Citation
Interorganizational relationships and nongovernmental organization institutional sustainability in Uganda

Material Information

Title:
Interorganizational relationships and nongovernmental organization institutional sustainability in Uganda
Creator:
Kagan, Jennifer A. ( author )
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
1 online resource (145 pages). : ;

Subjects

Subjects / Keywords:
Non-governmental organizations ( lcsh )
Interorganizational relations ( lcsh )
Interorganizational relations ( fast )
Non-governmental organizations ( fast )
Political science ( fast )
Politics and government -- Uganda ( lcsh )
Uganda ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Abstract:
This thesis examines a network of nongovernmental organizations (NGOs) operating in a suburban community in Uganda, and it explores the relationship between interorganizational partnerships and institutional sustainability. Most of the subject NGOs are members of a formal, locally developed network, called the Nansana Civil Society Network. Through examining both the formal and informal network, we can begin to understand how service delivery occurs through local initiatives and how the shift from first generation NGO work, including relief and welfare, to second and third generation work, involving local self-reliance and sustainable systems development, is occurring. This thesis sets out to understand a network of NGOs and the relationship between interorganizational partnerships and institutional sustainability. Organizations were surveyed regarding eight categories of institutional sustainability (services and programs, people, facilities, technology, finances, records, program evaluations, and human resource systems) and their partnerships with other organizations. Quantitative tests regarding correlations between measures of institutional sustainability and numbers of partners show that there is a stronger relationship between institutional sustainability and the number of times an organization was mentioned as a partner by other organizations than between institutional sustainability and the number of partners an organization mentioned. Specifically, there are significant positive correlations between the number of times an organization was mentioned as a partner and numbers of programs, numbers of beneficiaries, property ownership rates, numbers of technology systems, and numbers of computers. Furthermore, seven measures of institutional sustainability correlate with the strength of organizations' international ties, offering some evidence that institutional sustainability may be a function of international ties. While this suggests that international aid may be efficiently distributed through networks of organizations, it also suggests that much progress remains before local self-reliance and sustainable systems development are achieved.
Thesis:
Public affairs
General Note:
School of Public Affairs
Statement of Responsibility:
by Jennifer A. Kagan.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
868583185 ( OCLC )
ocn868583185

Downloads

This item has the following downloads:


Full Text
INTERORGANIZATIONAL RELATIONSHIPS AND NONGOVERNMENTAL
ORGANIZATION INSTITUTIONAL SUSTAINABILITY IN UGANDA
by
JENNIFER A. KAGAN
B.A., New York University, 2000
JD., University of San Diego, 2004
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado in partial fulfillment
of the requirements for the degree of
Master of Public Administration
Public Affairs and Administration
2013


11
2013
JENNIFER A. KAGAN
ALL RIGHTS RESERVED


Ill
This thesis for the Master of Public Administration degree by
Jennifer A. Kagan
has been approved for the
Public Affairs and Administration Program
by
Brian Gerber, Chair
Christine Martell
Betcy Jose
April 18, 2013


IV
Kagan, Jennifer A. (M.P.A., Public Affairs and Administration)
Interorganizational Relationships and Nongovernmental Organization Institutional
Sustainability in Uganda
Thesis directed by Associate Professor Brian Gerber.
ABSTRACT
This thesis examines a network of nongovernmental organizations (NGOs)
operating in a suburban community in Uganda, and it explores the relationship between
interorganizational partnerships and institutional sustainability. Most of the subject
NGOs are members of a formal, locally developed network, called the Nansana Civil
Society Network. Through examining both the formal and informal network, we can
begin to understand how service delivery occurs through local initiatives and how the
shift from first generation NGO work, including relief and welfare, to second and third
generation work, involving local self-reliance and sustainable systems development, is
occurring.
This thesis sets out to understand a network of NGOs and the relationship
between interorganizational partnerships and institutional sustainability. Organizations
were surveyed regarding eight categories of institutional sustainability (services and
programs, people, facilities, technology, finances, records, program evaluations, and
human resource systems) and their partnerships with other organizations. Quantitative
tests regarding correlations between measures of institutional sustainability and numbers
of partners show that there is a stronger relationship between institutional sustainability
and the number of times an organization was mentioned as a partner by other
organizations than between institutional sustainability and the number of partners an


V
organization mentioned. Specifically, there are significant positive correlations between
the number of times an organization was mentioned as a partner and numbers of
programs, numbers of beneficiaries, property ownership rates, numbers of technology
systems, and numbers of computers.
Furthermore, seven measures of institutional sustainability correlate with the
strength of organizations international ties, offering some evidence that institutional
sustainability may be a function of international ties. While this suggests that
international aid may be efficiently distributed through networks of organizations, it also
suggests that much progress remains before local self-reliance and sustainable systems
development are achieved.
The form and content of this abstract are approved. I recommend its publication.
Approved: Brian Gerber


VI
ACKNOWLEDGEMENTS
I would like to thank my committee members, Drs. Gerber, Martell, and Jose, for
their advice and support throughout my thesis process. Each of them enhanced my final
project and contributed to my experience and education more broadly. I also truly
appreciate the help of Drs. Varda and Sowa, who went beyond their calls of duty by
meeting with me on multiple occasions and helping me in the areas of social network
analysis and nonprofit evaluation, respectively. I would also like to thank the School of
Public Affairs for providing financial support for my thesis and for offering me the
opportunity to create my own path and explore my areas of interest.
There are many people in Uganda to whom I am deeply grateful. First, I would
like to thank Moses for so kindly spending so many days guiding me around Nansana,
setting up interviews for me, educating me, and being patient with me. I truly enjoyed
the time we spent together. Next, I would like to thank Grace for his friendship, for
introducing me to Moses, and for his support throughout my stay in Uganda. I would
also like to thank all of the people associated with Come Lets Dance for their warmth,
hospitality, and friendship. I am grateful to have had such a lovely home in Uganda.
Finally, I would like to thank all of the wonderful employees and volunteers of Nansana
civil society who so graciously took the time to speak with me and educate me regarding
their community and their organizations. I am amazed by the abundance of care and
concern, for both my wellbeing and the success of my research, that surrounded me
throughout my stay in Uganda. I have much to learn from the people there.


vii
TABLE OF CONTENTS
CHAPTER
I. BACKGROUND AM) OVERVIEW...........................1
II. EFFECTIVENESS AND NETWORKS.......................11
III. METHODS..........................................26
IV. OVERVIEW AND FINDINGS RELATED TO SAMPLE..........37
V. NETWORK MAPPING AND RESULTS......................50
VI. INSTITUTIONAL SUSTAINABILITY AND PARTNERSHIPS....63
VII. INTERNATIONAL TIES RESULTS.......................90
VIII. DISCUSSION, LIMITATIONS, CONCLUSION.............102
REFERENCES...............................................120
APPENDIX
A. Survey Instrument...............................124
B. Glossary of Social Network Analysis Terms.......134


Vlll
LIST OF TABLES
Table
III. 1. Measures of Institutional Sustainability..............................29
VI. 1. Comparisons based on Network membership................................64
VI.2. Comparisons based on numbers of out degrees.............................68
VI.3. Comparison of years of operation based on numbers of out degrees......69
VI.4. Comparison of programs based on numbers of out degrees................69
VI.5. Comparison of beneficiaries based on numbers of out degrees...........70
VI.6. Comparison of employees based on numbers of out degrees...............70
VI.7. Comparison of volunteers based on numbers of out degrees..............71
VI.8. Comparison of Nansana facilities based on numbers of out degrees......71
VI.9. Comparison of ownership rates based on numbers of out degrees.........72
VI. 10. Comparison of offices outside Nansana based on numbers of out degrees.72
VI. 11. Comparison of technology systems based on numbers of out degrees......73
VI.12. Comparison of computers based on numbers of out degrees...............73
VI. 13. Comparison of actual budget based on numbers of out degrees...........74
VI. 14. Comparison of budget ratio based on numbers of out degrees............74
VI.15. Comparison of record keeping based on numbers of out degrees..........75
VI. 16. Comparison of evaluations based on numbers of out degrees.............75
VI.17. Comparison of human resource systems based on numbers of out degrees..76
VI. 18. Comparisons based on numbers of in degrees............................78
VI. 19. Comparison of years of operation based on numbers of in degrees.......79
VI.20. Comparison of programs based on numbers of in degrees.................80


IX
VI.21. Comparison of beneficiaries based on numbers of in degrees.................81
VI.22. Comparison of employees based on numbers of in degrees.....................81
VI.23. Comparison of volunteers based on numbers of in degrees....................82
VI.24. Comparison of Nansana facilities based on numbers of in degrees............82
VI.25. Comparison of ownership rates based on numbers of in degrees...............83
VI.26. Comparison of offices outside Nansana based on numbers of in degrees.......83
VI.27. Comparison of technology systems based on numbers of in degrees............84
VI.28. Comparison of computers based on numbers of out degrees....................85
VI.29. Comparison of actual budget based on numbers of in degrees.................85
VI.30. Comparison of budget ratio based on numbers of in degrees..................86
VI.31. Comparison of record keeping based on numbers of in degrees................87
VI.32. Comparison of evaluations based on numbers of in degrees...................87
VI. 33. Comparison of human resource systems based on numbers of out degrees.....88
VII. 1. Comparisons based on strength of international ties.....................91
VII.2. Comparison of years of operation based on international ties...............92
VII.3. Comparison of programs based on international ties.........................93
VII.4. Comparison of beneficiaries based on international ties....................93
VII.5. Comparison of employees based on international ties........................94
VII.6. Comparison of volunteers based on international ties.......................94
VII. 7. Comparison of Nansana facilities based on international ties..............95
VII.8. Comparison of ownership rates based on international ties..................96
VII.9. Comparison of offices outside Nansana based on international ties..........96
VII. 10. Comparison of technology systems based on international ties.............97


X
VII. 11. Comparison of computers based on international ties...............................97
VII. 12. Comparison of actual budget based on international ties............................98
VII. 13. Comparison of budget ratio based on international ties.............................99
VII. 14. Comparison of record keeping based on international ties...........................99
VII. 15. Comparison of evaluations based on international ties.............................100
VII. 16. Comparison of human resource systems based on international ties..................101


XI
LIST OF FIGURES
Figure
IV. 1. Percentage of organizations with particular programs.....................40
IV. 2. Percentage of organizations serving each category of beneficiaries.......43
V. 1. All network ties.........................................................51
V.2. Bonded ties..............................................................53
V.3. The addition of partnerships over three years............................54
V.4. Formal partnerships......................................................55
V.5. Partnerships involving joint programs....................................57
V.6. Partnerships by levels of resource sharing...............................58
V.7. Partnerships by frequency of contact.....................................59
V.8. Network map including financial resources attribute......................60


1
CHAPTER I
BACKGROUND AND OVERVIEW
Development efforts in the worlds South have long been the victim of trial and
error and the subject of much debate. Governments and organizations across sectors
continually seek better and more effective ways to aid development efforts and are
simultaneously subjected to much criticism as their motives are questioned and the
negative side effects of their work are brought to light. Historically, development was
thought to occur through capital transfers and strengthening central governments (Korten,
1987). At least partially in response to government corruption issues in the developing
world, however, efforts have shifted to a focus on democratization and building local
capacity (Korten).
Similarly, the role that nongovernmental organizations (NGOs) play in the
development process has shifted over the course of the last few decades. While all
sectors of society (government, civil society, and the private sector) play a role in
international development, NGOs are increasingly at the helm of such efforts (Barr,
Fafchamps, & Owens, 2005).1 Furthermore, originally, the work of NGOs was largely
limited to relief, disaster assistance, and food distribution (Gellert, 1996). More recently,
however, NGOs have shifted their focus to alleviating poverty and improving the quality
of life in developing countries (Gellert).
The evolution of the role of NGOs in development efforts is thought to include
three generations (Korten, 1987; Atack, 1999). First generation NGOs focus on relief
and welfare and largely provide services to meet the immediate needs of those in the
1 NGOs are typically considered part of civil society, though some argue that they
actually occupy a fourth position in society, serving as a mediator between the other three
sectors (Fowler, 2002).


2
developing world. Second generation NGOs focus on local self-reliance, meaning that
they seek to develop the capacity of local people so that the benefits of an NGOs work
are sustainable even after the NGO leaves the region. In the third generation, the focus is
on sustainable systems development, or building the larger institutional and policy
context within which development takes place.
The increasingly central role of NGOs in development efforts is largely a reaction
both to the successes of various NGO initiatives globally and to perceived shortcomings
of governments abilities to successfully manage development (Barr et al., 2005).
Furthermore, when governments and the United Nations do engage in development
efforts, they now often use NGOs for their service delivery (Atack, 1999). Gellert (1996)
identified six attributes of NGOs that make them particularly well-suited to achieve
desirable outcomes in the developing world. First, NGOs have the ability to target and
reach areas in greatest need of assistance. Second, NGOs tend to promote local
involvement by working closely with local governments and enlisting local help. Third,
NGOs typically have relatively low costs of operations. Fourth, NGOs tend to be
adaptive and innovative. Fifth, NGOs are generally independent and free of political
involvement, and, finally, NGOs often seek sustainable outcomes.
While there are many benefits to NGO involvement in international development,
a number of challenges exist as well. Currently, the biggest challenge that NGOs face is
decreasing contributions from the international community (Gellert, 1996). This lack of
funds forces NGOs to become more creative in their service delivery, and, perhaps on the
positive side, limited funds are pushing NGOs to move toward more sustainable models
(Fowler, 2002). Furthermore, and of particular relevance to this thesis, with limited


3
funds, partnerships become an essential component of an organizations sustainability
because of the ability of partnerships to help an organization build capacity (Claiborne et
al., 2009). Indeed, since the early 1990s, there has been an increase in partnerships and
collaborations among NGOs (Gellert).
A second major challenge of NGOs is to maintain legitimacy as a facilitator of
international development. For example, many people have criticized NGOs for actually
hindering development efforts by undermining markets by providing goods and services
at less than market costs (Korten, 1987). Others question whether Western aid is actually
used in a way that is beneficial to the developing world, particularly in light of the
frequent disconnect between funders and beneficiaries of development efforts (Easterly,
2006). Indeed, as NGOs take on increasingly central roles in international development,
whether and how they should be involved in development processes has been the subject
of much debate.
Atack (1999) explores this issue by looking at NGO legitimacy, or the normative
basis for NGO involvement. He finds that four criteria are necessary to justify NGO
involvement in the developing world: representativeness, distinctive values, effectiveness,
and empowerment. Representativeness and distinctive values are formal-procedural
justifications. Representativeness refers to an organizations transparency and
accountability, particularly with respect to beneficiaries, rather than donors, and
distinctive values refers to an NGOs adherence to unified values, particular those with
which partners and constituents agree. The second two criteria, effectiveness and
empowerment, are substantive-purposive criteria. Effectiveness refers to an NGOs


progress toward development goals, and empowerment is the process by which
individuals are able to take control over their lives.
4
This thesis explores a particular aspect of legitimacy by looking at institutional
sustainability as a component of effectiveness. Organizational or institutional
sustainability places importance on building sustainable organizations to achieve
sustainable development benefit (Cannon, 2002, p. 363). As described in greater detail
in Chapter II, institutional sustainability refers to an NGOs ability to endure and to
continue to provide services. Theoretically, effectiveness and institutional sustainability
are related because more sustainable organizations will be better able to produce more
sustainable results, which is particularly important as organizations work toward local
empowerment and sustainable systems development. Of course, there are also
counterpoints to this argument, which are discussed in greater detail in Chapter II.
Purpose and Hypothesis
The purpose of this thesis is to examine the relationship between
interorganizational partnerships and institutional sustainability within a network of NGOs
operating in Uganda, and the hypothesis of this study is that organizations with greater
interconnectedness with other organizations in the subject network will score higher on
an array of institutional sustainability measures.
There are two primary components to this thesis. First, this thesis offers an in
depth look at a network of NGOs operating in the developing world. Through interviews
and network mapping, I describe the institutional sustainability, along a number of
dimensions, of the organizations involved in this study. I also explore the organizations
partnerships and relationships with one another and how the formal network operates and


5
has evolved. Understanding this network of organizations and its benefits and struggles
may offer insight to the international community regarding service delivery at the micro-
level. Looking deeply at a network of organizations that developed through local efforts
may help development efforts effectively move from relief and welfare to local
empowerment and sustainable systems development.
In the second part of this thesis, I test the hypothesis through a series of
quantitative tests based on survey results. Specifically, I look at differences in
institutional sustainability based on formal network membership and differences based on
organizations numbers of partners within the subject network. The purpose here is not
to rate these NGOs institutional sustainability, but rather, to look at the specific
characteristics of sustainability that correlate with increased numbers of network partners
(or network centrality). These findings may have important implications for networks
and partnerships abilities to build capacity through their relationships, particularly in
light of decreasing international aid.
Following a brief description of the subject network and its geographic context,
Chapter II presents a review of the literature regarding effectiveness generally,
institutional sustainability particularly, and partnerships and networks. Chapter III sets
forth the methods used in this study and includes a discussion of the particular measures
of institutional sustainability used in this study. Chapter IV consists of a detailed
description of the organizations in the study based on their interview responses, and
Chapter V uses network mapping to further understand the subject network and how
organizations within it interact. Chapter VI tests the hypothesis of this study through
statistical tests related to network membership and partnerships, and Chapter VII presents


6
an additional set of findings regarding the relationship between international ties and
institutional sustainability. Finally, Chapter VIII contains a discussion of the findings
and limitations of this study.
Uganda and Nansana Background
The network of organizations chosen for this study operates in the East African
nation of Uganda. As of July 2012, Ugandas estimated population was almost 34
million (Central Intelligence Agency, 2013). Despite its relative stability since 1986
when Yoweri Museveni assumed the presidency, Uganda continues to experience
difficulties related to violence, poverty, and widespread HIV/AIDS. The life expectancy
in Uganda is 53 years, and it has the fourth highest population growth rate in the world at
3.3% (Central Intelligence Agency). It is also eighth in the world for the rate of people
living with HIV/AIDS (Central Intelligence Agency). Ugandas per capita gross
domestic product is $1400, which is 202nd out of 228 ranked countries (Central
Intelligence Agency).
As with general recent patterns of NGO proliferation, the growth of the NGO
sector in Uganda began in the 1970s and 1980s following the collapse of the government
(Barr et al., 2005). In a descriptive study regarding a representative sample of NGOs
throughout Uganda, Barr et al. found that the three most common NGO activities in
Uganda are education and training, support to farming, and HIV/AIDS awareness and
prevention.
This study examines a network of NGOs operating in Nansana, Uganda. Nansana
is a suburban community in Wakiso District approximately 10 kilometers northwest of
Ugandas capital city, Kampala. Nansanas population was estimated at 89,900 in 2011,


7
up from 62,044 in 2002 (Uganda Bureau of Statistics, 2010a). Along with recent
population and development growth in Nansana (Nantakiika, 2011), there has been an
increase in the number of NGOs calling Nansana home. Nansana is currently home to at
least 40 NGOs, 35 of which are connected through the Nansana Civil Society Network
(M. Senyonjo, personal communication, January 18, 2013). In this study, I examine this
network of NGOs to assess the relationship between the levels of connectivity among
member organizations and the organizations institutional sustainability.
The Nansana Civil Society Network
Ugandan NGOs are well networked (Barr et al., 2005), and, in Nansana, there is a
formal, membership-based network of organizations called the Nansana Civil Society
Network (the NCSN). The NCSN began forming in 2008 when it was conceptualized by
three Ugandans working in the civil society sector in Nansana and received the support of
the Town Council.2 The NCSN officially began operating in 2010 as a community based
organization (CBO) in Wakiso District. CBOs in Uganda are generally single purpose
organizations, operating at subcounty levels to promote the well-being of a community
(Council on Foundations, 2012). The NCSN is a quasi-govemmental organization in that
it is technically part of civil society, but it has an office at the Nansana Town Council
(provided by the government) and is overseen by the Community Development Officer
for the Town Council.
The NCSN has 35 member organizations. To become members of the NCSN,
organizations submit an application form (describing items such as their programs,
2 Unless otherwise indicated, information throughout this section was obtained primarily
through personal communication with M. Senyonjo (the NCSN Coordinator) on January
18, 2013 and supplemented by additional interviews with members of Nansana civil
society throughout December 2012 and January 2013.


8
management, and operations) and letters of reference from other organizations. They
also submit to a site review by an NCSN representative. The only basic criterion for
membership is that organizations be located in Nansana (though exceptions may be made
for organizations located just outside the boundaries of Nansana). There are no
membership limitations based on organization size, services, or other organizational
characteristics. Organizations are generally admitted to the NCSN as long as they are
legitimate and well-intentioned.
The NCSN holds monthly members meetings, which give organizations the
opportunity to share, offer advice to one another, and support each other. The NCSN also
offers capacity building support, such as through trainings and by helping organizations
identify potential funding sources and apply for funds. Furthermore, a major emphasis of
the NCSN is on its referral system, a program through which organizations become aware
of each others programs and resources, enter into referral agreements, and refer cases
among themselves in order to best serve beneficiaries. The organizations also engage in
other periodic projects together, such as monthly sanitation (cleanup) efforts in Nansana.
The NCSN is entirely volunteer run. As described below, however, a volunteer
in Nansana is akin to an employee, except that instead of salaries, volunteers only receive
facilitation fees to cover their expenses. The NCSN has one permanent volunteer, who
serves as its coordinator, and seven rotating board members. With respect to the board
members, organizations are assigned to particular posts on the board of directors for one-
year terms. The selected organizations then appoint one of their employees (or
volunteers) to fill the post. The board of directors also meets monthly.


The NCSN is primarily funded through membership dues. Organizations pay an
annual fee to be part of the NCSN, and these membership dues comprise approximately
9
60% of the NCSNs budget. The remainder of the budget comes from additional
voluntary contributions from member organizations. The NCSN spends the majority of
its funds on its programs and coordinating activities, with a small portion of its fees going
toward supporting the Coordinator.
The NCSN Coordinator and representatives of member organizations generally
report positive results stemming from the NCSN, including improved service delivery
and enhanced capacity, particularly with respect to information and funding. The
Coordinator is also currently developing and promoting a volunteer and internship
program to encourage students and community members to volunteer for the various
Nansana organizations to further enhance the organizations capacities and to offer
valuable experience to local people.
While the NCSN appears to be largely successful based on member reports, it also
faces challenges, such as distrust among members stemming from perceived competition
for resources and duplication of services. Also, despite the large number of organizations,
many groups of people in Nansana do not receive support. The majority of organizations
in the NCSN target children as beneficiaries because it is relatively easy to get funding
for childrens programs and because there are so many children in need in Nansana (at
least in part reflecting the young population). Women are also frequent beneficiaries, but,
according to the Coordinator, few programs target other populations in need, including
men, widows and widowers, and people with disabilities. Furthermore, few


10
organizations focus on environmental issues because the cost of doing so is high and
because the environment is generally not considered a priority.
Following a discussion of the literature informing this study and the methods used
in this study, I move to a more detailed description of the organizations working in the
NCSN and in Nansana more broadly. I then look at whether the partnerships among
organizations in Nansana correlate with a variety of measures of the organizations
institutional sustainability.


11
CHAPTER II
EFFECTIVENESS AND NETWORKS
As described above, among the necessary ingredients for NGO legitimacy is
effectiveness (Atack, 1995). Effectiveness, however, typically refers to an NGOs
progress toward development goals (Atack), and increasingly, in the international
development realm, development goals focus on achieving sustainable results. In other
words, NGOs are indeed moving from first generation relief and welfare to second and
third generation work emphasizing the need to develop local capacity and sustainable
institutional and policy contexts.
This thesis looks at institutional sustainability as one dimension of effectiveness,
because if the organization itself is sustainable, it is in a better position to have lasting
effects on the community within which it works. Indeed, many organizations validly
focus on their institutional sustainability, because they want to be in a position to
continue to serve clients over an extended period of time (Edwards, 2002).
Organizational development implies a conscious process of acquiring and
strengthening the characteristics required by an [NGO] to position itself for maximum
impact and effectiveness (Edwards & Fowler, 2002, p. 3). Practically speaking, the
more stable an organization is, the more likely its employees will be able to turn their
attention outward and focus on the needs of beneficiaries. Also, an organization with
more characteristics of institutional sustainability (like adequate resources and reliable
record keeping) will be better poised to effectively serve a community, and as NGOs
transition from first generation to second and third generation work, this becomes
increasingly important.


12
At the same time, however, there is certainly an argument to be made that
organizations may have a self-interest in sustaining themselves at the expense of the
communities they serve. In other words, perhaps in some situations, institutional
sustainability undermines outcome effectiveness because organizations just want to
maintain a role for themselves. For these reasons, future studies exploring the
relationship between sustainable institutions and sustainable results may be warranted.
For purposes of the present study, however, it is reasonable to assume that
institutional sustainability is a component of effectiveness and that it may help an
organization achieve desirable results for its beneficiaries and move toward second and
third generation development work. I begin with a general discussion regarding
effectiveness.
Measuring Effectiveness Generally
Effectiveness in the nonprofit sector is an extremely broad and difficult to capture
concept, and there is little agreement regarding how to meaningfully measure it. This is
particularly important in the international NGO realm where effectiveness is often
assumed rather than tested (Atack, 1999). While in the for profit sector, effectiveness is
easily measured as it is synonymous with financial success, in the nonprofit sector,
missions and goals are typically more abstract and difficult to measure (Collins, 2005).
Like for profit institutions, however, to be successful and attract resources, nonprofits
must be able to demonstrate value to investors (Raiser, 2007).
Generally speaking, effectiveness can be measured at three levels: programmatic,
organizational, and network (Herman & Renz, 2008). Consistent with the discussion
above regarding outcome sustainability, the majority of studies evaluating nonprofit, and


13
particularly NGO, effectiveness do so at the program level. For example, in assessing the
effectiveness of an NGO health care program in Bangladesh, Mercer, Khan,
Daulatuzzaman, and Reid (2004) looked at programmatic outcomes, such as mortality
rates and the percentages of the populations being served who received certain treatments,
vaccinations, and types of care. Much work is also being done regarding measuring
effectiveness at the network level (Provan & Milward, 2001). The focus of this study is
effectiveness at the organizational level, which is an often side-lined yet essential
component of effectiveness, particularly as development efforts evolve.
Historically, measures of organizational effectiveness were inadequate because
they actually relied on programmatic measures (Herman & Renz, 2008). Thus,
recognizing that various models for measuring effectiveness often only target one aspect
of effectiveness (usually programmatic), recent nonprofit management literature supports
a multi-dimensional approach to measuring organizational performance (Herman &
Renz). For example, the Development Assistance Committee (1991) (a forum within
which member governments address economic, social, and environmental issues
associated with globalization) supports a framework that measures effectiveness at a
variety of levels and across issues. Similarly, recent measures of network success also
depend on multiple dimensions. For example, according to Kania and Kramer (2011),
each of five conditions (common agenda, shared measurement systems, mutually
reinforcing activities, continuous communication, and backbone support organizations)
must be present for a network of organizations to achieve collective success.
Sowa, Selden, and Sandfort (2004) developed a comprehensive framework for
measuring organizational effectiveness called the multidimensional, integrated model of


14
nonprofit organizational effectiveness (MIMNOE). Their model assumes that the
primary dimensions of effectiveness are management and program effectiveness, with
management referring to characteristics of the organization and its management, and
program referring to the organizations particular services. The two primary
dimensions (management and program) can then be subdivided into capacity and
outcomes, with capacity referring to the processes and structures of an organization and
outcomes referring to the results of programming or management efforts. Thus, there are
four categories within which to measure effectiveness: management capacity,
management outcomes, program capacity, and program outcomes. Furthermore, to
effectively measure each of these dimensions, Sowa et al. assert that both objective and
perceptual measures should be used.
Sowa et al. (2004) and Selden and Sowa (2004) (applying the MIMNOE in a
study of the effectiveness of 22 human service organizations providing early care and
education services) describe that objective measures of management capacity might
include items like whether the organization has a formal mission statement, a strategic
plan, human resources structures, employee training and performance management
systems, an independent financial audit, and information technology systems. Related
perceptual measures might include whether employees think the above systems serve a
functional role within the organization. Objective measures of management outcomes,
on the other hand, include items like revenue stability, the organizations ability to
maintain a financial surplus, and employee turnover. Perceptual measures might come
from managers self-reports regarding their organizations financial well-being or job


15
satisfaction reports among employees. These factors related to management capacity and
management outcome are particularly relevant to institutional sustainability.
On the program side, according to Sowa et al. (2004) and Selden and Sowa (2004),
objective capacity measures include items like how the organization transforms inputs
into outputs, the level of material resources assigned to programs, problems that have
occurred in implementing programs, staff experience levels, and diversity of services. A
perceptual capacity indicator might be how staff members assess the efficacy of their
programs or clients perceptions of staff competency or available resources. An objective
measure of program outcome might be the degree to which the program achieves its goals,
and a perceptual measure might look at client satisfaction.
Another model for measuring effectiveness at multiple levels comes from Sawhill
and Williamson (2001), who describe the Nature Conservancys experiences with
attempting to measure success. After a few rounds of trial and error, the Nature
Conservancy developed a model covering three broad areas: impact, activity, and
capacity. Impact asks whether an agency is making progress toward fulfilling its mission
and meeting its goals. Activity refers to whether an organizations activities are
achieving its programmatic objectives and implementing its strategies, and capacity
refers to whether the organization has the resources necessary to achieve its goals.
Most relevant to this study are capacity measures, which gauge the degree to
which the organization mobilized the resources necessary to fulfill the mission (Sawhill
& Williamson, 2001, p. 375). To keep its measure simple, the Nature Conservancy used
two to four measures for each of its three categories (impact, activity, and capacity). Its
capacity measures included fundraising performance, membership growth, and market


16
share. The authors also talked to other nonprofits about their efforts to measure
effectiveness, and, similarly, other organizations capacity measures typically include
items like financial activity, fundraising, and membership (Sawhill & Williamson).
To develop the institutional sustainability measure used in this study, I primarily
drew on the management measures described by Sowa et al. (2004) and the capacity
measures described by Sawhill and Williamson (2001). With these concepts and general
frameworks in mind, I now turn to a discussion of institutional sustainability.
Institutional Sustainability
Sustainability is a component of effectiveness that is increasingly gaining
attention, particularly in the NGO realm. This study looks specifically at institutional
sustainability rather than outcome sustainability, and, unlike the majority of studies
assessing programmatic outcomes, this study looks at correlations between
interorganizational relationships and institutional sustainability. As mentioned,
institutional sustainability may be particularly important as the emphasis in development
strategy moves from relief and welfare toward sustainable systems development (Atack,
1999).
Much has been written about outcome sustainability, as it is an increasingly
important measure of nonprofit effectiveness (Herman & Renz, 2008). According to the
Development Assistance Committee (1991), [t]he question of whether achievements are
sustainable in the longer run is of critical importance (p. 10). Lyons, Smuts, and
Stephens (2001) define sustainable development as the ability acquired and held by
communities over time, to initiate and control development, thus enabling communities
to participate more effectively in their own destiny (p. 1237), and they look at the


17
relationship between participation, empowerment, and sustainability. Also, Raiser (2007)
notes that nonprofits typically fall into three categories: traditional, sustainable, and cash
generating. While the traditional model tends to focus on charity, in sustainable and cash
generating models aid is viewed as an investment that will be used to help communities
become more sustainable and self-reliant.
As described above, institutional sustainability is theoretically related to outcome
sustainability, but less has been written about institutional sustainability directly. One
discussion comes from Lewis (2003), who looks at the relationship between
organizational culture and institutional sustainability in the sericulture (silk farming)
industry in Bangladesh. Drawing on Cannon (2002), Lewis explains that there are three
levels of institutional sustainability in the project setting: 1) Financial sustainability is a
projects ability to generate resources from a variety of sources; 2) organizational
sustainability refers to an organizational arrangements ability to continue to offer a
framework for delivering benefits or services; and 3) benefit sustainability refers to the
continuing availability of benefits or services beyond the life of the project.
In another study examining institutional sustainability in the NGO context,
Okorley and Nkrumah (2012) look at NGOs in Ghana and factors that might affect a
local NGOs sustainability. They look at factors that exist among sustainable
organizations, which they define as the ability or capacity of the NGOs to endure
(Okorley & Nkrumah, p. 331) and operationalize as organizations that have endured for
at least five years. They examine possible internal and external factors associated with
organizational sustainability, including leadership, funding, program development,


18
program management, and material resources, and they conclude that supportive
leadership and adequate funding are most correlated with organizational sustainability.
Here, I combine the above definitions and define institutional sustainability as the
ability of an NGO to endure and to continue to provide services. I include both
components because if institutional sustainability only refers to an organizations ability
to endure, we run the risk of having the organization endure at the expense of the
beneficiaries, and, if institutional sustainability only refers to a continuing ability to
provide services, the definition becomes too similar to a mere focus on programmatic
outcomes. Particularly in third generation development work, it is important to gauge the
sustainability of the institutional and policy systems surrounding development efforts.
Thus, the definition here is broader than that of Lewis (2003) because Lewis
discusses organizational sustainability in the context of a given project, and because, here,
I consider financial sustainability to be a component of institutional sustainability rather
than a separate type of sustainability. Furthermore, the operational concept used here is
also broader than that of Okorley and Nkrumah (2012), who submit that an organization
is sustainable if it has existed for five years. Here, I include elements of longevity,
service provision, capacity (including human, monetary, and capital resources), stability
of structures (both systems and physical), and planning for the future, in my
operationalized conception of institutional sustainability.
Interorganizational Relationships
Motivated to enhance their effectiveness, nonprofits increasingly rely on
partnerships and networks for their delivery of services (Sowa, 2009; Varda, Shoup, &
Miller, 2012). Networks have been shown to produce better outcomes for individuals,


19
organizations, and communities, and studies have shown positive associations between
networks and desirable outcomes like enhanced service delivery, organizational growth,
innovation and entrepreneurship, and increased civic engagement (Paarlberg & Varda,
2009). Furthermore, recent literature focuses on collective impact, which is based on the
idea that large-scale social change comes from better cross-sector coordination rather
than from the isolated intervention of individual organizations (Kania & Kramer, 2011,
p. 38). This study examines whether increased interorganizational connectivity is
associated with higher levels of institutional sustainability.
In the context of international development, partnerships among NGOs and
between NGOs and other sectors are considered important to improved service delivery
(Claiborne et al., 2009). As mentioned, this is particularly so as reduced funding persists.
Strong links, both among NGOs and between NGOs and other institutions build
organizational strength and social capital, and pool experience and other resources (p.
287). Indeed, NGOs are increasingly using collaboration to build capacity and enhance
their service delivery (Claiborne et al.).
Types of Relationships
Studies examining interorganizational relationships typically look at the
relationship between two organizations and the broader network of relationships. These
relationships can be categorized in a variety of ways. For example, networks can be
transmission networks, social support networks, and organizational networks (Varda,
Chandra, Stern, & Lurie, 2008). Relationships can also be classified based on the level at
which the arrangement occurs, including policy-centered integration, organization-


20
centered integration, program-centered integration, and client-centered integration
(Selden, Sowa, & Sandfort, 2006; Sowa, 2008).
A common way to think about interorganizational relationships is along a
continuum. For example, Kagan (1991) describes a continuum of relationships that
includes cooperation, coordination, and collaboration, with cooperation being the least
integrated relationship form and collaboration being the most integrated form. As one
progresses from cooperation through coordination to collaboration, interorganizational
relationships become more sophisticated, complex, and effective for problem solving
(Kagan, p. 2). Scholars typically agree that cooperation represents the least formal and
most prevalent type of relationship between organizations. Cooperations are grounded in
personal relationships and exist without clearly defined structures. Cooperating
organizations work only informally together, may have different goals, and generally do
not pool resources.
According to Kagan (1991), scholars agree less on definitions of coordination and
collaboration. Many do not see the difference between these two types of relationships,
as they share characteristics like mutuality of goals, resource sharing or pooling, and
dynamic, changing interactions. Those who separate the concepts see coordination as a
prerequisite for collaboration and as less complex and sophisticated than collaboration.
In coordination, agencies generally remain independent, but in collaboration, previously
separate organizations may come together in a new structure. Other important
distinctions include greater sharing of resources, more intense joint planning, and
increased sharing of power and authority.


21
More recently, Selden et al. (2006) described that cooperation consists of informal
and personal relationships between management and staff of different organizations,
while coordinating organizations try to calibrate their actions, and collaborating
organizations share existing resources, authority, and rewards. Others have also
expanded the continuum to include service integration where organizations work together
to provide a new package of services to shared clients (Selden et al.).
Along this continuum, Sowa (2008) further delineates the types of relationships.
She looked specifically at collaborations and found three different types of collaborations.
In collaborative contracts, agencies collaborate to produce services, but the relationship is
shallow in that it is mostly limited to sharing financial resources. In capacity-building
collaborations, agencies collaborate to provide services, but, beyond sharing financial
resources, they also share human resources, such as staff and professional development
resources. In community-building collaborations, the same types of resources are shared,
but the benefits go beyond the tangible organizational-centered benefits found in
capacity-building collaborations. In these relationships, the organizations involved also
gain an enhanced understanding of the community they are serving, and they extend their
role in that community.
Others describe a simpler continuum that varies depending on the formality of
relationships. At one end of the continuum are formal relationships, which are
characterized by joint programs, parent-subsidiary relationships, joint venture, or merger.
At the other end are informal collaborations, which are characterized by information
sharing, client referrals, sharing office spaces, and management service organizations
(Guo & Acar, 2005).


22
To understand the organizations in this study and their relationships, here I adopt
the language of Kagan (1991) and others, and describe relationships as cooperation,
coordination, or collaboration. These categories provide a useful framework for
understanding the types of relationships that exist in Nansana, particularly as they
describe varying levels of integration and organizations abilities to impact and pervade
communities. The literature that informs how to measure these relationships is described
below.
How to Measure Relationships
Because the intensity of relationships may have important impacts on a variety of
outcomes, scholars have sought to develop means of measuring this intensity (Selden et
al., 2006). Existing literature regarding interorganizational relationships and networks
describes a variety of means to measure the connections among organizations and
determine how to classify interorganizational relationships based on the above
frameworks. Connectivity is the measured interactions between partners in a
collaborative such as the amount and quality of interactions and how these relationships
might change over time (Varda et al., 2008, p. El).
To measure connectivity, studies often ask subjects to describe their various
partnerships and then also ask subjects about the relationships between the partners they
identify. For example, McPherson, Smith-Lovin, and Brashears (2006) examine
discussion networks of close confidants and include measures such as size of network
(number of names mentioned), density and range of network (which refers to the intensity
of relationships, operationalized as the mean intensity of tie strength among the partners
mentioned, where relationships were assigned a 0 if partners are strangers, a 1 if they are


23
close, and 0.5 otherwise), duration of relationship, frequency of contact, and diversity of
the interpersonal environment along various sociodemographic dimensions.
An often-mentioned method for measuring and distinguishing between
relationships is based on the amount of resource sharing involved. For example, Kagan
(1991) describes that informal relationships have little resource sharing, while formal
relationships may involve the formation of a new integrated organization. In between
these two extremes are relationships characterized by shared monetary or material
resources, shared programs, and shared administrative capacities.
Also, Sowa (2008) looked at three types of sharing: resources, staff, and rewards.
Resources are primarily conceptualized as financial, although nonfinancial resources,
such as professional or practice knowledge may also be shared. Staff can be shared either
through the co-use of a specialized position, or it can involve housing staff in a shared
office space. Finally, rewards are more difficult to define, but are considered the benefits
that accrue to the organizations involved in the collaboration or to the larger community
of which the collaboration is a part.
Others have described additional components that may be useful in examining
connectivity. For example, in measuring the connectivity of members of public health
collaboratives, Varda et al. (2008) identified seven core dimensions of connectivity in
public health collaboratives: network membership, network interaction, role of health
department, frequency of interaction, strategic value, trust, and reciprocity.
First, network membership refers to who is involved in the network and the
diversity of membership; it is measured by identifying organizations by name, type, and
other characteristics, such as size and mission. Second, network interaction goes beyond


24
merely counting numbers of members and looks at the quality of interactions among
subgroups by examining network patterns. Third, the role of health departments refers to
whether there is a coordinating entity within a network of organizations. Fourth, the
frequency of interactions can be operationalized by looking at the types of relationships
and levels of interactions among organizations, bearing in mind that more frequent
interactions do not necessarily equate with greater connectivity. Fifth, the strategic value
of partners looks at the extent to which each partner 1) has power and influence within a
community; 2) is actively involved in the network by participating in meetings,
accomplishing tasks, and so forth; and 3) has resources like money, food, physical space
for meetings, staff time, data, educational materials, knowledge, information, and
feedback. Sixth, trust depends on reliability, shared belief in mission, and opportunity for
frank discussion. Finally, reciprocity refers to mutual exchange of resources.
Other studies have examined characteristics of relationships to determine when
such relationships have particularly positive effects. According to Paarlberg and Varda
(2009), for a network to have positive synergistic effects, first, the exchanges must be
reciprocal, with resources flowing in both directions between members. Second, while
carrying capacity is typically determined by the human and financial capital in a
community, networks require particular capabilities necessary to facilitate these
relationships. Third, the process of exchange looks at whether network members are
directly connected or whether they are tied through a string of other actors. It also looks
at the strength of ties. Strong ties (exchanges that occur through frequent, extended, and
reciprocal activities) are common among homogeneous organizations and often develop
because of physical proximity or repeated interactions, and organizations with strong ties


25
tend to have similar missions and similar clients. Weak ties refer to ties between more
diverse network members, or those with different missions, clients, and resources streams.
Weak ties contribute to positive synergistic effects. Finally, network exchanges should
be coordinated, particularly via linkage to a governance structure, which can be either
formal or informal.
Each of the above characteristics and descriptive factors suggest methods for
measuring connectivity between various partners. This study examines a particular
network of organizations and measures the connectivity between member organizations
by looking at the duration of relationships and frequency of contact, the types and amount
of resources they share, whether they have joint programs, and the formality of the
relationships.


26
CHAPTER III
METHODS
To understand the relationships between NGOs in Nansana and their institutional
sustainability, I interviewed a sample of 33 organizations in Nansana. The majority of
the organizations interviewed (28) are members of the NCSN, thus comprising a sample
of 80% of the 35 organizations in the NCSN. Four organizations interviewed are not
members of the NCSN, and it is not clear how many organizations there are in Nansana
that are not members of the NCSN. The final organization interviewed is the NCSN
itself, which is a registered CBO. The NCSN, however, is not included in the data
analysis, but rather, is discussed separately in Chapter I.
Data Collection
Data collection occurred on site in Nansana in December 2012 and January 2013.
The NCSN Coordinator scheduled the majority of the meetings with organizations on my
behalf; however, I made clear to all interviewees that their participation was voluntary,
that their answers were confidential, and that their participation and responses would in
no way impact their status or finances. Whenever possible, I met with directors of
organizations, but, in all cases, I ensured that the person I spoke with was knowledgeable
regarding the organization. The Coordinator participated in the interviews only if the
interviewee specifically requested that he do so. Interviews took place at organizations
offices, except in the cases of those organizations without offices.
The interviews were semi-structured. I personally conducted all interviews and
used the same survey for each interview. The survey can be found in Appendix A.


27
Interviewees were asked to sign a consent form and were provided an opportunity to
provide personal data if they wished to receive a copy of the study outcomes.
The survey itself consists of two primary components: institutional sustainability
and interorganizational relationships. The institutional sustainability portion of the
survey is subdivided into eight sections, suggested by the literature: services and
programs, people, facilities, technology, finances, records, program evaluations, and
human resource systems. The interorganizational relationships section asks interviewees
to identify all of their partners according to specified parameters, and then, for each
organization named, interviewees were asked questions regarding the length of the
partnership, the extent of its formality, whether it involves joint programs, the types of
resource sharing involved, and the frequency of contact. In some situations, I attempted
to fill informational gaps by reviewing organizations websites or by contacting other
organizational representatives as recommended by interviewees.
Data Analysis
Data analysis was performed upon return to the United States. For purposes of
analysis and discussion, organizations were randomly assigned numbers from 1 to 32.
For the majority of the data analysis, I used Excel and SPSS to look at frequencies, means,
medians, minimum and maximum values, standard deviations, and significance tests. As
described below, a frequent method of analysis was to divide subject organizations into
two to four groups along various dimensions and compare mean answers for a variety of
institutional sustainability measures. Significances of the differences in means were
tested using one-way ANOVAs, independent samples t-tests, and chi-squares, as
appropriate.


28
Two methods were used to better understand the organizations in Nansana and
their relationships with one another. First, survey responses were analyzed using basic
descriptive measures as described below. Second, to further understand the relationships
between and among organizations in Nansana, network mapping was performed using
UCINET and NetDraw.
To test the hypothesis regarding the relationship between networks and
partnerships, on the one hand, and institutional sustainability, on the other hand, I
performed three sets of comparisons. For the first set of comparisons, I divided the
subject organizations based on whether they are members of the NCSN. For the second
set of comparisons, I divided organizations based on the number of partners they
mentioned, and, for the third set of comparisons, I divided organizations based on the
number of times they were mentioned as partners by other organizations.
Organizations were also divided into three groups according to the strength of
their international ties for further analysis. While conducting the research, it seemed that
organizations with strong international ties assumed a more central role among
organizations in Nansana. These international ties may largely explain network and
institutional sustainability patterns, and, as such, it seemed important to test these
relationships as well. According to Barr et al. (2005), NGOs in East Africa (including
Uganda) are generally funded through international sources, and most NGO funding is
funneled through a small number of Ugandan NGOs: Many Ugandan NGOs indeed de
facto or de jure operate as subcontractors for international donors (p. 676).


29
Measures of Institutional Sustainability
To test the hypothesis that organizations with more ties are likely to also score
higher on various measures of institutional sustainability, for each of the eight categories
of institutional sustainability, one to three measures of institutional sustainability were
selected to be compared across categories of organizations. The measures selected to
represent each category are described below and summarized in Table III. 1.
Table III.1. Measures of Institutional Sustainability.
Institutional Sustainability Measures
Services and Programs Years of operation
Number of programs
People Number of beneficiaries
Number of employees
Number of volunteers
Facilities Number of facilities
Ownership rate
Number of offices outside Nansana
Technology Number of technology items
Number of computers
Finances Actual budget
Actual: expected budget ratio
Records Points out of 5
Program Evaluations Rate conducting
Human Resource Systems Points out of 5
Services and Programs
Two measures of institutional sustainability were used in the services and
programs category: 1) the number of years the organization has been in existence and 2)
the number of programs run by each organization. With respect to the length of time an
organization has been in existence, while it may seem that an organizations years of
operation may itself be a proxy for institutional sustainability, and it has been used as
such (Okorley & Nkrumah, 2012), such a measure likely only captures a narrow view of


30
institutional sustainability. Specifically, an organization may technically operate for
many years, but without any resources or without actually serving any clients.
It should also be noted that some organizations indicated both the time that they
began internally and the time that the organization either began serving clients or
registered as an NGO or CBO. For consistencys sake, if an organization provided
multiple dates, I used the earliest date provided. Furthermore, if an organization
indicated the year in which it began, rather than the length of time it had been operational,
I subtracted the year from 2012.
The second measure of institutional sustainability in the area of services and
programs is the number of programs each organization conducts. While many of the
organizations claim to have a large number of programs to benefit their clients, it should
be noted that more programs does not make an organization more effective, and, in fact,
the converse may be true.
One of the strongest lessons to emerge from the research is that success is more
likely when organizations identify a clear long-term goal at the outset and stick to
it over time. Conversely, agencies which change their goals too often or try to
follow too many goals at the same time, often lose their way. (Edwards, 2002, p.
284)
Thus, it seems that organizations with fewer programs are likely more sustainable than
those attempting to carry out many programs because organizations with fewer programs
will have relatively greater resources to commit to those programs. Indeed, in Nansana,
many organizations appear to be attempting to operate numerous programs to serve their
communities, despite extremely limited budgets. As such, the expected direction of this
variable is unclear.


31
People
Three measures of institutional sustainability are used in the people category. The
first measure is the number of beneficiaries served by each organization. Here, I assumed
that more beneficiaries suggests greater institutional sustainability. While, on one hand,
an organization with fewer beneficiaries may be more sustainable because it will have
relatively greater resources to commit to each beneficiary, on the other hand, an
organization with more beneficiaries presumably has the resources to sustain more
beneficiaries. Furthermore, organizations goals and objectives are generally to help
people. Thus, if an organization is not helping many people it is not very sustainable
with respect to its purpose.
Organizations reported their numbers of beneficiaries differently, with some
counting individuals (such as children or women) as their beneficiaries, and others
counting the number of families they serve. For comparative purposes, I translated the
number of families into individual units by multiplying the number of families by five,
which is the average household size in Uganda (Uganda Bureau of Statistics, 2010b).
The second and third measures of institutional sustainability in this category are
number of volunteers and number of employees. Organizations with greater human
resources, as measured by numbers of volunteers and employees, are likely to have
greater institutional sustainability. Of course, it is possible that some organizations with
many employees may be spread too thinly and unable to sustain their employees, but
multiple measures of institutional sustainability are being used in this study in order to
compensate for the limitations of any single measure.


32
With respect to employees and volunteers, it should be noted that many of the
organizations are solely volunteer run due to a lack of funds. Local volunteers, however,
are only distinguishable from employees based on the compensation they receive.
Volunteers typically treat their posts as full-time jobs, yet they receive only a facilitation
fee, which might cover their transportation to and from and the job and minimal other
expenses. Thus, consistent with local terminology, I considered employees of
organizations to be only those individuals who receive a salary beyond a facilitation fee.
Volunteers include both local and international volunteers.
Facilities
The physical space or facilities dimension of institutional sustainability is
measured via three components. First, the number of facilities that an organization has
on its premises in Nansana may offer some indication of its sustainability. Examples of
facilities include offices, schools, health clinics, and housing. Of course, the number of
facilities may merely reflect an organizations goals or may not reflect the number of
facilities an organization can appropriately sustain, but, for purposes of this study, and
because it is, again, only one measure of sustainability, it will be assumed that an
organization with more facilities is more sustainable.
The second measure related to physical space is whether an organization rents or
owns its facilities in Nansana. Presumably, an organization that owns its facilities will
have greater institutional sustainability, because its location is more stable and because it
has more financial resources. Furthermore, at least among organizations interviewed,
those that own typically own their properties outright.


33
A third measure in this category relates to the number of offices (both in Uganda
and internationally) that an organization has outside of Nansana. Presumably,
organizations with a greater number of offices will have greater institutional
sustainability because they may have access to greater resources, such as informational
and material resources, and greater ability to compensate elsewhere when one office
location is struggling.
Technology
In the technology category, two measures of institutional sustainability are
employed. The first is the number of types of technology systems an organization has out
of the following 10 possible technology systems: website, network, electronic records,
domain name, organization e-mail, organization phone, computer/laptop, facsimile
machine, scanner, and printer. An organization with more of these resources will likely
be more sustainable and better positioned to reach its constituents. The second measure
is simply based on the number of computers or laptops that an organization owns.
Finances
Two measures of institutional sustainability related to finances will be used in this
study: 1) organizations actual budgets and 2) organizations actual budgets as a
percentage of their expected budgets. Financial sustainability and institutional
sustainability are often confused (Cannon, 2002), likely reflecting the fact that it may be
tempting to view an organizations sustainability as measurable by its finances. Indeed,
bigger NGOs are not necessarily more sustainable or better able to achieve sustainable
results. For example, according to Fowler (2002), large expenditures might suggest that
NGOs are more focused on ameliorative efforts than on sustainable results. Thus, it


34
should be borne in mind that this section is merely one component of institutional
sustainability, and, while it may be reasonable to assume that organizations with more
money are more sustainable, this is not necessarily the case. For example, an
organization with many capital resources and a small monetary budget may be at least as
sustainable as an organization with few capital resources and a larger monetary budget.
Furthermore, just because an organization has greater financial resources, does not mean
that it is using those resources well.
Financial information was obtained by asking organizations about their annual
finances for the year 2012. Organizations were invited to use estimates. If financial
information was not yet available for 2012, organizations were asked about their 2011
finances. Furthermore, a few of the organizations provided budgetary information for a
period of two or three years, in which case I used the mean annual budget for the period
covered. Many organizations reported their finances in Ugandan shillings, and one
organization reported its finances in Canadian dollars. To convert these figures to US
dollars, I used the December 31, 2012 Financial Management Service exchange rate of
2,686 Ugandan shillings and 0.995 Canadian dollars per 1 US dollar (Financial
Management Service, 2013). While there was some variation in the budgetary reporting
periods, for illustrative purposes, this exchange rate provides reasonable estimates of
organizations financial situations.
As mentioned, the first measure in this category is an organizations actual annual
budget. The second measure is the actual budget to expected (or ideal) budget ratio,
expressed as a percentage. Among the organizations in Nansana, there is a considerable
difference between the amount of money that many of the organizations budget for at the


35
beginning of each year (or budgetary period) and the amount of money they actually have
in a given year. Possibly reflecting the unpredictability of funding, rather than attempt to
predict what they may earn in a given year and budget accordingly, many organizations
create a budget based on a best-case scenario. While the anticipated budgets alone seem
to say little about an organizations institutional sustainability, the ratio may give some
indication regarding the predictability of an organizations budget or its ability to
realistically plan its programming and services, both of which logically relate to an
organizations institutional sustainability. By this logic, percentages closer to 100% are
more indicative of institutional sustainability.
Records
Record keeping in the developing world is often poor (Barr et al., 2005). To
measure institutional sustainability with respect to record keeping, organizations were
given a score out of five. The first four points relate to whether organizations produce or
maintain each of the following: periodic financial reports, client records, donor records,
and information regarding programs. Organizations received an additional point if they
maintain their records in both hard and soft format. While in the United States paperless
has become a more desirable mode of operation, the lack of networks and servers in
Nansana means that many organizations maintain their electronic files on memory sticks,
which could easily be erased or misplaced. Indeed, one organization reported having just
lost most of its electronic files due to a memory stick failure. Thus, it seems that
organizations that maintain files in both hard and soft form may be more sustainable.


36
Program Evaluations
Though perhaps more related to outcome sustainability, program evaluations may
also be an important component of institutional sustainability because they suggest some
level of continuity in programming in that organizations are periodically evaluating what
they are doing and making improvements. To measure institutional sustainability in this
category, I used a binomial variable to reflect whether organizations conduct some sort of
evaluations. Thus, organizations received a 1 in this category if there is some indication
that they conduct some sort of methodical evaluation of their programming. Limitations
of the research methods, including reliance on self-reports, made it difficult to determine
what organizations were referring to when they reported conducting internal or external
evaluations. For example, one organization reported conducting external evaluations, but,
when probed, it became clear that the organization representative was merely referring to
casual conversations with members of other organizations regarding her organizations
programming. In this situation, the organization did not receive a 1 due to the lack of
methodology associated with this type of evaluation.
Human Resource Systems
To measure institutional sustainability related to human resource systems,
organizations received a score out of five based on whether they have the following five
human resources systems: orientations for new hires, employment files, employee
manuals, policies regarding hiring and terminating, and employee access to ongoing
training. The presence of these items may offer some indication of an organizations
sustainability, in that they conduct some level of planning regarding their employees and
contribute to their employees growth.


37
CHAPTER IV
OVERVIEW AND FINDINGS RELATED TO SAMPLE
In addition to the NCSN, I interviewed 32 organizations. Twenty-eight of these
organizations are members of the NCSN and comprise a sample of 80% of its member
organizations. The other four organizations are other NGOs or CBOs operating in
Nansana but which are not part of the formal network. There appeared to be few
organizations operating outside the NCSN in Nansana, but I was not able to obtain data
related to the same.
Returning to the three generations of development, in Nansana, the majority of the
organizations interviewed remain focused on first generation work in that they continue
to strive to meet the immediate needs of vulnerable populations in their communities.
There is some evidence of a shift toward second generation work in that most of the
organizations are local, Ugandan-run NGOs, suggesting increased levels of local self-
reliance. Furthermore, many of the NGOs are seeking and focusing on sustainable
solutions to health and poverty issues, which is more in line with second generation work.
On the other hand, though, the vast majority of funding for all organizations in the NCSN
comes, either directly or indirectly, from international sources, suggesting that much
work remains before there will truly be local self-reliance.
In this chapter, I describe the organizations operating in Nansana based on data
obtained from the 32 organizations in the sample. Data are discussed according to the
eight categories of institutional sustainability described above; however, additional data
were obtained for descriptive purposes that are not included in the actual measures of
institutional sustainability.


38
Generally speaking, as noted above, there has been considerable growth in the
civil society sector in Nansana in recent years, and, while the vast majority of
organizations are based in Uganda and Ugandan run, funding often comes from outside
Nansana, and there is wide variation in organizations financial resources. Programs are
typically geared toward children, youth, and women, and volunteers play an important
role in Nansana, which bodes well for citizens commitments to their fellow community
members. Organizations tend to have good planning, record keeping, evaluation, and
human resource habits, and most are managed by boards of directors. Almost all
organizations have some physical space, and most are using technology systems to
varying degrees.
Thus, while there is certainly room for improvement given the relative youth of
many organizations (the average organization has been operating for 8.34 years),
Nansana organizations have fairly stable systems in place. Funding is an issue for many
organizations, but, particularly through the NCSN, smaller organizations are working to
build their capacities through other means. Below are more detailed survey and interview
results.
Services and Programs
As mentioned, organizations in Nansana have been operating for a mean length of
8.34 years, though some actually only began serving clients more recently. The newest
organization is two years old, and the oldest organization is 33 years old.
Of the 32 organizations, 27 are headquartered in Uganda, and four are
headquartered internationally (in Ireland, Japan, Switzerland, and the United States).
One organization has headquarters in both Uganda and Canada. All organizations are


39
registered as either NGOs in Uganda or CBOs in Wakiso District (the district in which
Nansana is located). Nineteen organizations are registered as NGOs (59.4%), and 13
(40.6%) are registered as CBOs. For purposes of the NCSN, however, one NGO is
considered a faith based organization (FBO) based on its primary emphasis on religion.
With respect to planning habits, 29 (90.6%) of the organizations periodically draft
strategic plans (defined as documents outlining longer term goals and how an
organization will achieve those goals). Organizations do this for an average of 25.15
months at a time, with some organizations planning just one month at time and other
organizations planning five years at a time. Furthermore, organizations in Nansana have
an average of 5.41 programs, with the minimum number of programs being 1 and the
maximum number being 9.
Organizations were asked in an open-ended question to describe their programs,
and these programs were then recoded into 12 distinct categories reflecting the types of
services frequently provided by many of the organizations. The results are shown in
Figure IV. 1. The most common type of program mentioned is educational programming,
with 23 organizations (71.9%) reporting having programs in this area. Here, education
refers to programs designed to assist children or youth in their educational pursuits. This
takes a variety of forms with some organizations directly providing education through
schools on their properties, and others paying full or partial school fees or providing
books, uniforms, or other scholastic materials, for children in need. The prevalence of
educational support programs is consistent with the findings of Barr et al. (2005), who
found that the most prevalent type of NGO activity in Uganda is education and training.


40
Program Areas
Education
Counseling
Vocational Training
Health
Recreation
Community Education
HIV/AIDS
Agriculture
Savings & Loans
Environment
Religious
Volunteers
0% 10% 20% 30% 40% 50% 60% 70% 80%
Figure IV. 1. Percentage of organizations with particular programs.
The second most common type of program among Nansana organizations is
counseling, which refers to programs offering a variety of counseling and psychosocial
support, including home visits. Seventeen organizations (53.1%) report having programs
in this area.
Third, 14 organizations (43.8%) report having programs in each of two areas:
vocational training and health. Vocational training refers to programs, typically targeting
women, designed to teach vocational skills, including tailoring, hairdressing, and baking.
Health refers to programs involving the actual provision of health care, referrals for
health care, or health education.
Fourth, 13 organizations (40.6%) report having programs related to recreation.
Recreation includes programs related to music, dance, drama, arts and crafts, culture,
sports, and the like. Some of these programs, particularly arts and crafts, are sometimes


41
designed to help raise funds for the organizations or their beneficiaries, though profits
derived from these programs are minimal.
Fifth, 12 organizations (37.5%) have programs related to community education.
Community education includes programs related to education, advocacy, or increasing
community awareness on a range of topics, such as substance abuse and domestic
violence.
Sixth, 11 organizations (34.4%) have HIV/AIDS programs, which typically
involve counseling, testing, and treatment related to HIV/AIDS. While these programs
may also be a subset of health education, here I broke them out separately because many
organizations specifically target HIV/AIDS, and organizations typically speak of
HIV/AIDS programs as being different from other health education or treatment efforts.
Indeed, it is common for organizations in Nansana to talk about their HCT (HIV
counseling and testing) programs.
Seventh, 8 organizations (25%) have agricultural programs, which refer to
programs involving farming or livestock. Many of these programs serve a dual role of
teaching beneficiaries agricultural skills and either raising revenues or reducing food
costs for the organizations.
Next, 6 organizations (18.8%) have programs in each of two areas: environmental
efforts and savings and loans. Environmental programs include those that provide
education regarding the environment, infrastructure support (such as constructing water
tanks in villages), or sanitation programs (such as periodic cleanup efforts). Savings and
loans programs are those related to microfinance, typically involving savings groups or


42
the provision of small loans or seed money. Savings and loans programs usually target
women.
Four organizations (12.5%) offer religious programming. While many of the
organizations in Nansana are religiously based, few specifically reported offering
religious programming, but, those that did, typically had programs involving Bible study
and evangelism.
Finally, three organizations (9.4%) reported having volunteer programs. These
programs refer to efforts related to volunteer coordination, in which organizations
actually recruit and place volunteers (usually volunteers from abroad).
People
Generally speaking, there are four groups of people associated with NGOs:
beneficiaries, employees, volunteers, and board members.
After converting beneficiaries into individual units, as described above, the mean
number of beneficiaries served is 699.03, with 10 being the lowest number of
beneficiaries and 10,894 being the highest number. The median is 198 beneficiaries. It
should be noted, however, that many of the organizations loosely defined beneficiaries,
meaning that the numbers of beneficiaries do not necessary reflect those individuals with
whom the organization has ongoing contact. For example, if an organization held an
informational program for the community and 100 people attended, the organization
likely counted those 100 people among its beneficiaries.
Categories of beneficiaries are shown below in Figure IV.2. The majority of the
organizations clients, or beneficiaries, are children, with 84.4% of organizations
claiming to benefit children. Twenty-five percent of the organizations also serve youth.


43
While definitions of youth vary across organizations, organizations tend to consider
youth as beginning in the mid- to late-teens and running into an individuals twenties
(though at least two organizations defined youth as including individuals up to age 35).
Twenty-five percent of organizations have programs specifically targeting women. Five
organizations (15.6%) include people living with HIV/AIDS among their beneficiaries,
and two organizations (6.3%) serve those with disabilities. Four organizations (12.5%)
consider the community at large to be among their beneficiaries.
Figure IV.2. Percentage of organizations serving each category of beneficiaries.
Organizations have a mean of 9.88 employees. With respect to employees and
volunteers, again, many organizations depend on full-time volunteers who receive only
facilitation fees. According to Barr et al. (2005), [f]ull- and part-time volunteers
account for most of the manpower available to surveyed NGOs (p. 669). Indeed, eight
organizations (25%) have no employees and the largest employs 72 individuals. On the
other hand, 29 organizations (90.6%) have volunteers. Among the 28 organizations that
reported having local volunteers, the mean number of local volunteers is 8.18. Only six


44
organizations reported having international volunteers, and the average number of
international volunteers for those organizations is 3.2.
Of the 32 organizations, 29 report that they have a board of directors. Twenty-
eight organizations have a board of directors in Uganda, and four have international
boards, meaning that three organizations have boards both in Uganda and in the country
where they are headquartered. Three of the Ugandan boards also have at least one
foreign member, while the rest consist solely of Ugandans. The mean size of the
Ugandan board is 7.54 people, with three people comprising the smallest board and 12
comprising the largest. Boards meet between once a month and once a year with the
average meeting occurring once every 4.61 months.
Facilities
Looking at the types of facilities and physical structures that house organizations
in Nansana, 30 of the 32 organizations reported having some sort of physical property,
and one of the two without property reported being between offices at the time of the
interview.
All organizations with property have an office, and schools and residential
facilities (generally for orphans) are the next most common type of facility with 25% of
organizations having each of these facilities. Other types of facilities include vocational
training centers, which 12.5% of organizations have; volunteer housing, which 9.4% of
organizations have; agriculture or livestock areas, health clinics, and libraries, each of
which 6.3% of organizations have; and a church, run by one (3.1%) organization.
Of the thirty organizations that had properties at the time of their interviews, 19
rent their space (63%) and 11 (37%) own their space. Furthermore, 14 of the 32


45
organizations have other offices or properties outside of Nansana as well. Of the 14, 13
have other offices in Uganda, and four have other offices internationally.
Technology
All but one of the organizations surveyed reported having some form of
technology system. Organizations were asked about 10 specific types of technology:
website, network, electronic records, domain name, organization e-mail account,
organization phone, computer/laptop, fax machine, scanner, and printer. The most
frequent type of technology reported by organizations was electronic records, with 31
organizations (96.9%) claiming to have at least some of their files in electronic format.
Next, 29 organizations (90.6%) have e-mail accounts. Twenty-three organizations
(71.9%) have at least one organizational computer or laptop, and among those
organizations with computers or laptops, the mean number of computers was 4.76, with
the numbers of computers per organization ranging from 1 to 36.
Twenty organizations (62.5%) have a website, though there is variation regarding
whether the websites are directly managed by the organization or by a free website
hosting service. Eighteen organizations (56.3%) have an organizational phone (either
mobile or landline), and the same number of organizations have a printer. Twelve
organizations (37.5%) have a scanner, and twelve have an organization domain name.
Six organizations (18.8%) have a network, and, finally, no organizations have a fax
machine.
Finances
With respect to annual budgets, again, there is a distinction between organizations
expected (or ideal) budgets and their actual budgets. Among those organizations that


46
reported their finances, the mean expected annual budget is $46,104, and the median is
$18,615. Expected budgets range from $856 to $297,840. The mean actual budget is
$40,274, and the median is $7,167. The lowest actual budget is $0, and the highest is
$297,840.
I also calculated the actual budget as a percentage of the expected budget given
the difference found between these figures. Indeed, the mean organizations actual
budget is 55% of its expected budget. While the lowest percent was 0%, reflecting the
fact that one organization reported having no money in 2012, one organizations actual
budget was 1% of its expected budget. Many organizations actual and expected budgets
were the same or similar, though.
As far as the sources of their finances, organizations were asked what percentage
of their funds come from each of seven categories: individual donors; foundation grants,
including grants from foreign government agencies, such as USAID; a government entity
in Uganda; membership fees; fees for services; commercial income, such as that from
farming or sale of goods; and an other, catchall category.
The largest source of funds for organizations in Nansana is individual donors,
both local and international. Organizations reported receiving an average of 41% of their
funds from individual donors, and many organizations reported receiving all of their
funds from individual donors. It should be noted that the percentages across
organizations were averaged.
The next most common source of income is the catchall category, with
organizations reporting receiving an average of 20% of their funds from other sources,
such as staff contributions, funds funneled from other NGOs, and other sources. While


47
many organizations received no foundation grants, on average organizations received 14%
of their funds from foundations or international government entities, such as USAID or
other countries equivalents. On average, organizations receive 10% of their money from
fees for services, most commonly from school fees. Membership fees account for an
average of 9% of organizations budgets, and commercial income accounts for 5% of
their budgets. Finally, very little money comes from government entities in Uganda.
Only one organization reported receiving a very small amount from the Ugandan
government for an overall average among organizations of 0.07%.
Organizations were also asked what percentage of their income is spent on each
of four categories: programs, staff, administrative expenses, and other items. The
majority of organizations finances is spent on their programs. Organizations reported
spending an average of 52% of their funds on programs, though the range is from 0 to
100%. Administrative expenses, such as rent, technology, and other administrative costs,
account for 28% of organizations budgets. Staff salaries average 15% of organizations
budgets, with many organizations having few or no staff expenses due to the widespread
use of volunteers, and 6% of budgets reportedly go toward other expenses, such as
payments directly to those in need.
Records
To understand the record keeping habits of the organizations in Nansana, the
organizations were asked whether they keep records in four areas: finances, clients,
donors, and programs. They were also asked whether they maintain their files in written
or electronic format, or both.


48
Organizations reported high rates of record keeping. All but one organization
reported maintaining periodic financial reports, and, on average, those organizations
producing periodic financial reports reported doing so 2.84 times per year. All
organizations reported maintaining records of their clients. The limitations of self-reports,
however, were obvious when one organization reported maintaining client records but
also claimed that its beneficiaries are uncountable. Twenty-seven organizations
(84.4%) reported maintaining donor files. All five of those without donor records,
however, claim to not have them not for want of record keeping, but rather because they
do not have donors. Twenty-nine organizations (90.6%) report maintaining some sort of
records regarding their programs.
With respect to how these records are kept, 30 organizations reported maintaining
both hard and soft copies of their files, though both sets of files were not necessarily
complete. One organization maintains only hard copies of its files, and one maintains
only soft copies. It seems difficult to say here whether one form of record keeping is
superior to another in the Ugandan context, particularly since the vast majority of
electronic records are not backed up by a server.
Program Evaluations
Twenty-seven organizations (84.4%) reported conducting some sort of evaluation
of their programs, though the formality of the evaluations are likely widely variable, with
some organizations considering discussions regarding programs at period staff meetings
to be evaluations. Eleven organizations (34.4%) reported conducting both internal and
external evaluations. Thirteen organizations (40.6%) reported conducting only internal
evaluations, and three organizations (9.4%) reported conducting only external evaluations.


49
It is also worth noting that, in their study, Barr et al. (2005) noted that Uganda NGOs
might over-report their external evaluations.
Human Resource Systems
Finally, organizations were asked a series of questions related to their human
resource systems. Perhaps reflecting the widespread use of and dependence on
volunteers, even among those organizations that are solely volunteer-run, many have
human resource systems in place.
Twenty-seven organizations (84.4%) reported conducting orientations for new
hires (including volunteers), though the formality of these trainings are likely widely
variable. The same number also reported having written policies regarding hiring and
termination, and offering their employees access to ongoing training. These high
numbers likely reflect the fact that organizational constitutions (required for registration
with the district or national government) typically include a section regarding grounds for
termination, and, with respect to training, one of the functions of the NCSN is to offer
training to employees of member organizations and to alert member organizations of
trainings hosted by partner organizations. Twenty-four organizations (75%) reported
maintaining employment files for employees or volunteers, and 17 organizations (53.1%)
reported having employee manuals that are distributed to new hires.


50
CHAPTER V
NETWORK MAPPING AND RESULTS
While the preceding chapter explored the institutional sustainability of NGOs in
Nansana, this chapter looks more deeply at the relationships among organizations in the
sample. Both through the formal network and through partnerships, organizations
develop relationships in order to enhance their capacity and service delivery. There is
also some evidence that when international aid is delivered to organizations that
participate in the network, those organizations in turn help to support other organizations
with more limited resources, thereby creating synergistic partnerships. Thus, through
better understanding the NCSN, and interorganizational relationships in Nansana more
broadly, we can begin to understand how efficient service delivery can occur through
local NGO initiatives.
To understand the relationships between organizations in the NCSN and, more
broadly, in Nansana, each interviewee was asked about his or her organizations partner
organizations. Organizations were first asked to list all of their partners and were then
asked a series of questions regarding what each relationship entails. Relationships were
mapped using UCINET and NetDraw.3 Figure V.l shows all of the relationships among
the organizations in the sample.
3 The discussion and analysis in this chapter is largely informed by Hanneman and Riddle
(2005).


51
O'
Green = no direct
international ties
Purple = limited
international ties
Red = strong
international ties
o = member of
NCSN
= nonmember
Figure V.l. All network ties.
The circles and square in the map are nodes representing organizations. Their
colors, shapes, and sizes represent attributes of the organizations, or certain relevant
characteristics. For example, organizations represented by circles are members of the
NCSN and organizations represented by squares are not members. The lines represent
relationships between the organizations, with arrows pointing from each interviewee to
each partner it mentioned. The four organizations on the left side of the map are isolates,
which are not connected to the other organizations in the sample. These organizations
did not report having any partnerships with organizations in the sample and were not
mentioned as partners by other organizations. The key is on the right side of the map.
Appendix B contains a glossary of terms related to network analysis.
To avoid overcomplicating the map, only two attributes, or characteristics, were
assigned to the nodes (which represent organizations) in the map. The first relates to
whether the organization is a member of the NCSN. As mentioned, organizations
represented by circles are members of the NCSN, and organizations represented by


52
squares are not members. Only four organizations in the sample are not part of the
NCSN, and, as can be seen in the map, two of these organizations are connected to the
map and two are isolates.
The second attribute relates to the level of an organizations international ties.
The organizations shown in red are those organizations with international offices, boards
of directors, or incorporation. In all but one instance, if an organization has one of these
characteristics, it has all three. Thus, this group consists of organizations that are
internationally based and run. The organizations shown in purple are those with a direct
international funding source, an international board member, or international volunteers.
As such, these organizations, though locally based, have meaningful international ties.
Finally, the organizations shown in green are those with no direct international ties. The
sizes of the nodes also depend on the strength of the organizations international ties.
Furthermore, the map includes all asymmetric, or directional, ties, meaning that
the lines represent each time an organization mentioned another, regardless of whether
the target organization reciprocated the mention. Thus, an organization may appear in
the map as a hub of activity because many organizations mentioned it as a partner or
because it mentioned many other organizations as a partner. Thus, for example,
organizations 15, 21, and 8 appear to be hubs in the map with many arrows coming to
them, while organization 30 appears to be a hub but only with arrows going out from it.
Despite the fact that I used the same definition of partner with each organization I
interviewed, the findings reveal that organizations have their own conceptions of what a
partner is because there was little symmetry in the relationships. In fact, while the overall
map consists of 91 ties, only 24 of these ties are reciprocal, meaning that there are only


53
12 bonded ties (or reciprocal ties in which both organizations mentioned each other as a
partner) in the sample. Figure V.2 shows all bonded ties in the sample.
Green = no direct
international ties
Purple = limited
international ties
Red = strong
international ties
o = member of
NCSN
= nonmember
Figure V.2. Bonded ties.
I discussed this asymmetry with one of the organizations that was often
mentioned as a partner by other organizations, but which mentioned few organizations in
Nansana as partners. This organization limited its conception of partner to strategic
partners, or those whom it specifically sought a partnership with in order to enhance its
capacity. Thus, at the risk of over generalizing, the smaller organizations that considered
this organization to be a partner might have been viewing it as a partner based on its
ability to enhance their capacity, while the organization at the hub largely only
considered its partners to be those outside of Nansana with greater resources and the
ability to enhance its own capacity.
Because of this difference in conception of partnership and because of the limited
number of bonded (or reciprocal) ties among organizations in the sample, in the next
chapter regarding correlations between numbers of partners and measures of institutional
sustainability, I separately analyze measures of in degrees, which refers to the number of


54
times organizations were mentioned as partners, and out degrees, or how many partners
each organization mentioned.
To gain a better understanding of the types of partnerships that organizations in
Nansana tend to engage in, for each partner, I asked organizations a series of questions
related to the strength of the partnerships. These questions included the length of the
partnerships, the formality of the partnerships, whether the organizations have joint
programs, what types of resources they share, and how often they communicate. Again,
the survey can be found in Appendix A.
Partnerships Lengths
Beginning with the lengths of the partnerships, I divided the lengths into three
categories: 1) less than one year, 2) one to three years, and 3) greater than three years. As
mentioned, the NCSN began forming in 2008 and became official in 2010. As Figure
V.3 shows, while few additional partnerships have formed in the last year, there was a
substantial increase in the number of partnerships formed among organizations in the
NCSN in the last three years, perhaps reflecting the effects of the NCSN.
Partnerships longer than three Partnerships of one to three years. All current partnerships.

* *

*
years.

Figure V.3. The addition of partnerships over three years.
The maps show that, along with an increase in the number of partners (possibly
related to the development of the NCSN), there was also a significant increase between
one and three years ago in the number of ways that organizations can connect with each


55
other. In other words, based on the new partnerships that formed, there was an increase
in the connectivity among organizations because, even if they are not directly connected
to one another, organizations are more closely connected indirectly. Furthermore, while
many more additional partnerships are possible in Nansana (only 91 of 992 possible ties
currently exist), there is still considerable connectivity and numerous opportunities or
ways in which various organizations can connect.
Next, with respect to formality, looking at Figure Y.4, it appears that few of the
ties are formal (36 of 91), and, while I asked about a variety of types of formality
(including joint ventures, parent-subsidiary relationships, and co-grant recipients), the
only type of relationship reported were other contractual based relationships, such as
memoranda of understanding. Also, many of the organizations reported having a
contractual relationship with one organization in particular (organization 15). Indeed,
this organization accounts for 18 of the 36 formal relationships (however, this
organization did not report having a formal relationship with any of the others). Thus,
the second map below shows what happens when formal relationships with this single
organization are removed.
Partnership Formality
All formal partnerships.
Formal partnerships excluding organization 15.
Figure V.4. Formal partnerships.


56
While the formality of relationships does not necessarily correlate with
relationship strength or quality, formality could contribute to the sustainability of
relationships. Thus, though the numbers are limited, it does appear that some
organizations are entering into more formal relationships with each other. Furthermore,
many of the organizations clearly perceive their relationships with organization 15 as
formal (though it is not clear whether the relationships are, in fact, formal), perhaps
suggesting that relationships with this organization may seem particularly important to
the organizations involved. If this is the case, organization 15 may be playing an
important role in connecting organizations and facilitating relationships.
Joint Programming Among Partners
The next question regarding the nature of interorganizational partnerships is
whether the organizations have joint programs. In 43 of the 91 ties, organizations
reported having joint programs with one another, and these joint programs primarily
centered around two organizations, as can be seen in Figure V.5. Again, there appeared
to be wide variation in what the organizations considered joint programs, and there was
little agreement among organizations regarding whether they were conducting joint
programs. The fact that two organizations with international ties are often involved in
joint programs, however, may suggest that international resources are being dispersed to
other organizations in the network and possibly used in synergistic relationships between
organizations with more and less resources.


57
Green = no direct
international ties
Purple = limited
international ties
Red = strong
international ties
o = member of
NCSN
= nonmember
Figure V.5. Partnerships involving joint programs.
Resource Sharing Among Partners
Next, I asked organizations about their levels of resource sharing with their
partners. Organizations were asked to indicate which of the following they share with
each of their partners: information, money, clients, physical space, materials, staff or
volunteers, board members, and administrative capacities. The most frequent type of
resource shared is information with 90 of 91 partnerships incorporating information
sharing (in fact, it is difficult to imagine how a partnership occurs without information
sharing). The second most frequent type of resource shared is beneficiaries, which
primarily involves client referrals for various services. Of the 91 partnerships, 45
reportedly include beneficiary sharing. Other types of resources shared include materials,
such as posters, books, and health care items (18 partnerships); staff or volunteers (11
partnerships); and physical space (7 partnerships). Administrative capacities and
finances were each only mentioned once, and no organizations reported sharing board
members.


58
The maps shown in Figure V.6 incorporate the strength of ties between
organizations based on the number of resources they reported sharing with their partners.
Thicker lines reflect more resources being shared (in bonded ties, the higher number of
reported resources dictates the line thickness). As can be seen, few organizations report
sharing more than two resources with any of their partners. Furthermore, only one
partnership reportedly incorporates more than three types of resource sharing, which,
interestingly, is in the partnership between the two organizations with the most in degrees
(or the most mentions by other organizations). One of these organizations reported
sharing five types of resources with the other, and the other reported that they share three
types of resources. Thus, while there are many partnerships within Nansana, there is
little collaboration in that organizations remain largely separate from one another.
All resource sharing. Two or more types of resource sharing.
Three or more types of resource sharing. Four or more types of resource sharing.
Figure V.6. Partnerships by levels of resource sharing.


59
Contact Frequency Among Partners
The final measure of connectivity relates to the frequency of contact between the
organizations. Contact was divided into five categories: variable (which merely indicates
the respondents inability to gauge any pattern in frequency of communication), at least
annually, at least monthly, at least weekly, and daily. Again, thicker lines represent more
frequent contact. In Figure V.7, the first map shows all types of contact between the
organizations; the second map shows all contact that occurs at least monthly (also
excluding variable contact); and the third map shows all contact that occurs at least
weekly. As can be seen, the majority of contact occurs at least monthly, but not weekly.
Only two organizations reported communicating daily, which may be attributable to these
organizations close proximity to one another. It should also be noted that, particularly in
large organizations, it may be difficult for a single organizational representative to
estimate frequency of contact.
All contact. At least monthly contact. At least weekly contact.
Figure V.7. Partnerships by frequency of contact.
Additional Analyses of Partnerships
As mentioned, for the majority of the maps in this chapter, I chose to include only
two node attributes (international ties and network membership) in order to maintain
readability. There are numerous other attributes that could have been included, however,
such as number of employees, financial resources, NGO/CBO/FBO classification,


60
number of beneficiaries, primary service area, and so forth. Many of these would offer
useful additional information.
The only additional attribute I looked at is financial resources. For this attribute, I
divided the organizations into quartiles based on their finances. Four organizations are
not included in this map because I was unable to obtain financial data for them. Red
organizations are those in the first quartile; yellow organizations are those in the second
quartile; green organizations are those in the third quartile; and blue organizations are
those in the fourth quartile. As can be seen in Figure V.8, the majority of organizations
with the most in degrees are in the fourth quartile based on their actual budgets, showing
that better funded organizations are more frequently mentioned as partners by other
organizations. There is some variation, though, perhaps reflecting other types of
resources at play or reluctance on the part of some organizations with substantial
resources to partner with other organizations in Nansana. Again, this suggests that
organizations in Nansana may have efficiently organized themselves to spread resources
and enhance service delivery.


61
Red = first
quartile
second quartile
Green = third
quartile
Blue = fourth
quartile
Figure V.8. Network map including financial resources attribute.
Looking back at the literature regarding cooperation, coordination, and
collaboration, it seems that most of the relationships represented in these network maps
are either cooperations or coordinations. Due to the lack of reciprocity in how various
organizations describe their relationships with one another, however, it is difficult to
categorize the relationships. It seems that most, if not all, of the organizations in the
network are engaged in some form of cooperation with each of the other organizations in
the network.
With respect to the partners specifically identified by organizations, except for
those that are merely information sharing, most of these organizations are likely engaged
in coordination. There appears to be little collaboration occurring between organizations
in that the organizations remain largely separate from one another. The NCSN, however,
is likely a collaboration in that it consists of organizations coming together to form a new
structure. The NCSN may even be a community-building collaboration in that it affects


62
and benefits the entire community in which it is embedded, not just its member
organizations.
Thus, the organizations in Nansana are well connected with one another, and they
have increased their connectivity considerably since the NCSN became active. While the
organizations actually share little in the way of resources, the fact that those with
substantial resources tend to have many partners suggests that productive partnerships
exist. Indeed, according to Claiborne et al. (2009), relationships between macro-level
organizations (including larger NGOs) and micro-level organizations are important to
capacity building.


63
CHAPTER VI
INSTITUTIONAL SUSTAINABILITY AND PARTNERSHIPS
This chapter tests the primary hypothesis of this thesis, which is that there is a
positive correlation between the number of partners an organization has and its
institutional sustainability. This thesis does not purport to rank organizations according
to their institutional sustainability. Rather, the question is to what extent networks and
partnerships correlate with certain measures of institutional sustainability.
The results from three separate sets of analyses are reported in this chapter. In the
first set, organizations that are members of the NCSN are compared with organizations
that are not members of the NCSN. In the second set, organizations are compared based
on their numbers of out degrees, and, in the third set, organizations are compared based
on their numbers of in degrees. For each set of analyses, organizations are compared
based on each of the eight categories and 15 measures of institutional sustainability
described in Chapter III.
Differences Between NCSN Members and Nonmembers
In this section, I compare organizations in Nansana that are members of the
NCSN with organizations that are not members of the NCSN. As mentioned, however,
28 of the organizations interviewed are members of the NCSN, while only four
organizations are not members of the NCSN. Given the small sample in the nonmember
group, t-tests and chi-squared measures did not indicate that the results in this section are
significant. The results are summarized in Table VI. 1.


64
Table VI.1. Comparisons based on NCSN membership.
Institutional Sustainability Measures NCSN members NCSN nonmembers
Services and Programs years of operation 8.89 4.50
# of programs 5.43 5.25
People # of beneficiaries 745.46 374.00
# of employees 9.43 13.00
# of volunteers 8.41 5.00
Facilities # of facilities 2.00 1.50
ownership rate 39.29% 0.00%
# of offices outside Nansana 1.93 2.50
Technology # of tech items 5.32 5.00
# of computers 4.85 3.00
Finances actual budget $44,366.36 $6,167.33
actual: expected 55.16% 50.43%
Records points out of 5 4.68 4.50
Program Evaluations rate conducting 82.14% 100.00%
Human Resource Systems points out of 5 4.15 4.67


65
Services and Programs
Beginning with the measures of institutional sustainability related to services and
programs, NCSN organizations have been operating for an average of 8.89 years, while
nonmembers have been operating for only 4.5 years. There is little difference between
the two groups with respect to their numbers of programs, with NCSN organizations
having an average of 5.43 programs and nonmembers having an average of 5.25
programs.
People
In the people categories, NCSN organizations have an average of 745.46
beneficiaries, while nonmembers organizations have an average of 374.00 beneficiaries.
NCSN organizations employ an average of 9.43 people, and nonmember organizations
employ an average of 13 people. Finally, NCSN organizations have an average of 8.41
volunteers, and nonmember organizations have an average of 5 volunteers.
Facilities
The next category of institutional sustainability relates to organizations physical
space or facilities. NCSN organizations have an average of two types of facilities at their
Nansana premises, while nonmember organizations have an average of 1.5 types of
facilities at their Nansana premises. Among NCSN organizations, 15 rent their facilities
in Nansana; 11 own their facilities in Nansana; and two do not have offices. All four of
the nonmember organizations rent their facilities in Nansana. With respect to offices
outside of Nansana, NCSN organizations have 1.93 offices outside of Nansana (either
within Uganda or internationally), and nonmember organizations have 2.5 offices outside
of Nansana.


66
Technology
The next category of institutional sustainability relates to an organizations
technology. Organizations received a score out of 10 related to the number of specified
technology systems that they have (website, network, electronic records, domain name,
organization e-mail account, organization phone, computer/laptop, fax machine, scanner,
and printer). NCSN organizations scored an average of 5.32 out of 10, and nonmember
organizations scored an average of 5. NCSN organizations also have an average of 4.85
computers or laptops, and nonmember organizations have an average of 3 computers or
laptops.
Finances
Here, I use two measures of institutional sustainability related to an organizations
finances: its actual budget and its actual budget as a percentage of its expected budget.
NCSN organizations have an average actual annual budget of $44,366.36, and
nonmember organizations have an average actual annual budget of $6,167.33. Also, the
actual budget to expected budget ratio for NCSN organizations is 55.16%, and the ratio
for nonmember organizations is 50.43%.
Records
There was little difference in record keeping between the two groups. Out of a
total possible score of 5, NCSN organizations received an average score of 4.68, while
nonmember organizations scored 4.50.
Program Evaluations
The next measure looks at whether organizations conduct some sort of evaluation
of their programs. All four of the nonmember organizations report conducting


67
evaluations of their programs, while 23 of 28 NCSN organizations report conducting
evaluations.
Human Resource Systems
Finally, organizations received a score out of five related to their human resource
systems. NCSN organizations scored an average of 4.15, while nonmember
organizations scored 4.67.
Significant Findings Regarding NCSN Membership
Though the results in this category were not significant in light of the small-n,
there do appear to be some differences between NCSN members and nonmembers.
Fairly large differences between the groups were found in many categories, and, in most
instances, member organizations are faring better than nonmember organizations.
Differences Based on Numbers of Out Degrees
The second set of analyses performed in relation to organizations connectivity is
based on their number of out degrees, or the number of partners that each organization
reports having among those organizations in the sample. All organizations in the sample
were included in this analysis. The mean number of out degrees is 2.84, with a median of
3.00 and a standard deviation of 2.13. To look at correlations between numbers of out
degrees and the various measures of institutional sustainability, organizations were
divided into three groups. The first group of organizations reported having no or one ties
with other organizations in the sample (n=9). The second group has two or three ties
(n=13), and third group has four or more ties (n=10). Unless otherwise stated, a one-way
ANOVA was used to test differences in mean scores between the groups. Table VI.2 is a
summary of the results in this section.


68
Table VI.2. Comparisons based on numbers of out degrees.
Institutional Sustainability Measures Group 1 (0-1 out degrees) Group 2 (2-3 out degrees) Group 3 (4+ out degrees)
Services and Programs years of operation 10.44 7.08 8.10
# of programs 5.22 5.08 6.00
People # of beneficiaries 496.89 394.85 1276.00
# of employees 19.11 3.85 9.40
# of volunteers 6.86 6.31 11.20
Facilities # of facilities* 1.89 1.42 2.60
ownership rate* 33.33% 7.69% 70.00%
# of offices outside Nansana 2.78 2.23 1.00
Technology # of tech items 5.67 5.00 5.30
# of computers 3.20 3.25 7.25
Finances actual budget $49,202.71 $27,797.27 $47,747.20
actual: expected 52.05% 56.76% 53.92%
Records points out of 5 4.56 4.54 4.90
Program Evaluations rate conducting 77.78% 76.92% 100.00%
Human Resource Systems points out of 5 4.13 4.00 4.50
* Differences are significant at p < .05


69
Services and Programs
Beginning with services, organizations in Group 1 (0 or 1 ties) have been
operating for an average of 10.44 years. Organizations in Group 2 (2 or 3 ties) have been
operating for an average of 7.08 years, and organizations in Group 3 (4 or more ties) have
been operating for 8.10 years. With respect to programs, organizations in Group 1 have
an average of 5.22 programs. Organizations in Group 2 have an average of 5.08
programs, and organizations in Group 3 have an average of 6 programs. For each of
these categories, the differences between the groups are not significant, as shown in
Tables VI.3 and VI.4.
Table VI.3. Comparison of years of operation based on numbers of out degrees.
N Mean Standard Deviation 95% Confide for iV nee Interval ean
Lower Bound Upper Bound
Group 1 9 10.44 10.631 2.27 18.62
Group 2 13 7.08 6.751 3.00 11.16
Group 3 10 8.10 4.701 4.74 11.46
F = 5.38, p = .590, Eta = .036
Table VI.4. Comparison of programs based on numbers of out degrees.
N Mean Standard Deviation 95% Confide for iV nee Interval ean
Lower Bound Upper Bound
Group 1 9 5.22 1.394 4.15 6.29
Group 2 13 5.08 2.253 3.72 6.44
Group 3 10 6.00 2.000 4.57 7.43
F = .676, p = .517, Eta = .045


People
70
Looking at the various people associated with organizations, organizations in
Group 1 serve an average of 496.89 beneficiaries. Organizations in Group 2 serve an
average of 394.85 beneficiaries, and organizations in Group 3 serve an average of 1,276
beneficiaries. With respect to employees, organizations in Group 1 have an average of
19.11 employees. Organizations in Group 2 have an average of 3.85 employees, and
organizations in Group 3 have an average of 9.40 employees. Finally, organizations in
Group 1 have an average of 6.86 volunteers. Organizations in Group 2 have an average
of 6.31 volunteers, and organizations in Group 3 have an average of 11.20 volunteers.
Again, these results are not significant, as shown in Tables VI.5, VI.6, and VI.7.
Table VI.5. Comparison of beneficiaries based on numbers of out degrees.
N Mean Standard Deviation 95% Confide for IV nee Interval ean
Lower Bound Upper Bound
Group 1 9 496.89 696.989 -38.86 1,032.64
Group 2 13 394.85 547.578 63.95 725.74
Group 3 10 1,276.40 3,388.294 - 1,147.44 3,700.24
F = 642, p = .534, Eta = 042
Table VI.6. Comparison of employees based on numbers of out degrees.
N Mean Standard Deviation 95% Confide for IV nee Interval ean
Lower Bound Upper Bound
Group 1 9 19.11 26.408 - 1.19 39.41
Group 2 13 3.85 5.505 .52 7.17
Group 3 10 9.40 17.128 -2.85 21.65
F = 2.099, p = 141, Eta = 126


71
Table VI.7. Comparison of volunteers based on numbers of out degrees.
N Mean Standard Deviation 95% Confide for iV nee Interval ean
Lower Bound Upper Bound
Group 1 7 6.86 6.176 1.15 12.57
Group 2 13 6.31 4.211 3.76 8.85
Group 3 10 11.20 14.367 .92 21.48
F = .873, p = .429, Eta = .061
Facilities
The next category of institutional sustainability relates to organizations offices
and facilities. Beginning with the number of facilities each organization has on its
premises in Nansana, organizations in Group 1 have an average of 1.89 facilities in
Nansana. Organizations in Group 2 have an average of 1.42 facilities in Nansana, and
organizations in Group 3 have an average of 2.60 facilities in Nansana. These results are
significant at p < .05 (p = .035), as shown in Table VI.8. In line with my general
hypothesis, Group 3 has the greatest number of facilities, but Group 1 has more facilities
than Group 2, perhaps suggesting a pattern wherein organizations in Group 1 have more
resources than those in Group 2.
Table VI.8. Comparison of Nansana facilities based on numbers of out degrees.
N Mean Standard Deviation 95% Confide for iV nee Interval ean
Lower Bound Upper Bound
Group 1 9 1.89 1.453 .77 3.01
Group 2 12 1.42 .669 .99 1.84
Group 3 10 2.60 .843 2.00 3.20
F = 3.805, p = .035, Eta = .214


72
Next, with respect to whether organizations rent or own their facilities, five of
nine organizations in Group 1 rent their facilities; three own; and one does not have an
office. In Group 2, 12 of 13 organizations rent their facilities, and one owns its facilities.
In Group 3, two organizations rent their facilities, seven own their facilities, and one does
not have an office. Chi-square tests indicate that these differences are significant (p
= .012), as shown in Table VI.9.
Table VI.9. Comparison of ownership rates based on numbers of out degrees.
Rent Own No office Row N
Group 1 55.6% 33.3% 11.1% 9
Group 2 92.3% 7.7% 0.0% 13
Group 3 20.2% 70.0% 10.0% 10
Chi2 = 12.773, cLf = 4, p = 012
Finally, looking at the number of offices each organization has outside Nansana
(both elsewhere in Uganda and internationally), organizations in Group 1 have an
average of 2.78 offices outside Nansana. Organizations in Group 2 have an average of
2.23 offices elsewhere, and organizations in Group 3 have an average of one office
elsewhere. These results are not significant, as shown in Table VI. 10.
Table VI.10. Comparison of offices outside Nansana based on numbers of out
degrees.
N Mean Standard Deviation 95% Confide for IV nee Interval ean
Lower Bound Upper Bound
Group 1 9 2.78 2.774 .65 4.91
Group 2 13 2.23 5.003 - .79 5.25
Group 3 10 1.00 1.414 - .01 2.01
F = 616, p = .547, Eta = 041


73
Technology
Out of 10 possible technology systems, organizations in Group 1 averaged 5.67
technology systems. Organizations in Group 2 averaged 5 technology systems, and
organizations in Group 3 averaged 5.3 technology systems. Organizations in Group 1
also have an average of 3.20 computers or laptops. Organizations in Group 2 have an
average of 3.25 computers or laptops, and organizations in Group 3 have an average of
7.25 computers or laptops. These results are not significant, as shown in Tables VI. 11
and VI. 12.
Table VI.11. Comparison of technology systems based on numbers of out degrees.
N Mean Standard Deviation 95% Confide for IV nee Interval ean
Lower Bound Upper Bound
Group 1 9 5.67 2.646 3.63 7.70
Group 2 13 5.00 2.309 3.60 6.40
Group 3 10 5.30 2.359 3.61 6.99
F = .202, p = .818, Eta = .014
Table VI.12. Comparison of computers based on numbers of out degrees.
N Mean Standard Deviation 95% Confide for IV nee Interval ean
Lower Bound Upper Bound
Group 1 5 3.20 1.095 1.84 4.56
Group 2 8 3.25 2.121 1.48 5.02
Group 3 8 7.25 11.961 -2.75 17.25
F = 694, p = .513, Eta = .072
Finances
Results with respect to financial measures were also not significant, as shown in
Tables VI. 13 and VI. 14. Organizations in Group 1 have an actual annual budget of


74
$49,202.71. Organizations in Group 2 have an actual budget of $27,797.27, and
organizations in Group 3 have an actual budget of $47,747.20. Furthermore,
organizations in Group 1 have an actual to expected budget ratio of 52.05%. The ratio
for organizations in Group 2 is 56.76%, and the ratio for organizations in Group 3 is
53.92%.
Table VI.13. Comparison of actual budget based on numbers of out degrees.
N Mean Standard Deviation 95% Confide for iV nee Interval ean
Lower Bound Upper Bound
Group 1 7 49,202.71 109,842.712 - 52,384.85 150,790.28
Group 2 11 27,797.27 44,037.646 - 1,787.62 57,382.17
Group 3 10 47,747.20 92,386.076 - 18,341.82 113,836.22
F = .210, p = .812, Eta = .017
Table VI.14. Comparison of budget ratio based on numbers of out degrees.
N Mean Standard Deviation 95% Confide for iV nee Interval ean
Lower Bound Upper Bound
Group 1 7 .5205 .46849 .0872 .9538
Group 2 11 .5676 .28710 .3747 .7604
Group 3 8 .5392 .38914 .2138 .8645
F = 036, p = .965, Eta = 003
Records
Looking at record keeping as a category of institutional sustainability, out of a
score of 5, organizations in Group 1 received an average score of 4.56. Organizations in
Group 2 received an average score of 4.54, and organizations in Group 3 received an
average score of 4.90. These results are not significant, as shown in Table VI. 15.


75
Table VI.15. Comparison of record keeping based on numbers of out degrees.
N Mean Standard Deviation 95% Confide for IV nee Interval ean
Lower Bound Upper Bound
Group 1 9 4.56 .527 4.15 4.96
Group 2 13 4.54 1.198 3.81 5.26
Group 3 10 4.90 .316 4.67 5.13
F = .617, p = .547, Eta = .041
Program Evaluations
Seven of nine organizations in Group 1 report conducting some sort of
evaluations of their programs. Ten of 13 organizations in Group 2 report conducting
some sort of evaluations, and all 10 of the organizations in Group 3 report conducting
evaluations. Chi-square tests were performed on these outcomes, and the results were not
significant, as shown in Table VI. 16.
Table VI.16. Comparison of evaluations based on numbers of out degrees.
Yes No Row N
Group 1 77.8% 22.2% 9
Group 2 76.9% 23.1% 13
Group 3 100.0% 0.0% 10
Chi2 = 2.697, d.f.=2,p = .260
Human Resource Systems
Finally, results with respect to differences among the groups related to human
resource systems were not significant, as shown in Table VI. 17. Out of a score of 5,
organizations in Group 1 received a score of 4.13. Organizations in Group 2 received a
score of 4, and organizations in Group 3 received a score of 4.5.


76
Table VI.17. Comparison of human resource systems based on numbers of out
degrees.
N Mean Standard Deviation 95% Confide for IV nee Interval ean
Lower Bound Upper Bound
Group 1 8 4.13 1.458 2.91 5.34
Group 2 11 4.00 1.265 3.15 4.85
Group 3 10 4.50 .707 3.99 5.01
F = .508, p = .607, Eta = .038
Significant Findings Regarding Out Degrees
Thus, results with respect to out degrees, or the number of partners mentioned by
each organization, were only significant in two categories: the number of facilities the
organizations have in Nansana and their rate of ownership of their properties in Nansana.
As mentioned in Chapter V, this may be reflective of a validity issue with respect to the
definition of partner, in that organizations have varied conceptions of partnerships.
Furthermore, the results suggest a parabolic relationship between numbers of partners
mentioned and institutional sustainability, perhaps reflecting reluctance on the part of
certain organizations to partner within Nansana out of fear of competition for resources.
These and other results are discussed in greater detail in Chapter VIII.
Differences Based on Numbers of In Degrees
For the final set of analyses in this chapter, I look at all organizations in the
sample and assign them to groups based on their numbers of in degrees, or the number of
times each organization was mentioned as a partner by other organizations in the sample.
The mean number of in degrees is 2.84, with a median of 1.00 and a standard deviation of
5.27 (considerably greater than the standard deviation of 2.13 related to numbers of out
degrees). Because of the greater variation between numbers of in degrees, in this section,


77
organizations are divided into four groups. The first group of organizations was never
mentioned as a partner by other organizations in the sample (n=13). The second group
was mentioned by 1 or 2 organizations (n=l 1). The third group was mentioned by 3 to 7
organizations as a partner (n=5), and the fourth group consists of 3 organizations that
were mentioned as partners 13, 18, and 22 times respectively. Again, unless otherwise
stated, a one-way ANOVA was used to test differences in mean scores between the
groups. An overview of the results in this section is presented in Table VI. 18.


78
Table VI.18. Comparisons based on numbers of in degrees.
Institutional Sustainability Measures Group 1 (no in degrees) Group 2 (1-2 in degrees) Group 3 (3-7 in degrees) Group 4 (13-22 in degrees
Services and Programs years of operation 5.77 10.09 7.00 15.33
# of programs 4.77 5.91 6.80 4.00
People # of beneficiaries* 235.54 473.82 180.20 4,398.00
# of employees 10.15 8.91 5.00 20.33
# of volunteers 7.00 7.18 12.80 7.50
Facilities # of facilities 1.42 2.18 2.60 2.00
ownership rate 15.38% 27.27% 60.00% 100.00%
# of offices outside Nansana 3.00 0.73 1.20 3.67
Technology # of tech items 4.54 4.91 6.60 7.67
# of computed 2.88 2.60 3.60 15.33
Finances actual budget $19,684.70 $41,667.10 $31,645.00 $118,639.33
actual: expected 37.81% 57.07% 65.64% 100.00%
Records points out of 5 4.62 4.55 5.00 4.67
Program Evaluations rate conducting 84.62% 72.73% 100.00% 100.00%
Human Resource Systems points out of 5 3.85 4.33 4.75 4.67
* Differences are significant at p < .05
| Differences are significant at p < 1


79
Services and Programs
Beginning with measures of institutional sustainability related to organizations
services and programs, the two measures discussed here are the number of years
organizations have existed and the number of programs each organization conducts.
Organizations in Group 1 (those with no in degrees) have been in existence for an
average of 5.77 years. Organizations in Group 2 (those with 1 or 2 in degrees) have been
in existence for an average of 10.09 years. Organizations in Group 3 (those with 3 to 7 in
degrees) have been in existence for an average of 7 years, and organizations in Group 4
(those with 13 to 22 in degrees) have been in existence for an average of 15.33 years.
These results are not significant, as shown in Table VI. 19.
Table VI.19. Comparison of years of operation based on numbers of in degrees.
N Mean Standard Deviation 95% Confide for IV nee Interval ean
Lower Bound Upper Bound
Group 1 13 5.77 5.600 2.39 9.15
Group 2 11 10.09 9.513 3.70 16.48
Group 3 5 7.00 4.062 1.96 12.04
Group 4 3 15.33 6.658 - 1.21 31.87
F = 1.790, p = .172, Eta = .161
With respect to numbers of programs, organizations in Group 1 have an average
of 4.77 programs. Organizations in Group 2 have an average of 5.91 programs.
Organizations in Group 3 have an average of 6.80 programs, and organizations in Group
4 have an average of 4 programs. These results are arguably significant at p < 1 (p
= .094), as shown in Table VI.20. Thus, interestingly, the number of programs per
organization tends to increase with the number of in degrees for the first three groups, but


80
then it decreases among those organizations with the most in degrees, perhaps reflecting a
recognition among organizations with the most in degrees of the need to focus attention
on a more limited number of programs. Also, many organizations with greater resources
tend to have contracts or grants for their programming, meaning that their only programs
are those that are specifically funded.
Table VI.20. Comparison of programs based on numbers of in degrees.
N Mean Standard Deviation 95% Confide for IV nee Interval ean
Lower Bound Upper Bound
Group 1 13 4.77 1.787 3.69 5.85
Group 2 11 5.91 1.578 4.85 6.97
Group 3 5 6.80 2.683 3.47 10.13
Group 4 3 4.00 1.000 1.52 6.48
F = 2.353, p = .094, Eta = .201
People
Looking next at measures of institutional sustainability related to the various
people associated with organizations, organizations in Group 1 serve an average of
235.54 beneficiaries. Organizations in Group 2 serve an average of 473.82 beneficiaries.
Organizations in Group 3 serve an average of 180.20 beneficiaries, and organizations in
Group 4 serve an average of 4398.00 beneficiaries. These results are significant (p
= .002), as shown in Table VI.21. This suggests that organizations with the most in
degrees serve considerably more beneficiaries than those with less in degrees, perhaps
reflecting that organizations with greater in degrees may be seen as partners because of
their increased capacity to serve clients and their ability to receive clients from other
organizations.


81
Table VL21. Comparison of beneficiaries based on numbers of in degrees.
N Mean Standard Deviation 95% Confide for iV nee Interval ean
Lower Bound Upper Bound
Group 1 13 235.54 262.570 76.87 394.21
Group 2 11 473.82 626.788 52.74 894.90
Group 3 5 180.20 236.622 - 113.61 474.01
Group 4 3 4,398.00 5,700.431 - 9,762.66 18,558.66
F = 6.109, p = .002, Eta = .295
The results were not significant with respect to differences in numbers of
employees and numbers of volunteers, as shown in Tables VI.22 and VI.23.
Organizations in Group 1 have an average of 10.15 employees. Organizations in Group 2
have an average of 8.91 employees. Organizations in Group 3 have an average of 5
employees, and organizations in Group 4 have an average of 20.33 employees.
Organizations in Group 1 have an average of 7 volunteers. Organizations in Group 2
have an average of 7.18 volunteers. Organizations in Group 3 have an average of 12.80
volunteers, and organizations in Group 4 have an average of 7.50 volunteers.
Table VI.22. Comparison of employees based on numbers of in degrees.
N Mean Standard Deviation 95% Confide for iV nee Interval ean
Lower Bound Upper Bound
Group 1 13 10.15 15.383 .86 19.45
Group 2 11 8.91 21.012 - 5.21 23.02
Group 3 5 5.00 4.690 - .82 10.82
Group 4 3 20.33 31.817 - 58.70 99.37
F = .457, p = 715, Eta = .047


82
Table VI.23. Comparison of volunteers based on numbers of in degrees.
N Mean Standard Deviation 95% Confide for IV nee Interval ean
Lower Bound Upper Bound
Group 1 12 7.00 8.464 1.62 12.38
Group 2 11 7.18 4.813 3.95 10.42
Group 3 5 12.80 17.978 -9.52 35.12
Group 4 2 7.50 .707 1.15 13.85
F = 5.06, p = .682, Eta = .055
Facilities
Looking at various measures of physical space as a component of institutional
sustainability, with respect to the number of facilities each organization has on its
premises in Nansana, organizations in Group 1 have an average of 1.42 facilities.
Organizations in Group 2 have an average of 2.18 facilities. Organizations in Group 3
have an average of 2.60 facilities, and organizations in Group 4 have an average of 2.00
facilities. These results are not significant, as shown in Table VI.24.
Table VI.24. Comparison of Nansana facilities based on numbers of in degrees.
N Mean Standard Deviation 95% Confide for IV nee Interval ean
Lower Bound Upper Bound
Group 1 12 1.42 .900 .84 1.99
Group 2 11 2.18 1.250 1.34 3.02
Group 3 5 2.60 .894 1.49 3.71
Group 4 3 2.00 1.000 - .48 4.48
F = 1.851, p = .162, Eta = .171
Next, in Group 1, 9 of 13 organizations rent their Nansana facilities; 2 own their
facilities; and 2 have no facilities. In Group 2, 8 of 11 organizations rent their facilities,


83
and 3 own their facilities. In Group 3, of the five organizations, 2 rent their facilities and
3 own their facilities. In Group 4, all three of the organizations own their facilities. Chi-
square tests were performed on these measures, and the results are arguably significant at
p < .1 (p = .067), as shown in Table VI.25.
Table VI.25. Comparison of ownership rates based on numbers of in degrees.
Rent Own No office Row N
Group 1 69.2% 15.4% 15.4% 13
Group 2 72.7% 27.3% 0.0% 11
Group 3 40.0% 60.0% 0.0% 5
Group 4 0.0% 100.0% 0.0% 3
Chi2 = 11.802, cLf = 6, p = 067
With respect to locations outside of Nansana (either in Uganda or internationally),
organizations in Group 1 have an average of 3.00 locations. Organizations in Group 2
have an average of 0.73 locations. Organizations in Group 3 have an average of 1.20
locations, and organizations in Group 4 have an average of 3.67 locations. These
differences are not significant, as shown in Table VI.26.
Table VI.26. Comparison of offices outside Nansana based on numbers of in
degrees.
N Mean Standard Deviation 95% Confide for IV nee Interval ean
Lower Bound Upper Bound
Group 1 13 3.00 4.916 .03 5.97
Group 2 11 .73 1.421 - .23 1.68
Group 3 5 1.20 1.643 - .84 3.24
Group 4 3 3.67 4.041 -6.37 13.71
F = 1.118, p = .359, Eta = .107


84
Technology
Differences in the organizations technological capabilities are arguably
significant. Organizations in Group 1 have an average of 4.54 (out of 10) technology
systems. Organizations in Group 2 have an average of 4.91 technology systems.
Organizations in Group 3 have an average of 6.60 technology systems, and organizations
in Group 4 have an average of 7.67 technology systems. These results are significant at p
< .1 (p = .096), as shown in Table VI.27.
Table VI.27. Comparison of technology systems based on numbers of in degrees.
N Mean Standard Deviation 95% Confide for IV nee Interval ean
Lower Bound Upper Bound
Group 1 13 4.54 2.402 3.09 5.99
Group 2 11 4.91 2.256 3.39 6.42
Group 3 5 6.60 2.074 4.03 9.17
Group 4 3 7.67 .577 6.23 9.10
F = 2.331, p = .096, Eta = .200
Also, organizations in Group 1 have an average of 2.88 computers or laptops.
Organizations in Group 2 have an average of 2.60 computers or laptops. Organizations in
Group 3 have an average of 3.60 computers or laptops. Organizations in Group 4 have an
average of 15.33 computers or laptops. These results are significant at p < .1 (p = .055),
as shown in Table VI.28.


85
Table VI.28. Comparison of computers based on numbers of out degrees.
N Mean Standard Deviation 95% Confide for IV nee Interval ean
Lower Bound Upper Bound
Group 1 8 2.88 1.458 1.66 4.09
Group 2 5 2.60 1.517 .72 4.48
Group 3 5 3.60 3.578 - .84 8.04
Group 4 3 15.33 18.009 - 29.40 60.07
F = 3.083, p = 055, Eta = 352
Finances
Again, the two measures used in this study to assess financial components of
institutional sustainability are organizations actual annual budgets and the percentage of
the expected budget that the actual budget represents. Organizations in Group 1 have an
average actual annual budget of $19,684.70. Organizations in Group 2 have an average
budget of $41,667.10. Organizations in Group 3 have an average budget of $31,645.00,
and organizations in Group 4 have an average budget of $118,639.33. These results are
not significant, as shown in Table VI.29.
Table VI.29. Comparison of actual budget based on numbers of in degrees.
N Mean Standard Deviation 95% Confide for IV nee Interval ean
Lower Bound Upper Bound
Group 1 10 19,684.70 46,164.650 - 13,339.50 52,708.90
Group 2 10 41,667.10 91,429.504 -23,737.63 107,071.83
Group 3 5 31,645.00 37,612.201 - 15,056.72 78,346.72
Group 4 3 118,639.33 155,408.339 -267,416.38 504,695.05
F = 1.243, p = .316, Eta = .134


86
With respect to the actual budget to expected budget ratio, organizations in Group
1 have an average percent of 37.81%. Organizations in Group 2 have an average percent
of 57.07%. Organizations in Group 3 have an average percent of 65.64%, and
organizations in Group 4 have an average of 100%. These results are not significant (p
= .108), as shown in Table VI.30.
Table VI.30. Comparison of budget ratio based on numbers of in degrees.
N Mean Standard Deviation 95% Confide for IV nee Interval ean
Lower Bound Upper Bound
Group 1 10 .3781 .32815 .1434 .6129
Group 2 9 .5707 .34562 .3051 .8364
Group 3 5 .6564 .36260 .2062 1.1066
Group 4 2 1.0000 .00000 1.0000 1.0000
F = 2.276, p = .108, Eta = .237
Records
The results are not significant with respect to differences in record keeping
between the four groups, as shown in Table VI.31. Out of a score of 5, organizations in
Group 1 scored an average of 4.62. Organizations in Group 2 scored an average of 4.55.
Organizations in Group 3 scored an average of 5.00, and organizations in Group 4 scored
an average of 4.67.


87
Table VL31. Comparison of record keeping based on numbers of in degrees.
N Mean Standard Deviation 95% Confide for iV nee Interval ean
Lower Bound Upper Bound
Group 1 13 4.62 .650 4.22 5.01
Group 2 11 4.55 1.214 3.73 5.36
Group 3 5 5.00 .000 5.00 5.00
Group 4 3 4.67 .577 3.23 6.10
F = .341, p = 796, Eta = .035
Program Evaluations
Of the 13 organizations in Group 1,11 conduct some sort of evaluations of their
programs. Of the 11 organizations in Group 2, 8 conduct evaluations of their programs.
In Groups 3 and 4, all of the organizations conduct evaluations of their programs. These
results are not significant according to chi-square tests, as shown in Table VI.32.
Table VI.32. Comparison of evaluations based on numbers of in degrees.
Yes No Row N
Group 1 84.6% 15.4% 13
Group 2 72.7% 27.3% 11
Group 3 100.0% 0.0% 5
Group 4 100.0% 0.0% 3
Chi2 = 2.614, d.f. = 3, p = .455
Human Resource Systems
Finally, the results with respect to human resource systems were also not
significant, as shown in Table VI.33. Out of a score of 5, organizations in Group 1
received an average score of 3.85. Organizations in Group 2 received an average score of


88
4.33. Organizations in Group 3 received an average score of 4.75, and organizations in
Group 4 received an average score of 4.67.
Table VI.33. Comparison of human resource systems based on numbers of in
degrees.
N Mean Standard Deviation 95% Confide for IV nee Interval ean
Lower Bound Upper Bound
Group 1 13 3.85 1.519 2.93 4.76
Group 2 9 4.33 .707 3.79 4.88
Group 3 4 4.75 .500 3.95 5.55
Group 4 3 4.67 .577 3.23 6.10
F = .919, p = .446, Eta = .099
Significant Findings Regarding In Degrees
Thus, with respect to groupings based on number of in degrees, or the number of
times each organization was mentioned as a partner by other organizations, results were
significant, at least at p < 0.1 in five categories: number of programs, number of
beneficiaries, property ownership rate in Nansana, number of technology items, and
number of computers. These results offer some support for the hypothesis that
organizations with higher levels of connectivity also have higher levels of institutional
sustainability.
While the above findings do suggest a relationship between institutional
sustainability and interorganizational partnerships, the lack of significant findings across
many categories begs the question of whether other factors might be affecting the results.
As I interviewed organizations in Nansana, I was struck by the importance of
international ties. Organizations with international ties play significant roles in the
community, and organizations without international ties actively seek international


89
connections. As such, the following chapter goes beyond testing my hypothesis and
looks at the role of international ties among NGOs in Nansana.


Full Text

PAGE 1

INTERORGANIZATIONAL RELATI ONSHIPS AND NONGOVERNMENTAL ORGANIZATION INSTITUTIONAL SUSTAINABILITY IN UGANDA by JENNIFER A. KAGAN B.A., New York University, 2000 J.D., University of San Diego, 2004 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Master of Public Administration Public Affairs and Administration 2013

PAGE 2

ii 2013 JENNIFER A. KAGAN ALL RIGHTS RESERVED

PAGE 3

iii This thesis for the Master of Public Administration degree by Jennifer A. Kagan has been approved for the Public Affairs and Administration Program by Brian Gerber, Chair Christine Martell Betcy Jose April 18, 2013

PAGE 4

iv Kagan, Jennifer A. (M.P.A., Public Affairs and Administration) Interorganizational Relationships and Nongovernment al Organization Institutional Sustainability in Uganda Thesis directed by Associate Professor Brian Gerber. ABSTRACT This thesis examines a network of nongovernmental organizations (NGOs) operating in a suburban community in Uganda and it explores the relationship between interorganizational partnerships and institu tional sustainability. Most of the subject NGOs are members of a formal, locally de veloped network, called the Nansana Civil Society Network. Through examining both the formal and informal network, we can begin to understand how service delivery o ccurs through local initiatives and how the shift from first generation NGO work, including relief and we lfare, to second and third generation work, involving loca l self-reliance and sustainabl e systems development, is occurring. This thesis sets out to understand a network of NGOs and the relationship between interorganizational pa rtnerships and institutional su stainability. Organizations were surveyed regarding eight categories of institutional sustainability (services and programs, people, facilities, technology, fi nances, records, program evaluations, and human resource systems) and their partnerships with other organizations. Quantitative tests regarding correlations be tween measures of instituti onal sustainability and numbers of partners show that there is a stronger relationship between inst itutional sustainability and the number of times an organization was mentioned as a partner by other organizations than between institutional sust ainability and the num ber of partners an

PAGE 5

v organization mentioned. Specifically, there ar e significant positive correlations between the number of times an organization was mentioned as a partner and numbers of programs, numbers of benefi ciaries, property ownership rates, numbers of technology systems, and numbers of computers. Furthermore, seven measures of institu tional sustainability correlate with the strength of organizationsÂ’ international ties, offering some evidence that institutional sustainability may be a func tion of international ties. While this suggests that international aid may be efficiently distribut ed through networks of organizations, it also suggests that much progress remains before local self-reliance and sustainable systems development are achieved. The form and content of this abstract are approved. I recommend its publication. Approved: Brian Gerber

PAGE 6

vi ACKNOWLEDGEMENTS I would like to thank my committee members, Drs. Gerber, Martell, and Jose, for their advice and support throughout my thesis pr ocess. Each of them enhanced my final project and contributed to my experience a nd education more broadly. I also truly appreciate the help of Drs. Varda and Sowa, who went beyond their calls of duty by meeting with me on multiple occasions and he lping me in the areas of social network analysis and nonprofit evaluation, respectively. I would also like to thank the School of Public Affairs for providing financial suppor t for my thesis and for offering me the opportunity to create my own path and explore my areas of interest. There are many people in Uganda to whom I am deeply grateful. First, I would like to thank Moses for so kindly spending so many days guiding me around Nansana, setting up interviews for me, educating me, a nd being patient with me. I truly enjoyed the time we spent together. Next, I woul d like to thank Grace for his friendship, for introducing me to Moses, and for his suppor t throughout my stay in Uganda. I would also like to thank all of the people associat ed with Come LetÂ’s Dance for their warmth, hospitality, and friendship. I am grateful to have had such a lovely home in Uganda. Finally, I would like to thank all of the w onderful employees and volunteers of Nansana civil society who so graciously took the time to speak with me and educate me regarding their community and their organizations. I am amazed by the abundance of care and concern, for both my wellbeing and the succe ss of my research, that surrounded me throughout my stay in Uganda. I have much to learn from the people there.

PAGE 7

vii TABLE OF CONTENTS CHAPTER I. BACKGROUND AND OVERVIEWÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…...1 II. EFFECTIVENESS AND NETWORKSÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…..11 III. METHODSÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…..26 IV. OVERVIEW AND FINDINGS RELATED TO SAMPLEÂ…Â…Â…Â….Â…Â…Â…37 V. NETWORK MAPPING AND RESULTSÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…..50 VI. INSTITUTIONAL SUSTAINABILITY AND PARTNERSHIPSÂ…Â…Â…Â….63 VII. INTERNATIONAL TIES RESULTSÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â….90 VIII. DISCUSSION, LIMITATIONS, CONCLUSIONÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…102 REFERENCESÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…120 APPENDIX A. Survey InstrumentÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â….Â…Â…Â….Â…Â…124 B. Glossary of Social Network Analysis TermsÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…..Â…...134

PAGE 8

viii LIST OF TABLES Table III.1. Measures of Institu tional Sustai nabilityÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…...Â…..29 VI.1. Comparisons based on Network membershipÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…64 VI.2. Comparisons based on numbers of out degreesÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â….68 VI.3. Comparison of years of operati on based on numbers of out degreesÂ…Â…Â…Â…Â…69 VI.4. Comparison of programs base d on numbers of out degreesÂ…Â…Â…Â…Â…Â…Â…Â…..69 VI.5. Comparison of beneficiaries based on numbers of out degreesÂ…Â…Â…Â…Â….Â…Â…70 VI.6. Comparison of employees ba sed on numbers of out degreesÂ…Â…Â…Â…..Â…Â…Â…..70 VI.7. Comparison of volunteers base d on numbers of out degreesÂ…Â…Â…Â…Â…..Â….Â….71 VI.8. Comparison of Nansana facilities based on numbers of out degreesÂ…Â…Â….....Â…71 VI.9. Comparison of ownership rates based on numbers of out degreesÂ…Â…Â…Â…Â…Â…72 VI.10. Comparison of offices outside Nans ana based on numbers of out degrees...Â…Â…72 VI.11. Comparison of technology systems based on numbers of out degreesÂ…Â…Â…Â…..73 VI.12. Comparison of computers ba sed on numbers of out degreesÂ…Â…Â…Â…Â…Â…Â…Â…73 VI.13. Comparison of actual budget based on numbers of out degreesÂ…Â…Â…Â…Â…Â…Â…74 VI.14. Comparison of budget ratio based on numbers of out degreesÂ…Â…Â…Â…Â…Â…Â…..74 VI.15. Comparison of record keeping based on numbers of out degreesÂ…Â…Â…Â…Â…Â….75 VI.16. Comparison of evaluations ba sed on numbers of out degreesÂ…Â…Â…Â…Â….Â…Â…..75 VI.17. Comparison of human resource systems based on numbers of out degreesÂ…Â…..76 VI.18. Comparisons based on numbers of in degreesÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…...Â…Â…78 VI.19. Comparison of years of operati on based on numbers of in degreesÂ…Â…Â…Â…Â…..79 VI.20. Comparison of programs base d on numbers of in degreesÂ…Â…Â…Â…Â…Â…Â…Â…Â…80

PAGE 9

ix VI.21. Comparison of beneficiaries based on numbers of in degreesÂ…Â…Â…Â…Â…...Â…Â…81 VI.22. Comparison of employees based on numbers of in degreesÂ…Â…Â…Â…Â…...Â…..Â….81 VI.23. Comparison of volunteers base d on numbers of in degreesÂ…Â…Â…Â…Â…Â….....Â….82 VI.24. Comparison of Nansana facilities based on numbers of in degreesÂ…Â…Â…Â…...Â…82 VI.25. Comparison of ownership rates based on numbers of in degreesÂ…Â…Â…Â…Â…Â…..83 VI.26. Comparison of offices outside Nans ana based on numbers of in degreesÂ…Â…Â….83 VI.27. Comparison of technology systems based on numbers of in degreesÂ…Â…Â…Â…Â…84 VI.28. Comparison of computers ba sed on numbers of out degreesÂ…Â…Â…Â…Â…Â…Â…Â…85 VI.29. Comparison of actual budget based on numbers of in degreesÂ…Â…Â…Â…Â….Â…Â….85 VI.30. Comparison of budget ratio based on numbers of in degreesÂ…Â…Â…Â…Â…Â…Â…Â…86 VI.31. Comparison of record keeping based on numbers of in degreesÂ…Â…Â…Â…...Â…Â…87 VI.32. Comparison of evaluations based on numbers of in degreesÂ…Â…Â…Â…Â…Â…Â…Â….87 VI.33. Comparison of human resource systems based on numbers of out degreesÂ…..Â…88 VII.1. Comparisons based on strength of international tiesÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â….91 VII.2. Comparison of years of operation based on international tiesÂ…Â…Â…Â…Â…Â…...Â…92 VII.3. Comparison of programs based on international tiesÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â….93 VII.4. Comparison of beneficiaries based on international tiesÂ…Â…Â…Â…Â…Â…Â…Â…...Â…93 VII.5. Comparison of employees based on international tiesÂ…Â…Â…Â…Â…Â…Â…Â…Â…...Â…94 VII.6. Comparison of volunteers based on international tiesÂ…Â…Â…Â…Â…Â…Â…Â…Â…...Â…94 VII.7. Comparison of Nansana facilities based on international tiesÂ…Â…Â…Â…Â…Â…...Â…95 VII.8. Comparison of ownership rates based on international tiesÂ…Â…Â…Â…Â…Â…Â…...Â…96 VII.9. Comparison of offices outside Nansana based on international tiesÂ…Â…Â…Â…..Â…96 VII.10. Comparison of technology systems based on international tiesÂ…Â…Â…Â…Â…Â…Â…97

PAGE 10

x VII.11. Comparison of computers based on international tiesÂ…Â…Â…Â…Â…Â…Â…Â…Â…...Â…97 VII.12. Comparison of actual budget based on international tiesÂ…Â…Â…Â…Â…Â…Â…Â…..Â…98 VII.13. Comparison of budget ratio based on international tiesÂ…Â…Â…Â…Â…Â…Â…Â…Â….Â…99 VII.14. Comparison of record keeping based on international tiesÂ…Â…Â…Â…Â…Â…Â…Â…....99 VII.15. Comparison of evaluations based on international tiesÂ…Â…Â…Â…Â…Â…Â…..Â…Â…..100 VII.16. Comparison of human resource systems based on international tiesÂ…...Â…Â…Â…101

PAGE 11

xi LIST OF FIGURES Figure IV.1. Percentage of organizations with particular programsÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…...40 IV.2. Percentage of organizations serv ing each category of beneficiariesÂ…Â…Â…Â…Â….43 V.1. All network tiesÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…..51 V.2. Bonded tiesÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â….53 V.3. The addition of partnerships over three yearsÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…54 V.4. Formal partnershipsÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…...55 V.5. Partnerships involving joint programsÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…...57 V.6. Partnerships by levels of resource sharingÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â….58 V.7. Partnerships by frequency of contactÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â….59 V.8. Network map including fina ncial resources attributeÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…60

PAGE 12

1 CHAPTER I BACKGROUND AND OVERVIEW Development efforts in the worldÂ’s South have long been the victim of trial and error and the subject of much debate. G overnments and organizations across sectors continually seek better and more effectiv e ways to aid development efforts and are simultaneously subjected to much criticism as their motives are questioned and the negative side effects of their work are brought to light. Histori cally, development was thought to occur through capit al transfers and strengtheni ng central governments (Korten, 1987). At least partially in response to gove rnment corruption issues in the developing world, however, efforts have shifted to a focus on democratization and building local capacity (Korten). Similarly, the role that nongovernmental organizations ( NGOs) play in the development process has shifted over the course of the last few decades. While all sectors of society (government civil society, and the privat e sector) play a role in international development, NGOs are increasingly at the helm of such efforts (Barr, Fafchamps, & Owens, 2005).1 Furthermore, originally, the work of NGOs was largely limited to relief, disaster assistance, and food distribution (Gellert, 1996 ). More recently, however, NGOs have shifted their focus to alleviating poverty and improving the quality of life in developing countries (Gellert). The evolution of the role of NGOs in development efforts is thought to include three generations (Korten, 1987; Atack, 1999). First generation NGOs focus on relief and welfare and largely provide services to meet the imme diate needs of those in the 1 NGOs are typically considered part of civil society, though some argue that they actually occupy a fourth position in society, serving as a mediator between the other three sectors (Fowler, 2002).

PAGE 13

2 developing world. Second generation NGOs fo cus on local self-relia nce, meaning that they seek to develop the capaci ty of local people so that the benefits of an NGOÂ’s work are sustainable even after the NGO leaves the region. In the third ge neration, the focus is on sustainable systems development, or bu ilding the larger institutional and policy context within which deve lopment takes place. The increasingly central role of NGOs in development efforts is largely a reaction both to the successes of various NGO initiatives globally and to perceived shortcomings of governmentsÂ’ abilities to successfully manage development (Barr et al., 2005). Furthermore, when governments and the United Nations do engage in development efforts, they now often use NGOs for their service delivery (Atac k, 1999). Gellert (1996) identified six attributes of NGOs that make them particularly well-suited to achieve desirable outcomes in the developing world. First, NGOs have the ability to target and reach areas in greatest need of assistan ce. Second, NGOs tend to promote local involvement by working closely with local gov ernments and enlisting local help. Third, NGOs typically have relatively low costs of operations. Fourth, NGOs tend to be adaptive and innovative. Fifth, NGOs are ge nerally independent and free of political involvement, and, finally, NGOs often seek sustainable outcomes. While there are many benefits to NGO invol vement in interna tional development, a number of challenges exist as well. Current ly, the biggest challenge that NGOs face is decreasing contributions from th e international community (Ge llert, 1996). This lack of funds forces NGOs to become more creative in their service delivery, and, perhaps on the positive side, limited funds are pushing NGOs to move toward more sustainable models (Fowler, 2002). Furthermore, and of particular relevance to this thesis, with limited

PAGE 14

3 funds, partnerships become an essential com ponent of an organizat ionÂ’s sustainability because of the ability of partne rships to help an organization build capacity (Claiborne et al., 2009). Indeed, since the early 1990s, there has been an increase in partnerships and collaborations among NGOs (Gellert). A second major challenge of NGOs is to maintain legitimacy as a facilitator of international development. For example, ma ny people have criticized NGOs for actually hindering development efforts by undermining markets by providing goods and services at less than market costs (Korten, 1987). Ot hers question whether We stern aid is actually used in a way that is benefi cial to the developing world, pa rticularly in light of the frequent disconnect between funders and bene ficiaries of development efforts (Easterly, 2006). Indeed, as NGOs take on increasingly cen tral roles in international development, whether and how they should be involved in development processes has been the subject of much debate. Atack (1999) explores this issue by looking at NGO legitimacy, or the normative basis for NGO involvement. He finds that four criteria are nece ssary to justify NGO involvement in the developing world: representa tiveness, distinctive values, effectiveness, and empowerment. Representativeness and distinctive values are formal-procedural justifications. Representativeness refers to an organizationÂ’s transparency and accountability, particularly with respect to beneficiaries, rather than donors, and distinctive values refers to an NGOÂ’s adherenc e to unified values, particular those with which partners and constituents agree. The second two criteria, effectiveness and empowerment, are substantive-purposive crit eria. Effectiveness refers to an NGOÂ’s

PAGE 15

4 progress toward development goals, and empowerment is the process by which individuals are able to take control over their lives. This thesis explores a particular aspect of legitimacy by looking at institutional sustainability as a component of effectiv eness. “Organizational or institutional sustainability places importance on ‘buildi ng sustainable organizations to achieve sustainable development benefit’” (Cannon, 2002, p. 363). As described in greater detail in Chapter II, institutional sustainability refers to an NGO’s ability to endure and to continue to provide services. Theoretically, effectiveness and institutional sustainability are related because more sustainable organizations will be better able to produce more sustainable results, which is particularly im portant as organizations work toward local empowerment and sustainable systems deve lopment. Of course, there are also counterpoints to this argument, which are di scussed in greater detail in Chapter II. Purpose and Hypothesis The purpose of this thesis is to examine the relationship between interorganizational partnershi ps and institutional sustainabi lity within a network of NGOs operating in Uganda, and the hypothesis of this study is that organizations with greater interconnectedness with other organizations in the subject network will score higher on an array of institutional sustainability measures. There are two primary components to this thes is. First, this thesis offers an in depth look at a network of NG Os operating in the developi ng world. Through interviews and network mapping, I describe the institutional sustai nability, along a number of dimensions, of the organizations involved in this study. I also expl ore the organizations’ partnerships and relationships with one anot her and how the formal network operates and

PAGE 16

5 has evolved. Understanding this network of organizations and its benefits and struggles may offer insight to the international commun ity regarding service delivery at the microlevel. Looking deeply at a network of orga nizations that develope d through local efforts may help development efforts effectively move from relief and welfare to local empowerment and sustainable systems development. In the second part of this thesis, I test the hypothesis through a series of quantitative tests based on survey results. Specifically, I look at differences in institutional sustainability based on formal network membership and differences based on organizationsÂ’ numbers of partners within th e subject network. The purpose here is not to rate these NGOsÂ’ institutional sustainabi lity, but rather, to look at the specific characteristics of sustainability that correlate with increased numbers of network partners (or network centrality). These findings may have important implications for networksÂ’ and partnershipsÂ’ abilities to build capacity through their relationships, particularly in light of decreasing in ternational aid. Following a brief descripti on of the subject network and its geographic context, Chapter II presents a review of the lite rature regarding effectiveness generally, institutional sustainability part icularly, and partnerships and networks. Chapter III sets forth the methods used in this study and includ es a discussion of th e particular measures of institutional sustainability used in this study. Chapter IV consists of a detailed description of the organiza tions in the study based on th eir interview responses, and Chapter V uses network mapping to furthe r understand the subj ect network and how organizations within it interact. Chapter VI tests the hypothesis of this study through statistical tests related to network membership and partnerships, and Chapter VII presents

PAGE 17

6 an additional set of findings regarding the relationship between international ties and institutional sustainability. Finally, Chapter VIII contains a disc ussion of the findings and limitations of this study. Uganda and Nansana Background The network of organizations chosen for this study operates in the East African nation of Uganda. As of July 2012, Uga ndaÂ’s estimated population was almost 34 million (Central Intelligence Agency, 2013). Despite its relative stability since 1986 when Yoweri Museveni assumed the presid ency, Uganda continues to experience difficulties related to violence, poverty, and wi despread HIV/AIDS. The life expectancy in Uganda is 53 years, and it has the fourth highes t population growth rate in the world at 3.3% (Central Intelligence Agency). It is also eighth in the world for the rate of people living with HIV/AIDS (Central Intelligen ce Agency). UgandaÂ’s per capita gross domestic product is $1400, which is 202nd out of 228 ranked countries (Central Intelligence Agency). As with general recent patterns of NGO proliferation, the growth of the NGO sector in Uganda began in the 1970s and 1980s following the collapse of the government (Barr et al., 2005). In a desc riptive study regarding a repr esentative sample of NGOs throughout Uganda, Barr et al. found that th e three most common NGO activities in Uganda are education and training, support to farming, and HIV/AIDS awareness and prevention. This study examines a network of NGOs operating in Nansana, Uganda. Nansana is a suburban community in Wakiso District approximately 10 kilometers northwest of UgandaÂ’s capital city, Kampala. Nansan aÂ’s population was estimated at 89,900 in 2011,

PAGE 18

7 up from 62,044 in 2002 (Uganda Bureau of Statistics, 2010a). Along with recent population and development growth in Nansan a (Nantakiika, 2011), there has been an increase in the number of NGOs calling Nansan a home. Nansana is currently home to at least 40 NGOs, 35 of which are connected th rough the Nansana Ci vil Society Network (M. Senyonjo, personal communication, January 18, 2013). In this study, I examine this network of NGOs to assess the relationship between the levels of connectivity among member organizations and the organizationsÂ’ institutional sustainability. The Nansana Civil Society Network Ugandan NGOs are well networked (Barr et al., 2005), and, in Nansana, there is a formal, membership-based network of orga nizations called the Nansana Civil Society Network (the NCSN). The NCSN began fo rming in 2008 when it was conceptualized by three Ugandans working in the civil society se ctor in Nansana and re ceived the support of the Town Council.2 The NCSN officially began oper ating in 2010 as a community based organization (CBO) in Wakiso District. CBO s in Uganda are generally single purpose organizations, operating at subc ounty levels to promote th e well-being of a community (Council on Foundations, 2012). The NCSN is a quasi-governmental or ganization in that it is technically part of ci vil society, but it has an offi ce at the Nansana Town Council (provided by the government) and is overseen by the Community Development Officer for the Town Council. The NCSN has 35 member organizations To become members of the NCSN, organizations submit an application form (describing items such as their programs, 2 Unless otherwise indicated, information thr oughout this section was obtained primarily through personal communication with M. Se nyonjo (the NCSN Coordinator) on January 18, 2013 and supplemented by additional interv iews with members of Nansana civil society throughout December 2012 and January 2013.

PAGE 19

8 management, and operations) and letters of re ference from other organizations. They also submit to a site review by an NCSN representative. The only basic criterion for membership is that organizations be located in Nansana (though exceptions may be made for organizations located ju st outside the boundaries of Nansana). There are no membership limitations based on organization size, services, or other organizational characteristics. Organizations are generally admitted to the NCSN as long as they are legitimate and well-intentioned. The NCSN holds monthly members mee tings, which give organizations the opportunity to share, offer advice to one anot her, and support each other. The NCSN also offers capacity building support, such as th rough trainings and by helping organizations identify potential funding sources and apply fo r funds. Furthermore, a major emphasis of the NCSN is on its referral system, a progr am through which organizations become aware of each other’s programs and resources, enter into referral agreements, and refer cases among themselves in order to best serve benefi ciaries. The organiza tions also engage in other periodic projects together such as monthly sanitation (c leanup) efforts in Nansana. The NCSN is entirely volunteer run. As described below, however, a “volunteer” in Nansana is akin to an employee, except that instead of salaries, volunteers only receive facilitation fees to cover their expenses. The NCSN has one permanent volunteer, who serves as its coordinator, and seven rotating board members. With respect to the board members, organizations are assigned to partic ular posts on the board of directors for oneyear terms. The selected organizations then appoint one of their employees (or volunteers) to fill the post. The board of directors also meets monthly.

PAGE 20

9 The NCSN is primarily funded through me mbership dues. Organizations pay an annual fee to be part of the NCSN, and these membership dues comprise approximately 60% of the NCSNÂ’s budget. The remai nder of the budget comes from additional voluntary contributions from member organizat ions. The NCSN spends the majority of its funds on its programs and coordinating activ ities, with a small portion of its fees going toward supporting the Coordinator. The NCSN Coordinator and representatives of member organizations generally report positive results stem ming from the NCSN, including improved service delivery and enhanced capacity, particularly with respect to information and funding. The Coordinator is also currently developing and promoting a volunteer and internship program to encourage students and commun ity members to volunteer for the various Nansana organizations to further enhance th e organizationsÂ’ capacities and to offer valuable experience to local people. While the NCSN appears to be largely successful based on member reports, it also faces challenges, such as distrust among members stemming from perceived competition for resources and duplication of services. Al so, despite the large number of organizations, many groups of people in Nansana do not receiv e support. The majority of organizations in the NCSN target children as beneficiaries because it is relativel y easy to get funding for childrenÂ’s programs and because there are so many children in need in Nansana (at least in part reflecting the young population). Wo men are also frequent beneficiaries, but, according to the Coordinator, few programs ta rget other populations in need, including men, widows and widowers, and people w ith disabilities. Furthermore, few

PAGE 21

10 organizations focus on environmental issues because the cost of doing so is high and because the environment is genera lly not considered a priority. Following a discussion of the literature in forming this study and the methods used in this study, I move to a more detailed desc ription of the organizations working in the NCSN and in Nansana more broadly. I then look at whether the partnerships among organizations in Nansana correlate with a variety of measures of the organizationsÂ’ institutional sustainability.

PAGE 22

11 CHAPTER II EFFECTIVENESS AND NETWORKS As described above, among the necessary ingredients for NGO legitimacy is effectiveness (Atack, 1995). Effectivene ss, however, typically refers to an NGO’s progress toward development goals (Atack), and increasi ngly, in the international development realm, development goals focus on achieving sustainable results. In other words, NGOs are indeed moving from first ge neration relief and welfare to second and third generation work emphasizing the need to develop local capacity and sustainable institutional and policy contexts. This thesis looks at institutional sustaina bility as one dimension of effectiveness, because if the organization itself is sustainable, it is in a better position to have lasting effects on the community within which it wo rks. Indeed, many organizations validly focus on their institutional sustainability, because they want to be in a position to continue to serve clients over an ex tended period of time (Edwards, 2002). “‘Organizational development’ implies a conscious process of acquiring and strengthening the characteristics required by an [NGO] to position itself for maximum impact and effectiveness” (Edwards & Fo wler, 2002, p. 3). Practically speaking, the more stable an organization is, the more like ly its employees will be able to turn their attention outward and focus on the needs of beneficiaries. Also, an organization with more characteristics of institutional sustaina bility (like adequate resources and reliable record keeping) will be better poised to effectively serve a community, and as NGOs transition from first generation to second and third generation work, this becomes increasingly important.

PAGE 23

12 At the same time, however, there is ce rtainly an argument to be made that organizations may have a self-interest in su staining themselves at the expense of the communities they serve. In other words, perhaps in some situations, institutional sustainability undermines outcome effectiven ess because organizations just want to maintain a role for themselves. For th ese reasons, future studies exploring the relationship between sustainable institutions a nd sustainable results may be warranted. For purposes of the present study, howeve r, it is reasonable to assume that institutional sustainability is a component of effectiveness and that it may help an organization achieve desirable results for its beneficiaries and move toward second and third generation development work. I be gin with a general discussion regarding effectiveness. Measuring Effectiv eness Generally Effectiveness in the nonprofit sector is an extremely broad and difficult to capture concept, and there is little agreement regardi ng how to meaningfully measure it. This is particularly important in the internati onal NGO realm where effectiveness is often assumed rather than tested (Atack, 1999). Wh ile in the for profit s ector, effectiveness is easily measured as it is synonymous with financial success, in the nonprofit sector, missions and goals are typically more abstract and difficult to measure (Collins, 2005). Like for profit institutions, however, to be successful and attrac t resources, nonprofits must be able to demonstrate valu e to investors (Ralser, 2007). Generally speaking, effectiveness can be measured at three levels: programmatic, organizational, and network (Herman & Renz 2008). Consistent with the discussion above regarding outcome sustainability, the majority of studies evaluating nonprofit, and

PAGE 24

13 particularly NGO, effectiveness do so at the pr ogram level. For example, in assessing the effectiveness of an NGO health care program in Bangladesh, Mercer, Khan, Daulatuzzaman, and Reid (2004) looked at programmatic outcomes, such as mortality rates and the percentages of the populations be ing served who received certain treatments, vaccinations, and types of care. Much work is also being done regarding measuring effectiveness at the network level (Provan & Milward, 2001). The focus of this study is effectiveness at the organizat ional level, which is an of ten side-lined yet essential component of effectivene ss, particularly as development efforts evolve. Historically, measures of organizationa l effectiveness were inadequate because they actually relied on programmatic measures (Herman & Renz, 2008). Thus, recognizing that various models for measuring effectiveness often only target one aspect of effectiveness (usually pr ogrammatic), recent nonprofit management literature supports a multi-dimensional approach to measuring organizational performance (Herman & Renz). For example, the Development A ssistance Committee (1991) (a forum within which member governments address economic social, and environmental issues associated with globalization) supports a fr amework that measures effectiveness at a variety of levels and across issues. Simila rly, recent measures of network success also depend on multiple dimensions. For example, according to Kania and Kramer (2011), each of five conditions (common agenda, shared measurement systems, mutually reinforcing activitie s, continuous communication, and backbone support organizations) must be present for a network of orga nizations to achieve collective success. Sowa, Selden, and Sandfort (2004) deve loped a comprehensive framework for measuring organizational effectiveness called the multidimensional, integrated model of

PAGE 25

14 nonprofit organizational effec tiveness (MIMNOE). Their model assumes that the primary dimensions of effectiveness are ma nagement and program effectiveness, with “management” referring to characteristics of the organization and its management, and “program” referring to the organization’s particular services. The two primary dimensions (management and program) can then be subdivided into capacity and outcomes, with capacity referring to the pro cesses and structures of an organization and outcomes referring to the results of programming or management efforts. Thus, there are four categories within which to measur e effectiveness: management capacity, management outcomes, program capacity, and program outcomes. Furthermore, to effectively measure each of these dimensions, So wa et al. assert that both objective and perceptual measures should be used. Sowa et al. (2004) and Selden and So wa (2004) (applying the MIMNOE in a study of the effectiveness of 22 human serv ice organizations provi ding early care and education services) describe that objectiv e measures of management capacity might include items like whether the organization ha s a formal mission statement, a strategic plan, human resources structures, employ ee training and performance management systems, an independent financial audit, and information technology systems. Related perceptual measures might include whether employees think the above systems serve a functional role within the organization. Objective meas ures of management outcomes, on the other hand, include items like revenue stability, the organization’s ability to maintain a financial surplus, and employee tu rnover. Perceptual measures might come from managers’ self-reports regarding their organization’s financia l well-being or job

PAGE 26

15 satisfaction reports among employees. These factors related to management capacity and management outcome are particularly re levant to institutional sustainability. On the program side, according to Sowa et al. (2004) and Selden and Sowa (2004), objective capacity measures include items like how the organization transforms inputs into outputs, the level of material resources assigned to programs, problems that have occurred in implementing programs, staff expe rience levels, and divers ity of services. A perceptual capacity indicator might be how staff members assess the efficacy of their programs or clients’ perceptions of staff comp etency or available resources. An objective measure of program outcome might be the degr ee to which the program achieves its goals, and a perceptual measure might look at client satisfaction. Another model for measuring effectiveness at multiple levels comes from Sawhill and Williamson (2001), who describe the Na ture Conservancy’s experiences with attempting to measure success. After a fe w rounds of trial and error, the Nature Conservancy developed a model covering three broad areas: impact, activity, and capacity. Impact asks whether an agency is making progress toward fulfilling its mission and meeting its goals. Activity refers to whether an organizat ion’s activities are achieving its programmatic objectives and implementing its strategies, and capacity refers to whether the organization has the resources necessary to achieve its goals. Most relevant to this st udy are capacity measures, which “gauge the degree to which the organization mobilized the resources necessary to fulfill the mission” (Sawhill & Williamson, 2001, p. 375). To keep its measure simple, the Nature Conservancy used two to four measures for each of its three categories (impact, activity, and capacity). Its capacity measures included fundraising perf ormance, membership growth, and market

PAGE 27

16 share. The authors also talked to othe r nonprofits about their efforts to measure effectiveness, and, similarly, other organiza tions’ capacity measures typically include items like financial activit y, fundraising, and membership (Sawhill & Williamson). To develop the institutional sustainability measure used in this study, I primarily drew on the management measures describe d by Sowa et al. (2004) and the capacity measures described by Sawhill and Williamson (2001). With these concepts and general frameworks in mind, I now turn to a di scussion of institutional sustainability. Institutional Sustainability Sustainability is a component of eff ectiveness that is increasingly gaining attention, particularly in the NGO realm. Th is study looks specifically at institutional sustainability rather than outcome sustaina bility, and, unlike the majority of studies assessing programmatic outcomes, this study looks at corr elations between interorganizational relationships and inst itutional sustainability. As mentioned, institutional sustainability may be particularly important as the emphasis in development strategy moves from relief and welfare toward sustainable systems development (Atack, 1999). Much has been written about outcome sust ainability, as it is an increasingly important measure of nonprofit effectivene ss (Herman & Renz, 2008). According to the Development Assistance Committ ee (1991), “[t]he question of whether achievements are sustainable in the longer run is of critical importance” (p. 10). Lyons, Smuts, and Stephens (2001) define sust ainable development “as the ability acquired and held by communities over time, to initiate and contro l development, thus enabling communities to participate more effectiv ely in their own destiny” ( p. 1237), and they look at the

PAGE 28

17 relationship between participat ion, empowerment, and sustainabi lity. Also, Ralser (2007) notes that nonprofits typically fa ll into three categories: trad itional, sustainable, and cash generating. While the traditiona l model tends to focus on char ity, in sustainable and cash generating models aid is viewed as an invest ment that will be used to help communities become more sustainable and self-reliant. As described above, institutiona l sustainability is theore tically related to outcome sustainability, but less has been written about institutional sustainabi lity directly. One discussion comes from Lewis (2003), w ho looks at the relationship between organizational culture and institutional sustai nability in the sericulture (silk farming) industry in Bangladesh. Dr awing on Cannon (2002), Lewis expl ains that there are three levels of institutional sustainability in the proj ect setting: 1) Financia l sustainability is a project’s ability to generate resources from a variety of sources; 2) organizational sustainability refers to an organizational ar rangement’s ability to continue to offer a framework for delivering benefits or services ; and 3) benefit sustai nability refers to the continuing availability of benefits or se rvices beyond the life of the project. In another study examining institutiona l sustainability in the NGO context, Okorley and Nkrumah (2012) look at NGOs in Ghana and factors that might affect a local NGO’s sustainability. They look at factors that exist among sustainable organizations, which they define as “the ability or capacity of the NGOs to endure” (Okorley & Nkrumah, p. 331) and operationalize as organizations that have endured for at least five years. They examine possible in ternal and external factors associated with organizational sustainabilit y, including leadership, funding, program development,

PAGE 29

18 program management, and material resour ces, and they concl ude that supportive leadership and adequate funding are most co rrelated with organiza tional sustainability. Here, I combine the above definitions and de fine institutional sustainability as the ability of an NGO to endure and to conti nue to provide services. I include both components because if institutional sustainabil ity only refers to an organizationÂ’s ability to endure, we run the risk of having the organization endure at the expense of the beneficiaries, and, if institutional sustainabi lity only refers to a continuing ability to provide services, the defin ition becomes too similar to a mere focus on programmatic outcomes. Particularly in third generation de velopment work, it is important to gauge the sustainability of the institutional and po licy systems surrounding development efforts. Thus, the definition here is broader th an that of Lewis (2003) because Lewis discusses organizational sustaina bility in the context of a give n project, and because, here, I consider financial sustainabili ty to be a component of inst itutional sustainability rather than a separate type of sustainability. Furt hermore, the operational concept used here is also broader than that of Okorley and Nk rumah (2012), who submit that an organization is sustainable if it has existe d for five years. Here, I include elements of longevity, service provision, capacity (including human, monetary, and capital resources), stability of structures (both systems and physical ), and planning for the future, in my operationalized conception of institutional sustainability. Interorganizational Relationships Motivated to enhance their effectiv eness, nonprofits increasingly rely on partnerships and networks for their deliver y of services (Sowa, 2009; Varda, Shoup, & Miller, 2012). Networks have been shown to produce better outco mes for individuals,

PAGE 30

19 organizations, and communities, and studies have shown positive associations between networks and desirable outcomes like enhan ced service delivery, organizational growth, innovation and entrepreneurship, and increase d civic engagement (Paarlberg & Varda, 2009). Furthermore, recent literature focuse s on collective impact, which is based on the idea that “large-scale social change comes from better cross-sector coordination rather than from the isolated intervention of indi vidual organizations” (Kania & Kramer, 2011, p. 38). This study examines whether increa sed interorganizational connectivity is associated with higher levels of institutional sustainability. In the context of international de velopment, partnerships among NGOs and between NGOs and other sectors are consider ed important to improved service delivery (Claiborne et al., 2009). As mentioned, this is particularly so as re duced funding persists. Strong links, both among NGOs and between NGOs and other institutions “build organizational strength and social capital, and pool experien ce and other resources” (p. 287). Indeed, NGOs are increasingly using co llaboration to build capacity and enhance their service delivery (Claiborne et al.). Types of Relationships Studies examining interorganizationa l relationships typically look at the relationship between two organizations and th e broader network of relationships. These relationships can be categorized in a variety of ways. Fo r example, networks can be transmission networks, social support netw orks, and organizational networks (Varda, Chandra, Stern, & Lurie, 2008). Relationships ca n also be classified based on the level at which the arrangement occurs, including policy-centered integration, organization-

PAGE 31

20 centered integration, program-centered inte gration, and client-centered integration (Selden, Sowa, & Sandfort, 2006; Sowa, 2008). A common way to think about interorg anizational relationships is along a continuum. For example, Kagan (1991) desc ribes a continuum of relationships that includes cooperation, coordina tion, and collaboration, with cooperation being the least integrated relationship form and collaboration being the most integrated form. “As one progresses from cooperation through coordina tion to collaboration, interorganizational relationships become more sophisticated, co mplex, and effective for problem solving” (Kagan, p. 2). Scholars typically agree that cooperation represents the least formal and most prevalent type of relationship between organizations. Cooperations are grounded in personal relationships and exist without cl early defined struct ures. Cooperating organizations work only informally together may have different goals, and generally do not pool resources. According to Kagan (1991), scholars agree less on definitions of coordination and collaboration. Many do not see the difference between these two types of relationships, as they share characteristics like mutualit y of goals, resource sharing or pooling, and dynamic, changing interactions. Those who se parate the concepts see coordination as a prerequisite for collaboration and as less complex and sophis ticated than collaboration. In coordination, agencies generally remain independent, but in collaboration, previously separate organizations may come together in a new structure. Other important distinctions include greate r sharing of resources, more intense joint planning, and increased sharing of pow er and authority.

PAGE 32

21 More recently, Selden et al. (2006) describe d that cooperation consists of informal and personal relationships between manageme nt and staff of different organizations, while coordinating organizations try to ca librate their actions, and collaborating organizations share existing resources, aut hority, and rewards. Others have also expanded the continuum to include service inte gration where organizations work together to provide a new package of services to shared clients (Selden et al.). Along this continuum, Sowa (2008) further delineates the types of relationships. She looked specifically at collaborations and found three di fferent types of collaborations. In collaborative contracts, agencies collabora te to produce services, but the relationship is shallow in that it is mostly limited to sharing financial resources. In capacity-building collaborations, agencies collaborate to pr ovide services, but, beyond sharing financial resources, they also share human resources, such as staff and professional development resources. In community-building collaboration s, the same types of resources are shared, but the benefits go beyond the tangible organizational-centered benefits found in capacity-building collaborations. In these re lationships, the organizations involved also gain an enhanced understanding of the commun ity they are serving, and they extend their role in that community. Others describe a simpler continuum th at varies depending on the formality of relationships. At one end of the conti nuum are formal relationships, which are characterized by joint programs, parent-subsidiary relationships, joint venture, or merger. At the other end are informal collaborati ons, which are characterized by information sharing, client referrals, sh aring office spaces, and management service organizations (Guo & Acar, 2005).

PAGE 33

22 To understand the organizations in this study and their relationships, here I adopt the language of Kagan (1991) and others, a nd describe relationships as cooperation, coordination, or collaboration. These cat egories provide a useful framework for understanding the types of relationships that exist in Nansana, particularly as they describe varying levels of integration and or ganizations’ abilities to impact and pervade communities. The literature th at informs how to measure these relationships is described below. How to Measure Relationships Because the intensity of relationships ma y have important impacts on a variety of outcomes, scholars have sought to develop mean s of measuring this intensity (Selden et al., 2006). Existing literature regarding inte rorganizational relationships and networks describes a variety of means to measure the connections among organizations and determine how to classify interorgani zational relationships based on the above frameworks. Connectivity is “the measur ed interactions between partners in a collaborative such as the amount and quality of interactions and how these relationships might change over time” (Varda et al., 2008, p. E1). To measure connectivity, studi es often ask subjects to describe their various partnerships and then also ask subjects about the relationships betw een the partners they identify. For example, McPherson, Sm ith-Lovin, and Brashears (2006) examine discussion networks of close confidants and include measures such as size of network (number of names mentioned), density and range of network (which refers to the intensity of relationships, operationalized as the mean intensity of tie strength among the partners mentioned, where relationships we re assigned a 0 if partners ar e strangers, a 1 if they are

PAGE 34

23 close, and 0.5 otherwise), durat ion of relationship, frequency of contact, and diversity of the interpersonal environment along vari ous sociodemographic dimensions. An often-mentioned method for meas uring and distinguishing between relationships is based on the amount of res ource sharing involved. For example, Kagan (1991) describes that informal relationships have little resource sharing, while formal relationships may involve the formation of a new integrated organization. In between these two extremes are relationships charac terized by shared monetary or material resources, shared programs, and shared administrative capacities. Also, Sowa (2008) looked at th ree types of sharing: resources, staff, and rewards. Resources are primarily conceptualized as financial, although nonfinancial resources, such as professional or practice knowledge may al so be shared. Staff can be shared either through the co-use of a specialized position, or it can involve housing staff in a shared office space. Finally, rewards are more difficult to define, but are considered the benefits that accrue to the organizations involved in the collaboration or to the larger community of which the collaboration is a part. Others have described additional components that may be useful in examining connectivity. For example, in measuring th e connectivity of memb ers of public health collaboratives, Varda et al. (2008) identified seven core di mensions of connectivity in public health collaboratives: ne twork membership, network in teraction, role of health department, frequency of interaction, stra tegic value, trust, and reciprocity. First, network membership refers to who is involved in the network and the diversity of membership; it is measured by identifying organizations by name, type, and other characteristics, such as size and mi ssion. Second, network interaction goes beyond

PAGE 35

24 merely counting numbers of members and l ooks at the quality of interactions among subgroups by examining network patterns. Third, the role of health departments refers to whether there is a coordinating entity within a network of organizations. Fourth, the frequency of interactions can be operationaliz ed by looking at the t ypes of relationships and levels of interactions among organizations, bearing in mind that more frequent interactions do not necessarily equate with greater connectivit y. Fifth, the strategic value of partners looks at the extent to which each partner 1) has power and influence within a community; 2) is actively i nvolved in the network by pa rticipating in meetings, accomplishing tasks, and so forth; and 3) ha s resources like money, food, physical space for meetings, staff time, data, educationa l materials, knowledge, information, and feedback. Sixth, trust depends on reliability, shared belief in mission, and opportunity for frank discussion. Finally, r eciprocity refers to mutual exchange of resources. Other studies have examined characteristic s of relationships to determine when such relationships have particularly positive effects. According to Paarlberg and Varda (2009), for a network to have positive synergis tic effects, first, the exchanges must be reciprocal, with resources fl owing in both directions between members. Second, while carrying capacity is typically determined by the human and financial capital in a community, networks require particular capa bilities necessary to facilitate these relationships. Third, the process of excha nge looks at whether network members are directly connected or whether they are tied th rough a string of other ac tors. It also looks at the strength of ties. Strong ties (excha nges that occur through frequent, extended, and reciprocal activities) are common among homo geneous organizations and often develop because of physical proximity or repeated in teractions, and organiza tions with strong ties

PAGE 36

25 tend to have similar missions and similar client s. Weak ties refer to ties between more diverse network members, or those with differe nt missions, clients, and resources streams. Weak ties contribute to positive synergistic e ffects. Finally, network exchanges should be coordinated, particularly via linkage to a governance structure, which can be either formal or informal. Each of the above characteristics and descriptive factors suggest methods for measuring connectivity between various partne rs. This study examines a particular network of organizations and measures the connectivity between me mber organizations by looking at the duration of relationships a nd frequency of contact, the types and amount of resources they share, whether they have joint programs, and the formality of the relationships.

PAGE 37

26 CHAPTER III METHODS To understand the relationships between NGOs in Nansana and their institutional sustainability, I interviewed a sample of 33 or ganizations in Nansana. The majority of the organizations interviewed (28) are memb ers of the NCSN, thus comprising a sample of 80% of the 35 organizations in the NCSN Four organizations interviewed are not members of the NCSN, and it is not clear how many organizations there are in Nansana that are not members of the NCSN. The fi nal organization interviewed is the NCSN itself, which is a registered CBO. The NC SN, however, is not included in the data analysis, but rather, is discusse d separately in Chapter I. Data Collection Data collection occurred on site in Nansana in December 2012 and January 2013. The NCSN Coordinator scheduled the majority of the meeti ngs with organizations on my behalf; however, I made clear to all interviewees that thei r participation was voluntary, that their answers were confid ential, and that their particip ation and responses would in no way impact their status or finances. Wh enever possible, I met with directors of organizations, but, in all cases, I ensured th at the person I spoke with was knowledgeable regarding the organization. Th e Coordinator participated in the interviews only if the interviewee specifically request ed that he do so. Interviews took place at organizationsÂ’ offices, except in the cases of t hose organizations without offices. The interviews were semi-structured. I personally conducted all interviews and used the same survey for each interview. The survey can be found in Appendix A.

PAGE 38

27 Interviewees were asked to sign a consent form and were provided an opportunity to provide personal data if they wished to receive a copy of the study outcomes. The survey itself consists of two primar y components: institu tional sustainability and interorganizational relationships. The institutional sustainability portion of the survey is subdivided into eight sections, suggested by the literature: services and programs, people, facilities, technology, fi nances, records, program evaluations, and human resource systems. The interorganizatio nal relationships section asks interviewees to identify all of their partners according to specified parameters, and then, for each organization named, interviewees were aske d questions regarding the length of the partnership, the extent of its formality, whet her it involves joint programs, the types of resource sharing involved, and the frequency of contact. In some situations, I attempted to fill informational gaps by reviewing organizationsÂ’ websites or by contacting other organizational representatives as recommended by interviewees. Data Analysis Data analysis was performed upon return to the United States. For purposes of analysis and discussion, orga nizations were randomly assigned numbers from 1 to 32. For the majority of the data analysis, I used Excel and SPSS to look at frequencies, means, medians, minimum and maximum values, standa rd deviations, and significance tests. As described below, a frequent method of analysis was to divide subjec t organizations into two to four groups along various dimensions and compare mean answers for a variety of institutional sustainability measures. Signi ficances of the differences in means were tested using one-way ANOVAs, independent samples t-tests, and chi-squares, as appropriate.

PAGE 39

28 Two methods were used to better under stand the organizations in Nansana and their relationships with one a nother. First, survey respons es were analyzed using basic descriptive measures as described below. Second, to further understand the relationships between and among organizations in Nansan a, network mapping was performed using UCINET and NetDraw. To test the hypothesis regarding th e relationship between networks and partnerships, on the one hand, and institutional sustainability, on the other hand, I performed three sets of comparisons. For the first set of comparisons, I divided the subject organizations based on whether they are members of the NCSN. For the second set of comparisons, I divided organizations based on the number of partners they mentioned, and, for the third set of compar isons, I divided organi zations based on the number of times they were mentione d as partners by other organizations. Organizations were also divided into three groups according to the strength of their international ties for fu rther analysis. While conducting the research, it seemed that organizations with strong international ties assumed a more central role among organizations in Nansana. These internat ional ties may largely explain network and institutional sustainability pa tterns, and, as such, it seemed important to test these relationships as well. According to Barr et al. (2005), NGOs in Ea st Africa (including Uganda) are generally funded through intern ational sources, and most NGO funding is funneled through a small number of Ugandan NGOs: “Many Ugandan NGOs indeed de facto or de jure operate as subcontractors for international do nors” (p. 676).

PAGE 40

29 Measures of Institutional Sustainability To test the hypothesis that organizations with more ties are likely to also score higher on various measures of institutional sust ainability, for each of the eight categories of institutional sustainability, one to three m easures of institutional sustainability were selected to be compared across categories of organizations. The measures selected to represent each category are described below and summarized in Table III.1. Table III.1. Measures of Institutional Sustainability. Institutional Sustainability Measures Services and Programs Years of operation Number of programs People Number of beneficiaries Number of employees Number of volunteers Facilities Number of facilities Ownership rate Number of offices outside Nansana Technology Number of technology items Number of computers Finances Actual budget Actual: expected budget ratio Records Points out of 5 Program Evaluations Rate conducting Human Resource Systems Points out of 5 Services and Programs Two measures of institut ional sustainability were used in the services and programs category: 1) the number of years the organization has been in existence and 2) the number of programs run by each organization. With respect to the length of time an organization has been in existence, while it may seem that an organizationÂ’s years of operation may itself be a proxy for institutional sustainability, and it has been used as such (Okorley & Nkrumah, 2012), such a measure likely only captures a narrow view of

PAGE 41

30 institutional sustainability. Specifically, an organization may technically operate for many years, but without any resources or without actually servi ng any clients. It should also be noted that some organi zations indicated both the time that they began internally and the time that the orga nization either began serving clients or registered as an NGO or CBO. For consis tencyÂ’s sake, if an organization provided multiple dates, I used the earliest date pr ovided. Furthermore, if an organization indicated the year in which it began, rather th an the length of time it had been operational, I subtracted the year from 2012. The second measure of institutional sustai nability in the area of services and programs is the number of programs each organization conducts. While many of the organizations claim to have a large number of programs to benefit their clients, it should be noted that more programs does not make an organization more effective, and, in fact, the converse may be true. One of the strongest lessons to emerge from the research is that success is more likely when organizations identify a clear long-term goal at the outset and stick to it over time. Conversely, agencies which change their goals too often or try to follow too many goals at the same time, often lose their way. (Edwards, 2002, p. 284) Thus, it seems that organizations with fewer programs are likely more sustainable than those attempting to carry out many programs because organizations with fewer programs will have relatively greater resources to commit to those programs. Indeed, in Nansana, many organizations appear to be attempting to operate numerous pr ograms to serve their communities, despite extremely limited budgets. As such, the expected direction of this variable is unclear.

PAGE 42

31 People Three measures of institutional sustainabi lity are used in the people category. The first measure is the number of beneficiaries served by each organization. Here, I assumed that more beneficiaries sugge sts greater institutional sust ainability. While, on one hand, an organization with fewer beneficiaries may be more sustainable because it will have relatively greater resources to commit to each beneficiary, on the other hand, an organization with more beneficiaries presum ably has the resources to sustain more beneficiaries. Furthermore, organizationsÂ’ goals and objectives are generally to help people. Thus, if an organization is not he lping many people it is not very sustainable with respect to its purpose. Organizations reported their numbers of beneficiaries differently, with some counting individuals (such as children or women) as thei r beneficiaries, and others counting the number of families they serve. For comparative purposes, I translated the number of families into individual units by multiplying the number of families by five, which is the average household size in Uga nda (Uganda Bureau of Statistics, 2010b). The second and third measures of institutiona l sustainability in this category are number of volunteers and number of empl oyees. Organizations with greater human resources, as measured by numbers of volunteers and employees, are likely to have greater institutional sustainability. Of course, it is possible that some organizations with many employees may be spread too thinly a nd unable to sustain their employees, but multiple measures of institutional sustainability are being used in this study in order to compensate for the limitations of any single measure.

PAGE 43

32 With respect to employees and volunteers, it should be noted that many of the organizations are solely volunteer run due to a lack of funds. Local volunteers, however, are only distinguishable from employees ba sed on the compensation they receive. Volunteers typically treat their posts as fulltime jobs, yet they receive only a facilitation fee, which might cover their transportation to and from and the job and minimal other expenses. Thus, consistent with local terminology, I considered employees of organizations to be only those individuals w ho receive a salary beyond a facilitation fee. Volunteers include both local a nd international volunteers. Facilities The physical space or facilities dimens ion of institutional sustainability is measured via three components. First, the num ber of facilities that an organization has on its premises in Nansana may offer some indi cation of its sustainabi lity. Examples of facilities include offices, schools, health clin ics, and housing. Of course, the number of facilities may merely reflect an organizationÂ’s goals or ma y not reflect the number of facilities an organization can appropriately sustain, but, fo r purposes of this study, and because it is, again, only one measure of sust ainability, it will be assumed that an organization with more facili ties is more sustainable. The second measure related to physical sp ace is whether an organization rents or owns its facilities in Nansana. Presumably, an organization that owns its facilities will have greater institutional sustainability, because its location is more stable and because it has more financial resources. Furthermore, at least among organizations interviewed, those that own typically ow n their properties outright.

PAGE 44

33 A third measure in this category relates to the number of offi ces (both in Uganda and internationally) that an organization has outside of Nansana. Presumably, organizations with a greater number of offices will have greater institutional sustainability because they may have access to greater resources, such as informational and material resources, and greater ability to compensate elsewhere when one office location is struggling. Technology In the technology category, two measures of institutional sustainability are employed. The first is the number of types of technology systems an organization has out of the following 10 possible technology system s: website, network, electronic records, domain name, organization e-mail, orga nization phone, computer/laptop, facsimile machine, scanner, and printer. An organizati on with more of these resources will likely be more sustainable and better positioned to reach its constituents The second measure is simply based on the number of computer s or laptops that an organization owns. Finances Two measures of institutional sustainability related to finances will be used in this study: 1) organizationsÂ’ actual budgets a nd 2) organizationsÂ’ actual budgets as a percentage of their expected budgets. Fi nancial sustainabil ity and institutional sustainability are often confused (Cannon, 2002), likely reflecting the f act that it may be tempting to view an organizationÂ’s sustainabil ity as measurable by its finances. Indeed, bigger NGOs are not necessarily more sustainabl e or better able to achieve sustainable results. For example, according to Fowler (2002), large expenditures might suggest that NGOs are more focused on ameliorative effort s than on sustainable results. Thus, it

PAGE 45

34 should be borne in mind that this section is merely one component of institutional sustainability, and, while it may be reasonable to assume that organizations with more money are more sustainable, this is not necessarily the case. For example, an organization with many capital resources and a small monetary budget may be at least as sustainable as an organization with few capita l resources and a larg er monetary budget. Furthermore, just because an organization ha s greater financial resources, does not mean that it is using those resources well. Financial information was obtained by asking organizations about their annual finances for the year 2012. Organizations we re invited to use estimates. If financial information was not yet available for 2012, or ganizations were asked about their 2011 finances. Furthermore, a few of the organi zations provided budgetary information for a period of two or three years, in which cas e I used the mean annual budget for the period covered. Many organizations reported their finances in Ugandan shillings, and one organization reported its finances in Canadian dollars. To convert these figures to US dollars, I used the December 31, 2012 Financia l Management Service exchange rate of 2,686 Ugandan shillings and 0.995 Canadian dollars per 1 US dollar (Financial Management Service, 2013). While there wa s some variation in the budgetary reporting periods, for illustrative purposes, this exchange rate provides reasonable estimates of organizationsÂ’ financial situations. As mentioned, the first meas ure in this category is an organizationÂ’s actual annual budget. The second measure is the actual budg et to expected (or ideal) budget ratio, expressed as a percentage. Among the organizat ions in Nansana, there is a considerable difference between the amount of money that many of the organizations budget for at the

PAGE 46

35 beginning of each year (or budgetary period) a nd the amount of money they actually have in a given year. Possibly reflecting the unpredic tability of funding, rath er than attempt to predict what they may earn in a given year and budget accordingly, many organizations create a budget based on a best-case scenari o. While the anticipated budgets alone seem to say little about an organizationÂ’s institu tional sustainability, the ratio may give some indication regarding the predic tability of an organizationÂ’ s budget or its ability to realistically plan its programm ing and services, both of wh ich logically relate to an organizationÂ’s institutional sustainability. By this logic, percentages closer to 100% are more indicative of inst itutional sustainability. Records Record keeping in the deve loping world is often poor (Barr et al., 2005). To measure institutional sustainability with resp ect to record keeping, organizations were given a score out of five. The first four poi nts relate to whether organizations produce or maintain each of the following: periodic financ ial reports, client reco rds, donor records, and information regarding programs. Organizati ons received an additi onal point if they maintain their records in both hard and soft format. While in the United States paperless has become a more desirable mode of opera tion, the lack of netw orks and servers in Nansana means that many organizations mainta in their electronic file s on memory sticks, which could easily be erased or misplaced. Indeed, one organization reported having just lost most of its electronic files due to a memory stick failure. Thus, it seems that organizations that maintain files in both ha rd and soft form may be more sustainable.

PAGE 47

36 Program Evaluations Though perhaps more related to outcome sustainability, program evaluations may also be an important component of institutio nal sustainability because they suggest some level of continuity in programming in that organizations are periodically evaluating what they are doing and making improvements. To m easure institutional sustainability in this category, I used a binomial variable to reflect whether organizations conduct some sort of evaluations. Thus, organizations received a 1 in this category if there is some indication that they conduct some sort of methodical evaluation of their programming. Limitations of the research methods, including reliance on self-reports, made it difficult to determine what organizations were referring to when th ey reported conducting in ternal or external evaluations. For example, one organization reported conducting exte rnal evaluations, but, when probed, it became clear that the organiza tion representative was merely referring to casual conversations with members of other organizations regardi ng her organizationÂ’s programming. In this situati on, the organization did not recei ve a 1 due to the lack of methodology associated with this type of evaluation. Human Resource Systems To measure institutional sustainabilit y related to human resource systems, organizations received a score out of five based on whether they have the following five human resources systems: orientations fo r new hires, employment files, employee manuals, policies regarding hiring and terminating, and employee access to ongoing training. The presence of thes e items may offer some indi cation of an organizationÂ’s sustainability, in that they conduct some leve l of planning regarding their employees and contribute to their employeesÂ’ growth.

PAGE 48

37 CHAPTER IV OVERVIEW AND FINDINGS RELATED TO SAMPLE In addition to the NCSN, I interviewed 32 organizations. Twenty-eight of these organizations are members of the NCSN and comprise a sample of 80% of its member organizations. The other four organiza tions are other NGOs or CBOs operating in Nansana but which are not part of the formal network. There appeared to be few organizations operating outside the NCSN in Nansana, but I was not able to obtain data related to the same. Returning to the three genera tions of development, in Nansana, the majority of the organizations interviewed remain focused on fi rst generation work in that they continue to strive to meet the immediate needs of vulnerable populations in their communities. There is some evidence of a shift toward s econd generation work in that most of the organizations are local, Ugandan-run NGOs, su ggesting increased levels of local selfreliance. Furthermore, many of the NGOs are seeking and focusing on sustainable solutions to health and poverty issues, which is more in line with second generation work. On the other hand, though, the vast majority of funding for all organizations in the NCSN comes, either directly or i ndirectly, from international sources, suggesting that much work remains before there will truly be local self-reliance. In this chapter, I describe the organi zations operating in Nansana based on data obtained from the 32 organizati ons in the sample. Data are discussed according to the eight categories of institutional sustainability described above; however, additional data were obtained for descriptive purposes that are not included in the actual measures of institutional sustainability.

PAGE 49

38 Generally speaking, as noted above, there has been considerable growth in the civil society sector in Nansana in recent years, and, while the vast majority of organizations are based in Uganda and Ug andan run, funding often comes from outside Nansana, and there is wide va riation in organizationsÂ’ financ ial resources. Programs are typically geared toward child ren, youth, and women, and volunteers play an important role in Nansana, which bodes well for citizen sÂ’ commitments to their fellow community members. Organizations tend to have good planning, record keeping, evaluation, and human resource habits, and most are manage d by boards of directors. Almost all organizations have some physical space, and most are using technology systems to varying degrees. Thus, while there is certainly room fo r improvement given th e relative youth of many organizations (the average organiza tion has been operating for 8.34 years), Nansana organizations have fair ly stable systems in place. Funding is an issue for many organizations, but, particularly through the NCSN, smaller organizations are working to build their capacities through othe r means. Below are more detailed survey and interview results. Services and Programs As mentioned, organizations in Nansana ha ve been operating for a mean length of 8.34 years, though some actually only began serv ing clients more recently. The newest organization is two years old, and the ol dest organization is 33 years old. Of the 32 organizations, 27 are headqu artered in Uganda, and four are headquartered internationally (in Ireland, Japan, Switzerland, and the United States). One organization has headquarters in both Ug anda and Canada. All organizations are

PAGE 50

39 registered as either NGOs in Uganda or CBOs in Wakiso Dist rict (the district in which Nansana is located). Nineteen organizati ons are registered as NGOs (59.4%), and 13 (40.6%) are registered as CBOs. For pur poses of the NCSN, however, one NGO is considered a faith based organization (FBO ) based on its primary emphasis on religion. With respect to planning habits, 29 (90.6%) of the organizations periodically draft strategic plans (defined as documents outlining longer term goals and how an organization will achieve those goals). Or ganizations do this for an average of 25.15 months at a time, with some organizations planning just one month at time and other organizations planning five years at a time. Furthermore, organizations in Nansana have an average of 5.41 programs, with the minimum number of programs being 1 and the maximum number being 9. Organizations were asked in an open-ende d question to describe their programs, and these programs were then recoded into 12 distinct categories reflecting the types of services frequently provided by many of th e organizations. The results are shown in Figure IV.1. The most common type of pr ogram mentioned is educational programming, with 23 organizations (71.9%) re porting having programs in th is area. Here, education refers to programs designed to assist children or youth in their educa tional pursuits. This takes a variety of forms with some organiza tions directly provid ing education through schools on their properties, and others paying full or partial school fees or providing books, uniforms, or other scholastic materials, for children in need. The prevalence of educational support programs is consistent with the findings of Barr et al. (2005), who found that the most prevalent type of NGO activ ity in Uganda is education and training.

PAGE 51

40 Figure IV.1. Percentage of organizations with particular programs. The second most common type of progr am among Nansana organizations is counseling, which refers to programs offeri ng a variety of counseling and psychosocial support, including home visits. Seventeen or ganizations (53.1%) report having programs in this area. Third, 14 organizations (43.8%) report having programs in each of two areas: vocational training and health. Vocational traini ng refers to programs, typically targeting women, designed to teach vocational skills, in cluding tailoring, hairdressing, and baking. Health refers to programs involving the act ual provision of health care, referrals for health care, or health education. Fourth, 13 organizations (40.6%) report having programs related to recreation. Recreation includes programs related to music, dance, drama, arts and crafts, culture, sports, and the like. Some of these programs, particularly arts and crafts, are sometimes 0%10%20%30%40%50%60%70%80% Volunteers Religious Environment Savings & Loans Agriculture HIV/AIDS Community Education Recreation Health Vocational Training Counseling EducationProgram Areas

PAGE 52

41 designed to help raise funds for the organiza tions or their beneficiaries, though profits derived from these programs are minimal. Fifth, 12 organizations (37.5%) have progr ams related to community education. Community education includes programs rela ted to education, advocacy, or increasing community awareness on a range of topics such as substance abuse and domestic violence. Sixth, 11 organizations (34.4%) have HI V/AIDS programs, which typically involve counseling, testing, and treatment re lated to HIV/AIDS. While these programs may also be a subset of health education, he re I broke them out separately because many organizations specifically ta rget HIV/AIDS, and organiza tions typically speak of HIV/AIDS programs as being diffe rent from other health educa tion or treatment efforts. Indeed, it is common for organizations in Nansana to talk about their “HCT” (HIV counseling and testing) programs. Seventh, 8 organizations (25%) have ag ricultural programs, which refer to programs involving farming or livestock. Many of these programs se rve a dual role of teaching beneficiaries agricultural skills a nd either raising revenues or reducing food costs for the organizations. Next, 6 organizations (18.8%) have program s in each of two areas: environmental efforts and savings and loans. Environm ental programs include those that provide education regarding the environment, infras tructure support (such as constructing water tanks in villages), or sanitati on programs (such as periodic cleanup efforts). Savings and loans programs are those related to microfin ance, typically involv ing savings groups or

PAGE 53

42 the provision of small loans or seed money. Savings and loans programs usually target women. Four organizations (12.5%) offer religi ous programming. While many of the organizations in Nansana are religiously based, few specifically reported offering religious programming, but, those that did, typically had programs involving Bible study and evangelism. Finally, three organizations (9.4%) repor ted having volunteer programs. These programs refer to efforts related to vol unteer coordination, in which organizations actually recruit and place volunteers (usually volunteers from abroad). People Generally speaking, there are four groups of people associated with NGOs: beneficiaries, employees, volunt eers, and board members. After converting beneficiaries into indivi dual units, as described above, the mean number of beneficiaries served is 69 9.03, with 10 being the lowest number of beneficiaries and 10,894 being the highest number. The median is 198 beneficiaries. It should be noted, however, that many of the or ganizations loosely de fined beneficiaries, meaning that the numbers of beneficiaries do not necessary reflect t hose individuals with whom the organization has ongoing contact. Fo r example, if an organization held an informational program for the community and 100 people attende d, the organization likely counted those 100 people among its beneficiaries. Categories of beneficiaries are shown belo w in Figure IV.2. The majority of the organizationsÂ’ clients, or beneficiaries, are children, with 84.4% of organizations claiming to benefit children. Twenty-five perc ent of the organizations also serve youth.

PAGE 54

43 While definitions of youth vary across orga nizations, organizations tend to consider youth as beginning in the midto late-teen s and running into an individual’s twenties (though at least two organizations defined youth as including individuals up to age 35). Twenty-five percent of organizations have programs specifically targeting women. Five organizations (15.6%) include people living with HIV/AIDS among th eir beneficiaries, and two organizations (6.3%) se rve those with disabilities. Four organizations (12.5%) consider the community at large to be among their beneficiaries. Figure IV.2. Percentage of organizations s erving each category of beneficiaries. Organizations have a mean of 9.88 employ ees. With respect to employees and volunteers, again, many organizations depe nd on full-time volunteers who receive only facilitation fees. According to Barr et al. (2005), “[f]ulla nd part-time volunteers account for most of the manpower available to surveyed NGOs” (p. 669). Indeed, eight organizations (25%) have no employees and th e largest employs 72 individuals. On the other hand, 29 organizations (90.6%) have volun teers. Among the 28 organizations that reported having local volunteers, the mean number of local vo lunteers is 8.18. Only six 0%20%40%60%80%100% Disabled Community HIV/AIDS Youth Women ChildrenBeneficiaries

PAGE 55

44 organizations reported having international volunteers, and the average number of international volunteers for those organizations is 3.2. Of the 32 organizations, 29 report that they have a board of directors. Twentyeight organizations have a board of director s in Uganda, and four have international boards, meaning that three organizations have boards both in Uganda and in the country where they are headquartered. Three of the Ugandan boards also have at least one foreign member, while the rest consist sole ly of Ugandans. The mean size of the Ugandan board is 7.54 people, with three peop le comprising the smallest board and 12 comprising the largest. Boards meet betw een once a month and once a year with the average meeting occurri ng once every 4.61 months. Facilities Looking at the types of facilities and physic al structures that house organizations in Nansana, 30 of the 32 organizations repor ted having some sort of physical property, and one of the two without property reported being between offices at the time of the interview. All organizations with property have an office, and schools and residential facilities (generally for orphans) are the next most common type of facility with 25% of organizations having each of these facilities. Other types of facilities include vocational training centers, which 12.5% of organizati ons have; volunteer housing, which 9.4% of organizations have; agriculture or livestock areas, health clin ics, and libraries, each of which 6.3% of organizations have; and a church, run by one (3.1%) organization. Of the thirty organizations that had properties at the time of their interviews, 19 rent their space (63%) and 11 (37%) own th eir space. Furthermore, 14 of the 32

PAGE 56

45 organizations have other offices or propertie s outside of Nansana as well. Of the 14, 13 have other offices in Uganda, and four have other offices internationally. Technology All but one of the organizations surv eyed reported having some form of technology system. Organizations were aske d about 10 specific types of technology: website, network, electronic records, do main name, organization e-mail account, organization phone, computer/laptop, fax machin e, scanner, and printer. The most frequent type of technology re ported by organizations was el ectronic records, with 31 organizations (96.9%) claiming to have at least some of their files in electronic format. Next, 29 organizations (90.6%) have e-mail accounts. Twenty-three organizations (71.9%) have at least one organizationa l computer or laptop, and among those organizations with computers or laptops, th e mean number of computers was 4.76, with the numbers of computers per organization ranging from 1 to 36. Twenty organizations (62.5%) have a webs ite, though there is variation regarding whether the websites are directly managed by the organization or by a free website hosting service. Eighteen organizations (56.3%) have an organi zational phone (either mobile or landline), and the same number of organizations have a printer. Twelve organizations (37.5%) have a scanner, and tw elve have an organization domain name. Six organizations (18.8%) have a network, and, finally, no organizations have a fax machine. Finances With respect to annual b udgets, again, there is a distin ction between organizationsÂ’ expected (or ideal) budgets a nd their actual budgets. Among those organizations that

PAGE 57

46 reported their finances, the mean expected annual budget is $46,104, and the median is $18,615. Expected budgets range from $856 to $297,840. The mean actual budget is $40,274, and the median is $7,167. The lowest actual budget is $0, and the highest is $297,840. I also calculated the actual budget as a percentage of the expected budget given the difference found between these figures. Indeed, the mean organization’s actual budget is 55% of its expected budget. While the lowest pe rcent was 0%, reflecting the fact that one organization reported having no money in 2012, one organization’s actual budget was 1% of its expected budget. Many organizations’ actual a nd expected budgets were the same or similar, though. As far as the sources of their finances, organizations were asked what percentage of their funds come from each of seven cat egories: individual donors; foundation grants, including grants from foreign government ag encies, such as USAID; a government entity in Uganda; membership fees; fees for servi ces; commercial income, such as that from farming or sale of goods; and an “other,” catchall category. The largest source of funds for organiza tions in Nansana is individual donors, both local and international. Organizations reported receivin g an average of 41% of their funds from individual donors, and many orga nizations reported receiving all of their funds from individual donors. It should be noted that th e percentages across organizations were averaged. The next most common source of income is the catchall category, with organizations reporting receiving an average of 20% of their funds from other sources, such as staff contributions, funds funneled from other NGOs, and other sources. While

PAGE 58

47 many organizations received no foundation gran ts, on average organizations received 14% of their funds from foundations or international government entities, such as USAID or other countriesÂ’ equivalents. On average, organizations receive 10% of their money from fees for services, most commonly from school fees. Membership fees account for an average of 9% of organizationsÂ’ budgets, and commercial income accounts for 5% of their budgets. Finally, very little money comes from government entities in Uganda. Only one organization reported receiving a very small amount from the Ugandan government for an overall average among organizations of 0.07%. Organizations were also asked what per centage of their income is spent on each of four categories: programs, staff, admi nistrative expenses, and other items. The majority of organizationsÂ’ finances is spen t on their programs. Organizations reported spending an average of 52% of their funds on programs, though the range is from 0 to 100%. Administrative expenses, such as rent technology, and other administrative costs, account for 28% of organizationsÂ’ budgets. St aff salaries average 15% of organizationsÂ’ budgets, with many organizations having few or no staff expenses due to the widespread use of volunteers, and 6% of budgets reporte dly go toward other expenses, such as payments directly to those in need. Records To understand the record keeping habits of the organizations in Nansana, the organizations were asked whether they keep records in four areas: finances, clients, donors, and programs. They were also asked wh ether they maintain th eir files in written or electronic format, or both.

PAGE 59

48 Organizations reported high rates of record keeping. All but one organization reported maintaining periodic financial repo rts, and, on average, those organizations producing periodic financial reports repor ted doing so 2.84 times per year. All organizations reported maintaining records of th eir clients. The limitations of self-reports, however, were obvious when one organization reported maintaining client records but also claimed that its benefi ciaries are “uncountable.” Twenty-seven organizations (84.4%) reported maintaining donor files. All five of those without donor records, however, claim to not have them not for want of record keeping, but rather because they do not have donors. Twenty-nine organizations (90.6%) report maintaining some sort of records regarding their programs. With respect to how these records are ke pt, 30 organizations reported maintaining both hard and soft copies of their files, t hough both sets of files were not necessarily complete. One organization maintains only hard copies of its files, and one maintains only soft copies. It seems difficult to say he re whether one form of record keeping is superior to another in the Ugandan context, particularly since the vast majority of electronic records are no t backed up by a server. Program Evaluations Twenty-seven organizations (84.4%) repor ted conducting some sort of evaluation of their programs, though the formality of the evaluations are likely widely variable, with some organizations considering discussions re garding programs at period staff meetings to be evaluations. Eleven organizations (34.4%) reported conducting both internal and external evaluations. Thirteen organizatio ns (40.6%) reported conducting only internal evaluations, and three organiza tions (9.4%) reported conducting only external evaluations.

PAGE 60

49 It is also worth noting that in their study, Barr et al. ( 2005) noted that Uganda NGOs might over-report their external evaluations. Human Resource Systems Finally, organizations were asked a series of questions related to their human resource systems. Perhaps reflecting the widespread use of and dependence on volunteers, even among those organizations that are solely volunteer-run, many have human resource systems in place. Twenty-seven organizations (84.4%) re ported conducting orientations for new hires (including volunteers), though the forma lity of these trainings are likely widely variable. The same number also reported having written policie s regarding hiring and termination, and offering their employees access to ongoing training. These high numbers likely reflect the fact that organiza tional constitutions (requ ired for registration with the district or nationa l government) typically include a section regarding grounds for termination, and, with respect to training, one of the functions of the NCSN is to offer training to employees of member organizati ons and to alert member organizations of trainings hosted by partner organizations. Twenty-four organizati ons (75%) reported maintaining employment files for employees or volunteers, and 17 organizations (53.1%) reported having employee manuals that are distributed to new hires.

PAGE 61

50 CHAPTER V NETWORK MAPPING AND RESULTS While the preceding chapter explored the in stitutional sustainability of NGOs in Nansana, this chapter looks more deeply at the relationships among organizations in the sample. Both through the formal networ k and through partnerships, organizations develop relationships in order to enhance th eir capacity and service delivery. There is also some evidence that when internationa l aid is delivered to organizations that participate in the network, those organizations in turn help to suppor t other organizations with more limited resources, thereby creating synergistic partnerships. Thus, through better understanding the NCSN, a nd interorganizati onal relationships in Nansana more broadly, we can begin to unde rstand how efficient service delivery can occur through local NGO initiatives. To understand the relationships between organizations in the NCSN and, more broadly, in Nansana, each interviewee was as ked about his or her organizationÂ’s partner organizations. Organizations were first asked to list all of their partners and were then asked a series of questions regarding what each relationship entails. Relationships were mapped using UCINET and NetDraw.3 Figure V.1 shows all of the relationships among the organizations in the sample. 3 The discussion and analysis in this chapte r is largely informed by Hanneman and Riddle (2005).

PAGE 62

51Green = no direct international ties Purple = limited international ties Red = strong international ties = member of NCSN = nonmember Figure V.1. All network ties. The circles and square in the map are “ nodes” representing organizations. Their colors, shapes, and sizes represent attributes of the organizations, or certain relevant characteristics. For example, organizations represented by circles are members of the NCSN and organizations represented by square s are not members. The lines represent relationships between the organizations, with arrows pointing from each interviewee to each partner it mentioned. The four organizations on the left side of the map are isolates, which are not connected to the other organiza tions in the sample. These organizations did not report having any partnerships with organizations in the sample and were not mentioned as partners by other organizations. The key is on the right side of the map. Appendix B contains a glossary of terms related to network analysis. To avoid overcomplicating the map, only two attributes, or characteristics, were assigned to the nodes (which represent organi zations) in the map. The first relates to whether the organization is a member of the NCSN. As mentioned, organizations represented by circles are members of th e NCSN, and organizations represented by 1 15 21 10 2 3 6 12 4 23 20 5 8 7 9 13 11 16 22 14 18 19 28 24 25 27 30 31 17 26 29 32

PAGE 63

52 squares are not members. Only four organi zations in the sample are not part of the NCSN, and, as can be seen in the map, two of these organizations are connected to the map and two are isolates. The second attribute relates to the level of an organizationÂ’s in ternational ties. The organizations shown in red are those orga nizations with international offices, boards of directors, or incorporation. In all but one instance, if an organization has one of these characteristics, it has all three. Thus, th is group consists of or ganizations that are internationally based and run. The organizati ons shown in purple are those with a direct international funding source, an international board member, or international volunteers. As such, these organizations, though locally ba sed, have meaningful international ties. Finally, the organizations shown in green are t hose with no direct inte rnational ties. The sizes of the nodes also depend on the strength of the organizations Â’ international ties. Furthermore, the map includes all asymmetr ic, or directional, ties, meaning that the lines represent each time an organizati on mentioned another, regardless of whether the target organization recipr ocated the mention. Thus, an organization may appear in the map as a hub of activity because many orga nizations mentioned it as a partner or because it mentioned many other organizations as a partner. Thus, for example, organizations 15, 21, and 8 appear to be hubs in the map with many arrows coming to them, while organization 30 appears to be a hub but only with arrows going out from it. Despite the fact that I used the same de finition of partner w ith each organization I interviewed, the findings reveal that organizations have thei r own conceptions of what a partner is because there was little symmetry in the relationships. In fact, while the overall map consists of 91 ties, only 24 of these ties are reciprocal, meaning that there are only

PAGE 64

53 12 bonded ties (or reciprocal ties in which both organizations mentioned each other as a partner) in the sample. Figure V.2 shows all bonded ties in the sample. Green = no direct international ties Purple = limited international ties Red = strong international ties = member of NCSN = nonmember Figure V.2. Bonded ties. I discussed this asymmetry with one of the organizations that was often mentioned as a partner by other organizations but which mentioned few organizations in Nansana as partners. This or ganization limited its conception of partner to “strategic partners,” or those whom it specifically sought a partnership with in order to enhance its capacity. Thus, at the risk of over generalizing, the smaller or ganizations that considered this organization to be a partner might have been viewing it as a partner based on its ability to enhance their capacity, whil e the organization at the hub largely only considered its partners to be those outsid e of Nansana with greater resources and the ability to enhance its own capacity. Because of this difference in conception of partnership and be cause of the limited number of bonded (or reciproc al) ties among organizations in the sample, in the next chapter regarding correlations between numbers of partners and measures of institutional sustainability, I separately analyze measures of in degrees, which refers to the number of 12 13 14 15 16 20 21 28 3 5 7 8 9

PAGE 65

54 times organizations were mentioned as partne rs, and out degrees, or how many partners each organization mentioned. To gain a better understanding of the types of partnershi ps that organizations in Nansana tend to engage in, for each partner, I asked organizations a series of questions related to the strength of th e partnerships. These questi ons included the length of the partnerships, the formality of the partnerships, whether the organizations have joint programs, what types of resources they shar e, and how often they communicate. Again, the survey can be found in Appendix A. Partnerships Lengths Beginning with the lengths of the partne rships, I divided the lengths into three categories: 1) less than one year, 2) one to thr ee years, and 3) greater than three years. As mentioned, the NCSN began forming in 2008 and became official in 2010. As Figure V.3 shows, while few additional partnerships have formed in the last year, there was a substantial increase in the num ber of partnerships formed among organizations in the NCSN in the last three years, perhap s reflecting the effects of the NCSN. Partnerships longer than three years. Partnerships of one to three years. All current partnerships. Figure V.3. The addition of partnerships over three years. The maps show that, along with an incr ease in the number of partners (possibly related to the development of the NCSN), th ere was also a signifi cant increase between one and three years ago in the number of ways that organizations can connect with each

PAGE 66

55 other. In other words, based on the new part nerships that formed, there was an increase in the connectivity among organizations because, even if they are not directly connected to one another, organizations are more closel y connected indirectly. Furthermore, while many more additional partnershi ps are possible in Nansana (only 91 of 992 possible ties currently exist), there is s till considerable connectivity and numerous opportunities or ways in which various organizations can connect. Partnership Formality Next, with respect to formality, looking at Figure V.4, it appears that few of the ties are formal (36 of 91), and, while I aske d about a variety of types of formality (including joint ventures, parent-subsidiary re lationships, and co-gra nt recipients), the only type of relationship repor ted were other contractual ba sed relationships, such as memoranda of understanding. Also, many of the organizations reported having a contractual relationship with one organization in particular (organization 15). Indeed, this organization accounts for 18 of the 36 formal relationships (however, this organization did not report having a formal re lationship with any of the others). Thus, the second map below shows what happens when formal relationships with this single organization are removed. All formal partnerships. Formal partnerships excluding organization 15. Figure V.4. Formal partnerships. 1 15 21 10 2 3 12 23 20 8 7 9 13 11 16 22 14 18 19 28 25 30 31 1 10 11 12 14 18 19 2 20 21 22 28 3 30 8 9

PAGE 67

56 While the formality of relationships does not necessarily correlate with relationship strength or qual ity, formality could contribute to the sustainability of relationships. Thus, though the numbers are limited, it does a ppear that some organizations are entering into more formal relationships with each other. Furthermore, many of the organizations clearly perceive their relationships w ith organization 15 as formal (though it is not clear whether the rela tionships are, in fact, formal), perhaps suggesting that relationships w ith this organization may seem particularly important to the organizations involved. If this is the case, organi zation 15 may be playing an important role in connecting organiza tions and facilitating relationships. Joint Programming Among Partners The next question regarding the nature of interorganizational partnerships is whether the organizations have joint program s. In 43 of the 91 ties, organizations reported having joint programs with one anot her, and these joint programs primarily centered around two organizations, as can be seen in Figure V.5. Again, there appeared to be wide variation in what the organizati ons considered joint programs, and there was little agreement among organi zations regarding whether they were conducting joint programs. The fact that two organizations with international ties are often involved in joint programs, however, may suggest that in ternational resources ar e being dispersed to other organizations in the network and possibl y used in synergistic relationships between organizations with more and less resources.

PAGE 68

57 Green = no direct international ties Purple = limited international ties Red = strong international ties = member of NCSN = nonmember Figure V.5. Partnerships involving joint programs. Resource Sharing Among Partners Next, I asked organizations about their levels of resource sharing with their partners. Organizations were asked to indicate which of the following they share with each of their partners: information, money, cl ients, physical space, materials, staff or volunteers, board members, and administrative capacities. The most frequent type of resource shared is information with 90 of 91 partnerships incorporating information sharing (in fact, it is difficult to imagine how a partnership occurs without information sharing). The second most fr equent type of resource shar ed is beneficiaries, which primarily involves client referrals for vari ous services. Of the 91 partnerships, 45 reportedly include beneficiary sh aring. Other types of resources shared include materials, such as posters, books, and health care items (18 partnerships); staff or volunteers (11 partnerships); and physical sp ace (7 partnerships). Ad ministrative capacities and finances were each only mentioned once, a nd no organizations reported sharing board members. 1 15 21 10 2 3 6 12 4 20 5 8 7 9 13 16 14 19 28 24 25 27 30

PAGE 69

58 The maps shown in Figure V.6 incor porate the strength of ties between organizations based on the number of resources they reported sharing with their partners. Thicker lines reflect more resources being shared (in bonded ties, the higher number of reported resources dictates the line thickness). As can be s een, few organizations report sharing more than two resources with any of their partners. Furthermore, only one partnership reportedly incorpor ates more than three types of resource sharing, which, interestingly, is in the partne rship between the two organizations with the most in degrees (or the most mentions by other organizations ). One of these organizations reported sharing five types of resources with the other, and the other reported that they share three types of resources. Thus, while there are ma ny partnerships within Nansana, there is little collaboration in that organizations re main largely separate from one another. All resource sharing. Two or more types of resource sharing. Three or more types of resource sharing. Four or more types of resource sharing. Figure V.6. Partnerships by levels of resource sharing.

PAGE 70

59 Contact Frequency Among Partners The final measure of connectivity relate s to the frequency of contact between the organizations. Contact was divi ded into five categories: vari able (which merely indicates the respondentÂ’s inability to gauge any pattern in frequency of communication), at least annually, at least monthly, at le ast weekly, and daily. Again, thicker lines represent more frequent contact. In Figure V.7, the first map shows all types of contact between the organizations; the second map shows all cont act that occurs at least monthly (also excluding variable contact); and the third ma p shows all contact that occurs at least weekly. As can be seen, the majority of cont act occurs at least monthly, but not weekly. Only two organizations reporte d communicating daily, which may be attributable to these organizationsÂ’ close proximity to one another. It should also be noted that, particularly in large organizations, it may be difficult for a single organizational representative to estimate frequency of contact. All contact. At least monthly contact. At least weekly contact. Figure V.7. Partnerships by frequency of contact. Additional Analyses of Partnerships As mentioned, for the majority of the maps in this chapter, I chose to include only two node attributes (internati onal ties and network membersh ip) in order to maintain readability. There are numerous other attrib utes that could have been included, however, such as number of employees, financia l resources, NGO/CBO/FBO classification,

PAGE 71

60 number of beneficiaries, prim ary service area, and so forth. Many of these would offer useful additional information. The only additional attribute I looked at is financial resource s. For this attribute, I divided the organizations into quartiles base d on their finances. F our organizations are not included in this map because I was unabl e to obtain financial data for them. Red organizations are those in the first quartile; yellow organizations ar e those in the second quartile; green organizations ar e those in the third quartile; and blue organizations are those in the fourth quartile. As can be seen in Figure V.8, the majority of organizations with the most in degrees are in the fourth quartile based on their actual budgets, showing that better funded organizations are more fr equently mentioned as partners by other organizations. There is some variatio n, though, perhaps reflecting other types of resources at play or reluct ance on the part of some orga nizations with substantial resources to partner with othe r organizations in Nansana. Again, this suggests that organizations in Nansana may have efficiently organized themselves to spread resources and enhance service delivery.

PAGE 72

61 Red = first quartile Yellow = second quartile Green = third quartile Blue = fourth quartile Figure V.8. Network map including financial resources attribute. Looking back at the literature re garding cooperation, coordination, and collaboration, it seems that most of the rela tionships represented in these network maps are either cooperations or coordinations. Due to the lack of recipr ocity in how various organizations describe their relationships w ith one another, however, it is difficult to categorize the relationships. It seems that mo st, if not all, of the organizations in the network are engaged in some form of coopera tion with each of the other organizations in the network. With respect to the partners specifica lly identified by organi zations, except for those that are merely information sharing, mo st of these organizations are likely engaged in coordination. There appears to be little collaboration occurring between organizations in that the organizations remain largely sepa rate from one another. The NCSN, however, is likely a collaboration in that it consists of organizations coming together to form a new structure. The NCSN may even be a commun ity-building collaboration in that it affects 1 15 21 10 2 3 6 12 4 23 20 5 8 7 9 13 11 16 14 19 28 24 25 27 30 17 26 29

PAGE 73

62 and benefits the entire community in wh ich it is embedded, not just its member organizations. Thus, the organizations in Nansana are we ll connected with one another, and they have increased their connectivity considerably since the NCSN became active. While the organizations actually share little in the wa y of resources, the fact that those with substantial resources tend to have many pa rtners suggests that productive partnerships exist. Indeed, according to Claiborne et al (2009), relationships between macro-level organizations (including larger NGOs) and micro-level orga nizations are important to capacity building.

PAGE 74

63 CHAPTER VI INSTITUTIONAL SUSTAINABILITY AND PARTNERSHIPS This chapter tests the primary hypothesis of this thesis, which is that there is a positive correlation between the number of partners an organization has and its institutional sustainability. This thesis doe s not purport to rank organizations according to their institutional sustainability. Rather, th e question is to what extent networks and partnerships correlate with certain measures of institutional sustainability. The results from three separate sets of anal yses are reported in th is chapter. In the first set, organizations that are members of the NCSN are compared with organizations that are not members of the NC SN. In the second set, orga nizations are compared based on their numbers of out degrees and, in the third set, orga nizations are compared based on their numbers of in degrees. For each se t of analyses, organizations are compared based on each of the eight categories and 15 measures of institutional sustainability described in Chapter III. Differences Between NCSN Members and Nonmembers In this section, I compare organizations in Nansana that are members of the NCSN with organizations that are not memb ers of the NCSN. As mentioned, however, 28 of the organizations interviewed are members of the NCSN, while only four organizations are not members of the NCSN. Given the small sample in the nonmember group, t-tests and chi-squared measures did not i ndicate that the results in this section are significant. The results ar e summarized in Table VI.1.

PAGE 75

64 Table VI.1. Comparisons based on NCSN membership. Institutional Sustainability Measures NCSN members NCSN nonmembers Services and Programs years of operation 8.89 4.50 # of programs 5.43 5.25 People # of beneficiaries 745.46 374.00 # of employees 9.43 13.00 # of volunteers 8.41 5.00 Facilities # of facilities 2.00 1.50 ownership rate 39.29% 0.00% # of offices outside Nansana 1.93 2.50 Technology # of tech items 5.32 5.00 # of computers 4.85 3.00 Finances actual budget $44,366.36 $6,167.33 actual: expected 55.16% 50.43% Records points out of 5 4.68 4.50 Program Evaluations rate conducting 82.14% 100.00% Human Resource Systems points out of 5 4.15 4.67

PAGE 76

65 Services and Programs Beginning with the measures of institutiona l sustainability related to services and programs, NCSN organizations have been ope rating for an average of 8.89 years, while nonmembers have been operating for only 4.5 ye ars. There is little difference between the two groups with respect to their numbers of programs, with NCSN organizations having an average of 5.43 programs and nonmembers having an average of 5.25 programs. People In the people categories, NCSN or ganizations have an average of 745.46 beneficiaries, while nonmembers organizations have an average of 374.00 beneficiaries. NCSN organizations employ an average of 9.43 people, and nonmember organizations employ an average of 13 people. Finally, NC SN organizations have an average of 8.41 volunteers, and nonmember organizations have an average of 5 volunteers. Facilities The next category of institutional sustaina bility relates to or ganizationsÂ’ physical space or facilities. NCSN organizations have an average of two types of facilities at their Nansana premises, while nonmember organiza tions have an average of 1.5 types of facilities at their Nansana pr emises. Among NCSN organizati ons, 15 rent their facilities in Nansana; 11 own their facilities in Nansana; and two do not have offices. All four of the nonmember organizations rent their facilitie s in Nansana. With respect to offices outside of Nansana, NCSN organizations ha ve 1.93 offices outside of Nansana (either within Uganda or internationa lly), and nonmember organizati ons have 2.5 offices outside of Nansana.

PAGE 77

66 Technology The next category of institutional sustai nability relates to an organizationÂ’s technology. Organizations receiv ed a score out of 10 related to the number of specified technology systems that they have (website, network, electronic records, domain name, organization e-mail account, organization phone, computer/laptop, fax machine, scanner, and printer). NCSN organizations scored an average of 5.32 out of 10, and nonmember organizations scored an average of 5. NCSN organizations also have an average of 4.85 computers or laptops, and nonmember organiza tions have an average of 3 computers or laptops. Finances Here, I use two measures of institutional sustainability related to an organizationÂ’s finances: its actual budget and it s actual budget as a percentage of its expected budget. NCSN organizations have an aver age actual annual budget of $44,366.36, and nonmember organizations have an averag e actual annual budget of $6,167.33. Also, the actual budget to expected budget ratio for NCSN organizatio ns is 55.16%, and the ratio for nonmember organizations is 50.43%. Records There was little difference in record k eeping between the two groups. Out of a total possible score of 5, NC SN organizations received an average score of 4.68, while nonmember organizations scored 4.50. Program Evaluations The next measure looks at whether orga nizations conduct some sort of evaluation of their programs. All four of the nonmember organizations report conducting

PAGE 78

67 evaluations of their programs, while 23 of 28 NCSN organizations report conducting evaluations. Human Resource Systems Finally, organizations received a score out of five related to their human resource systems. NCSN organizations scor ed an average of 4.15, while nonmember organizations scored 4.67. Significant Findings Re garding NCSN Membership Though the results in this category were not significant in light of the small-n, there do appear to be some differences between NCSN members and nonmembers. Fairly large differences between the groups were found in many categories, and, in most instances, member organizations are fari ng better than nonmember organizations. Differences Based on Numbers of Out Degrees The second set of analyses performed in relation to organizationsÂ’ connectivity is based on their number of out degrees, or th e number of partners that each organization reports having among those organizations in the sample. All organizations in the sample were included in this analysis The mean number of out de grees is 2.84, with a median of 3.00 and a standard deviation of 2.13. To l ook at correlations betw een numbers of out degrees and the various measures of instit utional sustainability, organizations were divided into three groups. The first group of organizations re ported having no or one ties with other organizations in the sample (n=9 ). The second group has two or three ties (n=13), and third group has four or more ties (n=10). Unless otherwise stat ed, a one-way ANOVA was used to test differences in mean scores between the groups. Table VI.2 is a summary of the results in this section.

PAGE 79

68 Table VI.2. Comparisons based on numbers of out degrees. Institutional Sustainability Measures Group 1 (0-1 out degrees) Group 2 (2-3 out degrees) Group 3 (4+ out degrees) Services and Programs years of operation 10.44 7.08 8.10 # of programs 5.22 5.08 6.00 People # of beneficiaries 496.89 394.85 1276.00 # of employees 19.11 3.85 9.40 # of volunteers 6.86 6.31 11.20 Facilities # of facilities* 1.89 1.42 2.60 ownership rate* 33.33% 7.69% 70.00% # of offices outside Nansana 2.78 2.23 1.00 Technology # of tech items 5.67 5.00 5.30 # of computers 3.20 3.25 7.25 Finances actual budget $49,202.71 $27,797.27 $47,747.20 actual: expected 52.05% 56.76% 53.92% Records points out of 5 4.56 4.54 4.90 Program Evaluations rate conducting 77.78% 76.92% 100.00% Human Resource Systems points out of 5 4.13 4.00 4.50 Differences are significant at p < .05

PAGE 80

69 Services and Programs Beginning with services, organizations in Group 1 (0 or 1 ties) have been operating for an average of 10.44 years. Organi zations in Group 2 (2 or 3 ties) have been operating for an average of 7.08 years, and orga nizations in Group 3 (4 or more ties) have been operating for 8.10 years. With respect to programs, organizations in Group 1 have an average of 5.22 programs. Organizati ons in Group 2 have an average of 5.08 programs, and organizations in Group 3 have an average of 6 programs. For each of these categories, the differences between th e groups are not significant, as shown in Tables VI.3 and VI.4. Table VI.3. Comparison of years of opera tion based on numbers of out degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 9 10.44 10.631 2.27 18.62 Group 2 13 7.08 6.751 3.00 11.16 Group 3 10 8.10 4.701 4.74 11.46 F = 5.38, p = .590, Eta = .036 Table VI.4. Comparison of programs based on numbers of out degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 9 5.22 1.394 4.15 6.29 Group 2 13 5.08 2.253 3.72 6.44 Group 3 10 6.00 2.000 4.57 7.43 F = .676, p = .517, Eta = .045

PAGE 81

70 People Looking at the various people associated with organizations, organizations in Group 1 serve an average of 496.89 beneficiarie s. Organizations in Group 2 serve an average of 394.85 beneficiaries, and organiza tions in Group 3 serve an average of 1,276 beneficiaries. With respect to employees, organizations in Group 1 have an average of 19.11 employees. Organizations in Group 2 have an average of 3.85 employees, and organizations in Group 3 have an average of 9.40 employees. Finally, organizations in Group 1 have an average of 6.86 volunteers. Organizations in Group 2 have an average of 6.31 volunteers, and organizations in Gr oup 3 have an average of 11.20 volunteers. Again, these results are not significant, as shown in Tables VI.5, VI.6, and VI.7. Table VI.5. Comparison of beneficiarie s based on numbers of out degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 9 496.89 696.989 38.86 1,032.64 Group 2 13 394.85 547.578 63.95 725.74 Group 3 10 1,276.40 3,388.294 1,147.44 3,700.24 F = .642, p = .534, Eta = .042 Table VI.6. Comparison of employees based on numbers of out degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 9 19.11 26.408 1.19 39.41 Group 2 13 3.85 5.505 .52 7.17 Group 3 10 9.40 17.128 2.85 21.65 F = 2.099, p = .141, Eta = .126

PAGE 82

71 Table VI.7. Comparison of volunteers based on numbers of out degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 7 6.86 6.176 1.15 12.57 Group 2 13 6.31 4.211 3.76 8.85 Group 3 10 11.20 14.367 .92 21.48 F = .873, p = .429, Eta = .061 Facilities The next category of institutional sustaina bility relates to organizationsÂ’ offices and facilities. Beginning with the number of facilities each organization has on its premises in Nansana, organizations in Gr oup 1 have an average of 1.89 facilities in Nansana. Organizations in Group 2 have an average of 1.42 facilities in Nansana, and organizations in Group 3 have an average of 2.60 facilities in Nansana. These results are significant at p < .05 (p = .035), as shown in Table VI.8. In line with my general hypothesis, Group 3 has the greate st number of facilities, but Group 1 has more facilities than Group 2, perhaps suggesting a pattern wher ein organizations in Group 1 have more resources than those in Group 2. Table VI.8. Comparison of Nansana facilities based on numbers of out degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 9 1.89 1.453 .77 3.01 Group 2 12 1.42 .669 .99 1.84 Group 3 10 2.60 .843 2.00 3.20 F = 3.805, p = .035, Eta = .214

PAGE 83

72 Next, with respect to whether organizations rent or own their facilities, five of nine organizations in Group 1 re nt their facilities; three own; and one does not have an office. In Group 2, 12 of 13 organi zations rent their facilities, and one owns its facilities. In Group 3, two organizations rent their facilities, seven own their facilities, and one does not have an office. Chi-squa re tests indicate that these differences are significant (p = .012), as shown in Table VI.9. Table VI.9. Comparison of ownership rates based on numbers of out degrees. Rent Own No office Row N Group 1 55.6% 33.3% 11.1% 9 Group 2 92.3% 7.7% 0.0% 13 Group 3 20.2% 70.0% 10.0% 10 Chi2 = 12.773, d.f. = 4, p = .012 Finally, looking at the numb er of offices each organization has outside Nansana (both elsewhere in Uganda and internati onally), organizations in Group 1 have an average of 2.78 offices outside Nansana. Or ganizations in Group 2 have an average of 2.23 offices elsewhere, and organizations in Group 3 have an average of one office elsewhere. These results are not si gnificant, as shown in Table VI.10. Table VI.10. Comparison of offices outs ide Nansana based on numbers of out degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 9 2.78 2.774 .65 4.91 Group 2 13 2.23 5.003 .79 5.25 Group 3 10 1.00 1.414 .01 2.01 F = .616, p = .547, Eta = .041

PAGE 84

73 Technology Out of 10 possible technology systems, organizations in Group 1 averaged 5.67 technology systems. Organi zations in Group 2 averaged 5 technology systems, and organizations in Group 3 averaged 5.3 tec hnology systems. Organizations in Group 1 also have an average of 3.20 computers or la ptops. Organizations in Group 2 have an average of 3.25 computers or laptops, and or ganizations in Group 3 have an average of 7.25 computers or laptops. These results are not significant, as s hown in Tables VI.11 and VI.12. Table VI.11. Comparison of technology syst ems based on numbers of out degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 9 5.67 2.646 3.63 7.70 Group 2 13 5.00 2.309 3.60 6.40 Group 3 10 5.30 2.359 3.61 6.99 F = .202, p = .818, Eta = .014 Table VI.12. Comparison of computers based on numbers of out degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 5 3.20 1.095 1.84 4.56 Group 2 8 3.25 2.121 1.48 5.02 Group 3 8 7.25 11.961 2.75 17.25 F = .694, p = .513, Eta = .072 Finances Results with respect to financial measures were also not significant, as shown in Tables VI.13 and VI.14. Organizations in Group 1 have an actual annual budget of

PAGE 85

74 $49,202.71. Organizations in Group 2 have an actual budget of $27,797.27, and organizations in Group 3 have an ac tual budget of $47,747.20. Furthermore, organizations in Group 1 have an actual to expected budget ratio of 52.05%. The ratio for organizations in Group 2 is 56.76%, and the ratio for organizations in Group 3 is 53.92%. Table VI.13. Comparison of actual budget based on numbers of out degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 7 49,202.71 109,842.712 52,384.85 150,790.28 Group 2 11 27,797.27 44,037.646 1,787.62 57,382.17 Group 3 10 47,747.20 92,386.076 18,341.82 113,836.22 F = .210, p = .812, Eta = .017 Table VI.14. Comparison of budget ratio based on numbers of out degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 7 .5205 .46849 .0872 .9538 Group 2 11 .5676 .28710 .3747 .7604 Group 3 8 .5392 .38914 .2138 .8645 F = .036, p = .965, Eta = .003 Records Looking at record keeping as a category of institutiona l sustainability, out of a score of 5, organizations in Group 1 received an average score of 4.56. Organizations in Group 2 received an average score of 4.54, a nd organizations in Group 3 received an average score of 4.90. These results are not significant, as shown in Table VI.15.

PAGE 86

75 Table VI.15. Comparison of record keeping based on numbers of out degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 9 4.56 .527 4.15 4.96 Group 2 13 4.54 1.198 3.81 5.26 Group 3 10 4.90 .316 4.67 5.13 F = .617, p = .547, Eta = .041 Program Evaluations Seven of nine organizations in Gr oup 1 report conducti ng some sort of evaluations of their progra ms. Ten of 13 organizations in Group 2 report conducting some sort of evaluations, and all 10 of the organizations in Group 3 report conducting evaluations. Chi-square tests were performed on these outcomes, and the results were not significant, as show n in Table VI.16. Table VI.16. Comparison of evaluation s based on numbers of out degrees. Yes No Row N Group 1 77.8% 22.2% 9 Group 2 76.9% 23.1% 13 Group 3 100.0% 0.0% 10 Chi2 = 2.697, d.f. = 2, p = .260 Human Resource Systems Finally, results with respect to diffe rences among the groups related to human resource systems were not significant, as s hown in Table VI.17. Out of a score of 5, organizations in Group 1 received a score of 4.13. Organizations in Group 2 received a score of 4, and organizations in Group 3 received a score of 4.5.

PAGE 87

76 Table VI.17. Comparison of human resource systems based on numbers of out degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 8 4.13 1.458 2.91 5.34 Group 2 11 4.00 1.265 3.15 4.85 Group 3 10 4.50 .707 3.99 5.01 F = .508, p = .607, Eta = .038 Significant Findings Regarding Out Degrees Thus, results with respect to out degrees, or the number of partners mentioned by each organization, were only signi ficant in two categories: the number of facilities the organizations have in Nansana and their rate of ownership of their properties in Nansana. As mentioned in Chapter V, this may be refl ective of a validity issue with respect to the definition of partner, in that organizations have varied c onceptions of partnerships. Furthermore, the results sugge st a parabolic relationship between numbers of partners mentioned and institutional sustainability, pe rhaps reflecting reluctance on the part of certain organizations to partner within Nansan a out of fear of competition for resources. These and other results are discussed in greater detail in Chapter VIII. Differences Based on Numbers of In Degrees For the final set of analyses in this ch apter, I look at all organizations in the sample and assign them to groups based on thei r numbers of in degrees, or the number of times each organization was mentioned as a partner by other organizations in the sample. The mean number of in degrees is 2.84, with a median of 1.00 and a standard deviation of 5.27 (considerably greater than the standard de viation of 2.13 related to numbers of out degrees). Because of the greate r variation between numbers of in degrees, in this section,

PAGE 88

77 organizations are divided into four groups. The first group of organizations was never mentioned as a partner by other organizations in the sample (n=13). The second group was mentioned by 1 or 2 organizations (n=11) The third group was mentioned by 3 to 7 organizations as a partner (n=5), and the f ourth group consists of 3 organizations that were mentioned as partners 13, 18, and 22 tim es respectively. Ag ain, unless otherwise stated, a one-way ANOVA was used to test differences in mean scores between the groups. An overview of the results in th is section is presen ted in Table VI.18.

PAGE 89

78 Table VI.18. Comparisons based on numbers of in degrees. Institutional Sustainability Measures Group 1 (no in degrees) Group 2 (1-2 in degrees) Group 3 (3-7 in degrees) Group 4 (13-22 in degrees Services and Programs years of operation 5.77 10.09 7.00 15.33 # of programs† 4.77 5.91 6.80 4.00 People # of beneficiaries* 235.54 473.82 180.20 4,398.00 # of employees 10.15 8.91 5.00 20.33 # of volunteers 7.00 7.18 12.80 7.50 Facilities # of facilities 1.42 2.18 2.60 2.00 ownership rate† 15.38% 27.27% 60.00% 100.00% # of offices outside Nansana 3.00 0.73 1.20 3.67 Technology # of tech items† 4.54 4.91 6.60 7.67 # of computers† 2.88 2.60 3.60 15.33 Finances actual budget $19,684.70 $41,667.10 $31,645.00 $118,639.33 actual: expected 37.81% 57.07% 65.64% 100.00% Records points out of 5 4.62 4.55 5.00 4.67 Program Evaluations rate conducting 84.62% 72.73% 100.00% 100.00% Human Resource Systems points out of 5 3.85 4.33 4.75 4.67 Differences are significant at p < .05 † Differences are significant at p < .1

PAGE 90

79 Services and Programs Beginning with measures of institutional sustainability related to organizationsÂ’ services and programs, the two measures discussed here are the number of years organizations have existed and the number of programs each organization conducts. Organizations in Group 1 (those with no in degrees) have been in existence for an average of 5.77 years. Organizations in Group 2 (those with 1 or 2 in degrees) have been in existence for an average of 10.09 years. Or ganizations in Group 3 (those with 3 to 7 in degrees) have been in existence for an aver age of 7 years, and or ganizations in Group 4 (those with 13 to 22 in degrees) have been in existence for an aver age of 15.33 years. These results are not significan t, as shown in Table VI.19. Table VI.19. Comparison of years of opera tion based on numbers of in degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 13 5.77 5.600 2.39 9.15 Group 2 11 10.09 9.513 3.70 16.48 Group 3 5 7.00 4.062 1.96 12.04 Group 4 3 15.33 6.658 1.21 31.87 F = 1.790, p = .172, Eta = .161 With respect to numbers of programs, organizations in Group 1 have an average of 4.77 programs. Organizations in Group 2 have an average of 5.91 programs. Organizations in Group 3 have an average of 6.80 programs, and organizations in Group 4 have an average of 4 programs. These re sults are arguably signi ficant at p < .1 (p = .094), as shown in Table VI.20. Thus, in terestingly, the number of programs per organization tends to increase with the number of in degrees for the first three groups, but

PAGE 91

80 then it decreases among those organizations with the most in degrees, perhaps reflecting a recognition among organizations with the most in degrees of the need to focus attention on a more limited number of programs. Als o, many organizations wi th greater resources tend to have contracts or grants for thei r programming, meaning that their only programs are those that are specifically funded. Table VI.20. Comparison of programs based on numbers of in degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 13 4.77 1.787 3.69 5.85 Group 2 11 5.91 1.578 4.85 6.97 Group 3 5 6.80 2.683 3.47 10.13 Group 4 3 4.00 1.000 1.52 6.48 F = 2.353, p = .094, Eta = .201 People Looking next at measures of institutiona l sustainability rela ted to the various people associated with organizations, orga nizations in Group 1 serve an average of 235.54 beneficiaries. Organizati ons in Group 2 serve an average of 473.82 beneficiaries. Organizations in Group 3 serve an average of 180.20 beneficiaries, and organizations in Group 4 serve an average of 4398.00 beneficiaries. These results are significant (p = .002), as shown in Table VI.21. This suggest s that organizations with the most in degrees serve considerably more beneficiarie s than those with le ss in degrees, perhaps reflecting that organizations with greater in degrees may be seen as partners because of their increased capacity to serve clients and their ability to receive clients from other organizations.

PAGE 92

81 Table VI.21. Comparison of beneficiarie s based on numbers of in degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 13 235.54 262.570 76.87 394.21 Group 2 11 473.82 626.788 52.74 894.90 Group 3 5 180.20 236.622 113.61 474.01 Group 4 3 4,398.00 5,700.431 9,762.66 18,558.66 F = 6.109, p = .002, Eta = .395 The results were not significant with respect to differences in numbers of employees and numbers of volunteers, as shown in Tables VI.22 and VI.23. Organizations in Group 1 have an average of 10.15 employees. Organizations in Group 2 have an average of 8.91 employees. Organi zations in Group 3 have an average of 5 employees, and organizations in Group 4 have an average of 20.33 employees. Organizations in Group 1 have an average of 7 volunteers. Orga nizations in Group 2 have an average of 7.18 volunteers. Organiza tions in Group 3 have an average of 12.80 volunteers, and organizations in Group 4 have an average of 7.50 volunteers. Table VI.22. Comparison of employees based on numbers of in degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 13 10.15 15.383 .86 19.45 Group 2 11 8.91 21.012 5.21 23.02 Group 3 5 5.00 4.690 .82 10.82 Group 4 3 20.33 31.817 58.70 99.37 F = .457, p = .715, Eta = .047

PAGE 93

82 Table VI.23. Comparison of volunteers based on numbers of in degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 12 7.00 8.464 1.62 12.38 Group 2 11 7.18 4.813 3.95 10.42 Group 3 5 12.80 17.978 9.52 35.12 Group 4 2 7.50 .707 1.15 13.85 F = 5.06, p = .682, Eta = .055 Facilities Looking at various measures of physical space as a component of institutional sustainability, with respect to the number of facilities each organization has on its premises in Nansana, organizations in Gr oup 1 have an average of 1.42 facilities. Organizations in Group 2 have an average of 2.18 facilities. Or ganizations in Group 3 have an average of 2.60 facilities, and organi zations in Group 4 have an average of 2.00 facilities. These results are not significant, as shown in Table VI.24. Table VI.24. Comparison of Nansana facili ties based on numbers of in degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 12 1.42 .900 .84 1.99 Group 2 11 2.18 1.250 1.34 3.02 Group 3 5 2.60 .894 1.49 3.71 Group 4 3 2.00 1.000 .48 4.48 F = 1.851, p = .162, Eta = .171 Next, in Group 1, 9 of 13 organizations rent their Nansana f acilities; 2 own their facilities; and 2 have no facili ties. In Group 2, 8 of 11 organi zations rent their facilities,

PAGE 94

83 and 3 own their facilities. In Group 3, of the five organizatio ns, 2 rent thei r facilities and 3 own their facilities. In Group 4, all three of the organi zations own their facilities. Chisquare tests were performed on these measures and the results are ar guably significant at p < .1 (p = .067), as shown in Table VI.25. Table VI.25. Comparison of ownership ra tes based on numbers of in degrees. Rent Own No office Row N Group 1 69.2% 15.4% 15.4% 13 Group 2 72.7% 27.3% 0.0% 11 Group 3 40.0% 60.0% 0.0% 5 Group 4 0.0% 100.0% 0.0% 3 Chi2 = 11.802, d.f. = 6, p = .067 With respect to locations outside of Nans ana (either in Uganda or internationally), organizations in Group 1 have an average of 3.00 locations. Organizations in Group 2 have an average of 0.73 locations. Organizations in Group 3 have an average of 1.20 locations, and organizations in Group 4 ha ve an average of 3.67 locations. These differences are not significan t, as shown in Table VI.26. Table VI.26. Comparison of offices outs ide Nansana based on numbers of in degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 13 3.00 4.916 .03 5.97 Group 2 11 .73 1.421 .23 1.68 Group 3 5 1.20 1.643 .84 3.24 Group 4 3 3.67 4.041 6.37 13.71 F = 1.118, p = .359, Eta = .107

PAGE 95

84 Technology Differences in the organizationsÂ’ t echnological capabil ities are arguably significant. Organizations in Group 1 have an average of 4.54 (out of 10) technology systems. Organizations in Group 2 have an average of 4.91 technology systems. Organizations in Group 3 have an average of 6.60 technology systems, and organizations in Group 4 have an average of 7.67 technology sy stems. These results are significant at p < .1 (p = .096), as shown in Table VI.27. Table VI.27. Comparison of technology syst ems based on numbers of in degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 13 4.54 2.402 3.09 5.99 Group 2 11 4.91 2.256 3.39 6.42 Group 3 5 6.60 2.074 4.03 9.17 Group 4 3 7.67 .577 6.23 9.10 F = 2.331, p = .096, Eta = .200 Also, organizations in Group 1 have an average of 2.88 computers or laptops. Organizations in Group 2 have an average of 2.60 computers or laptops. Organizations in Group 3 have an average of 3.60 computers or laptops. Organizations in Group 4 have an average of 15.33 computers or laptops. Thes e results are significant at p < .1 (p = .055), as shown in Table VI.28.

PAGE 96

85 Table VI.28. Comparison of computers based on numbers of out degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 8 2.88 1.458 1.66 4.09 Group 2 5 2.60 1.517 .72 4.48 Group 3 5 3.60 3.578 .84 8.04 Group 4 3 15.33 18.009 29.40 60.07 F = 3.083, p = .055, Eta = .352 Finances Again, the two measures used in this study to assess financial components of institutional sustainability are organizationsÂ’ actual annual budgets a nd the percentage of the expected budget that the actual budget repr esents. Organizations in Group 1 have an average actual annual budget of $19,684.70. Orga nizations in Group 2 have an average budget of $41,667.10. Organizations in Group 3 have an average budget of $31,645.00, and organizations in Group 4 have an av erage budget of $118,639.33. These results are not significant, as shown in Table VI.29. Table VI.29. Comparison of actual budget based on numbers of in degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 10 19,684.70 46,164.650 13,339.50 52,708.90 Group 2 10 41,667.10 91,429.504 23,737.63 107,071.83 Group 3 5 31,645.00 37,612.201 15,056.72 78,346.72 Group 4 3 118,639.33 155,408.339 267,416.38 504,695.05 F = 1.243, p = .316, Eta = .134

PAGE 97

86 With respect to the actual budget to e xpected budget ratio, organizations in Group 1 have an average percent of 37.81%. Organiza tions in Group 2 have an average percent of 57.07%. Organizations in Group 3 ha ve an average percent of 65.64%, and organizations in Group 4 have an average of 100%. These results are not significant (p = .108), as shown in Table VI.30. Table VI.30. Comparison of budget ratio based on numbers of in degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 10 .3781 .32815 .1434 .6129 Group 2 9 .5707 .34562 .3051 .8364 Group 3 5 .6564 .36260 .2062 1.1066 Group 4 2 1.0000 .00000 1.0000 1.0000 F = 2.276, p = .108, Eta = .237 Records The results are not significant with resp ect to differences in record keeping between the four groups, as shown in Table VI .31. Out of a score of 5, organizations in Group 1 scored an average of 4.62. Organizatio ns in Group 2 scored an average of 4.55. Organizations in Group 3 scored an average of 5.00, and organizati ons in Group 4 scored an average of 4.67.

PAGE 98

87 Table VI.31. Comparison of record keep ing based on numbers of in degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 13 4.62 .650 4.22 5.01 Group 2 11 4.55 1.214 3.73 5.36 Group 3 5 5.00 .000 5.00 5.00 Group 4 3 4.67 .577 3.23 6.10 F = .341, p = .796, Eta = .035 Program Evaluations Of the 13 organizations in Group 1, 11 conduc t some sort of evaluations of their programs. Of the 11 organizations in Group 2, 8 conduct evaluations of their programs. In Groups 3 and 4, all of the organizations c onduct evaluations of their programs. These results are not significant according to chisquare tests, as shown in Table VI.32. Table VI.32. Comparison of evaluation s based on numbers of in degrees. Yes No Row N Group 1 84.6% 15.4% 13 Group 2 72.7% 27.3% 11 Group 3 100.0% 0.0% 5 Group 4 100.0% 0.0% 3 Chi2 = 2.614, d.f. = 3, p = .455 Human Resource Systems Finally, the results with respect to human resource systems were also not significant, as shown in Tabl e VI.33. Out of a score of 5, organizations in Group 1 received an average score of 3.85. Organizations in Group 2 r eceived an average score of

PAGE 99

88 4.33. Organizations in Group 3 received an average score of 4.75, and organizations in Group 4 received an average score of 4.67. Table VI.33. Comparison of human resource systems based on numbers of in degrees. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 13 3.85 1.519 2.93 4.76 Group 2 9 4.33 .707 3.79 4.88 Group 3 4 4.75 .500 3.95 5.55 Group 4 3 4.67 .577 3.23 6.10 F = .919, p = .446, Eta = .099 Significant Findings Regarding In Degrees Thus, with respect to groupings based on nu mber of in degrees, or the number of times each organization was mentioned as a pa rtner by other organizations, results were significant, at least at p < 0.1 in five categories: numb er of programs, number of beneficiaries, property ownership rate in Nansana, number of technology items, and number of computers. These results offer some support for the hypothesis that organizations with higher levels of connectivity also have higher levels of institutional sustainability. While the above findings do suggest a relationship between institutional sustainability and interorganiza tional partnerships, the lack of significant findings across many categories begs the question of whether other factors might be affecting the results. As I interviewed organizations in Nans ana, I was struck by the importance of international ties. Organizations with inte rnational ties play significant roles in the community, and organizations without intern ational ties actively seek international

PAGE 100

89 connections. As such, the following ch apter goes beyond testing my hypothesis and looks at the role of internati onal ties among NGOs in Nansana.

PAGE 101

90 CHAPTER VII INTERNATIONAL TIES RESULTS As suggested by the network maps in Chapte r V, it seems that, at least in Nansana, many organizations that play a central role in the network (both formal and informal) have international ties. To be gin to explore this issue furt her and to attempt to understand the extent to which organizations are succe ssfully moving toward local self-reliance and sustainable systems development, I looked at the relationship between organizationsÂ’ institutional sustainability and th eir international ties. It se ems logical that organizations with stronger international ties will be more sustainable according to the measures in this study, but, given declining intern ational aid, they may be less sustainable in the longer run. To look at the relationships between international ties and institutional sustainability, I divided the organizations into three groups. Organi zations in the first group have no direct internationa l ties. Organizations in th e second group have at least one international volunteer, at least one intern ational board member, or at least some of their funds coming directly from an interna tional source. Finally, organizations in the third group have an internationa l office, international incorp oration, or an international board of directors. Organiza tions that qualified for both th e second and third groups were placed in the third group. In this sample, 14 organizations have no international ties; 13 organizations qualify for Group 2; and five qualify for Group 3. Table VII.1 provides an overview of the differences between the groups for each of the eight areas of institutional sustainability discussed in this thesis.

PAGE 102

91 Table VII.1. Comparisons based on strength of international ties. Institutional Sustainability Measures Group 1 (no direct int'l ties) Group 2 (limited int'l ties) Group 3 (strong int’l ties) Services and Programs years of operation 5.21 10.69 11.00 # of programs 5.50 5.15 5.80 People # of beneficiaries† 221.93 574.85 2428.00 # of employees* 2.50 12.23 24.40 # of volunteers 7.14 7.25 13.75 Facilities # of facilities 1.57 2.17 2.40 ownership rate 21.43% 38.46% 60.00% # of offices outside Nansana* 0.29 2.69 5.00 Technology # of tech items* 3.50 6.38 7.40 # of computers 1.71 3.89 10.60 Finances actual budget* $5,674.67 $44,973.83 $129,969.75 actual: expected* 31.20% 70.58% 76.62% Records points out of 5 4.50 4.69 5.00 Program Evaluations rate conducting 71.40% 92.30% 100.00% Human Resource Systems points out of 5* 3.50 4.67 4.80 Differences are significant at p < .05 † Differences are significant at p < .1

PAGE 103

92 Services and Programs First, looking at the length of time that organizations have been in existence, organizations in Group 1 have been in exis tence for an average of 5.21 years. Group 2 has been in existence for an average of 10.69 years, and Group 3 has been in existence for an average of 11 years. These results however, are not signi ficant (p = .107), as shown in Table VII.2. Table VII.2. Comparison of years of operation based on international ties. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 14 5.21 4.543 2.59 7.84 Group 2 13 10.69 9.187 5.14 16.24 Group 3 5 11.00 6.819 2.53 19.47 F = 2.412, p = .107, Eta = .143 Next, with respect to numbers of program s, there is little difference between the groups, as shown in Table VII.3. The averag e number of programs among those with no international ties is 5.50. Th e average number among those with some ties is 5.15, and the average among those with st rong international ties is 5.80. This might reflect, at least in part, uncertainty regarding whether few or many programs is more desirable. While it seems that organizations with access to more resources may wish to implement more programs to help their beneficiaries, as mentioned above, it is also important for organizations to keep their resources focuse d on particular programs and goals and not become too thinly spread.

PAGE 104

93 Table VII.3. Comparison of progr ams based on international ties. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 14 5.50 1.829 4.44 6.56 Group 2 13 5.15 1.676 4.14 6.17 Group 3 5 5.80 3.114 1.93 9.67 F = .216, p = .807, Eta = .015 People In the people categories, more significant results were found, at least with respect to beneficiaries and employees. Organizations with no international ties serve an average of 221.93 individuals. Organizatio ns with some international ties serve an average of 574.85 individuals, and organizatio ns with strong internationa l ties serve an average of 2,428 people. These results are arguably signif icant at p < .1 (p = .081). With regard to employees, organizations with no internationa l ties employ an average of 2.50 people. Organizations with some international ti es employ an average of 12.23 people, and organizations strong ties employ 24.40 people on average. These results are si gnificant at p < .05 (p = .045). Results in these two cate gories are shown below in Tables VII.4 and VII.5. Table VII.4. Comparison of benefici aries based on international ties. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 14 221.93 289.124 54.99 388.86 Group 2 13 547.85 732.115 105.43 990.26 Group 3 5 2,428.00 4,737.199 3,454.01 8,310.01 F = 2.747, p = .081, Eta = .159

PAGE 105

94 Table VII.5. Comparison of employ ees based on international ties. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 14 2.50 2.312 1.16 3.84 Group 2 13 12.23 21.218 .59 25.05 Group 3 5 24.40 24.825 6.42 55.22 F = 3.450, p = .045, Eta = .192 Results were not significant with respect to numbers of volunteers, as shown in Table VII.6. This is not su rprising, given the fact that many organizations with limited resources are completely dependent on their vol unteers, while those with more resources may be able to attract more local and in ternational volunteers based on a perceived desirability to work for them. Organizations with no international tie s have an average of 7.14 volunteers; those with limited ties have 7.25 volunteers; and those with strong ties have 13.75 volunteers on average. Table VII.6. Comparison of volunt eers based on international ties. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 14 7.14 7.574 2.77 11.52 Group 2 12 7.25 5.294 3.89 10.61 Group 3 4 13.75 20.370 18.66 46.16 F = .876, p = .428, Eta = .061 Facilities With respect to physical space, results were not significant with respect to the number of facilities an orga nization has on its premises, as shown in Table VII.7. Again, however, the number of facilities may not reflect differenc es in resources, but rather

PAGE 106

95 differences in objectives. Furthermore, organi zations were only asked about facilities at their Nansana site. Organizations with no international ties ha ve an average of 1.57 facilities at their Nansana si te. Organizations with some international ties have an average of 2.17 facilities at their Nansana site and organizations with strong ties have an average of 2.40 facilities at their Nansana site. Table VII.7. Comparison of Nansana fa cilities based on international ties. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 14 1.57 .938 1.03 2.11 Group 2 12 2.17 1.193 1.41 2.92 Group 3 5 2.40 1.140 .98 3.82 F = 1.550, p = .230, Eta = .100 While there appears to be some di fference between the rates at which organizations rent or own their offices in Nansana, these results were not significant according to chi-square tests, as shown in Table VII.8. Among those organizations with no international ties, nine of the 14 rent th eir office space; three own their office space; and two have no office space (as noted above, however, one of these organizations was between offices at the time of the intervie w). Among those organizations with some international ties, eight of 13 rent their o ffice space, and five own their office space. With respect to those organizations with st rong international ties, two rent their office space, and three own their office space.

PAGE 107

96 Table VII.8. Comparison of ownershi p rates based on international ties. Rent Own No office Row N Group 1 64.3% 21.4% 14.3% 14 Group 2 61.5% 38.5% 0.0% 13 Group 3 40.0% 60.0% 0.0% 5 Chi2 = 4.656, d.f. = 4, p = .324 Significant differences were found between the three groups with respect to the number of offices outside of Na nsana the organizations have. To calculate this figure, the number of offices both in Uganda and intern ationally were totaled. Organizations with no international ties have an average of 0.29 offices outside of Nansana, while organizations with some international tie s have an average of 2.69 offices outside Nansana, and organizations with strong international ties have an average of 5 offices outside Nansana. These results are signifi cant at p < .05 (p = .021), as shown in Table VII.9, though some of the difference betw een the groups can be explained by the definition of the categories themselves in th at organizations in th e third group typically have at least one o ffice outside Nansana. Table VII.9. Comparison of offices outs ide Nansana based on international ties. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 14 .29 1.069 .33 .90 Group 2 13 2.69 3.199 .76 4.63 Group 3 5 5.00 6.442 3.00 13.00 F = 4.411, p = .021, Eta = .233

PAGE 108

97 Technology Two measures of technology were used to compare the sustainability of the two groups. First, as described above, organizations received a score out of 10 for each type of technology it has among a lim ited set of options. Organiza tions with no international ties have an average of 3.50 of these items; organizations with some international ties have an average of 6.38 of these items; and organizations with strong ties have 7.40 of these items. These results are significant at p < .05 (p = .00), as s hown in Table VII.10. Table VII.10. Comparison of technology systems based on international ties. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 14 3.50 1.829 2.44 4.56 Group 2 13 6.38 1.758 5.32 7.45 Group 3 5 7.40 1.517 5.52 9.28 F = 13.356, p = .000, Eta = .479 Results regarding differences in the nu mber of computers or laptops between groups are not significant (p = .11), as shown in Table VII.11, with organizations with no international ties having 1.71 computers, orga nizations with some international ties having 3.89 computers, and organizations with strong ties having 10.60 computers. Table VII.11. Comparison of computers based on international ties. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 7 1.71 .756 1.02 2.41 Group 2 9 3.89 2.315 2.11 5.67 Group 3 5 10.60 14.398 7.28 28.48 F = 2.491, p = .111, Eta = .217

PAGE 109

98 Finances Financial differences between the thr ee groups are significant. Two measures were used to compare the finances of organi zations in the three gr oups: 1) their actual annual budgets, and 2) the percentage that or ganizationsÂ’ actual budgets represent of their anticipated budgets. Organizati ons with no international ties have an average actual budget of $5,674.67. Organizations with some in ternational ties have an average actual budget of $44,973.83, and organizations with strong ties have an average actual budget of $129,969.75. These results are significant at p < 05 (p = .019), as shown in Table VII.12. Table VII.12. Comparison of actual budget based on international ties. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 12 5,674.67 5,799.660 1,989.74 9,359.59 Group 2 12 44,973.83 83,171.929 7,871.09 97,818.76 Group 3 4 129,969.75 126,696.815 71,633.16 331,572.66 F = 4.695, p = .019, Eta = .273 Furthermore, the average percentage reflecting organizati onsÂ’ actual budgets over their planned (or ideal) budgets is 31.20% fo r organizations with no international ties, 70.58% for organizations with some internati onal ties, and 76.62% for organizations with strong ties. These results are significant at p < .05 (p = .010) as shown in Table VII.13.

PAGE 110

99 Table VII.13. Comparison of budget ratio based on international ties. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 11 .3120 .20248 .1760 .4480 Group 2 12 .7058 .35388 .4809 .9306 Group 3 3 .7662 .42999 .3020 1.8343 F = 5.618, p = .010, Eta = .328 Records Results with respect to record keepi ng are not significant, as shown in Table VII.14. Out of a score of 5, organizations with no international ties scored an average of 4.5 for record keeping. Organizations with some international ties scored an average of 4.69, and organizations with str ong international ties scored an average of 5 out of 5. Table VII.14. Comparison of record keeping based on international ties. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 14 4.50 1.160 3.83 5.17 Group 2 13 4.69 .480 4.40 4.98 Group 3 5 5.00 .000 5.00 5.00 F = .679, p = .515, Eta = .045 It is worth noting that there may actua lly be more variation between the groups than these results indicate, however, because the formality and types of record keeping that occur vary among organizations. For exam ple, all organizations report maintaining client records, but some keep these records on a memory stick, while others keep them on a server.

PAGE 111

100 Program Evaluations For this category, questions regarding ev aluations were collapsed into a single binomial measure reflecting whether or not or ganizations reported performing some type of evaluation of their programs. Among thos e organizations reporting no international ties, 71.4% of organizations reported conducting some sort of evaluations. Those with some international ties reported conducting ev aluations at a rate of 92.3%, and 100% of those with strong internationa l ties reported conducting eval uations. According to chisquare tests, these results are not si gnificant, as shown in Table VII.15. Table VII.15. Comparison of evaluations based on international ties. Yes No Row N Group 1 71.4% 28.6% 14 Group 2 92.3% 7.7% 13 Group 3 100.0% 0.0% 5 Chi2 = 3.326, d.f. = 2, p = .190 Human Resource Systems To measure the human resource systems of the organizations in the sample, organizations received a point for each of th e following: conducting orientations for new hires, maintaining employment files, di stributing employee manuals, having written policies regarding hiring and terminating, and providing employees access to ongoing training. Organizations with no international ties scored an average of 3.5 out of 5 on this measure. Organizations with some intern ational ties scored an average of 4.67, and organizations with strong international ties scored an average of 4.8. These results are significant at p < .05 (p = .014), as shown in Table VII.16.

PAGE 112

101 Table VII.16. Comparison of human resource systems based on international ties. N Mean Standard Deviation 95% Confidence Interval for Mean Lower Bound Upper Bound Group 1 12 3.50 1.382 2.62 4.38 Group 2 12 4.67 .651 4.25 5.08 Group 3 5 4.80 .447 4.24 5.36 F = 5.055, p = .014, Eta = .280 Significant Findings Regardi ng International Ties The findings in this chapter show that comparing organizations based on the strength of their internati onal ties produces significant re sults in seven categories of institutional sustainability: number of bene ficiaries, number of employees, number of offices outside Nansana, number of technol ogy items, actual budget, actual to expected budget ratio, and number of human resource system s. Thus, more significant results were found when comparing groups based on their international ties than when comparing them based on number of partners, perhap s suggesting that partnerships may be a function of resources more so than the ot her way around. The results are described in greater detail in the next chapter.

PAGE 113

102 CHAPTER VIII DISCUSSION, LIMITATIONS, CONCLUSION As mentioned, the primary hypothesis of th is thesis is that better connected organizations within a network will have higher levels of institutional sustainability, and to test this hypothesis, I used three different methods of comp aring organizations. First, I compared organizations that are members of the NCSN with organi zations that are not members of the formal network. Second, I a ssigned organizations to three groups based on their number of out degrees, or the number of organizations with whom they report having partnerships. Third, I assigned organi zations to four groups (because of the higher standard deviation) based on their num ber of in degrees, or the number of organizations that report havi ng partnerships with them. Detailed results regarding the differences between groups based on various measures of institutional sustainability are provided in Chapter VI, and here I desc ribe the significance of these results. Institutional Sustainability and NCSN Membership Beginning with the NCSN member vers us nonmember groupings, the results in this category are not significant; however, the lack of significance may reflect the fact that the nonmember group consis ts of only four organizations Furthermore, there may be opposing forces at play here in that some nonmember or ganizations may be struggling because they lack capacity and would bene fit from becoming members of the NCSN. Other organizations, though, may not be members of the NCSN because they have access to significant resources based on their internat ional ties or otherwise and do not believe they would benefit from the NCSN.

PAGE 114

103 Despite the lack of significance, it does s eem that there were differences between the groups. For example, NCSN members have been operating longer, serve more beneficiaries, have higher rates of propert y ownership, and have higher actual annual budgets than nonmembers. Also, more qualita tively, based on both my observations and the reports of member organizations, the NCSN provides numerous benefits to its member organizations by helping them conn ect with one another, by promoting better management systems, and through an array of additional capacity bui lding activities. Institutional Sustainability and Out Degrees Next, I compared organizations based on their number of out degrees, or the number of organizations they mentioned as partners. As described above, the organizations were divided into three groups with the first group having no or one ties, the second group having two or three ties, and the third group having four or more ties. Across all measures of institutional sustai nability, there were only two significant differences in the measures between these th ree groups. Specifically, the groups differed in the number of facilities at their Nansana site and their rates of ownership of their facilities in Nansana. While it is not clear why the groups differed on these two particular measures, the fact that they di d differ significantly on two measures suggests that there may be some differences in institutional sustainability between them. There also loosely appears to be a para bolic relationship between organizationsÂ’ out degrees and their institutional sustainabil ity. That is, organizations in Group 1 have 1.89 facilities in Nansana, and 33.33% own thei r facilities. In Group 2, organizations have only 1.42 facilities in Nansana, and only 7.69% own their facilities. In Group 3, the numbers go back up, with organizations having an average of 2.60 facilities in Nansana

PAGE 115

104 and 70% owning their facilities. Though other results were not significant, this pattern seems to repeat itself elsewhere, for example, with numbers of beneficiaries, numbers of employees, and actual budgets. This result might be somewhat explained by NGOsÂ’ hesitancy to partner with one another. Historically, NGOs were reluctant to partner with each other because they viewed other NGOs as competitors for scarce resources (Gellert, 1996). As such, perhaps organizations with very limited resources ar e motivated to partner with others because they will not survive without the partnerships while those with slightly more resources are less inclined to partner for fear of competition. Those with substantial resources may be secure enough in their instit utional sustainability that they are willing to partner and help others. Thus, practically speaking, ne tworks, including the NCSN, should continue to promote the benefits of network membersh ip as appropriate, and additional research should continue to explore the benefits of network memberships and, particularly, the potential for positive synergistic effects, as described by Paarlberg and Varda (2009). While the results here are somewhat informative, one issue with the out degree measure is that the definition of partners hip was not as objective as it was intended. Instead, organizations either each had their ow n conceptions of partne rships, or there was a similar conception of partne rship across organizations but which was more subjective in nature. In other words, many organizations may have similarly conceptualized partners as larger organizations who helped them build capacity. Thus, despite organizations having a similar definition of partnership, th ere was little reciprocity regarding which organizations each named as partners. While this finding certainly tells us something about how organizations relate to each ot her and choose their partnerships, a more

PAGE 116

105 objective definition would be useful for futu re empirical studies because it would allow for simpler comparisons across organizations. It should also be noted that out degrees he re only refer to out degrees in Nansana. Many organizations described pa rtnerships with organizations in Kampala, elsewhere in Uganda, and even internationally. Thus, it may be that organizations with few out degrees in Nansana are actually well connected elsewhere, meaning that their institutional sustainability is in fact more (o r less) affected by their partners hips than the results in this section suggest. Institutional Sustainability and In Degrees More significant differences in the meas ures of institutional sustainability were found when organizations were grouped base d on their number of in degrees, which, again, refers to the number of times each organization was mentioned as a partner by another organization. At least arguably signi ficant results were found in five areas: numbers of programs, numbers of beneficiar ies, property ownership rates in Nansana, numbers of technology systems, and numbers of computers. Results regarding differences in numbers of beneficiaries were the most significant at p = .002. Interestingly, though, wh ile those in Group 4 (organizations with the most in degrees) serve considerably more beneficiaries than the other groups, organizations in Group 2 serve more beneficiar ies than organizations in Groups 1 and 3. Thus, the primary conclusion to be drawn here is that the organizat ions in Group 4 serve considerably more beneficiaries than those in the other groups. Th is result seems logical given the high rate of beneficiary shar ing among partners. Specifically, 45 of 91 partnerships reportedly involve beneficiary sharing, which ty pically takes the form of

PAGE 117

106 organizations referring beneficiaries to each other for needed services. Also, a major emphasis of the NCSN is on its client referral system. As such, it is not surprising that those with more in degrees have more bene ficiaries because, among other factors, they are likely receiving more client referrals. Furthermore, the huge jump in beneficiaries from Group 3 to Group 4 and the more modest difference in numbers between Groups 1, 2, and 3 may reflect the differen ces in numbers of in degrees (organizations in Groups 1, 2, and 3 have between 0 and 7 in degrees, a nd organizations in Gr oup 4 have between 13 and 22 in degrees). Thus, the pattern in benefi ciaries reflects the pa ttern in in degrees. A secondary conclusion here is that there is not a linear pattern in number of beneficiaries with respect to Groups 1, 2, and 3. While, again, this may reflect the fact that differences in numbers of in degrees are small between these groups, it may also suggest that other factors may cau se organizations to either in crease their partners or their beneficiaries. For example, perhaps organi zations in Group 2 have more beneficiaries than Group 1 because they have more capacit y, but, as organizations increase their partners, they may be more in clined to refer more benefici aries, thus causing the number of beneficiaries to decrease between Groups 2 and 3. Another interesting finding is the vari ation between groups with respect to numbers of programs, which is significant at p = .094. Numbers of programs positively correlate with numbers of in degrees in Groups 1, 2, and 3, but Group 4 has less programs than any of the other three groups, suggesting that the organizations with the most in degrees are specializing more than those in the other groups. Thus, as mentioned previously, while having more programs ma y be somewhat desirable, having limited

PAGE 118

107 programs may allow organizati ons to truly specia lize and become more attractive as potential partners. Findings regarding programming might also suggest that organizations are naturally organizing themselves in such a way that those specia lizing assume more central roles in the network, while those with more programs are at th e periphery. It is likely that those organizations with fewe r programs are only conducting programs that are specifically funded through grants or c ontracts, while organizations with more programs are designing programs based on comm unity needs. If this is the case, partnerships between these two types of or ganizations may be essential because they reflect a union between organizations with resources that are connected to funding sources and organizations that are more deeply in touch with community needs. Finally, patterns regarding property ownership rates, numbers of technology systems, and numbers of computers are consis tent with the hypothesis in that each of these numbers positively correlates with numbers of in degrees (the only exception being that organizations in Group 2 have more co mputers than organiza tions in Group 1). Though these results are not consistent across al l categories of institutional sustainability, this does suggest that there may be a positive relationship between institutional sustainability and numbers of partners. Specifically, organiza tions with greater institutional capacity, particularly in the form of property and technology, also tend to have greater numbers of partners. Thus, patterns with respect to in degrees are generally clearer than patterns with respect to out degrees. Again, it seems that numbers of out degrees may be a function of organizationsÂ’ understandings of the concep t of partnership, a nd, in many instances,

PAGE 119

108 organizations with greater resources may ha ve named more organizations outside of Nansana as partners because they look up and out for strategic part ners. In degrees, however, may offer a better indi cation of which organizations in Nansana are perceived as having greater resources or are better positio ned to act as partners or help build the capacity of other organizations. While fu ture studies may work on better defining partnership, the variation in de finition found here may actually offer considerable insight into how partnerships are conceptualized and formed. Looking back at the hypothesis, this set of results offers the best indication of the areas of institutional sustainability most rela ted to levels of networking and partnering. Specifically, first, there is a correlation be tween numbers of part ners and numbers of beneficiaries, most pronounced among organizations with the most partners and the most beneficiaries. This finding may suggest that increased partnerships enhance organizationsÂ’ capacities and allow them to serv e more beneficiaries. It may also suggest that the NCSN is effectuating an effici ent method of service provision in that beneficiaries are being passed from organi zations with limited partners and possibly capacities to those with great er numbers of partners a nd possibly grea ter capacities. Second, organizations with more programs ma y be more desirable as partners to a degree, but those specializing in a limited nu mber of programs may actually have the most partners. Again, this would suggest th at the most beneficial partners are those organizations effectively carrying out or sp ecializing in smaller numbers of programs. Third, both in this set of comparisons and in comparisons based on numbers of out degrees, there is a correlation between numbers of partners and property ownership rates in Nansana. This might suggest that those organizations with greater resources or

PAGE 120

109 those with a more stable presence in Nans ana are more desirable partners, or it could mean that, as organizations increase their capacity, an important way that they increase their sustainability is through property owne rship. Similarly, measures of technology systems and numbers of computers positively correlate with numbers of partners, suggesting that these are impor tant components of institutional sustainability. Future studies may wish to explore the direction of causality between these measures. Institutional Sustainability and International Ties While this thesis does not purport to de scribe a causal relationship between partnerships and institu tional sustainability, to the extent there is a causal relationship, when I initially designed this study, though I was not testing the causal relationship, I thought that, if anything, measures of in stitutional sustainability would depend on numbers or strengths of partne rships. It seems logical that greater numbers of partners would mean greater resources, both materially and informationally (leading to better systems related to record k eeping, human resources, technology, and so forth). My data collection and analysis, however, led me to believe that, to the extent there is a causal relationship here, organizations might in f act be better connected because they have access to greater resources outside Nansana. If this is the case, the question then becomes why these organizations have better access to resources in the first place. While causal relationships are beyond th e scope of this thesis, I did some preliminary exploration of these questions by looking at the relationship between organizationsÂ’ international ties and their institutional sustainability. In looking at partnerships among organizations in Nans ana and throughout my conversations with those working in the civil soci ety sector in Nansana, it seem ed that those organizations

PAGE 121

110 playing a central role in Nansana tend to ha ve significant international ties, and a major focus of many smaller organizations is on bol stering their international ties. This suggests that organizationsÂ’ levels of conn ectivity depend on their existing institutional sustainability, which is, in turn, a func tion of their international ties. As described above, to test the relati onship between international ties and institutional sustainability, the organizations were divided into three groups based on the strength of their inte rnational ties. A larger number of significant results was found in comparing the means between these thr ee groups than was found when comparing organizations based on in degree and out degr ee groupings. In the groupings based on international ties, there were significant diffe rences in numbers of beneficiaries, numbers of employees, numbers of offices outside Nansana, numbers of technology systems, actual annual budgets, expected to actual budget ratios, and nu mbers of human resource systems. Except with respect to differences in numbers of beneficiaries, all of these results were significant at p < .05. For each of the seven categories where the between group differences are significant, there is a positive correlation be tween strength of international ties and institutional sustainability. This pattern also seems to repeat itself across other categories where the results are not signi ficant. This confirms my general sense, along with anecdotal evidence from interviews, regarding the importance of international ties to an organizationÂ’s institutional sustainability. On one hand, these results may prove that the system of international aid is working because the findings offer evidence that international aid, channeled through few orga nizations, is actually helping to support broader development efforts. In other word s, these few organizati ons are partnering and

PAGE 122

111 networking with other organizations in the developing world and helping to build their capacities. Thus, locally organized networks in the developing world, such as the NCSN, may help us understand how development can occur through local initiatives and may serve as models for systems of development and aid distribution in other areas. A more skeptical view of this finding re garding the importance of international ties, however, might suggest that little progr ess has actually been made to date with respect to local sustainability and self -reliance. Rather, the only sustainable organizations are those with international support, and NGOs have yet to figure out how to achieve sustainability through local effort s. This would suggest that much work remains before local sustainability and self-reliance are achieved, which is particularly urgent in light of dec lining international aid. An important note here is that the NC SN is relatively new, having only begun about three years ago. It is certainly possible that curren t institutional sustainability among organizations in the NCSN is more a re flection of pre-existi ng resources than the effects of network membership, but that afte r some period of time the effects of network membership will become more apparent. Futu re studies may look at other networks in the developing world that have been in exis tence longer, or longitudinal studies can look at the effects of networks over time. Limitations This study is the first of its kind, explor ing institutional sust ainability in the context of a particular netw ork of NGOs operating in the developing world. While there is much to be learned from this study, its li mitations are also worth noting so that they might be addressed in future studies.

PAGE 123

112 There is no single measure of institutional sustainability that can perfectly capture the concept. As mentioned, an organization may have existed for a long time but as a mere institutional shell without much benefi cial function, or an organization may have considerable financial res ources but it may be putting those resources toward administrative costs rather than programs Thus, numerous measures across eight categories were used here to attempt to capture the breadth of instit utional sustainability. The use of a variety of measures will hopefully help mitigate the limitations of any one measure. Even so, it seems that future studies may attempt to improve these measures. For example, as mentioned above, it is not clear whether more or fewer programs are more indicative of institutional sustainabi lity. Thus, rather than merely counting programs, future studies might develop a measure that accounts for how well funded programs are or whether programs actually re ach beneficiaries. Similarly, greater numbers of facilities and annual budgets may not necessarily mean greater institutional sustainability. Rather, resources should be considered in light of an organizationÂ’s mission and activities. Also, based on how the questions were frame d, particularly in the areas of records, program evaluations, and human resource system s, there was little variation between the organizations in their responses. Thus, future studies might probe more deeply into these areas to capture some of th e less obvious differences between the organizations. For example, if organizations claim to maintain electronic records, it might be worth asking how they store or backup these records, or if organizations claim to conduct program evaluations, appropriate follow-up probes might include who conducts these evaluations and what the organization subsequently does with them.

PAGE 124

113 There also may be other measures of institutional sustainability not included in this study that should be included in future studies. Most notably, effective leadership may be an important component of institutiona l sustainability. It is widely thought that inspirational and effective leadership is a key to an organizationÂ’s success (Edwards, 2002), and Okorley and Nkrumah (2012) found that supportive leadership is one of two factors most correlated with organizational sustainability. In ta lking with people in Nansana, I also had a general sense that various leaders were central to certain organizationsÂ’ successes and perhaps to Nansan a civil society more broadly. While there is certainly a counterargument that if an or ganization is overly dependent on its leader, it will collapse if and when that leader moves on, future studies should consider including some measure of strength of leadership as a component of institutional sustainability. With respect to partnerships, given the vast disagreement regarding whether partnerships exist, the extent of their formality, and what they entail, additional modifications should be made to questions in these areas. Most notably, it might be important to develop a better definition of pa rtnership for future studies. While the disagreement regarding what constitutes a partnership actually offered a learning opportunity in this study, if future studi es wish to more accurately measure the relationship between partnershi ps and institutional sustaina bility, it would be helpful to attain greater consistency acro ss organizations regarding the id entities of their partners. Additionally, while numbers of in degrees and out degrees offer some indication of interorganizational connectivity, future studies may wish to develop more complex measures of connectivity that go furt her in capturing relationship quality.

PAGE 125

114 Furthermore, a major limitation of this study more generally is that it depends on self-reports. While this methodology is effici ent given limited resources, more accurate data might be obtained through reviews of or ganizationsÂ’ records. Indeed, Barr et al. (2005) found that Ugandan respondents sometime s reported ideal situat ions rather than actual situations. While I do not believe that many, if a ny, organizations intentionally misled me regarding their institutional sustai nability or partnerships, I do believe that there were instances in which the informati on I sought through my que stions did not align with the information that organizations provide d to me due to differe nt understandings of certain concepts. An example of this, which I was able to correct for early in the process, was that some organizations understood my bud getary questions to refer to what they planned for, though what they planned for ofte n bore little relationship to their actual financial status. Viewing records and documentation would help reconcile any misunderstandings, though this would also cl early be a much more time and resource consuming endeavor. Finally, a major underlying issue here, which is worthy of mention, is that institutional sustainability in this study is conceptualized within a Western framework. I think it would be a travesty to take for grante d that institutional sustainability in the civil society sector in Nansana, Uganda can be fully understood through a Western lens and to assume that Western management systems will necessarily help the developing world. Future studies might focus on qualitativel y attempting to understand institutional sustainability in the context of the subject communities to better inform measures of institutional sustainability.

PAGE 126

115 Implications Existing domestic research suggests that networks and partners hips are important means of enhancing organizational capacity (Paarlberg & Varda, 2009). This study suggests, however, that, at least with respect to the subjec t network of NGOs, partnerships could be a function of existing resources, in that orga nizations with strong international ties and substant ial access to resources are more likely to attract partners within the network. This is not to say that t hose partnerships do not, in turn, help to build the capacity of existing organizations, though. Rather, it seems that a takeaway here may be that organizations should work toward improving their institutional sustainability for purposes of service delivery. As organizations do this, it is likely that they will attract more partners, and these partnerships will in turn help to build their capacities. “Success breeds support and commitment, which breeds even greater success, which breeds more support and commitment . ” (Collins, 2005). One way to do this might be for orga nizations with small budgets and limited resources to reduce their numbers of program s and focus on more limited service areas. Particularly since they are operating within a fairly cohesive ne twork, it might make sense for each organization to become more focused on a part icular service area and to complement, rather than duplicate, each other’s service areas. Indeed, avoiding duplication of services is one of the majo r goals of the NCSN (M. Senyonjo, personal communication, January 18, 2013). Organizations might choose to focus on a particular substantive area (such as education or health ) or they might choose to act as service facilitators, connecting resource-rich organi zations with more remote communities in need of available services.

PAGE 127

116 This study also shows that despite an existing objective in international development to help communities become more locally sustainable, there is still much room for improvement. While organizations in Nansana are largely locally run and many are focusing on more sustainabl e solutions to health and pove rty issues, the fact remains that these local organizations and their beneficiaries continue to be largely dependent on international resources. As such, organiza tions should continue to seek sustainable solutions and look for ways to increase thei r capacities beyond intern ational aid, such as through strategic partnerships. This is part icularly important in light of decreasing international aid. With respect to programs and management systems, this study may suggest certain areas in which there is room for impr ovement among organizations in the network. Specifically, the categories of institutional sustainability in which significant relationships were not found may be inform ative. While some of the lacks of significance may reflect imperfections in the m easures (which can be improved in future studies), other lacks of signi ficance may actually suggest areas in which the NCSN, or partnerships more broadly, could help or ganizations build their capacities. For example, there are differences in the areas of expected to actual budget ratios and numbers of human resources systems betw een organizations based on international ties but not based on numbers of partners. Thus, these might be areas that the NCSN could work toward improving among its members through education and capacity building. If organizations are better able to align their expected and actual budgets, they may be better poised to efficiently use their re sources and to more realistically plan their programming. Also, through the use of be tter human resource systems, even if

PAGE 128

117 organizations rely on volunteers, they might be able to enhance their capacities by promoting volunteer and employee training and ensuring accountability. Looking at partnerships in the network, th is study suggests that organizations are often only interested in partne ring up, or partnering with or ganizations they perceive as having greater resources than themselves. Th is was evidenced by the lack of symmetry in organizationsÂ’ identifications of their partners. As mentioned, despite the fact that organizations were given the same definition of partner, organizations tended to look up and out for their partners, such that many larg er organizations named less partners within the network and more partners among larger organizations outside of Nansana. There may be strategic value to partners at a vari ety of levels, though. This is suggested by both the literature and by anecdotal evidence w ithin the network. Claiborne et al. (2009) suggests that partnerships betw een microand macro-level or ganizations are important to capacity building, and, within the network, it does seem that some positive synergistic relationships currently exist. For example, certain organizations with considerable resources for actual service provision are curre ntly leveraging smalle r organizations with deeper connections to local communities to organize and reach potenti al beneficiaries. Organizations should recognize the benefits of these rela tionships and continue to develop them. Along similar lines, this study also show s that productive networks may exist in the developing world as a resu lt of local initiatives. Thus, when international NGOs come into an area, they might consider l ooking at what service delivery systems and networks are already in place. If existing systems are in place, not only might leveraging

PAGE 129

118 them lead to more efficient service delivery, but it might also help to promote movement toward second and third generation development work. Beyond its practical implications, this st udy makes a number of contributions to the development and NGO management literatu re by exploring a particular network of NGOs in the developing world and contributi ng to an understanding of the institutional sustainability of these organiza tions. The institutional sustai nability measures used in this study will hopefully serve as a reasonabl e starting point and should be continually refined based on this research and future studies, and additional qualitative measures should also be added. Future studies may bu ild off of this and related research to continue to develop measures of institutio nal sustainability, refine the definition of partner, explore the causal relationships be tween capacity and partnerships, and explore the connection between institutional sust ainability and outcome sustainability. Conclusion As described above, there is some eviden ce to support the hypothe sis in this study, which is that greater numbers of partnerships and levels of connectivity positively correlate with organizationsÂ’ ins titutional sustainability. Spec ifically, there is qualitative evidence that network membership enhances organizationsÂ’ capacitie s, and organizations that are frequently viewed as partners by ot her organizations specia lize in fewer programs and tend to have more beneficiaries, hi gher rates of property ownership, and more technology systems and computers. While there are some limited correlations between numbers of out degrees and measures of institutional sustainability, th e asymmetries between in degrees and out degrees may actually suggest more regardi ng how organizations understand the concept

PAGE 130

119 of partnership and how they value interorg anizational relationsh ips. There may be opportunities for enhancing partnerships and capacity build ing through helping organizations understand how relationships between larger and smaller organizations may be beneficial. Finally, this study shows that levels of institutional sustainability, at least in Nansana, remain largely a function of interna tional ties. On one hand, this suggests that international aid may be efficiently distribut ed through networks of organizations in the developing world, but it also s uggests that local se lf-reliance and sustai nability remain elusive. Future studies should look at the e ffects of networks over time and continue to explore how partnerships and networks can help NGOs build capacity and enhance their institutional sustainability.

PAGE 131

120 REFERENCES Atack, I. (1999). Four criteria of development NGO legitimacy. World Development, 27 855-864. Barr, A., Fafchamps, M., & Owens, T. (2005). The governance of non-governmental organizations in Uganda. World Development, 33 657-679. Cannon, L. (2002). Defining sustainability. In M. Edwards & A. Fowler (Eds.), The Earthscan reader on NGO management (pp. 363-365). London, UK: Earthscan Publications Ltd. Central Intelligence Agency. (2013). The world factbook Retrieved from https://www.cia.gov/library/publicati ons/the-world-factbook/geos/ug.html Claiborne, N., Junqing Liu, Vandenburgh, H., Hagen, J., Mera Rodas, A., Raunelli Sander, J. M., . Zurita Paucar, M. J. (2009). Northern Peruvian nongovernmental organizations. International Social Work, 52 327-341. Collins, J. (2005). Good to great and the social sectors Boulder, CO: Jim Collins. Council on Foundations, United States International Grantmaking. (2012). Country information: Uganda Retrieved from http://www.usig.org/countryinfo/uganda.asp Development Assistance Committee (1991). Principles for Evaluation of Development Assistance Retrieved from http://www.oecd.org/development/eva luationofdevelopmentprogrammes/2755284 .pdf Easterly, W. (2006). The white man’s burden New York, NY: The Penguin Group. Edwards, M. (2002). NGO performance: What breeds success? New evidence from South Asia. In M. Edwards & A. Fowler (Eds.), The Earthscan reader on NGO management (pp. 275-292). London, UK: Earthscan Publications Ltd. Edwards, M. & Fowler, A. (2002). In troduction: Changing challenges for NGDO management. In M. Edwards & A. Fowler (Eds.), The Earthscan reader on NGO management (pp. 1-10). London, UK: Eart hscan Publications Ltd. Financial Management Service. (2012). Treasury reporting rates of exchange Retrieved from http://fms.treas.gov/intn.html Fowler, A. (2002). NGO future s—Beyond aid: NGDO values and the fourth position. In M. Edwards & A. Fowler (Eds.), The Earthscan reader on NGO management (pp. 13-26). London, UK: Earthscan Publications Ltd.

PAGE 132

121 Gellert, G. A. (1996). Non-government organi zations in international health: Past successes, future challenges. International Journal of Health Planning and Management, 11 19-31. Guo, C. & Acar, M. (2005). Understanding collaboration among nonprofit organizations: Combining resource dependency, institutional, and network perspectives. Nonprofit and Voluntary Sector Quarterly, 34 340-361. Hanneman, R. A. & Riddle, M. (2005). Introduction to social network methods Riverside, CA: University of Calif ornia, Riverside. Retrieved from http://faculty.ucr.edu/~hanneman/ Herman, R. D & Renz, D. O. (2008). Adva ncing nonprofit organiza tional effectiveness research and theory: Nine theses. Nonprofit Management and Leadership, 18 399-415. Kagan, S. L. (1991). United we stand: Collaboration fo r child care and early education services New York, NY: Teachers College Press. Kania, J. & Kramer, M. (2011). Collective impact. Standford Social Innovation Review, Winter 2011 36-41. Korten, D. C. (1987). Third generation NGO strategies: A key to people-centered development. World Development, 15 145-159. Lewis, D. (2003). NGOs, organizational cult ure, and institutiona l sustainability. The ANNALS of the American Academy of Political and Social Science, 590 212-226. Lyons, M., Smuts, C., & Stephens, A. (2001). Participation, empowerment and sustainability: (How) do the links work? Urban Studies, 38 1233-1251. McPherson, M., Smith-Lovin, L., & Brashears, M. E. (2006). Social isolation in America: Changes in core discussion networks over two decades. American Sociological Review, 71 353-375. Mercer, A., Khan, M. H., Daulatuzzaman, M., & Reid, J. (2004). Effectiveness of an NGO primary health care programme in rural Bangladesh: Evidence from the management information system. Health Policy and Planning, 19 187-198. Nantakiika, J. (2011). Uganda: Na nsana is booming. Retrieved from http://allafrica.co m/stories/201109120040.html Okorley, E. L. & Nkrumah, E. E. (2012). Organi sational factors influencing sustainability of local non-governmental organisations: Lessons from a Ghanaian context. International Journal of Social Economics, 39 330-341.

PAGE 133

122 Paarlberg, L. E. & Varda, D. M. (2009) Community carrying capacity: A network perspective. Nonprofit and Voluntary Sector Quarterly, 38 597-613. Provan, K. G. & Milward, H. B. (2001). Do networks really work? A framework for evaluating public-sector organizational networks. Public Administration Review, 61 414-423. Ralser, T. (2007). ROI for nonprofits: The new key to sustainability Hoboken, NJ: John Wiley & Sons, Inc. Sawhill, J. C. & Williamson, D. (2001). Mission impossible? Measuring success in nonprofit organizations. Nonprofit Management and Leadership, 11 371-386. Selden, S. C. & Sowa, J. E. (2004). Testing a multi-dimensional model of organizational performance: Prospects and problems. Journal of Public Administration Research and Theory, 14 395-416. Selden, S. C., Sowa, J. E., & Sandfort, J. (2006). The impact of nonprofit collaboration in early child care and education on management and program outcomes. Public Administration Review, 66 412-425. Sowa, J. E. (2008). Implementing interagenc y collaborations: Expl oring variation in collaborative ventures in human service organizations. Administration and Society, 40 298-323. Sowa, J. E. (2009). The collaboration decision in nonprofit organizati ons: Views from the front line. Nonprofit and Voluntary Sector Quarterly, 38 1003-1025. Sowa, J. E., Selden, S. C., & Sandfort, J. R. (2004). No longer unmeasurable? A multidimensional integrated model of nonprofit organizational effectiveness. Nonprofit and Voluntary Sector Quarterly, 33 711-728. Uganda Bureau of Statistics. (2010a). TP5: 2010 mid-year projected populations for Town Councils Retrieved from http://www.ubos.org/onlinefiles/uploa ds/ubos/pdf%20documents/TP52010.pdf Uganda Bureau of Statistics. (2010b). Uganda national household survey report 2009/2010 Retrieved from http://www.ubos.org/UNHS0910/chapter 2_householdcharacteristics.html Varda, D. M., Chandra, A., Stern, S. A., & Lurie, N. (2008). Core dimensions of connectivity in public health collaboratives. Journal of Public Health Management and Practice, 14 E1-E7.

PAGE 134

123 Varda, D., Shoup, J. A., & Miller, S. (2012). A systematic review of collaboration and network research in the public affairs lit erature: Implications for public health practice and research. American Journal of Public Health, 102 564-571.

PAGE 135

124 APPENDIX A

PAGE 136

125

PAGE 137

126

PAGE 138

127

PAGE 139

128

PAGE 140

129

PAGE 141

130

PAGE 142

131

PAGE 143

132

PAGE 144

133

PAGE 145

134 APPENDIX B Glossary of Social Network Analysis Terms* Asymmetric tie: a connection between two organizat ions in which only one of the organizations mentioned the other as a partne r (also referred to as a directional tie) Attribute: a color, shape, or size assigned to a node based on a particular organizational characteristic Bonded tie: a connection between two organiza tions in which both organizations mentioned the other as a partner (als o referred to as a reciprocal tie) Directional tie: a connection between two organizations in which only one of the organizations mentioned the other as a partne r (also referred to as an asymmetric tie) In degrees: the number of times an organization is mentioned as a partner by other organizations in the network Isolate: an organization disconnected from th e network map because it mentioned no partners within the network and was not mentioned as a partner by any other organizations in the network Node: the points in a network map, which represent organizations Out degrees: the number of partners an organi zation mentions within the network Reciprocal tie: a connection between tw o organizations in which both organizations mentioned the other as a partner (also referred to as a bonded tie) These definitions are based on Hanneman and Riddle (2005).