Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
The institutionalization of nonprofit management: emergence, development, and legitimization
(USC Thesis Other)
The institutionalization of nonprofit management: emergence, development, and legitimization
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
THE INSTITUTIONALIZATION OF NONPROFIT MANAGEMENT:
EMERGENCE, DEVELOPMENT, AND LEGITIMIZATION
by
Youngmi Lee
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(POLICY, PLANNING, AND DEVELOPMENT)
August 2015
Copyright 2015 Youngmi Lee
i
DEDICATION
To my husband Jae young,
who has always been next to me through thick and thin during my PhD years,
and without whom I would not have been able to complete this work.
ii
TABLE OF CONTENTS
DEDICATION................................................................................................................................ i
LIST OF TABLES .........................................................................................................................v
LIST OF FIGURES ..................................................................................................................... vi
ABSTRACT ................................................................................................................................. vii
CHAPTER 1. INTRODUCTION .................................................................................................1
1.1. Why Nonprofit Management? ........................................................................................1
1.2. Purpose of the Study ........................................................................................................5
1.3. Research Questions ..........................................................................................................8
1.4. Overview of Methodology .............................................................................................10
1.5. Outline of Dissertation ...................................................................................................12
CHAPTER 2. THE STRUCTURATION OF THE NONPROFIT SECTOR ........................14
2.1. Public Management and the Nonprofit Sector ...............................................................14
2.2. The Nonprofit Sector and the Process of Structuration .................................................22
2.3. Sociological Institutionalism and Nonprofit Management Research and Education ....32
2.4. Discussion ......................................................................................................................37
CHAPTER 3. THE EXPANSION OF NONPROFIT MANAGEMENT RESEARCH
AND EDUCATION .....................................................................................................................39
3.1. Background ....................................................................................................................39
3.2. A Conceptual Model: Mechanisms of the Expansion....................................................41
3.3. Data and Methods ..........................................................................................................48
3.4. Results ............................................................................................................................58
3.5. Discussion ......................................................................................................................66
iii
CHAPTER 4. VARIATION IN THE DISCIPLINARY SETTING FOR
NONPROFIT MANAGEMENT STUDIES ..............................................................................70
4.1. Background ....................................................................................................................70
4.2. A Theoretical Framework: Dynamics surrounding the Disciplinary Location .............72
4.3. Data and Methods ..........................................................................................................80
4.4. Results ............................................................................................................................92
4.5. Discussion ....................................................................................................................101
CHAPTER 5. INSIDE THE BOX: THE GRADUAL INSTITUTIONALIZATION
PROCESS OF NONPROFIT MANAGEMENT STUDIES ..................................................105
5.1. Background ..................................................................................................................105
5.2. Nonprofit Studies as an Emerging Field: Formation and Early Development ............108
5.3. Data and Methods ........................................................................................................114
5.4. Findings........................................................................................................................129
5.5. Discussion ....................................................................................................................142
CHAPTER 6. CONCLUSION ..................................................................................................148
6.1. Summary of Findings ...................................................................................................148
6.2. Limitations and Additional Directions for Research ...................................................152
6.3. Contributions to the Field ............................................................................................154
6.4. Future Directions and Predictions ................................................................................156
BIBLIOGRAPHY ......................................................................................................................158
APPENDICES ............................................................................................................................181
Appendix A. Comparison between Original and Imputed Data (Chapter 3) ......................181
Appendix B. Piecewise Constant Models with Period-Specific Effects .............................182
Appendix C. Expanded Exponential Models for Structuration Variables ..........................186
Appendix D. Comparison between Original and Imputed Data (Chapter 4) ......................189
Appendix E. Expanded Competing-Risks Models for Institutional Context Variables .....190
Appendix F. Multinomial Logistic Approach to Competing Risks ....................................193
iv
Appendix G. Summary of Seven Cases ..............................................................................198
Appendix H. Interview Protocol and Instrument ................................................................199
Appendix I. Technical Interpretation of the QCA Results .................................................203
v
LIST OF TABLES
Table 2.1 Number of Charitable Nonprofits by Type of Service, 1989-2010 ...............................25
Table 2.2 Trends in Volunteering, 2002–2012 ..............................................................................28
Table 3.1 Definitions and Descriptive Statistics for Covariates ....................................................55
Table 3.2 Correlation Matrix of Covariates ...................................................................................56
Table 3.3 Exponential Transition Rate Models: Degree Program .................................................61
Table 3.4 Exponential Transition Rate Models: Research Center .................................................63
Table 3.5 Exponential Transition Rate Models: Any Type of Program ........................................65
Table 3.6 Exponential Transition Rate Models with Historical Context Effects ..........................67
Table 4.1 Definitions and Descriptive Statistics for Covariates ....................................................88
Table 4.2 Correlation Matrix of Covariates, 1971-2011 ................................................................89
Table 4.3 Correlation Matrix of Covariates, 1990-2011 ................................................................89
Table 4.4 Competing-Risks Models, 1971-2011 ...........................................................................94
Table 4.5 Competing-Risks Models with Regional Nonprofit-Sector Effects, 1990-2011 ...........95
Table 5.1 Fuzzy-Set Scores for Outcomes ...................................................................................125
Table 5.2 Crisp- and Fuzzy-Set Scores for Causal Conditions ....................................................128
Table 5.3 Descriptive Statistics and Correlation ..........................................................................129
Table 5.4 Configurations for the Timing of Formation ...............................................................132
Table 5.5 Configurations for the Institutional Stability ...............................................................138
vi
LIST OF FIGURES
Figure 2.1 Number of 501(c)(3) Charitable Organizations Registered with the Internal
Revenue Service, 1976-2010 ...................................................................................... 24
Figure 2.2 Trends in Revenues and Expenses of Reporting Public Charities and
Private Foundations, 1985-2008 ................................................................................. 26
Figure 2.3 Employment in the Nonprofit Sector and Selected Major U.S. Industries, 1998
and 2006 ......................................................................................................................27
Figure 3.1 The Growth of Academic Programs, Cumulative Count and Yearly Adoptions
(1971-2011) ...................................................................................................................59
Figure 4.1 Disciplinary Settings for Academic Programs, Cumulative Count (1971-2011) ..........92
vii
ABSTRACT
The purpose of this dissertation is to develop a better understanding of the emergence,
development, and institutionalization of nonprofit management in American society. Why has
nonprofit management expanded throughout the United States? How have various historical, social,
and organizational factors influenced the emergence and expansion of nonprofit management? In
what context and manner has nonprofit management been differently shaped and developed over
space and time?
To search for answers to these questions, this dissertation documents the evolution of
nonprofit management and explores various mechanisms associated with the evolution, drawing
on institutional sociology and other organizational theories. First, it provides a theoretical
framework for understanding the structuration of the nonprofit sector by closely examining the
development of the field and institutional processes behind the development. Additionally, in
moving the focus from the evolution of nonprofit management in a general context to that of the
higher education, this dissertation analyzes factors associated with the expansion of nonprofit
management research and education. More specifically, it utilizes both quantitative and qualitative
methods to identify various factors that explain the incorporation of nonprofit management studies
programs into higher education institutions and explores variations in its early development.
The empirical findings suggest that three broad sets of mechanisms lead to the emergence
and diffusion of nonprofit management research and education: (1) the structuration of the
nonprofit sector at the national level, which is observed by the nationwide growth in nonprofit
organizations, professional nonprofit associations, government grants to nonprofits, and nonprofit
publications and news articles; (2) a set of social and political conditions at the state level,
viii
including a large nonprofit sector, population heterogeneity, economic wealth, and political
conservatism; and (3) a set of university-specific organizational characteristics, such as being large,
private, less research oriented, and with resource-rich receptor sites of nonprofit management (e.g.,
public affairs, business, and social work schools).
The findings also indicate that historical changes in the three sets of mechanisms are closely
linked to different developmental processes of nonprofit management studies programs. In other
words, variation over time in the adoption rate, disciplinary setting, and organizational stability of
programs can be described as a product of historically-situated institutional environments, regional
social conditions, and organizational dynamics. For example, the adoption rate of programs has
increased over time as nonprofit management discourse has achieved more of a rule-like status—
especially after the early-2000s. Also, the preferred academic home for programs has varied across
time periods as nonprofit management discourse itself has gone through several historical stages—
including the most recent business-like model dominant stage. Lastly, the organizational stability
of adopted programs has fluctuated depending on ever-changing university dynamics—like a top
administrator change or funding stream change.
Taken together, my dissertation complements prior research on the expansion and
development of nonprofit management over the last few decades. By bringing attention to the
structuration of the nonprofit sector and critical junctures in history, while also considering the
role of university characteristics, my research highlights how theories like sociological
institutionalism matter for public management research. In addition, my research has the potential
to inform practice because it provides a context for further investigation on how the growth of
academic programs (and increased professionalism in the nonprofit sector) change the daily work
of nonprofits and the strategy of public affairs schools.
1
CHAPTER 1. INTRODUCTION
1.1. Why Nonprofit Management?
Nonprofit organizations, which are distinct from government agencies or for-profit firms,
make crucial contributions to both the global and community life of today. Indeed, nonprofits
deliver a variety of public services, give voice to a wide range of social, political, environmental,
ethnic, and community concerns, build social capital and connections among individuals, and help
to nurture and sustain essential national values such as the principles of individualism and
solidarity (Boris 2006; Frumkin 2002; Salamon 2002a, 2012). In other words, nonprofits not only
play a vital role in a mixed economy for public services, but also contribute to democratic pluralism
as vehicles for the expression of public interest and political mobilization. They have a dual
identity as service providers for public goods and democratic leaders for social and political
engagement (Clemens 2006; Suárez 2009).
Even though the dual identity of nonprofits has taken firm root in American history, it is only
recently that the role of the nonprofit sector has become visible and recognizable in the American
social and political scene. During the Great Society era of the 1960s, when government faced
severe criticism and pressure mounted to provide more efficient services, nonprofits came to be
viewed as an ideal and reliable alternative to government. The service role of nonprofits was
extended and strengthened in the early 1980s under what was described as the marketization of
welfare, which emphasized a greater level of privatization and devolution in delivering social
welfare programs through contracting-out (Frumkin 2002; Grønbjerg & Salamon 2012; Salamon
1995; Smith & Lipsky 1993). During the process of welfare reform in the mid-1990s, nonprofits
were further highlighted as an effective mechanism for providing noncash assistance and
2
customized social services to poor families through community-based service systems and based
on close relationships with local communities (Allard 2009; Smith 2012). In the American modern
welfare state, nonprofits came to be viewed as “the center of the policy debate and… central
instruments of development and welfare state reform” (Anheier & Salamon 2006, p. 93).
The increasing visibility and policy relevance of nonprofits was, however, not a wholly
American phenomenon, nor was it limited to service delivery functions. In the late 1970s, as New
Public Management (NPM) reforms took hold and a striking upsurge of organized private
voluntary activity known as the global associational revolution (Salamon 1994a) diffused broadly,
nonprofits became an indispensable part of governance systems and civil society in many countries.
The NPM reformers relied heavily on the contract relationship with nonprofits, trying to “replace
traditional rule-based, authority-driven processes [of government] with market-based,
competition-driven tactics” (Kettl 2005, p. 3), for a more efficient governance system (Anheier &
Salamon 2006; Boston et al. 1996; Hood 1991; Kaboolian 1998; Kettl 1997, 2005; Osborne &
Gaebler 1992). Moreover, the size and scope of nonprofit organizations increased as their virtues
of voluntarism and activism were acknowledged by civil society. As a result, they received a large
amount of attention over the process of searching for a third way (Giddens 1998) between social
democracy and free-market neoliberalism and a middle way between government and for-profit
firms (Salamon 1994b, 1995). Altogether, various social and political forces from both inside and
outside the country stimulated the growth of nonprofits and extended partnerships with
government, which led to the policy saliency of nonprofits’ dual identity in the 20
th
century.
Now, in the era of New Public Governance (NPG), which is characterized by cross-sectoral
arrangements and networks designed for the better provision of public service (Andrews &
Entwistle 2010; Bryson & Crosby 2008; Krueathep et al. 2008; McGuire 2006; Osborne 2010;
3
Osborne et al. 2012), nonprofits play a more prominent role in the development of social policy,
and their interactions with government are sometimes much more dynamic and based on a
stewardship rather than a principal-agent relationship (Lamothe & Lamothe 2012; Van Slyke
2007). In fact, government agencies operate in collaborative settings in which the extent of
interaction with and reliance on nonprofit organizations have increased over the years through
various forms of partnerships, alliances, committees, coalitions, and councils (Agranoff &
McGuire 2003; Posner 2009). Nonprofits now have a larger share of social and human service
delivery than do government agencies (Boris et al. 2010; Salamon 2003; Sowa 2008), and
government managers even predict that there would be a severe disruption in services if nonprofit
organizations were no longer willing or able to work with government to provide public services
(Denhardt et al. 2008).
As government and nonprofits share the role of public service provider, furthermore, the
careers of public-spirited people have increasingly moved between government and the nonprofit
sector. In fact, a growing number of public administration and policy school graduates have found
careers in the nonprofit sector (Light 2000). This growth has occurred not because nonprofits are
substitutes for governments that are composed of large, impersonal bureaucracies and are thus
difficult to work for, but because nonprofits have become a complementary venue for creating
public value and delivering public services at the front line (Cohen & Abbott 2000; Posner 2009;
Salamon 1998).
In this era of collaborative governance, government practitioners are increasingly expected
to understand both the similarities and differences between government and nonprofit operations
and find a better way of interacting with nonprofit partners, to minimize potential managerial
conflicts and further optimize the governance system. Government managers may need to be
4
trained not only as “experts of their particular institutions” (Mirabella & Wish 2000, p. 227), but
also as “experts on the different tools through which these institutions interact” (Salamon 1998, p.
144), so that they can address managerial conflicts derived from the different values and strategies
of each sector and make effective adaptations to the career switching that occurs between the two
sectors (Posner 2009; Salamon 1995, 2002b).
Practitioners and scholars concerned with the performance of public programs are also asked
to invent new mechanisms for managing and supporting the activities of nonprofit partners, such
as performance evaluations, purchase-of-service contracts, tax subsidies, and grants-in-aid
(O’Leary & Bingham 2009; Salamon 1995), to hold nonprofits accountable for achieving cost
efficiency and for delivering a quality service (Boris & Steuerle 2006; Smith & Smyth 2010).
These financial and institutional supports for enhancing the capacity of contracted nonprofits are
indeed viewed as another form of governmental investment for making collaborations operate
better for both sides (Mirabella 2001; Salamon 2003). Hence, a central task for the government
practitioners and scholars of today is to develop a deeper understanding of the goals, missions,
values, operations, procedures, cultures, and languages of nonprofit organizations to develop an
effective and more integrated public service system.
It is clear that within this new regime, the contemporary processes of public management
and public policy extend far beyond the traditional boundary of government—while government
remains central to the public service delivery system (DiIulio et al. 1993; Kettl 2006; Radin 1996).
In fact, the operational boundary of government has changed to include both the nonprofit sector
and parts of the for-profit sector (Agranoff & McGuire 2001; Salamon 1995); as O’Neill (2007, p.
171S) noted, “we’ve come a very long way from the days when public (sector) meant simply
government.” The transformed configuration of the U.S. public sector has called for not only the
5
re-conceptualization of the nature of public management, but also a change in the central unit of
analysis from government agent to public agent, which refers to people who are working on public
problems and service delivery regardless of whether the person is attached to government,
nonprofit, or even for-profit organizations (Salamon 1995, 1998). Public management has
expanded beyond governmental processes of policy making, implementation and the behavior of
civil servants. Public management reflects the multi-sectoral operating reality of the 21
st
century,
infusing its concepts with nonprofit and for-profit management (Kettl 2006; McGuire 2006). In
this regard, public management and policy scholars can benefit from a deeper understanding of
nonprofit management, which may strengthen collaborative management expertise and facilitate
complex cross-sectoral arrangements.
1.2. Purpose of the Study
Even though the issues surrounding the contemporary governance system make nonprofit
management an appropriate topic for those who are interested in the theory and practice of public
management, relatively little is known about the nature of nonprofit management itself. Prior
studies have contributed to an understanding of the growing discourse and practice of nonprofit
management. For instance, public and nonprofit management studies have explored (1) general
trends that have been termed the rationalization, professionalization, and commercialization of
nonprofit management and their consequences (e.g., Berman 1999; Eikenberry & Kluver 2004;
Hwang & Powell 2009; Kreutzer & Jäger 2011; Pearce 1993; Stone et al. 1999; Suárez 2011), (2)
the diffusion of nonprofit management ideas and practices through the establishment of academic
programs, professional associations, and management support agencies (e.g., Mirabella 2007;
Mirabella & Wish 2001; O’Neill 2005; O’Neill & Young 1988; Smith 1997), and (3) the growth
6
of academic recognition and interest in nonprofit management and its potential as a new academic
discipline and knowledge base (e.g., Bushouse & Sowa 2012; Graddy et al. 2011; Hall 1992; Katz
1999; Smith 1999). These studies provide a great deal of contextual detail, yet many questions
remain about where nonprofit management emerged, why and how it gained legitimacy, and how
it is evolving.
Therefore, the purpose of this dissertation is to develop a better understanding of the
emergence, development, and institutionalization of nonprofit management in American society.
In other words, my dissertation aims to document the expansion of nonprofit management ideas
and practices over time and identify the historical, social, and organizational mechanisms behind
its dramatic expansion in American society. In doing so, the dissertation specifically focuses on
the growth of nonprofit management research and education programs in higher education. The
study of social causes, social movements, and innovation diffusion is often based on the analysis
of higher education organizations, because universities and colleges are the primary social sites
that reflect changes in public life by continuously reconstituting their structure and knowledge
categories (Gumport & Snydman 2002). For this reason, it is expected that the dynamics
surrounding the development of nonprofit management studies programs in universities and
colleges mirrors the process through which general nonprofit management discourse is developed
and institutionalized in the broader society.
This dissertation offers a new way to think about the emergence, development, and
institutionalization of nonprofit management by drawing on both neoinstitutional theory (e.g.,
DiMaggio & Powell 1983; Meyer & Jepperson 2000; Meyer & Rowan 1977; Powell & DiMaggio
1991; Scott & Meyer 1994) and social constructivism (e.g., Berger & Luckmann 1967; Knorr-
Cetina 1994; Krasner 1983; March & Olsen 1984). Sociological institutionalism, which is a
7
sociological perspective within the neoinstitutional tradition, has not received sufficient attention
as an explanatory framework for understanding organizations in public management research.
However, this approach appears to provide important insight into our understanding of institutional
influences on organizational structure and behavior. Indeed, it captures the interplay between
(individual and organizational) actors and institutional environment and its impact on
organizational development, which has not been adequately addressed by the most prominent
approaches in the public management literature, such as resource dependency and contingency
theories (Binder 2007).
In addition, this dissertation provides an empirical explanation for why and how nonprofit
management research and education have emerged and spread over the last several decades. Public
management practitioners should pay attention to the empirical findings in this dissertation,
because the growth of nonprofit training and education programs is one of the indicators (and
facilitators) of the increased professionalism in the nonprofit sector (Bromley & Orchard
forthcoming; Hwang & Powell 2009). Based on the empirical work in this dissertation,
practitioners could develop a better understanding of how the development of academic programs
and increased professionalism changes the daily work of nonprofits and the relations between
public and nonprofit management. Public management scholars should also be interested in the
developmental process of academic programs for more immediate and practical reasons.
Considering the fact that public affairs schools have been the most ardent and enthusiastic
advocates of nonprofit management programs (Mirabella 2007; Mirabella & Wish 2000; Wish &
Mirabella 1998), understanding the factors that influence program creation and development could
inform the strategy of public affairs schools. If scholars and administrators want to understand
their field, they must be interested in the mechanisms that are driving change.
8
1.3. Research Questions
The overarching question that guides the proposed study is why and how nonprofit
management has emerged and expanded in the United States. The specific questions this study
addresses include the following:
1. How and under what conditions is nonprofit management legitimized as a distinct
knowledge base of American society?
2. How do various historical, social, and organizational factors influence the expansion of
nonprofit management?
3. In what context and manner is nonprofit management shaped and developed over space
and time?
To search for answers to these questions, this dissertation begins by developing a theoretical
argument (Chapter 2). It reviews the literature on management in the public sector and discusses
the relationship between public management and the nonprofit sector. The chapter then focuses
more specifically on the structuration of the nonprofit sector by documenting various aspects of
the structuration process. Finally, the chapter demonstrates how the structuration of the nonprofit
sector shapes the emergence and development of nonprofit management research and education
by emphasizing the institutional mechanisms that contribute to their diffusion.
In the following three chapters, this conceptual framework is elaborated through and
supported by empirical research on the development of nonprofit management studies programs
in higher education. More specifically, this dissertation explores how nonprofit management has
emerged and diffused across the country based on the analysis of longitudinal data (Chapter 3).
This work argues that the rapid and wide expansion of nonprofit management programs may be
explained by the following three factors: (1) the structuration of the nonprofit sector at the macro-
9
level, which may be observed in the increasing number of nonprofits and professional associations,
tax and policy reforms, and publications and news articles related to nonprofit management; (2)
the socioeconomic and political contexts surrounding the universities at the meso-level, with state-
level measures such as demographic setting, the economic and social climates, and political
orientation; and (3) university-specific organizational conditions at the micro-level, such as the
university size, mission, history, prestige, campus diversity, and departmental politics. In addition,
it emphasizes the historical context as a relevant factor in explaining the speed and extent of the
expansion.
This dissertation then investigates the different developmental processes of nonprofit
management (Chapter 4). It describes a variety of disciplinary settings in which nonprofit
management programs have been framed and offered. Furthermore, it explains how various
historical, social, and organizational forces shape different intellectual settings over space and time.
Based on an analysis of longitudinal data, this work empirically tests the assumption that variation
in the intellectual sources are conditioned by (1) the institutional context in which national
nonprofit and academic communities shape and reshape their understandings about what counts as
valid nonprofit management knowledge in the changing historical and social environments; (2) the
state-level context of the nonprofit sector in which nonprofit constituencies construct particular
needs and preferences related to their own managerial issues; and (3) university dynamics through
which faculty, administrations, students, and funders negotiate their different stakes in developing
the field of nonprofit management studies.
Finally, the present dissertation examines the gradual institutionalization of nonprofit
management in greater detail (Chapter 5). It addresses the formation and development processes
through which newly emerging nonprofit management studies are defined, justified, and evolved
10
in academic organizations. It assumes that the institutionalization process of nonprofit
management studies is governed by (1) university dynamics, including interactions among
university actors (e.g., deans, chairs, host faculties, administrators, students, and external funders)
and university organizational characteristics (e.g., the university mission, history, size, structure,
reputation, and student profiles) and (2) the social and political conditions surrounding universities,
such as the broader institutional environment, the state nonprofit sector setting, demographic
heterogeneity, economic status, and political climate. This dissertation verifies this type of
assumption based on a comparative case study of seven universities, and it further accounts for
variation in the institutionalization processes across universities in terms of differences in the two
broad sets of mechanisms.
1.4. Overview of Methodology
This dissertation relies on both quantitative and qualitative methods and involves multiple
levels of analysis to provide more comprehensive and in-depth insights into the emergence,
development, and legitimization of nonprofit management. It designs two quantitative studies
(Chapters 3 and 4) and one comparative case study (Chapter 5).
For the quantitative studies in Chapters 3 and 4, the present dissertation not only constructs
a new database on academic research and education programs for nonprofit management in U.S.
higher education institutions, but also compiles a variety of data sources at the federal, state, and
individual university level from 1971 to 2011, creating a 41-year spanning longitudinal dataset.
Data on academic research and education programs in nonprofit management
1
, including the
1
Academic research and education programs for nonprofit management are defined here as university-based degree
programs, non-degree programs, and research centers designed exclusively for general nonprofit management and
11
timing of the first program to be established and its disciplinary location in each institution, are
generated through a scrupulous review of 1,451 sample institutions’ online archives between 2011
and 2012 and follow-up email correspondence with more than 700 program directors, department
chairs, and school deans at selected institutions between 2012 and 2013. Data on national, state,
and individual university level indicators (1971-2011) are obtained from a variety of publicly
available data sources, including the Internal Revenue Service (IRS) Tax Stats, National Center
for Charitable Statistics (NCCS), U.S. Census Bureau, Higher Education General Information
Survey (HEGIS), Integrated Postsecondary Education Data System (IPEDS), ProQuest database,
and Lexis-Nexis database. As a methodology, this study employs techniques of event history
analysis, which is a special type of log-linear analysis that is useful for analyzing whether, when,
and under what conditions an event (or a qualitative change) was experienced by a sample of
individuals during a period of observation (Box-Steffensmeier & Jones 2004; Tuma & Hannan
1984; Yamaguchi 1991). More specifically, exponential transition rate models are used in Chapter
3 to estimate the transition rate of each university from not-adopting to adopting its first nonprofit
management program under various historical, social, and organizational conditions over time.
Exponential competing-risks models are used in Chapter 4 to understand the transition rate of each
university from not-adopting to adopting its first nonprofit management program within a certain
disciplinary setting under various conditions over time.
For the comparative case study in Chapter 5, seven university cases are purposely selected
by considering their similarities and differences in the gradual institutionalization process of
nonprofit management programs as well as historical, social, and organizational conditions that
contribute to the process. The primary data for this case study are collected through 35-90 minute
philanthropic studies—excluding those for industry-specific studies (e.g., arts management and human services
management).
12
semi-structured interviews with a total of 15 program directors, department chairs, and school
deans at the seven universities (which were conducted in 2014). Complementary secondary data
are gathered from a review of various types of historical documents that were obtained from each
institution’s online archives, e.g., college and school catalogs, department brochures, syllabi, and
other types of archival sources. A portion of the data used for the two quantitative studies are also
incorporated into the dataset as supplementary quantitative evidence. The collected data are
analyzed with a qualitative comparative analysis method based on the logic of crisp- and fuzzy-
sets. Qualitative comparative analysis is an analytic technique that allows for the statistical analysis
of qualitative evidence and small-N situations by using Boolean algebra rather than correlational
methods (Ragin 1987, 2000, 2008). This method also allows systematic, cross-case comparisons
and detailed within-case analysis simultaneously, by integrating the advantages of the variable-
oriented and case-oriented approaches (Ragin 2008; Rihoux & Ragin 2009). This study thus
employs qualitative comparative analysis to examine the different cross-university and within-
university patterns related to the development of nonprofit management studies in a more compact
and analytic way.
1.5. Outline of Dissertation
This dissertation is organized into six chapters. This first chapter, “Introduction,” provides
practical and theoretical rationales as to why public administration and policy scholars should care
about the nonprofit sector and nonprofit management in a collaborative governance context. The
chapter also outlines research questions and the conceptual frame, research design and methods,
and the overall structure of the dissertation. Chapter Two, “The Structuration of the Nonprofit
Sector,” develops my argument about the evolution of nonprofit management in American society.
13
This chapter documents the process of structuration in the nonprofit sector, focusing on recent
changes and developments in the public and nonprofit sectors. It also discusses the impact of this
structuration on the evolution of nonprofit management research and education, leveraging
sociological institutionalism. Chapter Three, “The Expansion of Nonprofit Management Research
and Education,” empirically tests a conceptual model that accounts for the rapid and wide
expansion of nonprofit management studies programs based on newly constructed event history
data (1971-2011) by using exponential transition rate models. This chapter explores whether
various historical, social, and organizational factors affect the emergence of university-based
nonprofit management programs. Chapter Four, “Variation in the Disciplinary Setting for
Nonprofit Management Studies,” examines why and how nonprofit management studies are
developed in a variety of forms and disciplinary venues based on an analysis of event history data
(1971-2011) using exponential competing-risks models. It empirically investigates an association
between variation in the disciplinary setting for nonprofit management programs and historical,
social, and organizational factors over space and time. Chapter Five, “Inside the Box: The
Institutionalization Process of Nonprofit Management Studies,” focuses on the institutionalization
of nonprofit management studies in the higher education context, with a comparative case study
of seven universities. Using qualitative comparative analysis techniques, it explains how different
historical, social, and organizational conditions influence the formation and early development of
the study field across university campuses. Chapter Six, “Conclusion,” closes the dissertation by
summarizing the major findings and limitations of this study, discussing the implications for theory
and practice, and suggesting directions for future study.
14
CHAPTER 2. THE STRUCTURATION OF THE NONPROFIT SECTOR
“The most important factor is simply the growth of the nonprofit sector after World
War II, especially the mid-1960s… The growth of the nonprofit sector was a necessary
condition. There were also sufficient conditions for making it happen—entrepreneur
activities of several people and organizational support from universities, foundations,
and national organizations.”
- Interview with faculty, July 2014
This chapter reviews the literature on management in the public sector and the relationship
between public management and the nonprofit sector. The chapter then focuses more specifically
on the structuration of the nonprofit sector by documenting various aspects of the structuration
process, such as the growth of the sector and the increased rationalization and professionalism in
the sector. Finally, the chapter demonstrates how the structuration of the nonprofit sector shapes
the emergence and development of nonprofit management research and education by emphasizing
the institutional mechanisms that contribute to their diffusion.
2.1. Public Management and the Nonprofit Sector
The relationship between nonprofits and the public sector has changed dramatically over the
last few decades, and interest in the nonprofit sector in both public management practice and the
literature has never been higher (Smith 2010). The key driving forces behind this recent trend
appear to be two of the broad reform movements in the public sector: (1) New Public Management
(NPM), which emerged in the late 1970s and early 1980s, and (2) the post-NPM movements, which
emerged in the beginning of the 21
st
century—including New Public Governance (NPG) and New
Public Service (NPS).
15
The NPM Reform and the Role of Nonprofits as Contractors
The NPM reform movement developed in part as a response to the deficiencies of
bureaucratic government, which was often regarded as inflexible, inefficient, and ineffective
(Alford & Hughes 2008; Salamon 2002b). Theoretically, NPM was influenced by two different
streams of ideas (Hood 1991; Kettl 1997; Terry 1998): first, business-type managerialism, which
is an updated version of an older scientific management tradition (also called Taylorism) that
advocates market-based thinking such as entrepreneurialism, innovation, competition, and
customer responsiveness; and second, rational choice theory and neoclassical economics —public
choice theory, principal-agent theory, and transaction cost theory, which posit “governmental
decision makers as self-interested subjects, working in an environment in which information
asymmetry, bounded rationality and opportunism lead to transaction costs and agency costs”
(Groot & Budding 2008, p. 2). Consequently, the NPM movement was associated with the
introduction of private sector management ideas and techniques in the public domain, thus
“[replacing] traditional rule-based, authority-driven processes with market-based, competition-
driven tactics” (Kettl 2005, p. 3). According to Hood (1991, pp. 4-5), seven key elements of NPM
can be summarized as follows:
1. The growth of hands-on top management—“in its own right and not as an off-shoot
of professionalism” (Osborne 2010, p. 3)—and a greater stress on entrepreneurial
leadership within public service organizations;
2. A greater emphasis on quantitative evaluation and performance management;
3. Greater attention to output rather than input controls;
4. The disaggregation of the public sector into business units organized along product
or service lines—“a shift from a unitary, functional form to a multidivisional
structure” (Anheier 2009, p. 1083);
16
5. A greater focus on the value of contract-based competition for public service
delivery—“the introduction of managed markets with public agencies as funders
and contract managers and private for-profit and nonprofit providers as contractors”
(Anheier 2009, p. 1083);
6. A greater use of private sector styles of management practices, such as flexible
hiring and rewards systems and advanced marketing techniques; and
7. Greater stress on discipline and parsimony in resource use, i.e., cost management.
With the rise of NPM, in sum, private sector mechanisms—including customer service,
competition, performance-based contracting, market incentives, deregulation, and
decentralization—emerged as dominant management tools in the public sector. In addition, public
managers were pressed to become entrepreneurial leaders who pursue the goals of economy,
efficiency, and effectiveness in the provision of public services (Kaboolian 1998; Terry 1998).
NPM resulted in significant effects that reach far beyond the domain of the public sector. It
engendered “a flow-on effect to the nonprofit sector and forced a fundamental challenge to the
existing roles and identities of professionals in both the public and nonprofit sectors” (Paulsen
2006, p. 17). Indeed, one obvious consequence of the reform movement was the growth of
competitive tendering and contracting for service provision—i.e., the rise of contractualism
(Alford & Hughes 2008). Under this contracting regime, government agencies became purchasers
of services and contract managers, and nonprofit organizations—along with for-profits—became
outsourced providers of direct services that were previously supplied by public sector
organizations (Paulsen 2006). Clearly, NPM changed the established role of nonprofits as
providers of government-funded services by developing government-nonprofit contractual
relationships. Under NPM, nonprofits were increasingly part of an experimentation with new
17
contracting models and came to be viewed as co-opted agents in the development and
implementation of public policy (Anheier 2009).
Accordingly, public management research generated robust scholarly literature that focused
on the nature of and the variation in the government-nonprofit contractual relationship (e.g., Dicke
& Ott 1999; Van Slyke 2007; Witesman & Fernandez 2013), the process by which the contractual
relationship is established and sustained (e.g., Ferris & Graddy 1998; Hefetz & Warner 2011), and
the efficiency and effectiveness of the use of nonprofit contractors for public service provision
(e.g., Van Slyke 2003; Van Slyke & Roch 2004). For instance, Witesman and Fernandez (2013)
examined whether systematic differences exist between government contracts with nonprofits and
for-profits at the local level and suggested that nonprofit contractors tend to be more trusted, to be
granted greater discretion, are less frequently monitored, are awarded longer contracts, and are
more preferred for services characterized by higher levels of task uncertainty than for-profit
contractors. Ferris and Graddy (1998) developed a model of the decision to contract out services.
Their empirical findings showed that local governments are likely to choose to contract with
nonprofit organizations—rather than other governments or for-profits—for services that are more
collective (as opposed to individual), that are related to health/human and recreation/arts fields,
and that have historically been provided by the nonprofit sector. Van Slyke (2003) examined
whether privatization and contracting for social services with nonprofit organizations meet the
expectations of cost savings and quality improvements. He found that in the context of social
services, the government-nonprofit contracting relationship may exist for politically symbolic
reasons rather than economic benefits associated with competition.
18
The Post-NPM Movements and the Role of Nonprofits as Partners
In recent years, research on the relationship between public agencies and nonprofits has
focused on collaboration, reflecting broader trends in the field. The emergence of additional
management perspectives in the public sector—New Public Governance (NPG) and New Public
Service (NPS)—developed as potential solutions to the perceived problems of NPM. NPM has
been commonly criticized for its adherence to the application of outdated market models to public
sector action, for its conception of citizens as individual customers and discrete users of public
services, for its downgrading of democratic values, and for its incomplete and distorted view of
human beings drawn from neoinstitutional economics (Box et al. 2001; Hood 1991; Terry 1998;
Osborne 2010).
More specifically, New Public Governance emphasizes collaborative, networked
arrangements that are designed to provide services based on the belief that a form of trust-based
governance can alleviate the transaction and agency costs of contracting out (Alford & Hughes
2008). In contrast to NPM, the movement has largely been influenced by the assumptions of
sociological and network theories. Osborne (2010, p. 9) characterized its main ideas as follows:
It posits both a plural state, where multiple interdependent actors contribute to the
delivery of public services, and a pluralist state, where multiple processes inform the
policy-making system. Drawing upon open natural systems theory, it is concerned with
the institutional and external environmental pressures that enable and constrain public
policy implementation and the delivery of public services within such a plural and
pluralist system. As a consequence of these two forms of plurality, its focus is very
much upon interorganizational relationships and upon the governance of processes,
stressing service effectiveness and outcomes that rely upon the interaction of public
service organizations with their environment.
19
With the rise of NPG, some form of network governance has become the dominant model of
public sector management. Moreover, the managerial emphasis has moved to “the design and
evaluation of enduring interorganizational relationships where trust, relational capital, and
relational contracts act as the core governance mechanisms” (Osborne 2006, p. 384). In the post-
NPM era, the established principal-agent relationship between public agencies and nonprofits
under a contracting regime has been complemented by a principal-steward perspective in which
nonprofit motives align with government goals (Van Slyke 2007). Indeed, nonprofit organizations
have come to play “a stronger role as partners, rather than just contractors of government” (Fosler
2002, p. 18).
Furthermore, nonprofit organizations have come to play an additional role as democratic
leaders to strengthen the accountability relationships between the government and citizens. New
Public Service is rooted in a variety of ideas from different disciplines, such as democratic
citizenship, community and civil society, and organizational humanism and discourse theory
(Denhardt & Denhardt 2000; Light 2001), which highlight “a citizen-inclusive, decentralized
practice of public service through which public authority is exercised in accordance with the
articulated values and needs of a community” (Alexander & Nank 2009, p. 364). Denhardt and
Denhardt (2000, pp. 553-557) summarized the seven principles of NPS:
1. Serve rather than steer—by helping citizens articulate and meet their shared
interests rather than by attempting to control or steer society in new directions;
2. The public interest is the aim, not the by-product—the goal is the creation of shared
interests and shared responsibility;
3. Think strategically, act democratically—through collective efforts and
collaborative processes;
4. Serve citizens, not customers—with a focus on building relationships of trust and
collaboration with and among citizens;
20
5. Accountability is not simple—public servants should attend to statutory and
constitutional law, community values, political norms, professional standards, and
citizen interests;
6. Value people, not just productivity—through processes of collaboration and shared
leadership based on respect for all people; and
7. Value citizenship and public service above entrepreneurship—with the
reconceptualization of the role of public managers as responsible participants.
Under NPS, a concern for democratic values—along with values such as efficiency and
productivity—has grown, and the use of mechanisms such as citizen advisory boards, community
councils, participatory budgeting, and the monitoring of service delivery has become pervasive in
the public sector. These changes, in turn, have strengthened calls for a greater role of nonprofits in
the larger context of democracy, the community, and the public interest. In the post-NPM era, the
nonprofit sector’s social-integrative and participatory function and its contribution to community
building have been greatly encouraged, and nonprofits have become considered “instruments of
greater transparency, heightened accountability, and improved governance of public institutions”
(Anheier 2009, p. 1083).
Reflecting these perspectives in the public management literature, the direction of
collaborative governance research has partly moved toward an emphasis on the interdependence
and continuing trust-based interactions between public agencies and nonprofit organizations. On
the one hand, numerous scholars of public management have examined the nature and extent of
government-nonprofit partnerships (e.g., Acar & Robertson 2004; Brinkerhoff & Brinkerhoff 2002;
Cho & Gillespie 2006; Gazley 2008; Gazley & Brudney 2007; Graddy & Chen 2006) and the
benefits and effectiveness of such partnerships (e.g., Andrews & Entwistle 2010; Chen & Graddy
2010; Gazley 2010; Graddy 2009; Shaw 2003). For example, Gazley (2008) found empirical
21
evidence that supports the existence of non-contractual relationships between local governments
and nonprofits, where funding, cultural norms, shared goals, organizational reputation, and trust
are used as the key mechanisms of public managers. Graddy and Chen (2006) noted that the size
and scope of networks for the delivery of publicly funded social services are importantly
influenced by environmental factors (e.g., the availability of potential partners) and programmatic
needs (e.g., the complexity of services and the diversity of clients). Andrews and Entwistle (2010)
tested the common belief that partnerships with nonprofit organizations may lead to more equitable
service provision. Unexpectedly, they found no statistically significant association between
government-nonprofit partnerships and public service effectiveness, efficiency, and equity, which
casts some doubt on the benefits of partnerships with nonprofit organizations.
On the other hand, public management scholars have explored how democratic governance
practices take shape and evolve in government-nonprofit partnerships and their effect on state-
citizen relationships (e.g., Alexander & Nank 2009; Kapucu 2006; LeRoux 2009). For instance,
Alexander and Nank (2009) showed that a trust-based partnership between government and
community-based nonprofits extends the democratic accountability of public agencies and
functions in the best interests of the marginalized population in the community. LeRoux (2009)
found that government-nonprofit partnerships create more opportunities for citizens to participate
in policy-making processes. The author also stressed the importance of government funding, which
shapes the practices of nonprofits in ways that promote participatory governance.
In short, a series of management reform movements in the public sector over the last few
decades have brought the nonprofit sector to the center of public policy debate and drawn greater
attention to the relationship between public and nonprofit management. As a result, the practice
and management of the nonprofit sector have come to be viewed as an important topic in both
22
public management practice and the literature. This tendency has caused practitioners and scholars
in the field of public management to attempt to gain a better understanding of the nonprofit
sector—its past, its current status, and its future development.
2.2. The Nonprofit Sector and the Process of Structuration
Associations in the United States have a long history, but the formalization of a nonprofit
sector is a relatively recent development. The ideological roots of the sector can be found in the
ancient traditions of charity, philanthropy, and voluntarism, such as the Greco-Roman emphasis
on community, citizenry, and social responsibility and the Judeo-Christian charitable ethos of self-
sacrifice and benevolence (Gies et al. 1990; Robbins 2006). The initial forms of voluntary
associations and philanthropy services were founded in the colonial era, when mutual care and
support through voluntary associations were essential for the survival of the Native American
communities (Hall 2010; Ott 2001). The legal foundations of the sector are also heavily influenced
by English laws passed in the 1600s (e.g., the Statute of Charitable Uses and the Poor Law), which
outlined the legitimate support activities and accountabilities of charitable organizations (Worth
2011).
However, the modern form of the nonprofit sector did not emerge until the early 20
th
century,
and it is only in recent years that the sector has been institutionally defined and structured. To
understand this process I borrow from DiMaggio and Powell (1983, p. 148), especially their
description of how organizational fields emerge and consolidate:
The process of institutional definition, or “structuration,” consists of four parts: an
increase in the extent of interaction among organizations in the field; the emergence of
sharply defined interorganizational structures of domination and patterns of coalition;
an increase in the information load with which organizations in a field must contend;
23
and the development of a mutual awareness among participants in a set of organizations
that they are involved in a common enterprise.
The notion of structuration denotes “[the] process of gradual maturity and specification of
roles, behaviors, and interactions of organizational communities” (Greenwood et al. 2002, p. 59).
Drawing on these neoinstitutional conceptions, I argue that the process of structuration in the
nonprofit sector is well underway. In particular, three broad trends, which not only indicate but
also facilitate more formal structuration in the sector, draw our attention: the growth of the
nonprofit sector in size and scope, a profound shift toward organizational rationalization in the
nonprofit sector, and the sector’s growing professionalization.
The Growth of the Nonprofit Sector
The nonprofit sector in the United States has grown significantly in both size and absolute
importance in the past few decades. Recent data estimate that as of 2012, the sector included
approximately 1.44 million organizations, an increase of 475 percent from 1975 when there were
approximately 0.25 million organizations; it contributed roughly $887.3 billion to the U.S.
economy, composing 5.4 percent of the country’s gross domestic product; and it encouraged 26.5
percent of U.S. adults (an estimated 64.5 million) to volunteer (McKeever & Pettijohn 2014). The
sector’s exact size and scope are virtually unknown because a large portion of the sector that is not
required to register with the IRS (e.g., religious organizations) is excluded from the data, and the
available data on the formally registered organizations are imperfect (O’Neill 1989; Salamon
2012). However, several different indicators have shown that there has been a continuous and
significant growth in the sector, at least since the 1970s.
24
Figure 2.1 shows a continuous increase in the number of 501(c)(3) charitable organizations
registered with the IRS. Over the past four decades, the number of formally registered nonprofit
organizations has grown by approximately 400 percent—from an estimated 0.26 million
organizations in 1976 to 1.28 million in 2010 (IRS 2011).
Figure 2.1. Number of 501(c)(3) Charitable Organizations Registered with the Internal Revenue Service,
1976-2010
Source: IRS Annual Report, 1976-1999; IRS Data Book, 2000-2010
Furthermore, an increasing number of nonprofit organizations have been engaged in the
provision of a wide variety of public services, thus expanding the scope of the sector’s activities.
Table 2.1 displays the number of registered charitable operating nonprofits by type of service.
Nonprofit organizations whose activities focus on the service fields of International & Foreign
Affairs, Public & Social Benefit, and Environment & Animals have grown the most rapidly, with
increases of 529, 425, and 395 percent from 1989 to 2010, respectively. Overall, the number of
nonprofit organizations whose primary purpose is to influence public policy within the eight major
0.26
1.28
0
0.2
0.4
0.6
0.8
1
1.2
1.4
Numbers in millions
25
service fields increased by 200 percent between 1989 and 2010, from 0.12 million to more than
0.36 million (Blackwood et al. 2008; Blackwood et al. 2012; Boris & Steuerle 2006).
Table 2.1. Number of Charitable Nonprofits by Type of Service, 1989-2010
Type 1989 1995 2000 2005 2010 Change 1989-2010
% change
Arts & Culture 13,817 21,277 27,302 34,514 39,536 25,719 186.14
Education 16,939 30,509 42,591 56,121 66,769 49,830 294.17
Environment & Animals 3,305 6,088 8,976 12,754 16,383 13,078 395.70
Health 23,039 32,289 36,057 40,638 44,128 21,089 91.54
Human Services 45,156 63,528 86,068 106,248 124,360 79,204 175.40
Int’l & Foreign Affairs 1,196 2,471 4,345 5,726 7,533 6,337 529.85
Public & Social Benefit 8,352 21,634 30,521 38,379 43,875 35,523 425.32
Religion 5,764 9,242 13,999 18,784 23,502 17,738 307.74
Total 123,687 187,038 249,859 313,164 366,086 242,399 195.98
Source: Boris and Steuerle, 2006, p. 9, “Table I.3. Growth in Nonprofit Organizations by Type of Service,
1989-2004”; Blackwood et al., 2008, p. 4, “Table 2. Number and Financial Scope of Reporting Public
Charities by Subsector, 2005”; Blackwood et al., 2012, p. 5, “Table 3. Number, Revenue, and Assets of
Reporting Public Charities by Subsector, 2000–2010”
Note: Only operating public charities that engage in the provision of services and products as well as the
production of information for the public and their members are included.
The nonprofit sector has also become a major part of the U.S. economy in terms of finance
and employment. The sector was estimated to contribute $887.3 billion to the U.S. economy in
2012, which accounted for nearly 5.4 percent of the U.S. gross domestic product. More specifically,
approximately 0.5 million reporting nonprofits—which are required to file financial data with the
IRS and comprise only 35 percent of the total registered nonprofits—had $2.16 trillion in revenue,
$2.03 trillion in expenses, and $4.84 trillion in assets (McKeever & Pettijohn 2014).
Figure 2.2 shows the overall financial growth of the sector from 1985 to 2008. During this
period, both the revenues and the expenses of reporting public charities and private foundations
26
grew at an increasing rate. Although revenues and expenses stayed at roughly similar proportions,
revenues grew slightly faster than expenses in the last decade of the period. Revenues outpaced
expenses during most of the period, but not in 2008, when the economic recession resulted in a
large decline in charitable giving (IRS 2012).
Figure 2.2. Trends in Revenues and Expenses of Reporting Public Charities and Private Foundations, 1985-
2008
Source: IRS, 2012, “Historical Table 16 (Expanded Version). Nonprofit Charitable Organization and
Domestic Private Foundation Information Returns, and Exempt Organization Business Income Tax Returns:
Selected Financial Data, 1985-2008”
Note: Reporting public charities include only organizations that both reported under Internal Revenue Code
section 501(c)(3)—excluding private foundations and most religious organizations—and had gross receipts
over $25,000. All amounts are in current dollars and are not adjusted for inflation.
The nonprofit sector also contributes substantial wages and employees to the U.S. economy.
In 2010, nonprofits paid $587.7 billion in wages and employed 13.7 million people, which was
equivalent to 9.2 percent of all wages and salaries paid in the United States and approximately 9
percent of the U.S. workforce. Although overall employment in the U.S. economy has decreased
1378.27
1396.37
49.67
60.32
0
200
400
600
800
1000
1200
1400
1600
1985 1988 1991 1994 1997 2000 2003 2006
Public charities (Total revenue)
Public charities (Total expenses)
Private foundations (Total revenue)
Private foundations (Total expenses)
Dollars in Billions
2008
27
since the economic recession of 2008, nonprofit employment and wages have continued to increase
and have even grown faster than those of government and for-profits. Indeed, employment and
wages in the nonprofit sector increased by 4 percent and 6.5 percent, respectively, from 2007 to
2010. However, the government sector reported an increase of only 1 percent in employment and
4.8 percent in wages, while the private sector reported a decrease of 8.4 percent in employment
and 8 percent in wages (Roeger et al. 2012).
Figure 2.3. Employment in the Nonprofit Sector and Selected Major U.S. Industries, 1998 and 2006
Source: Salamon, 2003, p. 9, “Figure 2-1. Nonprofit Employment in Relation to Employment in Major U.S.
Industries, 1998”; Salamon, 2012, p. 8, “Figure 1-2. Employment, Nonprofit Sector and Selected Industries,
2006”
As shown in Figure 2.3, the nonprofit workforce has become one of the three largest when
contrasted with major U.S. industries. Nonprofit organizations employed nearly 11 million paid
workers in 1998 and 13.5 million in 2006, which made the nonprofit paid workforce the second
largest of any U.S. industry in 1998 and the third largest in 2006, followed by the construction (8.5
and 7.6 million, respectively), finance/insurance/real estate (8.6 and 10.2 million, respectively),
16.6
20.5
5.1
8.5
8.6
3.4
18
14.1
15.4
7.6
10.2
1.2
0 5 10 15 20 25
Nonprofit Sector
Manufacturing
Retail trade
Construction
Finance, Insurance &
Real estate
Agriculture
1998 2006
Workers (Unit: Millions of full-time equivalent paid)
(5.7) (10.9)
(13.5) (4.5)
28
and agriculture industries (3.4 and 1.2 million, respectively). However, nonprofits also employed
the full-time equivalent of 5.7 million volunteers in 1998 and 4.5 million in 2006, which boosted
the total number of employees in the sector to 16.6 million in 1998 and 18 million in 2006. With
both paid workers and volunteers included, the nonprofit full-time equivalent workforce became
the largest when compared to U.S. industries in 2006, even outnumbering the workforces of retail
trade (15.4 million) and manufacturing (14.1 million) (Salamon 2003, 2012).
In addition to these economic contributions, nonprofits have significantly influenced
individuals and communities as a major organizing force, which has encouraged individuals to
actively participate in public affairs and volunteer for their communities. Table 2.2 illustrates that
almost 27 percent of U.S. adults (64.5 million individuals) volunteered at least once for or through
a nonprofit organization in 2012. They spent an average of 2.5 hours per day and an estimated total
of 13 billion hours during the year volunteering, which is the equivalent of 7.7 million full-time
employees and is worth nearly $260 billion in average private wages (Pettijohn 2013). Overall,
from 2002 to 2012, the number of volunteers and volunteering hours slightly increased.
Table 2.2. Trends in Volunteering, 2002–2012
2002 2004 2006 2008 2010 2012
Number and Hours
% of population volunteering 27.4 28.8 26.7 26.4 26.3 26.5
Number of volunteers (millions) 59.8 64.5 61.2 61.8 62.8 64.5
Total hours volunteered (billions) - 13.7 12.9 14.4 14.9 12.7
Hours per day per volunteer - 2.4 2.31 2.43 2.46 2.48
Value of volunteers
Full-time equivalent employment (millions) - 8.1 7.6 8.4 8.8 7.7
Assigned value of volunteer time (billion dollars) - 215.4 215.6 270.2 283.8 259.6
Source: Blackwood et al., 2008, p. 7, “Table 5. Number, Hours, and Dollar Value of Volunteers, 2002–
2006”; Pettijohn, 2013, p. 7, "Table 6. Number, Hours, and Dollar Value of Volunteers, 2006–12"
29
The Rationalization and Professionalization of the Nonprofit Sector
The nonprofit sector has rapidly grown not only in size but also in its economic and social
impact in the past half century, thus creating new pressures (Boris & Steuerle 2006; Frumkin 2002;
Young et al. 2012). More concretely, the sector has faced competitive pressures to secure financial
resources and expand its market share; the growing number of nonprofits has contributed to greater
competition for government funding, and the growing market for nonprofit services has attracted
for-profits to many traditionally nonprofit fields. Moreover, the sector has encountered political
pressures to improve its efficiency and accountability; the growing importance of the sector has
expanded the expectations of measurable performance on the part of key stakeholders, including
the government, foundations, and individual donors. These pressures, combined with the rise of
venture philanthropy and social entrepreneurship (Dees et al. 2001; Eikenberry & Kluver 2004),
have induced nonprofits to become more attuned to market ideologies and practices, which in turn
has been associated with a profound shift toward organizational rationalization and
professionalization in the nonprofit sector (Hwang & Powell 2009; Suárez 2011).
The rationalization of the nonprofit sector refers to “the integration of formalized roles and
rules around unified sovereignty, entailing the construction of nonprofits as actors with clear
identities” (Hwang & Powell 2009, p. 272). Institutional sociologists argue that rationalization
entails “the creation of cultural schemes defining means-ends relationships and standardizing
systems of control over activities and actors” (Scott & Meyer 1994, p. 3) and “the creation of
entities—identifiable social units—endowed with interests and having the capacity to take action”
(Scott 2008, p. 74). In other words, the concept of rationalization involves a great deal of activity,
such as “(1) continuing efforts to systematize social life around cultural schemes that explicitly
differentiate and then seek to link social means and social ends…; and (2) efforts to reconstruct all
30
social organization … as a means for the pursuit of collective purposes, these purposes themselves
subject to increasing simplification and systematization” (Jepperson 2002, p. 63). Therefore,
rationalization around the nonprofit sector compels a sharpened conception of nonprofits and a
delineation of the dimensions in which nonprofits’ interests and activities are to be sought, thus
enhancing the use of a broadening array of tools and routines (Scott & Meyer 1994). Some of the
actions associated with ongoing rationalization include developing strategic plans, using
quantitative performance measurements, performing financial audits, and hiring consultants
(Hwang & Powell 2009; Suárez 2011).
In addition to organizational rationalization, the nonprofit sector has been shifting from an
informal “amateur” orientation to a professional orientation over the last few decades. The
sociological literature on the professions (e.g., Abbott 1988, 1991; Caplow 1954; Wilensky 1964)
has observed that the process of professionalization involves several elements: “(1) the beginning
of full-time work; (2) the establishment of a training school; (3) the formation of a professional
association; (4) the protection of jurisdiction through state-sanctioned licensing; and (5) the
development of a formal code of ethics” (Hwang & Powell 2009, p. 273). A number of these
aspects of professionalization are indeed identified for nonprofits: the full-time, paid staff of
nonprofit organizations has grown at an increasing rate since the late 1970s (Salamon 2012);
nonprofit management training programs and higher education programs have exploded
(Mirabella 2007, 2014); scholarly professional associations and dedicated journals have been
established (Graddy et al. 2011; Shier & Handy 2014); and the adoption of codes and other
governance standards by nonprofits has increased (Bromley & Orchard, forthcoming). These
indicators all imply an increased professionalization of the sector, in which “the increasing
31
presence of specialized expertise in an organization and … shifts from volunteer labor to paid staff”
(Suárez 2011, p. 311) are observed.
Several studies have documented and explored these ongoing trends, which both indicate
and facilitate more formal structuration in the nonprofit sector. One group of studies has examined
the nature and extent of the sector’s rationalization and professionalization. For instance, Berman
(1999) compared the professional orientations among managers in local governments, nonprofit
social service organizations, and museums. He found that although public and nonprofit managers
have similar levels of orientation toward professionalism, nonprofit managers are more influenced
by organizational cultures and are less affected by participation in professional training and
development activities than public managers. Based on a meta-analysis of 65 empirical studies,
Stone et al. (1999) summarized the results of studies on strategic management in nonprofit
organizations. The authors reported that many nonprofits are more likely to rely on operational
and informal planning methods rather than formal strategic planning. Strategy formulation, content,
and implementation vary across nonprofit organizations, depending on their organizational size,
the characteristics of the board and management, prior agreements on organizational goals,
resource environments, and existing funder relationships.
Finally, the primary outcomes of strategic management are changes in the organizational
mission, structure, and board and management roles; meanwhile, the relationship between strategic
management and performance is not clear because little scholarly attention has been paid to
performance issues. Drawing on a random sample of San Francisco Bay Area 501(c)(3) operating
charities, Hwang and Powell (2009) maintained that positive relationships exist between higher
levels of organizational rationalization—e.g., the use of strategic planning, independent financial
32
audits, quantitative outcome evaluations, and consultants—and charities managed by paid, full-
time personnel and staffs who participate in professional training and education.
A second group of studies has considered the organizational consequences of the sector’s
rationalization and professionalization. Pearce (1993) and Kreutzer and Jäger (2011) focused on
the tension between paid employees and volunteers and its effect on nonprofits’ identity and
culture. They emphasized significant differences in motivations, commitments, expectations, and
perceptions of organizational identity between paid staff and volunteers and warned against
potential intraorganizational conflicts. Eikenberry and Kluver (2004) reviewed the major
professionalization and commercialization trends—including commercial revenue generation,
contract competition, and social entrepreneurship—occurring within the nonprofit sector. They
argued that nonprofits receive several benefits from marketization trends, such as more reliable
funding streams, greater managerial efficiency and innovation, better customer-oriented services,
increased legitimacy, and greater accountability. However, they also stressed potential harmful
effects on the nonprofit sector’s role in creating and maintaining a strong civil society. Suárez
(2011) found that professionalized nonprofits that rely a great deal on specialized experts and paid
staffs are more likely to receive government funding than other nonprofits, whereas the empirical
results did not find that rationalized nonprofits that use formalized tools such as performance
evaluations, strategic planning, audits, and consultants are more likely to acquire government
funding.
2.3. Sociological Institutionalism and Nonprofit Management Research and Education
As one aspect of the structuration of the nonprofit sector—more specifically, as an element
of professionalization in the nonprofit sector—we have witnessed the wide and rapid expansion of
33
nonprofit management research and education in the higher education context. This phenomenon,
the emergence and development of nonprofit management studies programs within university
organizations, can be explained by several approaches in organizational theory.
On the one hand, public management research often discusses the routines and work practices
of organizations in relation to outcomes such as efficiency and effectiveness. From this perspective,
innovation or reform occurs because of functional needs and demands. For example, resource
dependency theory argues that organizational changes are the result of a managerial response to
external dependencies and uncertainties in the resource environment (Pfeffer 1982; Pfeffer &
Salancik 1978). In the case of nonprofit management research and education, the argument is that
university organizations may strategically incorporate and develop nonprofit management
programs to acquire critical resources from their environment (e.g., funding or academic reputation)
and to reduce uncertainty. Another argument can be made based on contingency theory. In this
view, organizational behaviors are rationally and strategically oriented and are dependent on
internal and external resource environments (Thompson 1967). Thus, contingency theory may
assume the possibility for variation in the diffusion of nonprofit management programs across
different university organizations, thus reflecting the distinct goals and situations of each
organization.
On the other hand, research increasingly considers organizational practices from the lens of
neoinstitutional theory, which focuses more on the role of the norms and social expectations of the
institutional environment. Within this tradition, institutions are viewed as “comprised of regulative,
normative and cultural-cognitive elements that, together with associated activities and resources,
provide stability and meaning to social life” (Scott 2008, p. 48). That is, institutions are constituted
of three pillars—normative, regulative, and cultural-cognitive elements—and institutional
34
developments and changes in the three domains not only constrain but also enable organizational
practices (DiMaggio & Powell 1983). Sociological institutionalism consequently reasons that
organizations conform to the cultural belief systems operating in their environment as a way to
signal their legitimacy rather than just as a way to improve technical rationality or efficiency
(DiMaggio & Powell 1983; Meyer & Rowan 1977; Meyer & Scott 1992; Scott 2008).
Building on sociological institutionalism, the emergence and development of nonprofit
management research and education in the higher education context cannot be explained solely by
operational efficiency, technical rationality, or functional needs. Rather, it can be explained by the
results of the interplay between higher education organizations and the broader institutional
environment. In other words, the expansion of nonprofit management research and education can
be conceptualized in relation to developments and changes in the institutional environment—the
structuration of the nonprofit sector—by which university organizations are simultaneously
constrained and empowered.
More specifically, to begin with, it could be argued that institutional changes in the regulative
domain have arisen from a series of tax and policy reforms related to nonprofit management, which
have encouraged university organizations to incorporate nonprofit management discourse into
their systems. In general, the nation-state possesses coercive and regulatory powers for introducing
innovations, developing shared meanings and understandings, and adjudicating conflicting claims
for agreement (DiMaggio & Powell 1983; Greenwood et al. 2002; Ruef & Scott 1998; Scott &
Meyer 1994). In the United States, governmental bodies have established rules backed by sanctions,
incentives, and moral persuasion for the diffusion of nonprofit management ideas and practices
over the last few decades. For instance, the 1969 Tax Reform Act and the Sarbanes-Oxley Act of
2002 have contributed to an interest in financial management and the growth of financial experts
35
in the nonprofit sector as a way to comply with complex filing, reporting, and auditing provisions
(Salamon & Flaherty 1996; Scrivner 2001). In addition, reaching a certain professional staffing
level or acquiring professional licenses has often been dictated by government grant and contract
requirements (Frumkin 2002; Smith & Lipsky 1993). Such newly imposed legal and regulative
rules seem to have provided university actors with clear prescriptions to support the
professionalization of the nonprofit sector, which in turn may have given rise to the expansion of
nonprofit management studies programs in universities.
Second, efforts to create a distinct profession from nonprofit management as a normative
force also may influence the emergence and development of university-based academic nonprofit
management programs. Professions and occupational communities—in which members of an
occupation collectively struggle to define what knowledge and practices are relevant to their
occupational tasks (DiMaggio & Powell 1983; Khurana 2007)—normally form an allied source of
new ideas and practices and thus facilitate particular lines of innovative action (Strang & Soule
1998). Not surprisingly, communities of nonprofit management experts—including networks of
state and regional nonprofits (e.g., the Independent Sector and the National Council of Nonprofits)
and management support associations (e.g., the Alliance for Nonprofit Management) in the United
States—have constructed and distributed principles and guidelines for best managerial practices
and have built flow channels based on professional network ties and commitments. These shared
occupational understandings and well-developed flow channels may have provided university
actors with powerful inducements to develop knowledge and practices for nonprofit management,
which in turn may have provided an impetus for the diffusion of nonprofit management programs
across university campuses.
36
Finally, increased nonprofit management discourse in American society can be seen as a
cultural-cognitive force that may have contributed to the formation of a powerful institutional
framework that affects the behavior of university actors. In fact, nonprofit management has
become increasingly self-evident in American society over time. Mass media outlets such as
newspapers, magazines, books, TV, and radio have drawn nationwide attention to the
professionalization of the nonprofit sector. In addition, a significant amount of information on
nonprofit organizations and their management has been broadcast to the public through a variety
of scholarly books, journals, and magazines on nonprofit management (Katz 1999; O’Neill &
Fletcher 1998; Smith 1999). Nonprofit management has also become more central to academia
and has developed as a distinct study field; identified research communities (e.g., the Association
for Research on Nonprofit Organizations and Voluntary Action and the International Society for
Third-Sector Research) have started holding research workshops and conferences (O’Neill &
Fletcher 1998; O’Neill & Young 1988). This variety of evidence demonstrates the increasing
interest and organizing around nonprofit management and furthers the existence of rationalized
beliefs about it. By responding to such institutional changes in the cultural-cognitive domain,
universities may have developed more integrated and relevant academic programs within their
systems.
On the whole, I argue that higher education organizations not only have been responsive to
general institutional changes in society—the structuration of the nonprofit sector—but also have
actively engaged in the institutional process by trying to collect, conceptualize, and diffuse relevant
knowledge and practices. I suggest that this entire dynamic consequently constructs and reinforces
nonprofit management research and education in the United States.
37
2.4. Discussion
This chapter has presented a theoretical framework for understanding nonprofit management
education by linking public management practice to the ongoing structuration of the nonprofit
sector. To begin with, this chapter illustrated the changing relationship between government and
nonprofit organizations and identified several reform movements in the public sector—NPM, NPG,
and NPS—as driving forces of change. I argued that these reform movements have brought the
nonprofit sector—as a contractor or partner for service provision and a democratic leader for
improving social accountability—to the center of public policy debates. As a result, the changes
and developments in the nonprofit sector have become an important issue for both public
management practitioners and scholars.
This chapter then moved the focus to the nonprofit sector itself by documenting the notable
trends in recent years. As illustrated, the nonprofit sector seems to be institutionalized, similar to
the public and private sectors. Indeed, increased structuration in the sector has been empirically
observed based on several indicators. First, the sector has grown significantly in both size and
absolute importance in the past few decades. Second, the sector has employed rule-like tools and
routines, thus rationalizing its organizations. Third, the sector has been transformed from amateur
into professional by hiring well-educated, paid staff and infusing its management with specialized
ideas and practices.
Finally, this chapter discussed the effect of such structuration on the expansion of nonprofit
management research and education in the higher education context. Drawing on ideas from
sociological institutionalism—which has not received sustained attention in public management—
I argued that the diffusion of academic programs could be the result of the increased structuration
in the sector and developments within higher education organizations themselves. In other words,
38
the process of structuration in the nonprofit sector has been accompanied by a series of institutional
changes in the regulative, normative, and cultural-cognitive domains, which in turn seem to have
guided university organizations to the development of academic programs.
In the following chapters, this conceptual framework is illustrated through empirical studies
on the diffusion of nonprofit management studies programs in higher education institutions.
Chapters 3, 4, and 5 assess the importance of these macro-level (national) institutional processes
while also considering meso-level (state) and organizational-level (university) conditions. More
specifically, Chapter 3 explores how the macro-, meso-, and micro-level factors have contributed
to the expansion of nonprofit management programs. Chapter 4 examines the variations, over
space and time, in the macro-, meso-, and micro-level conditions and how such variations account
for the different developmental processes of nonprofit management programs. Chapter 5 digs
deeper into the dynamics among the macro-, meso-, and micro-level factors and investigates how
these dynamics affect the institutionalization of the study field of nonprofit management in
universities.
39
CHAPTER 3. THE EXPANSION OF NONPROFIT MANAGEMENT
RESEARCH AND EDUCATION
2
“If somebody asked me in 1983, ‘How many programs do you think will be of this kind
30 years from now?’ I might have said, ‘Well, I hope there would be 10 or 15 programs.’
Apparently, there are now more than 200 programs. The rapid growth of nonprofit
management [research and education] programs is just amazing to me.”
- Interview with faculty, July 2014
3.1. Background
University-based research and education programs in nonprofit management have emerged
and diffused at an increasing rate during the past few decades (Mirabella 2007; Mirabella & Wish
2001; O’Neill & Fletcher 1998; O’Neill & Young 1988). Even though the field was virtually
nonexistent before the late 1970s, academic programs—in the form of regular degrees,
concentrations, certifications, internships, and workshops—have become widely available in U.S.
higher education institutions, beginning with the most prominent research program, the Program
on Nonprofit Organizations at Yale University founded in 1978 (O’Neill 2005; Young 1988). The
growth of academic interest in nonprofit management is conventionally seen as responding to a
broader social and cultural atmosphere emerging after World War II, particularly through the
managerial revolution and remarkable development of the nonprofit sector. For example, studies
(e.g., Hall 1996; O’Neill 2005; Young 1999) claim that the advent and evolution of academic
programs was inspired by a series of historical events such as the rise of professionalism and
university-based education, the increasing emphasis on generic management skills and knowledge,
2
A previous version of this work was presented at the 43rd Annual Conference of the Association for Research on
Nonprofit Organizations and Voluntary Action, Denver, CO (November 2014).
40
and the explosive growth of nonprofit organizations. Others, like Hall (1992) and Young (1998),
view this phenomenon within the organizational context of higher education institutions; they
highlight the strong willingness of university system and various stakeholders—including
administrators, deans and host faculties, students and alumni, and external funders—as a key
driving force for the actual development of academic programs. Taken together, previous research
has identified several factors associated with the expansion of nonprofit management studies
programs, but to date no research has investigated the relative influence of these factors in one
empirical analysis.
This chapter thus aims to identify a variety factors that accelerated the wide and rapid
expansion of nonprofit management studies programs across the country. More ambitiously, it
aims to develop an integrative framework by bringing together historical, social, and
organizational factors that have not been discussed under one cohesive, conceptual frame. I
develop three primary lines of argument for understanding the emergence and growth of nonprofit
management education: (1) the structuration of the nonprofit sector at the macro-level, measured
over time as the number of nonprofits and professional associations, tax and policy reforms, and
nonprofit publications and news articles; (2) socioeconomic and political contexts surrounding the
universities at the meso-level, with state-level measures like demographic setting, economic and
social climate, and political orientations; and (3) university-specific organizational conditions and
dynamics at the local-level, such as university size, mission, history, location, prestige, diversity,
and departmental politics. The study also pays close attention to historical context as a relevant
factor in explaining the rate and extent of the expansion of programs.
The next section elaborates this conceptual framework, suggesting a series of hypotheses.
Subsequent sections describe the methods and empirically test the hypotheses using event history
41
data on the creation of nonprofit management studies programs in 1,451 universities and colleges
from 1971 to 2011, and then the chapter concludes by discussing the implications of the findings.
3.2. A Conceptual Model: Mechanisms of the Expansion
The Structuration of the Nonprofit Sector
Scholars working within the sociological neoinstitutional framework (e.g., March & Olsen
1989; March 1994; Powell & DiMaggio 1991; Scott & Meyer 1994) suggest that organizational
practices are enabled and constrained by an institutional environment that imposes shared
structures or practices on individual organizations within the relevant environment. From this
perspective, individual organizations conform to institutional logics—socially constructed,
historical patterns of values, beliefs, practices, and rules—because they provide legitimacy and
serve as guides for action (DiMaggio & Powell 1983; Meyer & Rowan 1977).
Aligning with this perspective, I argue that the structuration of the nonprofit sector
accelerates the diffusion of nonprofit management programs at universities by providing social
consensus regarding the overall utility and validity of programs. As reviewed in Chapter 2, the
nonprofit sector in the United States has been increasingly structured in recent years. Indicators of
structuration include a burgeoning number of nonprofits and professional associations, tax and
policy reforms, nonprofit management publications and news articles, and citizen memberships in
nonprofits (O’Neill 1998). All of these phenomena may create a shared belief that legitimizes
nonprofit management as a taken-for-granted institutional logic of society. This in turn may lead
to the expansion of nonprofit management programs across university campuses because
university are organizations that are subject to cultural belief systems in a society. Taken together,
it can be assumed that the structuration of the nonprofit sector confers regulative, normative, and
42
cultural-cognitive legitimacy—which is an institutional and cultural resource for diffusion (Scott
2008), so that nonprofit management programs become less controversial and more acceptable in
universities and colleges:
Hypothesis 1: The structuration of the nonprofit sector at the national level will have a
positive effect on the adoption rate of nonprofit management research and education
programs in universities and colleges.
Socioeconomic and Political Contexts
Though my core argument is that the structuration of the nonprofit sector in the United States
drives the rate at which nonprofit management programs are founded, the institutional environment
for universities is not limited to the national context. Universities are embedded within states and
other more local environments; I argue that these meso-level contexts could contribute to the
emergence of nonprofit management in universities. In other words, the regional properties of a
state or city in which a university is located may shape not only the university’s willingness and
capability for adoption but also the feasibility and benefits of adoption. Therefore, region-specific
socioeconomic and political conditions would be expected to either facilitate or constrain the
adoption of programs in universities and colleges.
To begin with, population heterogeneity in a state appears to be an important consideration
related to the founding of nonprofit management programs at universities in the state. Economic
theories of market and government failure (Hansmann 1987; Steinberg 2006; Weisbrod 1975)
argue that population heterogeneity—including age, gender, racial, ethnic, and religious
diversity—increases the demand for goods and services that cannot be adequately filled by the for-
profit or government sectors, but can be mainly met by the nonprofit sector; indeed, a number of
empirical studies (e.g., Ben-Ner & Van Hoomissen 1992; Grønbjerg & Paarlberg 2001; Matsunaga
43
& Yamauchi 2004) have found some evidence of a positive relationship between population
diversity and the size and scope of nonprofit service fields. According to this view, population
heterogeneity might result in bigger role of nonprofits in a region, which in turn could lead to more
interest in and demand for research and education programs targeting nonprofit managers in the
region. In other words, the more heterogeneous a state, the greater would be the concern by
universities that new curriculum will be needed to better equip nonprofit managers in dealing with
complex and diverse demands of the state:
Hypothesis 2a. Population heterogeneity in a state will have a positive effect on the adoption
rate of nonprofit management research and education programs in universities and colleges.
In addition to population heterogeneity, state support for social services may also affect the
motivation of universities and colleges to create research and education programs. This expectation
is driven by the idea of supply-side perspectives (James 1987; Salamon 1987) that more
government social welfare spending increases the supply of human and financial resources that are
needed for the survival and growth of the nonprofit sector; in fact, earlier empirical studies (e.g.,
(Ben-Ner & Van Hoomissen 1992; Corbin 1999; Grønbjerg & Paarlberg 2001; Lecy & Van Slyke
2013; Salamon et al. 2000) have suggested that more government spending on social services in a
region tend to be positively associated with the size of nonprofits that serve in the region. Viewed
in this light, it would make sense that the more regional resources stimulate the growth of the
nonprofit sector, the more regional needs for a better understanding of nonprofit activities and its
population are prevalent, and the regional needs facilitate the creation of programs at universities.
Such an idea leads to the following hypothesis:
Hypothesis 2b. Support for social services in a state will have a positive effect on the adoption
rate of nonprofit management research and education programs in universities and colleges.
44
Lastly, political liberalism in a state seems to facilitate the creation of academic programs at
universities located in the state. Previous diffusion studies note that innovation is often
ideologically driven (Bakardjieva 1992; Jensen 2003; Walker 1969). Brint et al. (2009), especially,
find that new study fields and innovative academic programs are more likely to be welcomed by
political liberalism. Relying on such an argument, it can be argued that whether a university is
located in an ideologically liberal region might be related with greater receptivity to nonprofit
management studies programs. Put differently, I expect that a more liberal climate in a state will
speed the establishment of university-based programs in the state, because creating new academic
programs for the emerging nonprofit sector cohere better with a liberal ideology. Taking into
account the impact of political climate on adoption rate, the following hypothesis is derived:
Hypothesis 2c. Political liberalism in a state will have a positive effect on the adoption rate
of nonprofit management research and education programs in universities and colleges.
University-Specific Conditions and Dynamics
Universities, as both producers and carriers of nonprofit management discourse, could shape
and reshape academic programs in order to meet the needs of wider society and their own
preferences and interests. Departing from institutional theory, which emphasizes the external
environment, university characteristics such as mission, history, location, size, prestige, diversity,
and internal politics also could influence the rate of adoption. Organizational change and diffusion
literatures indeed emphasize that adopter characteristics and internal dynamics are closely
associated with the adoption of new ideas and practices (Greenwood & Hinings 1996; Oliver 1991;
Soule 1997; Wejnert 2002).
45
A first factor that can be suggested as a determinant of the program adoption is the centrality
of some academic programs—including public affairs, business, and social work—in a university.
Sociological studies on the construction of new disciplines (e.g., Camic 1995; Small 1999; Wood
2012) maintain that different local conditions in and politics among key departments and schools
at universities affect the formation and institutionalization of new disciplines. The adoption of
nonprofit management programs within a university seems to be an outcome of internal politics,
through which various receptor sites (Frank et al. 2000) of nonprofit management—disciplines
that are able to receive and interpret the higher-order social logics of nonprofit discourse and
transmit the information to local university stakeholders—compete to adopt programs on their turf.
Given the findings of earlier studies that the study of nonprofit management has been closely
linked to public affairs, business, and social work (Mirabella 2007; Mirabella & Wish 2000, 2001;
Wish 1993), those study fields may act as receptor sites for macro nonprofit management agenda.
Hence, it is plausible to assume that public affairs, social work, and business schools in universities
serve as a context for the emergence of nonprofit management. Especially, public affairs, social
work, and business schools that have more power and resources available would be more likely to
support the founding of academic programs. Presented more formally:
Hypothesis 3a. The centrality of specific academic programs (public affairs, social work, and
business) in a university will have a positive effect on the adoption rate of nonprofit
management research and education programs in universities and colleges.
A second factor that may be linked to the adoption of programs is campus diversity. Scholars
note that the heterogeneous faculty and student composition on campus has a positive influence
on innovation; for example, the diversity of staff and student body increases the chance of adopting
interdisciplinary programs such as African American studies, race and ethnic studies,
46
environmental studies, and women’s studies (Brint et al. 2009; Olzak & Kangas 2008; Rojas 2006).
This line of argument explains curricular innovation and organizational change as an outcome of
the efforts of a constituency—with an assumption that faculties and students who are minorities
or women will benefit from such new programs and changes, and thus they are most likely to
support innovation as the constituency. By this reasoning, it is expected that campus diversity
would account for a greater probability of adopting nonprofit management programs. Because
nonprofit management studies encompasses the social issues involving minorities, marginalized
populations, and altruistic behaviors, the proportion of minorities and women on campus could
influence adoption. This idea suggests the following hypothesis:
Hypothesis 3b. Greater representation of minorities and women at a university will have a
positive effect on the adoption rate of nonprofit management research and education
programs in universities and colleges.
Historical Context
Historical context should be taken into consideration along with other factors, since it could
capture changing trends over time—for example, a shift in the effects of institutional logics and
historical contingency of diffusion determinants. The institutionalist research on innovation
diffusion (Ramirez et al. 1997; Thornton & Ocasio 1999; Tolbert & Zucker 1983; for educational
innovation, see Moon & Wotipka 2006; Suárez et al. 2009) indeed emphasizes the critical role of
historical time for understanding variation in the adoption of innovations. The main argument of
these studies is twofold. First, as innovative ideas and practices become more highly
institutionalized, the adoption rate will significantly increase from the earlier period to the later
period, because the legitimacy of the innovations themselves serves as the driving force for the
later adopters. And therefore, second, the internal organization-specific factors will become less
47
predictive than the external institutional factors over time. Later adopters will begin to accept the
taken for granted innovations regardless of any particular internal characteristics or conditions.
Drawing on such an institutionalist perspective, I argue that the post-2002 period may
constitute a critical juncture in the trajectory of nonprofit management programs. According to
Tolbert and Zucker (1996), there are two different phases of innovation diffusion: a phase of pre-
institutionalization, in which a limited number of organizations adopt innovative ideas and
practices for their own technical purposes, and a phase of full-institutionalization, in which
innovations gain socially constructed consensus on their general utility and thus diffuse more
rapidly and widely. By this periodization, the pre-2002 period can be viewed as a phase of pre-
institutionalization when the nonprofit sector was relatively under-structured and thus nonprofit
management discourse was seeking its legitimacy. In contrast, the post-2002 period can be seen as
a phase of full-institutionalization when the structuration of the nonprofit sector was accelerated
by the passage of the Sarbanes-Oxley Act of 2002 (and accompanied by tax and policy reforms)
which brought renewed attention to the professionalization of nonprofits and thus nonprofit
management began to achieve taken-for-granted status in the United States (Bromley & Orchard
forthcoming). Based on this periodization, it is expected that the adoption rate of programs during
the post-2002 period will be higher than that of the pre-2002 period, with accelerated structuration
of the nonprofit sector after the year 2002:
Hypothesis 4. A historical event in the United States, the passage of the Sarbanes-Oxley Act
of 2002, will alter the context for nonprofit management research and education, shifting the
baseline rate of program adoption upward.
48
3.3. Data and Methods
The quantitative dataset consists of a panel of 1,451 American universities and colleges from
1971 to 2011. The sample was drawn from the population of universities and colleges that offered
a Bachelor’s, Master’s, or Ph.D. degree based on Carnegie Classification of Institutions of Higher
Education for 1976, 1987, 1994, and 2000. Two-year, vocational, and other types of specialized
institutions that have remarkably different educational goals and organizational structures from
those of four-year institutions (Kraatz & Zajac 1996; Olzak & Kangas 2008) and institutions
located in U.S. territories (e.g., Guam, Puerto Rico, and the United States Virgin Islands) were
excluded from this population. Initially, 1,486 institutions were identified as a valid sample; the
number of institutions that were considered for the final analysis was reduced to 1,451 because
some lacked data on independent variables. The unit of analysis is the university-year; therefore,
the final sample includes 59,491 university-years of data—1,451 universities multiplied by 41
years of data for each university. The 1971 through 2011 timeframe is used not only because higher
quality data on the variables are available during this timeframe but also because a substantial
number of universities and colleges started establishing nonprofit management programs after the
mid-1970s.
3
Dependent Variables
I examined three types of academic program adoption at universities as follows: the
university’s first adoption of (1) a stand-alone nonprofit management degree program; (2) a
3
Because the data prior to 1971 are excluded from the analysis, the observations are subject to left censoring, which
arises when an event of interest has already occurred for a sample of individuals before the start of the study period,
because universities in the sample adopted nonprofit management programs before 1971. There are five left-censored
observations; all of them are certificate programs, not research centers or degree programs.
49
research center for nonprofit studies; and (3) any type of academic program, including stand-alone
degree programs, non-degree programs, and research centers. The first dependent variable consists
of Bachelor’s, Master’s, and doctoral degrees in nonprofit management (e.g., Bachelor of Arts in
Philanthropic Studies, Master of Arts in Nonprofit Leadership and Management, and Master of
Science in Nonprofit Organizations). The second includes within-university institutes for nonprofit
management education, research, training, and consulting. The third consists of all types of
nonprofit management programs, including undergraduate- and graduate-level degrees, non-
degrees (e.g., minor, track, specialization, concentration, focus, emphasis, and certificate), and
research centers, excluding non-credit internship and workshop programs. Each dependent
variable is coded as 1 if a university adopted a degree program, research center, or any type of
program in a given year (t) and 0 otherwise.
4
Once a university created its first program, it was no
longer included in the set of universities that were at risk of transition from not-adopting to
adopting a program. Therefore, universities remained in the risk set as long as they had not yet
adopted a nonprofit management program.
Data on the dependent variables were generated through a scrupulous review of each
institution’s online archives and follow-up email correspondences with program directors,
department chairs, and school deans at selected institutions. In 2011 and 2012, I first examined all
types of online resources—e.g., college/school catalogs, department brochures, syllabi, and other
archival sources—for each university and college. From this examination, I identified 342
universities and colleges that have ever established one or more nonprofit management programs.
In 2012 and 2013, I contacted more than 700 program directors and heads of departments/schools
of the 342 institutions via email, asking if and when they adopted their first program. In the end, I
4
The timing of program adoption was the academic year when a university or college launched its first program and
started offering it to students.
50
was able to collect information from 286 universities and colleges; the remaining 56 institutions
either refused to provide information or did not have accurate historical data on their programs.
To reduce the problem of retrospective bias, I communicated with at least two people at each
institution and compared their responses. Then, I verified the collected information, if possible,
through historical documents obtained from each institution’s online archives and a well-known
public database on university-based nonprofit management programs collected by Seton Hall
University (http://academic.shu.edu/npo).
5
Independent Variables
I included three sets of independent variables to capture the macro-, meso-, and local-level
concepts discussed earlier. The indicators were measured at the national, state, and university
levels, respectively. Thus, for each university, at each time point (each year), the macro-level
variables were taken from the U.S. as a whole and varied only by year; the meso-level variables
were from the state in which the university is located, and the local-level variables were from each
individual university. Data on independent variables were compiled from various publicly
available sources. Missing values were interpolated with the closest values available and the
average values around those that are missing; listwise deletion would have considerably reduced
the overall sample size, which, in turn, would have likely led to biased results.
6
Macro-Level: Structuration of the Nonprofit Sector. I measured a series of indicators of
the structuration of the nonprofit sector at the national level. I initially created six measures of the
5
I build upon and expand the Seton Hall University database by collecting historical data on various types of nonprofit
management studies programs, including a stand-alone degree program, non-degree program, and research center.
6
Appendix A shows descriptive statistics of imputed covariates, comparing imputed values with original ones.
51
extent of nonprofit discourse saturation: Nonprofit organization, Government grant, State
association, University program, Dissertation, and News article. The first three are measures that
capture changes in the size and scale of the U.S. nonprofit sector; Nonprofit organization is the
count of 501(c)(3) charitable organizations obtained from the Business Master Files provided by
the NCCS (2012); Government grant is the calculation of federal government grants to 501(c)(3)
organizations in thousands of U.S. dollars, which was obtained from the SOI Tax Stats provided
by the IRS (2011); and State association is the cumulative count of states with a statewide
professional association, which was calculated from the National Council of Nonprofits website
(http://www.councilofnonprofits.org/). The next two measures track the changing interest in
nonprofit studies within the academic community; University program is the cumulative count of
universities offering nonprofit management programs, which was drawn from the same dataset
used for the dependent variables, and Dissertation is the count of doctoral dissertations with the
keywords “nonprofit” or “nongovernmental” in their abstracts, which was collected from the
ProQuest database. The last sixth measure captures the ebb and flow of public attention on
nonprofit management issues; News article is the count of news articles with the keywords
“nonprofit,” “not-for-profit,” “charitable sector,” “philanthropic sector,” “voluntary sector,” “civil
society,” “NGO,” “non-governmental organization,” or “philanthropy” in their headings, which
were obtained from the Lexis-Nexis database. All six of the measures are time-varying, lagged by
one year and logged to reduce skewness, except for State association. In the analysis, however, I
combined all of the measures into one factor, Structuration, because the six measures are highly
correlated with a single underlying factor.
52
Meso-Level: Socioeconomic and Political Context. To determine the effects of state-level
socioeconomic and political conditions, I employed four variables. Urbanization, the percentage
of the population living in a Standard Metropolitan Statistical Area in a state, was used to measure
the demographic heterogeneity within a state. Welfare expenditure, the amount of social welfare
expenditures as a percentage of total state expenditures, captures support for social services in a
state. The data for these two socioeconomic variables were obtained from the U.S. Census Bureau
(2013). Citizen liberalism and Government liberalism reflect the political orientation of the citizens
and governments in a state. I used Berry et al. (2010)’s measure of citizen and state government
ideology, which ranges from 0 (high conservatism) to 100 (high liberalism). All four variables are
state-level time-varying variables and lagged by one year.
Local-Level: University Factors. I measured university-specific conditions and dynamics
using four variables. Data on these variables were collected from the HEGIS for the years 1970
through 1986 and the IPEDS for the years 1987 to 2010.
7
I included two indicators of the centrality
of specific academic programs in a university, as measured by the percentage of Bachelor’s,
Master’s, and Ph.D. degrees in public affairs/social work and business management conferred,
respectively. These two variables—PA/SW degree and BUS degree—capture different levels of
power and resources that a potential disciplinary venue for nonprofit management programs within
a university might have. To capture the influence of campus diversity, I also included
Black/Hispanic and Female, the proportion of Bachelor’s, Master’s, and doctoral degrees
conferred to African American/Hispanic students and female students, respectively. All four
variables are time-varying and lagged by one year.
7
I thank Dr. Ozan Jaquette for sharing his compiled data on accredited postsecondary institutions with me.
53
Historical Context. In addition to the three sets of independent variables, I included a
dummy variable, Post-2002, to capture the effect of historical context. I distinguished between a
phase of pre-institutionalization (1971-2001) and full-institutionalization (2002-2011), assigning
a value of 1 to the 2002-2011 timeframe and 0 to 1971-2011.
Control Variables
I controlled for university-level organizational characteristics using a measure of size,
structure, and reputation. First, to examine the effect of university size, I constructed University
revenue, the amount of university revenues in millions of U.S. dollars (logged), and Master’s
degree,
the percentage of total degrees granted at the Master’s level (logged). These two variables
are time-varying and lagged by one year. Typically, studies on universities and colleges (e.g., Brint
et al. 2009; Gumport & Snydman 2002; Lounsbury 2001; Robinson 2011) note that campus size
positively correlates with curricular innovation and organizational change. It is mainly because
large universities have substantial resources and capacities for innovation (e.g., funding and
staffing), and thus can absorb the risk of innovation. Therefore, it can be assumed that large
universities with more resource slack and capacity for action will have a higher rate of adopting
nonprofit management programs than small ones.
Second, a dummy variable, Public, was constructed to measure structural aspects.
8
Universities were scored 1 if they are public and 0 if private. Universities and colleges that are
chartered as public, religious, land grant, flagship, or minority-serving institutions tend to take a
different path toward new educational challenges than those are not. The reason behind this is that
the distinctive missions and structural characteristics that are built from their past characterize their
8
I tested a number of other measures, such as whether an institution is a religious, land grant, flagship, or historically
black college or university, but they did not improve model fit.
54
behaviors (Lounsbury 2001; McDowell 2003; Ostrander 2004; Tolbert 1985). Based on this line
of reasoning, it is plausible to assume that public universities will have a higher rate of adopting
programs because they are asked to fulfill their historical mission and obligation of public service
by responding to the needs of society for nonprofit management.
Lastly, I also measured university selectivity as a dummy variable, Doctoral/Research,
indicating whether a university is categorized as a doctorate-granting research university with a
very high level of research activity based on the 2005 Carnegie Classification of Institutions of
Higher Education.
9
Studies on universities and colleges (e.g., Brint et al. 2009; Kraatz & Zajac
1996; Olzak & Kangas 2008; Soule 1997) provide some evidence supporting a widely shared
belief that elite universities are more likely to take risks of adopting novel programs because
innovation is one of the ways to maintain their reputation. Extrapolating from this perception, it
can be argued that higher-prestige universities will have a higher rate of adopting programs than
lower-prestige ones.
Tables 3.1 and 3.2 report a summary of all independent and control variables, including basic
descriptive statistics and correlations.
Modeling
I used event history analysis to assess the factors that contribute to the first adoption of each
type of nonprofit management program in a given university from 1971 to 2011. Event history
analysis is a statistical methodology for analyzing whether, when, and under what
conditions an event (a qualitative change) was experienced by organizations (or another unit of
9
The 2005 edition of Carnegie Classification of Institutions of Higher Education has the following three categories
for doctorate-granting universities, which are based on a measure of research activity: RU/VH (Research
Universities—very high research activity), RU/H (Research Universities—high research activity), and DRU
(Doctoral/Research Universities). Here, only RU/VH were coded as 1, and RU/H, DRU, and non-doctorate-granting
universities were coded as 0.
55
Table 3.1. Definitions and Descriptive Statistics for Covariates (N=1,451 universities and 59,491 university years)
Category Variable Definition Mean SD Min Max
Structuration Structuration Factor of the six institutional context measures (logged) -4E-10 1.00 -0.98 2.34
- Nonprofit org. # of 501(c)(3) charitable organizations (logged) 14.14 0.53 13.47 15.06
- Gov. grant Government grants to 501(c)(3) organizations, in thousand USD (logged) 18.53 0.72 17.77 19.86
- State association Cum. # of states with a statewide association of nonprofits 19.24 13.90 1 43
- Univ. program Cum. # of universities with nonprofit management programs (logged) 4.42 1.76 1.00 6.61
- Dissertation # of doctoral dissertations with a keyword “nonprofit” in abstracts (logged) 5.39 1.93 1.00 7.66
- News article # of news articles with a keyword “nonprofit” in headings (logged) 7.92 1.64 5.06 10.78
Socioeconomic
& Political
context
Urbanization % of population living in a Standard Metropolitan Statistical Area, by state 72.77 14.00 32.20 100
Welfare expenditure Social welfare expenditures as a % of total state expenditures, by state 19.75 5.66 4.08 38.75
Citizen liberalism Citizen ideology index developed by Berry et al. (2010), by state 50.73 15.78 7.04 95.97
Gov. liberalism State government ideology index developed by Berry et al. (2010), by state 53.48 11.83 23.64 76.44
University
dynamics
PA/SW degree % of degrees in Public Affairs and Social Work conferred 2.08 4.95 0 100
BUS degree % of degrees in Business conferred 20.27 17.05 0 100
Black/Hispanic % of degrees conferred to African American and Hispanic students 12.57 21.14 0 100
Female % of degrees conferred to female students 53.05 20.51 0 100
Historical
context
Post-2002 Dummy for the 2002-2011 time period (1=2002-2011; 0=1971-2001)
0.22 0.41 0 1
University-
level controls
Univ. revenue Total university revenues, in million USD (logged) 4.83 1.37 1.00 10.30
Master’s degree % of total degrees granted at the master’s level (logged) 2.91 1.57 1.00 5.62
Public Dummy for public universities (1=public; 0=else) 0.35 0.48 0 1
Doctoral/Research Dummy for doctoral/research universities (1= RU/VH; 0=else) 0.07 0.25 0 1
Note: All the listed variables are lagged by one year, except time-invariant variables, Post-2002, Public, and Doctoral/Research.
56
Table 3.2. Correlation Matrix of Covariates
1 2 3 4 5 6 7 8 9 10 11 12 13
1. Structuration
2. Urbanization 0.14
3. Welfare expenditure 0.54 0.36
4. Citizen liberalism 0.28 0.39 0.46
5. Government liberalism -0.20 0.07 0.03 0.34
6. PA/SW degree 0.05 0.02 0.05 0.05 0.02
7. BUS degree 0.13 0.03 0.04 -0.03 -0.09 -0.01
8. Black/Hispanic 0.12 0.01 0.00 -0.14 -0.05 0.14 0.13
9. Female 0.27 0.02 0.21 0.11 -0.05 0.14 0.08 0.19
10. Post-2002 0.87 0.10 0.41 0.24 -0.13 0.03 0.06 0.10 0.19
11. University revenue 0.26 0.21 0.22 0.16 -0.03 0.06 -0.02 0.00 0.11 0.20
12. Master’s degree 0.21 0.22 0.16 0.12 -0.01 0.12 0.09 0.00 0.15 0.17 0.55
13. Public 0.00 -0.06 -0.08 -0.09 -0.02 0.04 -0.08 0.05 0.00 0.00 0.43 0.29
14. Doctoral/Research 0.00 0.10 0.01 0.03 0.01 0.03 -0.09 -0.07 -0.10 0.00 0.55 0.23 0.17
57
observation) during a period of observation. Compared to conventional regression models, event
history analysis is superior for studying diffusion because it is capable of predicting both the timing
of a transition across states and conditions under which the transition takes place—by including
time-varying covariates (covariates that change over time) and right-censored cases (observations
that have not experienced an event of interest by the end of the study timeframe) (Box-
Steffensmeier & Jones 2004; Tuma & Hannan 1984; Yamaguchi 1991).
More specifically, I utilized exponential transition rate models to estimate the transition rate
(or hazard rate) of each university from not-adopting to adopting an academic program under
various conditions over time. My argument about the expansion of university programs assumes
that the transition rate of each university from not-adopting to adopting an academic program is
the same over time, but contingent on covariates. For this reason, I employed the exponential
transition rate models in which the risk of a transition occurring is invariant with respect to time,
and all of the fluctuation in the transition rate comes from the covariates. In the exponential
transition rate models, the transition rate of the first program adoption, r(t), is log-linearly related
to the n covariates (Xn) as follows:
log r(t) = 𝛽 0
+ 𝛽 1
𝑋 1
+ 𝛽 2
𝑋 2
+ ⋯ + 𝛽 𝑛 𝑋 𝑛
where log r(t) is the log of the transition rate at time t, 𝛽 𝑛 are constants to be estimated, and 𝑋 𝑛 are
the covariates. The transition rate is the annual rate at which universities and colleges adopt their
first program, and I took the natural log of the transition function to prevent it from unrealistically
becoming less than zero.
10
10
As a robustness check, I compared exponential transition rate models to piecewise constant models with period-
specific effects (available in Appendix B) that assume the baseline transition rates and covariate effects can vary across
historical time periods. The results of the likelihood ratio tests showed that the exponential transition rate models are
a significant improvement over the piecewise constant models.
58
3.4. Results
Figure 3.1 presents three main dependent variables from 1971 to 2011, demonstrating the
number of universities that established their first stand-alone degree program, research center, and
any type of program each year and the cumulative number of universities that were established
since 1971. As the figure shows, the growth of academic programs was relatively slow until the
early 1990s—with approximately five universities having offered degree programs, about ten
universities having established research centers, and about fifty universities having adopted any
type of program. In fact, the majority of academic programs were created after the late-1990s. By
2011, as shown in the figure, 58 universities in the sample of 1,451 universities had created their
degree programs, 68 universities had established their first research centers, and 281 universities
had adopted at least one type of program.
Tables 3.3-3.5 provide the results of the exponential transition rate models for the three
dependent variables, respectively. I present the following four models for each dependent variable:
Model 1 includes local-level university variables and controls; Model 2 includes meso-level
socioeconomic and political variables and controls; Model 3 includes a factored macro-level
structuration variable and controls;
11
and Model 4 includes all covariates.
Degree Program
Table 3.3 presents the results concerning the creation of nonprofit management stand-alone
degree programs. As expected, the structuration of the nonprofit sector (Structuration) is
statistically significant and has a positive effect on the adoption. This finding provides support for
Hypothesis 1, indicating that universities and colleges are more likely to adopt programs as
11
Models separately testing each component of the factor are available in Appendix C.
59
< Degree Program Adoption >
< Research Center Adoption >
< Any Program Adoption >
Figure 3.1. The Growth of Academic Programs, Cumulative Count and Yearly Adoptions (1971-2011)
60
nonprofit management becomes legitimized and recognized as an important logic of American
society. More specifically, Table C.1 in Appendix C shows that the process of structuration in the
nonprofit sector—including the growth of nonprofits in size and scope (measured by Nonprofit
organization, Government grant, and State association), the increasing interest in nonprofit studies
within the academic community (measured by University program and Dissertation), and the
growing public attention on nonprofit management issues (measured by News article)—leads to
an overall increase in program adoption across universities.
Urbanization and Welfare expenditure, which are thought to increase the likelihood of degree
program adoption, do not have consistently positive or significant effects. Thus, Hypotheses 2a
and 2b, which assumed that population diversity and support for social services in a state would
impact adoption, are not verified. Interestingly, two variables for testing the effect of political
liberalism (Citizen liberalism and Government liberalism) have opposite directions. However,
both the variables have nonsignificant effects on the adoption, and thus the findings do not support
Hypothesis 2c.
Hypothesis 3a is strongly supported, as the existence of powerful and resource-rich receptor
sites for nonprofit management within a university (PA/SW degree and BUS degree) exerts a
positive impact. There is very little support for Hypothesis 3b, which proposed the demographic
diversity assumption; only Female has a significant positive effect. This partly supports the
assumption of Hypothesis 3b that universities with more female students may have a higher
adoption rate.
University size appears to be important. Both University revenue and Master’s degree
positively influence the degree program offerings, indicating that degree programs are more likely
to be adopted by larger universities and colleges. However, the results do not support the structure
61
Table 3.3. Exponential Transition Rate Models: Degree Program
Variable Model 1 Model 2 Model 3 Model 4
Structuration 0.836***
(0.131)
0.854***
(0.159)
Urbanization -0.016
(0.011)
-0.004
(0.012)
Welfare expenditure 0.087***
(0.027)
0.025
(0.031)
Citizen liberalism 0.004
(0.011)
-0.016
(0.012)
Government liberalism -0.013
(0.012)
0.004
(0.012)
PA/SW degree 0.034*
(0.017)
0.042***
(0.016)
BUS degree 0.026***
(0.008)
0.023***
(0.009)
Black/Hispanic -0.004
(0.008)
-0.012
(0.009)
Female 0.032***
(0.01)
0.019
(0.012)
Univ. revenue (log) 0.668***
(0.141)
0.484***
(0.142)
0.358***
(0.137)
0.486***
(0.151)
Master’s degree (log) 0.264**
(0.125)
0.364***
(0.124)
0.302**
(0.12)
0.222*
(0.125)
Public -0.725**
(0.305)
-0.731**
(0.313)
-0.664**
(0.301)
-0.619*
(0.326)
Doctoral/Research -0.74
(0.533)
-0.882*
(0.536)
-0.46
(0.544)
-0.397
(0.554)
Constant -13.606***
(1.144)
-10.708***
(1.143)
-10.110***
(0.729)
-11.773***
(1.479)
LRχ² 68.641 64.491 94.76 109.166
Log-likelihood 15.573 13.498 28.633 35.836
Number of adoptions 58 58 58 58
Number of universities 1,451 1,451 1,451 1,451
Yearly spells 59,065 59,065 59,065 59,065
* p<0.1; ** p<0.05; *** p<0.01. Standard errors are in parentheses.
62
and selectivity argument; none of the tested variables (Public and Doctoral/Research) are
positively associated with adoption. Rather, private and non-elite universities seem to be more
likely to have a degree program compared to public and elite universities.
Research Center
In Table 3.4, the results show a relationship between the establishment of research centers
and various independent variables. Hypothesis 1 is strongly supported here as well (see also Table
C.2 in Appendix C). This finding demonstrates that stronger structuration of the nonprofit sector
increases the rate of research center establishment.
The validity of Hypotheses 2a, 2b, and 2c is in question because none of the tested variables
(Urbanization, Welfare expenditure, Citizen liberalism and Government liberalism) are
statistically significant. Perhaps, there may be no significant relationship between a culturally
diverse, welfare-oriented, and liberal atmosphere at the state level and a higher adoption rate.
As expected in Hypothesis 3a, a resource-rich receptor site plays a key role in explaining the
rate at which universities establish a research center. That is, the likelihood of establishing a
research center increases by having larger public affairs/social work programs (PA/SW degree)
and business programs (BUS degree). Hypothesis 3b is partially supported here as well. Only
Female has a significantly positive effect, showing that demographic diversity within a university
seems to positively, but marginally, affect the adoption.
Both University revenue and Master’s degree are consistently positive, indicating that an
abundance of resources within a university seems to influence the adoption rate of a research center.
Here again, structural characteristics and high selectivity may not be associated with research
center creation; Public and Doctoral/Research have nonsignificant effects.
63
Table 3.4. Exponential Transition Rate Models: Research Center
Variable Model 1 Model 2 Model 3 Model 4
Structuration 0.398***
(0.114)
0.350**
(0.140)
Urbanization 0.007
(0.011)
0.016
(0.012)
Welfare expenditure 0.033
(0.026)
0.004
(0.029)
Citizen liberalism 0.012
(0.011)
0.001
(0.011)
Government liberalism -0.008
(0.011)
0.001
(0.011)
PA/SW degree 0.033
(0.023)
0.040*
(0.022)
BUS degree 0.031***
(0.009)
0.028***
(0.009)
Black/Hispanic -0.015
(0.011)
-0.024*
(0.013)
Female 0.033***
(0.011)
0.023*
(0.012)
Univ. revenue (log) 0.993***
(0.141)
0.800***
(0.142)
0.765***
(0.139)
0.824***
(0.150)
Master’s degree (log) 0.306*
(0.166)
0.399**
(0.165)
0.420**
(0.164)
0.301*
(0.165)
Public -0.103
(0.260)
0.004
(0.267)
-0.149
(0.251)
0.102
(0.282)
Doctoral/Research
-0.105
(0.391)
-0.266
(0.383)
-0.092
(0.395)
0.106
(0.402)
Constant -16.041***
(1.256)
-14.266***
(1.235)
-12.804***
(0.881)
-16.029***
(1.563)
LRχ² 152.334 140.547 146.196 162.956
Log-likelihood 65.234 59.341 62.165 70.545
Number of adoptions 68 68 68 68
Number of universities 1,451 1,451 1,451 1,451
Yearly spells 58,769 58,769 58,769 58,769
* p<0.1; ** p<0.05; *** p<0.01. Standard errors are in parentheses.
64
Any Type of Program
Table 3.5 illustrates the results for the first adoption of any type of academic program. The
results support Hypothesis 1, showing strong empirical evidence for the effects of the structuration
of the nonprofit sector on the adoption rate of programs (see also Table C.3 in Appendix C).
There is significant evidence supporting Hypotheses 2a and 2b, indicating that academic
programs are adopted at a more rapid rate in both more urbanized regions (Urbanization) and more
welfare-oriented regions (Welfare expenditure). Hypothesis 2c is not supported, as there are
inconsistent and negative effects of political liberalism on adoption rates. There is weak evidence
that liberal state government (Government liberalism) is negatively associated with the creation of
academic programs.
Hypothesis 3a is strongly supported, as the existence of powerful and resource-rich receptor
sites for nonprofit management within a university (PA/SW degree and BUS degree) exerts a
positive impact. The validity of Hypothesis 3b is in question because the analyses provide marginal
support for the hypothesis; there is a strong positive association between universities with more
female students (Female) and program adoption, while the presence of more African-American
and Hispanic students (Black/Hispanic) within a university has a negative relationship with the
adoption rates.
The results show that larger universities (University revenue) and universities with larger
graduate programs (Master’s degree) tend to be more involved in offering academic programs. As
expected, positive and significant parameter estimates for Public are found, supporting the idea
that public universities are more likely than private universities to have a program. The variable
that measured the university selectivity (Doctoral/Research) has a negative association, suggesting
that elite universities are less likely to adopt programs.
65
Table 3.5. Exponential Transition Rate Models: Any Type of Program
Variable Model 1 Model 2 Model 3 Model 4
Structuration 0.462***
(0.055)
0.368***
(0.070)
Urbanization 0.005
(0.005)
0.011**
(0.006)
Welfare expenditure 0.049***
(0.012)
0.019
(0.014)
Citizen liberalism 0.007
(0.005)
-0.002
(0.006)
Government liberalism -0.020***
(0.005)
-0.010*
(0.005)
PA/SW degree 0.024**
(0.011)
0.034***
(0.010)
BUS degree 0.024***
(0.004)
0.020***
(0.004)
Black/Hispanic -0.004
(0.004)
-0.009**
(0.004)
Female 0.029***
(0.005)
0.018***
(0.005)
Univ. revenue (log) 0.777***
(0.067)
0.604***
(0.067)
0.588***
(0.066)
0.622***
(0.071)
Master’s degree (log) 0.359***
(0.070)
0.431***
(0.070)
0.425***
(0.069)
0.346***
(0.070)
Public 0.116
(0.131)
0.203
(0.133)
0.142
(0.128)
0.322**
(0.140)
Doctoral/Research -0.356*
(0.207)
-0.490**
(0.206)
-0.333
(0.209)
-0.126
(0.213)
Constant -13.050***
(0.565)
-10.874***
(0.564)
-10.198***
(0.387)
-12.258***
(0.704)
LRχ² 419.496 404.234 428.77 477.358
Log-likelihood 576.509 568.878 581.146 605.44
Number of adoptions 281 281 281 281
Number of universities 1,446 1,446 1,446 1,446
Yearly spells 56,037 56,037 56,037 56,037
* p<0.1; ** p<0.05; *** p<0.01. Standard errors are in parentheses.
66
Historical Context Effects
Table 3.6 provides tests for Hypothesis 4 that suggested the assumption of historical time
dependence. In Table 3.6, two models for each dependent variable are presented: Model 5 includes
Post-2002 and controls; and Model 6 includes all covariates, except Structuration. Structuration
is excluded from this analysis because of a high correlation with Post-2002 (see the correlation
matrix in Table 3.2).
Overall, the results strongly support Hypothesis 4. The Post-2002 variable is statistically
significant and has a positive effect on the adoption of each type of nonprofit management program.
This indicates the adoption rates significantly increase from the first time period (1971-2001) to
the second period (2002-2011). In other words, when all the independent and control variables are
held constant, nonprofit management programs are much more likely to emerge in university
campuses in the post-2002 period than in the pre-2002 period. This finding aligns with the
observed adoptions in Figure 3.1, which indicate that the diffusion process began to unfold rapidly
around the time that a series of tax and policy reforms marked an important shift in the nonprofit
sector in the early 2000s.
3.5. Discussion
The analysis of the expansion of nonprofit management in universities offers insight into a
largely unexplored aspect of professionalization in the sector. Based on an event history data set
(1971-2011) with 1,451 American universities and colleges, this chapter explores the context and
manner in which nonprofit management studies programs have emerged and expanded in the
United States. Using exponential transition rate models, the chapter explores several lines of
argument to explain the adoption of nonprofit management programs across university campuses.
67
Table 3.6. Exponential Transition Rate Models with Historical Context Effects
Variable Degree Program Research Center Any Type of Program
Model 5 Model 6 Model 5 Model 6 Model 5 Model 6
Post-2002 1.368***
(0.280)
1.093***
(0.313)
0.482*
(0.254)
0.271
(0.282)
0.712***
(0.125)
0.407***
(0.140)
Urbanization -0.01
(0.011)
0.013
(0.012)
0.009
(0.006)
Welfare expenditure 0.047
(0.030)
0.018
(0.028)
0.032**
(0.013)
Citizen liberalism -0.005
(0.012)
0.006
(0.011)
0.002
(0.005)
Government liberalism -0.003
(0.012)
-0.004
(0.011)
-0.015***
(0.005)
PA/SW degree 0.039**
(0.016)
0.036
(0.023)
0.031***
(0.011)
BUS degree 0.024***
(0.008)
0.030***
(0.009)
0.022***
(0.004)
Black/Hispanic -0.008
(0.009)
-0.02
(0.013)
-0.007*
(0.004)
Female 0.022**
(0.011)
0.027**
(0.012)
0.022***
(0.005)
Univ. revenue (log) 0.420***
(0.137)
0.536***
(0.151)
0.838***
(0.139)
0.879***
(0.149)
0.639***
(0.066)
0.659***
(0.070)
Master’s degree (log) 0.329***
(0.121)
0.243*
(0.125)
0.417**
(0.166)
0.298*
(0.165)
0.429***
(0.069)
0.349***
(0.070)
Public -0.732**
(0.299)
-0.607*
(0.324)
-0.195
(0.250)
0.114
(0.281)
0.077
(0.127)
0.315**
(0.139)
Doctoral/Research -0.671
(0.538)
-0.499
(0.553)
-0.283
(0.385)
0.027
(0.401)
-0.502**
(0.206)
-0.186
(0.213)
Constant -10.546***
(0.715)
-12.596***
(1.441)
-13.162***
(0.871)
-16.695***
(1.539)
-10.518***
(0.382)
-12.779***
(0.695)
LRχ² 74.998 91.771 137.817 157.708 392.281 458.787
Log-likelihood 18.752 27.138 57.976 67.921 562.901 596.155
Number of adoptions 58 68 281
Number of universities 1,451 1,451 1,446
Yearly spells 59,065 58,769 56,037
* p<0.1; ** p<0.05; *** p<0.01. Standard errors are in parentheses.
68
Based on the analysis, between 1971 and 2011, 58 out of 1,451 sample universities created
their first degree program, 68 established their first research center, and 281 adopted their first
nonprofit-related program. Additionally the adoption rates of such programs in general have grown
rapidly since the mid-1990s. Exponential transition rate models support my argument that a key
factor contributing to the spread of academic programs, regardless of the type of program is the
structuration of the nonprofit sector. Universities are responding to the strongly institutionalized
logic of nonprofit management by creating academic programs and research centers. However,
there is less support for the socioeconomic and political context indicators. The adoption rates in
the models are occasionally positively influenced by demographic diversity, government support
for social services, and political liberalism in a given state. The findings support my argument
about university-specific conditions and dynamics; universities with more human and financial
resources, resource-rich receptor sites for nonprofit management (e.g., public administration,
social work, and business schools), more female students, and a less prestigious institutional status
tend to establish research and education programs. Lastly, the speed and extent of the expansion
vary by historical context. Overall, the adoption rates significantly increase from the pre-2002
period to the post-2002 period.
There are two core findings that merit further comment. First, nonprofit management is fairly
well established, at the late semi-institutionalization stage or early full-institutionalization stage in
the process suggested by Tolbert and Zucker (1996). The current analysis results show that six
different measures of the structuration of nonprofit management at the macro-level are highly
correlated with a single underlying factor, indicating that the transformation of the nonprofit sector
itself creates momentum for the emergence and growth of nonprofit management. Additionally,
structuration helps to explain the growth of research and education programs. After controlling for
69
other university- and region-specific factors, the structuration variable significantly affects the
creation of academic programs within universities and colleges.
Second, local university characteristics are quite relevant for the growth of nonprofit
management programs in the higher education. Universities that have been more active in offering
nonprofit programs tend to offer many graduate degrees and they tend to be large, private, less
prestigious, and less research-oriented. This finding provides some insight into the future
development of nonprofit management research and education. For example, greater commitment
of selective and research-oriented universities could be a final indicator of institutionalization, with
nonprofit programs moving from peripheral universities to the most prestigious and recognized
universities.
While this chapter analyzes various factors that are associated with the emergence and
expansion of nonprofit management studies programs, offering insights into a largely unexplored
aspect of the phenomenon, many more studies are needed to address the variations in the gradual
institutionalization processes. Why and how nonprofit management programs are differently
shaped and developed across university campuses merits further exploration. The following
chapters thus investigate the different developmental phenomena of nonprofit management and
associated factors, focusing mainly on variations in the disciplinary settings for and
institutionalization processes of nonprofit management studies.
70
CHAPTER 4. VARIATION IN THE DISCIPLINARY SETTING FOR
NONPROFIT MANAGEMENT STUDIES
“It is a huge field and it is a field of variety.”
- Interview with faculty, August 2014
“It is going to appear in every kind of field possible.”
- Interview with faculty, August 2014
4.1. Background
As described in the previous chapter, university-based research and education programs in
nonprofit management have expanded at an increased rate during recent decades (Mirabella 2007;
Mirabella & Wish 2001; O’Neill 2005). These programs have developed in a variety of forms and
venues—such as MPA degree programs housed in public affairs schools, MBA programs in
business schools, MSW programs in social work schools, and interdisciplinary or free-standing
degree programs in various other types of schools—but no single, dominant disciplinary affiliation
has yet emerged (Mirabella & Wish 2000; O’Neill 2007; Young 1999). Why and how are these
programs housed in different schools of universities? In what context and manner are these
programs differently framed and developed across university campuses? This chapter examines
these unanswered questions to gain a better understanding of the different developmental processes
of nonprofit management studies programs across universities and to further generalize the insights
obtained for the overall development of nonprofit management ideas and practices.
The present chapter discusses the variety of disciplinary settings in which nonprofit
management programs have emerged and developed and then develops several arguments about
how the external environment and historical context might influence where nonprofit management
71
programs are placed within universities. Building on and extending the work of Young (1988) and
Hall et al. (2001), this study proposes a comprehensive conceptual framework by arguing that the
disciplinary location of nonprofit management programs in universities is influenced not only by
the organizational conditions and politics within universities but also by market dynamics and a
changing institutional environment.
More specifically, using competing risk models with event history data that were collected
from 1,451 American universities and colleges (1971-2011), the present study explores how the
external environment and university characteristics influence the academic placement or home of
nonprofit management programs, and the study also considers if and how time affects the results.
The study contributes to the literature on public and nonprofit studies by extending prior work that
addresses normative questions about the most appropriate or beneficial academic setting for
nonprofit management programs. Rather that attempting to provide an answer as to where
nonprofit programs should be located, then, I contribute by concentrating on what influences where
they do get placed. The paper also contributes to the research literature by demonstrating how the
structuration of the nonprofit sector, and other aspects of the external environment, influence
universities. Though many contextual dynamics at the most local level matter for adoption, an
issue I address in the next chapter, my chapter also shows that history and cultural processes have
explanatory power of their own.
In the sections that follow, I first briefly review the literature and develop three major lines
of argument regarding the types of institutional, social, and organizational factors that are likely
to be associated with the disciplinary location of nonprofit management programs within a
university. In the next section, I describe the sample, data, variables, and empirical model
specification that are used to flesh out my general conceptual arguments. The results of a
72
competing risks analysis are then presented. Finally, I conclude by discussing the empirical
findings and answering unsettled questions regarding the development of nonprofit management
studies programs.
4.2. A Theoretical Framework: Dynamics surrounding the Disciplinary Location
Existing studies on the development of nonprofit management research and education have
generally explained the variety of academic homes for nonprofit management studies programs
from two distinct perspectives: the demand-side dynamics of nonprofit fields and the supply-side
politics within host universities. From the demand-side point of view, on the one hand, much has
been written about the association between disciplinary settings and substantive nonprofit fields
such as social work, the arts, education, and the human services. Besides segmentation in terms of
substantive professional work, nonprofit organizations vary in size, mission, structure, core
technology, revenue sources, labor relations, career mobility and many other characteristics (Cook
1988; DiMaggio 1988; Heimovics & Herman 1989; Rimer 1987). Given the great diversity of the
nonprofit sector, academic programs could emerge within specific fields like social work or they
could emerge in general business management programs (Haas & Robinson 1998; Hall 1996;
Hoefer 2003).
From the supply-side perspective, on the other hand, there is abundant literature (e.g., Bies
& Blackwood 2007; Hall et al. 2001) that argues that the disciplinary venues for nonprofit
management programs likely vary according to host universities’ traditions, structure, economics,
and culture. This variation mainly occurs because each university has different strengths, local
conditions, interests, and needs as well as different definitions of the scope and nature of nonprofit
management studies. Hall (1992) and O’Neill (2007) also note that established disciplinary
73
practices, incentives, and associations within a university are not always responsive to external
pressures or changing market dynamics; rather, they often prioritize their own needs, such as the
pursuit of organizational security and disciplinary tradition, which in turn leads to variation across
universities.
Although both perspectives have productively suggested reasonable explanations for the
issue, all of the historical, social, and organizational factors that they have identified could be
discussed simultaneously to obtain a holistic understanding of change over time. Several studies
indeed recognize that the disciplinary location is “not necessarily a logical outcome of nonprofit
management studies” (Larson et al. 2003, p. 179); it also is a social, cultural, and political outcome
constructed “within a layered context” (Young 1998, p. 121) in which different stakeholder groups
negotiate their preferences and needs for nonprofit management knowledge. Specifically, Young
(1998, pp. 121-122) states that the development of nonprofit management programs is shaped by
“a series of simultaneous, interrelated games” among various constituencies at several layers,
including “the school or department in which a given program may be embedded, the university
itself, the community where the university is located, and the national and international nonprofit
and scholarly communities.” In a similar vein, Hall et al. (2001, p. 74) note that in practice, the
disciplinary venue might be a function of the academic community, the particular university, and
real-world conditions and needs. Overall, these arguments indicate that the academic homes in
which nonprofit management programs are framed and offered are not only a reflection of the
interests of university decision makers but also a response to the broader social and institutional
environment surrounding the university. Additionally, varied disciplinary settings across
universities may result from differences in the organizational and environmental situations that
each university faces.
74
Building on and extending this idea, this study proposes an integrated conceptual framework
that accounts for the existence of various disciplinary settings for nonprofit management research
and education, including public affairs, political science, business, social work, and social sciences.
The present study develops three central lines of argument based on neoinstitutionalism in
sociology and organizational analysis: (1) the dynamics of structuration of the nonprofit sector at
the national level; (2) characteristics of state nonprofit sector; and (3) organizational features of
universities.
Institutional Context
The first line of argument draws attention to the influence of the broader institutional
environment on the decision-making process for the disciplinary location of nonprofit management
programs within a university. Neoinstitutional organization theory emphasizes that organizational
decisions are enabled and constrained by an institutional environment that imposes shared
structures or practices on individual organizations within the relevant environment (e.g., March &
Olsen 1989; March 1994; Powell & DiMaggio 1991; Scott & Meyer 1994). Taking a macro-
sociological approach, this line of work points to differences in regulatory requirements, normative
expectations, and cultural beliefs over space and time as important mechanisms that produce
variability and change. From the perspective of the organizational institutionalism, then, individual
organizations conform to institutional logics—socially constructed, historical patterns of values,
beliefs, practices, and rules—because they provide legitimacy and serve as guides for action
(DiMaggio & Powell 1983; Meyer & Rowan 1977).
Sociological institutionalism provides a conceptual for understanding the emergence and
growth of nonprofit management in universities, and it may help to explain the disciplinary settings
75
for programs as well. In American society, the growth of the nonprofit sector and the rise of
managerial professionalism in the late 20th century created an institutional environment that
legitimizes the expansion of nonprofit management knowledge and practices (Hwang & Powell
2009). Indeed, increased nonprofit management discourse across the country has been evident in
the state’s involvement in the discourse via law and regulations, the formation of professional
associations and management support organizations, and the growth of academic journals,
professional magazines, and book publishing (O’Neill 1998). Such a regulatory, normative, and
cultural atmosphere at the national level may lead to an overall increase in program adoption across
university campuses because universities are subject to broader social understandings about the
general importance of nonprofit management. As the general nonprofit management discourse
across a country increases, more research and education programs are likely to be found in all
disciplinary settings:
Hypothesis 1a. The structuration of the nonprofit sector at the national level will have a
positive effect on the rate of program adoption in all disciplinary settings.
The neoinstitutional perspective also claims that knowledge of the historical stages that
accompany institutional changes is critical to understanding the different patterns and outcomes of
organizational responses to the same underlying institutional pressures (Thornton & Ocasio 1999;
Tolbert & Zucker 1983). Based on this line of reasoning, it could be argued that historically
situated institutional logics in the United States provide different rationales for each disciplinary
setting across time periods. If this is the case, then disciplinary settings for programs are likely to
reflect the prevailing logic at the time. Historical evidence suggests that the nonprofit sector has
been shaped by a series of social and political processes. These changes could lead to different
emphases for management programs. For example, before the early 1990s, relatively little attention
76
was paid to business-like mechanisms such as strategic planning, revenue diversification,
performance management, and organizational restructuring; the major concern at that time was the
growing collaborative relationship with government in delivering social services, which was
triggered by the “Great Society” effort (Boris & Steuerle 2006; Salamon 1995). However, business
models and practices have gained significant attention from nonprofit and academic communities
over the last decade because of social and political movements toward greater levels of devolution,
privatization, and professionalization in the 1990s and 2000s—including the penetration of
marketization ideologies into social welfare, the retrenchment policy of the Bush administration,
and tax and policy reforms re-engineering nonprofit financial accountability—that have changed
the nonprofit scene (Allard 2009; Berry & Arons 2003; Frumkin 2002; Grønbjerg & Salamon 2012;
Smith & Lipsky 1993). Such a historical shift may influence the disciplinary setting for nonprofit
management programs. In particular, because business schools are more likely to emphasize
market-based mechanisms and tools in their nonprofit management curriculums than other schools
(Mirabella & Wish 2000), I expect the following:
Hypothesis 1b. Historical paradigm shift towards business-like in the American welfare state
will alter the context for nonprofit management education, contributing to a greater
likelihood of program adoption in business.
State-Level Context of the Nonprofit Sector
The second line of argument considers variation in the disciplinary venues for nonprofit
management studies programs in relation to the characteristics of the nonprofit sector at the state-
level. According to neoinstitutional organization theorists (Fligstein 2001; Scott 2008; Thornton
& Ocasio 1999), institutions that shape organizational actions and decisions are embedded within
not only higher-order social logics but also organization-field logics. An organizational field, as a
77
meso-level unit between systems of society at the macro level and individual organizations at the
micro level, constructs particular local social orders that structure the decision making and
practices of the actors in a given field. Those local social orders are often overlapping or
contradictory and have an independent effect on organizational behavior; therefore, a meso-view
of the construction and consequences of local social orders should also be part of the study of
organizations. Borrowing an idea from this perspective, this study suggests that dynamics of
nonprofit sector in a state could influence the distribution of nonprofit management studies
programs.
It is plausible to hypothesize that statewide constituencies drive university decisions about
the disciplinary location for programs. Prior research on the relationship between communities and
the nonprofit sector (e.g., Ben-Ner & Van Hoomissen 1992; Corbin 1999; Grønbjerg & Paarlberg
2001; Matsunaga & Yamauchi 2004) documents that the size and scope of the nonprofit sector
dramatically differ across place according to region-specific characteristics such as community
size, age, cohesion, wealth, and political culture. Moreover, substantial variations in nonprofit
organizations’ size, scale, service field, and revenue structure shape different educational goals
and curricular needs (Cook 1988; DiMaggio 1988; Haas & Robinson 1998; Heimovics & Herman
1989; Rimer 1987). Thus, it can be argued that each state has a specific configuration of the
nonprofit sector that is designed to address local and regional issues and residents’ interests;
therefore, the academic demands and preferences of nonprofit stakeholders vary by community.
Because a certain discipline may be better positioned to meet the particular needs of a region’s
nonprofit leaders and managers, the disciplinary settings preferred by community members and
universities will differ across regions.
78
Based on the discussion above, I argue that the size of some nonprofit service fields and the
degree of commercialization of nonprofit revenues will have a positive effect on the academic
setting for nonprofit management programs:
Hypothesis 2a. A larger public affairs sector will have a positive effect on the rate of program
adoption in public affairs/political science.
Hypothesis 2b. A larger human services sector will have a positive effect on the rate of
program adoption in public affairs/political science and social work/social sciences.
Hypothesis 2c. A larger proportion of commercial revenues will have a positive effect on the
rate of program adoption in business.
University Dynamics
The third line of argument emphasizes internal organizational drivers for the decision making
regarding where nonprofit management programs are housed within a university. One stream of
the institutionalist literature on organizations (e.g., Greenwood & Hinings 1996; Hirsch &
Lounsbury 1997; Oliver 1991) has focused on the internal dynamics of organizations in explaining
why organizational responses to the same institutional pressures diverge. In this line of research,
scholars argue that individual organizations define and interpret institutional processes within a
given institutional context differently due to “differences among organizations in the amount of
pressure they experience, in their characteristics, or in their location within the field” (Scott 2008).
Hence, even in the same institutional context, organizational responses to institutional pressures
could result in divergent rather than convergent outcomes. Taking such an argument into
consideration, this study predicts that the disciplinary setting for nonprofit management programs
perhaps varies because of differences in universities’ organizational characteristics and internal
politics.
79
More specifically, universities vary in their internal politics, through which disciplines claim
their own academic territories, compete for resources, and pursue their interests, and such variation
influences the establishment of nonprofit management programs and their disciplinary location.
This argument builds on sociological studies of the formation of new disciplines (e.g., Camic 1995;
Small 1999; Wood 2012), which maintain that different local conditions in and politics among key
departments/schools at universities result in different methods, scope, content, and theory for
newly created disciplines. In fact, the disciplinary location of a nonprofit management studies
program within a university seems to be an outcome of internal politics, through which various
“receptor sites” (Frank et al. 2000) of nonprofit management—disciplines that are able to receive
and interpret the higher-order social logics of nonprofit discourse and transmit the information to
local university stakeholders—compete to adopt programs on their turf.
Such a political process may differ across universities depending on the types of receptor
sites that are engaged in the process and the inequalities in power and resources among them. That
is, a certain receptor site’s participation in the competition may be determined by whether the dean,
faculty, students, and external funders of that discipline have any special interest in or commitment
to nonprofit management issues. In addition, each receptor site may have more or fewer
resources—including financial assets, infrastructure, market power, political support, reputation,
and relational networks—available for program adoption. As a result, each university is likely to
have its own well-developed, resource-rich receptor site for nonprofit management studies. To
summarize, local politics surrounding the adoption of nonprofit management programs should
vary with the conditions of the key disciplines in a given university, which might account for the
divergence in the disciplinary settings for these programs.
80
Based on the discussion above, I suggest that the size of some academic programs will have
a positive effect on the academic home for nonprofit management programs:
Hypothesis 3a. A larger public affairs/political science program will lead to a greater
likelihood of program adoption in public affairs/political science.
Hypothesis 3b. A larger business program will lead to a greater likelihood of program
adoption in business.
Hypothesis 3c. A larger social work/social sciences program will lead to a greater likelihood
of program adoption in social work/social sciences.
In summary, the conceptual framework presented here suggests that the broader institutional
context, the state-level context of the nonprofit sector, and university dynamics will
simultaneously influence the disciplinary location of nonprofit management studies programs
within a university. Furthermore, because of differences in these institutional, social, and
organizational features, I expect the disciplinary venue of nonprofit management programs to vary
significantly across space and time. In the following section, I describe the data, variables, and
empirical model specification used to examine these general theoretical arguments.
4.3. Data and Methods
The quantitative dataset consists of a panel of 1,446 American universities and colleges from
1971 to 2011.
12
The sample was drawn from the population of universities and colleges that
offered a Bachelor’s, Master’s, or Ph.D. degree based on Carnegie Classification of Institutions of
12
This dataset is essentially the same as that used in Chapter 3. Although the sample and most of the covariates are
exactly the same, the dataset used here includes a different set of dependent variables, an additional set of covariates,
and a different number of yearly spells.
81
Higher Education for 1976, 1987, 1994, and 2000. Two-year, vocational, and other types of
specialized institutions that have remarkably different educational goals and organizational
structures from those of four-year institutions (Kraatz & Zajac 1996; Olzak & Kangas 2008) and
institutions located in U.S. territories (e.g., Guam, Puerto Rico, and United States Virgin Islands)
were excluded from this population. A total of 1,451 universities and colleges were initially
identified as a valid sample, but I dropped five left-censored universities that adopted their first
nonprofit management programs before 1971, resulting in 1,446 universities and colleges.
13
The
unit of analysis is the university-year, and the final sample includes 59,286 university-years of
data—1,446 universities multiplied by 41 years of data for each university.
Dependent Variables
The dependent variable used in the analysis is a categorical variable that specifies the four
distinct patterns of nonprofit management program adoption in universities: (1) absence of
program adoption; (2) program adoption within the school of public affairs or political science
(PA/PS); (3) program adoption within the school of business (BUS); and (4) program adoption
within the school of social work or within other social sciences, including sociology, anthropology,
and psychology (SW/SC). I defined program adoption as the creation of the first stand-alone degree,
non-degree, or research center for nonprofit management studies.
14
The variable was coded as 0 if
a university had not yet adopted any program in a given year (t) and in the following year (t+1), 1
13
Because only a very small percentage of the universities in the sample were left-censored, the problem of selection
bias is unlikely to be serious here.
14
Similar to the procedure followed in Chapter 3, I identified Bachelor’s, Master’s, and doctoral degrees in nonprofit
studies (e.g., Bachelor of Arts in Philanthropic Studies, Master of Arts in Nonprofit Leadership and Management,
Master of Science in Nonprofit Organizations) as stand-alone degree programs; undergraduate- and graduate-level
minors, tracks, specializations, concentrations, and certificates in nonprofit studies (e.g., MPA/MBA/MSW
Concentration in Nonprofit Management, the Nonprofit Leadership Alliance Certificate, etc.) as non-degree programs;
and within-university institutes for nonprofit education, research, training, and consulting as research centers.
82
if a university established its first program housed by PA/PS in the following year (t+1), 2 if by
BUS in the following year (t+1), and 3 if by SW/SC in the following year (t+1). The timing of
adoption was regarded as the academic year when the university actually launched its first program
and started offering it to students rather than as when the university or college started designing
the program or obtained approval from the provost’s council or board of regents.
Data on the dependent variable were generated using the exact same procedure as that
employed in Chapter 3: a scrupulous review of each institution’s online archives and follow-up
email correspondences with program directors, department chairs, and school deans. The only
difference in the process of data collection is that my focus was not on when and what type of
programs the universities first adopted but on when and in which department/school their first
programs were housed. Finally, the collected data on 281 universities were cross-checked based
on historical documents obtained from each institution’s online archives, if possible, to reduce the
problem of retrospective bias.
Independent Variables
I included three sets of independent variables to capture the effects of macro-level
institutional, meso-level regional, and micro-level organizational factors. The data were compiled
from various publicly available sources. I estimated values for missing independent (and control)
variable data through interpolation using the closest values available and averaging values around
those that are missing; listwise deletion may drastically reduce the overall sample size and thus
likely lead to biased results.
15
15
Appendix D shows descriptive statistics of imputed covariates, comparing imputed values with original ones.
83
Institutional Context Factors. I measured a series of indicators of the structuration of the
nonprofit sector at the national level. I initially created six measures of the extent of nonprofit
discourse saturation: Nonprofit organization, Government grant, State association, University
program, Dissertation, and News article. The first three—Nonprofit organization, Government
grant, and State association—are measures that capture changes in the size and scale of the U.S.
nonprofit sector; Nonprofit organization is the count of 501(c)(3) charitable organizations obtained
from the Business Master Files provided by the NCCS (2012); Government grant is the calculation
of federal government grants to 501(c)(3) organizations in thousands of U.S. dollars, which was
obtained from the SOI Tax Stats provided by the IRS (2011); and State association is the
cumulative count of states with a statewide professional association, which was calculated from
the National Council of Nonprofits website (http://www.councilofnonprofits.org/). The next two
measures—University program and Dissertation—track the changing interest in nonprofit studies
within the academic community; University program is the cumulative count of universities
offering nonprofit management programs, which was drawn from the same dataset used for the
dependent variables, and Dissertation is the count of doctoral dissertations with the keywords
“nonprofit” or “nongovernmental” in their abstracts, which was collected from the ProQuest
database. The last measure, News article, captures the ebb and flow of public attention on nonprofit
management issues; News article is the count of news articles with the keywords “nonprofit,” “not-
for-profit,” “charitable sector,” “philanthropic sector,” “voluntary sector,” “civil society,” “NGO,”
“non-governmental organization,” or “philanthropy” in their headings, which were obtained from
the Lexis-Nexis database. All six of the measures are time-varying, lagged by one year and logged
to reduce skewness, except for State association. In the analysis, however, I combined all of the
84
measures into one factor, Structuration, because the six measures are highly correlated with a
single underlying factor.
I also computed a dummy variable, Business logic, to capture the effect of the prevailing
institutional logic in a particular historical period. I distinguished between an era of the state before
2002 and an era of entrepreneurial entities after 2002. Such a periodization was selected for
theoretical reasons because of a series of prominent social and political movements around the
year 2002—such as the 2002 enactment of the Sarbanes–Oxley Act and the welfare retrenchment
of the Bush administration (Grønbjerg & Salamon 2012)—that may have caused a substantial shift
in the management of nonprofits toward the use of business models and practices. Moreover, the
results of graphical and sensitivity analyses showed that the cutoff point of 2002 is the most
appropriate to periodize the historical time that I analyzed.
16
To test the period effect of business-
oriented logic, the Business logic variable was scored 1 for the 2002-2011 timeframe and 0 for
1971-2011.
State Nonprofit Sector Factors. To determine the effects of the regional nonprofit
community setting, I used indicators of the size of some nonprofit service fields and the degree of
commercialization of nonprofit revenues at the state level. Data on these indicators were obtained
from the Business Master Files provided by the NCCS (2012) and are only available from 1989 to
the present; to the best of my knowledge, publicly available, state-level data on the U.S. nonprofit
sector only date back to the early 1990s. Accordingly, I tested the state nonprofit sector setting
16
Using an exponential model including all of the covariates, I estimated six time periods set for four years before
2002 and two years after 2002. While the year 2002 is the best fit, the statistically significant differences between the
two historical time periods remain if I select cut-off points at any time after 2001.
85
indicators in a separate analysis with a shorter time frame—from 1990 to 2011—in the following
section.
I included two measures, Public affairs field and Human services field, to capture the extent
to which the size of a particular subfield of the nonprofit sector affected the disciplinary settings
for nonprofit management programs. Based on the categorization of the National Taxonomy of
Exempt Entities Classification System, I calculated the proportion of “Public/Social Benefit” and
“International/Foreign Affairs” nonprofits to create the Public affairs field variable and the
proportion of “Human Services” nonprofits to create the Human services field variable. In addition,
as a proxy for commercialization, I used Commercial revenue which is the proportion of earned,
non-donated income generated by nonprofits (e.g., program service fees, dues and assessments,
special events revenue, and profits from sales and inventory).
17
Likewise, all of the variables are
time-varying and lagged by one year.
University Dynamics Factors. As indicators of university dynamics, I focus on the size of
some academic programs at each institution. Data on these indicators were gathered from the
HEGIS for the years 1970 to 1986 and the IPEDS for the years 1987 to 2010. To measure
differences in power and resources among receptor sites of nonprofit management studies, I
constructed indicators of the relative size of each receptor site—including the schools of public
affairs/political science, business, and social work/social sciences. PA/PS degree, BUS degree, and
SW/SC degree are the logarithms of the proportions of conferred Bachelor’s, Master’s, and
doctoral degrees in public affairs/political science, business, and social work/social sciences,
respectively. All of the variables are time-varying and lagged by one year.
17
Because the time-series commercial income data were volatile from year to year, I smoothed the data by creating
weighted moving averages.
86
Control Variables
I controlled for state-level socioeconomic and political conditions using a measure of
demographic heterogeneity, support for social services, political orientation, and nonprofit sector
size. I expect that region-specific characteristics such as size, age, cohesion, wealth, and political
culture may be associated with the different educational goals and curricular needs of nonprofit
managers across regions. Urbanization is used as the population heterogeneity measure, which is
computed as the proportion of the population living in a Standard Metropolitan Statistical Area in
a state based on data from the U.S. Census Bureau (2013). To measure a state’s support for social
services, Welfare expenditure, which is the proportion of state government expenditures on social
welfare, is calculated from data obtained from the U.S. Census Bureau (2013). Citizen liberalism
captures the political orientation of the citizens in a state. I borrowed Berry et al.’s (2010) index,
which ranges from 0 (high conservatism) to 100 (high liberalism). As a state nonprofit sector size
measure, I constructed Nonprofit assets, which is the logarithm of nonprofit assets per capita and
obtained from the Business Master Files provided by the NCCS (2012). All of these variables are
time-varying and lagged by one year.
18
As university-level controls, I focus on each institution’s size, structure, reputation, and
campus diversity with an expectation that differences in universities’ organizational characteristics
are associated with the variation in disciplinary settings. University revenue, a proxy for size, is
the logarithm of total university revenue in millions of U.S. dollars. Public is a dummy variable
for the structural features of universities and colleges, which take a value of 1 if a university or
18
I also tested several other measures, such as the total population, per capita personal income, per capita state
expenditures, and citizen liberalism. However, these measures were omitted from the final models because they have
major correlational problems with the included variables or no significant effects across the estimated models.
87
college is public and 0 otherwise.
19
Doctoral/Research is a dummy variable that indicates whether
a university or college is highly selective, which is coded 1 if a university or college is categorized
as a doctorate-granting institution with a very high level of research activity according to the 2005
Carnegie Classification of Institutions of Higher Education and 0 otherwise. Black/Hispanic and
Female, the proportion of Bachelor’s, Master’s, and doctoral degrees conferred to African
American/Hispanic students and female students, respectively, are included as proxies for campus
diversity. All of the variables are time-varying and lagged by one year, except for two dummy
variables—Public and Doctoral/Research—which remain the same for each institution throughout
all of the time periods.
To control for the passage of time, I originally computed a continuous measure of the
calendar year. The expectation was that the passage of time would be positively related to the
increasing propensity to adopt programs within any type of disciplinary setting. However, I
excluded this measure from the main analyses due to its high correlation with Structuration. Thus,
I used Structuration as a substantive indicator.
Table 4.1 presents the definitions and descriptive statistics for the independent and control
variables used in the analysis. Tables 4.2 and 4.3 report correlation matrices for the two sets of
analyses: a 41-year timeframe analysis (1971-2011) and a 22-year analysis (1990-2011),
respectively.
19
I considered but dropped a number of other measures of structural characteristics, such as whether a university is a
religious, flagship, land grant, or historically black institution, due to multicollinearity problems with included
variables.
88
Table 4.1. Definitions and Descriptive Statistics for Covariates (N=1,446 universities and 59,286 university-years)
Category Variable Definition Mean SD Min Max
Institutional
context
Structuration Factor of the six institutional context measures (logged) -7E-08 1.00 -0.98 2.34
- Nonprofit org. # of 501(c)(3) charitable organizations (logged) 14.14 0.53 13.47 15.06
- Gov. grant Government grants to 501(c)(3) organizations, in thousand USD (logged) 18.53 0.72 17.77 19.86
- State association Cum. # of states with a statewide association of nonprofits 19.24 13.90 1 43
- Univ. program Cum. # of universities with nonprofit management programs (logged) 4.42 1.76 1.00 6.61
- Dissertation # of doctoral dissertations with a keyword “nonprofit” in abstracts (logged) 5.39 1.93 1.00 7.66
- News article # of news articles with a keyword “nonprofit” in headings (logged) 7.92 1.64 5.06 10.78
Business logic Dummy for the 2002-2011 time period 0.22 0.41 0 1
State nonprofit
sector setting
Public affairs field* % of NPOs categorized as “Public Benefit & International Affairs,” by state 27.19 3.93 16.04 47.24
Human services field* % of NPOs categorized as “Human Services,” by state 24.58 2.55 17.97 42.89
Commercial revenue* Commercial income of NPOs as a % of total revenues, by state 71.56 7.49 30.32 93.86
University
dynamics
PA/PS degree % of degrees in Public Affairs and Political Science conferred (logged) 1.60 0.76 1.00 5.62
BUS degree % of degrees in Business conferred (logged) 3.53 1.28 1.00 5.62
SW/SC degree % of degrees in Social Work and Social Sciences conferred (logged) 2.83 0.97 1.00 5.62
State-level
controls
Urbanization % of population living in a Standard Metropolitan Statistical Area, by state 72.78 14.00 32.20 100
Welfare expenditure Social welfare expenditures as a % of total state expenditures, by state 19.75 5.66 4.08 38.75
Citizen liberalism Citizen ideology index developed by Berry et al. (2010), by state 50.75 15.78 7.04 95.97
Nonprofit assets* Total assets of NPOs per capita, by state (logged) 9.94 0.63 8.50 12.65
Univ.-level
controls
University revenue Total university revenues, in million USD (logged) 4.83 1.37 1.00 10.30
Public Dummy for public universities 0.35 0.48 0 1
Doctoral/Research Dummy for highly prestigious doctoral/research universities 0.07 0.25 0 1
Black/Hispanic % of degrees conferred to African American and Hispanic students 12.58 21.17 0 100
Female % of degrees conferred to female students 53.04 20.54 0 100
* Data available from 1990 to 2011 only.
Note: All the listed variables are lagged by one year, except three dummy variables—Business logic, Public, and Doctoral/Research.
89
Table 4.2. Correlation Matrix of Covariates, 1971-2011
1 2 3 4 5 6 7 8 9 10 11 12
1. Structuration
2. Business logic 0.87
3. PA/PS degree -0.21 -0.15
4. BUS degree 0.20 0.12 0.02
5. SW/SC degree 0.05 0.03 0.35 0.18
6. Urbanization 0.14 0.10 0.06 -0.02 0.04
7. Welfare expenditure 0.54 0.41 -0.11 0.05 0.07 0.36
8. Citizen liberalism 0.28 0.24 -0.02 -0.08 0.03 0.39 0.46
9. University revenue 0.26 0.20 0.27 0.19 0.28 0.21 0.22 0.16
10. Public 0.00 0.00 0.11 0.09 0.04 -0.06 -0.08 -0.09 0.43
11. Doctoral/Research 0.00 0.00 0.16 0.00 0.06 0.10 0.01 0.03 0.55 0.17
12. Black/Hispanic 0.12 0.10 0.04 0.15 0.08 0.01 0.00 -0.14 0.00 0.05 -0.07
13. Female 0.27 0.19 0.00 0.22 0.42 0.02 0.21 0.11 0.11 0.00 -0.10 0.19
Table 4.3. Correlation Matrix of Covariates, 1990-2011
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1. Structuration
2. Public affairs field 0.04
3. Human services field 0.12 -0.47
4. Commercial rev. -0.08 -0.18 0.21
5. PA/PS degree -0.08 -0.05 -0.01 -0.08
6. BUS degree 0.01 0.01 -0.04 0.05 -0.01
7. SW/SC degree 0.02 0.03 -0.01 -0.04 0.26 0.05
8. Urbanization 0.04 0.03 -0.14 -0.34 0.11 -0.03 0.04
9. Welfare expenditure 0.31 0.26 -0.17 -0.10 -0.05 -0.08 0.03 0.16
10. Citizen liberalism 0.30 0.03 0.25 -0.14 0.01 -0.14 0.01 0.36 0.32
11. Nonprofit assets 0.56 0.19 0.13 -0.31 -0.02 -0.08 0.04 0.35 0.40 0.64
12. University revenue 0.14 0.02 0.00 -0.11 0.31 0.01 0.26 0.19 0.10 0.12 0.15
13. Public 0.00 -0.09 0.03 0.01 0.15 0.03 0.08 -0.07 -0.10 -0.09 -0.11 0.42
14. Doctoral/Research 0.00 0.01 -0.02 -0.04 0.18 -0.03 0.04 0.10 0.00 0.03 0.03 0.57 0.17
15. Black/Hispanic 0.10 -0.03 -0.14 -0.16 0.06 0.12 0.04 0.04 0.00 -0.11 -0.07 -0.05 0.05 -0.07
16. Female 0.10 0.03 0.01 -0.07 0.00 0.13 0.36 0.00 0.09 0.06 0.09 -0.06 -0.03 -0.16 0.18
90
Modeling
This study employed continuous-time, competing risks event history analysis to understand
not only the transition rate of each university or college from not adopting to adopting its first
nonprofit management program within a certain disciplinary setting over time but also the factors
that contributed to the different disciplinary settings in which the first program was housed.
Competing risks event history analysis is commonly used in situations in which an observation
can make a transition into one of several states at different points in time, and each state is assumed
to have its own transition model that governs both the timing of the occurrence and the factors that
lead to that state. It captures heterogeneity across different types of states by providing a separate
transition rate equation and a unique set of coefficients for each type of destination state (Allison
2010; Blossfeld et al. 2007; Box-Steffensmeier & Jones 2004). Through the use of competing risks
analysis, therefore, I can estimate different transition processes for three possible destination states
of the first founding of a university’s nonprofit management program—PA/PS, BUS, and SW/SC.
The main analyses were performed at the individual university level, and only the first
adoption of a research and education program within the university was considered. Although
program adoption could be a repeatable event from the perspective of universities and colleges, I
focus on first adoptions only because second and subsequent adoptions are likely to be affected by
and thus differ from first adoptions, possibly leading to biased results. In addition, I used
continuous-time models to define the time to first adoption as continuous rather than discrete-time
models in which the time to transition is measured on a discrete scale. I could not simply ignore
the discreteness of my time unit—an academic year, but I decided to take the continuous-time
approach because of the simplicity of its application and the fact that the results from the two
models are nearly the same in most cases (Allison 1982; Yamaguchi 1991). Taken together, in the
91
main analyses, it is assumed that each university is at risk of adopting its first nonprofit
management program within one of the three disciplinary settings at any point in time between
1971 and 2011 (or between 1990 and 2011). In addition, once an institution adopts its first program
within one disciplinary setting, the institution is assumed to no longer be at risk of adopting a
program within other types of disciplinary settings.
I modeled the shape of the transition rate using exponential models in which the risk of a
transition occurring is invariant with respect to time, and all of the fluctuation in the transition rate
comes from the covariates. The exponential model is particularly appropriate here because the
main focus of this study is to explain the different disciplinary settings for nonprofit management
programs using various social, cultural, and organizational factors rather than historical time.
Moreover, based on the assumption of time-constant baseline transition rates, I can approximate
the effects of a large number of theoretically identified covariates with ease; such an assumption
is standard in analyses that include numerous covariates (Box-Steffensmeier & Jones 1997). In the
competing risks exponential model, the transition rate of the first program adoption, r(t), is defined
under the three competing risks of disciplinary settings—r(t)1, r(t)2, and r(t)3, respectively—and
is log-linearly related to the n covariates (Xn) as follows:
log r(t)k = 𝛽 0,𝑘 + 𝛽 1,𝑘 𝑋 1
+ 𝛽 2,𝑘 𝑋 2
+ ⋯ + 𝛽 𝑛 ,𝑘 𝑋 𝑛 (k=1, 2, and 3)
where log r(t)k is the log of the transition rate for the type k disciplinary setting at time t, 𝛽 𝑛 ,𝑘 are
the type k-specific parameters to be estimated (with 𝛽 0,𝑘 representing the time-invariant baseline
transition rate for the type k disciplinary setting), and 𝑋 𝑛 are the covariates. Consequently, the
92
main analyses produce three sets of transition rates and parameter estimates, with one
corresponding to each of the three types of disciplinary settings.
20
4.4. Results
Figure 4.1 provides a graphic profile of the disciplinary settings in which the initially created
nonprofit management programs in universities are placed between 1971 and 2011. By 2011, a
total of 281 universities and colleges in the sample of 1,446 had established their first program
within one of the four disciplinary settings: 101 adopted programs (35.9%) are housed in PA/PS,
62 (22.1%) are in BUS, 38 (13.5%) are in SW/SC, and 80 (28.5%) are in other disciplines (Other)—
including arts and sciences, professional studies, or interdisciplinary studies. As Figure 4.1 shows,
Figure 4.1. Disciplinary Settings for Academic Programs, Cumulative Count (1971-2011)
20
I did not utilize piecewise exponential models with period-specific effects and instead used a time dummy. It was
because of the small number of cases for the dependent variables, as well as the complexity.
93
the rate of program adoption considerably varies for the different disciplinary settings. Specifically,
the adoption rates begin to diverge in the early 1980s, and this process accelerates after the 1990s,
with the highest rate, on average, for PA/PS, followed by Other, BUS, and SW/SC. The rate for
PA/PS increases more sharply than that for the other settings, and SW/SC has the lowest rate of
increase—although not essentially flat—throughout the time period.
Table 4.4 reports the results of exponential competing risks analyses with a 41-year
timeframe from 1971 to 2011, which provide tests for Hypotheses 1 and 3. In this table, I present
two full models (Models 1 and 2) that differ only by the specific institutional-context measure used
because those two measures—Structuration and Business logic—could not be included in the same
model due to their high correlation (see the correlation matrix in Table 4.2). Each model is
comprised of three sub-models; the three sub-models present the exponential estimates of the type-
specific transition rates corresponding to each of the three disciplinary settings. Table 4.5 presents
the results of the exponential competing risks analyses with a 22-year timeframe from 1990 to
2011, which provides a test of Hypothesis 2. Model 3 is a baseline model for comparison that
includes the state-level control variables and only the independent variables that are associated
with the state nonprofit sector setting. Model 4 is a nested model for testing Hypothesis 2 that
includes all of the identified control and independent variables.
21
Overall, the results in Table 4.4
provide strong support for the influence of institutional context and university dynamics as
identifying factors that are predictive of variation in the disciplinary setting for nonprofit
management research and education programs. However, the results in Table 4.5 provide little
support for the effects of the state nonprofit sector on the different patterns of adoption.
21
Similar to the procedure used for Table 4.4, I tested the full model with Structuration and Business logic separately.
However, I report only the results for Structuration here because the results for Business logic are quite similar to
those derived from Structuration.
94
Table 4.4. Competing-Risks Models, 1971-2011
Category Variable Model 1 Model 2
PA/PS BUS SW/SC PA/PS BUS SW/SC
Institutional
context
Structuration (log) 0.263**
(0.117)
0.657***
(0.149)
0.318*
(0.190)
Business logic 0.229
(0.239)
0.577*
(0.295)
0.420
(0.382)
University
dynamics
PA/PS degree (log) 0.951***
(0.163)
-0.000
(0.204)
-0.094
(0.269)
0.903***
(0.163)
-0.132
(0.205)
-0.138
(0.268)
BUS degree (log) 0.647***
(0.177)
0.558***
(0.176)
0.082
(0.186)
0.676***
(0.175)
0.603***
(0.175)
0.112
(0.185)
SW/SC degree (log) -0.199
(0.191)
-0.055
(0.181)
0.305
(0.296)
-0.195
(0.189)
-0.071
(0.178)
0.298
(0.293)
State-level
controls
Urbanization 0.017*
(0.009)
0.036***
(0.013)
-0.012
(0.013)
0.015*
(0.009)
0.033**
(0.013)
-0.013
(0.013)
Welfare expenditure 0.054**
(0.023)
-0.029
(0.031)
0.026
(0.037)
0.065***
(0.023)
0.000
(0.029)
0.036
(0.036)
Citizen liberalism -0.006
(0.008)
-0.017
(0.011)
0.001
(0.013)
-0.005
(0.008)
-0.012
(0.011)
0.003
(0.013)
University-level
controls
Univ. revenue (log) 1.027***
(0.128)
0.613***
(0.143)
1.014***
(0.179)
1.078***
(0.126)
0.700***
(0.140)
1.048***
(0.176)
Public 1.456***
(0.272)
-1.532***
(0.374)
0.335
(0.372)
1.440***
(0.271)
-1.548***
(0.373)
0.326
(0.371)
Doctoral/Research -0.376
(0.328)
-0.120
(0.553)
-0.885
(0.561)
-0.420
(0.326)
-0.299
(0.546)
-0.929*
(0.557)
Black/Hispanic -0.004
(0.006)
-0.016*
(0.010)
0.000
(0.009)
-0.003
(0.006)
-0.012
(0.009)
0.001
(0.008)
Female 0.032***
(0.012)
0.022**
(0.010)
0.022
(0.016)
0.037***
(0.011)
0.026***
(0.010)
0.025
(0.015)
Constant -20.716***
(1.710)
-14.161***
(1.636)
-15.218***
(2.126)
-21.512***
(1.675)
-15.367***
(1.606)
-15.823***
(2.078)
LRχ² 309.088 109.935 75.364 305.097 94.575 73.84
Log-likelihood 182.974 42.234 11.255 180.979 34.553 10.493
# of adoptions 101 62 38 101 62 38
# of universities 1,446 1,446 1,446 1,446 1,446 1,446
# of yearly spells 56,037 56,037 56,037 56,037 56,037 56,037
* p<0.1; ** p<0.05; *** p<0.01. Standard errors in parentheses.
95
Table 4.5. Competing-Risks Models with State Nonprofit-Sector Effects, 1990-2011
Category Variable Model 3 Model 4
PA/PS BUS SW/SC PA/PS BUS SW/SC
Institutional
context
Structuration (log) -0.084
(0.158)
0.645***
(0.199)
0.301
(0.267)
State nonprofit
sector setting
Public affairs field 0.050
(0.034)
-0.005
(0.044)
-0.021
(0.060)
0.057*
(0.034)
0.018
(0.044)
-0.018
(0.060)
Human services field 0.051
(0.049)
-0.007
(0.061)
-0.08
(0.101)
0.053
(0.044)
0.006
(0.072)
-0.095
(0.107)
Commercial revenue -0.016
(0.015)
0.029
(0.022)
0.041
(0.031)
-0.011
(0.016)
0.013
(0.022)
0.042
(0.032)
University
dynamics
PA/PS degree (log) 1.009***
(0.171)
-0.011
(0.228)
-0.016
(0.294)
BUS degree (log) 0.764***
(0.234)
0.487***
(0.184)
-0.034
(0.200)
SW/SC degree (log) -0.175
(0.216)
-0.136
(0.188)
0.326
(0.325)
State-level
controls
Urbanization 0.014
(0.010)
0.030**
(0.015)
-0.004
(0.016)
0.013
(0.010)
0.035**
(0.014)
-0.001
(0.016)
Welfare expenditure 0.016
(0.026)
-0.057*
(0.034)
0.02
(0.045)
0.042
(0.026)
-0.073**
(0.037)
0.015
(0.047)
Citizen liberalism -0.012
(0.012)
-0.013
(0.014)
-0.007
(0.018)
-0.013
(0.012)
-0.008
(0.015)
-0.006
(0.018)
Nonprofit assets (log) 0.125
(0.264)
0.382
(0.317)
0.346
(0.414)
0.236
(0.284)
-0.235
(0.400)
0.042
(0.519)
University-level
controls
Univ. revenue (log) 0.963***
(0.124)
0.565***
(0.145)
0.960***
(0.190)
1.060***
(0.141)
0.622***
(0.154)
0.930***
(0.202)
Public 1.618***
(0.313)
-1.587***
(0.412)
0.56
(0.418)
1.939***
(0.334)
-1.611***
(0.416)
0.554
(0.421)
Doctoral/Research -0.533
(0.346)
-0.661
(0.660)
-1.228*
(0.676)
-0.588
(0.359)
-0.531
(0.665)
-1.251*
(0.686)
Black/Hispanic 0.003
(0.006)
-0.006
(0.009)
0.003
(0.009)
-0.003
(0.006)
-0.015
(0.010)
0.000
(0.009)
Female 0.001
(0.011)
0.013
(0.010)
0.028*
(0.016)
0.016
(0.016)
0.021*
(0.012)
0.023
(0.018)
Constant -16.116***
(3.251)
-15.423***
(4.266)
-17.784***
(5.517)
-24.230***
(3.835)
-12.114**
(4.821)
-15.238**
(6.397)
LRχ² 204.923 40.801 44.136 260.518 63.226 46.387
Log-likelihood 173.818 39.222 15.677 201.616 50.435 16.803
# of adoptions 88 55 32 88 55 32
# of universities 1,407 1,407 1,407 1,407 1,407 1,407
# of yearly spells 28,785 28,785 28,785 28,785 28,785 28,785
* p<0.1; ** p<0.05; *** p<0.01. Standard errors in parentheses.
96
Institutional Context Effects
The findings in Table 4.4 strongly support Hypothesis 1a, i.e., the structuration of the
nonprofit sector at the national level is associated with increasing adoption rates across the
disciplinary settings. Model 1 shows a positive and statistically significant effect of Structuration
for all of the sub-models, which indicates that universities and colleges are more likely to adopt
programs in any part of their organization as nonprofit management becomes legitimized and
recognized as an important logic of American society. In other words, the process of structuration
in the nonprofit sector—including the growth of nonprofits in size and scope (measured by
Nonprofit organization, Government grant, and State association), the increasing interest in
nonprofit studies within the academic community (measured by University program and
Dissertation), and the growing public attention on nonprofit management issues (measured by
News article), as shown in Appendix E—appears to lead to an overall increase in program adoption
across university campuses.
Hypothesis 1b, which proposed the effect of the prevailing institutional logic in a particular
historical period, is also supported by the findings in Table 4.4. As reported in Model 2, adoptions
in BUS indeed accelerate more in the post-2002 period, when the ideas and practices of business-
like management are pervasive throughout the nonprofit sector. By contrast, adoptions in PA/PS
and SW/SC do not significantly increase compared with the pre-2002 period; both models show
potential increases in the adoption rates, but the estimates are not statistically significant at the 0.1
level. This result is consistent with the initial assumption that universities and colleges will
accommodate historically situated institutional logics by offering programs that are conceptually
related to such logics. As Mirabella and Wish (2000) empirically find, because programs in BUS
tend to emphasize private sector tools such as fundraising, marketing, and strategic management
97
in their curriculums more than do programs in PA/PS and SW/SC, adopting programs within BUS
might be preferred during the business-model-oriented period. It seems that the institutional
change in the discourse of nonprofit management, which was particularly facilitated by the passage
of laws and government regulations, is an important factor in shaping the dominant disciplinary
setting of the age.
State Nonprofit Sector Effects
None of the hypotheses regarding the state nonprofit sector setting—Hypotheses 2a, 2b, and
2c—are supported by the findings in Table 4.5. In Model 3, the two subsector size measures—
Public affairs field and Human services field—are positively related to the rate of program
adoption in PA/PS but are not statistically significant. The revenue structure measure—
Commercial revenue—is positively related to the rate of program adoption in BUS and SW/SC but
is also not statistically significant. Even when including the other main independent variables in
the model, these measures are not statistically significant. As Model 4 shows, the effects of the
newly added institutional and organizational variables are statistically strong and consistent with
the results of Model 1. However, the regional variables yield statistically nonsignificant effects
overall, which is contrary to my initial expectation that variation in the disciplinary setting can be
explained by regional differences in the nonprofit sector.
22
Such unexpected results may in part be attributed to three factors. One possibility is that
universities are no longer expected to mainly serve the needs of nonprofit stakeholders in a
geographic area. With the growth of the nonprofit sector worldwide and the development of online
22
We should be cautious in interpreting these findings, because the results of the multinomial logistic competing risks
models in Appendix F indicate that the state nonprofit sector factors might have significant effects that vary by time
period. The models show that adoption rates and patterns before 2006 were more likely to be influenced by the state
nonprofit sector conditions, while those after 2006 were less likely to be so influenced.
98
education tools in recent years, the targeted stakeholder group for nonprofit management research
and education might be expanded to include not only regional but also national and international
constituencies. In fact, one of the fastest-growing nonprofit management programs in the U.S. over
the last decade is an online curriculum—with an increase of 650% from 2002, when there were
only 10 universities offering 10 online courses, to 2014, when the numbers increased to 75
universities offering more than 380 courses (Mirabella 2014)—that attracts individuals of all
localities and nationalities. The nonsignificant effects of the regional variables may therefore be
the result of changes in the profile of stakeholders and their transregional demands for programs.
An additional possibility is that my data collection and variable operationalization should
have been conducted at a much more micro level—such as at the county, city, or community level
rather than the state level—to precisely measure the regional effects. If nonprofit management
programs have largely been designed for regional nonprofit constituencies, the main focus of the
programs might have been on issues that are important to the neighborhood rather than those of
the whole state. This difference may be partly because a majority of on-campus students tend to
work for local nonprofits that are located approximately one hour from the university and prefer
engagement in the program within their own local communities (Wilson & Larson 2002). Hence,
although data limitations prohibited me from observing statistically significant effects of regional
variables, such effects may have been observed with more accurate, micro-level measurements of
the local nonprofit sector setting.
The other possibility is that universities may have “an unusual degree of autonomy and... a
capacity to resist pressures from external constituencies” (Hall 1992, 403), as the supply-side
perspective reviewed earlier argues. Universities are often driven by powerful internal forces that
reinforce independence from the marketplace, and thus their own legitimate interests—such as
99
organizational security and disciplinary tradition—could be prioritized over the legitimate interests
of external stakeholders (Hall 1992; O’Neill 2007).
University Dynamics Effects
Partial support is found for Hypotheses 3a, 3b, and 3c, which captured the power and
resource imbalance among receptor sites. Even after controlling for other institutional and
organizational factors, there are strong and positive relationships between having a larger generic
management program (PA/PS degree and BUS degree) and the rate of program adoption in PA/PS
as well as between having a larger business management program (BUS degree) and the rate of
adoption in BUS. However, there is a positive but not statistically significant association between
having a larger social work and social sciences program (SW/SC degree) and the rate of adoption
in SW/SC. These results imply that adoptions in PA/PS and BUS may be largely determined by the
disciplinary home’s human and material resource abundance, while adoptions in SW/SC may be
determined by the disciplinary home’s special interest in or commitment to nonprofit management
issues regardless of the amount of resources available. Based on these findings, it can be argued
that larger and resource-rich receptor sites are often, but not always, better positioned to adopt
nonprofit management programs within their own boundaries. Internal politics and the dynamics
of universities seem to be crucial mechanisms that govern variations in the disciplinary settings,
but they are also complex mechanisms that cannot be easily understood.
State- and University-Level Controls
The effects of the state-level control variables on program adoption show some evidence that
socioeconomic and political conditions contribute to the divergence in disciplinary venues of
100
nonprofit management studies programs. Demographic heterogeneity (Urbanization) is positive
and statistically significant for adoptions in PA/PS and BUS and negative but not significant for
adoptions in SW/SC. This result implies that the rate of program adoption in PA/PS and BUS
increases when a university or college is located in a more demographically and culturally diverse
region. Welfare expenditure is positively related to adoptions in PA/PS, indicating that a state’s
support for social services increases the rate of program adoption, particularly in PA/PS. The
statistically insignificant effect of political liberalism (Citizen liberalism) on adopting programs in
all disciplinary settings demonstrates that a politically liberal atmosphere in a state in which a
university or college is located has no significant main effects.
Some empirical evidence is found in support of my assumption that predicted the impact of
various organizational attributes on an individual university’s choice among the three disciplinary
settings. Institution size as proxied by University revenue is positively significant across all of the
sub-models and thus does not explain the variation in disciplinary settings; large universities and
colleges that can draw greater resources into curricular innovation exhibit a greater propensity to
adopt programs regardless of the programs’ disciplinary home. Public affiliation as measured by
the Public variable has a substantively significant effect on program adoption in PA/PS and BUS
and shows a reverse direction between the two. The reverse sign demonstrates that public
institutions have a higher rate of adopting programs within PA/PS but a lower rate of adopting
programs within BUS. I find a marginal effect of Doctoral/Research, which measures an
institution’s selectivity; there may be a negative relationship between highly prestigious
institutions and rates of program adoption in general, but this relationship is not strongly supported
by the empirical evidence. Interestingly, two measures of campus diversity, Black/Hispanic and
Female, are opposite in direction; institutions with a higher proportion of degrees conferred to
101
African American and Hispanic students are less likely to adopt programs within BUS, while
institutions with a higher proportion of degrees conferred to female students are more likely to
adopt programs within PA/PS and BUS.
4.5. Discussion
Nonprofit management research and education programs have been housed in different
disciplines or academic settings within universities, and the variation in the academic homes for
such programs may be governed by a number of underlying historical, social, and organizational
forces. This chapter therefore aimed to provide a comprehensive examination of how various
factors in the real world shape the different disciplinary venues used to house nonprofit
management studies and further suggested some empirical insights into its developmental process.
To this end, I analyzed the rates and mechanisms by which American universities and colleges
adopted their first nonprofit management programs within a certain disciplinary setting over time.
The findings from this chapter indicate that the different disciplinary settings have grown at
increasing rates since the early 1970s, and the dominant setting among the three was the public
affairs and political science school setting—more than one-third of the initially established
university programs were housed in this setting between 1971 and 2011. The findings also provide
compelling evidence that the choice of disciplinary setting is a function of broader institutional
pressures, state nonprofit sector conditions, and organizational dynamics, which is essentially in
line with the arguments of Young (1998) and Hall et al. (2001). Although the factors associated
with the state nonprofit sector setting are relatively unhelpful in explaining how and why the
variations in the disciplinary settings occur, the historical shift in the institutional logics of
nonprofit management and university-specific dynamics are concrete mechanisms that account for
102
the disciplinary variations. More specifically, higher education institutions that placed their first
academic programs in the public affairs and political science school setting tended to be public, to
have a larger generic management program, and to be located in a more urbanized and welfare-
oriented region. Institutions that located such programs in the business school setting were
importantly influenced by the prevalence of the business-oriented nonprofit management discourse
in the 2000s and tended to be private, to have a larger business management program, and to be
located in a more heterogeneous and less welfare-oriented region. Institutions that adopted
programs in the social work and social sciences school setting were hardly explained by the factors
that I included and could only be characterized as less prestigious. However, different disciplinary
settings not only are shaped by different processes and mechanisms but also can be subject to the
same institutional and organizational forces. The findings demonstrate that the structure and theory
of the nonprofit management regime throughout American society triggered an increase in
adoption rates for all of the disciplinary settings. In addition, institutions that were resource-rich
had a higher propensity to adopt programs within all of the types.
One general implication of these findings is that the current body of academic knowledge on
nonprofit management originates from and relies on various intellectual sources, and its historical
development can be characterized by the process of creolization (Sahlin-Andersson & Engwall
2002). The term creolization is generally used to illustrate a cultural fusion among different
societies and refers to an active process of adaptation to and creation of new cultures or identities
without the loss of their original characteristics (Cohen 2007; Hall 2003; Hannerz 1987; Stewart
2007). In other words, the current field of nonprofit management studies may be a product of the
fertilizable mixture of various existing knowledge bases, including public administration and
policy, business, social work, sociology, psychology, anthropology, education, law, and
103
professional studies. In addition, it seems that the field continues to develop through a process in
which diverse intellectual sources either collaboratively or competitively create a new and
extended body of knowledge that is relevant to nonprofit management. An additional implication
is that one dominant disciplinary setting for nonprofit management research and education could
emerge in the future, although likely not in the near future because an overall trend toward
convergence among the academic homes for nonprofit management research and education is
empirically not found at this point in time—rather, I see a tendency toward greater divergence. In
addition, each university’s decision making on the academic home for nonprofit management
programs tends to be sensitive to its own unique historical, social, and organizational conditions
at this moment, when there is no overwhelming consensus regarding the best place for nonprofit
management studies.
The majority of leaders in this field—including Hall et al. (2001), O’Neill (2007), Rimer
(1987), and 15 former and current directors of nonprofit management programs at seven different
universities whom I interviewed—indeed claim that we will see continued experiments with
different intellectual bases and academic homes, at least until nonprofit management research and
education reaches a higher level of institutionalization. For instance, Rimer (1987, p. 52) maintains
that “It is not likely that any one will dominate for a long period of time. Rather, depending upon
political and power differences that will increase and decrease over time, there will be trends that
favor one [academic home] over another.” Likewise, Vinokur-Kaplan (in Hall et al. 2001, p. 79)
states that “I believe there will continue to be 1,000 flowers blooming, given the different
arrangements and cultures of various universities.” Nearly all of my interviewees also shared their
opinions, as follows:
“You will continue to see a fairly diverse set of offerings with great areas of
commonality focused on nonprofit management. I do not think we will see anytime
104
soon a consolidation of streaming within the nonprofit management [research and]
education world, like the MBA degree in business schools or the MPA degree in public
affairs schools. It is going to be more diverse for a long time.”
“I think it will continue to exist in different units of a university because the politics is
so different in every institution and the reasons why people want a program or do not
want a program are very different.”
“Like the expression, where you stand depends on where you sit, I believe it is going
to appear in every kind of field possible… the future will not change that much.”
At this moment, there seem to be certain benefits to having a variety of academic models and
homes. As Hall (in Hall et al. 2001, p. 78) notes, because the nonprofit sector engages in a wide
array of activities and provides a substantial range of services, “[to] reflect the great diversity of
nonprofitdom, [nonprofit management programs] may have to take a variety of forms and be
located in various parts of the university.” Moreover, each disciplinary setting may provide
different opportunities for the development of nonprofit management research and education, as
described by one of my interviewees:
“There are advantages for both ways. [Having programs housed by existing academic
units such as] public affairs and business [can easily] generate university support. But
having free-standing programs allows a lot of opportunities to develop the specifics of
nonprofit studies and concentrate on the actions of the nonprofit sector.”
Despite these benefits, it is expected that “the field itself will keep trying to search for its
own best way” (Interview with faculty, August 2014). In that process, lessons from the successes
and failures of the current ongoing experiments with various intellectual bases and academic
homes will play a major part.
105
CHAPTER 5. INSIDE THE BOX: THE GRADUAL INSTITUTIONALIZATION
PROCESS OF NONPROFIT MANAGEMENT STUDIES
“The main barrier that we overcome now was a lack of credibility for a new study field
to survive in an academic setting… It is still a fairly new field, so you have to convince
everybody in the university. You have to justify yourself.”
- Interview with faculty, August 2014
The previous two chapters analyzed national trends in the emergence and development of
academic programs in nonprofit management. This chapter narrows the discussion, focusing on
particular institutions of higher education that have established and developed nonprofit
management studies programs in different ways, to provide a more detailed exploration of the
gradual development process of nonprofit management. In addition to demonstrating the effects
of various historical and social contextual factors on the institutionalization of nonprofit
management studies, this chapter sheds more light on universities’ internal contingencies that seem
to be critical but that could not be properly captured by the previous two chapters due to the
limitations of quantitative methods.
5.1. Background
Nonprofit management, as an interdisciplinary area of knowledge that does not fit neatly into
existing academic disciplines, had been situated between or within existing fields of study—
including public administration and policy, business management, social work, sociology, political
science, urban affairs, economics, education, and law. Nonprofit management has only recently
achieved a certain degree of legitimacy as a distinct field of study. Several scholars have argued
that nonprofit management has become increasingly visible and widespread throughout the
106
academic community and that “there is no doubt that the field of philanthropy and nonprofit studies
has been established” (Graddy et al. 2011, p. 2). For instance, Katz (1999), Smith (1999), and
Young (1999) reviewed the historical development of the academic study of nonprofit
management in the United States. Based on anecdotal evidence, they concluded that there seems
to be a unique body of knowledge to support a separate field of nonprofit management studies, but
the field is relatively young, and thus needs to establish better theoretical foundations and
accumulate more knowledge to distinguish it from other disciplines. Drawing on an empirical
analysis of survey results and academic publications, Graddy et al. (2011) and Jackson et al. (2014)
made a similar argument. They maintained that although little intellectual coherence within the
study field has been observed, remarkable development in terms of knowledge production—such
as a larger number of journal articles and scholars, growing diversity in the institutional and
national affiliations, and increasing reliance on nonprofit-specific citation sources—has occurred
over the decades, which in turn supports the establishment of the field and shows that “steps toward
a shared body of knowledge have started to be taken” (Jackson et al. 2014, p. 807).
Although the development of nonprofit management studies as an emerging field is a
nationwide phenomenon, individual universities incorporate nonprofit management studies at
various times and in various ways. In other words, the time at which nonprofit management studies
programs have been adopted and academic homes for those programs vary widely by academic
institution, as shown by the empirical findings of the previous chapters. Moreover, the scope,
concepts, methods, and substantive areas of this field of study are defined differently across
institutions (Hall 1992). However, the manner in which the newly emerging nonprofit management
studies are differently defined, justified, and evolved in universities has not been well understood.
Although a substantial body of research has examined the emergence and evolution of the
107
academic study of nonprofit management (e.g., Block 1987; Graddy et al. 2011; Katz 1999; Shier
& Handy 2014; Smith 1999), much of that research has focused on the historical development of
this study field overall. As a result, there is widespread ambiguity about the nature and state of
nonprofit management studies at a diverse set of universities.
This chapter thus aims to explore how nonprofit management studies has emerged and
become a legitimized field of study in various university contexts. Drawing on a comparative case
study of seven universities with explicitly unique patterns of the formation and early development
of nonprofit management studies, this chapter explores how both local dynamics at universities
and larger macro-social factors affect variation in the gradual processes of institutionalization. The
underlying assumption of this chapter is that the emergence and institutionalization of nonprofit
management studies is largely governed by (1) university dynamics, including interactions among
university actors (e.g., deans, chairs, host faculties, administrators, students, and external funders)
and universities’ organizational characteristics (e.g., university mission, history, size, structure,
reputation, and student profiles) and (2) the social and political conditions surrounding universities,
such as the broader institutional environment, the state nonprofit sector setting, demographic
heterogeneity, economic status, and political climate. This assumption is empirically tested based
on interviews with faculty members, each university’s archival documents, and secondary data
compiled from various publicly available sources. Using a qualitative comparative analysis (QCA)
method, cross-university and within-university patterns related to the gradual institutionalization
process of nonprofit management studies are examined in a manner that is systematic and analytic.
This chapter proceeds as follows. I begin by providing a conceptual model that identifies
core elements of the formation and early development of nonprofit management studies at
universities, drawing on ideas from the sociology of scientific knowledge. Subsequent sections
108
focus on the issues of case selection, data collection, QCA coding procedure, and hypothesis
testing. In the conclusion, I summarize my findings and discuss the implications for the field’s
higher level of institutionalization and future development.
5.2. Nonprofit Studies as an Emerging Field: Formation and Early Development
Sociological research on scientific knowledge construction has argued that the formation and
early development of a new knowledge or discipline is fundamentally influenced by not only local
conditions at universities in which the knowledge or discipline first emerges and develops but also
the macro social and political conditions surrounding the universities (Camic 1995; Frickel 2004;
Small 1999; Wood 2012). Based on this line of reasoning, this section develops a theoretical
framework that argues that the local dynamics of universities and broader societal settings affect
the timing of emergence and the degree of institutionalization of nonprofit management studies
within university systems.
University Actors and Dynamics
Sociological research on disciplinary formation has emphasized the influence of university
conditions and politics on the intellectual and institutional development of new disciplines. For
instance, Camic (1995) showed how such a concept of localism can be applied to the development
of early visions of American sociology at Harvard University, Columbia University, and the
University of Chicago. He documented that the different patterns of interaction among faculty
members in newly established sociology departments and scholars in other departments on campus
shaped three distinct methodological foundations of sociology: analytical abstraction at Harvard,
statistical generalization at Columbia, and inductive methods at Chicago. Based on a comparative
109
case study of Temple University and Harvard University, Small (1999) showed that specific
interests and expectations related to the emerging field of African-American studies at each
university framed the different legitimation processes of the field of study. At Temple, the field
was developed as an independent, Afrocentric, and liberating discipline, primarily because of the
strong interest and leadership of a department chair and the support and involvement of the black
community in Philadelphia. By contrast, at Harvard, the primary constituencies were scholars in
established disciplines and a much wider collective of mainstream journalists, policymakers, and
philanthropists. Therefore, the field of study was developed as an interdisciplinary, scholarly but
policy- and culturally centered field.
Aligning with this perspective, I argue that the emergence and institutionalization of
nonprofit management studies has resulted largely from the values, interests, and practices of
major actors at universities—including host faculties, administrators, students and alumni, external
funders, nonprofit practitioners, and community leaders. Therefore, variation in the field’s
development process across universities may be caused by a different set of university actors and
their different needs and interests related to nonprofit management research and education. More
specifically, with respect to the field’s emergence, the primary impetus for incorporating the field
into university systems may come from either internal stakeholders—such as individual faculty
members, university administrators, and students—or external organized and vocal stakeholders—
such as foundation funders, nonprofit practitioners, and community leaders (Larson & Barnes-
Moorhead 2001). Moreover, major external funders for the field’s formation can be either large
grant-making foundations that have been publicly invested in an initiative to develop nonprofit
management studies programs—including the W. K. Kellogg Foundation, the Ford Foundation,
110
and the Lilly Endowment Inc.—or local family foundations and individual donors (O’Neill &
Fletcher 1998).
Reflecting on the experience of the field of study of public administration and business (for
more details, see O’Neill 2007), it can be assumed that universities that are either pushed by
external stakeholders or funded by well-known foundations are likely to incorporate nonprofit
management studies earlier than those that lack powerful external supporters or funders. This
makes sense because when there is little recognition of an emerging field of nonprofit management
studies in the higher education context, universities might need external validation—or at least
symbolic legitimacy for the field’s formation, which can be provided by highly visible external
stakeholders. With respect to the field’s institutionalization, however, it appears that the ongoing
commitment of internal leadership—top administrative leaders (e.g., presidents, provosts, and
deans) and host faculty members—and prominent external funders is associated with the field’s
strong institutionalization. According to Larson and Barnes-Moorhead (2001), strong support from
key administrators may lead to less resistance to the field’s development and, therefore, its
institutionalization can become more of a central priority for the university. Moreover, a stable
line of external funding from large, well-known foundations may not only encourage internal
actors to have higher expectations of the field’s development but also generate higher levels of
internal financial support to ensure the field’s security.
I also argue that variation in the rate and extent of the field’s development across universities
may be partially attributed to the organizational characteristics of host universities. In other words,
universities vary in their structural and institutional properties—including mission, history, size,
and prestige—and each university’s unique organizational features may differently characterize
the patterns of the field’s development. As elaborated in the previous two chapters, in fact, earlier
111
studies of higher education institutions (e.g., Brint et al. 2009; Kraatz & Zajac 1996; Lounsbury
2001; Olzak & Kangas 2008; Rojas 2006; Soule 1997) found a positive relationship between
university size and the adoption of new administrative structures, practices, and curriculums. They
also found some evidence to support a widely shared belief that highly prestigious universities
tend to be receptive to curricular innovation. Therefore, it can be argued that large or prestigious
universities are more likely to support the incorporation and institutionalization of an emerging
field of study of nonprofit management than are small or less-prestigious universities. Large
universities are more willing to develop the field within their systems than small universities
because the former may have higher levels of slack resources and capacities for action. A similar
tendency could be observed in elite universities because embracing an emerging field and, thus,
being at the cutting edge of academic development could enable them to maintain and enhance
their high reputations.
Social and Political Context
Sociological research on the construction of scientific knowledge has also maintained that
social, economic, and political factors at the national, regional, and local levels undoubtedly affect
how a new field of knowledge is defined, justified, and developed. On this point, for example,
Camic and Xie (1994) highlighted the influence of urban reformers and settlement house workers
who wished for informed solutions to contemporary social issues on the advance of sociology.
Moreover, Small (1999) argued that the evolution of African-American studies was closely tied to
political pressure from local black communities and demands from journalists, policymakers,
philanthropists, and others who were interested in addressing racial tensions in the United States.
Likewise, Wood (2012) noted that the rise of women’s studies was a consequence of changing
112
historical and social-structural conditions in the 1970s, such as increased opportunities for women
in coeducational universities, legal pressures to diversify student and faculty bodies under the
auspices of the Civil Rights Act, and state support for critical research fields in the social sciences.
Considering these findings, this study predicts that the emergence and early development of
nonprofit management studies may be partly explained by social and political conditions.
Furthermore, it assumes that variations in the field’s institutionalization process across universities
might be caused by the different social and political environments surrounding those universities.
Aligning with this line of argument and the neoinstitutional perspective (e.g., Ramirez et al.
1997; Thornton & Ocasio 1999; Tolbert & Zucker 1983), I expect that the broader institutional
environment and its historical change affect the field’s formation and institutionalization across
universities. A key feature of this institutional environment may be the structuration of the
nonprofit sector at the national level. As illustrated in Chapter 2, the nonprofit sector in the United
States has been increasingly institutionalized, similar to the public and private sectors, over the
last several decades. As the nonprofit sector becomes more widely accepted and taken-for-granted,
universities may be more likely to welcome the construction of the study field of nonprofit
management and the field’s strong institutionalization, regardless of internal demands. Indeed, the
institutionalist literature on higher education institutions (e.g., Moon & Wotipka 2006; Robinson
2011; Suárez et al. 2009) noted that “institutional forces often overwhelm functional explanations
once a reform or a movement achieves legitimacy” (Suárez et al. 2009, p. 208), and thus
universities with even low internal pressures for innovation tend to adopt innovative curriculum
or practices. Based on this discussion, it could be argued that the nationwide structuration of the
nonprofit sector over time will contribute to the creation and strong institutionalization of nonprofit
management studies programs within universities.
113
In addition, it is plausible to hypothesize that regional or local differences in the nonprofit
sector are associated with variations in the rate and extent of the field’s development across
universities. As discussed in Chapter 4, the size, scope, and composition of the nonprofit sector
dramatically differ from across place depending on region-specific characteristics such as
community size, age, cohesion, wealth, and political culture (Ben-Ner & Van Hoomissen 1992;
Corbin 1999; Grønbjerg & Paarlberg 2001; Matsunaga & Yamauchi 2004). Consequently, each
regional or local nonprofit sector shapes the distinct needs, goals, and agendas of teaching and
researching nonprofit management (Cook 1988; DiMaggio 1988; Haas & Robinson 1998;
Heimovics & Herman 1989; Rimer 1987). Extrapolating from this line of argument, there may be
reason to believe that universities located in a region with a large nonprofit sector tend to
incorporate the study field earlier and to institutionalize the incorporated field more strongly than
those located in a region with a small nonprofit sector. This seems reasonable because larger
regional nonprofit sectors can strongly push universities to create and develop academic programs
by providing more visible political support, more funding opportunities, and a larger pool of
consumers of the programs—namely, current and potential nonprofit managers as students.
Lastly, it is assumed that region-specific demographic settings and political orientations will
either facilitate or constrain the field’s formation and institutionalization at different universities.
Population heterogeneity and political liberalism appear to be particularly related to the early
formation and strong institutionalization of the study field. In other words, universities that are
located in a demographically heterogeneous region may be more likely to be early adopters and
strong advocates of the field than universities that are located in a homogeneous region. This is
because demographically diverse regions tend to have larger nonprofit sectors to address their
populations’ complex and varied service demands (Ben-Ner & Van Hoomissen 1992; Grønbjerg
114
& Paarlberg 2001; Matsunaga & Yamauchi 2004), which, in turn, can be linked to an intense
interest in and need for research and education programs in nonprofit management. Furthermore,
universities that are embedded in a politically liberal climate may be more open to integrating the
field earlier and further developing the field than those in a conservative climate. This is because
political liberalism often welcomes the construction of new disciplines and adoption of innovative
academic programs (Brint et al. 2009). Overall, the key idea here is that regional differences in
receptivity to nonprofit management research and education may lead to variation in the rate and
extent of the field’s development across universities.
5.3. Data and Methods
To flesh out my conceptual arguments, a comparative case study of seven universities was
conducted. Data were collected mostly through interviews, and the collected data were analyzed
with a QCA method.
The Cases
The cases include nonprofit management studies research and education programs at seven
universities in the U.S.: a public flagship university in the Mountain West region, a private research
university in the Great Lakes region, a large public university in the Great Lakes region, a public
flagship university in the South Atlantic region, two private Catholic universities in the Pacific
region, and a private research university in the Pacific region. Here, I refer to these universities as
Mountain West University, Private Great Lakes University, Public Great Lakes University, South
Atlantic University, Northern Gold Coast University, Southern Gold Coast University, and Central
115
Gold Coast University, respectively.
23
These seven universities were intentionally selected based
on their similarities and differences (for more details, see Appendix G) and on accessibility to key
informants.
More specifically, the seven cases were selected because they capture various aspects of the
development process of nonprofit management studies. With respect to the timing of the adoption
of a stand-alone degree program in nonprofit management studies (hereafter referred to as a degree
program),
24
Private Great Lakes and Northern Gold Coast are early adopters that have offered
their degree programs since the 1980s, whereas Mountain West, Southern Gold Coast, and Central
Gold Coast are relatively late adopters that founded their first programs after the 2000s. Public
Great Lakes and South Atlantic are neither early nor late adopters because they introduced degree
programs to their students in the 1990s. With respect to the types of degree programs that they
create and confer, Public Great Lakes grants Bachelor’s, Master’s, and doctoral degrees; Mountain
West grants Bachelor’s and Master’s degrees; and the other five universities grant only Master’s
degrees.
With respect to the institutional arrangement for nonprofit management studies, Public Great
Lakes and Mountain West have freestanding departments that hire full-time, tenure-track faculty
independently and are financially self-sustaining with minimal external funding. By contrast, the
other five universities house nonprofit management studies within traditional academic disciplines
including social work, public affairs, management, education, and professional studies; thus, their
programs are materially and financially dependent on existing departments or schools. Taken
23
I refer to the case institutions using pseudonyms to protect the identities of the faculty members who I interviewed.
24
Stand-alone degree programs for nonprofit management studies here include undergraduate and graduate degrees
in general nonprofit and philanthropic studies (e.g., Bachelor of Arts in Nonprofit Leadership and Management,
Master of Arts in Philanthropic Studies, Master of Science in Nonprofit Organizations, Ph.D. in Nonprofit Studies,
and so on) but exclude industry-specific programs such as human services management and arts management.
116
together, the seven cases show clear patterns of similarity and difference in terms of the formation
and early development of nonprofit management studies, making them ideal for comparison.
Furthermore, each of these cases has unique organizational characteristics in terms of
university mission, history, size, structure, reputation, and student profiles. More specifically,
Mountain West is a public flagship research university and one of the largest universities in terms
of enrollment; Private Great Lakes is a private university that is highly research oriented; Public
Great Lakes is a large public university and has a larger number of underrepresented minority
students than other universities in the state; South Atlantic is a public flagship research university
and is one of the oldest and largest universities in the state; Northern Gold Coast is a private
Catholic university and is one of the oldest universities in the state; Southern Gold Coast is also a
private Catholic university and is the smallest and newest of the seven case universities; and
Central Gold Coast is a private research university and has the largest enrollment of the seven case
universities. Accordingly, it was possible to use these seven universities to conduct a close
investigation of the influence of unique organizational features on the evolution of nonprofit
management studies.
Finally, these cases reflect variation in regional contexts. Private Great Lakes and Public
Great Lakes are located in the Midwest region, which has dominated the field since the mid-1990s,
offering nearly 40 percent of nonprofit education programs. The Midwest region has also been
well known for its large amounts of external funding from several of the nation’s largest
foundations, such as the W. K. Kellogg Foundation and the Lilly Endowment Inc. (Mirabella 2007;
Mirabella & Wish 2001; Wish 1991; Wish & Mirabella 1998). The other five universities are
located in the South or West regions, where approximately 15 percent of nonprofit education
programs have been offered since the 1990s; the growth of programs in those regions has remained
117
relatively stable compared to the Midwest and Northeast regions (Wish & Mirabella 1998;
Mirabella 2007). In addition to differences in the educational environments, there are variations in
social and political environments surrounding the seven universities. Indeed, Private Great Lakes
is located in a Midwestern state that is more urbanized, wealthier, and more politically liberal than
the state in which Public Great Lakes is located; South Atlantic is located in the state that is the
least urbanized and that has a small nonprofit sector; Mountain West is located in a state that tends
to be urbanized and politically conservative and has a small nonprofit sector; Northern Gold Coast,
Southern Gold Coast, and Central Gold Coast are located in a state that is one of the most populous,
urbanized, and liberal states and that has a large nonprofit sector. In sum, a comparison of these
seven universities located in different parts of the country provides a better understanding of the
impact of different regional characteristics on the development of nonprofit management studies.
Data
The data for this study are composed of 15 personal interviews with faculty members at the
seven universities, each university’s archival documents, and secondary data compiled from
various publicly available sources. First, interviews were semi-structured and lasted from 35 to 90
minutes. The interviews were conducted either in person or over the telephone in 2014. The 15
interview participants from seven different universities were identified primarily through snowball
sampling techniques, and participants were limited to faculty members with extensive knowledge
of the emergence and development of nonprofit management studies at their institutions. The
interview participants were mostly former or current directors of their institution’s research center
or education program in nonprofit management and philanthropy.
25
Interview protocols were
25
I refer to interview participants anonymously to protect their identities.
118
approved by the Human Subjects Protection Program at the University of Southern California
(Approval ID: #UP-14-00375). Based on a 21-question interview instrument, I asked each
interviewee to provide a brief career biography related to nonprofit research and teaching; to
explain the current offerings of nonprofit management research and education programs at his/her
institution; to describe the historical, social, and organizational conditions under which those
programs were established; to illustrate the intellectual and institutional evolution of nonprofit
management studies at his/her institution; and to evaluate his/her nonprofit management program
and the state of the field in general (for more details, see Appendix H). All of the questions were
open-ended and primarily required recollections of past events. To reduce the problem of
retrospective bias, I interviewed at least two people at each institution. Each interview was tape-
recorded and later transcribed following the interview participants’ consent to these methods.
Second, historical documents were collected to supplement and facilitate interpretation of
the interview data. The historical records dated back to the creation of nonprofit management
studies programs at each university and were obtained from each university’s online archive. They
consist of program proposals and self-evaluations, annual reports, college/school catalogs,
department brochures, syllabi, and lecture notes.
Third, secondary data on organizational characteristics of universities and social and political
conditions at the state level (1980-2010) were compiled from various publicly available sources.
Data on university characteristics include measures of university size, status, and campus diversity
collected from the HEGIS for the years 1980 through 1986 and the IPEDS for the years 1987 to
2010. Data on state-level social and political conditions include measures of nonprofit sector size,
demographic and economic status, and political orientation, which came from the NCCS (2012),
the U.S. Census Bureau (2013), and Berry et al. (2010), respectively.
119
Analytical Methods: Qualitative Comparative Analysis
Transcribed interviews, university archival documents, and combined secondary data were
analyzed using a QCA method. QCA is an analytic technique that aims to analyze how causal
conditions (or a combination of conditions) contribute to an outcome in question based on set
theory and Boolean algebra (Ragin 1987, 2000, 2008). It pursues “a middle path between
quantitative and qualitative research” (Ragin 2008, p. 1) by integrating the advantages of the
variable- and case-oriented approaches (Rihoux & Ragin 2009)—even though it can be claimed
that this technique tends more toward the side of case-oriented approaches (Rihoux & Lobe 2009).
Moreover, it allows “systematic cross-case comparisons, while at the same time giving justice to
within-case complexity, particularly in small- and intermediate-N research designs” (Rihoux &
Ragin 2009, p. xviii). In this regard, this technique is uniquely suited for comparative case studies.
QCA is fundamentally an analysis of set relations because it is based on a set-theoretic
understanding of how causal conditions combine to produce a given outcome (Fiss 2007, 2011).
Crisp-set and fuzzy-set methods, as logical tools for assessing necessary and sufficient causal
relations, provide powerful tools for the statistical analysis of set relations. Crisp-set (or Boolean)
logic is based on the dichotomous presence or absence of hypothesized causes and outcomes across
cases. With crisp-set methods, each case is assigned one of two possible membership scores in
each set—a causal or outcome category—included in a study: 1 (a member of the set) or 0 (not a
member of the set). Overcoming the binary nature of crisp-set logic, fuzzy-set logic permits cases
to exist partially in the set. With fuzzy-set methods, each case is scored between 0 (full non-
membership) and 1 (full membership) based on its degree of membership in a set. More
specifically, a fuzzy membership score of 1.0 is a qualitatively defined state indicating full
membership in a set; scores close to 1 and greater than 0.5 (e.g., 0.8 or 0.75) represent partial
120
membership indicating that a case is located not quite in a set; a score of 0.5 is qualitatively
anchored and indicates the point of maximum fuzziness in the assessment of whether a case
belongs to a set; scores of less than 0.5 but greater than 0.0 (e.g., 0.4 or 0.25) represent partial
membership indicating that a case is located more out of a set; and a score of 0.0 is a qualitatively
defined state indicating full non-membership in a set. Overall, a crisp- and fuzzy-set can be viewed
as a dichotomous and continuous variable, respectively. However, they are much more than simple
dichotomous and continuous variables because the process of calibration—which refers to the
transformation of all causal conditions and outcomes into crisp- and/or fuzzy-sets—is heavily
grounded in a detailed, in-depth qualitative knowledge of the cases included in a study (Ragin
2000, 2008; Rihoux & Ragin 2009).
Crisp- and fuzzy-set methods entail an analysis of necessary and sufficient causal
relationships. A necessary condition is a condition that typically or always is required to produce
the outcome. A necessary relationship exists if the outcome set is composed of a subset of the
causal condition set; that is, membership scores on the outcome set will be less than or equal to
scores on the causal conditions across all cases. By contrast, a sufficient condition is a condition
that typically or always leads to the outcome. A sufficient relationship is present when the causal
condition set is a subset of the outcome set; that is, membership scores on the outcome set will be
greater than or equal to scores on the causal conditions across cases. To assess how closely a
perfect set relationship is approximated and what proportion of cases with the outcome is explained,
QCA uses two measures: consistency and coverage. Consistency measures the degree to which a
necessary or sufficient causal relationship between a cause (or causal combination) and an outcome
in question is met within a given data set, which is analogous to the concept of significance in
statistical models. Coverage measures the degree to which a cause (or causal combination)
121
explains cases of a particular outcome, which resembles the measure of R
2
in statistical models.
Both values range between 0 and 1, with 0 indicating no consistency or coverage and 1 indicating
perfect consistency or coverage. In general, a consistency or coverage value greater than 0.75
denotes a well-specified model (Ragin 2006, 2008).
In practice, the identification of necessary and sufficient causes can be accomplished through
several steps. Once causal conditions and outcomes are transformed into crisp- and/or fuzzy-sets,
a test for necessary conditions is conducted before the analysis of sufficiency. Although necessary
conditions are rarely identified empirically, it is useful to determine what causal conditions might
be necessary for the outcome because they are often not observed in the most logically simplified
causal combinations. A truth table is then constructed and analyzed to determine which causal
recipes are sufficient for the outcome. The truth table is a data matrix with 2
k
number of causal
combinations (where k is the number of causal conditions used in the analysis), listing all logically
possible combinations. Boolean algebra is employed to logically minimize the number of causal
combinations in the truth table and to further identify more simple and general patterns of
combinations that are sufficient for the outcome. For the minimization process, it is recommended
to use simplifying assumptions based on logical remainders (or counterfactual cases)—causal
combinations that lack empirical instances and, thus, need to be imagined. Different ways of using
logical remainders when theorizing how a given condition is causally related to the outcome yield
three different solutions: (1) a complex solution, which avoids the use of any logical remainders;
(2) a parsimonious solution, which allows all logical remainders without any substantial
consideration of their plausibility; and (3) an intermediate solution, which permits only the logical
remainders that seem to be plausible given the existing theoretical and substantive knowledge. In
general, the intermediate solution is preferred as the main point of reference for interpreting QCA
122
results because it is superior to the other two solutions in terms of the wise use of logical
remainders; the intermediate solution falls between complexity and parsimony as a superset of the
most complex solution and a subset of the most parsimonious solution. By contrast, the complex
solution is the least preferred among the three because it is unnecessarily complex and offers little
insight (Ragin 2008; Rihoux & Ragin 2009).
In this study, I rely on the Fuzzy-Set/Qualitative Comparative Analysis 2.5 (Ragin & Davey
2014) software package—which is the most widely used package and is specially designed to
compute QCA results.
Outcome Measures
The two outcomes of interest in this study are the timing of emergence and the degree of
institutionalization of nonprofit management studies at universities. Accordingly, I created two
fuzzy-set measures of the field’s emergence—Early formation and Late formation—and two of
the field’s institutionalization—High stability and Low stability.
The Early formation and Late formation measures were operationalized as the year that each
university established its first degree program. Data on the establishment year came from semi-
structured interviews and historical documents. These measures were calibrated using the direct
method of fuzzy-set calibration introduced by Ragin (2008, ch.5). The direct method requires
researchers to identify the values of an interval scale that correspond to the three qualitative
anchors—full membership, full non-membership, and a crossover point—based on both
theoretically derived, substantive knowledge and inductively derived, sample-specific knowledge.
It then rescales continuous variables to range from 0.0 to 1.0, “using the crossover point as an
anchor from which deviation scores are calculated [and] taking the values of full membership and
123
full non-membership as the upper and lower bounds” (Fiss 2011, p. 406-407).
26
Following the
procedure, I specified three qualitative breakpoints for the first set measure, membership in the set
of universities with Early formation, based on sample-dependent anchors (i.e., the mean, median,
upper-quartile, and lower-quartile) for the seven case universities: a threshold of full membership
value of the year 1985 (the 1
st
percentile), a crossover value of the year 2000 (approximately the
50
th
percentile), and a threshold of full non-membership value of the year 2010 (the 90
th
percentile).
In other words, the Early formation measure was coded as 0.5 < Xi ≤ 1 if a university created its
first degree program between 1985 and 1999, 0.5 if in 2000, and 0 ≤ Xi < 0.5 if between 2001 and
2010.
27
The second set measure, membership in the set of universities with Late formation, is the set
negation of the Early formation measure. In crisp- and fuzzy-set analysis, the membership of a
case in a negation set is calculated by simply subtracting its membership in the unnegated set from
1.0; for example, if a university received a 0.85 membership score in the set Early formation, its
score in the set Late formation is 0.15 (i.e., 1.0−0.85 = 0.15). Therefore, the Late formation
measure was coded 0.5 < Xi ≤ 1 if a university founded its first degree program between 2001 and
2010, 0.5 if in 2000, and 0 ≤ Xi < 0.5 if between 1980 and 1999.
28
The High stability and Low stability measures were operationalized as the degree of
institutional stability of nonprofit management studies research and education programs. The
degree of stability was evaluated in terms of whether nonprofit management studies programs
26
For a detailed description of the direct calibration procedure, see Ragin (2008, p. 86-94).
27
To select the best cut-off point, I additionally estimated four time periods set for two years before 2000 and two
years after 2000. The results showed that the year 2000 makes the most sense—the year 2002 was the best cut-off
point for both Chapters 3 and 4, though.
28
It is necessary to separately construct and test the absence of an outcome of interest (here, the absence of Early
formation) because causal conditions leading to the presence of a given outcome may be quite different from those
leading to the absence of the outcome paths (Borgna 2013; Ragin 2008). QCA assumes that set relations can be
asymmetrical, and in this sense “stand in contrast to the common correlational understanding of causality, in which
causal symmetry is assumed” (Fiss 2011, p. 394).
124
follow organizationally independent patterns whereby separate bureaucratic systems (e.g., a free-
standing department or school and a self-sustaining system for human and financial resources) and
programmatic mechanisms (e.g., degree-granting status, a self-governing curriculum, and an in-
house journal) exist for nonprofit management studies. Using data obtained from semi-structured
interviews, I calibrated the measure of membership in the set of universities with High stability as
a seven-value fuzzy-set. The fuzzy-set therefore embraces seven different levels of the institutional
stability of nonprofit management studies within the context of the seven case universities. I
assigned a value of 1.0 (fully in the set) to universities that have a free-standing school for nonprofit
management studies, grant degrees in nonprofit management studies at both the undergraduate
(e.g., B.A. or B.S.) and graduate levels (e.g., M.A., M.S., or Ph.D.), and employ more than ten
tenure-track, full-time faculty for researching and teaching nonprofit management studies. A value
of 0.9 (mostly but not fully in the set) was assigned to universities that have a separate department,
grant degrees at the both undergraduate and graduate levels, and employ more than five tenure-
track, full-time faculty, and a value of 0.7 (more in than out of the set) to universities that have a
separate unit or division, not a department. I assigned a value of 0.5 (neither in nor out of the set)
to universities that house nonprofit management studies programs within an established
division/department/school, grant degrees only at the Master’s level but offer more than three non-
degree programs (e.g., minor, track, specialization, and certification), and employ fewer than five
tenure-track, full-time faculty. A value of 0.3 (more out than in the set) was assigned to universities
that house nonprofit management studies within an established division/department/school, grant
degrees only at the Master’s level and offer fewer than three non-degree programs, and employ
fewer than three tenure-track, full-time faculty, and a value of 0.1 (mostly but not fully out of the
set) to universities that offer only certificate programs as non-degree programs. I assigned a value
125
of 0.0 (fully out of the set) to universities that house nonprofit management studies within an
established division/department/school, grant degrees only at the Master’s level and offer only
certificate programs, and employ only non-tenure-track or part-time faculty.
The measure of membership in the set of universities with Low stability is the set negation
of the High stability measure, which was calculated by subtracting each case’s membership score
in the High stability set from 1.0.
29
Table 5.1 provides a summary of these outcome measures and
their fuzzy-set scores for the seven case universities.
Table 5.1. Fuzzy-Set Scores for Outcomes
Mountain
West
Private
Great
Lakes
Public
Great
Lakes
South
Atlantic
Southern
Gold
Coast
Northern
Gold
Coast
Central
Gold
Coast
Emergence
Early formation 0.14 0.93 0.8 0.55 0.35 0.95 0.01
Late formation 0.86 0.07 0.2 0.45 0.65 0.05 0.99
Institutionalization
High stability 0.7 0.3 1 0.3 0 0.1 0.5
Low stability 0.3 0.7 0 0.7 1 0.9 0.5
Causal Condition Measures
I included five measures of university dynamics and three measures of social and political
environments. Measures that are categorical in nature were calibrated as crisp-sets and all other
measures were calibrated as fuzzy-sets using the direct calibration method. In general, fuzzy-sets
are preferable to crisp-sets because crisp-set solutions tend to have a lower standard of set-theoretic
consistency, which makes perfect set-theoretic consistency easier to achieve (Ragin 2008, p. 141).
29
Note again that with QCA, the absence of an outcome of interest (here, the absence of High stability) needs to be
separately tested because of the asymmetry of set relationships assumed by QCA.
126
I assessed the effect of major university actors using three measures—External champion,
Iconic leader, and Leading funder. All three of these measures were based on semi-structured
interview questions and historical documents and calibrated as crisp-sets. First, External champion
is a measure of whether the most influential advocates for the creation of nonprofit management
studies programs were a university’s external stakeholders. Membership in the set of External
champion was coded 1 if demands from foundation funders, nonprofit leaders and managers, or
community leaders jumpstarted the establishment of academic programs for nonprofit
management studies and 0 if major pushes came from the interests of individual faculty members,
students and alumni, or university administrators. Second, Iconic leader is a measure of the
existence of strong and consistent leadership for the development of nonprofit management studies
after a program’s creation. I coded universities in which the most influential founding member or
iconic faculty member remains affiliated with the university and directs nonprofit management
studies programs as fully in the set of Iconic leader (a membership score of 1), with universities
that lack such an iconic leader due to his/her retirement or transfer as fully out of the set (a
membership score of 0). Third, Leading funder is a measure of whether a major external funder
for nonprofit management studies research and education was a large grant-making foundation
that had sought to encourage the development of academic programs for nonprofit management
studies as one of its major missions. I created a measure of membership in the set of Leading
funder, assigning a value of 1 to universities that were largely funded by large, national foundations
such as the W. K. Kellogg Foundation, the Ford Foundation, and the Lilly Endowment Inc. and a
value of 0 to universities that were largely funded by local family foundations or individual donors.
127
I operationalized the characteristics of university using two measures—Large university and
Elite university—based on secondary data from the HEGIS and IPEDS.
30
Large university
measures university size based on an average number of Bachelor’s, Master’s, and doctoral
degrees conferred between 1980 and 2010. The threshold for full membership in the set of Large
university was placed at the 99
th
percentile (8,500 or more degrees); the threshold for non-
membership was placed at the 75
th
percentile (fewer than 1,270 degrees); and the crossover point
was set at the 95
th
percentile (3,200 degrees). Elite university is a measure of organizational status,
which was coded as a crisp-set. A value of 1 was assigned to universities that were categorized as
doctorate-granting research universities with a “very high” level of research activity by the 2005
Carnegie Classification of Institutions of Higher Education, and a value of 0 was assigned to
universities that were categorized as research universities with a “high” level of research activity.
Three measures of social and political context—Large nonprofit sector, High urbanization,
and High liberalism—were based on state-level secondary data and calculated as averages for the
period of 1980-2010. To calibrate these measures, I used statistical properties of the 50 U.S. states
and Washington D.C. rather than properties of the sample itself. The first measure, Large nonprofit
sector, is a measure of the size of the nonprofit sector based on a state’s average number of
nonprofit organizations—which was obtained from the NCCS (2012).
31
Universities that are
located in states with 124,000 or more nonprofit organizations between 1980 and 2010 (the 99
th
30
Several other measures—such as university revenues and expenditures as a proxy for university size, whether a
school is a public, land-grant, flagship, or religious university as a proxy for university status, and the proportion of
minority and female students as a proxy for campus diversity—were originally included to measure a variety of aspects
of university organization. However, they provided either no significant or less consistent results and finally were
excluded from the analysis.
31
I originally also considered two other measures, i.e., the number of grant-making foundations and nonprofit assets
per capita. However, the three measures were highly correlated to each other with a single underlying factor. Based
on the result of principal component factor analysis, therefore, I used the number of nonprofit organizations as a
factored variable in the analysis; it has a smallest uniqueness value of 0.001 and, thus, might be the most informative
measure of the three.
128
percentile) were coded as fully in the set of Large nonprofit sector and those with fewer than
15,000 (the 25
th
percentile) were fully out; the crossover point was set at 70,000 (the 90
th
percentile). The second measure, High urbanization, determines demographic heterogeneity based
on an average percentage of population living in a SMSA—which was collected from the U.S.
Census Bureau (2013). For the fuzzy-set of universities located in highly urbanized states,
universities in the 99
th
percentile (100% of population) were coded as fully in and universities in
the 1
st
percentile (37% of population) were coded as fully out; as the crossover point, I chose the
50
th
percentile value (74% of population). The last measure, High liberalism, is a measure of
political orientation based on Berry et al.’s (2010) indices for citizen and state government
ideology, which ranges from 0 (high conservatism) to 100 (high liberalism). The fuzzy-set measure
of universities that are located in highly liberal states was coded as fully in the set of High
liberalism for the index score of 60 (the 80
th
percentile) and fully out of the set for the index score
of 30 (the 1
st
percentile); I used the index score of 50 (the 50
th
percentile) as the crossover point.
Table 5.2. Crisp- and Fuzzy-Set Scores for Causal Conditions
Mountain
West
Private
Great
Lakes
Public
Great
Lakes
South
Atlantic
Southern
Gold
Coast
Northern
Gold
Coast
Central
Gold
Coast
University context
External champion 1 1 1 0 0 0 0
Iconic leader 1 0 1 0 1 0 1
Leading funder 1 1 1 0 0 1 0
Large university 0.95 0.12 0.4 0.86 0.05 0.15 0.91
Elite university 1 1 0 1 0 0 1
Social/political context
Large nonprofit sector 0.15 0.68 0.49 0.38 0.95 0.95 0.95
High urbanization 0.8 0.56 0.38 0.37 0.9 0.9 0.9
High liberalism 0.17 0.53 0.31 0.69 0.86 0.86 0.86
129
Table 5.2 summarizes these causal condition measures and their crisp- and fuzzy-set scores
for the seven case universities.
5.4. Findings
Table 5.3 shows descriptive statistics and correlations for all of the outcome and causal
condition measures. Overall, as expected, university factors are positively correlated with Early
formation and High stability and negatively correlated with Late formation and Low stability. By
contrast, there are unexpected negative correlations between social/political factors and Early
formation and High stability. There are also unexpected, but partly aligned with the results in Table
3.3 in Chapter 3, positive correlations between social/political factors and Late formation and Low
stability.
Table 5.3. Descriptive Statistics and Correlation
Variable Mean S.D. 1 2 3 4 5 6 7 8 9 10 11
Outcomes
1. Early formation 0.53 0.38
2. Late formation 0.47 0.38 -1.00
3. High stability 0.41 0.35 -0.13 0.13
4. Low stability 0.59 0.35 0.13 -0.13 -1.00
Causal conditions
5. External champion 0.43 0.53 0.22 -0.22 − −
6. Iconic leader 0.57 0.53 − − 0.49 -0.49 0.17
7. Leading funder 0.57 0.53 0.57 -0.57 0.40 -0.40 0.75 -0.17
8. Large university 0.49 0.40 -0.68 0.68 0.48 -0.48 0.00 0.27 -0.27
9. Elite university 0.57 0.53 -0.41 0.41 0.13 -0.13 0.17 -0.17 -0.17 0.67
10. Large NP sector 0.65 0.32 0.10 -0.10 -0.60 0.60 -0.61 -0.06 -0.32 -0.56 -0.43
11. High urbanization 0.69 0.24 -0.44 0.44 -0.46 0.46 -0.41 0.29 -0.14 -0.11 -0.15 0.56
12. High liberalism 0.61 0.28 -0.02 0.02 -0.78 0.78 -0.91 -0.27 -0.63 -0.30 -0.21 0.86 0.45
130
Tables 5.4 and 5.5 present the results of truth table analysis for the formation and
institutionalization of nonprofit management studies, respectively. They show multivariate
configurations that are sufficient for the outcome of interest in this study, exhibiting acceptable
frequency (≥ one case) and consistency (≥ 0.75); a frequency threshold of one case and consistency
threshold of 75% were selected to represent relevant and consistent cases; causal conditions and
their combinations passing these thresholds are considered as usually sufficient to produce a given
outcome. Building upon the notion of configuration tables suggested by Ragin and Fiss (2008), I
used black circles (“
”) to indicate the presence of a condition and circles with a strikethrough
(“
”) to indicate the absence of a condition. Here, large circles (“
” and “
”) refer to core causal
conditions that are identified by parsimonious solutions and small circles (“
” and “
”) refer to
complementary causal conditions that are additionally identified by intermediate solutions.
32
In
addition, blank spaces in each solution indicate a “don’t care” situation in which the causal
condition shows no consistent pattern and thus is assumed to be either present or absent.
Given the small number of cases (N=7) in this study, I attempted to maintain a certain ratio
of variables to cases to obtain more robust QCA results; according to Marx (2006), increasing the
number of variables relative to cases increases the risk of finding random configurations that
falsely pass sufficiency tests and, therefore, the use of four or less independent variables is
recommended for an analysis of fewer than 10 cases. Accordingly, two separate models for each
of the four outcomes were estimated: one that includes four university factors only and another
that includes two compounding sets of university factors and three social and political factors. To
estimate the second model, I created three new sets by combining existing sets of university causal
32
Core conditions are “the decisive ingredients that distinguish combinations of conditions that are consistent subsets
of the outcome from those that are not,” whereas peripheral conditions are ingredients that “make sense as important
contributing factors and can be removed from the solution only if the researcher is willing to make assumptions that
are at odds with existing substantive and theoretical knowledge” (Ragin 2008, 204).
131
factors
33
: (1) Champion or Funder, the set union of External champion and Leading funder based
on the assumption that these two conditions might offer equivalent bases for a given outcome; (2)
Leader or Funder, the set union of Iconic leader and Leading funder; and (3) Large and Elite, the
set intersection of Large university and Elite university based on the conjecture that these two
conditions together lead to a given outcome. Using these compounding sets would be beneficial
not only to cope with the variable number restrictions but also to understand how university factors
combine to bring about outcomes.
Formation of the Study Field
Table 5.4 summarizes multiple combinations of measures (so-called solutions) that lead to
the outcomes Early formation and Late formation. Solutions 1 and 3 evaluate the effect of each of
the four university factors, and solutions 2 and 4 evaluate the effect of all university and
social/political factors. Solutions 2 and 4 show that there are two different paths to Early formation
(with solution 2a as the most empirically relevant path) and three different paths to Late formation
(with solution 4a as the most empirically relevant path). In both solutions 2 and 4, two
compounding sets of university factors are identified as core conditions and three social/political
factors are identified as peripheral conditions, indicating that university dynamics factors tend to
play more critical roles than do social and political context factors (for more detailed technical
interpretation of the QCA results, see Appendix I). Overall, the QCA results in Table 5.4 suggest
two distinct, broad configurations of causal factors regarding the timing of the field’s formation.
33
Because no combinations of social and political causal factors passed the test of sufficiency, compounding sets for
university factors only were created and tested here.
132
Table 5.4. Configurations for the Timing of Formation
Configuration Early formation Late formation
1a 1b 2a 2b 3 4a 4b 4c
University context
External champion
− − − − −
Leading funder
− − − − −
Large university
− − − − −
Elite university
− −
− − −
Champion or Funder
−
− −
Large and Elite
− − −
Social/political context
Large nonprofit sector − −
−
High urbanization − −
−
High liberalism − −
−
Best cases Private Great Lakes,
Public Great Lakes,
Northern Gold Coast
Mountain West, South Atlantic,
Southern Gold Coast,
Central Gold Coast
Consistency 1.00 1.00 1.00 1.00 0.81 0.90 0.96 0.75
Raw coverage 0.39 0.41 0.58 0.32 0.70 0.57 0.39 0.15
Unique coverage 0.23 0.25 0.29 0.03 0.70 0.45 0.27 0.02
Overall consistency 1.00 1.00 0.81 0.87
Overall coverage 0.64 0.62 0.70 0.87
Note:
= core causal condition (present);
= core causal condition (absent);
= complementary causal
condition (present);
= complementary causal condition (absent). Blank spaces indicate a “don’t care”
situation in which the causal condition may be either present or absent, and dashes (“−”) indicate a situation
in which the causal condition is purposely omitted from the analysis.
Early Formation. The first pattern is found in an Early-formation group, including the cases
of Private Great Lakes, Public Great Lakes, and Northern Gold Coast. For this group, the major
efforts for the field’s formation came from the requests of external stakeholders—such as
foundation funders, nonprofit leaders and managers, and community leaders—and the major
external funders were large grant-making foundations that demonstrated a commitment to
133
developing nonprofit management academic programs. Early on, when there was little recognition
of the field of study of nonprofit management, external university stakeholders (nonprofit and
community champions) seemed to educate and convince internal university actors of the need for
nonprofit management studies. On the one hand, they provided some direct guidelines to
universities, as illustrated by one interviewee:
“[A wealthy philanthropist] who was very concerned about nonprofit organizations in
our region and across the country approached our institution to start programs for
educating nonprofit leaders. The university did not respond initially, but eventually
agreed that [his idea] was excellent.”
On the other hand, the expressed interest in nonprofit management studies itself indirectly
stimulated the field’s formation by attracting internal actors’ attention. In a typical comment, one
of the interviewees provided the following explanation:
“Some of us inside were aware of the issues of philanthropy and the nonprofit sector
[and the need for academic programs]. However, it was true that… a sizable outside
grant from [one of the national foundations] was the catalyst in getting the attention of
senior faculty members and the administration, and that helped us move this forward
and institutionalize.”
Meanwhile, universities in the Early-formation group are mostly small, non-elite institutions.
This unexpected finding may be because large, elite universities tended to have stronger
boundaries between traditional disciplines and more complex bureaucracies, which, in turn,
inhibited the field’s early formation. To illustrate, one interviewee noted the following:
“Dr. Richard Cyert was a major figure in management education and a dean of the
Business school at Carnegie Mellon University. He said at the first nonprofit
134
management education conference in 1986, “Do not expect the prestige [universities
and] business schools to get interested in nonprofit programs”… I think that was
insightful at that time and as I look over the field, that is still the case. [Large and well-
recognized universities] have not been very interested in opening up to new ideas.
Many nonprofit programs came from smaller and less recognized universities like us.”
Another interviewee explained the advantages of being small and non-elite as follows:
“I found that I was very fortunate because of the lack of bureaucracy here—a small and
private university. I had a lot less bureaucracy and a lot of leeway to be entrepreneurial.”
The Early-formation group also shares some social and political factors at the state level,
such as a relatively large nonprofit sector, demographic homogeneity, and politically conservative
ideology. A large nonprofit sector in a region could help universities to recruit students and
practitioner faculty members with ease. One interviewee described his/her blessed experience as
follows:
“In [our city], nonprofits are very important and have a very long history. We have one
of the top five art museums and one of the big five orchestras in the country… Many
people here work for nonprofit organizations and contribute to and care about them…
[Therefore, we could easily] get people with senior leadership experience in the field
to come and teach our students. [For example,] we had courses on nonprofit law taught
by lawyers who were very prominent nationally for their special work with nonprofit
organizations. And we had some of the highly respected fundraisers who taught our
courses on fundraising.”
Although the QCA results report that demographic homogeneity and politically conservative
ideology are partly associated with early formation, interviewees in this study emphasized that
nationwide social and political pressures were much stronger than any regional pressures. In other
135
words, universities in the Early-formation group were more likely to be influenced by the overall
growth of the nonprofit sector and research initiatives at the national level, as depicted by multiple
interviewees:
“The main thing was the tremendous growth of the nonprofit sector, the number of
people employed, and the need for management training across the county. So we
needed a management program that tailors to the needs of nonprofit organizations
across the country.”
“[Rather, the major driving force was] national initiatives that were stimulated by the
interest being developed nationally—such as the Independent Sector’s research
committee for studying nonprofit organizations.”
Late Formation. The second pattern is found in the Late-formation group, including the
cases of Mountain West, South Atlantic, Southern Gold Coast, and Central Gold Coast. The
common feature among the universities in this group is that the field’s formation was mainly
motivated by the needs of internal university actors, including faculty, students, and administrators.
Clearly, because of the increased structuration of the nonprofit sector over time, faculty research
interest and the number of students who specifically wanted to study nonprofits increased
substantially, which, in turn, paved the way for the field’s formation. Interviewees shared details
on this dynamic as follows:
“The facilitating factor was an interest that I and other faculties had in doing this…
[After] I became interested in nonprofit management…, I decided to go outside of my
school and looked for allies and colleagues. After talking to lots of people all across
our campus, I found twelve faculty members who were interested in voluntarily
teaching courses and coordinating them… We put together a proposal to target the
university to ask for a right to grant a Master’s degree to students.”
136
“It came out of the interest of faculty members. The first program was just two general
nonprofit courses at the graduate and undergraduate level… But, particularly students
in the MPA program expressed a lot of interest in knowing more about nonprofit
management. So, we started adding courses [and creating degree programs] to respond
to that interest.”
Except for Mountain West, all three of the universities in this group had local family
foundations and individual donors as major external funders. Different funding streams are
observed between the Early-formation and Late-formation groups; the former relied more on large,
national grant-making foundations, whereas the latter relied more on small, local family
foundations and several wealthy donors. This variation may not be specific to the cases in this
study but, rather, represent a nationwide trend, according to one interviewee:
“At least for a while, the support of national foundations was important. But, I think
that time has largely passed… Most of the foundations that played a major role in the
development of nonprofit programs in the 1980s and the 1990s have disappeared…
And then, some smaller foundations have come down and played an important role
locally. Now, sort of local funders are helpful in supporting [the development of]
programs.”
In terms of organizational characteristics, the Late-formation universities are characterized
as large and elite. Large, elite institutions took more time to formally adopt nonprofit management
studies than did small, non-elite institutions, perhaps (as noted earlier) due to their rigid
administrative structures and conservative attitudes toward a novel field of study. One interviewee
reported the following:
“It took several years of work with our faculty, the university administration, the board
of trustees, and the state commission for higher education because we are a large public
137
university. We had to go through extra steps that perhaps [small, private universities]
did not have to go through.”
The Late-formation universities are commonly embedded in the regional context in which a
small nonprofit sector, a heterogeneous population, and politically liberal citizens and
governments existed. However, the QCA results show marginal effects of these regional context
factors, identifying them as peripheral causal conditions. In my interviews, faculty members also
stressed that the national mood or their own personal interest, rather than the region’s social and
political conditions, was the major driving force that led to the field’s formation:
“It happened with the growing interest in the nonprofit sector throughout the scholarly
and professional communities. I do not think there was a particular social or political
event [within the community] that I can articulate.”
“I just was very interested in doing it and finding other people who wanted to join it.
There were no social and political reasons behind the creation of the program.”
Institutional Stability of the Study Field
Table 5.5 shows various causal combinations that lead to the High-stability and Low-stability
outcomes.
34
Solutions 5 and 7 evaluate the effect of each of the four university factors, and
solutions 6 and 8 evaluate the effect of all university and social/political factors. Solutions 6 and 8
show that there are two different paths to High stability (with solution 6a as the most empirically
relevant causal recipe) and another two different paths to Low Stability (with solution 8b as the
most empirically relevant causal recipe). In both solutions 6 and 8, two compounding sets of
university factors are identified as core conditions and three social/political factors are identified
34
Central Gold Coast is not covered by any of these solutions and, thus, remains a puzzle in this analysis.
138
as peripheral conditions, indicating that social and political context factors are sufficient for the
outcomes only to a limited extent. Overall, the QCA results in Table 5.5 suggest that there are two
central pathways toward the field’s institutionalization.
Table 5.5. Configurations for the Institutional Stability
Configuration High stability Low stability
5 6a 6b 7a 7b 7c 8a 8b
University context
Iconic leader
− −
− −
Leading funder
− − − −
Large university
− −
− −
Elite university − −
− −
Leader or Funder
−
− −
−
Large and Elite
− − − −
Social/political context
Large nonprofit sector −
− − −
High urbanization −
− − −
High liberalism −
− − −
Best cases Mountain West,
Public Great Lakes
Private Great Lakes, South Atlantic,
Southern Gold Coast, Northern Gold Coast
Consistency 0.85 0.94 0.87 0.75 0.90 1.00 1.00 0.89
Raw coverage 0.59 0.38 0.30 0.17 0.41 0.23 0.17 0.62
Unique coverage 0.59 0.26 0.17 0.14 0.38 0.23 0.13 0.58
Overall consistency 0.85 0.89 0.87 0.91
Overall coverage 0.59 0.55 0.78 0.75
Note:
= core causal condition (present);
= core causal condition (absent);
= complementary causal
condition (present);
= complementary causal condition (absent). Blank spaces indicate a “don’t care”
situation in which the causal condition may be either present or absent, and dashes (“−”) indicate a situation
in which the causal condition is purposely omitted from the analysis.
High Stability. The first path leads to High stability and is followed by a group of
universities, including Mountain West and Public Great Lakes. In this group, strong and consistent
139
leadership from founder faculty or powerful external funders exists. Universities in this group do
not share common organizational features and, thus, are either large and elite or small and non-
elite institutions. Against all expectations, social and political environments surrounding these
universities are characterized by a small nonprofit sector, less demographic diversity, and
conservative political ideology.
More specifically, the institutional stability of nonprofit management studies in universities
seems to depend heavily on the degree of interest and enthusiasm of university actors. Indeed,
interviewees affiliated with the universities in the High-stability group emphasized that their iconic
champions’ effective leadership was the most critical factor that led to the strong
institutionalization of nonprofit management studies at their institutions. One interviewee who has
been regarded as one of the iconic leaders at his/her university addressed this point as follows:
“The facilitating factors were leadership, leadership, and leadership... I have to give a
lot of personal credit to a couple of former deans. [Especially, one of them] was
enormously not driven by the tradition of academic disciplines and silos… We had
permission to think across boundaries… And, I have not gone away—it is very unusual.
In most academic settings, faculties move on to other universities… [Also, there was]
a gentlemen [as a community champion] who brought a lot of energy and passion to
this work… [In short, our achievements all are due to] some enlightened administrators,
some champions on the faculty side, and willing community sponsors and funders.”
In most cases, the university’s organizational characteristics and regional contexts did not
exert a significant influence on the institutionalization process of nonprofit management studies.
In my interviews, respondents generally agreed on this and further suggested internal politics
within universities as a more important determinant for the institutional stability of the field of
study. This mirrors the idea of games universities play suggested by Young (1998), who argued
140
that the long-term stability of nonprofit academic centers hinges on a few powerful stakeholders—
including top administrators and faculties of various traditional schools and departments—and,
thus, that centers should play games with those stakeholders when appealing for their support.
Universities in the High-stability group clearly attempted to address “powerful, but largely
indifferent and sometimes antagonistic, internal university interests” (Young 1998, p. 135) to
ensure the viability of the field of study. They successfully reconciled the field’s own needs with
those of the powerful stakeholders and attracted considerable support from stakeholders, including
university presidents, provosts, deans, and senior tenured faculties, as I learned from one
interviewee:
“[In an effort to institutionalize nonprofit management studies,] we depended on very
senior faculty members [at the school of law, history, sociology, and so on]… We did
not want to marginalize this study inside the university. We wanted it to be seen as
something that was endorsed, promoted, and studied by most senior faculty members,
[so that it could be] fully institutionalized.”
Low Stability. The second path contributes to Low stability and is followed by a group of
universities including Private Great Lakes, South Atlantic, Southern Gold Coast, and Northern
Gold Coast. In contrast to the High-stability group, universities in this group typically do not have
charismatic leadership involved in the effort to institutionalize nonprofit management studies, and
this is due mainly to the absence of an iconic faculty or funder. As small, non-elite institutions,
they also have relatively little human and material resources. Surprisingly, they are located in
regions that have a large nonprofit sector, a diverse population, and political liberalism—and thus
a strong demand for this highly institutionalized field of study is expected.
141
Specifically, for universities in the Low-stability group, their iconic champions’ effective
leadership did not exist or has disappeared over the years. Consequently, those universities could
not establish a stable base for nonprofit management studies, or they had to discontinue a portion
of their established programs. One interviewee described his/her experience as follows:
“[Our research center is closed because] it lost funding from its principal sponsor. The
foundation, which was the lead funder of the center, became increasingly unclear about
what it was actually getting for its investments. It started to ask questions about the
value of its investments. At that time, our university had gone through an interim
president period, so there was a lack of key leadership at the university to really defend
the center’s interests effectively. Finally, the foundation elected to withdraw the
funding.”
Another interviewee who is now retired from his director position frankly told me the
following:
“After I left [the directorship of nonprofit management programs], my successor
became a dean. As dean, he was absolutely an opponent of the study of nonprofit
organizations. The dean refused to allocate any money to the programs... I am just
watching it with sadness as [the field of study] begins to collapse.”
My argument may not be limited to the cases in this study; rather, it can be generalized to
other university contexts. For example, one interviewee summarized the situation at Yale
University as follows:
“The first university-based nonprofit research institute, the PONPO at Yale, was shut
down several years ago. The PONPO generated a lot of great research. So why is it
gone? This is what happens in universities. The PONPO started with not only some
generous foundation supports but also the strong personal backing of the president at
142
Yale at that time. But over time, those conditions changed... It is about money and
people. When the financial support of foundations disappears and when the personal
support of key administrators and faculty disappears, the programs unfortunately tend
to go away.”
In contrast to the High-stability group, universities in the Low-stability group mostly failed
to negotiate games; they were unsuccessful in either convincing administrators or faculty members
who were indifferent or in promoting interdisciplinary collaborations. As a result, they were unable
to mobilize sufficient human and material resources, change structural systems, or cultivate
powerful allies for the strong institutionalization of nonprofit management studies. One
interviewee who was a program director affiliated with the department of social work phrased the
challenge he/she had faced when asking for support from various parts of the university as follows:
“I met a chairman of the department of public administration and then asked, “Could
you identify some [faculty in your department] to join me?” He said, “No, in fact we
do not want to be any part of your collaborative effort. We will run our own program
later.” [Since then,] we just kept degree programs going without any representative
person from the public administration department. My personal suspicion is that the
chairman could be quite arrogant… [because] he knew more people on campus than
me and he could run his own program without any help from me—I was a faculty
member [who is affiliated with a department of social work,] a very low-status
department in the university system.”
5.5. Discussion
Nonprofit management studies as an emerging field has been incorporated into university
systems at various times and in various ways. However, factors that affect variation in the patterns
of the development of the field of study have not been explored in much detail. This chapter thus
143
investigated two broad sets of explanations—university dynamics and social and political
contexts—for the evolution of nonprofit management studies at different universities, drawing on
a comparative case study of seven universities.
The QCA results show that university dynamics factors tend to more importantly shape the
evolutionary process than do social and political context factors. The results also demonstrate a
variety of pathways by which nonprofit management studies is first introduced into higher
education institutions and then institutionalized as a legitimized study field. More specifically,
with respect to the formation of the field of study, the pathways differ markedly for early-adopter
and late-adopter universities, as expected. For early-adopter universities, the combination of the
active engagement of external stakeholders (e.g., nonprofit and community leaders and foundation
funders), organizational characteristics of being small and non-elite, and the existence of a well-
developed nonprofit sector at the state level is the most common pathway to establishing the field.
For late-adopter universities, however, the combination of strong demand from internal
stakeholders (e.g., top administrators, faculty members, and students), university status as a large
and elite institution, and a demographically diverse and politically liberal atmosphere in the region
is sufficient to create the field. Based on my interviews with faculty members, I speculate that such
a difference may reflect a nationwide institutional change over time in the recognition of nonprofit
management research and education.
In addition, with respect to the institutionalization of the field of study, the pathways vary
considerably with the degree of institutional stability. In line with expectations, the pathway for
universities with highly stable institutional systems for the field is rooted primarily in iconic
leaders’ charismatic leadership and powerful stakeholders’ support. The pathway for universities
with less stable systems contains elements such as a lack of strong leadership and support from
144
influential stakeholders and the organizational characteristics of being small and non-elite.
Meanwhile, social and political context factors in my analysis have surprisingly few—and even
unexpected—effects on the institutionalization of the field. This may indicate that internal
university politics and dynamics are more salient than larger macro-social environments for
nonprofit management studies to be welcomed by and institutionalized within universities.
35
Although the field of nonprofit management has been increasingly visible in the higher
education context, overall, it remains a discipline that is neither fully recognized nor fully
institutionalized. All of my interviewees agreed on this point, speaking with one voice as follows:
“I do think that [nonprofit management studies] is unique enough in its characteristics
to get some special attention, but… I do not think it is a separate discipline yet.”
“We are still struggling with finding a real focus of nonprofit management studies, still
debating where nonprofit studies should be housed, and still trying to create
competencies for [its research and education programs]… And we are still talking
about its name—some people do not like the name nonprofit studies.”
“There are issues about whether there is a conceptual, intellectual, and philosophical
foundational view of the field and whether our students, alumni, and faculty can be
very secure… It is not a fully acknowledged discipline yet. It is still evolving.”
In this sense, they evaluated the maturity of their current research and education programs in
nonprofit management as an average of seven on a scale from one to ten; more specifically, the
35
Even though the findings of this chapter more emphasize internal university dynamics as a key factor contributing
to the development of academic programs than the broader institutional environment, we should be cautious in
interpreting these findings. It is because the effect of institutional context is not directly tested here due to the inherent
limitation of QCA techniques. It is also because the seven universities I purposely selected can be regarded as leading
universities in this field—considering a short history of the field—and thus the findings may show the behavior of
“pioneer” institutions rather than that of “general” institutions.
145
main reason for this rating was that their programs had certainly passed the start-up stage, but there
was much more work to be done to achieve full institutionalization.
What might accelerate the field’s institutional development, then? In my interviews, three
broad sets of recommendations were proposed. First, in terms of systemic mechanisms for the field,
it is important for universities to develop internal resources and strengths, as one interviewee noted:
“I think one of the mistakes the field made is that it is too comfortable with depending
on some major outside funders [and top administrators], who enter in a big way and
then disappear to some extent. You cannot control changes at the top, but you can to
some extent by developing your own internal strength in terms of faculty lines and your
own financial support… Leaders in the nonprofit field have to be street smart about
getting internal support in a way that hopefully will be protected against… what they
have no control over.”
The experiences of seven universities in this study indeed show that the field’s institutional
development has largely hinged on certain “money and people,” and therefore it has been
extremely vulnerable to changes in funding streams and leadership. At first, a sizable outside grant
or heroic individual’s entrepreneurship was a blessing to the field; however, later, such features
became a curse, at least in some cases. Lessons from these universities’ past experiences suggest
that it is critical for the field to develop its own self-sustaining system for financial and human
resources. Efforts to do so may include diversifying sources of financial resources and preparing
the next generation of faculty members.
Second, given the field’s lack of intellectual coherence, it is important for the field to advance
the quality of research and to establish theoretical and methodological foundations that are more
rigorous. When I asked the interviewees whether their research and education programs in
nonprofit management take a particular conceptual and philosophical approach, all of them
146
responded as follows: “No, but we are working toward that.” Considering the fact that the selected
seven universities are often regarded as leading institutions in the field, their answers are surprising
and somewhat frustrating. For the field’s future development, therefore, interviewees highlighted
the need to continue to advance the research frontier and develop more theories and rigorous
analyses of the nonprofit sector so that the knowledge base of the field could be both expanded
and cohesive. The field can justify itself as a distinct field of study with clear lines delineating it
from other fields of scientific inquiry only after it achieves more obvious and definable intellectual
foundations. One interviewee captured this idea as follows:
“One of the things that holds us back from what we should accomplish in the world of
nonprofits is doing high-quality research. As we get more people doing high-quality
research, more people will take [nonprofit management studies] seriously as an
academic field, not just something for social good.”
Third, there is a need for the field to develop its curriculum in a manner that maintains a fine
balance between different types of educational needs. In most cases, interviewees reported that
their curriculums tend to be more oriented toward professional practice, primarily focused on
students at the Master’s level, and generally taught by part-time adjunct faculties composed of
practitioners. They acknowledged this practitioner orientation as their weaknesses to some extent:
“If students want us to prepare them for the Ph.D. work, it would be our weakness
because our [current education programs] are more applied ones… There are no
doctoral-level classes now—you just can take Master’s classes and work independently
with one of our nonprofit professors... I would love to have a really defined doctoral
program for the field.”
147
“I think at the doctoral level, we are weak and we can do a much better job… The
undergraduate level is much tougher. I do not think there is a well-developed model
yet of undergraduate education [in the field].”
“For many years, our [Master’s and certificate] programs have been taught mostly by
adjunct faculty members with a variety of fields of expertise and specializations… I
always value the importance of adjunct faculty, but we need to have a better balance
between a full-time and adjunct faculty—maybe half and half.”
This might call for greater attention to achieving a balance between practitioner orientation
and academic orientation and institutionalizing that balance by, for example, expanding research-
oriented doctoral programs and hiring more full-time faculty who can teach theories of the
nonprofit sector. Additionally, more efforts to meet demands from different marketplaces for
nonprofit management education are required. Several interviewees noted that the field now has
three different marketplaces for education at the local, national, and international levels, and the
traditional contents and formats are not equally effective for every marketplace. For the field’s
long-term competitiveness and sustainability, there may be a need for continuing efforts to devise
various educational contents and formats—for example, online programs or short
seminar/workshop series—that can make it possible to deliver a high-quality education to students
in different cities, states, and nations.
How the field might evolve in the future remains unpredictable. However, the future of
nonprofit management studies seems to be promising—at least at this moment in time. The reason
for this is that the majority of interviewees predict that the nonprofit sector will continue to grow,
and there are many important changes and innovations in the sector that remain to be studied. More
importantly, they are relatively optimistic about our own potential and enthusiasm for developing
the field. As one interviewee noted, “We will continue to get better.”
148
CHAPTER 6. CONCLUSION
6.1. Summary of Findings
The dramatic and ongoing growth of nonprofit management programs in the United States
merits research attention. Over the last several decades, scholars and practitioners have advocated
for more, and better, nonprofit management training. A lack of management training in the sector
certainly explains some of the growth of nonprofit management programs in universities, but I
argue that demand or needs within the sector are insufficient for explaining growth. To begin with,
management training in the nonprofit sector presumably could have been addressed through
professional development programs run by intermediary organizations, implying that the
emergence of university programming is not obvious or somehow required for satisfying demand.
The growth of management programs also needs to be contextualized in relation to evidence of
efficacy. Individuals derive salary benefits from management credentials like an MBA, but the
effect of MBAs on profits and revenue growth is uncertain even in the business sector (Khurana
2007). University programs in nonprofit management could lead to gains in nonprofit efficiency
and effectiveness, but growth has preceded empirical evidence, justifying studies that
contextualize demand and provide more nuanced accounts for growth.
To that end, this dissertation documents the evolution of nonprofit management in American
society and explores various historical, social, and organizational mechanisms associated with the
evolution, drawing on institutional sociology and other organizational theories. First, it provides a
theoretical framework for understanding the structuration of the nonprofit sector by closely
examining the development of the field and institutional processes behind the development.
Additionally, in moving the focus from the evolution of nonprofit management in a general context
149
to that of the higher education, this dissertation analyzes factors associated with the emergence,
development, and institutionalization of nonprofit management research and education. More
specifically, it utilizes both quantitative and qualitative methods to identify various factors that
explain the incorporation of nonprofit management studies programs into higher education
institutions and explores variations in its early development process.
The empirical findings in Chapters 3, 4, and 5 provide compelling evidence of the rapid,
wide expansion of nonprofit management studies nationwide. My longitudinal data show that
university-based research and education programs in nonprofit management were virtually
nonexistent before the mid-1970s, but over the last four decades, they have expanded dramatically
and developed in a variety of forms and venues. In fact, of the sample of 1,451 American
universities and colleges, approximately 350 created new programs in nonprofit management
between 1971 and 2011. The research and education programs that have developed have taken
various forms, including stand-alone degrees, concentrations, tracks, certificates, and related
options at the graduate and undergraduate levels; it is estimated that 58 universities granted stand-
alone degrees in nonprofit management, 270 universities offered minor, specialization, emphasis,
and certificate programs in nonprofit management, and 68 universities established research centers
for nonprofit studies. Research and education programs have emerged in diverse academic settings
as well. When considering only the very first program created at each university, more than one-
third of the programs (35.9%) were found in a public affairs or political science setting, slightly
more than one-quarter (28.5%) were in arts and sciences, interdisciplinary studies, or professional
studies, slightly less than one-quarter (22.1%) were in business, and the remainder (13.5%) were
in social work or other social sciences.
36
36
Additionally, it is estimated that 180 universities offered more than one nonprofit management program and 29
universities had more than one academic home for nonprofit management studies between 1971 and 2011.
150
Moreover, my empirical findings support the argument that social and cultural processes
shape the rate at which nonprofit management studies programs develop. In other words, the
findings demonstrate that the institutional environment, measured at the macro- and meso- level,
provides important explanatory power for the emergence and growth of nonprofit management
programs in universities. With my integrated framework I demonstrate that organizational
characteristics also matter a great deal for the emergence and evolution of university-based
programs, demonstrating that the institutional explanations and strategic or resource dependence
perspectives are not mutually exclusive. Case studies provide extensive, important details on
political dynamics and individual actors that contribute to the creation of university programs. I
recognize and appreciate the importance of these micro-level mechanisms and individual-level
actors, but I suggest that the narrow focus of case studies often overlooks broader social conditions
and environmental dynamics. My quantitative analyses consequently offer a wide lens perspective
on nonprofit management programs in universities, informing and contributing to the growing
literature on nonprofit management.
At the national level, a series of nationwide trends—including the growth in nonprofit
organizations, professional associations, government grants, management support and training
programs, scholarly publications, and news articles—created institutional conditions for the
diffusion of nonprofit management. Over time, research and education in nonprofit management
have become more legitimate in higher education as a distinctive field, triggering additional
growth. At the regional level, constant and dynamic characteristics of states are relevant for
understanding nonprofit management programming in universities. In general, I find some
evidence that time-varying indicators like population heterogeneity, urbanization, and political
conditions in states contribute to the emergence and development of programs. I also find that
151
university location, a measure that does not vary over time, also informs program adoption.
Universities in the Northeast and the West led the field of nonprofit management studies until the
mid-1990s, but differences have become less pronounced since then. At the organizational level,
structural and demographic properties and internal politics of universities have a strong influence
the rate and extent of program growth. In the initial stage of program development, large, non-
elite, demographically diverse universities with relevant “receptor sites” tended to be the most
engaged in creating and developing programs.
Though this is a long list of university-level indicators, collectively they show a general
pattern. Nonprofit management programs emerged initially in large and accessible universities that
privileged practical or work-relevant training, universities with a breadth of curricular programs.
It is important to emphasize, though, that historical changes have contributed to different patterns
in program creation over time. That is, universities are embedded in an institutional environment
that influences how, when and why they create university programs. University-level conditions
are pivotal for understanding the variation observed in the program creation rate, preferred
disciplinary setting, and institutionalization process, but historical processes situate or
contextualize that variability. Many university-level programs were created before 2000 but my
analyses offer evidence that the post-2000 period constitutes a critical juncture in the trajectory of
nonprofit management programs. Adoption rates increased markedly, and degree programs
became much more common.
I have offered several explanations for this critical juncture. The first is the Sarbanes-Oxley
Act and associated legitimacy concerns in the business sector. I suggest that nonprofit management
gained importance in the post-2000 period in the United States, both as a spillover reaction to
corruption scandals in the for-profit sector and as a more positive appreciation of management
152
itself. A second explanation, which can co-exist with the first, is the rise of a “business logic”
within the nonprofit sector. The NPM reform movement was premised on the belief that
bureaucracies needed to embrace competition and market-based strategies; a similar phenomenon
took off in the nonprofit sector after 2000.
Taken together, my dissertation complements prior research on the expansion and
development of nonprofit management over the last few decades. Rather than constructing my
argument around resources, internal demands, and organizational politics within universities, I
have argued that the institutional environment situates adoption patterns and offers a cultural
explanation for nonprofit program growth in universities. By bringing attention to the structuration
of the nonprofit sector and critical junctures in history, while also considering the role of university
characteristics, my research highlights how theories like sociological institutionalism matter for
public management research.
6.2. Limitations and Additional Directions for Research
Given the scope of my research questions and the availability of the data, I should
acknowledge that there are limitations to this study. A first limitation concerns the potential source
of bias in the process of data collection. In this study, I define university-based nonprofit
management studies programs as generic management programs using the field’s name
(“nonprofit”) in their titles. Therefore, programs narrowly focusing on one subgroup of nonprofit
organizations and thus using the specific subfield’s name in their titles (e.g., “human services” or
“arts”) are excluded from the analysis. This implies the potential underrepresentation of subfield-
specific programs that address nonprofit management issues within their curriculums. In particular,
social work (human or social services management) programs are often recognized as “nonprofit
153
management programs by another name” (Mirabella 2014) but are excluded from this study. Thus,
a closer examination of these subfield-specific programs would (in my view) be a prime avenue
for future research on the history of nonprofit management studies programs.
A second limitation concerns the fact that my event history analysis in Chapters 3 and 4 is
solely focused on the first adoption of nonprofit management studies programs at each university
as a dependent variable. Of course, program adoption is a repeatable event from the perspective of
universities and colleges; in other words, universities and colleges can incorporate more than one
program at different times; thus, second, third, or fourth adoptions may exist. However, I take only
first adoptions into consideration not only because it is difficult to collect precise historical data
on all of the existing programs but also because multiple events at universities remain relatively
rare. For these reasons, an analysis of multiple adoptions is beyond the scope of this study, but it
might be an interesting venue for future research.
A third limitation concerns possible misleading results due to omitted variable bias and
measurement problems with the indicators I use. In Chapters 3, 4, and 5, I tested many indicators,
and I attempted to control for all of the organizational characteristics that I thought might affect
the adoption of university programs. Still, new work may demonstrate the relevance of concepts
and measures that I did not consider. In my quantitative models, social and political setting factors
such as the size and composition of the nonprofit sector, population diversity, economic status,
and political ideology are measured at the state level. I was surprised that these indicators had only
a weak relationship with the evolution of nonprofit management studies programs across all of the
models estimated. Though I attempted to identify and use the best longitudinal measures available
at the state level, better alternatives may exist. A lack of longitudinal data at the county, city, and
community levels prevents me from conducting additional analyses, but it is possible that those
154
social and political contexts could be more meaningful. Thus, one direction for future research
would involve conducting an analysis with data at the county, city, or community level.
Finally, as an early attempt to apply QCA to the study of nonprofit management, my analysis
in Chapter 5 is limited both in the number of its cases and the sophistication of its analysis.
Specifically, my findings are based on a relatively small sample of seven cases and a relatively
simple statistical modeling with five sets of causal conditions. The seven cases that I selected are
not likely to capture all of the interesting dynamics in the development process of nonprofit
management studies. Moreover, the five sets of causal conditions that I identified as being
important drivers of the field’s development are not likely to be exhaustive, and there might be
other important conditions (or combinations of conditions) that lead to the field’s development.
Despite the limitations I have noted, I am confident that my work offers a solid foundation for
future research while also contributing to the extant research literature.
6.3. Contributions to the Field
This dissertation makes several theoretical, empirical, and methodological contributions to
the field of nonprofit management. First, it contributes to the theoretical literature by offering a
comprehensive conceptual framework for a better understanding of the nature of nonprofit
management development, and it expands the presence of neoinstitutionalism and social
constructivism in public and nonprofit management. Previous studies on the evolution of nonprofit
management have discussed a variety of carriers of nonprofit management and a variety of
historical, social, and organizational context factors in isolation (e.g., for a discussion of individual
carriers, see Brilliant 2000; Connor et al. 1999; Smith 1997; for historical and social context factors,
see Hall 1996; O’Neill 2005; for organizational context factors, see Hall 1992; Young 1998).
155
However, some studies focus on describing and documenting the phenomenon rather than on
developing a causal argument based on theory; other studies posit a limited set of mechanisms.
Unlike those previous studies, my dissertation offers a macro-level conceptual frame that attends
to sources and flows of nonprofit management in the institutional environment. In doing so, my
study sheds light on the underappreciated dynamics that contribute to the emergence and
development of nonprofit management.
Second, this study contributes to the development of empirical research by constructing a
new longitudinal database on nonprofit management studies programs in U.S. higher education
institutions. The database itself complements the most well-known source of information on
university-based nonprofit programs, maintained by Seton Hall University (http://academic.shu.
edu/npo). My database builds upon and extends this Seton Hall University database, but the most
important difference is that my database is not limited to the dependent variable, nonprofit
management programs.
37
With my sample of universities and colleges, I am able to assess the
importance of several different variables on the development of nonprofit management studies
programs.
Finally, this study’s methodological contribution is the application of two techniques—event
history analysis and QCA—that have great promise for public and nonprofit scholarship. In
Chapters 3 and 4, this dissertation adopts event history analysis to overcome the limitations of
37
There are some important differences between my datasets. First, the two databases draw their samples from
different populations of American universities and colleges. As a result, the Seton Hall University database has
approximately 1,360 universities as a valid sample, whereas my database has 1,451 universities. Second, the two
databases have different time frames. The Seton Hall University database consists of data collected from 1990 to 2014,
whereas my database has a longer time frame, from 1971 to 2011. Third, although the two databases essentially define
the scope and type of university-based programs in a manner that is very similar, my database includes information
on research centers for nonprofit management studies, whereas the Seton Hall University database does not. Fourth,
my database contains more detailed historical information that is not provided in the Seton Hall University database,
such as the year that each type of program—stand-alone degree, non-degree, and research center, respectively—was
established.
156
existing studies, characterized by cross-sectional or retrospective research designs and case studies
(e.g., Katz 1999; Smith 1999; O’Neill 2005). For studying innovation diffusion in particular, event
history analysis also is superior to alternatives because it allows for the inclusion of time-varying
covariates and right-censored cases in the analysis. By using event history models, therefore, both
the timing of program adoption and the conditions under which the adoption takes place can be
more precisely estimated. In Chapter 5, furthermore, the dissertation employs the relatively new
approach of QCA for a comparative case study. QCA enables this study to analyze cross- and
within-university patterns related to the gradual development of nonprofit management studies
more systematically, bridging the gap between case studies and statistical analysis. This technique
also allows a new method of understanding cause-effect relationships—namely, a set-theoretic
understanding of how causal factors combine to produce a given outcome—inherent in the study
field’s development. In summary, this dissertation contributes to methodological development by
attracting renewed attention to these two methods as powerful techniques for a diffusion analysis
or multiple case analysis, along with showing how these methods can be applied to the study of
nonprofit management.
6.4. Future Directions and Predictions
The findings of this dissertation permit several tentative hypotheses for nonprofit
management. Nonprofit management now appears to be at the end of its early institutionalization
stage and is moving towards full-institutionalization. Tolbert and Zucker (1996) suggested that
there are three phases of the institutionalization process: pre-institutionalization, semi-
institutionalization, and full-institutionalization. According to Greenwood et al. (2002), a semi-
institutionalization stage involves the process of theorization and diffusion of new ideas and
157
practices. Indeed, nonprofit management is successfully theorized in the U.S. context, as the highly
correlated six different measures of the structuration of nonprofit management imply. In addition,
nonprofit management ideas and practices are diffused throughout the country, as indicated by the
wide expansion of research and education programs in nonprofit management. I expect nonprofit
management to become more fully institutionalized over time. Evidence from this dissertation
suggests that expansion is ongoing and unlikely to stop in the near future.
With respect to nonprofit management in the higher education context, empirical findings
indicate that nonprofit management studies is now largely accepted as an emerging field of study
in universities and colleges, but it remains in its early stage of development. More specifically,
few elite research universities have created degree programs specifically for nonprofit
management. I expect growth in this segment of the university landscape, a trend that would offer
a powerful measure of legitimacy for nonprofit studies. Finally, the field remains fragmented
despite the growth I have documented, relying on a variety of intellectual sources including the
social sciences, management, and humanities. A final hypothesis I have is that nonprofit
management will become less fragmented in universities. Though nonprofit management
programs may continue to be embedded in specialized professional programs like business
management, public management, or social work, I predict that degree programs increasingly will
reflect the breadth of the nonprofit sector itself; for instance, I would expect nonprofit degree
programs housed in social work to approve nonprofit elective courses offered in fields like
education, the arts, and religion—not just courses with clear ties to human services. Overall, a
specialized degree in nonprofit management will become more common, constructed around
unique knowledge claims that will standardize core curricular content.
158
BIBLIOGRAPHY
Abbott, Andrew. 1988. The System of Professions: An Essay on the Division of Expert Labor.
Chicago, IL: University of Chicago Press.
_____________. 1991. The Order of Professionalization: An Empirical Analysis. Work and
Occupations 18(4), 355-384.
Acar, Muhittin and Peter J. Robertson. 2004. Accountability Challenges in Networks and
Partnerships: Evidence from Educational Partnerships in the United States. International
Review of Administrative Sciences 70(2): 331-344.
Agranoff, Robert and Michael McGuire. 2001. Big Questions in Public Network Management
Research. Journal of Public Administration Research and Theory 11(3): 295-326.
_________________________________. 2003. Collaborative Public Management: New
Strategies for Local Governments. Washington, DC: Georgetown University Press.
Alexander, Jennifer and Renée Nank. 2009. Public-Nonprofit Partnership: Realizing the New
Public Service. Administration & Society 41(3): 364-386.
Alford, John and Owen Hughes. 2008. Public Value Pragmatism as the Next Phase of Public
Management. The American Review of Public Administration 38(2): 130-148.
Allard, Scott W. 2009. Out of Reach: Place, Poverty, and the New American Welfare State. New
Haven, CT: Yale University Press.
Allison, Paul D. 1982. Discrete-Time Methods for the Analysis of Event Histories. Sociological
Methodology 13(1): 61-98.
_____________. 2010. Survival Analysis Using SAS: A Practical Guide, 2
nd
ed. Cary, NC: SAS
Institute Inc.
Andrews, Rhys and Tom Entwistle. 2010. Does Cross-Sectoral Partnership Deliver? An Empirical
Exploration of Public Service Effectiveness, Efficiency, and Equity. Journal of Public
Administration Research and Theory 20(3): 679-701.
159
Anheier, Helmut K. 2009. What Kind of Nonprofit Sector, What Kind of Society? American
Behavioral Scientist 52(7): 1082-1094.
Anheier, Helmut K. and Lester M. Salamon. 2006. The Nonprofit Sector in Comparative
Perspective. In The Nonprofit Sector: A Research Handbook, 2
nd
ed., edited by Walter W.
Powell and Richard Steinberg, 89-114. New Haven, CO: Yale University Press.
Bakardjieva, Maria. 1992. Home Satellite TV Reception in Bulgaria. European Journal of
Communication 7(4): 477-489.
Ben-Ner, Avner and Theresa Van Hoomissen. 1992. An Empirical Investigation of the Joint
Determination of the Size of the For-profit, Non-profit and Voluntary Sectors. Annals of
Public and Cooperative Economics 63(3): 391-415.
Berger, Peter L. and Thomas Luckmann. 1967. The Social Construction of Reality: A Treatise in
the Sociology of Knowledge. New York, NY: Doubleday Anchor.
Berman, Evan M. 1999. Professionalism among Public and Nonprofit Mangers: A Comparison.
American Review of Public Administration 29(2): 149-166.
Berry, Jeffrey M. and David F. Arons. 2003. A Voice for Nonprofits. Washington, DC: Brookings
Institution Press.
Berry, William D., Richard C. Fording, Evan J. Ringquist, Russell L. Hanson, and Carl Klarner.
2010. Measuring Citizen and Government Ideology in the American States: A Re-appraisal.
State Politics and Policy Quarterly 10(2): 117-135.
Bies, Angela L. and Amy S. B. Blackwood. 2007. Accountability, Ethics, Evaluation, and
Governance in Nonprofit Management Education: Trends and Treatment. Journal of Public
Affairs Education 13(3/4): 519-547.
Binder, Amy. 2007. For Love and Money: Organizations’ Creative Responses to Multiple
Environmental Logics. Theory and Society 36(6): 547-571.
Blackwood, Amy S., Katie L. Roeger, and Sarah L. Pettijohn. 2012. The Nonprofit Sector in Brief:
Public Charities, Giving and Volunteering, 2012. Washington, DC: The Urban Institute.
160
Accessed Oct 22, 2012. http://www.urban.org/UploadedPDF/412674-The-Nonprofit-
Sector-in-Brief.pdf.
Blackwood, Amy S., Kennard T. Wing and Thomas H. Pollak. 2008. The Nonprofit Sector in
Brief—Facts and Figures from the Nonprofit Almanac 2008: Public Charities, Giving, and
Volunteering. Washington, DC: The Urban Institute. Accessed Oct 22, 2012.
http://www.urban.org/UploadedPDF/411664_facts_and _figures.pdf.
Block, Stephen R. 1987. The Academic Discipline of Nonprofit Organization Management: Past,
Present and Future. PhD Diss., Graduate School of Public Affairs, University of Colorado at
Denver.
Blossfeld, Hans-Peter and Götz Rohwer. 1995. Techniques of Event History Modeling: New
Approaches to Causal Analysis. Mahwah, NJ: Lawrence Erlbaum.
Blossfeld, Hans-Peter, Katrin Golsch, and Götz Rohwer. 2007. Event History Analysis with STATA.
Mahwah, NJ: Lawrence Erlbaum.
Borgna, Camilla. 2013. Fuzzy-Set Coincidence Analysis: The Hidden Asymmetries. COMPASSS
Working Paper Series 2013-72. Accessed January 30, 2015.
http://www.compasss.org/wpseries/Borgna2013.pdf.
Boris, Elizabeth T. 2006. Introduction: Nonprofit Organizations in a Democracy—Roles and
Responsibilities. In Nonprofits & Government: Collaboration & Conflict, 2
nd
ed, edited by
Elizabeth T. Boris and C. Eugene Steuerle, 1-35. Washington, DC: The Urban Institute Press.
Boris, Elizabeth T. and C. Eugene Steuerle, ed. 2006. Nonprofits and Government: Collaboration
and Conflict, 2
nd
ed. Washington, DC: Urban Institute Press.
Boris, Elizabeth T., Erwin de Leon, Katie L. Roeger, and Milena Nikolova. 2010. Human Service
Nonprofits and Government Collaboration: Findings from the 2010 National Survey of
Nonprofit Government Contracting and Grants. Washington, DC: The Urban Institute.
Accessed October 9, 2012. http://www.urban.org /uploadedpdf/412228-nonprofit-
government-contracting.pdf.
161
Boston, Jonathan, John Martin, June Pallot, and Pat Walsh. 1996. Public Management: The New
Zealand Model. New York, NY: Oxford University Press.
Box, Richard C., Gary S. Marshall, B.J. Reed, Christine M. Reed. 2001. New Public Management
and Substantive Democracy. Public Administration Review 61(5): 608-619.
Box-Steffensmeier, Janet M. and Bradford S. Jones. 1997. Time is of the Essence: Event History
Models in Political Science. American Journal of Political Science 41(4): 1414-1461.
__________________________________________. 2004. Event History Modeling: A Guide for
Social Scientists. New York, NY: Cambridge University Press.
Brinkerhoff, Jennifer M. and Derick W. Brinkerhoff. 2002. Government–Nonprofit relations in
Comparative Perspective: Evolution, Themes and New Directions. Public Administration
and Development 22(1): 3-18.
Brint, Steven G., Lori Turk-Bicakci, Kristopher Proctor, and Scott Patrick Murphy. 2009.
Expanding the Social Frame of Knowledge: Interdisciplinary, Degree-Granting Fields in
American Colleges and Universities, 1975–2000. Review of Higher Education 32(2): 155-
183.
Bromley, Patricia and Charlene Orchard. Forthcoming. Managed Morality: The Rise of Codes of
Conduct in US Nonprofit Associations, 1994-2011. Nonprofit and Voluntary Sector
Quarterly.
Bryson, John. M. and Barbara C. Crosby. 2008. Falling into Cross-Sector Collaboration
Successfully. In Big Ideas in Collaborative Public Management, edited by Lisa B. Bingham
and Rosemary O’Leary, 55-78. New York, NY: M.E. Sharpe.
Bushouse, Brenda K. and Jessica E. Sowa. 2012. Producing Knowledge for Practice: Assessing
NVSQ 2000-2010. Nonprofit and Voluntary Sector Quarterly 41(3): 497-513.
Camic, Charles. 1995. Three Departments in Search of a Discipline: Localism and
Interdisciplinary Interaction in American Sociology, 1890-1940. Social Research 62(4):
1003-1033.
162
Camic, Charles and Yu Xie. 1994. The Statistical Turn in American Social Science: Columbia
University, 1890 to 1915. American Sociological Review 59(5): 773-805.
Caplow, Theodore. 1954. The Sociology of Work. New York, NY: McGraw-Hill.
Castilla, Emilio J. 2007. Dynamic Analysis in the Social Sciences. Burlington, MA : Academic
Press.
Chen, Bin and Elizabeth A. Graddy. 2010. The Effectiveness of Nonprofit Lead‐Organization
Networks for Social Service Delivery. Nonprofit Management and Leadership 20(4), 405‐
422.
Cho, Sungsook and David F. Gillespie. 2006. A Conceptual Model Exploring the Dynamics of
Government–Nonprofit Service Delivery. Nonprofit and Voluntary Sector Quarterly 35(3):
493-509.
Clemens, Elisabeth S. 2006. The Constitution of Citizens: Political Theories of Nonprofit
Organizations. In The Nonprofit Sector: A Research Handbook, 2
nd
ed., edited by Walter W.
Powell and Richard Steinberg, 207-220. New Haven, CT: Yale University Press.
Cohen, Steven and Tracie Abbott. 2000. Integrating Nonprofit Management Education into
Graduate Programs in Public Policy and Administration. Accessed May 30, 2012.
http://www.columbia.edu/~sc32/documents/Integrating%20Nonprofit%20Management.pdf.
Cohen, Robin. 2007. Creolization and Cultural Globalization: The Soft Sounds of Fugitive Power.
Globalizations 4(3): 369-373.
Cook, Jonathan B. 1988. Managing Nonprofits of Different Sizes. Educating Managers of
Nonprofit Organizations, edited by Michael O’Neill and Dennis R. Young, 101-116. New
York, NY: Praeger.
Corbin, John J. 1999. A Study of Factors Influencing the Growth of Nonprofits in Social Services.
Nonprofit and Voluntary Sector Quarterly 28(3): 296-314.
Dees, J. Gregory, Jed Emerson, and Peter Economy. 2001. Enterprising Nonprofits: A Toolkit for
Social Entrepreneurs. New York, NY: Wiley.
163
Denhardt, Kathryn G., Maria P. Aristigueta, Lynne Foote, Deborah Auger, Lauren Miltenberger,
Catherine Dodds, and Carey Addison. 2008. Forward Together Project: Achieving Better
Performance in Nonprofit-State Government Contracting for Human Services. Newark, DE:
The University of Delaware. Accessed October 9, 2012. http://soc.kuleuven.be/io/
performance/paper/WS3/WS3_Denhardt_Aristigueta_Foote.pdf.
Denhardt, Robert B. and Janet Vinzant Denhardt. 2000. The New Public Service: Serving Rather
Than Steering. Public Administration Review 60(6): 549-559.
Dicke, Lisa A. and J. Steven Ott. 1999. Public Agency Accountability in Human Services
Contracting. Public Productivity & Management Review 22(4): 502-516.
DiIulio, John J. Jr., Gerald Garvey, and Donald F. Kettl. 1993. Improving Government
Performance: An Owner’s Manual. Washington, DC: Brookings Institution Press.
DiMaggio, Paul J. 1988. Nonprofit Managers in Different Fields of Service: Managerial Tasks and
Management Training. In Educating Managers of Nonprofit Organizations, edited by
Michael O’Neill and Dennis R. Young, 51-69. New York, NY: Praeger.
DiMaggio, Paul J. and Walter W. Powell. 1983. The Iron Cage Revisited: Institutional
Isomorphism and Collective Rationality in Organizational Fields. American Sociological
Review 48(2): 147-160.
Eikenberry, Angela M. and Jodie Drapal Kluver. 2004. The Marketization of the Nonprofit Sector:
Civil Society at Risk? Public Administration Review 64(2): 132-140.
Ferris, James M. and Elizabeth A. Graddy. 1998. A Contractual Framework for New Public
Management Theory. International Public Management Journal 1(2): 225-240.
Fiss, Peer C. 2007. A Set-Theoretic Approach to Organizational Configurations. Academy of
Management Journal 32(4), 1180-1198.
_________. 2011. Building Better Causal Theories: A Fuzzy Set Approach to Typologies in
Organization Research. Academy of Management Journal 54(2): 393-420.
Fligstein, Neil. 2001. Social Skill and the Theory of Fields. Sociological Theory 19(2): 105-125.
164
Fosler, R. Scott. 2002. Working Better Together: How Government, Business, and Nonprofit
Organizations Can Achieve Public Purposes through Cross-Sector Collaboration,
Alliances, and Partnerships. Washington, DC: Independent Sector.
Frank, D. John, Ann Hironaka, and Evan Schofer. 2000. The Nation-State and the Natural
Environment over the Twentieth Century. American Sociological Review 65(1): 96–116.
Frickel, Scott. 2004. Building an Interdiscipline: Collective Action Framing and the Rise of
Genetic Toxicology. Social Problems 51(2): 269-287.
Frumkin, Peter. 2002. On Being Nonprofit: A Conceptual and Policy Primer. Cambridge, MA:
Harvard University Press.
Gazley, Beth. 2008. Beyond the Contract: The Scope and Nature of Informal Government–
Nonprofit Partnerships. Public Administration Review 68(1): 141-154.
Gazley, Beth. 2010. Linking Collaborative Capacity to Performance Measurement in
Government–Nonprofit Partnerships. Nonprofit and Voluntary Sector Quarterly 39(4):
653–673.
Gazley, Beth and Jeffrey L. Brudney. 2007. The Purpose (and Perils) of Government-Nonprofit
Partnership. Nonprofit and Voluntary Sector Quarterly 36(3): 389-415.
Giddens, Anthony. 1998. The Third Way: The Renewal of Social Democracy. Cambridge, UK:
Polity Press.
Gies, David L., J. Steven Ott, and Jay M. Shafritz. 1990. The Nonprofit Organization: Essential
Readings. Pacific Grove, CA: Brooks/Cole Publishing Company.
Graddy, Elizabeth A. 2009. Cross-Sectoral Governance and Performance in Service Delivery.
International Review of Public Administration 13(supplement1): 61-73.
Graddy, Elizabeth A. and Bin Chen. 2006. Influences on the Size and Scope of Networks for Social
Service Delivery. Journal of Public Administration Research and Theory 16(4): 533-552.
Graddy, Elizabeth A., James M. Ferris and Yu J. Sohn. 2011. The Evolution of Research on
Philanthropy and Nonprofit Organization as an Intellectual Field: An Analysis of Leading
165
Journals. University of Southern California, the Center on Philanthropy & Public Policy,
Research Paper, no. 35.
Greenwood, Royston and C. R. Hinings. 1996. Understanding Radical Organizational Change:
Bringing together the Old and the New Institutionalism. The Academy of Management
Review 21(4): 1022-1054.
Greenwood, Royston, Roy Suddaby and C. R. Hinings. 2002. Theorizing Change: The Role of
Professional Associations in the Transformation of Institutionalized Fields. The Academy of
Management Journal 45(1): 58-80.
Grønbjerg, Kirsten A. and Laurie Paarlberg. 2001. Community Variations in the Size and Scope
of the Nonprofit Sector: Theory and Preliminary Findings. Nonprofit and Voluntary Sector
Quarterly 30(4): 684-706.
Grønbjerg, Kirsten A. and Lester M. Salamon. 2012. Devolution, Marketization, and the Changing
Shape of Government-Nonprofit Relations. In The State of Nonprofit America, 2
nd
ed., edited
by Lester M. Salamon, 549-586. Washington, DC: Brookings Institution Press.
Groot, Tom and Tjerk Budding. 2008. New Public Management's Current Issues and Future
Prospects. Financial Accountability & Management 24(1): 1-13.
Gumport, Patricia J. and Stuart K. Snydman. 2002. The Formal Organization of Knowledge: An
Analysis of Academic Structure. Journal of Higher Education 73(3): 375-408.
Haas, Peter J. and Maynard G. Robinson. 1998. The Views of Nonprofit Executives on Educating
Nonprofit Managers. Nonprofit Management and Leadership 8(4): 349–362.
Hall, Peter D. 1992. Teaching and Research on Philanthropy, Voluntarism, and Nonprofit
Organizations: A Case Study of Academic Innovation. Teachers College Record 93(3): 403-
435.
____________. 1996. No One Best Way: Management Careers and Curricula in an Era of
Institutional Crisis. Nonprofit Management Education 1996 Conference (March), Berkeley,
CA, 14-16.
166
____________. 2010. Historical Perspectives on Nonprofit Organizations in the United States. In
The Jossey-Bass Handbook of Nonprofit Leadership and Management, 3
rd
ed., edited by
David O. Renz and Associates, 3-41. San Francisco, CA: Jossey-Bass.
Hall, Peter D., Michael O’Neill, Diane Vinokur-Kaplan, Dennis R. Young, and Frederick S. Lane.
2001. Panel Discussion: Where You Stand Depends on Where You Sit: The Implications of
Organizational Location for University-Based Programs in Nonprofit Management. Public
Performance and Management Review 25(1): 74-87.
Hall, Stuart. 2003. Creolization, Diaspora, and Hybridity in the Context of Globalization. In
Créolité and Creolization, edited by Okwui Enwezor, 183-198. Ostfildern-Ruit, DE: Hatje
Cantzpp.
Hannerz, Ulf. 1987. The World in Creolisation. Africa 57(4): 546-559.
Hansmann, Henry. 1987. Economic Theories of Nonprofit Organization. In The Nonprofit Sector:
A Research Handbook, edited by Walter W. Powell, 27-42. New Haven, CT: Yale University
Press.
Hefetz, Amir and Mildred E. Warner. 2011. Contracting or Public Delivery? The Importance of
Service, Market and Management Characteristics. Journal of Public Administration
Research and Theory 22(2): 289-317.
Heimovics, Richard D. and Robert D. Herman. 1989. The Salient Management Skills: A
Conceptual Framework for a Curriculum for Managers in Nonprofit Organizations. The
American Review of Public Administration 19(4): 295-312.
Hirsch, Paul M. and Michael Lounsbury. 1997. Ending the Family Quarrel Toward a
Reconciliation of “Old” and “New” Institutionalisms. American Behavioral Scientist 40(4):
406-418.
Hoefer, Richard. 2003. Administrative Skills and Degrees: The “Best Place” Debate Rages On.
Administration in Social Work 27(1): 25-45.
Hood, Christopher. 1991. A Public Management for All Seasons? Public Administration 69(1): 3-
19.
167
Hwang, Hokyu and Walter W. Powell. 2009. The Rationalization of Charity: The Influences of
Professionalism in the Nonprofit Sector. Administrative Science Quarterly 54(2): 268-298.
Internal Revenue Service (IRS). 2011. SOI Tax Stats: Charities and Other Tax-Exempt
Organizations (various years). Retrieved from http://www.irs.gov/uac/SOI-Tax-Stats-
Charities-and-Other-Tax-Exempt-Organizations-Statistics.
___________________________. 2012. Historical Table 16 (Expanded Version): Nonprofit
Charitable Organization and Domestic Private Foundation Information Returns, and Exempt
Organization Business Income Tax Returns—Selected Financial Data, 1985-2008. Retrieved
from http://www.irs.gov/uac/SOI-Tax-Stats---Historical-Table-16.
Jackson, Stephen K., Santiago Guerrero, and Susan Appe. 2014. The State of Nonprofit and
Philanthropic Studies Doctoral Education. Nonprofit and Voluntary Sector Quarterly 43(5):
795–811.
James, Estelle. 1987. The Nonprofit Sector in Comparative Perspective. In The Nonprofit Sector:
A Research Handbook, edited by Walter W. Powell, 397-415. New Haven, CT: Yale
University Press.
Jensen, Jason L. 2003. Policy Diffusion through Institutional Legitimation: State Lotteries.
Journal of Public Administration Research and Theory 13(4): 521-541.
Jepperson, Ronald L. 2002. Political Modernities: Disentangling Two Underlying Dimensions of
Institutional Differentiation. Sociological Theory 20(1): 61-85.
Kaboolian, Linda. 1998. The New Public Management: Challenging the Boundaries of the
Management vs. Administration Debate. Public Administration Review 58(3): 189-93.
Kapucu, Naim. 2006. Public-Nonprofit Partnerships for Collective Action in Dynamic Contexts
of Emergencies. Public Administration 84(1): 205-220.
Katz, Stanley N. 1999. Where Did the Serious Study of Philanthropy Come From, Anyway?
Nonprofit and Voluntary Sector Quarterly 28(1): 74-82.
168
Kettl, Donald F. 1997. The Global Revolution in Public Management: Driving Themes, Missing
Links. Journal of Policy Analysis and Management 16(3): 446-62.
_____________, ed. 2005. The Global Revolution in Public Management. Washington, DC:
Brookings Institution Press.
_____________. 2006. Managing Boundaries in American Administration: The Collaboration
Imperative. Public Administration Review 66(Special Issue):10-19.
Khurana, Rakesh. 2007. From Higher Aims to Hired Hands: The Social Transformation of
American Business Schools and the Unfulfilled Promise of Management as a Profession.
Princeton, NJ: Princeton University Press.
Knorr-Cetina, Karin. 1994. Primitive Classification and Postmodernity: Towards a Sociological
Notion of Fiction. Theory, Culture and Society 11(3): 1-22.
Kraatz, Matthew S. and Edward J. Zajac. 1996. Exploring the Limits of the New Institutionalism:
The Causes and Consequences of Illegitimate Organizational Change. American
Sociological Review 61(5): 812-836.
Krasner, Stephen D., ed. 1983. International Regimes. Ithaca, NY: Cornell University Press.
Kreutzer, Karin and Urs Jäger. 2011. Volunteering Versus Managerialism: Conflict Over
Organizational Identity in Voluntary Associations. Nonprofit and Voluntary Sector
Quarterly 40(4): 634–661.
Krueathep, Weerasak, Norma M. Riccucci, and Charas Suwanmala. 2008. Why Do Agencies
Work Together? The Determinants of Network Formation at the Subnational Level of
Government in Thailand. Journal of Public Administration Research and Theory 20(1): 157-
85.
Lamothe, Meeyoung and Scott Lamothe. 2012. To Trust or Not to Trust? What Matters in Local
Government-Vendor Relationships? Journal of Public Administration Research and Theory
22(4): 867-892.
169
Larson, R. Sam, and Sonia Barnes-Moorhead. 2001. How Centers Work: Building and Sustaining
Nonprofit Academic Centers. Battle Creek, MI: W.K. Kellogg Foundation. Accessed
November 8, 2014. http://files.eric.ed.gov/fulltext/ED459648.pdf.
Larson, R. Sam, Mark I. Wilson, and Donghun Chung. 2003. Curricular Content for Nonprofit
Management Programs: The Student Perspective. Journal of Public Affairs Education 9(3):
169-180.
Lecy, Jesse D. and David M. Van Slyke. 2013. Nonprofit Sector Growth and Density: Testing
Theories of Government Support. Journal of Public Administration Research and Theory
23(1): 189-214.
LeRoux, Kelly. 2009. Paternalistic or Participatory Governance? Examining Opportunities for
Client Participation in Nonprofit Social Service Organizations. Public Administration
Review 69(3): 504-517.
Light, Paul C. 2000. Making Nonprofits Work: A Report on the Tides of Nonprofit Management
Reform. Washington, DC: Aspen Institute/Brookings Institution Press.
___________. 2001. The New Public Service. Washington, DC: Brookings Institution Press.
Lounsbury, M. 2001. Institutional sources of practice variation: Staffing college and university
recycling programs. Administrative Science Quarterly 46(1): 29-56.
March, James G. 1994. A Primer on Decision Making: How Decisions Happen. New York, NY:
Free Press.
March, James G. and Johan P. Olsen. 1984. The New Institutionalism: Organization Factors in
Political Life. American Political Science Review 78(3): 734-749.
_______________________________. 1989. Rediscovering Institutions: The Organizational
Basis of Politics. New York, NY: Free Press.
Marx, Axel. 2006. Towards More Robust Model Specification in QCA: Results from A
Methodological Experiment. COMPASSS Working Paper Series 2006-43. Accessed January
30, 2015. http://www.compasss.org/wpseries/Marx2006.pdf.
170
Matsunaga, Yoshiho, and Naoto Yamauchi. 2004. Is the Government Failure Theory still Relevant?
A Panel Analysis Using US State Level Data. Annals of Public and Cooperative Economics
75(2): 227-263.
McDowell, George R. 2003. Engaged Universities: Lessons from the Land-Grant Universities and
Extension. Annals of the American Academy of Political and Social Science 585(1): 31-50.
McGuire, Michael. 2006. Collaborative Public Management: Assessing What We Know and How
We Know It. Public Administration Review 66(special issue): 33-43.
McKeever, Brice S. and Sarah L. Pettijohn. 2014. The Nonprofit Sector in Brief 2014: Public
Charities, Giving, and Volunteering. Washington, DC: The Urban Institute. Accessed
December 2, 2014. http://www.urban.org/publications/413277.html.
Meyer, John W. and Ronald L. Jepperson. 2000. The ‘Actors’ of Modern Society: The Cultural
Construction of Social Agency. Sociological Theory 18(1): 100-120.
Meyer, John W. and Brian Rowan. 1977. Institutionalized Organizations: Formal Structure as
Myth and Ceremony. American Journal of Sociology 83(2): 364-385.
Meyer, John W. and W. Richard Scott, eds. 1992. Organizational Environments: Ritual and
Rationality. Newbury Park, CA: Sage.
Mirabella, Roseanne M. 2001. A Symposium Introduction: Filling the Hollow State: Capacity-
Building within the Nonprofit Sector. Public Performance & Management Review 25(1): 8-
13.
_________________________. 2007. University-Based Educational Programs in Nonprofit
Management and Philanthropic Studies: A 10-Year Review and Projections of Future Trends.
Nonprofit and Voluntary Sector Quarterly 36(4): 11S-27S.
_________________________. 2014. Nonprofit Management Education and Philanthropic
Studies. Unpublished Manuscript, Seton Hall University, South Orange, NJ.
171
Mirabella, Roseanne M. and Naomi B. Wish. 2000. The “Best Place” Debate: A Comparison of
Graduate Education Programs for Nonprofit Managers. Public Administration Review 60(3):
219-29.
____________________________________. 2001. University-Based Educational Programs in
the Management of Nonprofit Organizations: An Updated Census of U.S. Public
Performance & Management Review 25(1): 30-41.
Moon, Hyeyoung and Christine Min Wotipka. 2006. The Worldwide Diffusion of Business
Education, 1881–1999: Historical Trajectory and Mechanisms of Expansion. In
Globalization and Organization: World Society and Organizational Change, edited by Gili
Drori, John W. Meyer, and Hokyu Hwang, 121–136. Oxford, UK: Oxford University Press.
National Center for Charitable Statistics (NCCS). 2012. Business Master Files (BMF) [Data file].
Retrieved from http://nccs.urban.org/nccs/database/.
O’Leary, Rosemary and Lisa Blomgren Bingham, ed. 2009. Collaborative Public Manager: New
Ideas for the Twenty-First Century. Washington, DC: Georgetown University Press.
O’Neill, Michael. 1989. The Third America: The Emergence of the Nonprofit Sector in the United
States. San Francisco, CA: Jossey-Bass Publishers.
______________. 1998. Nonprofit Management Education: History, Current Issues, and the
Future. In Nonprofit Management Education: U.S. and World Perspectives, edited by
Michael O’Neill and Kathleen Fletcher, 1-12. Westport, CT: Praeger.
______________. 2005. Developmental Contexts of Nonprofit Management Education. Nonprofit
Management and Leadership 16(1): 5–17.
______________. 2007. The Future of Nonprofit Management Education. Nonprofit and
Voluntary Sector Quarterly Supplement 36(4), 169S-176S.
O’Neill, Michael and Kathleen Fletcher. 1998. Nonprofit Management Education: U.S. and World
Perspectives. Westport, CT: Praeger.
172
O’Neill, Michael and Dennis R. Young. 1988. Educating Managers of Nonprofit Organizations.
Westport, CT: Praeger.
Oliver, Christine. 1991. Strategic Responses to Institutional Processes. Academy of Management
Review 16(1): 145-179.
Olzak, Susan and Nicole Kangas. 2008. Ethnic, Women’s, and African American Studies Majors
in U.S. Institutions of Higher Education. Sociology of Education 81(April): 163-188.
Osborne, David and Ted Gaebler. 1992. Reinventing Government: How the Entrepreneurial Spirit
is Transforming the Public Sector. Reading, MA: Addison-Wesley.
Osborne, Stephen P. 2006. The New Public Governance? Public Management Review 8(3): 377-
387.
________________., ed. 2010. The New Public Governance: Emerging Perspectives on the
Theory and Practice of Public Governance. London, UK: Routledge.
Osborne, Stephen P., Zoe Radnor, and Greta Nasi. 2012. A New Theory for Public Service
Management? Toward a (Public) Service-Dominant Approach. American Review of Public
Administration 43(2): 135-158.
Ostrander, Susan A. 2004. Democracy, Civic Participation, and the University: A Comparative
Study of Civic Engagement on Five Campuses. Nonprofit and Voluntary Sector Quarterly
33(1): 74-93.
Ott, J. Steven, ed. 2001. The Nature of the Nonprofit Sector. Boulder, CO: Westview Press.
Paulsen, Neil. 2006. New Public Management, Innovation, and the Non-Profit Domain: New
Forms of Organizing and Professional Identity. In Organizing Innovation: New
Approaches to Cultural Change and Intervention in Public Sector Organizations, edited
by Marcel Veenswijk, 15-28. Amsterdam: IOS Press.
Pearce, Jone L. 1993. Volunteers: The Organizational Behavior of Unpaid Workers. London, UK:
Routledge.
173
Pettijohn, Sarah L. 2013. The Nonprofit Sector in Brief: Public Charities, Giving, and
Volunteering, 2013. Washington, DC: The Urban Institute. Accessed Dec 2, 2014.
http://www.urban.org/sites/default/files/alfresco/publication-pdfs/412923-The-Nonprofit-
Sector-in-Brief-Public-Charities-Giving-and-Volunteering-.PDF.
Pfeffer, Jeffrey. 1982. Organizations and Organization Theory. Marshfield, MA: Pitman.
Pfeffer, Jeffrey and Gerald R. Salancik. 1978. The External Control of Organizations: A Resource
Dependence Perspective. New York, NY: Harper and Row.
Posner, Paul L. 2009. A Public Administration Education for the Third-Party Governance Era:
Reclaiming Leadership of the Field. In Collaborative Public Manager: New Ideas for the
Twenty-First Century, edited by Rosemary O’Leary and Lisa Blomgren Bingham, 233-253.
Washington, DC: Georgetown University Press.
Powell, Walter W. and Paul J. DiMaggio, eds. 1991. The New Institutionalism in Organizational
Analysis. Chicago, IL: University of Chicago Press.
Radin, Margaret J. 1996. Contested Commodities. Cambridge, MA: Harvard University Press.
Ragin, Charles C. 1987. The Comparative Method: Moving Beyond Qualitative and Quantitative
Strategies, With a New Introduction. Berkeley, CA: University of California Press.
____________. 2000. Fuzzy Set Social Science. Chicago, IL: University of Chicago Press.
____________. 2006. Set Relations in Social Research: Evaluating their Consistency and
Coverage. Political Analysis 14(3): 291-310.
____________. 2008. Redesigning Social Inquiry: Fuzzy Sets and Beyond. Chicago, IL:
University of Chicago Press.
Ragin, Charles C. and Sean Davey. 2014. Fuzzy-Set/Qualitative Comparative Analysis 2.5. Irvine,
CA: Department of Sociology, University of California.
Ragin, Charles C. and Peer C. Fiss. 2008. Net Effects versus Configurations: An Empirical
Demonstration. In Redesigning Social Inquiry: Fuzzy Sets and Beyond, by Charles C. Ragin,
190-212. Chicago, IL: University of Chicago Press.
174
Ramirez, Francisco O., Yasemin Soysal, and Suzanne Shanahan. 1997. The Changing Logic of
Political Citizenship: Cross-National Acquisition of Women's Suffrage Rights, 1890 to 1990.
American Sociological Review 62(5): 735-745.
Rihoux, Benoit and Bojana Lobe. 2009. The Case for Qualitative Comparative Analysis (QCA):
Adding Leverage for Thick Cross-Case Comparison. In The Sage Handbook of Case-Based
Methods, edited by David Byrne and Charles C. Ragin, 222-242. Thousand Oaks, CA: Sage.
Rihoux, Benoî t and Charles C. Ragin, eds. 2009. Configurational Comparative Methods:
Qualitative Comparative Analysis (QCA) and Related Techniques. Thousand Oaks, CA:
Sage.
Rimer, Edward. 1987. Reconceptualizing the Conflict with MPA, MBA, and MPH Programs.
Administration in Social Work 11(2): 45-55.
Robinson, Karen Jeong. 2011. The Rise of Choice in the U.S. University and College: 1910–2005.
Sociological Forum 26(3): 601–622.
Robbins, Kevin C. 2006. The Nonprofit Sector in Historical Perspective: Traditions of
Philanthropy in the West. In The Nonprofit Sector: A Research Handbook, 2
nd
ed., edited by
Walter W. Powell and Richard Steinberg, 13-31. New Haven, CT: Yale University Press.
Roeger, Katie L., Amy S. Blackwood, and Sarah L. Pettijohn. 2012. The Nonprofit Almanac 2012.
Washington, DC: Urban Institute Press.
Rojas, Fabio. 2006. Social Movement Tactics, Organizational Change and the Spread of African-
American Studies. Social Forces 84(4): 2147-2166.
Ruef, Martin and W. Richard Scott. 1998. A Multidimensional Model of Organizational
Legitimacy: Hospital Survival in Changing Institutional Environments. Administrative
Science Quarterly 43(4): 877-904.
Sahlin-Andersson, Kerstin and Lars Engwall, eds. 2002. The Expansion of Management
Knowledge: Carriers, Flows, and Sources. Stanford, CA: Stanford Business Books.
175
Salamon, Lester M. 1987. Partners in Public Service: The Scope and Theory of Government-
Nonprofit Relations. In The Nonprofit Sector: A Research Handbook, edited by Walter W.
Powell, 99-117. New Haven, CT: Yale University Press.
________________. 1994a. The Rise of the Nonprofit Sector. Foreign Affairs 73(4): 109-122.
________________, ed. 1994b. Global Civil Society: Dimensions of the Nonprofit Sector.
Bloomfield, CT: Kumarian Press.
________________. 1995. Partners in Public Service: Government-Nonprofit Relations in the
Modern Welfare State. Baltimore, MD: The Johns Hopkins University Press.
________________. 1998. Nonprofit Management Education: A Field Whose Time Has Passed?
In Nonprofit Management Education: U.S. and World Perspectives, edited by Michael
O’Neill and Kathleen Fletcher, 137-145. New York, NY: Praeger.
________________. 2002a. The State of Nonprofit America. Washington, DC: Brookings
Institution Press.
________________. 2002b. The Tools of Government: A Guide to the New Governance. New
York, NY: Oxford University Press.
________________. 2003. The Resilient Sector: The State of Nonprofit America. Washington, DC:
Brookings Institution Press.
________________, ed. 2012. The State of Nonprofit America, 2
nd
ed. Washington, DC:
Brookings Institution Press.
Salamon, Lester M. and Susan LQ Flaherty. 1996. Nonprofit Law: Ten Issues in Search of
Resolution. Johns Hopkins Institute for Policy Studies, the Center for Civil Society Studies,
Working Papers of the Johns Hopkins Comparative Nonprofit Sector Project, no. 20.
Salamon, Lester M., S. Wojciech Sokolowski, and Helmut K. Anheier. 2000. Social Origins of
Civil Society: An Overview. Working Papers of the Johns Hopkins Comparative Nonprofit
Sector Project 38. Baltimore, MD: The Johns Hopkins Center for Civil Society Studies.
176
Scott, W. Richard. 2008. Institutions and Organizations: Ideas and Interests, 3
rd
ed. Thousand
Oaks, CA: Sage.
Scott, W. Richard and John W. Meyer. 1994. Institutional Environments and Organizations:
Structural Complexity and Individualism. Thousand Oaks, CA: Sage.
Scrivner, Gary N. 2001. A brief History of Tax Policy Changes Affecting Charitable Organizations.
In The Nature of the Nonprofit Sector, edited by J. Steven Ott, 126-142. Boulder, CO:
Westview Press.
Shaw, Mary M. 2003. Successful Collaboration between the Nonprofit and Public Sectors.
Nonprofit management and leadership 14(1): 107-120.
Shier, Micheal L. and Femida Handy. 2014. Research Trends in Nonprofit Graduate Studies A
Growing Interdisciplinary Field. Nonprofit and Voluntary Sector Quarterly 43(5): 812-831.
Small, Mario L. 1999. Departmental Conditions and the Emergence of New Disciplines: Two
Cases in the Legitimation of African-American Studies. Theory and Society 28(5): 659-707.
Smith, David Horton. 1999. Researching Volunteer Associations and Other Nonprofits: An
Emergent Interdisciplinary Field and Possible New Discipline. The American Sociologist
30(4): 5-35.
Smith, Rick. 1997. Building the Nonprofit Sector Knowledge Base: Can Academic Centers and
Management Support Organizations Come Together? Nonprofit Management and
Leadership 8(1): 89-97.
Smith, Steven R. 2010. Nonprofits and Public Administration Reconciling Performance
Management and Citizen Engagement. The American Review of Public Administration
40(2): 129-152.
_______________. 2012. Social Services. In The State of Nonprofit America, 2nd ed., edited by
Lester M. Salamon, 192-228. Washington, DC: Brookings Institution Press.
Smith, Steven R. and Judith Smyth. 2010. The Governance of Contracting Relationships: “Killing
the Golden Goose”. In The New Public Governance?: Emerging Perspectives on the Theory
177
and Practice of Public Governance, edited by Stephen P. Osborne, 270-300. New York, NY:
Routledge.
Smith, Steven R. and Michael Lipsky. 1993. Nonprofits for Hire: The Welfare State in the Age of
Contracting. Cambridge, MA: Harvard University Press.
Soule, Sarah A. 1997. The Student Divestment Movement in the United States and Tactical
Diffusion: The Shantytown Protest. Social Forces 75(3): 855-882.
Sowa, Jessica E. 2008. Implementing Interagency Collaborations: Exploring Variation in
Collaborative Ventures in Human Service Organizations. Administration & Society 40(3):
298-323.
Steinberg, Richard. 2006. Economic Theories of Nonprofit Organization. In The Nonprofit Sector:
A Research Handbook, 2
nd
ed., edited by Walter W. Powell and Richard Steinberg, 117-139.
New Haven, CT: Yale University Press.
Stewart, Charles, ed. 2007. Creolization: History, Ethnography, Theory. Walnut Creek, CA: Left
Coast Press.
Stone, Melissa M., Barbara Bigelow, and William Crittenden. 1999. Research on Strategic
Management in Nonprofit Organizations: Synthesis, Analysis, and Future Directions.
Administration & Society 31(3): 378-423.
Strang, David and Sarah Soule. 1998. Diffusion in Organizations and Social Movements: From
Hybrid Corn to Poison Pills. Annual Review of Sociology, 24(1): 265-290.
Suárez, David F. 2009. Nonprofit Advocacy and Civic Engagement on the Internet. Administration
& Society 41(3): 267-289.
_______________. 2011. Collaboration and Professionalization: The Contours of Public Sector
Funding for Nonprofit Organizations. Journal of Public Administration Research and Theory
21(2): 307-326.
178
Suárez, David F., Francisco O. Ramirez and Jeong-Woo Koo. 2009. UNESCO and the Associated
Schools Project: Symbolic Affirmation of World Community, International Understanding,
and Human Rights. Sociology of Education 82(3): 197-216.
Terry, Larry D. 1998. Administrative Leadership, Neo-Managerialism, and the Public
Management Movement. Public Administration Review 58(3): 194-200.
Thompson, James D. 1967. Organizations in Action. New York, NY: McGraw Hill.
Thornton, Patricia H. and William Ocasio. 1999. Institutional Logics and the Historical
Contingency of Power in Organizations: Executive Succession in the Higher Education
Publishing Industry, 1958-1990. American Journal of Sociology 105: 801-843.
Tolbert, Pamela S. 1985. Resource dependence and institutional environments: Sources of
administrative structure in institutions of higher education. Administrative Science Quarterly
30: 1-13.
Tolbert, Pamela S. and Lynne G. Zucker. 1983. Institutional Sources of Change in the Formal
Structure of Organizations: The Diffusion of Civil Service Reform, 1880-1935.
Administrative Science Quarterly 28(1): 22-39.
__________________________________. 1996. Institutionalization of Institutional Theory. In
Handbook of Organization Studies, edited by Stewart R. Clegg, Cynthia Hardy, Thomas
Lawrence, and Walter R. Nord, 175-190. Thousand Oaks, CA: Sage.
Tuma, Nancy Brandon and Michael T. Hannan. 1984. Social Dynamics: Models and Methods.
Orlando, FL: Academic Press.
U.S. Census Bureau. 2013. Statistical Abstract. Retrieved from http://www.census.gov/
compendia/statab/past_years.ht.
Van Slyke, David M. 2003. The Mythology of Privatization in Contracting for Social Services.
Public Administration Review 63(3): 296-315.
179
_________________. 2007. Agents or Stewards: Using Theory to Understand the Government-
Nonprofit Social Service Contracting Relationship. Journal of Public Administration
Research and Theory 17(2): 157-187.
Van Slyke, David M. and Christine H. Roch. 2004. What Do They Know, and Whom Do They
Hold Accountable? Citizens in the Government-Nonprofit Contracting Relationship.
Journal of Public Administration Research and Theory 14(2): 191-209.
Walker, Jack L. 1969. The Diffusion of Innovations among the American States. American
Political Science Review 63(3): 880-899.
Weisbrod, Burton A. 1975. Toward a Theory of the Voluntary Non-Profit Sector in a Three-Sector
Economy. In Altruism, Morality, and Economic theory, edited by Edmund S. Phelps, 171-
195. New York, NY: Russell Sage Foundation.
Wejnert, Barbara. 2002. Reviewed Integrating Models of Diffusion of Innovations: A Conceptual
Framework. Annual Review of Sociology 28: 297-326.
Wilensky, Harold L. 1964. The Professionalization of Everyone? American Journal of Sociology
70(2): 137-158.
Wilson, Mark I. and R. Sam Larson. 2002. Nonprofit Management Students: Who They Are and
Why They Enroll. Nonprofit and Voluntary Sector Quarterly 31(2): 259-270.
Wish, Naomi B. 1991. University- and College-Based Nonprofit Management Programs in the
United States. Washington, DC: National Association of Schools of Public Affairs and
Administration.
_____________. 1993. Graduate Programs in Nonprofit Management: An Update. Journal of the
National Association of Graduate Admissions Professionals 5(2): 15-20.
Wish, Naomi B. and Roseanne M. Mirabella. 1998. Nonprofit Management Education: Current
Offerings and Practices in University-Based Programs. In Nonprofit Management Education:
U.S. and World Perspectives, edited by Michael O’Neill and Kathleen Fletcher, 13-22.
Westport, CT: Praeger.
180
Witesman, Eva M. and Sergio Fernandez. 2013. Government Contracts with Private Organizations:
Are There Differences between Nonprofits and For-profits? Nonprofit and Voluntary Sector
Quarterly 42(4): 689–715.
Wood, Christine V. 2012. Knowledge Practices, Institutional Strategies, and External Influences
in the Making of an Interdisciplinary Field: Insights From the Case of Women’s and Gender
Studies. American Behavioral Scientist 56(10): 1301–1325.
Worth, Michael J. 2011. Nonprofit Management: Principles and Practice, 2
nd
ed. Thousand Oaks,
CA: Sage Publications, Inc.
Yamaguchi, Kazuo. 1991. Event History Analysis. Newbury Park, CA: Sage.
Young, Dennis R. 1988. Nonprofit Studies in The University. Nonprofit World 6(5): 35-36.
______________. 1998. Games Universities Play: An Analysis of the Institutional Contexts of
Centers for Nonprofit Study. In Nonprofit Management Education: U.S. and World
Perspectives, edited by Michael O’Neill and Kathleen Fletcher, 119-136. Westport, CT:
Praeger.
______________. 1999. Nonprofit Management Studies in the United States: Current
Developments and Future Prospects. Journal of Public Affairs Education 5(1): 13-23.
Young, Dennis R., Lester M. Salamon, and Mary Clark Grinsfelder. 2012. Commercialization,
Social Ventures, and For-Profit Competition. In The State of Nonprofit America, 2
nd
ed.,
edited by Lester M. Salamon, 521-548. Washington, DC: Brookings Institution Press.
181
APPENDIX A.
Comparison between Original and Imputed Data (Chapter 3)
Table A.1. Descriptive Statistics for Imputed Covariates: Original and Imputed Values
Variable Years available Mean SD Min Max N. Cases
Nonprofit org. 1976-2010 (Original) 645934 310118 259523 1280739 50785
1971-2011 (Imputed) 589386 317415 259523 1280739 59491
Gov. grant 1982-2010 (Original) 71166 39454 19181 155125 39177
1971-2011 (Imputed) 53674 40204 19181 155125 59491
Urbanization 1971-2009 (Original) 72.50 13.85 32.2 94.4 57600
1971-2011 (Imputed) 72.73 13.93 32.2 94.4 59491
Welfare exp. 1971-2008 (Original) 0.20 0.06 0.04 0.39 56578
1971-2011 (Imputed) 0.20 0.06 0.04 0.39 59491
Citizen liberalism 1971-2008 (Original) 49.81 15.33 7.04 95.97 56160
1971-2011 (Imputed) 50.73 15.78 7.04 95.97 59491
Gov. liberalism 1971-2008 (Original) 53.40 11.66 23.64 76.44 56160
1971-2011 (Imputed) 53.48 11.83 23.64 76.44 59491
PA/SW degree 1971-2010 (Original) 2.17 4.95 0 100 56577
1971-2011 (Imputed) 2.08 4.95 0 100 59491
BUS degree 1971-2010 (Original) 21.19 16.83 0 100 56577
1971-2011 (Imputed) 20.27 17.05 0 100 59491
Black/Hispanic 1971-2010 (Original) 9.53 18.87 0 100 58096
1971-2011 (Imputed) 12.57 21.14 0 100 59491
Female 1971-2010 (Original) 54.30 19.04 0 100 58096
1971-2011 (Imputed) 53.05 20.51 0 100 59491
Univ. revenue 1971-2009 (Original) 138.78 349.58 0 10909 55927
1971-2011 (Imputed) 136.95 352.71 0 10909 59491
Master’s degree 1971-2010 (Original) 15.71 17.57 0 100 58096
1971-2011 (Imputed) 15.66 17.80 0 100 59491
182
APPENDIX B.
Piecewise Constant Models with Period-Specific Effects
Piecewise exponential models with period-specific effects allow the baseline transition rates
and effects of covariates to vary nonmonotonically between the predetermined time intervals. In
other words, in the piecewise exponential models with period-specific effects, the transition rate
from a given origin state j (not-adopting a program) to a destination state k (adopting a program)
is specified as:
rjk (t) = exp {α
𝑝 (𝑗𝑘 )
+ β
𝑝 (𝑗𝑘 )
A
(𝑗𝑘 )
} if t ∈ 𝐼 𝑝
where Ip denotes a given pth period in which different transition rates are estimated, αp
(jk)
is a
constant coefficient associated with the pth time period, βp
(jk)
is an associated vector of coefficients
for the pth time period, and A
(jk)
is a row vector of covariates (Blossfeld & Rohwer 1995; Blossfeld
et al. 2007; Castilla 2007).
Based on an exploratory analysis and substantive historical consideration, I split the time
axis into two historical periods: 1971-2001 and 2002-2011. To select the exact time periods, I
started with theory. In theory, a split in 2002 makes sense because it is expected to capture mimetic
governance changes resulting from such as the Sarbanes-Oxley Act of 2002, a prominent
regulatory change that imposed restrictions and requirements on for-profit organizations
(Grønbjerg & Salamon 2012). I also conducted a graphical analysis and sensitivity analysis,
examining variations in the transition rate and the sensitivity of specific cutoff periods. These
analyses did not provide support for choosing a year other than 2002, the year I chose for
substantive reasons.
183
Table B.1. Piecewise Constant Models with Period-Specific Effects: Degree Program
Variable Model 1 Model 2 Model 3 Model 4
1971-
2001
2002-
2011
1971-
2001
2002-
2011
1971-
2001
2002-
2011
1971-
2001
2002-
2011
Structuration 1.95***
(0.53)
0.79**
(0.33)
1.92***
(0.61)
0.81**
(0.36)
Urbanization -0.01
(0.02)
-0.01
(0.01)
0.01
(0.02)
-0.01
(0.01)
Welfare expenditure 0.10**
(0.05)
0.03
(0.04)
0.00
(0.06)
0.04
(0.04)
Citizen liberalism -0.05**
(0.02)
0.02
(0.01)
-0.05**
(0.02)
0.01
(0.02)
Government liberalism 0.01
(0.02)
-0.01
(0.01)
0.03
(0.02)
-0.01
(0.01)
PA/SW degree 0.03
(0.04)
0.04**
(0.02)
0.04
(0.03)
0.05**
(0.02)
BUS degree 0.03**
(0.01)
0.02*
(0.01)
0.02
(0.01)
0.02**
(0.01)
Black/Hispanic -0.01
(0.02)
-0.01
(0.01)
-0.02
(0.02)
-0.01
(0.01)
Female 0.03*
(0.02)
0.02
(0.02)
0.02
(0.02)
0.02
(0.02)
Univ. revenue (log) 0.64**
(0.26)
0.44**
(0.18)
0.57**
(0.25)
0.34**
(0.17)
0.29
(0.25)
0.34**
(0.17)
0.50*
(0.27)
0.44**
(0.18)
Master’s degree (log) 0.24
(0.21)
0.23
(0.16)
0.34*
(0.21)
0.32**
(0.15)
0.31
(0.20)
0.30**
(0.15)
0.24
(0.21)
0.22
(0.16)
Public -1.00*
(0.55)
-0.39
(0.37)
-1.35**
(0.57)
-0.43
(0.38)
-0.91*
(0.54)
-0.50
(0.36)
-1.08*
(0.58)
-0.36
(0.39)
Doctoral/Research -0.31
(0.89)
-0.57
(0.70)
-0.76
(0.88)
-0.58
(0.70)
-0.02
(0.91)
-0.58
(0.69)
-0.10
(0.93)
-0.52
(0.70)
Period Effect -13.91***
(1.86)
-10.45***
(1.56)
-11.09***
(1.96)
-9.17***
(1.54)
-9.54***
(1.22)
-10.02***
(1.11)
-11.61***
(2.58)
-11.85***
(2.04)
LRχ² 2245.04 2255.16 2194.95 2093.88
Log-likelihood 27.75 26.65 31.76 41.30
Number of adoptions 58 58 58 58
Number of universities 1,451 1,451 1,451 1,451
Yearly spells 59,065 59,065 59,065 59,065
* p<0.1; ** p<0.05; *** p<0.01. Standard errors are in parentheses.
Note: 19 adoptions in the first period (1971-2001) and 39 adoptions in the second period (2002-2011).
184
Table B.2. Piecewise Constant Models with Period-Specific Effects: Research Center
Variable Model 1 Model 2 Model 3 Model 4
1971-
2001
2002-
2011
1971-
2001
2002-
2011
1971-
2001
2002-
2011
1971-
2001
2002-
2011
Structuration 1.77***
(0.38)
-0.01
(0.34)
1.82***
(0.45)
-0.28
(0.38)
Urbanization 0.00
(0.02)
0.01
(0.02)
0.01
(0.02)
0.02
(0.02)
Welfare expenditure 0.01
(0.04)
0.05
(0.05)
-0.06
(0.04)
0.05
(0.04)
Citizen liberalism -0.00
(0.02)
0.02
(0.02)
-0.00
(0.02)
0.02
(0.02)
Government liberalism -0.03**
(0.02)
0.02
(0.02)
-0.01
(0.02)
0.02
(0.02)
PA/SW degree 0.02
(0.04)
0.05*
(0.03)
0.03
(0.04)
0.04
(0.03)
BUS degree 0.04***
(0.01)
0.01
(0.02)
0.04***
(0.01)
0.01
(0.01)
Black/Hispanic -0.01
(0.02)
-0.02
(0.02)
-0.03
(0.02)
-0.02
(0.02)
Female 0.04**
(0.02)
0.01
(0.02)
0.02
(0.02)
0.01
(0.02)
Univ. revenue (log) 1.04***
(0.21)
0.80***
(0.20)
0.84***
(0.21)
0.69***
(0.20)
0.62***
(0.20)
0.77***
(0.20)
0.81***
(0.22)
0.72***
(0.21)
Master’s degree (log) 0.29
(0.20)
0.37
(0.26)
0.39*
(0.20)
0.4
(0.24)
0.43**
(0.20)
0.44*
(0.25)
0.33
(0.20)
0.34
(0.25)
Public -0.31
(0.36)
0.21
(0.38)
-0.5
(0.37)
0.5
(0.39)
-0.31
(0.34)
0.15
(0.37)
-0.3
(0.39)
0.62
(0.41)
Doctoral/Research 0.34
(0.56)
-0.45
(0.59)
-0.03
(0.53)
-0.36
(0.59)
0.53
(0.57)
-0.49
(0.59)
0.62
(0.57)
-0.35
(0.60)
Period Effect -17.08***
(1.76)
-13.17***
(1.95)
-11.90***
(1.61)
-16.30***
(2.09)
-11.73***
(1.19)
-12.41***
(1.46)
-13.30***
(2.22)
-16.91***
(2.51)
LRχ² 2230.62 2265.97 2223.35 2086.31
Log-likelihood 70.36 67.66 71.73 85.7
Number of adoptions 68 68 68 68
Number of universities 1,451 1,451 1,451 1,451
Yearly spells 58,769 58,769 58,769 58,769
* p<0.1; ** p<0.05; *** p<0.01. Standard errors are in parentheses.
Note: 36 adoptions in the first period (1971-2001) and 32 adoptions in the second period (2002-2011).
185
Table B.3. Piecewise Constant Models with Period-Specific Effects: Any Type of Program
Variable Model 1 Model 2 Model 3 Model 4
1971-
2001
2002-
2011
1971-
2001
2002-
2011
1971-
2001
2002-
2011
1971-
2001
2002-
2011
Structuration 1.42***
(0.18)
-0.25
(0.16)
1.21***
(0.22)
-0.22
(0.18)
Urbanization 0.00
(0.01)
0.01
(0.01)
0.01
(0.01)
0.01
(0.01)
Welfare expenditure 0.06***
(0.02)
-0.02
(0.02)
0.01
(0.02)
-0.02
(0.02)
Citizen liberalism 0.00
(0.01)
0.00
(0.01)
0.00
(0.01)
0.00
(0.01)
Government liberalism -0.02***
(0.01)
-0.00
(0.01)
-0.00
(0.01)
-0.01
(0.01)
PA/SW degree 0.03*
(0.02)
0.03**
(0.01)
0.03**
(0.01)
0.03**
(0.01)
BUS degree 0.03***
(0.01)
0.01**
(0.01)
0.02***
(0.01)
0.01**
(0.01)
Black/Hispanic -0.00
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
Female 0.03***
(0.01)
0.01
(0.01)
0.01**
(0.01)
0.01
(0.01)
Univ. revenue (log) 0.79***
(0.10)
0.57***
(0.10)
0.61***
(0.10)
0.52***
(0.10)
0.53***
(0.09)
0.54***
(0.09)
0.58***
(0.10)
0.56***
(0.10)
Master’s degree (log) 0.39***
(0.10)
0.32***
(0.10)
0.47***
(0.10)
0.37***
(0.10
0.48***
(0.10
0.38***
(0.10
0.41***
(0.10
0.32***
(0.10
Public -0.07
(0.18)
0.52***
(0.20)
0.00
(0.18)
0.43**
(0.20)
-0.01
(0.17)
0.42**
(0.19)
0.17
(0.19)
0.53***
(0.20)
Doctoral/Research -0.06
(0.28)
-0.49
(0.34)
-0.25
(0.27)
-0.56
(0.34)
0.05
(0.28)
-0.59*
(0.34)
0.21
(0.29)
-0.5
(0.34)
Period Effect -13.65***
(0.79)
-10.20***
(0.88)
-10.84***
(0.78)
-9.02***
(0.88)
-9.83***
(0.55)
-8.71***
(0.62)
-11.84***
(0.99)
-9.91***
(1.12)
LRχ² 5941.73 6026.78 5869.76 5741.16
Log-likelihood 599.9 589.8 609.82 626.69
Number of adoptions 281 281 281 281
Number of universities 1,446 1,446 1,446 1,446
Yearly spells 56,037 56,037 56,037 56,037
* p<0.1; ** p<0.05; *** p<0.01. Standard errors are in parentheses.
Note: 146 adoptions in the first period (1971-2001) and 135 adoptions in the second period (2002-2011).
186
APPENDIX C.
Expanded Exponential (Transition Rate) Models for Structuration Variables
Table C.1. Models for Each Component of the Factored Variable: Degree Program
Variable 1 2 3 4 5 6
Nonprofit org. (log) 2.32***
(0.45)
Gov. grant (log) 1.58***
(0.31)
State association 0.08***
(0.02)
Univ. program (log) 1.10***
(0.23)
Dissertation (log) 1.09***
(0.24)
News article (log) 0.70***
(0.13)
Urbanization -0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.00
(0.01)
Welfare expenditure 0.01
(0.03)
0.01
(0.03)
0.01
(0.03)
0.01
(0.03)
0.01
(0.03)
0.01
(0.03)
Citizen liberalism -0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
Government liberalism 0.01
(0.01)
0.01
(0.01)
0.01
(0.01)
0.01
(0.01)
0.01
(0.01)
0.01
(0.01)
PA/SW degree 0.04***
(0.02)
0.04***
(0.02)
0.04***
(0.02)
0.04**
(0.02)
0.04**
(0.02)
0.04***
(0.02)
BUS degree 0.02**
(0.01)
0.02**
(0.01)
0.02**
(0.01)
0.02**
(0.01)
0.02**
(0.01)
0.02**
(0.01)
Black/Hispanic -0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
Female 0.02
(0.01)
0.02
(0.01)
0.02
(0.01)
0.01
(0.01)
0.01
(0.01)
0.02
(0.01)
Univ. revenue (log) 0.45***
(0.15)
0.46***
(0.15)
0.45***
(0.15)
0.44***
(0.15)
0.45***
(0.15)
0.46***
(0.15)
Master’s degree (log) 0.23*
(0.12)
0.23*
(0.12)
0.23*
(0.12)
0.24*
(0.12)
0.24*
(0.12)
0.23*
(0.12)
Public -0.60*
(0.33)
-0.60*
(0.33)
-0.60*
(0.33)
-0.59*
(0.32)
-0.60*
(0.32)
-0.60*
(0.33)
Doctoral/Research -0.36
(0.56)
-0.36
(0.56)
-0.36
(0.56)
-0.35
(0.56)
-0.37
(0.56)
-0.36
(0.56)
Constant -44.39***
(6.47)
-40.91***
(5.78)
-13.14***
(1.50)
-16.76***
(1.84)
-17.94***
(2.00)
-17.14***
(1.74)
LRχ² 110.85 110.34 109.35 110.96 109.73 110.53
Log-likelihood 36.68 36.42 35.93 36.73 36.12 36.52
* p<0.1; ** p<0.05; *** p<0.01. Standard errors are in parentheses. Total number of adoptions=58; total
number of universities=1,451; total number of yearly spells=59,065.
187
Table C.2. Models for Each Component of the Factored Variable: Research Center
Variable 1 2 3 4 5 6
Nonprofit org. (log) 1.17***
(0.35)
Gov. grant (log) 0.79***
(0.24)
State association 0.04***
(0.01)
Univ. program (log) 0.55***
(0.16)
Dissertation (log) 0.48***
(0.15)
News article (log) 0.35***
(0.11)
Urbanization 0.02
(0.01)
0.02
(0.01)
0.02
(0.01)
0.02
(0.01)
0.02
(0.01)
0.02
(0.01)
Welfare expenditure -0.01
(0.03)
-0.01
(0.03)
-0.01
(0.03)
-0.01
(0.03)
-0.01
(0.03)
-0.01
(0.03)
Citizen liberalism 0.00
(0.01)
0.00
(0.01)
0.00
(0.01)
0.00
(0.01)
0.00
(0.01)
0.00
(0.01)
Government liberalism 0.00
(0.01)
0.00
(0.01)
0.00
(0.01)
0.00
(0.01)
0.00
(0.01)
0.00
(0.01)
PA/SW degree 0.04**
(0.02)
0.04*
(0.02)
0.04**
(0.02)
0.04**
(0.02)
0.04*
(0.02)
0.04*
(0.02)
BUS degree 0.03***
(0.01)
0.03***
(0.01)
0.03***
(0.01)
0.02***
(0.01)
0.02***
(0.01)
0.03***
(0.01)
Black/Hispanic -0.02*
(0.01)
-0.02*
(0.01)
-0.02*
(0.01)
-0.02*
(0.01)
-0.02*
(0.01)
-0.02*
(0.01)
Female 0.02
(0.01)
0.02
(0.01)
0.02
(0.01)
0.01
(0.01)
0.02
(0.01)
0.02
(0.01)
Univ. revenue (log) 0.78***
(0.15)
0.79***
(0.15)
0.78***
(0.15)
0.76***
(0.15)
0.77***
(0.15)
0.79***
(0.15)
Master’s degree (log) 0.31*
(0.17)
0.31*
(0.17)
0.31*
(0.17)
0.32*
(0.17)
0.32*
(0.17)
0.31*
(0.17)
Public 0.1
(0.28)
0.1
(0.28)
0.1
(0.28)
0.1
(0.28)
0.1
(0.28)
0.1
(0.28)
Doctoral/Research 0.16
(0.40)
0.15
(0.40)
0.16
(0.40)
0.16
(0.40)
0.14
(0.40)
0.15
(0.40)
Constant -31.99***
(4.80)
-30.19***
(4.36)
-16.32***
(1.54)
-17.77***
(1.61)
-18.11***
(1.65)
-18.33***
(1.61)
LRχ² 168.72 168.07 168.32 171.84 169.25 167.52
Log-likelihood 73.43 73.1 73.23 74.99 73.69 72.83
* p<0.1; ** p<0.05; *** p<0.01. Standard errors are in parentheses. Total number of adoptions=68; total
number of universities=1,451; total number of yearly spells=58,769.
188
Table C.3. Models for Each Component of the Factored Variable: Any Type of Program
Variable 1 2 3 4 5 6
Nonprofit org. (log) 1.13***
(0.17
Gov. grant (log) 0.77***
(0.12
State association 0.04***
(0.01
Univ. program (log) 0.47***
(0.07
Dissertation (log) 0.42***
(0.07
News article (log) 0.32***
(0.05
Urbanization 0.01**
(0.01)
0.01**
(0.01)
0.01**
(0.01)
0.01*
(0.01)
0.01*
(0.01)
0.01**
(0.01)
Welfare expenditure 0.00
(0.01)
0.00
(0.01)
0.01
(0.01)
0.00
(0.01)
0.01
(0.01)
0.01
(0.01)
Citizen liberalism -0.00
(0.01)
-0.00
(0.01)
-0.00
(0.01)
-0.00
(0.01)
-0.00
(0.01)
-0.00
(0.01)
Government liberalism -0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
PA/SW degree 0.03***
(0.01)
0.03***
(0.01)
0.03***
(0.01)
0.03***
(0.01)
0.03***
(0.01)
0.03***
(0.01)
BUS degree 0.02***
(0.00)
0.02***
(0.00)
0.02***
(0.00)
0.02***
(0.00)
0.02***
(0.00)
0.02***
(0.00)
Black/Hispanic -0.01**
(0.00)
-0.01**
(0.00)
-0.01**
(0.00)
-0.01**
(0.00)
-0.01**
(0.00)
-0.01**
(0.00)
Female 0.01***
(0.01)
0.02***
(0.01)
0.01***
(0.01)
0.01**
(0.01)
0.01**
(0.01)
0.02***
(0.01)
Univ. revenue (log) 0.59***
(0.07)
0.59***
(0.07)
0.59***
(0.07)
0.57***
(0.07)
0.58***
(0.07)
0.60***
(0.07)
Master’s degree (log) 0.35***
(0.07)
0.35***
(0.07)
0.35***
(0.07)
0.36***
(0.07)
0.36***
(0.07)
0.35***
(0.07)
Public 0.33**
(0.14)
0.33**
(0.14)
0.33**
(0.14)
0.32**
(0.14)
0.32**
(0.14)
0.33**
(0.14)
Doctoral/Research -0.08
(0.21)
-0.09
(0.21)
-0.09
(0.21)
-0.08
(0.21)
-0.1
(0.21)
-0.1
(0.21)
Constant -27.84***
(2.34)
-26.16***
(2.14)
-12.64***
(0.69)
-13.82***
(0.73)
-14.13***
(0.75)
-14.51***
(0.74)
LRχ² 497.31 494.62 493.51 502.06 494.9 489.17
Log-likelihood 615.41 614.07 613.51 617.79 614.21 611.35
* p<0.1; ** p<0.05; *** p<0.01. Standard errors are in parentheses. Total number of adoptions=281; total
number of universities=1,446; total number of yearly spells=56,037.
189
APPENDIX D.
Comparison between Original and Imputed Data (Chapter 4)
Table D.1. Descriptive Statistics for Imputed Covariates: Original and Imputed Values
Variable Years available Mean SD Min Max N. Cases
Nonprofit org. 1976-2010 (Original) 645934 310118 259523 1280739 50610
1971-2011 (Imputed) 589386 317415 259523 1280739 59286
Gov. grant 1982-2010 (Original) 71166 39454 19181 155125 39042
1971-2011 (Imputed) 53674 40204 19181 155125 59286
Public affairs field 1989, 1995-2010 (Original) 27.58 4.09 16.04 47.24 24582
1990-2011 (Imputed) 27.19 3.93 16.04 47.24 31812
Human services field 1989, 1995-2010 (Original) 24.64 2.63 17.97 42.89 24582
1990-2011 (Imputed) 24.58 2.55 17.97 42.89 31812
Commercial revenue 1989, 1995-2010 (Original) 71.56 7.50 30.32 93.86 31812
1990-2011 (Imputed) 71.56 7.50 30.32 93.86 31812
PA/PS degree 1971-2010 (Original) 1.60 2.85 0 100 57500
1971-2011 (Imputed) 1.59 2.83 0 100 59286
BUS degree 1971-2010 (Original) 21.11 16.81 0 100 56436
1971-2011 (Imputed) 20.21 17.02 0 100 59286
SW/SC degree 1971-2010 (Original) 11.14 10.15 0 100 57500
1971-2011 (Imputed) 11.11 10.12 0 100 59286
Urbanization 1971-2009 (Original) 72.51 13.85 32.2 94.4 57400
1971-2011 (Imputed) 72.74 13.93 32.2 94.4 59286
Welfare expenditure 1971-2008 (Original) 0.20 0.06 0.04 0.39 56383
1971-2011 (Imputed) 0.20 0.06 0.04 0.39 59286
Citizen liberalism 1971-2008 (Original) 49.83 15.34 7.04 95.97 55965
1971-2011 (Imputed) 50.75 15.78 7.04 95.97 59286
Nonprofit assets 1989, 1995-2010 (Original) 10798 9161 1809 114781 23136
1990-2011 (Imputed) 9555 8563 1809 114781 31812
Univ. revenue 1971-2009 (Original) 139 350 0 10910 55727
1971-2011 (Imputed) 137 353 0 10910 59286
Black/Hispanic 1971-2010 (Original) 9.53 18.90 0 100 57891
1971-2011 (Imputed) 12.58 21.17 0 100 59286
Female 1971-2010 (Original) 54.30 19.07 0 100 57891
1971-2011 (Imputed) 53.04 20.54 0 100 59286
190
APPENDIX E.
Expanded Competing-Risks Models for Institutional Context Variables
Table E.1. Models for Each Component of the Factored Variable: PA/PS
Category Variable 1 2 3 4 5 6
Institutional
Context
Nonprofit org. (log) 1.13***
(0.27)
Gov. grant (log)
0.77***
(0.19)
State association 0.04***
(0.01)
Univ. program (log) 0.54***
(0.12)
Dissertation (log) 0.49***
(0.12)
News article (log) 0.29***
(0.09)
University
Dynamics
PA/PS degree (log) 1.02***
(0.16)
1.02***
(0.16)
1.01***
(0.16)
1.03***
(0.16)
1.01***
(0.16)
0.99***
(0.16)
BUS degree (log) 0.58***
(0.18)
0.60***
(0.18)
0.59***
(0.18)
0.53***
(0.17)
0.53***
(0.17)
0.60***
(0.18)
SW/SC degree (log) -0.19
(0.19)
-0.19
(0.19)
-0.19
(0.19)
-0.16
(0.20)
-0.15
(0.20)
-0.19
(0.19)
State-level
Controls
Urbanization 0.02*
(0.01)
0.02*
(0.01)
0.02*
(0.01)
0.02*
(0.01)
0.02*
(0.01)
0.02*
(0.01)
Welfare expenditure 0.03
(0.02)
0.03
(0.02)
0.04
(0.02)
0.03
(0.02)
0.03
(0.02)
0.04*
(0.02)
Citizen liberalism -0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
University-
level
Controls
Univ. revenue (log) 0.94***
(0.13)
0.95***
(0.13)
0.95***
(0.13)
0.91***
(0.13)
0.93***
(0.13)
0.98***
(0.13)
Public 1.49***
(0.27)
1.49***
(0.27)
1.49***
(0.27)
1.49***
(0.27)
1.48***
(0.27)
1.48***
(0.27)
Doctoral/Research -0.32
(0.33)
-0.32
(0.33)
-0.33
(0.33)
-0.33
(0.33)
-0.35
(0.33)
-0.35
(0.33)
Black/Hispanic -0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
Female 0.02*
(0.01)
0.02*
(0.01)
0.02*
(0.01)
0.02
(0.01)
0.02
(0.01)
0.03**
(0.01)
Constant -35.37***
(3.74)
-33.78***
(3.44)
-20.28***
(1.67)
-21.16***
(1.64)
-21.63***
(1.65)
-22.13***
(1.65)
LRχ² 321.75 320.79 319.75 328.33 325.79 315.72
Log-likelihood 189.31 188.82 188.31 192.59 191.33 186.29
* p<0.1; ** p<0.05; *** p<0.01. Standard errors in parentheses.
Note: Total number of adoptions=101; total number of universities=1,446; total number of yearly spells=56,037.
191
Table E.2. Models for Each Component of the Factored Variable: BUS
Category Variable 1 2 3 4 5 6
Institutional
Context
Nonprofit org. (log) 1.71***
(0.38)
Gov. grant (log) 1.18***
(0.26)
State association 0.06***
(0.01)
Univ. program (log) 0.67***
(0.17)
Dissertation (log) 0.73***
(0.19)
News article (log) 0.51***
(0.12)
University
Dynamics
PA/PS degree (log) 0.06
(0.20)
0.05
(0.20)
0.04
(0.20)
0.04
(0.20)
0.04
(0.20)
0.03
(0.20)
BUS degree (log) 0.52***
(0.17)
0.54***
(0.17)
0.53***
(0.17)
0.50***
(0.17)
0.50***
(0.17)
0.53***
(0.17)
SW/SC degree (log) -0.05
(0.18)
-0.05
(0.18)
-0.05
(0.18)
-0.04
(0.18)
-0.04
(0.18)
-0.05
(0.18)
State-level
Controls
Urbanization 0.03**
(0.01)
0.03**
(0.01)
0.03**
(0.01)
0.03**
(0.01)
0.03**
(0.01)
0.03***
(0.01)
Welfare expenditure -0.04
(0.03)
-0.04
(0.03)
-0.04
(0.03)
-0.04
(0.03)
-0.04
(0.03)
-0.04
(0.03)
Citizen liberalism -0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
University-
level
Controls
Univ. revenue (log) 0.58***
(0.14)
0.58***
(0.14)
0.58***
(0.14)
0.58***
(0.14)
0.58***
(0.14)
0.59***
(0.14)
Public -1.51***
(0.37)
-1.51***
(0.37)
-1.51***
(0.37)
-1.51***
(0.37)
-1.52***
(0.37)
-1.52***
(0.37)
Doctoral/Research -0.06
(0.56)
-0.07
(0.56)
-0.08
(0.56)
-0.09
(0.55)
-0.08
(0.55)
-0.08
(0.56)
Black/Hispanic -0.02
(0.01)
-0.02
(0.01)
-0.02
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.02
(0.01)
Female 0.02*
(0.01)
0.02*
(0.01)
0.02*
(0.01)
0.02
(0.01)
0.02
(0.01)
0.02*
(0.01)
Constant -37.88***
(5.32)
-35.51***
(4.81)
-14.89***
(1.61)
-16.72***
(1.68)
-17.75***
(1.80)
-17.83***
(1.70)
LRχ² 112.54 112.1 110.33 109.18 111.74 110.92
Log-likelihood 43.54 43.31 42.43 41.85 43.14 42.73
* p<0.1; ** p<0.05; *** p<0.01. Standard errors in parentheses.
Note: Total number of adoptions=62; total number of universities=1,446; total number of yearly spells=56,037.
192
Table E.3. Models for Each Component of the Factored Variable: SW/SC
Category Variable 1 2 3 4 5 6
Institutional
Context
Nonprofit org. (log) 0.99**
(0.46
Gov. grant (log) 0.65**
(0.32
State association 0.04**
(0.02
Univ. program (log) 0.45**
(0.19
Dissertation (log) 0.36*
(0.18
News article (log) 0.30**
(0.14
University
Dynamics
PA/PS degree (log) -0.05
(0.27)
-0.06
(0.27)
-0.05
(0.27)
-0.03
(0.26)
-0.06
(0.26)
-0.06
(0.27)
BUS degree (log) 0.05
(0.18)
0.06
(0.19)
0.05
(0.19)
0.02
(0.18)
0.04
(0.18)
0.05
(0.19)
SW/SC degree (log) 0.31
(0.30)
0.3
(0.30)
0.31
(0.30)
0.32
(0.30)
0.32
(0.30)
0.31
(0.30)
State-level
Controls
Urbanization -0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.01
(0.01)
Welfare expenditure 0.01
(0.04)
0.01
(0.04)
0.01
(0.04)
0.01
(0.04)
0.02
(0.04)
0.01
(0.04)
Citizen liberalism 0.00
(0.01)
0.00
(0.01)
0.00
(0.01)
0.00
(0.01)
0.00
(0.01)
0.00
(0.01)
University-
level
Controls
Univ. revenue (log) 0.98***
(0.18)
0.99***
(0.18)
0.98***
(0.18)
0.96***
(0.18)
0.98***
(0.18)
0.98***
(0.18)
Public 0.35
(0.37)
0.35
(0.37)
0.35
(0.37)
0.36
(0.37)
0.35
(0.37)
0.35
(0.37)
Doctoral/Research -0.84
(0.56)
-0.85
(0.56)
-0.84
(0.56)
-0.83
(0.56)
-0.86
(0.56)
-0.85
(0.56)
Black/Hispanic -0.00
(0.01)
0.00
(0.01)
-0.00
(0.01)
0.00
(0.01)
0.00
(0.01)
-0.00
(0.01)
Female 0.02
(0.02)
0.02
(0.02)
0.02
(0.02)
0.02
(0.02)
0.02
(0.02)
0.02
(0.02)
Constant -28.66***
(6.22
-26.82***
(5.70
-15.41***
(2.07
-16.48***
(2.09
-16.70***
(2.11
-17.15***
(2.11
LRχ² 77.42 76.8 77.41 78.99 77.31 77.2
Log-likelihood 12.28 11.97 12.28 13.07 12.23 12.17
* p<0.1; ** p<0.05; *** p<0.01. Standard errors in parentheses.
Note: Total number of adoptions=38; total number of universities=1,446; total number of yearly spells=56,037.
193
APPENDIX F.
Multinomial Logistic Approach to Competing Risks
Table F.1. Multinomial Logistic Competing-Risks Models: 1971-1986 and 1971-1991
Variable 1971-1886 1971-1991
PA/PS BUS SW/SC PA/PS BUS SW/SC
Structuration (log) -1.47**
(0.64)
-6.07***
(0.85)
-3.19***
(0.74)
-1.34***
(0.24)
-4.01***
(1.11)
-2.44***
(0.53)
Public affairs field
Human services field
Commercial revenue
PA/PS degree (log) 0.93*
(0.54)
0.21
(1.08)
-0.38
(0.51)
0.69*
(0.36)
0.52
(0.76)
-0.31
(0.38)
BUS degree (log) -0.25
(0.46)
1.43***
(0.43)
0.37
(0.62)
0.28
(0.41)
1.11***
(0.35)
0.56
(0.55)
SW/SC degree (log) -0.36
(0.65)
0.75
(0.53)
0.58
(0.48)
-0.42
(0.42)
0.14
(0.52)
0.4
(0.43)
Urbanization 0.01
(0.03)
-0.02
(0.03)
-0.06**
(0.02)
0.03
(0.02)
0.00
(0.03)
-0.06**
(0.03)
Welfare expenditure -0.1
(0.06)
0.25***
(0.06)
-0.02
(0.09)
-0.07
(0.04)
0.19***
(0.06)
-0.00
(0.05)
Citizen liberalism 0.01
(0.02)
-0.04***
(0.01)
0.05***
(0.02)
0.02
(0.01)
-0.03**
(0.01)
0.06***
(0.02)
Nonprofit assets (log)
Univ. revenue (log) -0.12
(0.73)
0.84***
(0.26)
0.78
(0.53)
0.43
(0.32)
0.56**
(0.27)
1.22***
(0.35)
Public 0.00
(1.08)
-1.73*
(0.98)
-1.89
(1.35)
0.71
(0.67)
-1.05
(0.77)
-0.27
(0.93)
Doctoral/Research 3.52
(2.35)
0.28
(1.26)
2.66
(2.69)
1.67*
(0.90)
1.13
(1.13)
-0.15
(1.00)
Black/Hispanic -0.00
(0.01)
-0.01
(0.02)
-0.05
(0.06)
-0.00
(0.01)
-0.01
(0.02)
0.00
(0.01)
Female 0.02
(0.02)
0.04*
(0.02)
0.02
(0.03)
0.04**
(0.02)
0.04**
(0.02)
0.01
(0.02)
Constant -10.74***
(2.83)
-29.30***
(2.26)
-17.60***
(3.99)
-16.77***
(3.65)
-24.66***
(2.32)
-19.57***
(4.72)
LRχ² 2099.99 1117.53
Pseudo R
2
0.23 0.18
# of adoptions 8 6 4 18 8 7
* p<0.1; ** p<0.05; *** p<0.01. Standard errors in parentheses. “Non-adoption (0)” is the reference category.
Total number of adoptions=18 and 33; Total number of universities=1,446; Total number of yearly spells=59,286.
Note: Nonprofit-sector factors (Public affairs field, Human services field, Commercial revenue, and Nonprofit assets)
could not be tested for years 1986 and 1991, because both failed to meet the minimum required number of “adoption”
events.
194
Table F.2. Multinomial Logistic Competing-Risks Models: 1971-1996
Variable 1971-1996 1971-1996 with Nonprofit-Sector Effects
PA/PS BUS SW/SC PA/PS BUS SW/SC
Structuration (log) -0.94***
(0.15)
-1.72***
(0.46)
-1.40***
(0.39)
-2.76***
(0.55)
-2.96***
(0.48)
-3.15***
(0.89)
Public affairs field
0.05
(0.05)
-0.04
(0.10)
-0.07
(0.09)
Human services field
0.01
(0.09)
0.16
(0.14)
-0.16
(0.23)
Commercial revenue
-0.00
(0.03)
0.11***
(0.04)
0.14*
(0.08)
PA/PS degree (log) 1.09***
(0.26)
0.33
(0.46)
0.03
(0.33)
1.25***
(0.27)
0.34
(0.51)
0.54
(0.34)
BUS degree (log) 0.69*
(0.42)
0.70***
(0.22)
0.4
(0.54)
1.08***
(0.38)
0.48**
(0.22)
0.03
(0.71)
SW/SC degree (log) -0.37
(0.31)
-0.4
(0.47)
0.37
(0.29)
-0.26
(0.29)
-0.85
(0.75)
0.56
(0.42)
Urbanization 0.01
(0.01)
0.02
(0.03)
-0.05**
(0.02)
0.00
(0.01)
0.04
(0.04)
-0.02
(0.03)
Welfare expenditure 0.02
(0.04)
0.15***
(0.03)
0.03
(0.04)
0.02
(0.06)
0.13**
(0.07)
0.09*
(0.05)
Citizen liberalism 0.01
(0.01)
-0.02
(0.01)
0.02
(0.02)
0.01
(0.02)
-0.01
(0.01)
-0.06
(0.05)
Nonprofit assets (log)
-0.53
(0.64)
0.63
(1.23)
0.35
(0.35)
Univ. revenue (log) 0.79***
(0.21)
0.48*
(0.29)
1.05***
(0.29)
1.03***
(0.21)
0.33
(0.52)
0.48
(0.34)
Public 0.91**
(0.42)
-0.78
(0.67)
-0.33
(0.60)
0.99*
(0.54)
-0.46
(1.07)
0.39
(0.96)
Doctoral/Research 0.12
(0.57)
1.02
(0.86)
-0.29
(0.90)
-0.54
(0.73)
1.11
(1.22)
-0.43
(1.53)
Black/Hispanic -0.02
(0.01)
-0.02
(0.02)
-0.01
(0.01)
-0.02
(0.02)
-0.02
(0.03)
-0.02
(0.01)
Female 0.05***
(0.01)
0.02
(0.02)
0.01
(0.02)
0.04***
(0.01)
0.01
(0.03)
-0.02
(0.03)
Constant -20.06***
(3.31)
-17.91***
(2.27)
-16.44***
(4.19)
-19.10**
(8.15)
-33.14***
(7.61)
-18.70**
(8.75)
LRχ² 567.87 3737.24
Pseudo R
2
0.14 0.23
# of adoptions 36 14 11 36 14 11
* p<0.1; ** p<0.05; *** p<0.01. Standard errors in parentheses. “Non-adoption (0)” is the reference category.
Total number of adoptions=85; Total number of universities=1,446; Total number of yearly spells=59,286.
195
Table F.3. Multinomial Logistic Competing-Risks Models: 1971-2001
Variable 1971-2001 1971-2001 with Nonprofit-Sector Effects
PA/PS BUS SW/SC PA/PS BUS SW/SC
Structuration (log) -0.75***
(0.19)
-0.95***
(0.22)
-0.73***
(0.21)
-1.23***
(0.30)
-1.29***
(0.32)
-1.13***
(0.28)
Public affairs field
0.10**
(0.05)
0.03
(0.08)
0.13***
(0.04)
Human services field
0.13**
(0.06)
-0.15
(0.14)
0.20***
(0.07)
Commercial revenue
0.05
(0.04)
0.01
(0.05)
0.01
(0.03)
PA/PS degree (log) 0.03
(0.31)
-0.02
(0.28)
-0.26
(0.30)
0.08
(0.32)
0.24
(0.33)
-0.25
(0.44)
BUS degree (log) 0.60***
(0.16)
0.50
(0.47)
0.42*
(0.25)
0.41**
(0.17)
0.23
(0.58)
0.99***
(0.36)
SW/SC degree (log) -0.28
(0.26)
0.81**
(0.34)
0.36
(0.31)
-0.47
(0.29)
1.28**
(0.51)
0.95**
(0.44)
Urbanization 0.03*
(0.02)
-0.03*
(0.02)
0.02
(0.02)
0.04*
(0.02)
-0.01
(0.02)
0.01
(0.01)
Welfare expenditure 0.09**
(0.04)
0.06**
(0.03)
0.06*
(0.04)
0.03
(0.05)
0.05
(0.04)
-0.02
(0.05)
Citizen liberalism -0.04***
(0.01)
0.00
(0.02)
-0.02
(0.01)
-0.04
(0.03)
-0.02
(0.02)
-0.02
(0.02)
Nonprofit assets (log)
0.10
(0.67)
-0.59
(0.62)
-0.47
(0.45)
Univ. revenue (log) 0.77***
(0.18)
1.29***
(0.22)
0.49**
(0.23)
0.75***
(0.19)
1.12***
(0.32)
0.72***
(0.24)
Public -1.45***
(0.53)
0.21
(0.52)
-0.12
(0.40)
-1.49**
(0.65)
0.76
(0.65)
0.08
(0.50)
Doctoral/Research -0.28
(0.58)
-1.16*
(0.69)
-0.38
(0.69)
-0.71
(0.61)
-1.87*
(1.07)
-0.61
(0.63)
Black/Hispanic -0.01
(0.01)
-0.01
(0.01)
-0.02**
(0.01)
-0.00
(0.01)
-0.01
(0.01)
-0.03**
(0.01)
Female 0.03**
(0.01)
0.02
(0.02)
0.02*
(0.01)
0.02*
(0.01)
0.01
(0.03)
0.04**
(0.02)
Constant -16.61***
(1.57)
-20.21***
(4.44)
-14.66***
(2.08)
-24.67***
(8.09)
-11.54
(7.57)
-23.46***
(7.39)
LRχ² 228.39 609.64
Pseudo R
2
0.08 0.12
Number of adoptions 61 29 18 61 29 18
* p<0.1; ** p<0.05; *** p<0.01. Standard errors in parentheses. “Non-adoption (0)” is the reference category.
Total number of adoptions=146; Total number of universities=1,446; Total number of yearly spells=59,286.
196
Table F.4. Multinomial Logistic Competing-Risks Models: 1971-2006
Variable 1971-2006 1971-2006 with Nonprofit-Sector Effects
PA/PS BUS SW/SC PA/PS BUS SW/SC
Structuration (log) -0.19*
(0.10)
-0.14
(0.15)
-0.19
(0.17)
-0.69***
(0.13)
-0.40*
(0.21)
-0.27
(0.23)
Public affairs field
0.07**
(0.03)
0.09**
(0.04)
-0.01
(0.07)
Human services field
0.06
(0.05)
0.05
(0.06)
-0.06
(0.11)
Commercial revenue
-0.01
(0.01)
0.04
(0.03)
0.03
(0.05)
PA/PS degree (log) 0.76***
(0.17)
-0.12
(0.27)
-0.25
(0.21)
0.84***
(0.17)
-0.13
(0.30)
-0.14
(0.24)
BUS degree (log) 0.54**
(0.26)
0.68***
(0.17)
0.01
(0.23)
0.60*
(0.32)
0.54***
(0.19)
-0.19
(0.25)
SW/SC degree (log) -0.16
(0.20)
-0.21
(0.23)
0.66***
(0.25)
-0.10
(0.19)
-0.34
(0.24)
0.76**
(0.31)
Urbanization 0.01
(0.01)
0.03**
(0.01)
-0.02
(0.01)
0.01
(0.01)
0.04**
(0.02)
-0.00
(0.02)
Welfare expenditure 0.07***
(0.02)
0.02
(0.03)
0.05
(0.03)
0.05**
(0.02)
-0.05
(0.04)
0.04
(0.05)
Citizen liberalism -0.01
(0.01)
-0.03**
(0.01)
-0.00
(0.01)
-0.02
(0.01)
-0.03
(0.02)
-0.01
(0.02)
Nonprofit assets (log)
0.24
(0.25)
0.23
(0.51)
0.00
(0.61)
Univ. revenue (log) 0.86***
(0.12)
0.68***
(0.13)
0.99***
(0.16)
0.87***
(0.12)
0.66***
(0.13)
0.83***
(0.20)
Public 1.35***
(0.25)
-1.14***
(0.42)
0.39
(0.38)
1.79***
(0.31)
-1.05**
(0.47)
0.71
(0.46)
Doctoral/Research -0.27
(0.32)
-0.24
(0.49)
-1.02*
(0.59)
-0.55
(0.34)
-0.6
(0.55)
-1.42*
(0.79)
Black/Hispanic -0.00
(0.00)
-0.01
(0.01)
0.00
(0.01)
-0.00
(0.00)
-0.01
(0.01)
0.00
(0.01)
Female 0.03**
(0.01)
0.03***
(0.01)
0.02
(0.01)
0.01
(0.02)
0.02**
(0.01)
0.02
(0.02)
Constant -18.82***
(2.26)
-15.43***
(1.50)
-15.93***
(2.45)
-21.70***
(4.35)
-21.80***
(5.53)
-15.16*
(8.96)
LRχ² 413.09 375.51
Pseudo R
2
0.13 0.13
Number of adoptions 89 42 31 89 42 31
* p<0.1; ** p<0.05; *** p<0.01. Standard errors in parentheses. “Non-adoption (0)” is the reference category.
Total number of adoptions=221; Total number of universities=1,446; Total number of yearly spells=59,286.
197
Table F.5. Multinomial Logistic Competing-Risks Models: 1971-2011
Variable 1971-2011 1971-2011 with Nonprofit-Sector Effects
PA/PS BUS SW/SC PA/PS BUS SW/SC
Structuration (log) 0.02
(0.10)
0.56***
(0.15)
0.17
(0.18)
-0.36***
(0.13)
0.56***
(0.21)
0.16
(0.23)
Public affairs field
0.05*
(0.03)
0.02
(0.04)
-0.03
(0.07)
Human services field
0.05
(0.04)
0.00
(0.06)
-0.09
(0.10)
Commercial revenue
-0.01
(0.01)
0.01
(0.02)
0.04
(0.04)
PA/PS degree (log) 0.74***
(0.16)
-0.09
(0.22)
-0.24
(0.21)
0.79***
(0.16)
-0.11
(0.24)
-0.17
(0.25)
BUS degree (log) 0.52**
(0.23)
0.51***
(0.12)
0.02
(0.20)
0.58**
(0.26)
0.43***
(0.12)
-0.12
(0.21)
SW/SC degree (log) -0.18
(0.18)
-0.05
(0.18)
0.31
(0.30)
-0.12
(0.16)
-0.13
(0.19)
0.31
(0.35)
Urbanization 0.02**
(0.01)
0.03***
(0.01)
-0.01
(0.01)
0.01
(0.01)
0.03**
(0.01)
-0.00
(0.01)
Welfare expenditure 0.05**
(0.02)
-0.03
(0.03)
0.03
(0.04)
0.04*
(0.02)
-0.07**
(0.04)
0.02
(0.05)
Citizen liberalism -0.01
(0.01)
-0.02*
(0.01)
0.00
(0.01)
-0.01
(0.01)
-0.01
(0.01)
-0.01
(0.02)
Nonprofit assets (log)
0.19
(0.23)
-0.25
(0.42)
0.02
(0.50)
Univ. revenue (log) 0.88***
(0.11)
0.50***
(0.12)
0.92***
(0.16)
0.89***
(0.11)
0.50***
(0.12)
0.82***
(0.20)
Public 1.47***
(0.24)
-1.48***
(0.38)
0.36
(0.35)
1.90***
(0.28)
-1.57***
(0.43)
0.56
(0.41)
Doctoral/Research -0.47
(0.29)
-0.29
(0.48)
-1.01*
(0.56)
-0.75**
(0.30)
-0.71
(0.56)
-1.34*
(0.72)
Black/Hispanic -0.00
(0.00)
-0.01**
(0.01)
0.00
(0.01)
-0.00
(0.01)
-0.01*
(0.01)
0.00
(0.01)
Female 0.03**
(0.01)
0.02**
(0.01)
0.02
(0.01)
0.00
(0.02)
0.02*
(0.01)
0.02
(0.02)
Constant -18.59***
(2.00)
-13.11***
(1.15)
-14.15***
(2.09)
-20.34***
(3.69)
-10.67**
(4.61)
-13.22*
(7.11)
LRχ² 527.39 372.57
Pseudo R
2
0.13 0.12
Number of adoptions 101 62 38 101 62 38
* p<0.1; ** p<0.05; *** p<0.01. Standard errors in parentheses. “Non-adoption (0)” is the reference category.
Total number of adoptions=281; Total number of universities=1,446; Total number of yearly spells=59,286.
198
APPENDIX G.
Summary of Seven Cases
Table G.1. Detailed Description of University Cases
Mountain
West
Private
Great Lakes
Public Great
Lakes
South
Atlantic
Southern
Gold Coast
Northern
Gold Coast
Central Gold
Coast
Formation of Nonprofit Studies
Degree mid-2000s late-1980s early-1990s late-1990s early-2000s early-1980s mid-2010s
Minor/Track/Certificate early-1980s late-1980s early-1990s early-2000s late-1980s mid-1990s mid-1990s
Research center late-1990s mid-1980s late-1980s late-1990s mid-2000s early-1980s early-2000s
Organizational Setting for Nonprofit Studies
Types of degrees offered BS/MA MA BA/MA/PhD MA MA MA MA
Free-standing system Yes No Yes No No No No
Socioeconomic & Political Context
Region West Midwest Midwest South West West West
Total # of nonprofits 23,719 68,298 37,703 41,013 166,725 166,725 166,725
Total # of grantmaking foundations 929 3,306 1,128 1,433 7,342 7,342 7,342
% People living in a SMSA 88.2 77.4 70.8 71.6 94.4 94.4 94.4
Per cap income (USD) 34,539 36,162 33,981 34,747 42,514 42,514 42,514
Citizen liberalism (index) 30.85 59.16 46.90 53.98 65.40 65.40 65.40
Government liberalism (index) 47.79 53.05 32.83 24.89 48.87 48.87 48.87
Organizational Characteristics
Total revenues (million USD) 790.52 569.85 910.34 890.37 181.47 174.03 1,528.45
Total # of degrees conferred 8,530 1,897 2,950 6,399 1,274 2,086 7,200
Public institution Yes No Yes Yes No No No
Religious institution No No No No Yes Yes No
Land-grant institution No No No Yes No No No
Flagship institution No No No Yes No No No
Very-high research activity Yes Yes No Yes No No Yes
% Black/Hispanic students 10.79 6.76 9.50 6.77 12.22 13.94 14.19
% Female students 49.40 41.52 55.70 53.75 55.01 58.18 44.49
Note: Socioeconomic & Political Context items are based on state-level secondary data in 2010; Organizational Characteristics items are based on university-
level secondary data and take averages for the period of 1980-2010—to protect the anonymity of universities.
199
APPENDIX H.
Interview Protocol and Instrument
University ID #: ___________________________________________
Interviewee ID #: ______________________
Date of Interview: ______________________
Time Begin: _________________ Time End: _________________
Thank you very much for taking the time today. This interview is designed to study the
institutionalization of nonprofit management studies at universities. This is part of my
doctoral dissertation project, and I am interested in what factors are associated with the
development of nonprofit research and education programs and why different conceptions
of nonprofit studies exist across universities. For a comparative case study, I will interview
several other faculties and directors at different universities as well.
I promise that all information you provide me will be handled as confidentially and used for
research purposes only. Also, your name and other personally identifiable information will
be kept anonymous in my dissertation and any other types of publications and presentations.
With your permission, I will audiotape and take notes during the interview. The recording
is to accurately record the information you provide, and will be used for transcription
purposes only. If you choose not to be audiotaped, I will take notes instead. If you agree to
being audiotaped but feel uncomfortable at any time during the interview, I can turn off the
recorder at your request. Or if you don’t wish to continue, you can stop the interview at any
time.
200
A. Interviewee’s Personal Background
[1] What is your current position at your institution?
[2] How did you come to that position? Could you briefly describe your career biography?
B. Current Nonprofit Management Research and Education Programs
*Note: Here, I define the term “nonprofit management research and education programs” as a
broad one that includes all types of university-based education, training, and research programs
for those who (potentially) working in or with the nonprofit sector, not necessarily for nonprofit
managers.
[1] Does your institution offer undergraduate- and graduate-level degree programs (e.g.,
BA/BS, MA/MS, and PhD)?
1. If so, what types of degree programs does your institution have?
2. If so, when was the first degree program created?
[2] Does your institution offer undergraduate- and graduate-level non-degree programs (e.g.,
minor, track, emphasis, concentration, specialization, and certificate)?
1. If so, what types of non-degree programs does your institution have?
2. If so, when was the first non-degree program created?
[3] Does your institution have nonprofit research centers?
1. If so, how many research centers does your institution have?
2. If so, when was the first research center established?
C. Emergence: Historical, Social, and Organizational Antecedents
[1] Could you briefly describe under what conditions the very first nonprofit management
program was created at your institution? If you do not know, could you direct me to
someone who could discuss the emergence of nonprofit programs at this institution?
1. Did any historical or social events motivate the creation of program?
2. Were there any requests from local communities or nonprofits?
3. Were there any pressures from an academic community or rival university?
4. Did any specific university mission or status (e.g., public/private, religious, land grant,
flagship, or minority-serving) affect the creation of the program?
5. Was there any pivotal faculty for the creation of the program?
201
6. Which school/department played a leading role? Did different types of schools and
departments cooperate for designing the program?
7. Overall, what were the barriers and facilitating factors for creating the new program?
[2] Could you briefly describe under what historical, social, and organizational conditions the
subsequent programs were created?
[3] How important have foundations been for the creation of your programs?
D. Evolution: Institutional and Intellectual Features
[1] Institutional Elements
1. By which school/department (e.g., public administration and policy, business and
management, social work, and professional studies) within your institution have the
programs been offered?
2. Were there any influential faculty or stakeholders in the early-to-middle phases of the
development of programs at your institution?
3. How has your institution staffed the programs? What is the proportion of full-time,
tenure-track faculty? And what disciplinary background do faculty have?
4. How has your institution funded the programs? What is the major source of funding?
Is it mainly tuition or do you have support from foundations?
[2] Intellectual Conceptions
1. How does your institution define the scope and nature of nonprofit management studies?
Does your institution define it as an interdisciplinary study field or independent
discipline?
2. What is the main objective of your programs? What is the key to the uniqueness of
your programs?
a. Do your programs have any teaching and research priorities?
b. Do your programs target any specific stakeholders?
c. Do your programs have any particular theoretical/methodological foundations?
d. How do your programs differentiate themselves from traditional study fields?
3. Have the conceptions and objectives of nonprofit studies at your institution been
changed since the very first program was created? If so, how?
202
E. Future Development
[1] Regarding nonprofit management research and education programs at your institution,
1. What do you think about your current programs?
a. What are the strengths and weaknesses of your current programs?
b. On a scale from 1-10, how would you rate the maturity of nonprofit studies at
your institution?
2. Does your institution have any future plans? What are the potential barriers and
opportunities for your future plans?
3. Who do you see as your major competitors in nonprofit management?
[2] Regarding nonprofit management programs in U.S. universities and colleges,
1. Do you think nonprofit management will continue to exist in different academic
departments like public administration, business, and social work, or will it become
more common in one field?
2. Are you satisfied with the current status of nonprofit management research and
education? If not, what kind of efforts should be made for the future development?
3. What do you see as the future of nonprofit management studies? Will nonprofit degrees
become more common? How about research centers? There have been some examples
of nonprofit centers failing, why do you think that is taking place?
Thank you very much for your time, again.
I expect to conduct only one interview, but follow-ups may be needed for clarification. If
so, would you mind if I contact you to request it?
And, would you be interested in receiving a copy of the findings of this research? If you
have any questions or concerns regarding my research, please contact me at anytime.
203
APPENDIX I.
Technical Interpretation of the QCA Results
Formation of the Study Field (Table 5.4)
Table 5.4 summarizes multiple combinations of measures (so-called solutions) that lead to
the outcomes Early and Late formation. With respect to Early formation, first, solution 1 (1a and
1b) evaluates the effect of each of the four university factors. The parsimonious solution indicates
that universities funded by a large national foundation (Leading funder) in a small size (~Large
university) tend to be early adopters of degree programs. In addition, the intermediate solution
indicates that having an external supporter (External champion) and a non-elite status (~Elite
university) are peripheral conditions and there are trade-offs between them. In other words,
solution 1a shows that being non-elite allows for a university’s early adoption regardless of
whether it has an external champion, as indicated by the blank space for External champion that
signals a “don’t care” situation for that causal condition. By contrast, solution 1b presents the
opposite pattern: in the presence of an external champion, an early-adopter university may or may
not be highly selective. Solution 2 (2a and 2b) evaluates the effect of all university and
social/political factors, indicating that there are two different paths to Early formation. Along both
paths, two compounding sets of university factors are identified as core conditions and three
social/political factors are identified as peripheral conditions. On the one hand, solution 2a
specifically indicates that universities with either an external champion or a leading funder as a
major actor (Champion or Funder), that are not large and elite (~Large and Elite), and that are
located in a region with a large nonprofit sector (Large nonprofit-sector) are likely to have
established degree programs before 2000. This solution covers the cases of Private Great Lakes
204
and Northern Gold Coast. On the other hand, solution 2b indicates that universities with either an
external champion or leading funder as a major actor (Champion or Funder), that are not large and
elite (~Large and Elite), and that are located in a less urbanized (~High urbanization) and more
conservative (~High liberalism) region are also likely. This solution covers the case of Public
Great Lakes. The two solutions show trade-offs among social/political factors, with the existence
of a large nonprofit sector and the absence of a high level of urbanization and liberalism
substituting for each other. Overall, solution 2a appears to cover more cases in the data set (with a
raw coverage of 0.58) and has more explanatory power (with a unique coverage of 0.29) than
solution 2b (with a raw coverage of 0.32 and unique coverage of 0.03).
With respect to Late formation, solution 3 indicates that large size (Large university) as a
core condition combining with high selectivity (Elite university) as a peripheral condition is
sufficient for the outcome. Solutions 4a, 4b, and 4c indicate the existence of three distinct
configurational groupings for Late formation when including all university and social/political
factors. Solution 4a covers the cases of Southern Gold Coast and Central Gold Coast, indicating
that the absence of an external champion or a leading funder (~Champion or Funder) combining
with demographic heterogeneity (High urbanization) and the liberal political orientation (High
liberalism) of a region pushes universities toward adopting degree programs after 2000. Solution
4b covers the case of Mountain West, indicating that the combination of being large and elite
(Large and Elite) and located in a region with a small nonprofit sector (~Large nonprofit-sector)
and that is highly urbanized (High urbanization) contributes to a university’s late adoption of a
degree program. Solution 4c covers the case of South Atlantic, indicating that not having an
external champion or a leading funder (~Champion or Funder) combining with being large and
elite (Large and Elite) and being located in a region with a small nonprofit sector (~Large
205
nonprofit-sector) and liberal political ideology (High liberalism) leads universities to late adoption.
Among these three solutions, solution 4a shows the highest raw and unique coverage scores (0.57
and 0.45) and, therefore, is the most empirically relevant causal recipe.
Institutional Stability of the Study Field (Table 5.5)
Table 5.5 shows various causal combinations that lead to the High- and Low-stability
outcomes. With respect to High stability, solution 5 identifies Iconic leader and Leading funder as
core conditions and suggests that the combination of these two factors, regardless of a university’s
organizational characteristics, is sufficient for the outcome.
38
Solution 6 (6a and 6b) also identifies
the set union of these two factors (Leader or Funder) as a core condition, along with a small
nonprofit sector at the state level (~Large nonprofit-sector). Specifically, solution 6a, which
encompasses the case of Public Great Lakes, shows that the combination of the constant
involvement of an iconic leader or powerful external funder (Leader or Funder) and being located
in a region with a small nonprofit sector (~Large nonprofit-sector), that is less urbanized (~High
urbanization), and that is politically conservative (~High liberalism) is one of the two paths to
High stability. Solution 6b, which encompasses the case of Mountain West, presents another path
that combines the two core conditions (Leader or Funder and ~Large nonprofit-sector), with being
large and elite (Large and Elite) and being located in a more conservative region (~High liberalism)
as peripheral conditions. In terms of raw and unique coverage scores, solution 6a (0.38 and 0.26)
seems to constitute a slightly better causal recipe than solution 6b (0.30 and 0.17).
38
The test results for necessary conditions indicate the existence of two possible necessary conditions, namely, Iconic
leader and Leading funder. However, it is difficult to say these two factors are necessary conditions for High stability
because the solutions do not cover all possible paths and, indeed, there are other paths that are not captured here given
the limited diversity of cases but that lead to High stability.
206
With respect to Low stability, solution 7 (7a, 7b, and 7c) shows three different combinations
of university factors as a sufficient condition. Solutions 7a and 7b rely on the absence of an iconic
leader (~Iconic leader) in combination with either having a local foundation or individual donor
as an external funder (~Leading funder) or being small (~Large university); ~Leading funder and
~Large university can be treated as substitutes here. The case of South Atlantic is covered by
solution 7a (raw coverage 0.17; unique coverage 0.14), and the cases of Private Great Lakes and
Northern Gold Coast are covered by solution 7b (raw coverage 0.41; unique coverage 0.38). The
case of Southern Gold Coast is covered by solution 7c, which involves having a local foundation
or individual donor as an external funder (~Leading funder) and being small (~Large university)
as core conditions and being non-elite (~Elite university) as a peripheral condition. Solution 8 (8a
and 8b) indicates the existence of two different configurations for Low stability when including all
university and social/political factors. Solution 8a indicates that universities without an iconic
leader or leading funder (~Leader or Funder) but that are located in a liberal state (High liberalism)
tend to have less organizationally independent systems for nonprofit management studies.
However, this solution covers only one case, South Atlantic, and has low raw and unique coverage
scores (0.17 and 0.13, respectively). Solution 8b covers the cases of Private Great Lakes, Southern
Gold Coast, and Northern Gold Coast, with a raw coverage of 0.62 and unique coverage of 0.58
and, therefore, is a more empirically relevant recipe than solution 8a. The solution indicates that
small, non-elite universities (~Large and Elite) that are located in a region with a large nonprofit
sector (Large nonprofit-sector) and that is highly urbanized (High urbanization) and highly liberal
(High liberalism) tend to have less stable systems for nonprofit management studies.
Abstract (if available)
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
The functions of the middleman: how intermediary nonprofit organizations support the sector and society
PDF
Intradepartmental collaboration in the public organizations: implications to practice in an era of resource scarcity and economic uncertainty
PDF
Looking to the stars: perceptions of celebrity influence in the nonprofit sector
PDF
Institutional contestation, network legitimacy and organizational heterogeneity: interactions between government and environmental nonprofits in South Korea
PDF
China-Africa cooperation: an assessment through the lens of China’s development experience
PDF
Elements of a successful multi-sectoral collaborative from a local government perspective: a framework for collaborative governance – dimensions shared by award winning multi-sectoral partnerships
PDF
Embedding sustainability: a change management guide for ports
PDF
The impact of social capital: a case study on the role of social capital in the restoration and recovery of communities after disasters
PDF
Governing regional collaboratives: institutional design, management and leadership
PDF
Emerging catastrophes in slums of the developing world: considerations for policy makers
PDF
Urban universities' campus expansion projects in the 21st century: a case study of the University of Southern Calfornia's "Village at USC" project and its potential economic and social impacts on...
PDF
The use of mobile technology and mobile applications as the next paradigm in development: can it be a game-changer in development for women in rural Afghanistan?
PDF
Individual differences in trust development: Assessing the effects of trustor attributes on trust building, stability, and dissolution
PDF
The context of leadership in the development of California’s innovation hubs
PDF
Essays on fiscal outcomes of cities in California
PDF
Nonprofit organizations and the environmental policy outcomes: a systematic inquiry into the role of different types of nonprofits to influence the processes and outcomes of the environmental policy
PDF
A miracle or a mirage? A study to evaluate the impacts of microfinance
PDF
Structure, agency, and the Kuznets Curve: observations and implications for sustainability planning in U.S. cities
PDF
Quango reforms and challenges in South Korea: social relations, informal networks, and hidden actions
PDF
Collaboration: is it worth it? The Magnolia Community Initiative from the perspective of initiative partner participants
Asset Metadata
Creator
Lee, Youngmi
(author)
Core Title
The institutionalization of nonprofit management: emergence, development, and legitimization
School
School of Policy, Planning and Development
Degree
Doctor of Policy, Planning & Development
Degree Program
Policy, Planning, and Development
Publication Date
07/16/2015
Defense Date
05/20/2015
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
institutionalization,nonprofit management,nonprofit management studies,nonprofit sector,OAI-PMH Harvest,Public Management,social sciences,structuration
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Graddy, Elizabeth A. (
committee chair
), Robertson, Peter John (
committee member
), Suárez, David F. (
committee member
)
Creator Email
youngmil@usc.edu,youngmilee15@gmail.com
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c3-596604
Unique identifier
UC11300336
Identifier
etd-LeeYoungmi-3617.pdf (filename),usctheses-c3-596604 (legacy record id)
Legacy Identifier
etd-LeeYoungmi-3617.pdf
Dmrecord
596604
Document Type
Dissertation
Format
application/pdf (imt)
Rights
Lee, Youngmi
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
institutionalization
nonprofit management
nonprofit management studies
nonprofit sector
structuration