initial meeting, two baseline decisions were made. First, the research was
deemed useful. Second, to accommodate the role of the insider, ongoing dialogue would

be necessary.

59

Standing committee recruitment. The research was recommended to the full
membership of the standing committee as a self-examination method to explore the
relationships between and among members and the strength of those relationships, a need
identified at the May 2001 retreat. The inside researcher proposed that network analysis
would be useful because such analysis, conducted as a participatory action study, would
help inform the group about individual relationships and network function, as well as
generate potential action that could be taken to expand or to enhance the network
(Committee minutes, September 2001). Executive Committee members summarized
their recommendation by saying, “This will help us realize our connections” (3001).

Sixteen members of the committee attended the meeting; one was absent. The
group agreed to serve as the stakeholder group for the study. There was little discussion
at this meeting about the research. The group was “willing to go along with it,”
according to one member (3101). No dissention was expressed. The leadership group
that had met with the outside researcher gave positive comments about what could be
learned. One member indicated that, given the leadership’s recommendation, it “made
sense to do it” (1901).

Stakeholder design meeting. The insider and the outside researcher sent a
preliminary letter (Appendix C) to all members of the stakeholder group to frame the
work. Fourteen of the 17 members attended the October meeting. The work of Frank
and Zhao (2001) was used as an illustration. In that study of teachers,’ sociograms similar
to those that would frame the proposed collaborative work were produced. The group
was most engaged with the sociogram of teachers in a school-diffusing innovation. The
sociograms pictured the individual teachers by number, the subgroups that emerged from

the association, and the ties between them.

60

Three threshold issues emerged during this discussion: confidentiality, the
evaluative nature of position, and the practical advantage of position. When the teacher-
study sociogram was presented to the group, a group member immediately asked whether
results of the FCC survey would plot each person (2302). Could “I be identified?” one
person asked (3202). It was explained that individual actors’ names would be coded.
“Who knows the codes?” another asked (1201). The insider said she would do the
coding. The group had a lengthy discussion about confidentiality. The concern centered
ion the pictures exposing individuals’ identities. There was clear direction that no names
should be reported next to dots on the FCC sociogram until the group reviewed the coded
pictures.

The next discussion was about an actor’s position within the picture. What
position would be the preferred one? Many in the group indicated that being “in the
middle” of the picture was favorable. “This would be a way to tell if you are being
collaborative,” commented one member (2302). Another agreed with that comment by
adding, “It would not look very good if you were outside” (4062). They were concerned
that position within the network could be used to evaluate an individual or that person’s
organization. After some discussion, the group agreed that, without names, there would
not be a serious threat of vulnerability.

Some group members, however, did not think that being in the middle was a
preferred position. One member said she “did not mind being an ‘outlier’ at all” (3001).
“If you’re in the middle you can’t move,” she said. The group generally agreed that
position within the network would vary, depending on “how much work you had to do

with the other people” (3202). Referring to the school sociogram, a member pointed to a

61

teacher in the middle and said, “We don’t have to do a study. We just have to find
teacher 21, take her out for a drink, and have her explain it” (1101).

The remainder of the meeting was used to develop definitions and parameters for
the study. In defining the network parameters, the committee settled on the “four critical
issues” around which the FCC originally had organized. They talked about how the FCC
defined its work and to what extent that definition reflected the critical issues. The group
had significant conversation about this point. Some FCC members were involved with
the critical issues through a collaborative (formal groups created by the FCC); others
were not members of the collaborative but were still involved with the issue. Some FCC
member agencies did not deliver services (related to teen pregnancy, for example) but
funded such programs. Some committee members did not personally deliver the services
because their staff did that. Some members did not have any involvement with certain
issue and received only general information. Thus, the group agreed that it was important
to ask both “who helps you deliver services” and “who gives you information.”

It was explained that the Kliquefinder software would use the ties as reported in
the survey to create a sociogram, which would be a social structure map of “importance”
in the FCC network. The resulting “importance ties” network would then be used to
observe and analyze patterns around each critical issue. After some discussion, the group
agreed that frequency was an indicator of the strength of a tie, but they also wanted the
type of tie to be included. Participants added that an indication of “formally mandated, a
formal partnership, or a collaboration” should be included in the survey to further qualify
relational ties. The stakeholder group agreed that identification and demographic items

could also be included in the survey.

62

Application of data to research cmestions.

Question 1: How does the concept of social capital relate to members’ ability
to build, maintain, and access resources in a collaborative network?

The stakeholder group meeting verified that social ties were influential in
accessing and allocating resources. At the Executive Committee meetings, for example,
the relationships with university outreach and with the inside researcher influenced
stakeholders’ willingness to participate in this work. Social capital, accessing actual or
potential resources through social ties, was present then in gaining access and
participation.

At the stakeholder meeting, remarks about the insider’s access to the data and
about identification indicated that the group had concerns about trust and the implications
of social ties. They thought that “collaboration” would be evaluated and that, given their
relative position, they also would be evaluated. From this early juncture, the power that
is embedded in the relationships within the network was clear.

The group’s decision to organize the network study around the critical issues was
evidence that they believed that how resources moved through the network was based on
what type of work was being done. They saw the relationships as adapting to the nature
of the work at hand. Different work, the group hypothesized, brought one to select
different sets of people.

Question 2: How do network members respond to social capital, network
concepts, and the research process? Does action result?

Members discussed the concept of position within a network by looking at the
school-study illustration and applying that to their own network. They had some ideas
about their own positions and the potential of a preferred “collaborative” position. They

decided that being in the middle was not preferred for some, although it was for others.

63

They responded with interest and caution. The decision not to plot names in the
sociogram related to the level of trust within the group.

The personal aspect of the research was uncomfortable for some group members.
The group was interested in learning about network analysis and the sociograms. The
conversation that took place concerning their work and relationships was evidence that

they found some relevance in the concepts.

Point 2: Communig Delivery Survey (Appendix A)

Of the 16 surveys distributed, 15 were returned. The survey asked respondents to
choose individuals, by name, in each of the nine networks. In the first round, 45 different
individuals or groups were chosen. Hence, 45 codes were assigned, one for each of the
individuals or organizations listed on the respondents’ surveys. The respondents’ names
were removed, and then the surveys were forwarded to the insider for coding. The data
were then entered into the Kliquefinder program, and the following five sociograms were

produced:

Importance Ties

0 Violence Overlaid on Importance (using dotted lines to represent
the information network and solid lines to represent the service
delivery network, a method repeated in all overlays)

0 Substance Abuse Overlaid on Importance
Teen Pregnancy Overlaid on Importance

- Parenting Overlaid on Importance

Other variables, such as descriptive information (i.e., gender, length of time in job, length
of time in the area) and group identification, were not analyzed at this point in the study
due to the small size of the sample.

The assignment of codes proved to be a job for the insider because knowledge of
the community and of the general associations within it was required to code the

selections. Two coding strategies were adopted for use in this study, consistent with
64

selections. Two coding strategies were adopted for use in this study, consistent with
other research done with Kliquefinder (F oster-F ishrnan et al, 2001; Frank & Zhao, 2001).
First, when coding names of those chosen, the insider coded people who worked in the
same organization with a sequence of numbers; i.e., all who worked for agency X were
coded in the 1000 series, like 1001, 1002, 1003. This allowed the stakeholders to observe
the cross-linkages in the sociograms without knowing individuals’ names. If a subgroup
included five chosen numbers like 1002, 3001, 2005, 1003, and 1004, an observer could
conclude that three organizations were represented in that subgroup. Second, the insider
“cleaned” the data by carefully looking for duplicate names. In choosing network
members, the “chooser” may have misspelled a name or used an acronym to refer to an
organization. The Health Department, for example, was referred to as the county health
department or MPHD, the Public Health Department, and by county name. These are all
names for the same organization, so they were coded with the same number.

Both of these coding strategies constituted a sorting process based on an
undeniable insider bias that, given the confidentiality constraints of this study, was
impossible to avoid. In a study in which the names of both chooser and chosen were
included, the coding could be validated through the peer-review process to reduce the
influence of insider bias. In this study, it was more important to maintain the autonomy
of both chooser and chosen. The stakeholder group was willing to show the chosen
names to the inside researcher, but this meant that bias would have resulted.

While stakeholders were completing the survey, a number of questions surfaced
about the scope of the study. In telephone calls and email messages, the insider was
asked whom stakeholders “should” be thinking about when they completed the survey.

They asked, “DO I think about people in town only?” (2302, 1101, 2101) and “I get these

65

things from people in my organization; do I list them?” (3202, 1601,1701). Stakeholders

were instructed to choose the people and agencies that were most important to them and

their agency (Community Delivery Survey, Appendix A) and to define that as they saw

fit. Choosing who was included in their personal network then was left to them.
Application of data to research questions.

Question 1: How does the concept of social capital relate to members’ ability
to build, maintain, and access resources in a collaborative network?

When faced with choosing the most important relationship, stakeholders varied in
the criteria they used. Some indicated that they selected individuals who were important
to the work of the FCC, whereas others indicated that they selected individuals who were
important, as stakeholder 1901 said, in “all the things I do.” Theoretically, the
individuals selected represented the chooser’s social structure. Social capital exists in
relationships that have a high level of reciprocity, trust, and mutual values and actions.
Such relationships were likely to exist in the social structure produced through the
survey/sociogram process. The egocentrically created social structure needed chooser
clarification before one could determine how social capital resided within it.

When the coded survey results were then analyzed using Kliquefinder, the
individual choosers’ selections were combined with all others, and using an algorithm
process, relative positions for the chooser were determined. At that point, the social
capital of the chooser moved to the network level as the individual and his or her
selections were embedded within the entire social structure.

The conceptual complexity of social network analysis became evident at this
point in the study. A number of stakeholders wanted clarity but, given their level of
experience with social capital and network concepts, it seemed just too “abstract” (1003,

3001). Despite that frustration, the stakeholders filled out the survey, but on two

66

 

occasions indicated that they did it to comply with the task (1601, 1701, 3202). These
stakeholders were influenced to do so, not because they saw value in the work but
because they saw value in complying with the norms of the group, either because it
helped someone else (the inside researcher) or because it kept them in the process in case
something of value came from it.

Question 2: How do network members respond to social capital, network
concepts, and the research process? Does action result?

Some stakeholders thought that the scope of the survey was not clear (2302,
1103), raising questions about “how to think about all the people” (1503, 1601). It
seemed ambiguous to a number of them because the network was not predefined. In
addition, some stakeholders expressed frustration with the complexity of the work. As

one completed the survey, she remarked that she could not see “how this would tell
anything” (1001).
Point 3: Stakeholder Meeting: Interpret Sociograms
agd Inform About Kefitageholder Interviews

The research team presented the sociogram “pictures” to the stakeholder group at
a special meeting. The presentation included a brief description of how the pictures had
been created from the survey results. The sociograms included the code for each
“chooser” or participant and those codes assigned to people who were “chosen” or named
as part of a network. They did not show names, consistent with the design’s
confidentiality agreement. Codes that appeared on the sociogram represented all
individuals who either chose or were chosen in the initial survey. A total of 40 different
points appeared, with a variety of lines connecting each of the four emergent subgroups.
To ease interpretation of the sociograms, the insider assigned a category to each of the

subgroups based on the general orientation of the services members delivered. In

67

 

subgroup A, for example, most of the members were involved in justice-related activities.
In each of the four groups, however, there were people who seemed inconsistent to the
inside researcher. It seemed that within the network identified in this survey sample,
those inconsistent stakeholders were probably the cross-linkage people. Preliminarily,
those people were considered the subgroup bridgers.

The “Importance Ties” sociogram was used to illustrate the emergence of
subgroups from the survey data and to begin discussion of the strength of ties, the
position of various groups, and the issues involved with unlabeled interpretation. The
relationship ties, shown in Figure 4.2, were combined to represent those that connected
the 21 connections represented in the sociogram. For ease of interpretation, the ties

between individuals were merged into solid lines between subgroups.

Funding

 

 

C
Human
Services

 

Legend

Subgroups that
O emerged from

the initial survey

Health _ General orientation
of subgroup

\ Group
Ties

 

 

 

Figure 4.2: Illustration of importance of ties in the collaborative.

68

The insider and the outside researcher reviewed the sociograms with stakeholders
to elicit their reactions, thoughts, and general interpretations. The comments can be
grouped as general reactions to the “Importance Ties” sociogram and comments about
differences among the four critical-issue sociograms.

The general reaction to the “Importance Ties” sociogram was confusion about
how to read the picture and puzzlement about where the subgroups came from and what
the lines and subgroups meant. The core response from the stakeholders was curiosity
about the network picture. They wanted to know whether the researchers thought it
reflected anything “significant” (1101, 1301). Stakeholder comments included the
following:

0 “These [the pictures] are really overwhelming”(3001, 1103).

o “I must have missed the meeting where we learned about what this
means” (1601).

0 “We must just spend time talking to one another”(1201, 4062).

o “Is this good, all of these lines?” (1301).

The network, according to the outside researcher, appeared to have density,
symmetry, and inclusion (LeCompte et al., 1999). There were, for example, a number of
connections between subgroups, which could indicate that people did talk between
groups (density). Also, there were multiple connections between groups, so it appeared
that the connections were not limited to certain individuals (symmetry). Some
individuals did have more connections than others, indicating that they served as critical
connectors (linkers) between groups (1901, 1001, 3001, for example). Ideas shared in the
discussion included the following:

0 “There is balance in the connections” (1301).

69

0 “We look like we are working with each other.”

0 “Makes sense that the funding agencies are not central to the
conversations and only connect at certain points” (3 001).

o The connector in juvenile justice would include (various names were
identified) (1503, 1201).

0 “There are a couple of people that are very central” (1103).

o The differences between the critical issues seem to indicate that we can
mobilize the basic network to organize around a service delivery network
when there are resources (3001 , 1101).

The individual critical-issue sociograms elicited the following types of reactions:
Violence:

0 Everyone is connected to juvenile justice people, which makes
sense because of recent planning (1201, 1701).

o The connection to human service relates to FIA’S being a point of
referral for the courts (1103, 1601).

o The human service connection would be the Child and Family
youth interventions and Webster House for Runaways (program of
Every Woman’s Place) being the contractual provider for the
PIA/court clients (1701, 1101).

Substance Abuse:

0 “There is a lot of connection going on, because substance

abuse is at the root of a lot of other interventions” (1103).

70

“This reflects that there is inadequate funding for substance
abuse, that we need each other to piece together services”
(1201,1301)

“Everybody has to deal with substance abuse” (3001, 1701).
“Justice, healthcare, education all are using whatever public
sanctions can be used to make people comply with treatment”
(3101).

“I can ask the court to order someone to treatment or as the

FIA to make treatment a condition of parental rights” (1601).

Teen Pregnancy:

Parenting:

“This shows the Teen Pregnancy network created by the state
funding” (1101).

“The delivery network is lined up because the individuals are
organized into a specific, coordinated system.”

“We all know who those lines all run to” (1101, 1301, 3001).

“Lots of lines for information, few lines for delivery of service.
That is because nobody knows what anybody else is doing”
(1701)

“Parenting has been hard for the FCC to organize into any
network” (1201).

“I am amazed that there are so many connections” (3001).

“I wonder what all of us are talking about” (1101).

71

Interpretation was difficult for the stakeholder group without having individual
names, yet they reiterated that the names should not be reported. They did agree that
interviews with the key stakeholders would shed light on their initial reactions to the
sociogram pictures. They were interested in individual positions (most speculated aloud
about where they were located in the picture) in dyads (comments like “I talk to you all
the time” were common) and in the network (naming an individual who was known to
bridge between subgroups, recognizing the variation between topic networks).

This stakeholder meeting was held on a Friday afternoon the week before mid-
winter break. The general tenor of the meeting was an “end of the day,” “need a break”
kind of mood. The fact that all stakeholders who were not already out of town attended
this meeting is evidence of the group’s commitment to the research. They were intrigued
about the work and were curious at what point the work would inform the FCC. They
wondered why individuals might be positioned in various locations in the network and,
given that, how they might access and distribute resources. The group generated a
number of possible questions for the key informant interviews including the following:

0 Why do we select the people we do?

0 What are the differences in our relationships with one another?
0 Do we respond to opportunities with creativity?

0 What can we do differently?

The group also agreed to extend the survey and network study to include a
snowball sample of the individuals or organizations identified in the initial stakeholder
group sample. The group agreed to take personal responsibility for contacting those
people they had identified, explaining the study, and asking them to complete and return

the survey instrument.

72

Application of dago research questions.

Question 1: How does the concept of social capital relate to members’ ability
to build, maintain, and access resources in a collaborative network?

Stakeholders continued to express interest in the concept of social capital by
asking questions about position, density of ties, and connectedness. “We are really
talking to each other” (1301), one said. The group could “make some sense” of the
sociograms as they collectively agreed that the “network was stable” (3001, 1301) and
adaptable because the flow of information and delivery adjusted to different topics.
“That is a good thing, that we move based on opportunities” was the conclusion of one
member (3001), and others concurred. One member commented that he was “shocked
that we work that much around these issues” (1201). Group members indicated that the
sociograms verified their perceptions that the FCC moved to meet those opportunities.
“Because of the FCC, we were able to organize and get teen pregnancy funding” (3001),
one commented, whereas another said, “This kind of collaboration got us the juvenile
justice grant” (1601). Others concurred. The enduring nature of the relationships
allowed stakeholders to adapt with “speed’ (1601, 1701), which means that social capital
in the network allowed the group to access resources.

The strength of the network also moved scare resources around. Substance abuse,
members agreed, was everybody’s business and nobody’s business. “There are no
resources to pay for treatment, but it is at the root of most of the problems we all face,”
one member remarked (1701). With regard to substance abuse, another remarked,
concurring with the first, that the systems “really need each other” (1103). The FIA can
threaten to remove the kids, the court can issue an order, and the school can expel, she
explained; “the force is greater if we all can make the point together.” The systems

individually wanted a client to comply with a substance abuse plan. They were most

73

 

successful if they approached the client collectively. One of the resources accessed
through the collaborative system was the sanction that each system could bring to bear on
a given situation. Mutual action, to sanction client compliance, was clearly seen in the
density of ties in the substance abuse sociogram.

Question 2: How do network members respond to social capital, network
concepts, and the research process? Does action result?

Although the stakeholders responded with enthusiasm to the conversation and the
concepts, the tools were frustrating for stakeholders to interpret with any confidence or to
apply to their situation. There was a need for the outside researcher to interpret the
results. They wanted him to diagnose the network. In this meeting it was evident that the
concepts were relevant to the stakeholders’ work. It was, however, difficult for
stakeholders to reflect on results or to advise on further research because the results were

constrained by confidentiality requirements and the concepts were new and frustrating to

apply-

Point 4: Snowball Simple

At point 4, the insider sent community delivery surveys to those “chosen” in the
original survey, drafted the interview protocol, and coded the snowball surveys. A total
of 65 snowball surveys were distributed and 31 were returned, expanding the survey
responses to 46. As a result of those surveys, the code book of organizations and
individuals expanded to 87. The descriptive and inferential variables were analyzed to
describe the study participants and to look for relationships between various items.

Descriptive and inferential results of variables. Two separate versions of the
survey existed when the snowball sample survey was distributed. One included the

demographic items; the other did not. Unfortunately, of the 46 surveys that were

74

completed, approximately 35% had some missing data, which made a meaningful
analysis of any demographic data impossible.

The “identification” questions were, however, available for analysis. Pearson
correlation coefficients indicated that all scales had an acceptable level of reliability. The
results are as follows:

Identify with the community = totm =.79
Iden01 “My values fit in the community.”
Iden02 “I belong in the community.”
Iden04 “I identify with other people in community.”
Iden07 “People in the community like what I like.”
Identify with FCC=totf=.74
Iden01 “My values fit in the FCC.”
Iden02 “I belong in FCC.”
Iden04 “I identify with other people in FCC.”
Iden08 “People in FCC like what I like.”
Identify with My Organization = toto =.81
Iden01 “My values fit in my organization.”
Iden02 “I belong in my organization.”
Iden04 “I identify with other people in my organization.”
Iden08 “People in my organization like what I like.”

Pearson correlation coefficients showed a significant correlation between
identification with the community and identification with both the FCC and the
participant’s organization. Given the low number of participants in the sample, however,
these results can only be used to suggest that the higher the level of identification an
individual has with the community, the higher the level of identification there might be
with the collaborative and the organization.

Sociogram emergent subgroups. The snowball sample produced 12 subgroups.
The sociograms of the critical issues were the result of the Kliquefinder analysis of the

snowball sample, total analysis. They could not be included in this dissertation because

they did not conform to dissertation guidelines.

75

 

In looking at the sociograms, a number of things could be observed about the
network. The network has density (a number of connections between subgroups) and
symmetry (an evenness or balance to the distribution of ties within the network). A

closer look at the subgroups indicates a number of patterns in the data (Table 4.1).

Table 4.1: Snowball sample subgroups.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A B C D E F G H I J K
#inthegroup 8 12 12 8 8 18 ll 17 16 17 9
#ofFCC in l l l 3 0 6 2 l 0 8 1
group
# of agencies 7 8 6 6 7 10 7 9 7 ll 8
ingroup
h- : t:

t '- '0 1.. .5 "" ..... -_=
.. a 8 s e e a E E 8 E a
9. w a a) a) 99 4.. o u“ 4:: 3’: H o
v x a < w '5 So a a a: = 2 = < "~-
5 3 a o .3 ‘5 8 <6 .5 5 '9 3 LE 3 o E '3
o a a z < e. e. 2 o m a": .‘3 O m z 0 a.

 

 

 

The groups are basically the same size, and there is some balance between them.
They all share significant diversity of members as a large percentage of the members
work in different agencies. These are importance ties, so there is a high level of cross-
agency linkage among the subgroups of this network. For most of the groups, the FCC
members are in the minority; that makes it highly likely that they could serve to bridge
their subgroup to others. This also means that those bridging ties serve to bring resources
into the subgroup, as described by Granovetter (1973) as a function of weak ties. The
individuals (in groups A, B, C, G, H, and K) are in strong competitive positions (Burt,
2001) because they are on the outside of the network and have maximum flexibility to

move, as opposed to those who are in the network interior with many obligations

76

(Wasserman & Faust, 1995). Groups E and I have no FCC members, so there is no
obvious bridge, but these groups have a high level of diversity in their memberships,
which gives them a number of ties to a variety of the other subgroups.

No quantitative analysis could glean much meaning from the sociograms or Table
4.1. The surveys that generated these illustrations were completed to produce a
framework through which ties could be observed, explained, and understood. This was
accomplished through the key informant interviews conducted with all individual
stakeholders.

Application of data to research questions.

Question 1: How does the concept of social capital relate to members’ ability
to build, maintain, and access resources in a collaborative network?

The number of ties that appear on the sociogram indicates the selections made by
members of the network. These ties can be interpreted as the manifestation of social
capital as they represent the route actors took to access and distribute resources. FCC
members of subgroups, for example, interpreted the connections as a way (1901, 1601) to
access resources through weak ties and get them to get work done. Subgroups had a
number of pathways between individuals and subgroups, and the patterns adjusted based
on the issue showing how the network adapted to maneuver around various work
demands and needs. “We move around to find the money,” one stakeholder remarked
(3101). The group also saw the enduring relational structure as a way to judiciously
move quickly. “It is because we talk that we can get in line” was one remark (1301).
The network also served to move resources to niche groups (A, B, and C) and
nontraditional players (E and I). “We all know who that is in Group C,” said a

stakeholder. “They have a strong niche in substance abuse” (1701).

77

Question 2: How do members respond to social capital, network concepts,
and the research process? Does action result?

The snowball sample did not generate any data for this research question.

Point 5: Data Analysis of Keflnfomant Interviews

All members of the stakeholder group were asked to participate in key informant
interviews. They were given the option to be interviewed by the insider or the outside
researcher. A total of 16 interviews were conducted. Thirteen of the interviewees had
already been coded and appeared on the sociogram. Two stakeholders did not appear by
name (coded by name), but the organization of one of them was coded and in the picture.
The 16th interview was with a person who originally had been involved with the FCC but
had retired. After a number of interviews had been conducted, it became clear that she
might offer a particular perspective to the study; thus, she was added. Her interview was
used as archival information.

The interviews followed a prescribed protocol (Appendix B). They took an
average of an hour and a half and were tape recorded to provide transcripts for analysis.
An informed consent process, approved by the human subjects review committee at the
university, was followed with each interview.

Although the interviews were structured, they were also designed to be
exploratory. Before this, social capital had not been defined in a community
collaborative network by its members. In the interviews, the goal was to capture
specifics about relational ties in order to discern at what point and under what conditions
collaborative members make social capital investments. Data from the interviews were

analyzed using a compare-and-contrast process to garner themes from the responses

78

(Miles & Huberrnan, 1999). The findings for each interview question are presented in the
following pages.

Interview Question l--Relationship between the researcher and the stakeholder.
All stakeholders acknowledged that they had some level of relationship with the inside
researcher. The strength of relationship varied. Some stakeholders indicated that they
would not be influenced (either positively or negatively) by the relationship (2200, 2302,
1201). Most respondents, however, said that they were motivated to cooperate with the
interview because of a sense that to do so would be beneficial for them because they
wanted to help out the insider (1601, 1901, 4602), because they might learn something
(1503, 1301), or because they could be frank because oftrust (1103, 3101). One person
did say that he “would be more cautious with an active member of the network” (1301)
but was comfortable moving forward with the interview.

Most people thought that relationships within the collaborative varied a “great
deal” (4062, 1101, 3001). Most agreed that the insider would not hinder participation in
the study (1901, 1503, 2201, 2303).

Interview Question 2--Involvement with the collaborative. Involvement with the
collaborative varied. Two stakeholders said that they used to be involved but that their
job assignments had changed (1103, 3202). People mainly indicated that they saw the
FCC as a place to “connect” (2201, 1503, 3101), to access resources in the form of
information (1601, 1201) or funding, or to demonstrate a collaborative effort (1701,
2101). Stakeholders talked about the convenience of having people in the FCC because it
provided access and proximity: “You see them and then can say later, ‘Saw you at the
F CC’” (2101). It also saved “leg-work” (2201) and allowed for smaller organizations

(those that have a niche market) to hear what the grant makers were saying (2201) so that

79

one could “line up” the organization (1503, 3101). The FCC was a place where you
could “tell your story” (1402, 4062).

Interview Ouestion 3--Are there rules and, if sophow are they enforced?
Members believed there were rules but no real capacity to sanction. One member
indicated, “If you only look out for yourself it will catch up with you” and people will
ignore you when you need help (1701). Some members expressed frustration because
they thought the collaborative rules were unspoken (2101, 2201) and that you discovered
them only by making a mistake or if someone told you (or you asked). The rule most
referenced was that members were “supposed to come to the table to give something”
(3001, 1101, 1701, 1503, 1901), that “something” being money (1301) or a willingness to
get involved in the work (1101, 1601), and they thought that the group could do
something (1901). “Collaboration is when you have to sit down and figure something out
with people.”

There was a rule about attending meetings (3101); if you did not show up, people
would start mentioning it to you (2201). The reason to go, in that case, was that “if you
don’t show up, you will look like you are part of the problem” (3101, 1101). Some
people did not show up on a regular basis. Others viewed that organization as not
collaborative (1503, 1103, 3001, 3101). A stakeholder from that organization indicated
that the FCC was not meeting the cross-linkage needs of the community (2302) and
“nothing is done about it.”

Interview Question 4-Compare and contrast relationships with FCC members.
Stakeholders asserted that relationships with FCC members were very different. They
varied based on experience, where individuals had the chance to see people act or to

engage in collective work that influenced their relationships. Relationships also varied

8O

based on the similarity of work. If individuals shared similar work, their relationship
could be positive, but relationships between competing individuals or organizations were
less positive unless the work was something in which they do not compete. “I connect
best when I am ‘value-added,”’ commented one stakeholder (1503).

Positive experiences in the FCC arena did not necessarily change the relationships
between competitors. One stakeholder explained that, in her relationships with the FCC,
she had worked with her strongest community competitor on an issue they both cared
about but did not directly control. This work had been a positive experience, but it had
not “changed the fact that I hate the way he competes” (3101). Two stakeholders (1503,
1301) described their relationship with an FCC member (1901) as being “limitless.”
There was nothing either could imagine that they would not “do for that guy.” When
asked where such loyalty came from, each separately indicated (a) that he had observed
loyal behavior in the other (in 1901), (b) that he was repeatedly observed as honest and
trustworthy, and (c) that he had complimentary, not competitive, concerns in the
community.

A relationship with trust, according to stakeholders, exists when one person
intemalizes the well-being of another person at any of a number of levels. For example,
person A goes to a meeting (1503) and something is going on that does or might affect
person B (1301). In a trusting relationship, person A could act on behalf of person B and
know that the action will be supported. Person A would act “for my agency with the
same regard he would act for his own” (1503). Another example is when someone calls
and asks you about something and you have a contact that could help. If you have a
relationship with the caller, you might suggest they call and use your name, but if it was a

trusting relationship, you would probably call for them (1503).

81

Collaborative relationships reflect three levels of trust and engagement. At the
start it is about people “being in a room and hearing what someone else does” (2201).
“You move toward people who have similar ways as they fit with what you want”
(1301). If people “know what you do, then they will think about you when they are
doing something” (4062). At a more engaged level, another “adjusted his schedule to
include me. He did not just say we could do it sometime. He said, ‘Let’s sit down and

9”

figure it out (1601). At this level of engagement, a member internalized the well-being
of one member. At the third and highest level of trust, stakeholders’ senses of well-being
converge. The members act on behalf of each other, knowing that each would have the
same regard for the other as he would for himself.

Consideration was a main distinction in the stakeholders’ sense of reciprocity. At
the first level, a member believes that if he does his part others will do the same (2101,
2302); for example, “Other organizations have sanctioning ability. I expect them to use
it” (1201). It is “like in the theatre, where everybody does the part to make the whole
thing work like magic,” one member suggested metaphorically (9000). At the second
level, one believes that others will exercise their influence in concert with him. “I know
that I can call him and we will figure out what to do, together” (1102). At the third level,
members come together to figure out how each of their parts can create a new
configuration. It is a convergence of activity where one knows that the other will do
whatever needs to be done. “There are people who just get it,” said one stakeholder
(1101), “who are about changing the world.” These relationships are based on a sense
that those involved see more than there own vision, either independently or in tandem

with certain others. These individuals see “changing things for the better” (1901). This

consideration is a source of trust and increased collaboration.

82

Interview Question 5--How do you select whom you interact with? According to
stakeholders, one involves oneself with someone who “brings something beyond what
they know” (1103). Get involved with peOple who have “common interests” (1601), who
believe that “you can make things better” (1901, 1101). People clearly selected others
based on personal preference (1701, 1601), those who were like them and worked hard
(1102, 2302, 1201), because, as one stakeholder described it, there are lots of people and
“I cannot resent spending time” with them (3001). Selection occurs because people have
similar positions, directing a similar size budget, for example (1001, 1301, 2302), that
can be mobilized when there is a need (3202, 1602). People select some over others to
inoculate themselves from harm (9000), to gain access to information (2101, 2201), to be
with people like themselves (1001, 3001), or to be with people who are different (1901)
from themselves.

Selection within the collaborative was based on attraction. Two members (1402,
3101) were attracted to “powerful people,” looking for information that could be relayed
(to parlay their own personal power in other settings), whereas two others’ (1901, 1301)
selections were based on gaining access to and credibility with people who were not like
themselves. “If people see me working on family issues,” one member commented,
“they will see me differently, more like them, and that helps me” (1901). “I look for
people that will move my thinking” (1301), another said. “Selection” is based on the
need to “tell your story” (4062), or “getting to know people who will then think about
you because they know what you do” (2101).

Interview Question 6--Do you access resources throughyour relationships? There
was universal agreement on this question. Resources included access itself, being willing

to make a call for someone. Other resources included information, money, and staff

83

(2201, 1201, 1301). A distinction was made, based on the strength of the relationships.
Some accessed resources from casual contacts, but if there was regularity in the contact,
resources were more likely. Monthly meeting, for example, helped people make resource
connections (“saw you at the FCC”) (2101). In a very strong relationship, however,
stakeholders said they would offer resources to people because if others could benefit, the
“whole deal does” (1601, 1201, 1101).

One member said be accessed resources by being where people were not like him
because if he was around at all the events, they got used to him. Proximity and
familiarity will bring people to tell you things that they otherwise might not, or at least
tell someone to tell you (1901). The notion was that certain relationships had a level of
honesty that allowed one to talk frankly (1701, 1602, 1901) and to “get the skinny” on
something (2302), a job applicant, for example. A

Interview Question 7--Are there people in the FCC tflt are central to the work?
Are there others who argnarginal? There was general agreement that the collaborative
had “inside” and “outside” groups. Some members thought that the inside group was
formed by the original members (1101, 1103, 1301), who internalized the mission of the
collaborative. One person thought that the community was made up of “doers” and
“mucky-mucks” (9000). She said that some people carried information between the
groups. The group was “full of very complex relationships” (1301). One stakeholder
thought that it was not a closed group because if someone showed up and wanted to be
involved, they just had to “do the work” (1101).

Interview @estion 8--Are M influenced by the collaborative? The FCC did
influence stakeholders’ work (1301, 1901). Information helped people set program

priorities (1402, 1601) or make referrals (3202). One stakeholder asserted that “money

84

makes one influential” and that he had been known to change his thinking based on what
he heard a grant maker say at the FCC (2101). The collaborative meetings gave those
with resources the chance to exercise a level of influence they would not have without
that platform (1101, 1201). A number of members indicated that they were influenced
when people showed “intensity and passion” about making things better (1901, 1001,
3202)

Interview Question 9-- at do you think of the sociograms? Overall,
stakeholders expressed frustration with the sociograms (1601, 1701, 1103, 1901).
Although they were sure that they did not want names in the pictures, they did not think
much could be learned from the pictures without names (1201 , 3101, 4062). One
stakeholder, despite frustration, pointed at the sociogram and remarked, “Those are the
lines of our experience.” 1

Application of the damp the research questions.

Question 1: How does the concept of social capital relate to members’ ability
to build, maintain, and access resources in a collaborative network?

Social capital did relate to stakeholders’ ability to build, maintain, and access
resources. Four themes emerged from the stakeholder interviews:

1. Social capital resides in relationships of trust, where individuals are considered
by others in ways that protect them and advance them because those others have at some
level internalized the individual’s well-being.

2. Building social capital requires some measure of proximity and experience,
which, in turn gives members access to resources.

3. People select others because they are “like themselves” or because they are not

“like themselves” and can, therefore, offer information, power, or other similar resources.

85

Selection is based on opportunity (either past experience, intuition or mandate), the
nature of business, and the capacity to build reciprocity and trust.

4. The collaborative influences members’ work with information and access to
each other. However, the group has no formal method to sanction those who are not
influenced. Informally, those who are not influenced are not integrated into the collective
work. To be a part of the collaborative vision, one must make a contribution beyond
one’s own place.

Question 2: How do members respond to social capital, network concepts,
and the research process? Does action result?

Generally, by this point, stakeholders were struggling with the intellectual labor
involved in the study. The concepts were relatively new to people, the tools were
complicated, and the time required to consistently engage in the work was unavailable.
The sociograms were limited by anonymity and, at this point, the researchers were seen
as equally indecisive. Group members remained interested in the effort but were not
engaged in a reflective process. There was no method to apply the research results in a
practical way. Also, stakeholders felt a tension between not wanting to disclose names

(network position and contacts) and yet wanting the study to be informative.

Point 6: Sociogram/Interview Triangulation

At this point, the results of the survey and sociogram process were combined with
the results of the key informant interviews to verify and explain social capital within the
collaborative network. It is important to remember that key informants were interviewed
without any knowledge of their position in the sociogram.

The triangulation method framed the insights gained from both the interviews and

stakeholder meetings within the social context that the stakeholders themselves defined.

86

The “Importance Ties” sociogram (Figure 4.2) is the context within which the following
verification and explanations emerged.

Group A: One key informant (3202) was interviewed. This group was positioned
at the bottom left of the network, with a limited number of ties to other groups. That
position would lead one to expect members to play an entrepreneurial or niche role in the
network. The interviewee described herself as being slightly “detached” from the FCC.
She maintained relationships with people of similar responsibilities so that resources
could be mobilized in an emergency. She also said that, although those kinds of
connections could get results for people, systems change makes people run back home.
PeOple also run home when their house is under attack. Her system had been under such
attack for the past few years, which may also have contributed to her position outside the
center of the network. The connections recorded on the picture were consistent with her
assessment.

_G_rpu_p_B: One key informant (4062) was interviewed. This individual said she
saw her job as being visible and found the FCC network useful for that purpose. She was
confident that she could identify and access the leadership of the community because it
was important to have that access. She was positioned in the network with connections to
grant makers, but it should be noted that the ties to grant makers were not identified by
individuals, just by organizations.

Gr_orrp_§: One key informant (1402) was interviewed. This group was very
interrelated. The individuals and organizations were involved with each other as
contractors or board members. Many of the ties around this group came from the
organization (1400) or this key informant. Group C was filling a structural hole in the

network (Burt, 2001), having connections to all the other groups. This informant

87

indicated that she had become involved with the FCC as an opportunity to align her
organization with people who made decisions. Her ties were to individuals who managed
contracts or grant firnds in organizations.

Group D: Four key informants (1103, 1601, 2200, 2101) were interviewed. One
informant (1103) described her association with the FCC as being dependent on the role
of others in her organization. She was involved to “do the work” of her organization, not
the community’s work. She said she relied on relationships developed long ago. She
made it a point never to lose a connection. Her position in Group D was consistent with
her interview.

Another informant (1601) in Group D defined his job as getting additional
resources into his organization. He built relationships with people by working with them
and finding common interests. He believed that collaboration meant you “sit down and
figure stuff out” with others. He had strong ties in a group of people who did similar
work and had many resource-transferring weak ties throughout, the network.

Another informant in this group (2200) was not identified by name, but her
organization was positioned in this group. In her interview she said that she selected
peOple who could help her further the organization’s mission. She had an active and
strong board, which constituted her primary network. This would explain her position
somewhat outside the network center.

The other informant from this group (2101) asserted strongly in the interview that
at the FCC he was in the business of parlaying relationships into resources for his
organization. That would explain the number of ties involving him. He said that his job
was to sell himself. When people began to trust him, they would think of his

organization.

88

MILE: No key informants were interviewed.

M: Five key informants (1701, 1301, 3101, 1201, 1503) were interviewed.
One of them (1701) thought that in his work he needed to be involved with a number of
agencies that, upon review, surrounded him in the sociogram. His ties did not extend
beyond that group, which allowed him flexibility. In the interview, he said that was
important.

Another informant (1301) in this group also considered flexibility important. He
indicated that he “shared” on a need-to-know basis and selected people based on clear
expectations. He said trust required that people communicate directly, which was
consistent with the direct ties shown in the sociogram.

A third informant in this group (3101) said she needed flexibility for protection of
the point of view she represented at the collaborative. Experience had taught her, she
said, whom her philosophy “meshed with” and whom it did not, so she picked her time
and people based on presenting circumstances.

A fourth informant (1201) from Group F said his role with the FCC was to
support innovation and to encourage his staff to form relationships outside the
organization. He and his organization were central in the network, which would be
consistent with his expressed goal.

The fifih informant (1503) indicated an interest in both flexibility and supportive
connections. He said that he did not include people in his network just because he “had
to do it.” In some situations, “you still hold your cards close” because you do not trust
the person will consider your interests as their own. He thought his relationships had

developed over time through involvement with the FCC.

89

Group G: One key informant was interviewed. This individual said his job was
to connect to people who were not like himself, to learn from people and connect them.
His position in the network verified that bridging role. He said that if people were going
to talk to you and tell you things, they had to see you. He was the central connector to
most other network groups from his subgroup, which was consistent with his ascribed
goal to be out in a variety of settings.

Group H: Two key informants were interviewed. One of them (2302) said her
role was to provide the best specific services possible in the community. She indicated
that it was critical that she not duplicate the work of other organizations. Further, she
said she picked her community partners very carefully, which would explain why most of
her ties in this network were with individuals in her organization or with grant makers.
She indicated that she did not see “value” in community collaboration unless it was
around providing specific services to fill an existing gap. Her organization and her
expressed philosophy resonated with her position in the “Importance Ties” sociogram
and that of her organization. They rested on the outside, with ties to specific grant
makers or big systems.

Another informant (3 001) in this network said she “picked partners” carefully.
She described the FCC as a cocktail party to which she could go and pick whom she
wants to work with based on personal priorities. She explained that much of her work
was “non-F CC” work. She would give preference to FCC members with whom she had
made a personal connection on either FCC or non-F CC work. Her position in the
network was consistent with this assessment. She was in a subgroup with individuals and
organizations that did similar work, but she had ties to a variety of peOple she had listed

as personal connections that had developed into resource connections (a connection to

90

law enforcement, for example, that started as a personal conversation and developed into
a community strategy on substance abuse).

Group J: One individual was interviewed. This informant indicated that the FCC
work (collaborative problem solving and community development) was her work. She
said she had passion for the vision. Her position in the network was central, and the ties
gave her great Opportunities for influence and obligations. Her position was consistent
with her expressed central role in the collaborative.

Group(s) K g I: No one was interviewed.

One informant did not appear on the sociogram as a chooser or a chosen person or
organization. This might be because she did not complete a survey or submitted one that
was not identified. For analysis purposes, she was coded 9000. She said in her interview
that her role in the FCC was to be a “carrier pigeon” between the “mucky-mucks” and the
“doers.” This she considered an important role that she could perform because her
agency served a particular niche and was not competitive with other area agencies for
resources. She indicated that her relationships with other FCC members had expanded
when she was involved in projects that gave her personal connections with others.

Application of the data to research qpestions.

Question 1: How does the concept of social capital relate to members’ ability
to build, maintain, and access resources in a collaborative network?

The data triangulation verified that stakeholders in this network were using their
network relational ties to build, maintain, and access resources. Further, the data
indicated some specific ways in which those ties were being used: to parlay a
competitive edge, to support innovation, and to influence outcomes based on passion for

a vision.

91

Question 2: How do members respond to social capital, network concepts,
and the research process? Does action result?

The triangulation of data did not provide data for this question.

Point 7: Stakeholder Meeting: Reactions,
Reflectiorfirand More Questions

The stakeholder meeting at point 6 was conducted shortly after the key informant
interviews were completed because of significant changes in the community leadership
structure. During the study, three of the five grant makers in the county had a change of
executive leadership. This change pushed ahead the reporting of data to the stakeholder
group.

The erosion of the stakeholders’ leadership base brought the potential for
informed action on the study to a screeching halt. A stakeholder meeting was held to
review the study data for stakeholders’ insights, potential action, and direction. The
previous enthusiasm for the research was no longer apparent. The inside researcher was
directed to complete the analysis for future consideration when the leadership had
stabilized and strategic activities could realistically resume.

Faced with the contextual reality, the inside researcher continued with the
reflection-action model by reporting to stakeholder meetings and informing individual
stakeholders of the findings, when they appeared contextually relevant.

Application of the data to research questions.

Question 1: How does the concept of social capital relate to members’ ability
to build, maintain, and access resources in a collaborative network?

The decision to suspend group action based on this work was soundly grounded
on the concept of social capital. The three community leaders who changed in 2002 (one

relocated and two retired) were all key stakeholders in the network as leaders who

92

distributed resources. Such change caused systemic disruption that will require some
time to stabilize. The structural holes created by these changes may be filled over time.

In addition, changes in the economy seriously shifted the financial situation
private nonprofit organizations in the community. A number of organizations indicated
that their relationships needed to change. One director remarked that, with money so
tight, “collaboration is just not going to be worth it for awhile” (2301). Another indicated
that he would not “go looking for new friends,” but saw the collaborative ties he did have
as important. “I could use them before; now I need them,” he said. When asked what he
meant, he said, “Before, we were making progress in new areas. Now we need to figure
out how not to lose what we made.” (1601). One stakeholder (1102), who was promoted
as part of the change, indicated that the network information might be useful in strategic
positioning. First, she was going to “deal with my own organization. Then I will figure
the next piece out.”

Question 2: How do members respond to social capital, network concepts,
and the research process? Does action result?

Although the network study can inform that effort, it is not reasonable to think
that a participatory design can be the method under such circumstances. The stakeholder
group decided that the data analysis should be completed and that they should be kept

informed with updates on what was being learned.

93

CHAPTER V

FINDINGS AND IMPLICATIONS

Restatement of Goals and Objectives

This research was undertaken to examine the relationships between and among
members of a community collaborative network. The study had two purposes. The first
was to learn how members build, maintain, and access resources in a collaborative
network and the relevance of applying social capital as a concept to explain the
relationships. This brought the stakeholders to analyze their relationships to one another
and to the work of the collaborative along a number of conceptual dimensions, including
mandates and proximity, common work, selection, and influence. Social capital was used
as a framework for this consideration. The strength of social capital as the organizing
framework is that it allows analyses at the individual, the dyad, and the network levels of
engagement and it affords, therefore, an opportunity to articulate the dimensions at
particular points.

Next, the distinctions among levels that emerged in these concepts also verified
both stakeholders’ positions in the network and the work they did within it. By analyzing
the interview insights within the importance sociogram, both data-gathering techniques
were validated. This analysis enabled me to compare and contrast stakeholders’
comments, allowing for a contextual interpretation that not only verified each
stakeholder’s position but also explained the position. This study resulted in five major
findings that articulate the relevance of social capital as a conceptual framework for

community collaborative network relationships.

94

 

The second purpose of the study was to pilot the use of quantitative network
analysis techniques and qualitative interviews and observations to inform participants and
encourage their action. How do members of the collaborative respond to information
about relational patterns? Are the network survey results consistent with key informant
interviews? DO members (participants) apply the results in community work? In the last
finding, this study informs the use of a participatory action design in a community

collaborative network.

Maior Findings: Ouestion One

Finding 1: Stakeholder relationships vary and reflect the economic realities of the
community at large.

Two themes emerged from stakeholder meetings and interviews around individual
involvement with other stakeholders and the collaborative network. First, stakeholders
said that the relationships varied in terms of the quality of the relationship involved;
second, external forces as well as personal preference and experience, in part, determined
that quality.

When asked to describe relationships with others, all key stakeholders who were
interviewed indicated that they had a variety of ties with others (as conceptualized in
Figure 1.2, p. 8) and agreed that the quality of the ties could be considered as the granting
of preferential consideration to some individuals and not to others. “I would do for him
because he expects me to and we know that,” one remarked (1601). “I know that he will
consider what will help me, just as he would for himself,” another stated (1503). When
asked how he knew of this regard, the individual said, “We talked about it.” A number of
people described this quasi-formal contracting process as a method to develop

collaborative relationships intentionally between either dyads of members or groups.

95

This type of norm-setting activity is important in-group process and, according to
collaborative researchers, is evident in collaboratives that are considered successful
(Chavis, 1995; Wolfe, 2001; Chaskin, 1999). The norm-setting process is also consistent
with the concept of social capital, in which trust supports the development of norms
(Coleman, 1990). The reward for following norms was described in this study as
increased quality in the ties. “You learn, over time, how a person behaves behind the
scenes and develop trust in the consistency” (1701). However, as regards norms, if one
does not follow them, the opposite is true. Stakeholders indicated that you “renegotiate
the contract” if someone “stiffs you,” either with a conversation (1301) or by
withdrawing from mutual action (1901, 1701).

Stakeholders were forthcoming about the presence and value of a network that
includes a variety of ties. Preferential consideration does occur, and as more than one
person indicated, it is “neither good or had” (1102, 1201, 1601). This variation allowed
the FCC member to “shop for relationships” (2302, 1401), promoting the concept of the
FCC as “a cocktail party” (3001). Variation also allowed members to come and go in the
network. “I connect with the people I need to but always have access to others if I have a
reason to develop a stronger linkage” (1301) was the way one member described the
value; another quipped, “It is one big pool of possible people” (4062, 1503).
Stakeholders were clear, however, that the FCC network is an egocentric one (Knoke &
Kuklinski, 1982; Wasserman & Faust, 1996), where the network is defined from the
perspective of the individual actor. The theme that emerged is that many stakeholders
were positioned with a variety of ties that held the potential to “parlay value” (2101) by

intentionally maximizing the quality of one’s network ties.

96

An individual’s relationship with the collaborative (FCC) often was an economic
decision based on community realities. Some stakeholders indicated that they joined the
collaborative because they saw it as a source of money. The FCC distributes money for
programming, and some said they went to “get at some of that” (1401, 4062). Others
said that grant makers required “collaborative” involvement to access other funding
(9000, 1601). This is more of a contingency involvement; that is, they were involved
because it would eventually “get them resources.” Thus, these two sets of people saw the
FCC as a source of money.

The second group saw FCC involvement as a source of information to make them
more competitive. Some used FCC involvement to sharpen the edges of a niche (2302,
1401), reinforcing Burt’s (2001) concept of structural holes. One individual said he was
involved to expand his sphere of influence beyond his own system and into groups that
normally had nothing in common with him. “Who would have thought,” he said, “that I
can go to a neighborhood meeting for breakfast and the community foundation for their
annual lunch? I would never have been invited before” (1901). Others saw the
information as a rich source of context for their own particular work (1301, 1001). These
individuals asserted that each of the systems (i.e., education, health care, and juvenile
justice) could “better” achieve its “own” outcomes if those were defined in an integrated
community context. They, therefore, came to the FCC for information and linkages that
colored in a richer community picture.

The central theme is that people described their involvement in economic rather
than social or political terms. It was clearly a decision about participation as having a
payback in resources into their systems (2301, 1401, 1601). Stakeholders indicated that

there was not always economic advantage. One stakeholder indicated that community

97

 

work diluted her organization’s position, so she did not work in community collaboration
unless there was direct benefit. She occasionally showed up to get the “lowdown” and
then acted on her own. (2302). Some people said that, in a competitive environment, that
participating in community work is not “cost effective” because it takes so much time
(1701, 4062). Also, directors of small organizations indicated that there must be work
that is done “for me” at meetings. “If I can kill two birds with one stone and connect
with people, then the FCC is worthwhile for me,” one commented (3101). The economic
advantage was, at least for some members, about “getting the work of my organization”
done through the community table. Others thought the economic advantage to a
community table created outcomes that collectively got her organization’s work and that
of others done.

As mapped in the conceptual framework (Figure 1.2), then, the collaborative
network gives structure to a variety of relationships seen as ties between stakeholders in
the network and engagement (involvement) between stakeholders and the network itself.
Within the network structure, various dimensions work for the stakeholders in a number
of different, individually defined ways.

Finding 2: Stakeholder relationships hold actual and potential access to and
distribution of resources. Members get and give resources through the relationships that
form the collaborative network, and the access/distribution process is related to the
individual actor’s position in the network.

Stakeholders reported a full range of resources that were accessed as well as
distributed through the collaborative network, where, because of an individual’s

membership, either information or other forms of preferential access were granted. Being

98

Social Capital

in a Community Collaborative Network

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
   
  
 
 
   
       
 

 

 

 

Network Social Ties :I'i':
'Sharédvalues s—eaames ____SociaITies
-Rec1proc1ty
-Mutual Action
/ l \J Legend
Network Structure .
-Mandates and Proximity 309131
aTmst ties
-Reciprocity
°Selection N k
-Influence etwor

A -Work D structure

Increased '
social capital 1 Connection
Increased social 3:: between
collaboration ties

° Access to
resources
° OUICOUJC l Direction
of ties
Decreased social capital

Decreased collaboration B O _ _

Decrsron

 

 

 

Figure 1.2: Conceptual map.

in a room with public policy makers and private grant makers gives members access.
Because the access is regular and semi-structured, members know they might see
someone else, can plan on that connection, and can use it as a point of familiarity. At the
most basic level, members reported that membership in the network moved you “one step
up from a cold call” (3101), granting an implied level of familiarity. As one member
explained, “It is like if I see you in my neighborhood, I at least know that we have that in
common, even if you were lost.” So, two stakeholders who have an FCC affiliation have

that in common. Should there be interest in something more, the affiliation supports that

99

someone else, can plan on that connection, and can use it as a point of familiarity. At the
most basic level, members reported that membership in the network moved you “one step
up from a cold call” (3101), granting an implied level of familiarity. As one member
explained, “It is like if I see you in my neighborhood, I at least know that we have that in
common, even if you were lost.” So, two stakeholders who have an FCC affiliation have
that in common. Should there be interest in something more, the affiliation supports that
development. The “something more” often does happen, according to the stakeholders.
“You see people around certain values,” according to one member, “and you connect
because you talk about important issues for families” (1301, 1601).

This point of collective work is another important theme in collaborative
relational ties. In order for relationships to develop social capital and collective
consideration, stakeholders need to be engaged in some joint work. “This table is Open
for people who will show up and do the work,” according to one of the FCC’s original
members (1101). That is particularly true when there are disagreements. “You have to
say what you mean, right in the meeting [not outside or when it is over] if the relationship
is going to move forward” (1901). Being able to agree on joint projects and act
collectively was a core theme among stakeholders in terms of relationships. “It is the
only way to build trust [to be able to disagree], and that is core to doing anything between
systems” (1501). “I tell people all the time,” one stakeholder remarked, “the key to long-
terrn relationships in a community is sticking with it and figuring stuff out” (1601).
“Figuring stuff out,” according to this stakeholder and others, are not just conflict
resolution but also more often a process of mediation. Through the discussion process,
commonalities are chiseled from what may have appeared “rock-hard” positions (1503,

3101). This type of process builds trust. According to key informants, then, common

100

 

 

work is about outcomes (outcomes that have value to either individuals or the collective)

and an engaged process.

Finding 3: There are distinguishable differences in each dimension of the

collaborative network across the individual, the dyad, and the network levels, and all

levels Operate at any point in time in a community collaborative network (Table 5.1).

The three levels can be interpreted as “levels of collaborative engagement” as

well as levels of social capital. It is at the network level that the expressions of trust,

reciprocity, and work (mutual action) reach the level conceptualized by social capital

theorists (Bourdieu, 1995; Coleman, 1994; Portes, 1998). Table 5.1 summarizes the

themes that emerged from this study by dimensions and levels.

Table 5.1: Collaborative dimensions.

 

 

 

 

 

 

 

 

 

 

 

 

Individual Dyad Network
Mandate Involved to meet Involved to meet Involved to meet collective
organizational mandates/access for mandates and access to
mandates/access certain others resources
resources
Proximity Access to “Tell My Access to hear/see others Access to create a new story
Story” to @sition my story
Trust Consider “myself" Consider certain others Consider others as I would
Confidence in “myself” Confidence in certain consider myself
others
Reciprocity I will do for you if you do I will do for you if you do I will do for us
for me for me or our mutual goals
Selection I select to inoculate I select to expand myself I select because you are like
nyself me/not like me
Influence I am not influenced Certain others influence Others thoughts change my
my work behavior
Work I do my work I do my work through you We do our work
I will do my part I will do my part to help I will redefine my part
you

 

101

 

Finding 4: The emergence of the levels of engagement indicates that not only is
there a relationship between social capital and collaborative work, but that the
relationship can gain momentum and expand over time or can be diluted and lose
momentum over time. Take trust, for example. Informants explained how experience
with others and opportunities to witness them in various settings and with various others
facilitated the development of trust. It would seem, then, that the more such
opportunities are presented, the higher would be the network’s levels of trust. That is not,
however, the case, according to the stakeholder interviews and meetings. There is
another dimension that influences relationships and work. Some stakeholders did report
trust building and relationship building based on the experiences previously described,
but others indicated the opposite. “I used to think he was a jerk,” one stakeholder said
candidly, and “then I have to go to meetings and find out that for sure he is” (1101). The
study results indicated that the collaborative process does not naturally develop levels of
social capital but that the process provides an infrastructure for potential development.
The nexus where the work and the relationships come together with resources and
outcomes will take one of two trajectories. Either the relational path will build and
regenerate the system, or it will exit the system. An individual will either engage in an
increasingly collective effort or will go elsewhere.

Collaborative researchers would concur with this study’s finding that networks
are made up of many relational ties (Bruner, 1995; Bruner & Chavez, 1996; Chaskin,
1995). The problem for community collaboratives is that if social capital develops, it
might close the network to the variety that many members prefer. The challenge, much
like Frank (2002) suggested, is that social capital is dynamic in nature. Collective

energy, created by the relational ties and action, spirals the network actors’ individual

102

actions toward collective actions. This convergence of individuals to a collective point
ejects those who are not congruent with the collective force. It can close the system.
That closure eliminates the strength of a network that has a high level of connection but
balance and symmetry that allow for movement and inclusion (Knoke & Kuklinski,
1982). There is a danger that the cohesion of a particular subgroup could spiral the FCC.

Stakeholders made numerous references to the diversity of the collaborative in
terms of relationships (Figure 5.1) and types of work (Figure 3.2). There was
organizational diversity in the subgroups of the subgroups (Figure 4.3). However,
insider/outsider distinctions were also made (4062, 1001, 3101). Stakeholders who
described themselves as “at the core”(1001) or on the inside (1503) were the stakeholders
who also described themselves as coming to the table to “lean in” (1901) or to adapt
“individual protocol for collaborative practice” (1503). All six of these individuals
described the FCC work with the same words--“come to the table and give more than you
take”--and all said some others agreed with them and some did not. They all asserted that
they knew who was who.

A second group of stakeholders described themselves as “outside” (2301) or as
emerging (4062) and saw the collaborative as the point of access to the “mucky-mucks”
for the “doers” (9000). All seven of these peOple believed that the FCC helped them do
the work of their organizations. They did not indicate that they were being excluded
from inside involvement but that they did not need to be inside.

The collaborative did have distinct groups, and both reported being comfortable
with the lack of cohesion with the entire membership. There is no danger of this system

closing around a single collective vision, which can create an entirely different danger.

103

The literature on collaboratives does indicate that for organizations to move
forward successfully requires that there be a single, collective vision (F oster-F ishman et
al., 2001). The stakeholders did not have just one collective vision, but rather two.

Finding 5: In interviews and meetings, stakeholders verified how they
interacted within the collaborative network and explained about those interactions. When
interpreted with the actual position the stakeholder occupied on the sociogram, a number
of roles emerged. In Figure 5 .1, the stakeholders are identified in the subgroups where
they appeared in the “Importance Ties” sociogram. By comparing and contrasting the
interviews of these stakeholders, the study findings explain how these stakeholders

explained their interaction with the network.

9000

 

Figure 5.1: Position of key stakeholders within the importance sociogram.

104

The Bridgers (this includes 1901, 1601 and 1402) stakeholders use weak ties to
access resources that can be brought back to group of strong ties (Granovetter, 1973:
Knoke & Kuklinski, 1982). These individuals explained in interviews that their jobs were
to “connect the people in their system with other systems” and can be seen in Appendix
D-l. They talked specifically about accessing resources and bringing them back into their
system.

The Gatekeepers (1001) mandated obligations, positioned in central location of a
subgroup where systemic resources are obligated by public requirements.

The Flexors (1301, 1201, 1701, 3001, 1503) are not mandated into the system or
the work but add to the work/to the collaborative as value added (either the FCC adds
value to their efforts or their efforts add value to the FCC) there to be value added into
the system. They reported wanting fluidness and flexibility in their relations.

The Players (3202, 1003) are outside because their work is not “directly” related
to FCC work. They retain connections but define their collaborative efforts within their

organizational context.

Major Finding: Ouestion Two

The value of participatory action research in this study was that it gave a
conceptual framework to a previously undefined phenomenon. It gave stakeholders a
way to think about and to talk about resource access and allocation within the context of
their relationships. The assumption that social capital is an operational reality in the
collaborative’s everyday existence proved to be true. The assumption that that was an
undeniably engaging process was not true.

From the initial meetings, community stakeholders articulated that the study was

relevant to their felt needs. They wanted a means to discuss their relationships, a way to

105

bring the connections “out in the open” (1901). Throughout the stakeholder meetings and
interviews, there was a consensus that the sociograms were interesting (but challenging to
interpret) and validated a group sense that they did a number of things. One, the social
ties validated that they were talking to one another and, given the diversity of the
subgroups, that they were communicating across systems. Two, the differences between
the sociograms indicated to stakeholders that the collaborative could mobilize around
critical issues to respond to opportunities. It was clearly proven in this study that,
conceptually, social capital and network analysis are both relevant and informing for
community collaborative members.

The stumbling block for integrating the network study at the collaborative level
was the unwillingness of the group to agree to reveal names and stakeholder positions.
They agreed that there was a personal and/or collective vulnerability to making those
connections public. “I don’t want my reactions or behavior to be predictable,” one
stakeholder said (1301), and his assertion was more common than unique (1001, 2302,
3101, 3202). The network method was, in fact, more concrete than the group was willing
to put out as public knowledge.

The stakeholder initiated a change in their role from interpreters of data to
recipients. They indicated that the change was because of leadership instability (1003,
3202, 1503), priorities (3101, 1402, 4062), and a general sense that, without

interpretation, the data were not useful to the group (1101, 1301, 1901).

How Research Advances Scholarship and Collaborative Practice
This research used network methods to further define the concept of social
capital. It advances the work around social capital at two levels: method and

collaborative practice.

106

Social network analysis was effective in this study to define social capital along
conceptual dimensions. In the study, stakeholders were asked to identify their personal
networks, which created a concrete framework in which those same stakeholders could
describe and, in so doing, validate and explain their work. Then, the triangulation of the
data was a useful data-analysis technique. The sociograms were used as a physical
structure on which interview comments and subgroup perspectives could be mapped.

The process advances the concept of qualitative/quantitative data analysis by
validating the quantitative calculation of relative distance created in the sociogram with
the quantitative responses of key stakeholder interviews.

At the collaborative level, this research has advanced practice in two ways. It
asserts that variations in relationships provide a balanced and rich network for
community collaboration and that such a network distributes resources through social ties

among members, increasing the network’s capacity to mobilize to access additional

resources.

Influence on Personal Practice
This research has had significant influence on the personal practice of the “inside”
researcher in five ways. One, there was an assumption that the primary work of a
community collaborative “organizer” is to work with building relationship. This study
clearly indicates that social ties are not only the reason stakeholders engage in the work
but they are also the reason stakeholders stay engaged or not. This means that, for the
organizer, the work is much more about process than it is about outcome. The outcomes

will emerge as the group determines, but with no process the outcomes might likely not

come.

107

 

 

The second influence on practice has to do with accepting the diversity of issues
and the subgroups that must exist. The variety of relationships within a collaborative is
the collaborative’s strength. As stakeholders converge around common interests, they
will build social capital and will move forward collectively. To have such movement,
groups need to be able to group around values and goals. If they are given such
flexibility, they will find work to do (1301, 1201). The key for an organizer is to be open
enough to allow the groups to emerge from the work rather than to predeterrnine them.

Each stakeholder has a reason for his or her position within the collaborative
network. This is the third finding from this study that will influence community
organization practice. There is a predisposition in collaborative work to believe that
those who come to the community table to do the “community’s wor ” are more valuable
than those who are interested in their organization’s work. This study verified that both
foci could exist within the same network.

The fourth influence from this study is that community organization is about
learning. This study verified the "reflective-action—reflective” process defined by Freire
(1970). The changes in design reinforced the dynamic nature of knowledge and
reinforced the study’s preliminary framework that the knowledge would emerge with the
stakeholder’s willingness and need to know.

The fifth influence on personal practice from the study is a refining of the
definition of collaborative work. For a number of the members of the FCC, their
involvement is mandated in that they are required by either state departments or by grant
makers to be involved at “some” level. That level may be just involved enough to secure

appropriate documentation when applying for funding or some more advanced

108

involvement (i.e. primary firnder). This mandated element separates the collaborative

effort from a completely voluntary group that has formed around an issue or activity.

Research Implications

The results of this study have implications for those who conduct participatory
action research. The stakeholder group was selected by the Executive Committee of the
FCC and the researchers because of the group’s interest in learning about the
collaborative relationships. The stakeholders were engaged in the work, participating
with the researchers in defining questions and concepts until the presentation of the
sociograms. It was at that point that a level of cognitive dissonance developed for
members of the stakeholder group. Comments like “I must have missed the meeting
where we learned about this” (1601) were common. It was, however, difficult to
determine how much the apparent disconnect was because of an inability of the
researchers to articulate the pictures and overarching themes and how much was a lack of
relevance for the stakeholders. The group directed the researchers to report back with
findings. For applied researchers interested in conducting meaningful action research,
there is a need to more carefully consider how data can be shared in a way that facilitates
stakeholder involvement rather than frustrates it.

Replication of this study, with input from stakeholders on how the data can be
presented and explained, would expand the value of this type of study as an action-

generating process.

Policy Implications
This research has an implication for public policy. Simply stated, collaboration

builds relationships between community actors and supports a community infrastructure

109

for mobilization. The research further demonstrated that public mandates and funding
opportunities helped bring stakeholders to the community table, where, with experience,
social capital can be built. State policies, including the executive orders in Michigan in
1994, are instrumental in bringing community stakeholders to collaborative tables. This
research is documentation that such community tables result in organizational integration
and movement toward a collective vision, movement that might not have happened
without the incentive of grant makers and state government.

This study could be used to support an intentional infrastructure of relationships
in communities between individuals representing various organizations. These
relationships would be intentional, as they would be established for the purpose of
building social capital between these individuals. As this network study showed, not all
relationships or networks of relationships have social capital, but when certain
dimensions are present, social capital can form. Policy leaders would do well to consider
the strength of such a relational infrastructure when seeking to solve complex community
problems.

Community collaboratives do not function in horizontal isolation. The member
organizations include public county agencies that are part of a state hierarchy, private
organizations with pubic missions that are shared by statewide associations and
stakeholder groups. These agencies and organizations are, therefore, influenced by others.
The impact of these vertical structures was not accounted for in this study. The
community collaborative network members all belong to some other networks that affect

them and their collaborative relationships.

llO

 

 

Questions for Future Research

Two questions emerged from this work. The first is, what would happen if this
study were conducted in another collaborative? The second is, what will happen to the
FCC members when the concepts learned in this study are reinforced over time?

As for the first question, when this study is replicated it is likely that similar roles
would emerge in the collaborative. In addition, it is likely that those roles are filled by
individuals (and organizations) that are structurally similar to those found in the FCC. It
will be important to look at how individuals position themselves, however, and if those
positions tend to be based on opportunities for resources as was the case in this
collaborative. The stakeholders in this collaborative indicated that their organization
generated in part from mobilizing to access resources. It could be implied that such
would be the case in other situations, but it is not clear.

Second, given the leadership change with this stakeholder group it is important
that the FCC be observed over time. Such study would be useful in determining how
learning about social capital and the FCC social network might influence the stakeholders

and their efforts.

Benefits g1 Limitations of Insider Research
The benefits and limitations to being an insider in a research project can be
summed up with one concept: social capital. An insider researcher can use relationships
to gain access to individuals. In this study, stakeholders indicated that participation was
in some part motivated by a willingness to help the inside researcher. There was an
expressed sense of mutual benefit. That consideration of the research as shared work is a
key dimension in relationships with social capital. The insider brings that dimension. In

addition, the inside researcher brings a contextual understanding that can push the work

111

from inside the process. In interviews, for example, the researcher had a number of
shared experiences with the stakeholders so could ask informed, direct questions
capitalizing on opportunities that someone without the context would likely have missed.

Social capital is also a limitation for an insider. As documented in this study, it is
not likely that an individual has the same level of engagement with all potential
stakeholders. That means, then, that some people will talk clearly and frankly and some
will not, or at least might not. This study of social capital in a community collaborative
network was designed with this limitation in mind. Triangulation of the data neutralized
the risk of the insider’s bias. Stakeholders were given the opportunity in interviews to
explain their collaborative involvement and the surveys also documented that

involvement.

112

APPENDICES

113

APPENDIX A

Delivery of Community Services
in Muskegon

We are asking you and other community service agents to respond to the
following survey. We are trying to learn about the processes through which peOple share
information and help each other to delivery community services.

We are not here to address the effectiveness of the Family Coordinating Council.
We can use the results of the research to help the community implement innovations while
cultivating a sense of ownership. Background and organizational openness questions have
been included to allow researchers to evaluate the relative importance of social ties to
those factors.

We plan to collect the following data as part of this research project:

1. Field notes based on interviews and working sessions.
2. This 20 minute survey regarding networks, capacity to deliver services, identification
with Muskegon, and background information.

Completing this survey indicates your consent as a participant in this study in so
far as your responses will be analyzed. Participating in this study is voluntary, and we
will keep all data collected confidential. Your privacy will be protected to the maximum
extent allowable by law. Given the size of the Muskegon networks, however, it may
be possible to determine your identity.

How much will we protect confidentiality? A full report of our results will be
given to the Family Coordinating Council including identifying information if the
Council agrees to such distribution. Results used for research and for use by the general
public will not include identifying information. Note: Nothing will be published from
your response to this survey until 2002 or without the consent of the FCC.

You may contact Ashir Dumar, M.D. chairperson of UCRIHS (the human
subjects review board) at Michigan State University (517-355-2180), in case you have
concerns or questions about your rights in participating in this human subjects research.

After completing the survey, please return it to Ken Frank at the address listed
below. If you have any questions or comments, please contact either of us at the
addresses provided below. I appreciate your taking the time to respond to this survey.

Mary B. McDonald
460 Erickson Hall Muskegon, MI 49440
East Lansing, MI 48823 Phone: (231) 722-4538
Phone: (517) 355-8538 Fax: 517-353-6393 Fax: (231) 722-4616
E-mail: kenfrank@msu.edu E-mail: mmcdonald@cffmc.org

114

9.

APPENDIX B

Key Informant Interview Protocol: Interview Questions

“You and I know each other. How might that relationship influence your
participation in the interview or your remarks”

Probe: does the participant indicate that the relationship holds any

social capital...reciprocity/access to resources/value of mutual

action/trust
Please briefly describe your involvement with the collaborative including
how you began and how your involvement might have evolved?

Probe: why involvement began/who might have influenced the

involvement/how contacts were made/how were “rules learned”

How are rules enforced? What if someone violates a norm?Give example of
type of rule and exactly what you mean by “enforcement” How about asking
- what happens if someone violates a norm, and give example of a norm?
Please describe your relationships with various collaborative members.

Probe: compare and contrast...subgroups...type of ties
How do you select whom you interact with?

Probe: what makes people attractive/is there different types of access?
Do you have social capital with certain members of the collaborative?
(Defined as resources that you may access through social ties (Frank and
Yasumoto, 1998) and reciprocity, trust, adaptability and flexibility among
partners where there is mutual benefit (Gittell and Vidal, 2000)

Probe: what resources can you access through the relationship?

Stories that give example of trust/reciprocity/value
Are there people who are central to the work of the collaborative? Are there
people who are marginal?

Probe: does centrality vary depending on kind of work? Do some

people hold the same spot in all situations?

Are you influenced by the collaborative personally or in your job
performance? '

Probe: would you change your mind because to agree with the
collaborative or because of how your decision would influence the
collaborative?

What do you think of the sociograms? Are there cliques?

Do you want to cover in the interview questions regarding the value of using the

sociograms, or will interviews be conducted before viewing sociograms?

Do you want a question about clusters or cliques in the network?

115

APPENDIX C

To: FCC Program Committee
Annette Abrams
Ken Franks

From: Mary Mc Donald

Re: Network Project

Date: September 10, 2001

Dear Friends,

The attached memo dated September 10th was sent to me by Ken Frank. It
summarizes the initial Network Project design questions. I have met with Ken
and he has agreed to come to met with us on October 9th at the end of the
regularly scheduled FCC Program meeting. We both agreed that an hour to an hour and a
half conversation would bring initial agreement on a timeline for the work,
communication methods, schedule and other parameters.

Consider the questions and issues noted in the memo fi'om Ken. In doing so, I ask
you to think about how we will define our network. Do we want to select a list of
individuals (or organizations) and then ask them to add others to insure the network is
discovered? Do we want instead to select an issue (such as emergency needs) and
discover the network that responds to that issue? Another option is to select a particular
program and focus on the network that defines that program.

Three questions can help when thinking about networks. First, what individuals
(organizations) have resources that are brought to bear within the network? Second, who ‘
has an opportunity for action and how flexible is that opportunity? Third, who is aware of
the resources and action within the network?

The purpose of this project is to use the knowledge of our social network
(structure) to inform the work of the FCC by creating a social map that documents the
ties among individuals and organizations. Our relationships, our values and our
identification as a group will then be by intention rather than chance. Or, more precisely,
when they are by chance we will be able to document them accordingly. In addition, the
map will be a baseline for comparison as the collaborative process moves forward in
Muskegon and is consistent with the commitment to review relationships between
agencies made at the Children’s Summit.

Please feel flee to give feedback on the ideas presented here prior to the October
9th meeting. You can call/email me or can email Ken at 1W.

Let me hear from you and please plan to meet on the 9th fi'om around 3:30pm to
around 5PM. Thanks.

116

September 10, 2001

First, we’d like to thank the members of the FCC again for agreeing to participate
in the study and to work with us as we conduct the study. We’re looking forward helping
the people of understand, access and extend relationships among community
service agencies.

In some form or another, the study will include a survey of people and agencies in

The survey will include questions about relationships with other agencies,
the basis of those relationships, etc. The exact questions will be determined as we go.

We have a number of questions and issues we’d like to raise as we begin the study
and hope that you, the members of the Program Committee, will help us address them.

Research questions:

How are resources (e.g., information financial, physical etc) transferred and allocated
among agencies?

How is the allocation of resources affected by underlying and stable relationships/ties?

How does this distribution affect the delivery of services to the community?

1. Participation and respondents

Who should be in the study? Should it be just members of the FCC? The transfer
and distribution of resources depends on relationships in a whole system. You can’t
understand fully why one agency shares with another unless you know how many
partners they have in common, where they reside in a larger social structure, etc.
Therefore we need to study a well-defined system. Any agency or person who has
extensive impact on the system, even if not from . . should be considered for
inclusion.

Question for the group: Can we assemble a list of agencies and people that
captures most of the action? Who should we ask to confirm and extend the list?

2. Agencies versus people

Should we study agencies or people? Our goal is to understand why and how
resources are allocated between agencies. Ifallocation is a function of personal contacts,
then we should be studying people. Ifit is a function of institutionalized relationships
between agencies, then we should study agencies. Perhaps we should study both?

Question for the group: How much is the allocation of resources a firnction of
specific personal ties between members of the agencies?

117

3. Types of relationships to measure

A) Allocation of resources among agencies or peOple. This might include who
helped whom in terms of loaning S, providing information, etc.
Question for the group: What are the most currently active forms of

helping/resource allocation? Is there a current program around which many agencies are
coordinating and sharing resources?

B) Stable ties

The key to our analyses is to link the immediate and current allocation of
resources to some underlying pattern of stable ties, with the goal of using the stable ties
to predict the next set of resource sharing. When we focus on peOple, ties might Consist
of who went to school with whom, who belongs to the same church, etc.

Question for the group: What are the bases of stable social ties among people?
How do we know who is connected to whom in a way that would affect resource

allocation? Is there a source of information about historical and institutionalized ties
among agencies?

4. Link to service delivery and neighborhoods

Knowing relationships among agencies and how they allocate resources to each
other doesn’t directly inform service delivery. Do we need to measure service delivery
and links to neighborhood groups?

Questzons for the group: How would we study service delivery. Is there existing
data and surveys? Do they contain identifying information to make the link?

5. Use of the results

A) The first use of the data might in a feedback discussion oriented around our
construction of a map of the social structure and recent transfer/allocation of resources.
Where are there gaps? Are they sensible and necessary, or should they be filled? How
would these gaps affect fixture coordination among the agencies/people as a system?

Question for the group: How do you see using the data, hypothetically or actual
application? What information would be most strategically informative?

B) Modifying relationships

We would like to return to assess how the social structure has changed, and
whether knowledge of the social structure has helped members of the system act more
deliberately to modify relationships.

Question for the group: What information would be indicative of changes in
relationships?

5. Revealing information/confidentiality

The more information about specific people and agencies that we reveal, the more
usefill will be the data to the group. But this will also compromise confidentiality.
There must be agreement on the maximal amount of information that can be revealed

118

about any one person or agency. Even if everyone agrees, how will this level of feedback
affect the validity of the responses?

Question for the group: How much information should be revealed in reporting
back?

6. Timeline ,
Once we have established the scope of the study, we can talk about various
timeline needs, from implementing current programs to coordinating action for new
programs to reporting at conferences.
Question for the group: what deadlines apply, what time goals should we have?

These questions are meant just to start the conversation. Many others will emerge as we
construct the study.

7. How should we communicate?

These questions are meant to start a conversation. How should we get feedback?
Should we exchange some e-mail, with the idea of setting up a meeting time to work
together as a group and finalize some decisions?

MICHIGAN STATE
11 N I v E R 51 T Y

 

January 4, 2002 APPENDIX D
TO: Kenneth FRANK '

115 Erickson Hall

MSU

RE: Ines 01.325 CATEGORY: EXEMPT 1-C,D,E '
' APPROVAL DATE: December 31 2001

TITLE: INDIVIDUAL AND ORGANIZATIONAL NETWORKS AS DEFINED BY FOUR
COMMUNITY SELECTED CRITICAL QUALITY OF LIFE RISK FACTORS

The University Committee on Research Involving Human Subjects' (UCRIHS) review of this
project is complete and I am pleased to advise that the rights and welfare of the human
subjects appear to be adequately protected and methods to obtain informed consent are
appropriate. Therefore the UCRIHS approved this project.

RENEWALS: UCRIHS approval' 15 valid for one calendar year, beginning with the approval date
shown above. Projects continuing beyond one year must be renewed with the green renewal
form. A maximum of four such expedited renewals possible. Investigators wishing to continue _a
project beyond that time need to submit it again for a complete review.

REVISIONS: UCRIHS must review any changes' In procedures involving human subjects, prior
to initiation of the change. If this is done at the time of renewal, please use the green renewal
form. To revise an approved pretocol at any other time during the year, send your written
request to the UCRIHS Chair, requesting revised approval and referencing the project‘s IRB#
and title. Include in your request a description of the change and any revised Instruments,
consent forms or advertisements that are applicable.

PROBLEMSICHANGES: Should either of the following arise during the course of the work,
notify UCRIHS promptly. 1) problems (unexpected side effects, complaints, etc.) involving
human subjects or 2) changes in the research enVironment or new lnforrnation indicating
greater risk to the human subjects than existed when the protocol was previously reviewed and

approved.

 

OFFICEOF
RESEARCH If we can be of further assistance, please contact us at (517) 355-2180 or via email:

AND UCRIHS@msu. edu. Please note that all UCRIHS forms are located on the web:
http. [lwww. msu. eduluser/ucrihs

GRADUATE
STUDIES

university Committee on
Research Involving
Human Subjects

Michigan Slate University . ’
2.46 MDIHISII'ZIIOH Building UCRIHS Chair

East Lansing. Michigan
48824-1046

517/355-2180

FAX: 517/353—2976

Web: WNW/1mm
E—Maii: unrihsOrIIsuedu AK; bd

CC: Mary McDonald
Human Ecology Building

 

The Michigan Slate Unhusly
I054 is mar/ram! Diversi):
Bee/lance In Actim.

MSU is an Jimmie-scrim,

equal-opportunity hstitualm.

120

 

BIBLIOGRAPHY

121

REFERENCES

Andrews, M., Bubolz, M., & Paolucci, B. (1980, Spring/Summer). An ecological
approach to study of the Family. Marriage and the F amilv Review,3(1): 2946.

Ashman, D., Brown, L. D., & Zwick, E. (1998, Winter). The strength of strong and weak
ties: Building social capital for the formation of governance of civil society
resource organizations. Nonprofit Management and Leadership, 9(2), 153-171.

Bartunek, J. M. (1993, October). Scholarly dialogues and participatory action research.
Human Relations, 46(10), 1221-1231.

Bartunek, J ., Foster-Fishman, P., & Keys, C. (1996). Using collaborative advocacy to

foster intergroup cooperation: A joint insider-outsider investigation. Human
Relations 4(6), 701-731.

 

Bartunek, J. M.,& Louis, M. R. (1996). Insider/outsider team research. Thousand Oaks,
CA: Sage.

Bearman, P. (1997, March). Generalized theory. Americgn Journgl of Sociology. 102(5),
1383-1415.

Benson, P. L. (1997). Allfids are our kids. San Francisco: Jossey-Bass.

Benson, P. L., Leffert, N., Scales, P. C., & Blyth, D. A. (1998). Beyond the village
rhetoric: Creating healthy communities for children and adolescents. Applied
Developmental Science, 2, 138-159.

Blackman, E., & Smith, S. (2000, Summer). Healthy organizations, unhealthy
communities. Nonprofit Manggement and Leadership. 10(4), 355-373.

Blau, P. (1994). The structural contexts of opportunities. Chicago: University of Chicago
Press.

Bogenschneider, K., Small, 8., & Tsay, J. (1997, May). Child, parent and contextual
influences on perceived parenting competence among parents of adolescents.
Journal of Marriage and Family. 59, 345-362.

Bourdieu, P. (1985). The forms of capital. In J. G. Richardson (Ed.), Handbook of theory

and research for sociology of education (pp. 241-25 8). New York: Greenwood
Press.

122

Brehm, J ., & Rahn, W. (1997). Individual-level evidence for the causes and consequences
of social capital. American Journ_al of Politicpl Sciencp, 41(3), 999-1023.

Bronfenbrenner, U. (1986). Ecology of the family as a context for human development:
Research perspectives. American Psychological Association, 22(6), 723-742.

Brooks-Gunn, J ., Duncan, G., Klebanov, P. K., & Sealand, N. (1993). Do neighborhoods
influence child and adolescent development? American Journ_al of Sociology,
92(2), 69-93.

Brown, L. D. (1993). Developing bridging organizations and strategic management for
social change. Advances in Strategic Mianagement, 9, 381-405.

Bruner, C. (1995). Toward govemment’s role as catalyst: Building social capital—in
disinvested neighborhoods. Des Moines, IA: Child and Family Policy Center.

Bruner, C., & Chavez, M. (1996). Getting to the grassroots: Neighborhood organizing
and mobilization. Community collaborative guidebooks, Vol. 6.

Bubolz, M., & Sontag, S. (1993). Human ecology theory. In P. Boss, W. Doherty, R.
LaRossa, W. Schumm, & S. Steinmetz (Eds), Sourcebook of family theories and
methods: A conceptual approach. (pp. 419-445). New York: Plenum.

Burt, R. (1997). The contingent value of social capital. Administrative Science Quarterly,
4;, 339-365.

Burt, R. (2001). The social capital of structural holes. In M. Guillen, R. Collins, P.
England, & M. Meyer (Eds), New directions in economic sociology. New York:
Russell Sage Foundation.

Butterfoss, F. D., Goodman, R. M., & Wandersman, A. (1996). Community coalitions for
prevention and health promotion. Health Education Quarterly. 23(3), 263 -274.

Catalano, R. F., & Hawkins, J. D. (1996). The social development model: A theory of
antisocial behavior. In J. D. Hawkins (Ed.), Delinquency and crime: Current

theories (pp. 149-197). New York: Cambridge University Press.

Chaskin, R. (1992). The Ford Foundation’s Neighborhood and Family Initiative: Toward
a model of comprehensive neighborhood-based development. Chicago: University
of Chicago, Chapin Hall Center for Children.

Chaskin, R. (1999). Defininmmmunity capacity: A framework and implications from p
comprehensive communng initiative. Chicago: University of Chicago, Chapin
Hall Center for Children.

123

Chaskin, R., Brown, R, Venkatesh, S., & Vidal, A. (2001). Building community
capacifl. New York: Aldin de Gruyter Press.

Chavis, D. M. (1995). Building community capacity to prevent violence through
coalitions and partnerships. Journal of Healthcare for the Poor and Underserved,
6, 234-245.

Chinman, M., Anderson, C., Irnm, P., 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

Communipl Psychology, 24(3), 263-274.

Cigler, B. (1999). Pre-conditions for the emergence of multicommunity collaborative
projects. Policy Studies Review, 16(1), 86-102.

Cohen, D., & Prusak, L. (2001). In good company. Boston: Harvard Business School
Press.

Coleman, J. S. (1986). Social theory, social research and a theory of social action.
American Journg of Sociolgy, 91(6), 1309-1335.

Coleman, J. S. (1988). Social capital in the creation of human capital. American Joumai
of Sociology, 94. 95-119.

Coleman, J. S. (1990). Foundations of social theory. Cambridge, MA: Belknap Press.

Cox, D. (2000). Developing a framework for understanding university-community
partnerships. Cityscape: A Journal of Policy Development and Research, 5(1), 9-
l8.

Creswell, J. (1994). Research design: Qualitative and quantitative apalysis. Thousand
Oaks, CA: Sage.

deGroot, W. T. (1988). A grounded theory for human ecology: A general research design.
In R. J. Bordon, J. Jacob, & G. L. Yound (Eds), Human ecology: Resefichapp
application (pp. 293-303). College Park, IL: The Society of Human Ecology.

Denner, J ., Kirby, D., Coyle, K., & Brindis, C. (2001, February). The protective role of

social capital and cultural norms in Latino communities: A study of adolescent
births. Hispanic Joumial of Behavioral Sciences, 23(1), 3-21.

Edwards, 3., & Foley, M. W. (1998). Civil society and social capital beyond Putnam.
America; Behavioral Scientist, 44(2), 124-39.

124

 

 

Florin, P., Stevenson, J., & Mitchell, R. (1995, October). Empowerment praxis in
community coalition. American Journal of Community Psycholgy, 23(5), 699-
715.

Foley, M. W., & Edwards, B. (1997). Escape from politics? Social theory and the social
capital debate. American Behavioral Scientist, 40(5), 550-61.

Foster-Fishman, P., Berkowitz, S., Lounsbury, D., Jacobsen, S., & Allen, N. (2001,
April). Building collaborative capacity in community coalition: A review and

integrated framework. American Journ_al of Community Psychology, 29(2), 241-
261.

Foster—Fishman, P., Salem, D., Allen, N., & F ahrbach, K. (2001). Facilitating
interorganizational collaboration: The contributions of interorganizational
alliances. American Journal of Community Psychology, 29(6) 875-905.

Frank, K. A. (1995). Identifying cohesive subgroups. Social Networks, 17, 27-56.

Frank, K. A., & Fahrbach, K. (1997, February). Explofingjntenelzuid structures: A
system of longitudinal models of selection and interpersonal influence in a social

network. Paper presented at the annual meeting of the Sunbelt Social Networks
Conference, San Diego, CA.

Frank, K. A., & Yasumoto, J. (1998). Linking action to social structure within a system.

Social capital within and between groups. American Journal of Sociology, 104
(3), 642-686.

Frank, K. A., & Zhao, Y.(2001, October). Subgroups a_s Meso-level Entity in the
Organization of Schools. Paper presented at Charles Bidwell’s retirement
conference, Chicago, II.

F reire, P. (1970). Edmion for critical consciousness . New York: Continuum Press.

Furstenburg, F. (1993). How families manage risk and opportunity in dangerous
neighborhoods. In W. J. Wilson (Ed), Sociology and the public aggnda (pp. 231-
258). Newbury Park, CA: Sage.

Furstenburg, F.,& Hughes, H. (1995, August). Social capital and successful development
among at-risk youth. Journ_al of Marriage and thiamily. 57, 580-592.

Garbarino, J. (1995). Raising children in a socially toxic environment. San Francisco:
Jossey-Bass.

125

Garmezy, N. (1991). Resiliency and vulnerability to adverse developmental outcomes
associated with poverty. American Behavioral Scientist, 34, 416-430.

Geertz, C. (1973). Interpretation of cultures. Princeton, NJ: Princeton University Press.

Geertz, C. (2000). Available light: Anthropological reflections on philosophical topics.
Princeton, NJ: Princeton University Press.

Gittell, R. J ., & Vidal, A. (1998). Community organizing: Buildingysocigapitalapa
development strategy. Thousand Oaks, CA: Sage.

Goodman, R., Wandersman, A., Chinman, M., Imm, P., & Morrisey, E. (1996, February).
Ecological assessment of community-based interventions. Journal of Community
Psychology, 24(1), 33-56.

Granovetter, M. (1973). The strength of weak ties. American Journal of Sociology. 78.
1360-80.
Greeley, A. (1997). Coleman revisited. American Behavioral Scientist, 40(5), 587-594.

Gulati, R. & Gargiulo, M. (1999). Where do interorganizational networks come from?
American Journal of Sociology, 104(5), 1439-1493.

Hofferth, S., Boisjoly, J ., & Duncan, G. J. (1999). The development of social capital.
Rationality and Society, 11(1), 79-110.

Knoke, D., & Kuklinski, J. (1982). Network analysis: Quantitative applications in the
social sciences. Newbury Park, CA: Sage.

Krackhard, D. (1992). The strength of strong ties. In R. Nohria & R. Eccles (Eds),
Networks and organizations. Cambridge, MA: Harvard University Press.

Krackhard, D., & Porter, L. (1985). When friends leave: A structural analysis of the
relationship between turnover and stayers’ attitudes. Administrative Science
Quarterly. 30. 242-261.

LeCompte, M., Schensul, J ., Weeks, M., & Singer, M. (1999). Ethnographer’s tool kit:
Book 6. Walnut Creek CA: AltaMira Press.

Lin, N. (1999). Building a network theory of social capital. Connections, 22(1), 28-51.

Mandell, M. (2001, Fall). Collaboration through network structures for community
building efforts. National Civic Review, 90(3).

126

Marshall, C., & Rossman, G. (1995). Designing Qualitative research. Thousand Oaks,
CA: Sage.

McKnight, J. (1989, Summer). Do no harm: Policy options that meet human needs.
Social Policy, pp. 5-15.

McKnight, J. (1995). Careless society. New York: Basic Books.

McKnight, J. (1997, March/April). A let century map for healthy communities. Families
in Sociegy, pp. 117-27.

McKnight, J ., & Kretzrnan, J. (1993). Building communities from the inside out.
Chicago: ACTA Publications.

Miles, M., & Huberrnan, A. (1994). Qualitative data analysis. Thousand Oaks, CA: Sage.

Mitchell, J. (1969). The concept and use of social networks. In C. Mitchell (Ed.), Social
networks in urban situations (pp. 1-50). Manchester, England: Manchester
University Press.

National Commission on the Role of School and Community in Promoting Adolescent
Health. (1990). Code blue: Uniting for healthier youth. Alexandria, VA: National
Association of State Boards of Education.

Oliver, A., & Ebers, M. (1998). Networking network studies: An analysis of conceptual
configurations in the study of inter-organizational relationships. Organizational
Studies 19(4), 549-583.

 

Park, P. (1999, June). Characteristics of participatory research. Management Learning,
$0), 141-157.

Paxton, P. (2002). Social capital and democracy: An integrated relationship. American
Sociological Review, 67(2), 254-277.

Portes, A. (1998). Social capital: Its origin and applications in modern society. Annual
Review of Sociology. 24, 1-25.

Portney, K. E., & Berry, J. (1997). Mobilizing minority communities. American
Behavioral Scientist, 40(5), 632-44.

Putnam, R. (1995). Bowling alone: America’s declining social capital. Journal of
Democracy. 6(1), 65-78.

127

Putnam, R. (2002). Saguaro seminar: www.cfsv.org/communitysurvey/working together.

Reason, P., & Bradbury, H. (1998). Handbook of action research. Thousand Oaks, CA:
Sage.

Rubin, H., & Rubin, I. (2001). Communig organizing and development. Needham
Heights, MA: Allyn & Bacon.

Sampson, R. J ., Morenoff, S., & Felton, E. (1999, August). Contextual influences on
parenting. Science, 27L 96-133.

Schensul, M., Schensul, J ., & LeCompte, M. (1999). Ethnographer’s tool kit. Book;
Walnut Creek, CA: AltaMira Press.

Schorr, L. (1988). Within our reach: Breaking the cycle of disadvantage. New York:
Anchor Press.

Small, S. (1995). Action-oriented research: Models and methods. Journal of Marriage and
Family, 57(4), 941-969.

Stevenson, W., & Greenberg, D. (2000, December). Agency and social networks:

Strategies of action in a social structure. Administrative Science Quarterly, 45(4),
651-678.

Taylor, M. (2002). Communities that lead: Power, organizational capacity and social
capital. Urban Studies. 37(5/6), 1019-1035.

Temkin, K., & Rohe, W. (1996). Neighborhood change and urban policy. Journal of
Planning Education and Research, 15(3), 159-170.

Tesch, R. (1990). Qualitative research: Analysis types and software tools. Bristol, PA:
Palmer Press.

Uphoff, N. (1992). Learning from Gal Oya, Ithaca, NY: Cornell University Press.

Wasserman, S., & Faust, K. (1994). Social network analysis: Methods and applications.
New York: Cambridge University.

Wellman. B. (1988). Structural analysis: From method and metaphor to theory and
substance. In B. Wellman & S. D. Berkowitz (Eds), Social structures: A network
approach (pp. 19-61). Cambridge, England: Cambridge University Press.

Werner, E. E. (1987). Overcoming the odds: High risk children from birth to adulthood.
Ithaca, NY: Cornell University Press.

128

Wolff, T. (2001). A practitioner’s guide to successfill coalitions. American Journal of
Community Psycholpgy. 29(2), 173-191.

Woolcock, M., & Narayan, D. (1995). Social capital: Implications for development
theory, research and policy. The World Bank Research Observer, 15(2), 225-249.

129

   

l111111111111111