University of Southern California Dissertations and Theses

Collaborative social networks in student affairs: an exploration of the outcomes and strategies associated with cross‐institutional collaboration

(USC Thesis Other)

Collaborative social networks in student affairs: an exploration of the outcomes and strategies associated with cross‐institutional collaboration

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content
i

COLLABORATIVE SOCIAL NETWORKS IN STUDENT AFFAIRS: AN EXPLORATION
OF THE OUTCOMES AND STRATEGIES ASSOCIATED WITH CROSS-INSTITUTIONAL
COLLABORATION

by

Sean Gehrke

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
(URBAN EDUCATION POLICY)

August 2015

Copyright 2015 Sean Gehrke

ii
Acknowledgements
The journey to earn my Ph.D. has been supported along the way by many important
people in my life. To begin, I have had the pleasure of working alongside and learning from an
outstanding scholar in higher education over these past four years. Dr. Adrianna Kezar has
provided a perfect combination of challenge and support to help me through the process of
developing as a scholar and researcher. I have valued your candor, mentorship, and friendship
along the way as you showed me what it means to be a professional and scholar of integrity and
determination. I will always be proud of what I accomplished with you and what I learned from
you.
I was also fortunate to work with other outstanding faculty members through this
dissertation process. Dr. Darnell Cole, thank you for supporting me in my research all along the
way and for always being a sounding board for me to process through my ideas. Dr. Patricia
Tobey, thank you for helping me remain connected to student affairs through teaching and for
always being a champion of my ideas and endeavors. Dr. Tom Valente, thank you for giving me
the tools to explore and utilize social network analysis, both for this dissertation and for my work
that lies ahead. You have all offered me continual support through the dissertation process, and
for that I will always be grateful.
I have had the privilege and honor to learn alongside fantastic scholars through the USC
Ph.D. program, and my experience here would not have been nearly as fulfilling without the
friendships I developed, especially in my cohort. To the PICCLES crew – Andrea Bingham,
Sophie Hiss, and Kristen Fong – thank you for the laughs and support we shared over many
hours of Korean food, boba, and fun times. Julia Duncheon, Dave Knight, Elena Son, Stephani
Wrabel, and Sable Manson – thank you for all the good times throughout the program, both here

iii
in LA and at our conferences. And Dan Maxey, it wouldn’t have been the same journey without
you along the way with me. I have valued having you as a colleague and friend through this
Ph.D. experience. I will value everything I learned from all of you in these four years at USC and
look forward to continuing to strengthen our friendships.
This dissertation reflects the passion I have for my original professional calling in student
affairs. While this list could be much longer, I am compelled to acknowledge the friendship and
mentorship I have received from so many individuals in student affairs over the years. My
Maryland Terrapins – notably Marlena Love, Paige Haber-Curran, Seth Christman, Darren
Pierre, Dan Tillapaugh, Julie Own, John Garland, John Dugan, Wendy Wagner, and Susan
Komives – have had such an impact on who I am as a professional and a person, and I would not
be the professional I am today without them. And my Whitman colleagues – especially Nancy
Tavelli, Chuck Cleveland, and Brian Dohe – pushed me to grow and allowed me to thrive in my
work as a student affairs professional during my time there. Thanks to all of you for grounding
me in this profession and for your continued friendship.
I could not have completed this journey without the love and support of my families.
Andrene, Tim, Thomas, and Drew – thank you for accepting me so quickly as a member of your
family and for constantly encouraging me on this journey. To my sister, Annie, your strength and
perseverance inspire me in all I do, and I so value our relationship as we continue to grow older
and closer. To my parents, Robert and Dorothy, I wouldn’t be who I am today without your
guidance, your support, your love, and your constant encouragement to strive to be the best
person I can be in everything I do. I am the scholar, professional, brother, son, husband, and
person I am today because of all that I learned from you. I love you both very much, and this
degree is honor of both you.

iv
Finally, nearly eight years ago I began another journey with my wife, Colleen. You have
always been such a rock of support for me and an encouraging force in all that I and we have
done. You are the love of my life, and I look back on these past four years in Los Angeles as
ones of tremendous growth for us and our lives together. I cannot wait to start the next chapter
and look forward to all the journeys to come with you by my side. I love you, boo.

v
TABLE OF CONTENTS

ACKNOWLEDGEMENTS ......................................................................................................... ii
TABLE OF CONTENTS ..............................................................................................................v
LIST OF TABLES & FIGURES ............................................................................................... vii
ABSTRACT ................................................................................................................................ viii
CHAPTER 1 – An Examination of Collaborative Social Networks in Student Affairs ..........1
1.1 Introduction ..........................................................................................................................1
1.2 Literature Review.................................................................................................................4
Social Networks in Organizations ....................................................................................5
Social Network Research in Education ...........................................................................11
1.3 Methods..............................................................................................................................12
Data Collection & Sample ..............................................................................................12
Variables .........................................................................................................................14
Data Analysis ..................................................................................................................18
Limitations ......................................................................................................................19
1.4 Findings..............................................................................................................................20
SA Professionals’ Collaborative Social Networks .........................................................20
Predictors of Cross-Boundary Collaboration in SA........................................................22
1.5 Discussion ..........................................................................................................................25
Behavioral vs. Structural Factors in Cross-Boundary Collaboration ..............................26
Social Identities and Collaboration .................................................................................31
Implications.....................................................................................................................32
1.6 Conclusion .........................................................................................................................35
1.7 Appendix 1A ......................................................................................................................36
CHAPTER 2 – Collaborative Social Networks and Student Affairs Competencies:
The Strength of Strong Ties ........................................................................................................37
2.1 Introduction ........................................................................................................................37
2.2 Research on Student Affairs Competencies and Collaboration in Higher Education ......39
Student Affairs Competencies ........................................................................................40
Collaboration in Higher Education .................................................................................43
2.3 Conceptual and Theoretical Framework: Social Network Theory ....................................44
2.4 Methods..............................................................................................................................47
Data Collection & Sample ..............................................................................................47
Variables .........................................................................................................................49
Data Analysis ..................................................................................................................54
Limitations ......................................................................................................................55

vi
2.5 Findings..............................................................................................................................57
Professional Competencies by Rank, Functional Unit, and Experience .........................57
Random Effects Models for Student Affairs Competencies ...........................................59
2.6 Discussion ..........................................................................................................................61
Strength of Strong Ties and Other Network Indicators ..................................................62
Competencies and Professional Characteristics ..............................................................64
Implications for Practice .................................................................................................66
Implications for Research ...............................................................................................68
2.7 Conclusion .........................................................................................................................69
2.8 Appendix 2A ......................................................................................................................70
CHAPTER 3 – Navigating the Rapids: Network Behaviors and Strategies of Systemic
Leaders in Three Collaborative Social Networks in Student Affairs......................................72
3.1 Introduction ........................................................................................................................72
3.2 Theoretical Frameworks ...................................................................................................75
Systemic Leadership .......................................................................................................75
Social Network Theory ...................................................................................................77
3.3 Leadership in Student Affairs ............................................................................................82
3.4 Methods..............................................................................................................................84
Data Collection & Sample ..............................................................................................85
Variables for Quantitative Analysis ................................................................................87
Analyses ..........................................................................................................................90
Limitations ......................................................................................................................92
3.5 Findings..............................................................................................................................93
Predictors of Systemic Leadership in SA Collaborative Networks ................................95
Strategies of Systemic Leaders .......................................................................................97
A Profile of a Systemic Leader .....................................................................................103
3.6 Discussion ........................................................................................................................106
Behaviors and Strategies for Systemic Leadership .......................................................106
Implications...................................................................................................................110
3.7 Conclusion .......................................................................................................................113
3.8 Appendix 3A ....................................................................................................................115
REFERENCES ...........................................................................................................................116

vii
LIST OF TABLES & FIGURES

CHAPTER 1
Table 1.1 Key Ego Network Measures for Organizational Research .................................10
Table 1.2 Descriptive Statistics for Student Affairs Professionals’ Collaborative
Social Networks ..................................................................................................21
Table 1.3 Random Effects Models Predicting Cross-Departmental and Cross-
Institutional Collaboration among Student Affairs Professionals ......................23
Appendix 1A Descriptive Statistics for Other Variables in this Study .....................................36

CHAPTER 2
Table 2.1 Composite Scales and Descriptive Statistics for Dependent Variables ..............51
Table 2.2 Relationships of SA Professionals Perceived Competencies to Professional
Rank, Functional Unit, and Years of Experience in Higher Education ..............58
Table 2.3 Random Effects Models Predicting Student Affairs Administrators’
Perceived Competencies .....................................................................................60
Appendix 2A Descriptive Statistics for Other Variables in this Study .....................................70

CHAPTER 3
Table 3.1 Social Network Theory and Systemic Leadership ...............................................81
Table 3.2 Response Rates for Study of Three Collaborative Social Networks in
Student Affairs .....................................................................................................87
Table 3.3 Descriptive Statistics for Student Affairs Professionals’ Collaborative
Social Networks from Three Student Affairs Divisions.......................................89
Figure 3.1 Collaborative Social Networks among Student Affairs Professionals at Three
Mid-Size Masters Universities ............................................................................94
Table 3.4 Random Effects Models Predicting Leadership Measures in Three
Collaborative Student Affairs Networks .............................................................96
Figure 3.2 Model of Network Behaviors and Strategies for Systemic Leadership
in Student Affairs Collaborative Networks .......................................................108
Appendix 3A Descriptive Statistics for Other Variables in this Study ...................................115

viii
ABSTRACT
The research in this dissertation examines cross-institutional collaboration among student affairs
administrators utilizing social network analysis as a theoretical and methodological framework.
Student affairs professionals are increasingly being called to collaborate across organizational
boundaries, yet the extent to which they engage in these practices and the outcomes and
strategies associated with these practices is under-studied in the student affairs and higher
education literature. Social network analysis provides a useful theoretical and methodological
framework for examining collaboration, as well as other practices, in student affairs.
The three chapters in this dissertation are each written as standalone studies utilizing data from
one of two surveys. I designed these surveys to examine collaborative social networks among
student affairs administrators. The first survey was administered to a large, representative sample
of student affairs administrators and focused on these professionals’ collaborative ego networks.
The second survey was administered to professionals in three divisions of student affairs at
comparable higher education institutions and was designed to capture data about the
collaboration network among the professionals in each of these institutions.
In Chapter 1, I explore the collaborative social networks of student affairs professionals and
identify network behaviors associated with cross-boundary collaboration, both across student
affairs departments and with others outside of student affairs. The student affairs professionals in
this study engage in collaboration with other student affairs departments more than with others
outside of their student affairs divisions. Additionally, they tend to collaborate in closed
networks and rarely collaborate with faculty. Findings from the random effects models reveal
that collaborating in closed networks and with those who are physically nearby on campus is

ix
negatively associated with cross-boundary collaboration, and that student affairs professionals
are drawn to collaborate more with others who share their gender and racial/ethnic identity.
In Chapter 2, I examine the extent to which student affairs professionals’ collaborative network
behaviors are associated with organizationally-related student affairs competencies, as well as
trends in student affairs professionals’ perceived proficiencies in these areas. The findings from
this study reveal that most of the variance in perceived proficiency in these competencies is due
to individual characteristics, but collaborative behaviors, specifically developing strong ties to
one’s collaborators, is significantly associated with proficiency in these competency areas.
In Chapter 3, I utilize network analysis to measure systemic leadership among student affairs
professionals from three institutions and examine collaborative behaviors and strategies
employed by systemic leaders on these campuses. The resulting model informed by the analyses
in this study highlights several collaborative network behaviors – including unconstrained
personal networks, stronger ties, and less homophily based on propinquity – as well as strategies
– including a big-picture orientation, openness to new ideas, political savvy, and collaborating
with a student-focus – which provide a roadmap of sorts for student affairs professionals who
seek to practice and foster systemic leadership.
Together, these three studies highlight the important contributions that can be made to the
student affairs and higher education literature by utilizing social network analysis as a theoretical
and methodological framework. A social network paradigm holds the potential to allow student
affairs professionals to think about their work in new ways and to examine student affairs
practice from a new perspective, allowing the field to reframe important issues and conceptualize
new approaches to understanding and engaging in its work.

1
CHAPTER 1 – An Examination of Collaborative Social Networks in Student Affairs
1.1 Introduction
Higher education faces an increasing set of challenges and uncertainties. Withering
budgets, increasing calls for accountability, diversifying student bodies, increased demand,
safety and mental health concerns, globalization, and disruptive communication technology are
just some of the factors that are affecting higher education institutions (Kuk, Banning, & Amey,
2010; Task Force, 2010; Woodard, Love, & Komives, 2000). Amidst these challenges, scholars
have likened the current context of higher education to permanent white water, in which
conditions are constantly changing and complex problems are the norm (Manning et al., 2006;
Task Force, 2010). Scholars contend that responding to these conditions requires higher
education institutions that are flexible and adaptive, with professionals who engage in
relationships that cut across organizational boundaries and who understand the interconnectivity
of all functions and individuals within the organization (Allen & Cherrey, 2000; Kuk et al., 2010;
Love & Estanek, 2004).
In the midst of these turbulent times, critics are directing increasing scrutiny toward the
proliferation of non-academic administrators on college campuses, including student affairs (SA)
professionals, suggesting that higher education institutions are misguided in their proliferation of
administrators whose purview is outside of the college classroom (Leslie & Rhoades, 1995;
Macchio, 2012; Marcus, 2014). In response, student affairs leaders suggest that the increasing
role played by student affairs administrators comes as a result of the unique challenges and
complexity facing higher education (Kerr, Porterfield, Davenport, & Roberts, 2014). They
further contend that despite the relatively small portion of university budgets directed toward
student services (Kerr et al., 2014), the role these administrators play related to college student

2
success and organizational functions is crucial to colleges and universities (Macchio, 2012;
Smith, 2013; Task Force, 2010).
Concurrently with this increased scrutiny, the SA professional community continues to
engage in dialogue about the purpose of the profession and the roles SA administrators should
play on their campuses. Over the past two decades, leading professional associations have
sponsored reports bolstered by research that call on SA administrators to collaborate and work
across functional siloes while emphasizing student learning as the main focus of the profession.
The message conveyed in these philosophical statements and calls-to-action suggest that in order
to meet the challenges of uncertain times, SA professionals must coordinate services, form
partnerships, and cultivate relationships across hierarchical boundaries to enhance learning for all
students (ACPA, 1994; Joint Task Force, 1998; Keeling, 2004; Task Force, 2010). These calls
for collaboration are echoed by scholars in the field who increasingly argue for SA professionals
to reach beyond their functional siloes in order to meet the demands of these turbulent times
(Kuk et al., 2010; Love & Estanek, 2004; Manning et al., 2006; Winston, Creamer, & Miller,
2001). As Kuk and colleagues (2010) aptly state:
Many SA organizations have outgrown the utility of functional hierarchical structures.
For the most part, SA organizations are increasingly complex, multifocused organizations
that need to be able to respond to students, parents, the external community, and other
constituent groups quickly and from multiple perspectives simultaneously. Collaboration
and sharing information, resources, and expertise are critical components of everyday
life. It is no longer effective to stay within one’s functional unit and expect to meet the
demands and challenges that are being presented (p. 22, emphasis added).
Despite these calls for collaboration that cuts across organizational boundaries in higher
education institutions, few are examining the extent to which SA professionals are engaging in
this type of behavior. Research on the practices of SA professionals is generally hard to come by,
as most inquiry in the field is focused on student outcomes as opposed to effective practice in
administration (Kuk et al., 2010). Research on collaboration in higher education suggests a

3
plethora of outcomes that result from effective partnerships across institutional boundaries,
including better coordination of services, gains in student engagement and learning, and
identification of best practices such as learning communities (see Kezar & Gehrke, in press;
Whitt, 2011). However, much of this research focuses on intentionally-designed programs that
are meant to partner academic and student affairs colleagues in joint efforts and tends to focus on
exemplary efforts or best practices. Furthermore, these types of collaboration are not widespread
in the academy (Kezar & Lester, 2009).
As the SA profession faces increased scrutiny for the resources dedicated to employing
these administrators, it is important to better understand and document the collaborative
behaviors of these professionals. Given the positive educational and institutional benefits that
come through collaboration, documenting the extent to which SA professionals are collaborating
and the nature of these relationships in their institution is integral to highlighting behaviors
associated with this type of work and advance the scholarship on collaboration in higher
education. The purpose of this paper is to describe the collaborative behavior of SA professionals
in their institutions and identify behaviors that are associated with cross-boundary collaboration.
This inquiry meets this purpose by answering the following research questions:
1. How are student affairs administrators’ collaborative social networks structured?

2. To what extent are student affairs administrators’ collaborative social network behaviors
associated with collaboration a) across student affairs departmental boundaries, and b)
across broader institutional functional boundaries after controlling for professional
characteristics, personal demographics, and institutional characteristics?
These questions address the gaps in the literature pertaining to the extent to which SA
administrators collaborate across their institutions. By examining the structure of SA
administrators’ collaborative social networks, we can begin to understand whether or not SA
professionals are heeding the profession’s call to collaborate across organizational boundaries.

4
Additionally, identifying behaviors among SA professionals that are associated with cross-
organizational collaboration can provide guidance to other administrators in higher education
who are seeking to break down boundaries and work across their institutions.
1.2 Literature Review
This research is primarily informed by the literature pertaining to social network theory
and analysis (SNA). SNA is an appropriate theoretical and empirical lens for this inquiry because
it moves beyond examining the traits and behaviors of individuals to focus on the nature of
relationships between and among individuals (Borgatti & Ofem, 2010); an examination of
collaboration must inherently examine relationships between and among individuals since
collaborating requires social interaction and cooperation. Also, SNA is underutilized in the
higher education literature (Kezar, 2014), yet its focus on relationships in organizations is a good
fit for the relational nature and approach to work in student affairs. The use of SNA in this paper,
therefore, serves not only as a means to better understand and operationalize collaborative
behavior in student affairs, but it also serves to illustrate the utility of SNA in studies of
administrative behavior in higher education and student affairs.
SNA emphasizes the concept of a network, which “consists of a set of nodes or actors,
along with a set of ties of a single type that connect the nodes” (Borgatti & Ofem, p. 19). Social
network research can either focus on individual networks surrounding an individual (ego
network analysis, in which ego is the person at the center of the network surrounded by the
individuals ego is connected to, referred to as alters) or an entire system of relationships in an
organization (whole network analysis – Daly, 2010a). This study relies on the structures of ego
networks as the main unit of analysis, allowing me to gather information from a broad, random
sample of individuals to make generalizations about a larger population (Valente, 2010). I draw

5
by-and-large from research on social networks in organizations and applied to K-12 education
contexts to inform the study since SNA is underutilized in higher education (Kezar, 2014). I
begin by reviewing the literature regarding social networks in organizations, drawing special
attention to inter-unit networks, followed by descriptions of pertinent constructs in network
research for this study.
Social Networks in Organizations
Social network theory and research exhibits four core principles that distinguish it from
rival theoretical and empirical approaches to studying organizations (Balkundi & Kilduff, 2006;
Kilduff & Brass, 2010; Kilduff, Tsai, & Hanke, 2006). First, social network theory emphasizes
that the nature of relationships in organizations are just as, if not more important to
understanding organizational effectiveness (Kilduff & Brass, 2010). Second, “human behavior is
seen as embedded in networks of interpersonal relationships” of friendship and acquaintance
(Balkundi & Kilduff, 2006, p. 420). Interactions in organizations are not only dictated by formal
relationships defined by organizational hierarchy but are influenced by personal affinity and
prior relationships; relationships form for both formal and informal reasons (a phenomenon
known as multiplexity – Kadushin, 2012). Third, social network theory emphasizes the
structured patterning of social relations and the importance of understanding the interplay
between individuals’ positions in the larger structure and the overall structure of individuals in
organizations (Balkundi & Kilduff, 2006; Monge & Contractor, 2003). Finally, social network
research emphasizes the utility of social network connections for individuals and organizations:
“social networks provide the opportunities and constraints that affect outcomes of importance to
individuals and groups” (Kilduff & Brass, p. 328).

6
The social network literature is replete with research examining consequences or
outcomes associated with social networks in organizations. One area in particular that is most
pertinent to this study is the literature that examines the influence of social networks on
performance in organizations (Borgatti & Foster, 2003; Brass et al., 2004). As Brass and
colleagues (2004) aptly state: “As is the case with interdependent tasks in organizations,
relationships with others affect performance, especially if those relationships involve the ability
to acquire necessary information and expertise” (p. 799). This research focuses on a variety of
benefits associated with network ties (reviewed below).
A strand of network research highlights the distinct benefits that come from inter-unit
network ties in organizations, including better coordination in times of crises (Krackhardt &
Stern, 1988), better individual (Cross & Cummings, 2004) and unit performance (Mehra,
Kilduff, & Brass, 2001), higher productivity (Reagans & Zuckerman, 2001), improved unit
efficiency (Reagans, Zuckerman, & McEvily, 2004) and reduced inter-unit conflict (Nelson,
1989). The extents to which groups and organizations encourage and are designed to facilitate
inter-unit ties also influence group performance (Pearce & David, 1983). The importance of
inter-unit ties is particularly relevant to this study of collaboration in SA, indicating types of
outcomes that have been associated with cross-boundary collaboration as a way to support such
behavior in SA. I now turn to specific indicators of network behavior identified in the SNA
literature that are relevant to this study of collaborative social networks in SA. These are
homophily, density, and strength of ties (summarized in Table 1.1).
Homophily. Homophily refers to the extent to which individuals form relationships with
others who are similar to themselves (Lazarsfeld & Merton, 1954; McPherson, Smith-Lovin, &
Cook, 2001). For organizational research, homophily pertaining to functional unit is particularly

7
useful for understanding the extent to which individuals are forming relationships across
boundaries in organization (Kilduff & Tsai, 2003). Homophilous networks in organizations
based on functional unit and department, in which individuals in one’s ego network share the
same departmental affiliation, suggest that information flowing to other departments will be
limited because connections are not being made with other members in different functional areas.
Beyond access to information, research also shows that more diversity in one’s network based on
departmental affiliation is positively linked to productivity and performance measures (Papa,
1990). Homophily in one’s network based on departmental affiliation is a key measure to
consider when studying collaboration in higher education institutions, as it is an indicator of the
extent to which SA professionals are reaching across organizational boundaries.
Homophily related to other personal and professional characteristics is also important for
this study. Research identifies how relationships formed around other organizational
characteristics, such as physical propinquity, can dictate who is connected to whom; formal
structures in organizations place individuals in specific physical locations, which can restrict
interaction with some in the organization while facilitating interactions with others (Borgatti &
Cross, 2003; Brass et al., 2004; Festinger, Schachter, & Back, 1950; Kadushin, 2012; Monge &
Contractor, 2003). Homophily research also focuses on personal characteristics, such as gender
(Brass, 1985; Ibarra, 1992) and race (DiMaggio & Garip, 2012). These scholars note the
tendency of individuals to associate with others who share their identities, while also suggesting
how doing so can influence individuals’ power and rewards in organizations, largely connected
to societal structures of power. For example, homophilous networks were much more likely to
lead to promotions for men than women in Brass’s (1985) study of differences in men’s and
women’s networks in organizations. The extent to which SA professionals exhibit homophily

8
can therefore indicate how their collaborative relationships may influence their effectiveness in
their organizations.
Density. Density in ego networks refers to the extent to which individuals’ personal
networks are comprised of people who are also connected to one other (Balkundi & Kilduff,
2006; Kezar, 2014; Valente, 2010). Someone whose connections are not also connected to one
another has a less dense network, whereas an ego network comprised of individuals who are
more connected to one another is a denser network (i.e., more closure). Denser ego networks in
which members are predominantly connected to one another tend to share similar attitudes and
beliefs and exhibit more trust but are closed off to the rest of the organization (Balkundi &
Kilduff, 2006; Coleman, 1988), suggesting difficulty accessing new information from others in
the organization who are not part of the network. However, research has also shown that working
groups (e.g., departments or offices) with more closure outperform groups with less closure
(Mehra, Dixon, Brass, & Robertson, 2006), and denser networks have also been associated with
greater ease of information and knowledge transfer (Fritsch & Kauffeld-Monz, 2008). Ego
network density is relevant to understanding collaborative behavior among SA professionals
because it indicates the extent to which someone may have access to other parts of the institution
and one’s potential ability to collaborate beyond one’s local network.
Strength of ties. Finally, tie strength describes the quality (based on frequency of
interaction and perception of the relationship) of connections between individuals and their
network connections (Brass & Krackhardt, 1999; Gillis, 2008; Krackhardt, 1992; Valente, 2010).
Information is transmitted across organizations more freely through strong ties, in which
individuals have more frequent interaction and trust with their network connections (Granovetter,
1973, 1982; Hoppe & Reinelt, 2010); altruism and interpersonal citizenship have been associated

9
with stronger ties in organizational research (Bowler & Brass, 2006; Settoon & Mossholder,
2002; Sparrowe, Liden, Wayne, & Kraimer, 2001). However, weak ties can serve as bridges of
communication across organizations, connecting disparate groups and facilitating diffusion of
information, practice, and innovation (Burt, 1992). They can also facilitate information transfer
particularly in dynamic and changing environments (Krackhardt, 1990) and have been associated
with more creativity in organizations (Brass, 1995; Perry-Smith & Shalley, 2003). In the context
of SA administrators’ collaborative networks, the strength of collaborative relationships can
serve as an indicator of the nature of collaborative relationships and the kind of trust, interaction,
and access to information that can come through these connections.
Individual and organizational antecedents of network measures. Beyond the relevant
network measures reviewed above, SNA research suggests the necessity of considering other
factors when engaging in network research. Abundant research on social networks suggests that
network behavior varies across a host of individual characteristics (e.g., race, gender),
professional characteristics (e.g., position in organizational hierarchy, tenure at an organization)
and organizational context (Borgatti & Foster, 2003; Bowler & Brass, 2006; Brass, 1985; Brass
et al., 2004; Ibarra, 1992; Kadushin, 2012; McPherson et al., 2001; Mehra et al., 2001;Mehra,
Kilduff, & Brass, 1998; Perry-Smith & Shalley, 2003). Much of the SNA literature pertaining to
individual characteristics examines homophily and the extent to which individuals form
relationships with similar others (Brass et al., 2004). Related to professional characteristics,
informal networks form within an organizational context that has a formal structure, making it
not uncommon for social networks to form based on the formal position and hierarchy of
individuals in the organization (Brass et al., 2004). A related dimension to formal position is
tenure in an organization, which “can be regarded as a proxy for work experience” (Carboni &

10
Table 1.1: Key Ego Network Measures for Organizational Research
EGO NETWORK
MEASURES
DEFINITION INDICATORS
Density The extent to which egos’
network is comprised of alters
who are also connected to one
other
Denser networks: Indicate more closure, trust, and
shared beliefs and values among members; greater
ease of information transfer among network
members; better performance in units with more
closure
Less dense networks: Suggest more access to
information and resources in the rest of the
organization, as denser networks tend to be more
insular and closed off to others throughout an
organization
Strength of Ties The quality (based on
frequency of interaction and
perception of the relationship)
of connections between egos
and their alters
Strong ties: Information passed more freely due to
more frequent interaction and trust; associated with
greater altruism and citizenship in organizations
Weak ties: More access to new information; serve as
bridges or connectors across organizations connecting
disparate groups and facilitating diffusion of
information, practice, and innovation; associated with
more creativity in organizations
Homophily:
Department &
Division
The extent to which ego is
connected to alters from ego’s
department/
division/functional unit in an
organization
Greater homophily: Suggests that access to
information and resources across the organization
will be limited because connections are not being
made with other members in different functional
areas
Less homophily: Suggests interaction across
boundaries and more access to information and
resources across organizations; associated with host
of outcomes including individual and unit
productivity, increased efficiency, and reduced inter-
unit conflict
Homophily:
Propinquity
The extent to which ego is
connected to alters who share
the same physical space or are
in close proximity to ego
Greater homophily: Suggests that individuals are
inhibited by physical structures of the organization
and are not able to make connections beyond their
physical location
Less homophily: Suggests that individuals are able to
form connections across physical boundaries to form
connections with units physically located beyond
their local area
Homophily:
Race & Gender

The extent to which ego is
connected to alters who share
the same racial or gender
identity as ego
Greater homophily: Suggests that individuals are
more likely to associate with individuals who share
their same social identities; can inhibit those with
marginalized identities in gaining access to rewards
and other benefits in organizations
Less homophily: Suggests that individuals are able to
form relationships with others who share a variety of
social identities; increased diversity in personal
network can provide greater perspective for problem
solving and job performance

11
Ehrlich, 2013, p. 515) and influences the extent to which an individual is seen as knowledgeable
about organizational functions and connections (Carboni & Ehrlich, 2013; Ibarra & Andrews,
1993; Soda & Zaheer, 2012). Finally, organizational context, including institutional
characteristics, can influence how individuals form relationships within them and should be
considered when trying to understand network behavior (Ostroff, Kinicki, & Tamkins, 2003).
Social Network Research in Education
Most educational research informed by social network theory is in the K-12 sector; social
network analysis and theory is rarely utilized to evaluate organizational dynamics in higher
education (Kezar, 2014), although some studies of leadership suggest the importance of informal
social networks in contributing to change (e.g., Kezar & Lester, 2011). Studies utilizing social
network analysis in higher education examine phenomena such as career pathways to the senior
student affairs position (Biddix, 2011); campus’ disaster preparedness (Stein, Vickio, Fogo, &
Abraham, 2007); governance participation and performance of faculty (Dose, 2012); and the role
of social networks relating to student outcomes integration and persistence (Smith, 2011;
Thomas, 2000). Pertinent to this research, Stein and colleagues (2007) found that the university
in their study was not best-structured to handle major crises given the relative isolation of
different units in the campus’ network. This study suggests the benefits that come with interunit
connections, but it only focuses on departments as the unit of analysis, not individual
administrators. With the exception of Biddix (2011), these studies are also limited as they are
primarily situated within a single institution rather than seeking to understand behavior across
the profession. Additionally, no work to-date has utilized social network theory to better
understand collaboration and boundary-crossing relationships outcomes among administrators.

12
The work in education that most closely relates to this study examines informal networks
of teachers and staff within schools (Daly, 2010b; Moolenar, 2012). Informal networks within
schools often do not conform to their formal structure and hierarchy of schools, whether based
on formal position (administration versus staff) or administrative area (teacher versus coach), as
teachers and other staff form relationships based on expertise and personal similarity not
ascribed by organizational structure (Moolenar, 2012). Research has found relationships between
teacher and staff network structures and a host of outcomes, such as positive associations
between teachers’ strong ties and student performance (e.g., Leana & Pil, 2006); teachers’ sub-
group ties and teaching practice (e.g., Penuel, Frank, & Krause, 2010); dense teacher networks,
trust among teachers, and openness to new practices (Moolenar & Sleegers, 2010); and nature of
ties and success in school reform (Coburn, Choi, & Mata, 2010; Coburn, Russell, Kaufman, &
Stein, 2012; Penuel, Sun, Frank, & Gallagher, 2012). The use of social network analysis in K-12
organizational contexts to identify individual and organizational outcomes suggests utility for
these frameworks and methods for a student affairs context.
1.3 Methods
Data Collection & Sample
Data for this study were collected through the Collaborative Social Networks in Student
Affair (CSNSA) survey. This survey was distributed to members of the American College
Personnel Association (ACPA-College Student Educators International) who identified as
professional staff working in a college or university in the United States in the summer of 2014.
ACPA was selected due to its national representation of SA administrators from all position
levels and functional areas in higher education institutions.

13
The purpose of the CSNSA survey was to gather data pertaining to SA professionals’
collaborative networks in their institutions. It contained questions asking participants to identify
up to seven individuals with whom they collaborate in their work, followed by questions about
each individual they identified. Questions for each collaborator included items assessing the
strength of their relationship, professional and personal characteristics of the collaborator, and
the collaborative relationships of each individual with others in the participants’ ego network.
Additional questions on the survey pertained to personal demographics of participants (e.g.,
racial/ethnic identity, gender identity), professional demographics (e.g., institution, position rank,
functional area, years of experience), and perceived proficiency in several student affairs
competencies. Survey validity was assessed in two primary ways. First, survey design was
informed by the literature on social network theory and prior research, which suggested methods
for obtaining information related to individuals’ ego networks. For example, I utilized a
psychometrically sound scale to assess tie strength between participants and each of their
collaborators (Gillis, 2008). Second, the survey was piloted by eight individuals a) with expertise
in research in higher education and b) who either currently or formerly worked in an
administrative position in SA. I gathered qualitative feedback from each individual regarding the
content and structure of the survey, ensuring both face and content validity of the items on the
survey.
Data were collected over a four-week period in July and August 2014. The target
population for this study is SA administrators. Therefore, ACPA identified individuals from
among their members who were current administrators at a higher education institution in the
United States from all levels in the hierarchy (entry, mid, and senior-level). In order to ensure
generalizability across the sample, individuals who identified as current faculty members,

14
undergraduate and graduate students, ACPA members who work in institutions outside of the
United States, and affiliate members who work for organizations that are not colleges or
universities were excluded from the sample. The initial invitation to participate in the survey was
sent by the ACPA central office to 3,205 participants from 888 institutions who fit this
description, with two follow-up emails sent two weeks apart after the initial invitation.
Participation was incentivized with the option for participants to be entered into a drawing to win
a grand prize of an iPad Mini or one of 10 gift cards to an online retailer. Of the 3,205
individuals who were invited to participate, 913 responded to the survey invitations, indicating a
28.5% initial response rate. Of those who responded, 640 participants completed the entire
survey (70.1% completion rate) from 322 institutions, resulting in a final response rate of 20.0%.
Descriptive statistics pertaining to individual demographics, professional characteristics, and
institutional characteristics for the sample are listed in Appendix A1.
Variables
Focal variables. The focal variables for this study are those pertaining to social network
measures of participants’ collaborative networks. These include ego network size, density,
average tie strength, homophily based on racial/ethnic identity and gender identity, homophily
based on propinquity (i.e., working in the same building), extent of collaboration with faculty,
and homophily based on department and student affairs. Ego network size is simply the number
of individuals participants identified as collaborators in their work (up to seven). For ego
network density, participants were asked to identify whom each member of their ego networks
collaborated with among the other individuals in the network. Density was calculated as
proportion of actual ties in the network compared to the potential ties in the network (not
including ties to ego), ranging from 0 (least dense) to 1 (most dense). For example, in an ego

15
network of five individuals, the maximum number of ties is 10. If only six of those possible
relationships exist, network density for that ego network would be 0.6.
I assessed tie strength using a composite scale developed by Gillis (2008). Participants
responded to three items for each collaborative relationship they identified assessing their
perceived closeness to that individual (on a five-point Likert-like scale, with 1 = “Very Distant”
and 5 = “Very Close) and the frequency with which they went to that individual for both work-
related and other types of advice (on a five-point Likert-like scale, with 1 = “Rarely” and 5 =
“Often”). I calculated scale scores by calculating the mean of the three items, and the scale was
internally reliable for this sample (α=0.79). The extent that each participant collaborated with
faculty was calculated as the proportion of relationships with faculty members to the total
number of relationships with individuals in the participants’ ego networks.
I calculated homophily as the proportion of ties in participants’ ego networks who shared
a given trait with the participant compared to the total number of ties in the network. Participants
were asked to identify, to the best of their knowledge, whether or not collaborators shared their
gender identity and racial/ethnic identity; these items were used to calculate homophily based on
race/ethnicity and gender identity. Participants were also asked a series of questions identifying
whether each collaborator worked in their department, worked in student affairs, and worked in
their same building. These items were used to calculate homophily based on department, student
affairs, and propinquity. Descriptive statistics and descriptions of each of these focal variables
are listed in Table 1.2.
The variables pertaining to homophily by department and student affairs were
transformed prior to their inclusion as the dependent variables in the random effects models
described below. The purpose of these models in addressing the second research question were

16
to examine the extent to which collaborative network behavior and institutional, personal, and
professional characteristics are associated with cross-departmental and cross-institutional
collaboration. Therefore, the homophily scores by department and student affairs, which
represent the extent to which participants collaborate with others in their departments and within
student affairs, were subtracted from 1 to create dependent variables indicating the extent to
which participants collaborate across departments and across the institution (i.e., with others
outside of student affairs).
Control variables. In addition to the focal variables listed above, other independent
variables were included in random effects models as controls. These include variables pertaining
to institutional characteristics, professional characteristics, and personal demographics.
Organizational contexts influence network behavior (Ostroff et al., 2002), which is why I
included information pertaining to participants’ institutions, including size, Carnegie type,
control (i.e., public vs. private) and perceived institutional climate for collaboration. Participants
were asked to identify their current institution on the survey, and information relating to
institution size, control, and Carnegie type was identified using 2013-2014 Institutional
Postsecondary Education Data Systems (IPEDS) data. I calculated climate for collaboration from
an item on the survey asking participants to assess the climate for collaboration in their
institution on a five-point Likert-like scale, with 1 = “Completely Unsupportive of
Collaboration” and 5 = “Completely Supportive of Collaboration.” When more than one
individual from an institution participated in the survey, I aggregated these scores to the
institutional level to signify the climate; when only one individual from an institution
participated in the survey, I utilized that participants’ perception of the climate to represent the
institutional climate.

17
Individuals’ network behavior also varies by personal and professional characteristics
(Borgatti & Foster, 2003; Bowler & Brass, 2006; Brass, 1985; Brass et al., 2004; Ibarra, 1992;
Kadushin, 2012; McPherson et al., 2001; Mehra et al., 1998; 2001; Perry-Smith & Shalley,
2003). Therefore, I also included several individual-level control variables to account for these
characteristics. Professional control variables include participants’ rank in their institution (e.g.,
entry-level, mid-level, senior-level, and senior student affairs officer), functional area, and years
of experience in both higher education and at their current institution. Participants were asked to
identify the functional area in which they worked from a list of 22 potential functional areas,
including an “Other” option in which they were allowed to enter an area not listed from among
the other 21 options. For the sake of parsimony in the analyses, these functional areas were
collapsed into nine umbrella functional areas which perform similar functions on college and
university campuses: activities/student union/recreational activities; advising/career
services/academic support; administrative leadership/generalist; admissions/orientation;
multicultural/diversity programs; wellness/health programs; assessment; residence life/student
conduct; and other student affairs areas (representing other areas that do not fit in the other
categories). Personal characteristics included as control variables in the analyses below include
racial/ethnic identity and gender identity. Again, descriptive statistics for these variables are
listed in Appendix A1.
Data Analysis
Prior to running the models for this paper, I explored the data to identify any patterns in
missing data. Fewer than 10% of participants in the study exhibited missing data, with fewer
than 1% of individual values missing in the dataset, which are well within the suggested range of
missing values to be included in the analyses (Newton & Rudestam, 1999). Most of the missing

18
values pertained to institutional variables, as some participants did not identify the institution in
which they worked, with few cases missing personal or professional demographic variables.
These were accounted for by listwise deletion in the random effects models below. Missing
continuous variables were imputed using the expected maximization (EM) algorithm, which
utilizes maximum likelihood techniques to impute missing values.
In order to answer the first research question pertaining to the structure of SA
professionals’ collaborative networks, I calculated descriptive statistics for each participant’s
collaborative networks pertaining to the focal network variables described above. Given that
each participant answered questions about each individual in their network, I aggregated these
measures for each participant based on the individual relationships described resulting in
measures of their ego networks. Two of the network measures pertain to homophily based on
social identity (e.g., gender identity, racial/ethnic identity). It is difficult to interpret these two
variables out of context. So, in order to better understand the dynamics of these two variables, I
utilized ANOVA to identify whether or not homophily based on gender identity was
significantly related to participants’ gender identity, as well as whether or not homophily based
on racial/ethnic identity was significantly related to participants’ racial/ethnic identity.
In order to answer the second research question pertaining to variables associated with
collaboration outside of one’s department and outside of student affairs, I utilized random effects
modeling. The sample exhibits clustering, as some participants work in the same institutions as
other participants in the sample. Network behavior is influenced by this kind of clustering,
violating the assumption of independence (Valente, 2010), which can result in biased standard
errors if calculated using ordinary least squares regression (Rabe-Hesketh & Skrondal, 2012).
Therefore, network analysts recommend using multilevel random effects models to account for

19
this nonindependence. I utilized two multilevel models to examine the associations of
collaborative network behaviors, institutional characteristics, and personal and professional
characteristics with two dependent variables: collaboration across departments and collaboration
outside of student affairs. I entered the variables into the models in three blocks in order to
examine the variance accounted for by each block. The first block pertained to institutional
characteristics, followed by the next block pertaining to personal and professional characteristics.
I calculated the change in variance (i.e., R
2
) for each block prior to entering the third block with
the focal network variables. This allowed me to identify the amount of unique variance
accounted for by these variables after controlling for the institutional and personal/professional
characteristics. I also standardized all continuous variables prior to including them in the models
so that their coefficients could be interpreted as effects sizes.
Limitations
As with all social science research, this study exhibits several limitations. The first
limitation pertains to the cross-sectional nature of the data, which precludes me from making
causal inferences about network behavior. Utilizing longitudinal and/or experimental study
designs could have allowed me to make these claims. However, given the focus of studying
collaboration among SA administrators in their natural institutional settings, it was not possible
to design an experimental condition to control for all of the various individual and institutional
factors that could influence this behavior. A longitudinal approach to studying collaboration
networks could allow me to make claims about the dynamic nature of these networks over time
(Borgatti & Halgin, 2011; Monge & Contractor, 2003), and this is a future direction for research
I describe below. The second primary limitation is the self-reporting nature of survey
instruments. Participants may not be able accurately assess their behavior and that of others they

20
identify. While criticisms abound about the potential for self-report surveys to produce biased
estimates of organizational behavior (see Podsakoff & Organ, 1988; Podsakoff, Mackenzie, &
Lee, 2003; Spector, 1994), these authors also acknowledge the importance of self-report surveys
in advancing the literature on organizational behavior. Spector (1994) in particular suggests that
the use of self-report questionnaires should be used when appropriate to the research. Given the
goal of this research to gather data from a representative sample of administrators and make
inferences about their collaborative behavior, survey self-report is the most appropriate method.
1.4 Findings
SA Professionals’ Collaborative Social Networks
Descriptive statistics for participants’ collaborative social networks reveal patterns of
collaborative behavior for these administrators (see Table 1.2). When given the option to identify
up to seven potential collaborators, participants on average identified between six and seven.
Participants’ networks are also relatively dense, with more than 60% of potential ties existing in
their networks. Additionally, the relationships formed in these networks skew slightly toward
stronger than weaker ties, on average. Roughly half of participants’ collaborative relationships
are with others from their departments, but a majority of their collaborative relationships are with
others in SA. These participants also indicate that roughly half of the individuals they collaborate
with work in their building. Finally, despite calls for more collaboration between academic
affairs and SA, relatively few collaborative relationships are being formed with faculty.
Participants in this study also exhibit homophily by gender and race, with more than half
of their collaborative relationships formed with individuals who share their same gender identity
and nearly three-quarters of their relationships formed with individuals who share their racial
identity. As mentioned above, these variables are difficult to understand without the context of

21

participants’ own racial/ethnic and gender identities. ANOVA results reveal a significant
relationship between participants’ gender identity and network homophily by gender [F(2, 636) =
124.789, p < .001]. Bonferroni post-hoc statistics reveal that participants who identify as female
exhibit significantly greater homophily based on gender (M =.65, SD = .21) compared to
participants who identify as male (M = .38, SD = .21) or transgender (M = .35, SD = .35).
ANOVA results also identify a significant relationships between participants’ racial/ethnic
identity and network homophily by race/ethnicity [F(4, 631) = 146.28, p < .001]. Bonferroni
post-hoc statistics reveal that White participants exhibit significantly greater homophily by
racial/ethnic identity (M = .79, SD = .22) than Latino (M = .26, SD = .29), Asian (M = .13, SD =
.25), Black (M = .33, SD = .28), and multiracial participants (M = .15, SD = .23). Additionally,
Table 1.2: Descriptive Statistics for Student Affairs Professionals’ Collaborative Social Networks
M (SD) Description
Ego Network Size 6.30 (1.31)
Participants could nominate up to 7 individuals with whom they
collaborate in their work; ranges from 0 (no collaborators) to 7
(7 collaborators)
Ego Network Density 0.63 (0.24)
Proportion of number of connections among ego’s alters to the
total possible number of connections; ranges from 0 (least
dense) to 1 (most dense)
Average Tie Strength 3.15 (0.55)
The average strength of relationships ego has to alters in
network; 5-point scale, with 1 = weak tie and 5 = strong tie
Homophily: Propinquity 0.49 (0.31)
The proportion of ego’s alters from ego’s building; ranges from 0
(no alters from ego’s building) to 1 (all alters from ego’s
building)
Homophily: Gender 0.55 (0.25)
The proportion of ego’s alters who share ego’s gender identity;
ranges from 0 (no alters share gender identity with ego) to 1 (all
alters share gender identity with ego)
Homophily: Race/Ethnicity 0.69 (0.32)
The proportion of ego’s alters who share ego’s race/ethnicity;
ranges from 0 (no alters share race/ethnicity with ego) to 1 (all
alters share race/ethnicity with ego)
Collaborate with Faculty 0.04 (0.11)
The proportion of ego’s alters who identify as faculty; ranges
from 0 (no faculty in ego network) to 1 (all faculty in ego
network)
Homophily: Department 0.53 (0.32)
The proportion of ego’s alters from ego’s department; ranges
from 0 (no alters from ego’s department) to 1 (all alters from
ego’s department)
Homophily: Student Affairs 0.70 (0.28)
The proportion of ego’s alters who also work in student affairs;
ranges from 0 (no alters from student affairs) to 1 (all alters
from student affairs)

22
Black participants also exhibit significantly greater homophily by racial/ethnic identity than
Asian and multiracial participants.
Predictors of Cross-Boundary Collaboration in SA
Cross-Department Collaboration. Results of the two random effects models identify
several collaborative network behaviors associated with cross-boundary collaboration in SA, as
well as other associations with control variables (see Table 1.3). Beginning first with the model
for collaboration outside of one’s department, several collaborative network variables are
significant predictors after controlling for other variables. Network density, mean tie strength,
and homophily by building are all significantly and negatively associated with collaboration
outside of one’s department. In other words, participants whose networks are more dense,
contain weaker ties, and involve more others who work in the same building as the participant
are expected to exhibit less collaboration outside of one’s department. Homophily by gender is a
significant positive predictor of the dependent variable, suggesting that collaborating more with
others who share one’s gender identity is associated with more collaboration beyond one’s
department.
Beyond network predictors, several personal and professional characteristics are
associated with cross-departmental collaboration. Participants who identify as male or
transgender indicate more cross-department collaborative relationships compared to those who
identify as female. Additionally, rank in the organizational hierarchy is associated with cross-
department collaboration, as mid-level, senior-level, and senior student affairs officers all exhibit
more cross-departmental collaboration compared to entry-level professionals. One’s functional
unit is also associated with cross-departmental collaboration, as participants who work in
residence life and student conduct departments exhibit significantly less collaboration across

23
Table 1.3: Random Effects Models Predicting Cross-Departmental and Cross-Institutional
Collaboration among Student Affairs Professionals
Cross-Departmental
Collaboration
Cross-Institutional
Collaboration
R
2
β SE R
2
β SE
Institutional Characteristics .03 .01
Institution Size: < 1,000
a
.11 .30 .26 .35
Institution Size: 1,000-4,999
a
.29 .15 .31 .18
Institution Size: 5,000-9,999
a
.36** .13 .13 .15
Institution Size: 10,000-19,999
a
.28** .09 .16 .11
Carnegie: Masters
b
-.08 .10 -.15 .11
Carnegie: Baccalaureate
b
.06 .15 -.24 .17
Carnegie: Associates
b
-.03 .16 -.06 .18
Carnegie: Other
b
.36 .20 -.01 .23
Control: Public .12 .09 .10 .10
Institutional Climate for Collaboration .01 .03 -.06 .04
Individual Characteristics .14 .19
Male
c
.19* .08 -.06 .09
Transgender
c
.64* .30 .25 .35
Latino/Hispanic
d
.02 .19 .32 .22
Asian/Asian American
d
-.08 .21 .40 .24
Black/African American
d
-.04 .13 .46** .15
Multiracial
d
-.25 .18 .19 .21
Mid-level Professional
e
.30** .09 .16 .11
Senior-Level Professional
e
.56*** .14 .24 .16
Senior Student Affairs Officer
e
.52** .20 .61** .23
Functional Area: Activities/Union/Rec
f
.63*** .10 .18 .11
Functional Area:
Advising/Careers/Academic Support
f

.46*** .10

.87*** .11
Functional Area: Administrative
Leadership/Generalist
f

.42** .13

.33* .15
Functional Area: Admissions/Orientation
f
.61*** .15 .41* .17
Functional Area: Multicultural/ Diversity
Programs
f

.70*** .15

.43* .18
Functional Area: Wellness/Health
f
.62** .22 -.08 .26
Functional Area: Assessment
f
.89*** .19 .23 .22
Functional Area: Other
f
.91*** .25 .85** .30
Years at Institution -.04 .04 .01 .05
Years in Higher Education .04 .06 .14* .07
Collaborative Network Variables .32 .09
Ego Network Size .05 .03 .04 .04
Ego Network Density -.23*** .03 -.19*** .04
Ego Network Mean Tie Strength -.07* .03 .00 .04
Ego Network Homophily: Propinquity -.44*** .03 -.13** .04
Ego Network Homophily: Gender .10** .04 -.06 .04
Ego Network Homophily: Race/Ethnicity .00 .05 .18** .05
TOTAL R
2
.47 .29
NOTES:* p < .05; ** p < .01; *** p < .001; Reference Categories for Categorical Variables:
a
20,000+;
b
Doctoral
Institutions;
c
Female;
d
White;
e
Entry-level;
f
Residence Life/Student

24
departments when compared to all other functional units in this study. Finally, institutional size
is a significant predictor of cross-department collaboration in SA; specifically, participants who
work at mid-size institutions (5,000-9,999 and 10,000-19,999 enrollment) exhibit more cross-
departmental collaboration compared to their peers from larger institutions (20,000+ enrollment).
Most of the variance in the dependent variable for this model is explained by the collaborative
network variables (32%), followed by individual (14%) and institutional characteristics (3%).
Cross-Institutional Collaboration. Turning to collaboration with others outside of
student affairs, several network variables are significant predictors. Network density and
homophily by building are significantly and negatively associated with collaboration outside of
student affairs, again suggesting that participants who exhibit more dense networks and who
collaborate with more individuals from their buildings are expected to exhibit less collaboration
outside of student affairs. In contrast to the cross-department model, homophily on racial/ethnic
identity is a significant positive predictor of collaboration outside of student affairs while
homophily on gender is not significantly associated with this dependent variable.
Several personal and professional characteristics are also associated with cross-
institutional collaboration. First, participants’ racial/ethnic identity is significantly associated
with cross-institutional collaboration, as Black participants exhibit significantly greater
collaboration with others outside of SA compared to their White colleagues. Rank in the
organizational hierarchy is also related to cross-institutional collaboration, as participants in the
highest ranks of their divisions (i.e., senior student affairs officers) report more cross-
institutional collaborations than their entry-level counterparts. As for functional unit, participants
in residence life and student conduct exhibit significantly less cross-institutional collaboration
compared to their colleagues in advising and academic support, administrative

25
leadership/generalists, admissions/orientation, multicultural programs, and other areas. Finally,
experience in the field (based on years) is a positive predictor of cross-institutional collaboration
after controlling for other variables. In contrast to the previous model, most of the variance in the
dependent variable for this model can be explained by individual characteristics (19%), followed
by collaborative network variables (9%), and institutional characteristics (1%).
1.5 Discussion
SA professionals are increasingly being called to collaborate across organizational
boundaries to influence organizational effectiveness (ACPA, 1994; Joint Task Force, 1998;
Keeling, 2004; Kuk et al., 2010; Love & Estanek, 2004; Manning et al., 2006; Task Force, 2010;
Winston et al., 2001), yet the extent to which they engage in this behavior is not well-
documented in the literature. This study provides evidence of collaboration for a large,
generalizable sample of SA professionals, as well as highlights behaviors and other
characteristics that may influence cross-boundary collaboration. In general, SA administrators’
personal collaborative networks indicate that they are successfully reaching across departmental
and physical campus boundaries to collaborate but are not collaborating as much with others
outside of SA. Findings from the random effects models indicate some commonalities, but they
also suggest that cross-departmental collaboration is much more accounted for by personal
behavior, while cross-institutional collaboration is more accounted for by structural factors. I
begin with a discussion of these commonalities and distinctions to highlight the differences in
these two phenomena, followed by analyzing specific findings pertaining to social identities and

26
collaboration that add further complexity to understanding collaboration in SA.
1
I conclude with
implications for practice and future research.
Behavioral vs. Structural Factors in Cross-Boundary Collaboration
Findings from this study reveal that SA professionals are engaging in collaborative
relationships across departments. In the midst of increasing calls for collaboration to enhance
student learning, Fried (2007) describes how many SA organizations have fallen victim to “silo”
thinking as organizational structures limit professionals to only think within the bounds of their
given functional area, resulting in “limited awareness of the activities, needs, and resources of
other divisions and departments” (p. 6). My research suggests that despite these structural
limitations, SA professionals are at least collaborating across departmental boundaries in their
work, as roughly half of the relationships formed in SA collaborative networks are with others
who work in different SA departments. It appears that SA professionals are working in structures
that necessitate these types of cross-departmental partnerships as part of everyday practice.
Despite the encouraging finding pertaining to cross-departmental collaboration, SA collaborative
networks also indicate that most collaboration is still occurring within SA. While cross-
departmental collaboration appears to be the norm, the relative lack of cross-institutional
collaboration suggests that SA professionals struggle with forming these partnerships in their
work and are either still trapped in siloed thinking or inhibited by institutional structures; the
findings from this study suggest it is probably both.
The random effects models indicate differences in cross-departmental and cross-
institutional collaboration and the behavioral or structural factors that relate to them, which could
highlight why SA professionals are engaging in more cross-departmental collaboration but not as

1
The dynamics of collaborative networks and social identity could be considered under behavioral and structural
factors influencing collaboration, but the overall trends and significance of social identities in SA work led me to
pull these out

27
much cross-institutional collaboration. On the one hand, most of the variance in cross-
departmental collaboration is accounted for by collaborative behavior (i.e., engaging in dense
networks, strong ties with collaborators, and collaborating within their building), not individual
or institutional characteristics, suggesting that SA professionals hold much of the agency for
being able to reach outside of their departments to form collaborative relationships. On the other
hand, more variance in cross-institutional collaboration is accounted for by individual
characteristics (both personal and professional) rather than collaborative behavior, which points
to the power of structural factors that could inhibit these partnerships. I discuss both behavioral
and structural variables below, highlighting trends and distinctions across the two models.
Network behaviors. Two specific network behaviors are negatively associated with both
types of collaboration – network closure and propinquity homophily. Descriptive statistics for
density indicate that SA collaborative networks tend toward more closure, suggesting that SA
professionals are by-and-large collaborating in closed of groups of professionals, which can
inhibit access to information and resources throughout the institution (Balkundi & Kilduff, 2006;
Coleman, 1988; Granovetter, 1973; 1982). The models in this study provide further evidence for
the inhibitive nature of closed networks. While SA professionals are likely to develop shared
attitudes and values with their colleagues through their closed collaborative networks (Balkundi
& Kilduff, 2006), they are limiting their potential for collaborating across departments and the
institution more broadly by doing so. The tendency of SA professionals to collaborate in closed
networks points to a behavioral barrier for the SA profession to engage in the kind of
collaboration advocated by its leaders.
Propinquity also exerts a strong pull inhibiting relationships from forming beyond
physical boundaries, which is reinforced by my models. While descriptive statistics suggest that

28
SA professionals are not necessarily limited by this pull, as roughly half of the collaborative
relationships in SA professionals’ networks are beyond one’s building, the models in this study
indicate that when SA professionals form collaborative relationships based on propinquity they
will be less likely to engage in cross-boundary collaboration. These findings point to both
behavioral and structural barriers to cross-boundary collaboration in SA. The behavioral
dimension suggests that individuals who do not extend effort to reach beyond their physical
boundaries and engage with other parts of a campus are inhibiting themselves from forming
these important partnerships. However, one’s physical location in an organization is often
dictated by hierarchical and organizational structures (Borgatti & Cross, 2003; Brass et al., 2004;
Festinger et al., 1950; Kadushin, 2012; Monge & Contractor, 2003), which can reinforce
structural silos (Fried, 2007) by placing individuals in geographic based on functional units and
hierarchy. So SA professionals in this case may be limited by physical (and by extension
organizational) structures, but also might not be pushing themselves to network and connect with
others across campus.
A final behavioral factor in predicting cross-departmental collaboration is tie strength,
with SA professionals being more likely to collaborate with others with whom they have weaker
relationships. While strong ties can foster trust (Granovetter, 1973, 1982; Hoppe & Reinelt,
2010; Krackhardt, 1992), weak ties serve a purpose in organizations by connecting disparate
parts of the institution (Burt, 1992; Krackhardt, 1990). SA professionals who have weaker ties in
their collaborative networks are more likely to collaborate with others outside of their
department, reinforcing the findings from the research about the potential for weak ties to
connect people across organizations.

29
Formal organizational structures. Results pertaining to professional characteristics
reveal insights into specific structural factors relating to collaboration. Regarding professional
rank, entry-level professionals exhibit significantly less cross-departmental collaboration
compared to colleagues at all other ranks. However, these effects do not carry over to cross-
institutional collaboration; no differences are evident among entry-level, mid-level, and senior-
level professionals, with only senior student affairs officers exhibiting more tendencies toward
collaborative relationships outside of student affairs. These findings highlight two structural
inhibitors to cross-boundary collaboration, reinforcing the literature on organizational hierarchy
and social networks (Brass et al., 2004). First, entry-level professionals appear to struggle with
forming relationships both beyond their department and across the institution. Despite
suggestions that all professionals can and should create cross-boundary collaborations (Kuk et
al., 2010; Love & Estanek, 2004; Manning et al., 2006; Winston et al., 2001), these findings
suggest that professionals lower in the organizational hierarchy are either not able or are not
actively pursuing such relationships. Second, professionals at the top of the organizational
hierarchy (i.e., senior student affairs officers) are expected to collaborate significantly more with
others across the institution compared to entry-level professionals, and further analyses reveal
that they engage in this type of collaboration more than all other professionals at lower ranks.
2

So the ability to collaborate outside of SA and across institutions appears to be inhibited by
organizational structures, as those at the top of the hierarchy are the ones most able to engage in
this boundary-crossing behavior. In a related finding, professionals with more years of

2
The random effects models only compare the three groups to entry-level professionals. In order to understand the
relationships of professional rank to cross-institutional collaboration among all groups, I utilized ANOVA analyses,
which reveal that these upper administrators exhibit more cross-institutional collaboration than all professionals at
lower levels below them, including even senior-level administrators [F(3,636) = 19.775, p < .001]

30
experience in higher education, which frequently coincides with professional rank, are also more
likely to engage in cross-institutional collaboration, reinforcing this notion.
A final structural factor in cross-boundary collaboration is evident in the associations of
functional unit with the dependent variables. For both models, professionals who work in
residence life and student conduct programs exhibit significantly less cross-boundary
collaboration, and in the case of cross-department collaboration they score lower than all other
functional areas. Despite the fact that residence life and housing operations in particular must
coordinate efforts with other departments to communicate about enrollment and increasing
partnerships with academic affairs being developed through living learning programs (Dungy,
2003; Golde & Pribbenbow, 2000), these professionals lag far behind their colleagues in cross-
boundary collaboration. This likely suggests some insularity in the nature of housing and conduct
work, with professionals tending to collaborate more with others in their departments in order to
accomplish their work.
Taken together, these findings suggest that SA professionals may have some agency in
pursuing behaviors to increase cross-boundary collaboration, such as trying to expand their
personal networks beyond tight-knit groups and reaching beyond the confines of their buildings.
However, organizational structures seem to also play influential roles in these behaviors, with
organizational hierarchy, functional unit, and even physical campus location (which is often
dictated by organizational structures) associated with collaboration. Network behaviors account
for much more variance in cross-departmental collaboration, yet they account for relatively little
variance in cross-institutional collaboration, which further suggests that professional
characteristics tied to organizational structures play a more influential a role in cross-institutional
collaboration.

31
Social Identities and Collaboration
The descriptive statistics and models in this study reveal complex dynamics of
collaborative network behavior and social identity. Homophily based on social identity is
associated with cross-boundary collaboration, but the nature of this relationship varies depending
on the type of collaboration. Gender identity homophily is associated with cross-departmental
collaboration, whereas racial/ethnic identity homophily is associated with cross-institutional
collaboration. These findings suggest that SA professionals are more likely to collaborate across
departments with others who share their same gender identity and more likely to collaborate
across institutional boundaries with those who share their same racial/ethnic identity. As other
analyses revealed, those who identify with a female gender identity exhibit much more gender
homophily in their networks than their male and transgender colleagues, and White SA
professionals exhibit much more racial/ethnic homophily than their colleagues of color. These
patterns led me to examine if there was an interaction effect of gender homophily by gender on
cross-departmental collaboration, as well as an interaction effect of racial/ethnic homophily by
race/ethnicity on cross-institutional collaboration. The coefficients for the interaction effects in
these two models were not significant, suggesting that the association of homophily on these
collaborative behaviors is not moderated by individuals’ social identities. In other words, there
are no differential effects of homophily on collaboration based on one’s social identity; SA
professionals are drawn to collaborate across departments with others who share their gender
identity and collaborate outside of SA with others who share their racial/ethnic identity
regardless of their own social identities.
Other covariates in the model help further unpack the role of social identities in
collaboration. Other analyses for this paper revealed that individuals in the majority in SA (i.e.,

32
female professionals and White professionals – Taub & McEwen, 2006) exhibit more homophily
on these characteristics than their colleagues who comprise the minority in the field. Research
suggests this is due to the fact that individuals in the minority in organizational settings are
forced to connect with others with different identities more than their own due to the skewed
representation in these organizations (McPherson et al., 2001). However, the models in this study
suggest that individuals in the minority in SA – namely male and Black participants – exhibit
more cross-boundary collaboration.
3
Despite the fact that female professionals are in the
majority in the SA profession, male professionals seem better able to reach across organizational
boundaries, coinciding with the literature on differential returns for males compared to females
in same sex networks (Brass, 1985; Ibarra, 1992).
4
Further, homophily by race and ethnicity is
one of the strongest factors in network behavior generally (McPherson et al., 2001), and research
identifies how racial minorities in organizations tend to make more decisions about relationships
based on identity in order to make connections and find support from those with shared
experiences (Mehra et al., 1998). The models indicate that Black professionals are more likely to
reach across institutional boundaries than their White colleagues, suggesting they are possibly
seeking out connections from across institutions who share their racial identity, thus using that
shared identity as a means to make connections and find support within the institution (Mehra et
al., 1998).
Implications
This study holds implications for both practice and research. First, the combination of
behavioral and structural factors associated with cross-boundary collaboration suggests

3
Given the relatively small sample size of transgender participants in this study, it is difficult to draw meaningful
conclusions based on these findings strictly for transgender participants, despite the fact that they also are expected
to collaborate more across departments compared to female participants based on these models.
4
Literature on social networks tends to examine patterns by sex as opposed to gender. Despite the fact that these
studies focus on sex and not gender, they are the closes approximation to the effects observed in this study.

33
intentional strategies for improving this type of collaboration. Specifically, engaging in closed
networks and being limited by geographic location are two potential inhibitors to cross-boundary
collaboration. SA professionals can work to combat these forces in their work by seeking out
opportunities to engage with other beyond their local networks and buildings. Seeking out
involvement opportunities on campus, such as committee work, and creating new programs that
intentionally bring together members from disparate parts of campus are strategies that can help
SA professionals reach beyond their closed networks and their geographic locations. These
strategies can also provide opportunities for professionals to interact more regularly in their
work, which can serve to strengthen relationships and bonds across campus.
The structural barriers identified in this study related to rank and functional unit, as well
as propinquity, highlights the need for structural considerations to increase collaboration.
Campus leaders, both within and outside of student affairs, should consider strategies for
improving collaboration across campuses, such as those highlighted by Kezar and Lester (2009).
Mission and vision statements communicate meaning and goals for organizations (Bolman &
Deal, 2008) and provide a guide of sorts for organizational members on how to engage in the
organization. Campus leaders and stakeholders who desire more collaboration can influence this
by not only including collaboration in these statements but also through intentionally reminding
the campus community about the mission and its value in public documents and forums (Kezar
& Lester, 2009). Reward structures can also be utilized to enhance collaboration by making this
behavior rewarded in promotion and other personnel policies. Given the structural factors
pertaining to rank, these structures could also include rewards for senior leaders in SA to mentor
and provide opportunities for those lower in the hierarchy to engage in cross-boundary
relationships. Restructuring or integrating departments to connect functional units across campus

34
can also help combat some of the structural inhibitors to collaboration (Kezar & Lester, 2009).
Given that residence life and student conduct departments exhibit more insularity in this study,
campus leaders might specifically pursue ways to connect these departments with others across
the campus to ensure structured connections are occurring. Finally, the SA profession exhibits its
own homogeneity in the fact that a majority of its members are White females (Taub &
McEwen, 2006). Given the tendency for these individuals to engage in more homophilous
collaborative networks, those responsible for hiring as well as recruiting in graduate preparation
programs should intentionally seek to broaden the diversity of the profession in order to increase
the potential for more heterophilous collaboration based on these identities.
Finally, this study suggests some future directions for research pertaining to collaboration
in higher education. First, this study is from the perspective of SA administrators, and while the
results can be useful to other professionals in higher education, research utilizing SNA as a
theoretical and methodological framework can help advance this area of study. While SA
professionals are called to collaborate across organizational boundaries, others have highlighted
how the fragmented learning environment is detrimental to student outcomes (Kezar, 2005;
2006), suggesting the need for more research from multiple perspectives in higher education,
including academic administrators and faculty. Utilizing social network analysis to identify
patterns in collaborative networks for these individuals will help to paint a more complete picture
of the possibilities and barriers to meaningful cross-boundary collaboration. Additionally, this
study is from the perspective of individuals about their social networks, relying on and trusting
self-report data. Future research could utilize follow-ups with individuals in the study exhibiting
more cross-boundary collaboration in order to identify their collaborators and utilize them in
future survey research; this could allow researchers to better understand the accuracy of

35
participants’ perceptions of their network connections. This would fit in with the social cognition
strand of SNA research (Balkundi & Kilduff, 2006) and identify the merits and potential
constraints of self-report in ego network studies.
1.6 Conclusion
The calls for more cross-boundary collaboration in higher education and SA are well-
documented, but the extent to which it occurs and the complexities of this behavior is less well-
known. This study is the first to highlight that SA professionals are more successful at
collaborating across SA departments as opposed to across institutions and specific network
behaviors that both contribute to and inhibit this collaboration. In order to contribute to student
learning and respond to increasing challenges and uncertainties in higher education, SA
professionals will have to engage in more of this behavior in order to successfully guide their
institutions through the rough water that is facing higher education (Manning et al., 2006). This
research points to behavioral and structure dimensions that can contribute to cross-boundary
collaboration, which in the long run can contribute to student success and improved institutional
effectiveness.

36
1.7 Appendix 1A: Descriptive Statistics for Other Variables in this Study
Mean S.D. Min Max Description
Institutional Characteristics
Institution Size: < 1,000 0.01 0.00 1.00 1 = <1,000; 0 = All others
Institution Size: 1,000-4,999

0.25 0.00 1.00 1 = 1,000-4,999; 0 = All others
Institution Size: 5,000-9,999 0.11 0.00 1.00 1 = 5,000-9,999; 0 = All others
Institution Size: 10,000-19,999

0.23 0.00 1.00 1 = 10,000-19,999; 0 = All others
Institution Size: 20,000+ 0.39 0.00 1.00 1 = 20,000+; 0 = All others
Carnegie: Doctoral 0.49 0.00 1.00 1 = Doctoral; 0 = All others
Carnegie: Masters 0.28 0.00 1.00 1 = Masters; 0 = All others
Carnegie: Baccalaureate

0.14 0.00 1.00 1 = Baccalaureate; 0 = All others
Carnegie: Associates 0.05 0.00 1.00 1 = Associates; 0 = All others
Carnegie: Other

0.04 0.00 1.00 1 = Other; 0 = All others
Control: Public 0.61 0.00 1.00 1 = Public; 0 = All others
Control: Private 0.39 0.00 1.00 1 = Private; 0 = All others
Institutional Climate for
Collaboration
3.99 0.88 1.00 5.00 4-point scale; 1 = Newcomer; 4 =
Sporadic; 3 = Episodic; 4 = Continuous
Individual Characteristics
Female 0.61 0.00 1.00 1 = Female; 0 = All others
Male 0.37 0.00 1.00 1 = Male; 0 = All others
Transgender 0.01 0.00 1.00 1 = Transgender; 0 = All others
Latino/Hispanic 0.03 0.00 1.00 1 = Latino/Hispanic; 0 = All others
Asian/Asian American 0.03 0.00 1.00 1 = Asian/Asian American; 0 = All others
Black/African American 0.09 0.00 1.00 1 = Black/African American; 0 = All others
White

0.80 0.00 1.00 1 = White; 0 = All others
Multiracial

0.04 0.00 1.00 1 = Multiracial; 0 = All others
Entry-level Professional 0.21 0.00 1.00 1 = Entry-level; 0 = All others
Mid-level Professional 0.51 0.00 1.00 1 = Mid-level; 0 = All others
Senior-Level Professional 0.18 0.00 1.00 1 = Senior-level; 0 = All others
Senior Student Affairs Officer
(SSAO)
0.10 0.00 1.00 1 = SSAO; 0 = All others
Functional Area: Residence
Life/Student Conduct
0.28 0.00 1.00 1 = Residence Life/Student Conduct; 0 =
All others
Functional Area:
Activities/Union/Rec
0.19 0.00 1.00 1 = Activities/Union/Rec; 0 = All others

Functional Area: Advising/
Careers/ Academic Support
0.17 0.00 1.00 1 = Advising/Careers/Academic Support;
0 = All others
Functional Area: Administrative
Leadership/Generalist
0.18 0.00 1.00 1 = Administrative Leadership/Generalist;
0 = All others
Functional Area:
Admissions/Orientation
0.05 0.00 1.00 1 = Admissions/Orientation; 0 = All others
Functional Area: Multicultural/
Diversity Programs
0.05 0.00 1.00 1 = Multicultural/Diversity Programs; 0 =
All others
Functional Area:
Wellness/Health
0.02 0.00 1.00 1 = Wellness/Health; 0 = All others
Functional Area: Assessment 0.03 0.00 1.00 1 = Assessment; 0 = All others
Functional Area: Other

0.02 0.00 1.00 1 = Other Functional Area; 0 = All others
Years at Institution 7.16 7.19 1.00 41.00 Units in years
Years in Higher Education 14.10 9.94 1.00 41.00 Units in years

37
CHAPTER 2 – Collaborative Social Networks and Student Affairs Competencies:
The Strength of Strong Ties

2.1 Introduction

Ever since its emergence in the early 20
th
century, the student affairs (SA) profession has
charted its direction through philosophical statements and position papers, which detail the
evolving roles of SA administrators on college campus (Evans & Reason, 2001; Wilson, Comes,
& Dannells, 2006). For nearly 60 years, from the release of the Student Personnel Point of View
(ACE, 1937/2012) to the Student Learning Imperative (ACPA, 1994), leading voices in the field
have offered perspectives on the nature of SA work and its supplementary status to academics in
the mission of higher education institutions. The release of the several key documents in the
1990’s appeared to indicate a shift in thinking for the profession, as they were “hailed for
introducing a new student affairs philosophy focused on student learning and encouraging
collaboration” (Evans & Reason, 2001, p. 359, emphasis added). The message conveyed in these
statements and papers suggest that SA professionals must continue to expand their roles on
campus by coordinating services, forming partnerships, and cultivating relationships across
hierarchical boundaries to enhance learning for all students (ACPA, 1994; Joint Task Force,
1998; Keeling, 2004; Task Force, 2010). These calls for collaboration are echoed by scholars in
the field who increasingly argue for SA professionals to reach beyond their functional siloes in
order to contribute to increased effectiveness in higher education (Fried, 2007; Kuk, Banning, &
Amey, 2010; Love & Estanek, 2004; Manning et al., 2006; Winston, Creamer, & Miller, 2001).
As the role of SA professionals has expanded and grown over the past century, so too has
the list of competencies deemed necessary for effective practice in SA. In the early years of the
profession, SA administrators could be successful in their work with well-developed counseling
skills and knowledge of student development (Sriram, 2014), but the nature of student affairs

38
work in the 21
st
century necessitates a broader array of competencies in areas such as
supervision, leadership, and administration in order to successfully work within complex higher
education institutions (Porterfield, Roper, & Whitt, 2011). While the knowledge and skills
necessary for effective SA practice have been included and referenced in many of the guiding
documents of the profession, leaders and scholars from the preeminent SA professional
associations recently came together to identify and codify the necessary competency areas for
SA practice in their document, Professional Competency Areas for Student Affairs Practitioners
(Bresciani, Todd, & Associates, 2010). Only one study to-date measures SA professionals’
perceived competencies in these areas (Sriram, 2014), and no studies have related these
competencies to professional practice or individual and professional characteristics (such as
functional area, rank, and experience), which is an important next step in the evolution of
competencies research to inform professional practice
In conjunction with this need for more research pertaining to SA competencies, more
research is also needed related to the recent calls for collaboration cited above. Despite
increasing calls for collaboration that cuts across organizational boundaries in higher education
institutions, few are examining the extent to which SA professionals are engaging in this type of
behavior. Research on structured, formal collaborations connecting academic and student affairs
in higher education point to many institutional outcomes that arise from these partnerships (such
as better coordination of services, gains in student engagement and learning, and identification of
best practices such as learning communities – Kezar & Gehrke, in press; Whitt, 2011); however,
these types of programs are not widespread in the academy (Kezar & Lester, 2009), and this
research does not focus on the benefits of collaboration for individual administrators. Little is
generally known about the extent to which SA professionals collaborate in their everyday work

39
and how their collaborative behavior may relate to their own individual effectiveness in SA
work. Given the increased necessity for SA professionals to serve as administrators and leaders
in complex higher education institutions, specifically understanding the extent to which
collaboration relates to competencies pertaining to administration and leadership can provide one
of the first insights into the relationship of collaborative professional behavior to important
competencies for SA practice.
Given the gaps in the literature cited above, this paper seeks to answer two research
questions:
1. How do SA professionals assess their proficiency in three organizationally-related
competency areas (human and organizational resources; leadership for
personal/community development; and organizational leadership), and are there
significant differences in these competencies by professional rank and functional
unit?

2. To what extent is collaborative behavior in student affairs is associated with
proficiency in these competency areas after controlling for institutional, personal, and
professional characteristics.
In answering these questions, this is one of the first studies to assess perceived competency in
SA across a broad sample of professionals from various institutions, functional units, and
professional ranks. It is also the first study to attempt to identify relationships between student
affairs competencies and organizational behavior in SA from a large, generalizable sample.
2.2 Research on Student Affairs Competencies and Collaboration in Higher Education
The empirical literature on student affairs competencies and collaboration lay the
groundwork for this study in identifying the evolution of research on competencies in student
affairs and benefits associated with collaboration in higher education and. I review these bodies
of research in order to illustrate the ways in which this study advances the research on
competencies as well as suggest why collaboration is an important phenomenon to examine
given its benefits to students and institutions.

40
Student Affairs Competencies
The research literature on competencies in SA tends to fall into two categories, one
identifying competencies of SA work and the other examining competencies by position (Sriram,
2014). The body of work identifying relevant competencies for SA work has been summarized
by two meta-analyses (e.g., Herdlein, Riefler, & Mrowka, 2014; Lovell & Koston, 2000), who
combined reviewed 45 years of research (over 50 studies) identifying the types of skills and
competencies required for SA work. These meta-analyses reveal that early studies of SA
competencies focused mostly on interpersonal relationships, human development, and
programming (Sriram, 2014). Later research also included competency related to management
and administration, with the most recent work adding the importance of multicultural proficiency
and research/assessment skills.
This body of research prompted leaders of the profession to collaborate on a project,
which was sponsored by the two preeminent professional associations in SA, to synthesize this
research and identify the core competencies for SA practice (Bresciani et al., 2010). These
scholars identified 10 competency areas they deemed necessary for all SA professionals to
exhibit at least basic proficiency in. These areas are: advising and helping; assessment,
evaluation, and research; equity, diversity, and inclusion; ethical professional practice; history,
philosophy, and values of the profession; human and organizational resources; law, policy, and
governance in higher education; leadership; personal foundations; and student learning and
development. Given the sponsorship of these competencies by the two primary associations in
this profession, they provide the most comprehensive and agreed upon set of core competencies
for this work by which professionals can inform practice and research.

41
Recent scholars focus on the difficulties of professionals in navigating some of the
complex demands of administration and leadership in higher education institutions specifically,
suggesting that while they may feel prepared in areas traditionally considered part of SA work
(e.g., student development), developing a better understanding of the behaviors and
characteristics associated with increased competencies in more organizationally-related areas can
inform practice and advance the scholarship regarding SA competencies (Cuyjet, Longwell-
Grice, & Molina, 2009; Sriram, 2014). Therefore, the two competency areas identified by
Bresciani and colleagues (2010) that are the focus for this inquiry are 1) human and
organizational resources, and 2) leadership. The human and organizational resources competency
area encapsulates knowledge, skills, and attitudes important for staff supervision and
management, as well as additional subareas including conflict resolution, financial and facilities
management, crisis and risk management, and sustainable resources (Bresciani et al.). The
leadership competency area encapsulates knowledge, skills, and attitudes to be a leader, both
individually and for the good of an institution, regardless of one’s position in the organizational
hierarchy (Bresciani et al.); this competency area entails understanding one’s own leadership
style and interpersonal skills, as well as ability to understand and effect larger organizational
changes.
5

Competencies research. Much of the research examining competencies does so by
professional rank or position in institutional hierarchies. This work focuses on the skills required
for new professionals entering the profession (Burkard, Cole, Ott, & Stoflet, 2005; Kretovics,
2002; Kuk, Cobb, & Forrest, 2007; Waple, 2006). These studies by-and-large suggest that most
entry-level professionals are underprepared in areas relating to administration (such as

5
In the methods section, I describe the factor analyses utilized in this study that revealed two distinct constructs
within the leadership competency area, resulting in two leadership areas: leadership for personal/community
development, and leadership for organizational change.

42
budgeting, supervision, etc.), which are deemed to be important and necessary for entry-level
work as perceived by faculty in professional preparation programs and supervisors of new
professionals. As for other levels in institutional hierarchy, research suggests that skills related to
leadership and working with others are most important for mid-level and senior professionals
(Gordon, Strode, & Mann, 1993; Randall & Globetti, 1992; Saunders & Cooper, 1999). These
studies identify the importance of these competencies based on perceptions of others in relation
to these different position levels (i.e., supervisors, faculty) but do not attempt to assess
individuals’ levels of competencies in these areas.
Few studies have attempted to measure competency levels of SA professionals, but most
tend to focus on narrow competencies or subsets of SA professionals (e.g., Castellanos, Gloria,
Mayorga, & Salas, 2007; Kuk et al., 2007; Waple, 2006). Only one study to-date has examined
the competency levels of SA professionals across a broad range of competencies and
professional levels in SA (Sriram, 2014). Utilizing the Bresciani et al. (2010) framework, Sriram
(2014) developed a psychometrically sound instrument consisting of 15 subscales that mapped
onto the 10 competency areas identified by national leaders and scholars, suggesting that these
competencies are valid for assessing SA professionals’ professional competencies. Sriram found
that SA professionals rate themselves strongest in areas pertaining to diversity, ethical practice,
leadership, and advising/helping, while their weakest competency areas pertained to research
skills/values/behaviors, governance, and collaboration with academic colleagues. However,
Sriram did not disaggregate this data nor perform multivariate analyses to assess differences by
professional rank, functional area, or individual behavior, so we still do not know how these
characteristics relate to competency in SA work. One type of behavior that SA professionals are
increasingly being called upon to engage in order to improve institutional outcomes is

43
collaboration across higher education institutions, which suggests it is a type of professional
behavior worth examining and that may be associated with these competencies. I now highlight
some of the benefits of collaboration in higher education.
Collaboration in Higher Education
Collaboration holds the potential to not only improve student learning and teaching (Kuh,
Kinzie, Schuh, & Whitt, 2005; Schroeder 1999a; 1999b) but also contribute to organizational
benefits in higher education (Bensimon & Neumann, 1993; Bourassa & Kruger, 2001; Kezar,
Carducci, & Contrerars-McGavin, 2006). In their work detailing frameworks for collaboration,
Kezar and Lester (2009) define partnerships and collaboration as involving mutual goals among
individuals who rely on each other to accomplish them. They further identify several key
advantages from the research literature that result from collaborative action in organizations. One
of the most cited benefits of collaboration is that it leads to “more interaction, information
sharing, communication, and collective problem solving [resulting] in innovation and learning”
(Kezar & Lester, p. 10). Collaboration also increases cognitive complexity in problem solving
(Bensimon & Neumann, 1993; Denison, Hart, & Kahn, 1996), leads to more efficiency and cost
effectiveness (Hagadoorn, 1993), and results in increased employee motivation (Denison et al.,
1996; Googins & Rochlin, 2000) and better service to institutional stakeholders (Schroeder,
1999a).
Research on collaboration generally identifies benefits for students and institutions
(Kezar & Lester, 2009; Rodems, 2011), yet relatively few studies focus on individual benefits
that come from these collaborative relationships. Some individual benefits identified in the
research include increased cognitive complexity through exposure to multiple perspectives
(Kezar & Lester, 2009; Nidiffer, 2006; Rodems, 2011) and increased satisfaction and access to

44
resources (Boardman & Ponomariov, 2007). The research on improved institutional management
through collaboration (Bensimon & Neumann, 1993; Kezar & Lester, 2009) coupled with the
increasing necessity for SA administrators to possess skills related to administration and
management (Porterfield et al., 2011; Sriram, 2014) suggests that another benefit of
collaboration for SA professionals could come in increased competency in these areas, yet no
research to-date has examined these relationships. This study seeks to fill this gap, and social
network theory provides a relevant conceptual and theoretical framework for examining these
relationships.
2.3 Conceptual and Theoretical Framework: Social Network Theory
Collaboration is inherently relational as it involves interactions and connections among
two or more individuals. Social network theory is an appropriate theoretical and empirical lens
for this inquiry because it moves beyond examining the traits and behaviors of individuals to
focus on the nature of relationships between and among individuals (Borgatti & Ofem, 2010); an
examination of collaboration must inherently examine relationships between and among
individuals since collaborating requires social interaction and cooperation. Social network theory
and research exhibits four core principles that distinguish it from rival theoretical and empirical
approaches to studying organizational effectiveness (Balkundi & Kilduff, 2006; Kilduff & Brass,
2010; Kilduff, Tsai, & Hanke, 2006). These core principles are: 1) an emphasis on the nature of
relationships in organizations as being just as, if not more important, than individual attributes
for understanding organizational effectiveness; 2) the acknowledgment that human behavior is
embedded in broader systems and networks of friendship and acquaintance relationships; 3) an
emphasis on the structured patterning of social relations and the importance of understanding

45
human relations at multiple levels; and 4) social networks afford both opportunities and
constraints that influence outcomes relevant to both individuals and organizations.
Social network theory utilizes many theoretically and empirically-tested concepts to
understand and operationalize human behavior. Several of these concepts are particularly
important to understanding human behavior in organizations (Balkundi & Kilduff, 2006; Borgatti
& Cross, 2003; Brass et al., 2004) and can contribute to understanding collaborative behavior in
SA. These are tie strength, density, and homophily. Strength of ties describes the quality (based
on frequency of interaction and perception of the relationship) of connections between an
individual and his or her network connections (Brass & Krackhardt, 1999; Gillis, 2008;
Krackhardt, 1992; Valente, 2010). Information is transmitted across organizations more freely
through strong ties, in which individuals have more frequent interaction and trust with their
network connections (Granovetter, 1973, 1982; Hoppe & Reinelt, 2010; Krackhardt, 1992).
However, weak ties can serve as bridges of communication across organizations, connecting
disparate groups and facilitating diffusion of information, practice, and innovation (Burt, 1992;
Granovetter, 1973; 1982). The strength of collaborative relationships among SA professionals
can indicate the kind of trust, interaction, and access to information that can come through these
connections.
Density refers to the extent to which individuals’ personal networks are comprised of
people who are also connected to one other (Balkundi & Kilduff, 2006; Kezar, 2014; Valente,
2010). Dense personal networks (i.e., exhibiting more closure) are associated with sharing
similar attitudes and beliefs and exhibiting more trust, but are also closed off to the rest of the
organization (Balkundi & Kilduff, 2006; Coleman, 1988), suggesting difficulty accessing new
information from others in the organization who are not part of the network. Ego network density

46
is relevant to understanding collaborative behavior among SA professionals because it indicates
the extent to which someone may have access to other parts of the institution and one’s potential
ability to collaborate beyond one’s local network.
Homophily refers to the extent to which individuals form relationships with others who
are similar to themselves (Lazarsfeld & Merton, 1954; McPherson, Smith-Lovin, & Cook, 2001).
These tendencies can be related to personal demographics (i.e., gender, race) and professional
characteristics (i.e., hierarchical rank, functional unit, physical geographic location in an
institution). Lack of diversity in a personal network can inhibit access to resources, information
transfer, and workplace performance for individuals in those networks (Borgatti & Cross, 2003;
Brass, 1985; Brass et al., 2004; Ibarra, 1992; Papa, 1990). Homophily is a key measure to
consider when studying collaboration, as it can indicate of the extent to which SA professionals
are collaborating with others outside of their departments and divisions, as well as their
tendencies to be drawn to collaborate with those who share their own identities.
The social network literature is replete with research examining consequences or
outcomes associated with these network behaviors in organizations. The research that is most
pertinent to this study examining associations of collaborative network behavior to SA
competency is that focusing on the associations of social networks with performance in
organizations (Borgatti & Foster, 2003; Brass et al., 2004). Research in this vein focuses on a
variety of benefits associated with network ties, including associations of weak ties with
creativity (Brass, 1995; Perry-Smith & Shalley, 2003), network centrality and homophily with
formal performance evaluation (Baldwin, Bedell, & Johnson, 1997; Brass, 1985; Mehra, Kilduff,
& Brass, 2001), network diversity and size with productivity (Papa, 1990), and network
centrality and strong ties with interpersonal citizenship and altruism (Bowler & Brass, 2006;

47
Settoon & Mossholder, 2002; Sparrowe, Liden, Wayne, & Kraimer, 2001). An additional strand
of network research highlights the distinct benefits that come from inter-unit network ties in
organizations, including better coordination in times of crises (Krackhardt & Stern, 1988), better
individual (Cross & Cummings, 2004) and unit performance (Mehra et al., 2001), higher
productivity (Reagans & Zuckerman, 2001), improved unit efficiency (Reagans, Zuckerman, &
McEvily, 2004) and reduced inter-unit conflict (Nelson, 1989). The extents to which groups and
organizations encourage and are designed to facilitate inter-unit ties also influence group
performance (Pearce & David, 1983). This research collectively identifies how network behavior
can contribute to a host of outcomes for individuals and their organizations. The extent to which
collaborative network behavior is associated with competency in SA work is the focus of this
study.
2.4 Methods
Data Collection & Sample
Data for this study were collected through the Collaborative Social Networks in Student
Affair (CSNSA) survey. This survey was distributed to members of the American College
Personnel Association (ACPA-College Student Educators International) who identified as
professional staff working in a college or university in the United States in the summer of 2014.
ACPA was selected due to its national representation of SA administrators from all position
levels and functional areas in higher education institutions.
The purpose of the CSNSA was to gather data pertaining to SA professionals’
collaborative networks in their institutions and SA professionals’ perceived proficiencies in
several SA competency areas. It contained questions asking participants to identify up to seven
individuals with whom they collaborate in their work, followed by questions about each

48
individual they identified. Questions for each collaborator included items assessing the strength
of their relationship, professional and personal characteristics of the collaborator, and the
collaborative relationships of each individual with others in the participants’ ego network.
Additional questions on the survey pertained to perceived proficiency in several student affairs
competencies, personal demographics of participants (e.g., racial/ethnic identity, gender
identity), and professional demographics (e.g., institution, position rank, functional area, years of
experience). Survey validity was assessed in three primary ways. First, survey design was
informed by the literature on social network theory and prior research, which suggested methods
for obtaining information related to individuals’ ego networks. For example, I utilized a
psychometrically sound scale to assess tie strength between participants and each of their
collaborators (Gillis, 2008). I also utilized the literature pertaining to SA competencies to inform
items to include in the survey. Second, the survey was piloted by eight individuals a) with
expertise in research in higher education and b) who either currently or formerly worked in an
administrative position in SA. I gathered qualitative feedback from each individual regarding the
content and structure of the survey, ensuring both face and content validity of the items on the
survey. Third, I ensured construct validity of the competency measures through principal axis
factor analysis with pairwise deletion and Promax rotation. The Kaiser-Meyer-Olin measure of
sampling adequacy (KMO = 0.94) and Bartlett’s Test of Sphericity (p < .001) suggest that the
proper assumption were met for factor analysis with these items.
Data were collected over a four-week period in July and August 2014. The target
population for this study is SA administrators. Therefore, ACPA identified individuals from
among their members who were current administrators at a higher education institution in the
United States from all levels in the hierarchy (entry, mid, and senior-level). In order to ensure

49
generalizability across the sample, individuals who identified as current faculty members,
undergraduate and graduate students, ACPA members who work in institutions outside of the
United States, and affiliate members who work for organizations that are not colleges or
universities were excluded from the sample. The initial invitation to participate in the survey was
sent by the ACPA central office to 3,205 participants from 888 institutions who fit this
description, with two follow-up emails sent two weeks apart after the initial invitation.
Participation was incentivized with the option for participants to be entered into a drawing to win
a grand prize of an iPad Mini or one of 10 gift cards to an online retailer. Of the 3,205
individuals who were invited to participate, 913 responded to the survey invitations, indicating a
28.5% initial response rate. Of those who responded, 640 participants completed the entire
survey (70.1% completion rate) from 322 institutions, resulting in a final response rate of 20.0%.
Descriptive statistics pertaining to individual demographics, professional characteristics, and
institutional characteristics for the sample are listed in Appendix 2A.
Variables
Dependent variables. The three dependent variables for this study represent proficiency
in three SA competency areas. The CSNSA contained 50 items I developed to assess
participants’ perceived proficiency in sub-areas of six broad competency areas: advising and
helping; assessment, evaluation, and research; equity, diversity, and inclusion; human and
organizational resources; leadership; and personal foundations, which were identified from and
informed by the work of Bresciani et al. (2010). Rather than attempt to assess perceived
proficiency in all 10 competency areas, I chose six areas that most closely pertained to the types
of job performance outcomes identified in the SNA literature in order to limit the length of the
CSNSA in an effort to reduce survey dropout. I utilized principal axis factor analysis with

50
pairwise deletion and Promax rotation on these 50 items to identify underlying constructs of SA
competencies. After removing items that loaded on multiple constructs, I identified seven
constructs from the remaining 43 items. Five of the seven constructs mapped onto five
competency areas assessed on the survey; however, the items pertaining to leadership loaded on
two separate factors, suggesting two forms of leadership that are necessary for SA work.
This study utilizes three of the seven factors identified through the factor analyses above
as the dependent variables in the random effects models described below. These three factors
represent organizationally-related competencies, which are of increasing importance in
competency research (Cuyjet, Longwell-Grice, & Molina, 2009; Sriram, 2014) and are areas
where new professionals in particular exhibit some of the weakest proficiency levels (Burkard et
al., 2005; Kretovics, 2002; Kuk et al., 2007; Waple, 2006). Therefore, identifying predictors for
these three specific competency areas can inform practice for SA professionals specifically
regarding important organizationally-related competency areas, especially for entry-level and
new professionals. The first is “Human and Organizational Resources,” a five-item scale of
proficiency in practices related to important competencies for organizational functions and
administrations, such as risk and crisis management, conflict resolution, sustainability issues, and
resource management.
The other two factors are the two leadership factors identified through the factor
analyses; Bresciani and colleagues (2010) identified leadership as a competency area for SA
professionals, but the factor analyses for this study reveal two distinct factors that relate to
leadership. The first, “Leadership for Personal and Community Development,” contains items
pertaining to team and community building, mentoring, and developing others and can be
considered more of an individual-level leadership construct. The second leadership factor,

51

“Organizational Leadership,” contains items pertaining to one’s knowledge of the institution’s
culture/mission, ability to contribute to organizational change, ability to utilize networks and
partnerships, and decision-making. These three factors are all internally reliable, with Cronbach
α’s ranging from 0.78 to 0.89. Scale score for each factor are the mean score for all the items in
the construct, with scores ranging from 1 = “Not Proficient” to 7 = “Very Proficient.” Factor
loadings and descriptive statistics for the three dependent variables are listed in Table 2.1.
Focal independent variables. The focal variables for this study are those pertaining to
social network measures of participants’ collaborative networks. These include ego network size,
density, average tie strength, homophily based on racial/ethnic identity and gender identity,
homophily based on propinquity (i.e., working in the same building), extent of collaboration with
faculty, and homophily based on department and student affairs. Ego network size was simply
the number of individuals participants identified as collaborators in their work (up to seven). For
ego network density, participants were asked to identify whom each member of their ego
Table 2.1: Composite Scales and Descriptive Statistics for Dependent Variables
Factor
Loading
Eigenvalues Cronbach α Mean (SD)
Human and Organizational Resources 1.52 0.85 5.17 (1.00)
Risk management 0.89
Crisis management 0.84
Sustainability issues 0.69
Conflict resolution 0.68
Resource management 0.55
Leadership for Personal & Community Development 3.40 0.89 5.68 (0.91)
Building teams 0.90
Mentoring 0.89
Development of others 0.86
Community building 0.86
Organizational Leadership 1.01 0.78 5.79 (0.83)
Knowledge of institutions’ culture/mission 0.96
Ability to contribute to organizational change 0.81
Utilizing networks and partnerships 0.46
Decision-making 0.45
NOTE: 7-point scale; 1 = Not proficient; 3 = Somewhat proficient; 5 = Proficient; 7 = Very
proficient

52
networks collaborated with among the other individuals in the network. Density was calculated
as proportion of actual ties in the network compared to the potential ties in the network (not
including ties to the participant), ranging from 0 (least dense) to 1 (most dense). For example, in
an ego network of five individuals, the maximum number of ties is 10. If only four of those
possible relationships exist, network density for that ego network would be 0.4.
I assessed tie strength using a composite scale developed by Gillis (2008). Participants
responded to three items for each collaborative relationship they identified assessing their
perceived closeness to that individual (on a five-point Likert-like scale, with 1 = “Very Distant”
and 5 = “Very Close) and the frequency with which they went to that individual for both work-
related and other types of advice (on a five-point Likert-like scale, with 1 = “Rarely” and 5 =
“Often”). I calculated scale scores by calculating the mean of the three items, and the scale was
internally reliable for this sample (α=0.79). The extent that each participant collaborated with
faculty was calculated as the proportion of relationships with faculty members to the total
number of relationships with individuals in the participants’ ego networks.
I calculated homophily as the proportion of ties in participants’ ego networks who shared
a given trait with the participant compared to the total number of ties in the network. Participants
were asked to identify, to the best of their knowledge, whether or not collaborators shared their
gender identity and racial/ethnic identity; these items were used to calculate homophily based on
race/ethnicity and gender identity. Participants were also asked a series of questions identifying
whether each collaborator worked in their department, worked in student affairs, and worked in
their same building. These items were used to calculate homophily based on department, student
affairs, and propinquity. Descriptive statistics and descriptions of each of these focal variables
are listed in Appendix 2A.

53
Control variables. In addition to the variables listed above, I included several sets of
control variables pertaining to institutional characteristics, professional characteristics, and
personal demographics (Borgatti & Foster, 2003; Bowler & Brass, 2006; Brass, 1985; Brass et
al., 2004; Ibarra, 1992; Kadushin, 2012; McPherson et al., 2001; Mehra et al., 1998; 2001; Perry-
Smith & Shalley, 2003). Institutional characteristics include information pertaining to
participants’ institutions, including size, Carnegie type, control (i.e., public vs. private) and
perceived institutional climate for collaboration. Participants were asked to identify their current
institution on the survey, and information relating to institution size, control, and Carnegie type
was identified using 2013-2014 Institutional Postsecondary Education Data Systems (IPEDS)
data. I calculated climate for collaboration from an item on the survey asking participants to
assess the climate for collaboration in their institution on a five-point Likert-like scale, with 1 =
“Completely Unsupportive of Collaboration” and 5 = “Completely Supportive of Collaboration.”
When more than one individual from an institution participated in the survey, I aggregated these
scores to the institutional level to signify the climate; when only one individual from an
institution participated in the survey, I utilized that participants’ perception of the climate to
represent the institutional climate. Professional control variables include participants’ rank in
their institution (e.g., entry-level, mid-level, and senior-level/senior student affairs officer),
functional area, and years of experience in both higher education and at their current institution.
Participants were asked to identify the functional area in which they worked from a list of 22
potential functional areas, including an “Other” option in which they were allowed to enter an
area not listed from among the other 21 options. For the sake of parsimony in the analyses, these
functional areas were collapsed into nine umbrella functional areas which perform similar
functions on college and university campuses: activities/student union/recreational activities;

54
advising/career services/academic support; administrative leadership/generalist;
admissions/orientation; multicultural/diversity programs; wellness/health programs; assessment;
residence life/student conduct; and other student affairs areas (representing other areas that do
not fit in the other categories). Personal characteristics included as control variables in the
analyses below include racial/ethnic identity and gender identity. Again, descriptive statistics for
these variables are listed in Appendix 2A.
Data Analysis
Prior to running the models for this paper, I explored the data to identify any patterns in
missing data. Fewer than 10% of participants in the study exhibited missing data, with fewer
than 1% of individual values missing in the dataset, which are well within the suggested range of
missing values to be included in the analyses (Newton & Rudestam, 1999). Most of the missing
values pertained to institutional variables, as some participants did not identify the institution in
which they worked, with few cases missing personal or professional demographic variables.
These were accounted for by case-wise deletion in the random effects models below. Missing
continuous variables were imputed using the expected maximization (EM) algorithm, which
utilizes maximum likelihood techniques to impute missing values.
I utilized principal axis factor analysis, ANOVA, and Pearson r correlations to answer the
first research question regarding SA competencies and differences by professional rank and
functional unit and the relationship between competencies and years of experience in higher
education. As described above, I first utilized factor analysis to identify the broad competency
constructs from the items on the CSNSA. I then used ANOVA with Bonferroni post-hoc tests to
identify any differences in the three competency variables based on professional rank and

55
functional unit. Finally, I calculated the Pearson r bivariate correlation statistics to examine the
relationships between professional competency and years of experience.
I utilized hierarchical linear modeling to answer the second research question regarding
the associations of collaborative behavior with SA competencies after controlling for
institutional, personal, and professional characteristics. The sample exhibits clustering, as some
participants work in the same institutions as other participants in the sample. Network behavior
is influenced by this kind of clustering, violating the assumption of independence (Valente,
2010), which can result in biased standard errors if calculated using ordinary least-square
regression (Rabe-Hesketh & Skrondal, 2012). Therefore, network analysts recommend using
multilevel random effects models to account for this nonindependence. I utilized three multilevel
models to examine the associations of collaborative network behaviors, institutional
characteristics, and personal and professional characteristics with the three SA competency
variables described above. I entered the variables into the models in three blocks in order to
examine the variance accounted for by each block. The first block pertained to institutional
characteristics, followed by the next block pertaining to personal and professional characteristics.
I calculated the change in variance (i.e., R
2
) for each block prior to entering the third block with
the focal network variables. This allowed me to identify the amount of unique variance in the
dependent variables accounted for by these variables after controlling for the institutional and
personal/professional characteristics. I also standardized all continuous variables prior to
including them in the models so that their coefficients could be interpreted as effects sizes.
Limitations
This study exhibits several limitations. The first limitation pertains to the cross-sectional
nature of the data, which precludes me from making causal inferences about the influence of the

56
model covariates on SA competencies. Utilizing longitudinal and/or experimental study designs
could have allowed me to make these claims. However, given the focus of studying
competencies of and collaboration among SA administrators in their natural institutional settings,
it was not possible to design an experimental condition to control for all of the various individual
and institutional factors that could influence these behaviors and perceptions. A longitudinal
approach to studying collaboration networks could allow me to make claims about the dynamic
nature of these networks over time (Borgatti & Halgin, 2011; Monge & Contractor, 2003) and
causal relationships between them and SA competencies. A second limitation is the self-
reporting nature of survey instruments. Participants may not be able accurately assess their true
proficiency in competencies or their behavior and that of others with whom they collaborate.
Scholars have suggested criticisms of self-report in social science research (see Podsakoff &
Organ, 1988; Podsakoff, Mackenzie, & Lee, 2003; Spector, 1994), yet these authors also
acknowledge the importance of self-report surveys in advancing the literature on organizational
behavior. Spector (1994) in particular suggests that the use of self-report questionnaires should
be used when appropriate to the research. Given the goal of this research to gather data from a
representative sample of administrators and make inferences about their collaborative behavior,
survey self-report was the most appropriate method. Finally, this study utilized a specific
approach to assessing SA professionals’ level of competency in their work. Only one other study
has measured competency levels for SA administrators (Sriram, 2014), and this researcher
identified more constructs due the goal of identifying constructs for all 10 SA competencies
(Bresciani et al., 2010). While the factors identified in this study mirror those identified by
Sriram, other approaches to measuring and assessing SA competencies might uncover more
complex and underlying processes involved in professional work.

57
2.5 Findings
Professional Competencies by Rank
Table 2.1 lists the three competency variables for this study – human and organizational
resources, leadership for personal/community development, and organizational leadership – with
their factor loadings, scale reliabilities, and descriptive statistics for the entire sample. On the
whole, the participants in this study perceive themselves to be proficient in these three
competency areas, with means ranging from 5.17 to 5.79, falling in the proficient to very
proficient range of the scale. This is reinforced by the standard deviations for these means, which
indicate little variation and suggest that most participants tend to rate themselves highly in these
three areas. Descriptive statistics, ANOVA results comparing perceived competency in these
three areas, and Pearson r correlations of years of experience to competencies are listed in Table
2.2. First, proficiency level is related to professional rank, as entry-level participants indicate
lower scores than mid-level participants, who in turn indicate lower scores than senior-level
participants, for two of the competency areas – human and organizational resources and
organizational leadership. As for leadership for personal/community development, entry-level
and mid-level participants do not exhibit significant differences in perceived proficiency, but
senior-level participants score significantly higher than both groups.
Turning to differences by functional unit, ANOVA results indicate significant
relationships between functional unit and proficiency in all three competency areas. Bonferroni
post-hoc analyses identify specific group differences. First, participants who work as
administrative leaders or generalists responsible for multiple functional areas score significantly
higher in human and organizational resources than participants from all other functional units
except one (wellness & health programs). Additionally, participants who work in residence life

58

and student conduct programs score significantly higher for this competency compared to three
groups (student activities/recreational programs; multicultural/diversity programs; and
advising/career/academic programs). Turning to leadership for personal/community
development, most of the functional units do not exhibit significant differences in scores.
However, administrative leaders and generalists score significantly higher than participants in
advising/career/academic services and assessment. Finally for organizational leadership,
generalists and administrative leaders score significantly higher than participants from four
functional units – residence life and student conduct; student activities/recreational programs;
multicultural/diversity programs; and advising/career/academic programs. Finally, years of
experience is significantly correlated with all three of the competencies in this study, ranging
from r = 0.25 to r = 0.38, p < .001.
Table 2.2: Relationships of SA Professionals Perceived Competencies to Professional Rank,
Functional Unit, and Years of Experience in Higher Education
Human &
Organizational
Resources
Leadership for
Personal/Community
Development
Organizational
Leadership
Professional Rank
1. Entry-Level 4.71 (0.99) 5.39 (0.95) 5.31 (0.85)
2. Mid-Level 5.01 (0.95) 5.58 (0.91) 5.74 (0.79)
3. Senior-Level/SSAO 5.83 (0.74) 6.07 (0.75) 6.25 (0.64)
F 69.12*** 27.00*** 60.29***
Bonferroni Post-Hoc 1 < 2 < 3 1,2 < 3 1 < 2 < 3
Functional Unit
1. Administrative Leadership/Generalist 5.78 (0.77) 5.96 (0.77) 6.23 (0.70)
2. Wellness/Health Programs 5.46 (0.71) 5.79 (0.85) 5.98 (0.87)
3. Residence Life/Student Conduct 5.42 (0.85) 5.68 (0.83) 5.74 (0.81)
4. Admissions/ Orientation 5.00 (0.91) 5.84 (0.92) 6.03 (0.75)
5. Activities/Student Union/Recreation 4.92 (0.98) 5.70 (0.90) 5.61 (0.84)
6. Multicultural/ Diversity Programs 4.82 (1.06) 5.77 (0.84) 5.69 (0.69)
7. Advising/Career Services/Academic
Support
4.74 (1.01) 5.37 (1.05) 5.56 (0.85)
8. Assessment 4.43 (1.09) 5.24 (9.92) 5.92 (0.64)
9. Other Student Affairs Areas 4.78 (1.29) 5.36 (1.33) 5.57 (1.23)
F 15.02*** 4.02*** 6.93***
Bonferroni Post-Hoc
3, 4, 5, 6, 7, 8, 9 < 1;
5, 6, 7 < 3
7, 8 < 1 3, 5, 6, 7 < 1
Years of Experience
Pearson r Correlations 0.38*** 0.25*** 0.34***
NOTE: *** p < .001

59
Random Effects Models for Student Affairs Competencies
Results from the random effects models predicting proficiency in three organizationally-
related competencies can be found in Table 2.3. After controlling for other variables in the
model, two variables pertaining to collaborative behavior are significantly associated with
proficiency in human and organizational resources – network density and strong ties. Aside from
these collaboration variables, several individual characteristic variables are also significant. First,
male participants are expected to perceive greater proficiency compared to female participants.
Consistent with the prior analyses, both mid-level and senior-level participants are expected to
score higher than entry-level participants after controlling for the other variables. Likewise,
residence life and student conduct professionals are expected to score higher than several other
functional units. None of the institutional characteristics are significantly associated with SA
proficiency in human and organizational resources. The model explains a total of 30% of
variance in this dependent variable, with the majority of that variance accounted for by
individual characteristics (26%).
Turning to the model for leadership for personal and community development, mean tie
strength is positively associated with the dependent variable and homophily by propinquity is
negatively associated with the dependent variable after controlling for other variables. Unlike the
previous model, only one professional characteristic is significantly associated with leadership
for personal and community development after controlling for other variables; senior-level
participants are expected to score significantly higher than entry-level participants. As with the
previous model, no institutional characteristics are significantly associated with leadership for
personal and community development. This model explains a total of 19% of variance in this

60
Table 2.3: Random Effects Models Predicting Student Affairs Administrators’ Perceived Competencies

Human & Organizational
Resources
Leadership for
Personal/Community
Development
Organizational
Leadership
R
2
β SE R
2
β SE R
2
β SE
Institutional Characteristics .02 .02 .03
Institution Size: < 1,000
a
.63 .35 .38 .38 .38 .35
Institution Size: 1,000-4,999
a
.09 .17 -.01 .19 .02 .18
Institution Size: 5,000-9,999
a
-.05 .15 -.12 .17 .04 .16
Institution Size: 10,000-19,999
a
.03 .11 -.12 .12 .08 .11
Carnegie: Masters
b
-.01 .11 .10 .13 -.02 .11
Carnegie: Baccalaureate
b
-.09 .17 .06 .19 .20 .17
Carnegie: Associates
b
-.13 .18 -.08 .21 -.17 .19
Carnegie: Other
b
.05 .22 -.05 .25 .19 .23
Control: Public .07 .10 .00 .12 .08 .11
Institutional Climate for Collaboration .00 .04 .05 .04 .07 .04
Individual Characteristics .26 .11 .21
Male
c
.23* .09 -.09 .09 .05 .09
Transgender
c
-.60 .35 -.68 .38 -1.33
***
.35
Latino/Hispanic
d
-.04 .22 -.13 .23 .14 .22
Asian/Asian American
d
-.20 .24 -.03 .25 .01 .24
Black/African American
d
-.02 .15 .04 .16 -.05 .15
Multiracial
d
-.18 .21 -.24 .22 -.15 .21
Mid-level Professional
e
.25
*
.10 .21 .11 .45
***
.11
Senior-Level Professional/SSAO
e
.61
***
.16 .56
**
.17 .80
***
.16
Functional Area: Activities/Union/Rec
f
-.38
**
.11 .21 .12 -.10 .12
Functional Area:
Advising/Careers/Academic Support
f

-.57
***
.12

-.19 .13

-.17 .12
Functional Area: Administrative
Leadership/Generalist
f

-.21 .13

-.15 .14

-.01 .14
Functional Area: Admissions/Orientation
f
-.30 .17 .34 .18 .37
**
.18
Functional Area: Multicultural/ Diversity
Programs
f

-.34 .18

.15 .19

.13 .18
Functional Area: Wellness/Health
f
.17 .26 .27 .27 .38 .26
Functional Area: Assessment
f
-1.03
***
.22 -.30 .23 .14 .22
Functional Area: Other
f
-.77
**
.30 -.47 .32 -.27 .30
Years at Institution .07 .05 -.05 .06 .13
*
.05
Years in Higher Education .12 .07 .13 .07 -.09 .07
Collaborative Network Variables .02 .06 .04
Ego Network Size -.01 .04 -.03 .04 .03 .04
Ego Network Density .09
*
.04 .03 .04 -.02 .04
Ego Network Mean Tie Strength .10
**
.04 .19
***
.04 .20
***
.04
Ego Network Homophily: Propinquity -.03 .04 -.19
***
.05 -.08 .04
Ego Network Homophily: Department -.04 .05 .07 .05 -.03 .05
Ego Network Homophily: Student Affairs -.03 .04 -.07 .05 -.04 .05
Ego Network Homophily: Gender .00 .04 -.02 .05 .06 .04
Ego Network Homophily: Race/Ethnicity .00 .05 -.08 .06 .05 .05
Collaborate with Faculty -.07 .04 -.00 .04 -.02 .04
R
2
.30 .19 .28
NOTES:
*
p < .05;
**
p < .01;
***
p < .001; Reference Categories for Categorical Variables:
a
20,000+;
b
Doctoral Institutions;
c
Female;
d
White;
e
Entry-level;
r
Residence Life/Student Conduct

61
dependent variable, with the majority of the variance accounted for by individual characteristics
(11%).
Finally, only one collaborative behavioral variable is significantly associated with
organizational leadership after controlling for other variables; mean tie strength in an ego
collaborative network is positively associated with organizational leadership. Several individual
characteristics are significantly associated with this dependent variable. First, transgender
participants are expected to score lower than female participants for organizational leadership
proficiency. Similar to organizational and human resources, mid-level and senior-level
professionals are expected to score higher than entry-level participants. Professionals in
admissions/orientation programs are expected to score higher than residence life and student
conduct professionals, and years worked in higher education is positively associated with this
dependent variable. This model explains a total of 28% of variance in this dependent variable,
with the majority of the variance, once again, accounted for by individual characteristics (21%).
2.6 Discussion
Student affairs professionals are increasingly being urged to assess and improve their
proficiency in the relevant competency areas for SA (Bresciani et al., 2010). Researchers are just
beginning to measure these competencies (Sriram, 2014), and in order for this area of SA
research and practice to evolve, more research is needed to examine the associations of personal
characteristics and professional practice to these competencies. This study addresses this need by
assessing SA professionals’ competencies in three organizationally-related competency areas and
identifying relationships of these competencies to collaborative behavior and individual
characteristics. Much of the work prior to this study simply identified relevant competencies
needed for different professional levels (Burkard et al., 2005; Gordon et al., 1993; Kretovics,

62
2002; Kuk et al., 2007; Randall & Globetti, 1992; Saunders & Cooper, 1999; Waple, 2006) and
functional areas (Pope & Reynolds, 1997; Reynolds, 2011). This study is the first to identify
differences in perceived competency by these characteristics, as well as the first to relate these
competencies to an aspect of professional practice. I highlight the most notable findings below.
Strength of Strong Ties and Other Network Indicators
Personal characteristics account for largest the proportion of variance in the dependent
variables in this study, and the collaborative network behaviors accounted for relatively little
variance in these three outcomes. SA professionals are being called to specifically collaborate
across boundaries in their institutions, yet collaborating outside of one’s department, outside of
SA, and with faculty are not significantly related to these three competencies. While these
behaviors likely influence other aspects of institutional effectiveness, it is clear from this
research that they are not associated with SA professionals developing proficiency in these three
organizationally-related competencies.
Despite relatively little variance explained by these variables compared to personal
characteristics, the random effects models reveal valuable information about a specific form of
collaborative behavior. Having a greater proportion of strong ties in SA professionals’
collaborative networks is positively associated with all three competencies in this study. Social
network researchers identify benefits of both strong and weak ties in organizations. In the case of
SA professionals, it appears that forming stronger relationships on average with one’s
collaborators, based on perceived closeness and advice seeking, is a benefit as it relates to SA
professionals’ competencies to lead and manage resources important to organizational
functioning. Strong ties foster trust and allow information to flow freely among individuals who
have these ties (Granovetter, 1973; 1982; Hoppe & Reinelt, 2010) and are associated with

63
altruism and interpersonal citizenship (Bowler & Brass, 2006; Settoon & Mossholder, 2002;
Sparrowe, Liden, Wayne, & Kraimer, 2001). The findings in this study specifically related to the
two leadership competencies reinforce these prior findings from the literature, as they speak to
one’s ability to mentor and develop others and groups while being able to influence change
through these relationships in one’s organization. Additionally, as SA professionals develop
stronger relationships with their collaborators, they likely feel better equipped to handle
organizational issues that arise given the trust they feel through their connections across the
organization and potentially learn more about practice through more frequent interaction with
their network connections. This finding reveals a complexity of collaboration for SA
professionals and the benefits that can come to individuals if they invest the time and energy to
strengthen their collaborative connections.
Beyond strong ties, two other collaborative behaviors are related to specific
competencies. First, engaging in denser collaborative networks is related to greater perceived
competencies in human and organizational resources, suggesting that SA professionals who
engage in more closed networks of collaborators feel more able to handle important issues
pertaining to the organization. Closed networks exhibit more trust among individuals within the
network compared to more open networks (Balkundi & Kilduff, 2006), which coupled with the
strong ties finding above could suggest the importance of trust in collaborative relationships in
contributing to SA professional feeling proficient in their work specifically related to
organizational competencies.
While dense ego networks are associated with the human and organizational competency
area, greater propinquity homophily is negatively associated with leadership for personal and
community development. Other findings from this research (Gehrke, 2015) reveal that

64
propinquity homophily is also negatively associated with collaboration across organizational
boundaries. Propinquity exhibits a strong pull on social behavior and can inhibit connections
forming beyond geographic and physical boundaries in organizations (Borgatti & Cross, 2003;
Brass et al., 2004; Festinger, Schachter, & Back, 1950; Kadushin, 2012; Monge & Contractor,
2003), and this finding reinforces this pull among SA professionals. If administrators find that
they are unable to reach beyond the confines of their physical location on a campus, they will
likely not feel they are as able to have an influence on the personal and community development
of those around them. These findings in tandem begin to paint a picture of the detriment of
propinquity in contributing to important behaviors and outcomes important to the profession.
Competencies and Professional Characteristics
The next notable finding is that student affairs competencies are related to professional
rank, functional unit, and years of experience in higher education. For all three competencies
assessed in this study, senior-level professionals scored significantly higher than those in mid-
level and entry-level positions. Additionally, mid-level professionals scored significantly higher
than entry-level professionals in two of the competency areas in this study. These relationships
are inherently intertwined with professional experience in higher education. Years of experience
is significantly correlated with all three competency areas, and further analyses reveal that on
average, entry-level participants in this study have worked in higher education for 4.35 years,
while mid-level participants have 12.32 years of experience and senior-level participants have
24.58 years of experience on average, which is also a statistically significant relationships [F(2,
637) = 356.92, p < .001]. So while years of experience exhibit a moderately weak relationship
with the three competencies, these further analyses reveal substantial differences in the years of
experience for each of these levels in the organizational hierarchy.

65
When these sets of variables are included in random effects models examining
proficiency in these competency areas, years of experience in higher education is no longer
significantly associated with the outcomes but professional rank remains significant with fairly
large effects, especially for senior-level professionals compared to entry-level professionals.
These results suggest that the higher one moves in the organizational hierarchy, which requires
substantial years of experience to do based on the analyses above, the more proficient one
becomes in these competency areas. This result is not all that surprising, but empirically
documenting such relationships is important to understand the dynamics of developing
competency in student affairs. Interestingly, despite the significant association between rank and
competencies, differences in these competencies still occur toward the higher end of the
measurement scale, with entry-level professionals still indicating feeling proficient or nearly
proficient in these competency areas on average. The phenomenon of social desirability suggests
that survey participants tend to over-state their abilities in survey research (DeMaio, 1984;
Kreuter, Presser, & Tourangeau, 2008), which suggests that more variation probably exists
between these participants in reality, but these findings are still promising for a profession in
which we expect all professionals to exhibit some level of competency in these areas.
Functional unit is also related to competencies in student affairs in a way that is
intertwined with professional experience. For all of the three competencies in this study,
professionals who identify as administrative leaders and generalists exhibited significantly higher
proficiency levels compared to professionals from between two and seven other functional units.
Of the 112 participants in the sample who identify with this functional unit, 96 (85.7%) identify
as senior-level professionals. Most positions in student affairs at the entry and mid-level of the
organizational hierarchy work within a specific functional area, but as professionals work their

66
way up in the hierarchy they tend to oversee multiple functional areas or are generally
responsible for multiple duties in their work. Again, these findings document significant trends
in competencies based on organizational structures to better understand dynamics of competency
development in SA.
The relationships of human and organizational resources to working in residence life and
student conduct programs identified in the ANOVA models play out in the random effects
models once I control for other variables. These professionals are expected to score significantly
higher than their colleagues in student activities/recreational programs; advising/career/academic
programs; and assessment after controlling for other variables. This suggests that working in this
area probably requires much more navigation of organizationally-related issues compared to
these other areas, which are beneficial to these professionals. Residence life work in particular
requires professionals at all levels to manage budgets and crises on a fairly regular basis, and the
benefits of such demands are evident in the results of the ANOVA and random effects models.
Implications for Practice
This research reveals several implications for student affairs practice. First, the
importance of strong ties and the detriment of propinquity homophily in particular suggest
implications for practice. Developing stronger ties at work can increase trust in organizations,
and this research highlights further benefits of these ties on SA professionals’ perceived
proficiency in their work related to organizational effectiveness. Practitioners from across the
organizational hierarchy should consider the ways in which they could strengthen their
relationships at work, and unit leaders and supervisors in particular should foster professional
development and provide opportunities for their administrators to develop stronger relationships.
These strategies could include increased committee involvement opportunities or retreats that

67
bring together individuals from across campus who may not have formal reasons to work
together collaboratively in order to strengthen these relationships. These strategies can also serve
to break down geographical barriers on campus by bringing professionals outside of their
buildings and locations in order to interact more frequently and form connections across their
physical campus.
Individuals who are higher in the hierarchy, work as generalists or oversee multiple
functions, and have more years of experience in higher education report greater competency in
human and organizational resources and two leadership areas. In one sense, practitioners could
view these results as ones of maturation and assume that they will develop more in these
competencies as they work their way up in the hierarchy and work longer in SA. However, this
research identifies the wisdom that could come from these more experienced individuals. Entry-
level professionals and those who work in singular functional units have individuals on their
campus who can serve as mentors and role models. For those who feel like they have less
proficiency in these areas, seeking out mentorship from these more senior members of their
campus community could provide additional development opportunities for people to strengthen
their proficiencies in SA practice. This research also has implications for graduate preparation
programs, which can perhaps focus more intentionally on these outcomes, knowing that entry-
level professionals feel less proficient in these organizationally-related competencies. Given
these results, in tandem with prior research indicating that entry-level professionals in particular
are less prepared to handle administrative aspects of SA work (Burkard et al., 2005; Kretovics,
2002; Kuk et al., 2007; Waple, 2006), providing opportunities for students to specifically reflect
on the skills needed to effectively manage organizational resources and lead in organizations in

68
tandem with internship and practicum experiences could further prepare them for these important
areas of practice.
Implications for Research
The findings for this study also carry several implications for research related both to SA
competencies and collaborative network behavior. The findings in this study specifically related
to the strength of strong ties for competency identify a specific aspect of collaborative behavior
that may influence how SA professionals develop competency in this work. While the
relationships were identified in this quantitative study, future qualitative research should seek to
better understand the dynamics of strong collaborative relationships. Questions such as how do
strong relationships develop among collaborators in higher education, and what underlying
processes inherent in these strong relationships contribute to SA professionals perceived
proficiency in their work will help scholars and practitioners better understand why developing
strong relationships in can help them in their practice beyond the personal benefits of close
relationships. Similar studies could also seek to understand other phenomenon from this study,
such as the relationship of dense collaborative networks with greater proficiency in human and
organizational resources. Ultimately, mixed methods research in which researchers can identify
quantitative relationships and explore underlying causes through qualitative analyses with those
participants will contribute to uncovering the dynamics of these network behaviors and
competencies.
Turning to competencies, this study contributes to and advances the research of
measuring and evaluating SA competencies, and it is the first one to compare several of the
competency areas agreed upon by the preeminent SA professional associations based on
individual rank, functional area, and years of experience in higher education. The findings in this

69
study reveal that organizationally-related SA competencies are related to structural aspects of SA
work and highlight how those higher in the hierarchy and in generalist positions perceive greater
proficiency, yet these are only a subset of the competencies necessary for effective SA practice
(Bresciani et al., 2010). Future research should seek to examine if these structural relationships
exist for the remaining competencies identified in order to better understand if other factors may
influence perceived competency beyond these aspects of professional work. Additional research
should also identify other factors that may influence perceived competency, such as graduate
preparation and professional development, to identify other experiences that may help those
lower in the hierarchy and in other functional units to make strides in their work.
2.7 Conclusion
SA professionals in recent years have been made more aware of the expectation for them
to master a core set of professional competencies in order to be effective practitioners. These
professionals are also being called to collaborate in their work in order to influence institutional
effectiveness. The research in this study advances knowledge about these two movements in SA
work by measuring and identifying differences in perceived organizational competencies among
administrators while also identifying aspects of collaborative behavior related to these
competencies. While those who are higher in the hierarchy and who have more experience in the
profession communicate more proficiency in these competency areas, this study highlights the
benefit of strong relationships with collaborators to contribute to one’s perceptions and feelings
of competency. While this study highlights trends and identifies certain relationships, the study
of student affairs competencies is still in its infancy, with much more inquiry necessary to
identify other professional behaviors that may relate to these and other competencies important
for SA practice.

70
2.8 Appendix 2A: Descriptive Statistics for Other Variables in this Study
Mean S.D. Min Max Description
Institutional Characteristics
Institution Size: < 1,000 0.01 0.00 1.00 1 = <1,000; 0 = All others
Institution Size: 1,000-4,999

0.25 0.00 1.00 1 = 1,000-4,999; 0 = All others
Institution Size: 5,000-9,999 0.11 0.00 1.00 1 = 5,000-9,999; 0 = All others
Institution Size: 10,000-19,999

0.23 0.00 1.00 1 = 10,000-19,999; 0 = All others
Institution Size: 20,000+ 0.39 0.00 1.00 1 = 20,000+; 0 = All others
Carnegie: Doctoral 0.49 0.00 1.00 1 = Doctoral; 0 = All others
Carnegie: Masters 0.28 0.00 1.00 1 = Masters; 0 = All others
Carnegie: Baccalaureate

0.14 0.00 1.00 1 = Baccalaureate; 0 = All others
Carnegie: Associates 0.05 0.00 1.00 1 = Associates; 0 = All others
Carnegie: Other

0.04 0.00 1.00 1 = Other; 0 = All others
Control: Public 0.61 0.00 1.00 1 = Public; 0 = All others
Control: Private 0.39 0.00 1.00 1 = Private; 0 = All others
Institutional Climate for
Collaboration
3.99 0.88 1.00 5.00 5-point scale; 1 = Completely
unsupportive; 5 = Completely supportive
Individual Characteristics
Female 0.61 0.00 1.00 1 = Female; 0 = All others
Male 0.37 0.00 1.00 1 = Male; 0 = All others
Transgender 0.01 0.00 1.00 1 = Transgender; 0 = All others
Latino/Hispanic 0.03 0.00 1.00 1 = Latino/Hispanic; 0 = All others
Asian/Asian American 0.03 0.00 1.00 1 = Asian/Asian American; 0 = All others
Black/African American 0.09 0.00 1.00 1 = Black/African American; 0 = All others
White

0.80 0.00 1.00 1 = White; 0 = All others
Multiracial

0.04 0.00 1.00 1 = Multiracial; 0 = All others
Entry-level Professional 0.21 0.00 1.00 1 = Entry-level; 0 = All others
Mid-level Professional 0.51 0.00 1.00 1 = Mid-level; 0 = All others
Senior-Level Professional/SSAO 0.28 0.00 1.00 1 = Senior-level; 0 = All others
Functional Area: Residence
Life/Student Conduct
0.28 0.00 1.00 1 = Residence Life/Student Conduct; 0 =
All others
Functional Area:
Activities/Union/Rec
0.19 0.00 1.00 1 = Activities/Union/Rec; 0 = All others

Functional Area: Advising/
Careers/ Academic Support
0.17 0.00 1.00 1 = Advising/Careers/Academic Support;
0 = All others
Functional Area: Administrative
Leadership/Generalist
0.18 0.00 1.00 1 = Administrative Leadership/Generalist;
0 = All others
Functional Area:
Admissions/Orientation
0.05 0.00 1.00 1 = Admissions/Orientation; 0 = All others
Functional Area: Multicultural/
Diversity Programs
0.05 0.00 1.00 1 = Multicultural/Diversity Programs; 0 =
All others
Functional Area:
Wellness/Health
0.02 0.00 1.00 1 = Wellness/Health; 0 = All others
Functional Area: Assessment 0.03 0.00 1.00 1 = Assessment; 0 = All others
Functional Area: Other

0.02 0.00 1.00 1 = Other Functional Area; 0 = All others
Years at Institution 7.16 7.19 1.00 41.00 Units in years
Years in Higher Education 14.10 9.94 1.00 41.00 Units in years

71
Mean S.D. Min Max Description
Collaborative Network Variables
Ego Network Size 6.30 1.31 0.00 7.00 Number of collaborators in ego network.
Ego Network Density 0.63 0.24 0.00 1.00 Proportion of number of connections
among ego’s collaborators to the total
possible number of connections; ranges
from 0 (least dense) to 1 (most dense)
Ego Network Mean Tie Strength 3.15 0.55 1.00 5.00 The average strength of relationships ego
has to collaborators in network; 5-point
scale; 1 = Weak Tie; 5 = Strong Tie
Ego Network Homophily:
Propinquity
0.49 0.31 0.00 1.00 The proportion of collaborators who work
in ego’s building; 0 = No collaborators
from ego’s building; 1 = All collaborators
from ego’s building
Ego Network Homophily:
Department
0.53 0.32 0.00 1.00 The proportion of collaborators from
ego’s department; 0 = No collaborators
from ego’s department; 1 = All
collaborators from ego’s department
Ego Network Homophily:
Student Affairs
0.70 0.28 0.00 1.00 The proportion of collaborators who work
in student affairs; 0 = No collaborators
from student affairs; 1 = All collaborators
from student affairs
Ego Network Homophily:
Gender

0.55 0.25 0.00 1.00 The proportion of collaborators who
share ego’s gender identity; 0 = No
collaborators share gender identity with
ego; 1 = All collaborators share gender
identity with ego
Ego Network Homophily:
Race/Ethnicity
0.69 0.32 0.00 1.00 The proportion of collaborators who
share ego’s race/ethnicity; 0 = No
collaborators share race/ethnicity with
ego; 1 = All collaborators share
race/ethnicity with ego
Collaborate with Faculty 0.04 0.11 0.00 1.00 The proportion of collaborators who
identify as faculty; 0 = No faculty in ego
network; 1 = All faculty in ego network

72
CHAPTER 3 – Navigating the Rapids: Network Behaviors and Strategies of Systemic
Leaders in Three Collaborative Social Networks in Student Affairs

3.1 Introduction

An underlying assumption of student affairs practice is that the context of national higher
education is dynamic and ever changing. The pace of education has escalated and promises to
stay at or increase in speed and intensity. Some have characterized this state as ‘permanent
white water’ (Manning, Kinzie, & Schuh, 2006, p. 148).

Higher education institutions today are faced with an increasing set of uncertainties,
brought about by factors such as diversifying student bodies, globalization, withering budgets,
increasing calls for accountability, safety and mental health concerns, and disruptive technology
(Kuk, Banning, & Amey, 2010; Task Force, 2010; Woodard, Love, & Komives, 2000). The
image of permanent white water is an apt metaphor for the current context in which higher
education and student affairs professionals find themselves, as they are forced to navigate, adapt,
and respond to these complex problems in order to ensure student and institutional success
(Manning et al., 2006; Task Force, 2010). In order to respond to these conditions, higher
education institutions must be flexible and adaptive, necessitating that professionals engage in
systemic leadership that cuts across organizational boundaries and understand the
interconnectivity of all functions and individuals within the organization (Allen & Cherrey,
2000; Kuk et al., 2010; Love & Estanek, 2004). These systemic leaders can serve as river guides,
helping their institutions navigate through the rough waters facing higher education today.
The preeminent professional associations in student affairs (SA) have issued calls-to-
action over the past two decades calling for SA professionals to exhibit this type of leadership,
advocating for practice that cuts across the campus by coordinating services, forming
partnerships, and cultivating relationships to enhance learning for all students (ACPA, 1994;
Joint Task Force, 1998; Keeling, 2004; Task Force, 2010). These calls for leadership are also

73
echoed by scholars in the field (Amey, 2006; Caple & Newton, 1991; Kuk et al., 2010; Manning,
1996; Manning et al., 2006; Roberts, 2001; Rogers, 2003; Roper, 2002; Winston, Creamer, &
Miller, 2001). As Roberts (2001) states: “There is little hope to deal effectively with campus
challenges unless the goal of leadership becomes one of shared responsibility for decisions and
actions. Sharing responsibility means including all voices of fellow professionals…regardless of
position, authority, or power” (p. 384). Consequently, new SA practitioners identify developing
leadership skills as a top professional development goal and concern (Cilente et al., 2007; Renn
&Hodges, 2007), suggesting that understanding factors and behaviors that contribute to effective
leadership can inform practice and professional development for these and all SA professionals.
While SA professionals are increasingly being called to engage in effective cross-
institutional leadership, little is known about these leaders and the behaviors which contribute to
their practice. In general, little attention is paid to SA staff and organizations in research; most
research emphasizes the college student experience (Kuk et al., 2010). Also, despite pervasive
calls for leadership in student affairs (Clement & Rickard, 1992), research on SA leadership is
“virtually non-existent” (Smith & Hughey, 2006). When Clement & Rickard (1992) published
their influential study of effective leadership in SA, they made several observations about the
state of research on leadership that still hold relatively true today – the literature was fragmented;
it focused on career paths and program administration, “separating functional roles from
leadership and viewing style as independent of situation” (p. 9); and it did not take into account
the experience of leaders from across the organizational hierarchy, focusing mainly on upper-
level administrators. Since their study, the landscape for leadership research in SA is not much
improved, and the same critiques can be applied to current research.

74
Collaboration is one of the most common ways for professionals to connect across higher
education institutions and entails individuals agreeing on mutual goals and relying on each other
to accomplish these goals (Kezar & Lester, 2009). Calls for effective leadership cited above
often frame it in terms of collaboration across institutions and espouse the necessity of
collaboration in higher education for institutions to be effective. As Kuk and colleagues (2010)
aptly state: “Collaboration and sharing information, resources, and expertise are critical
components of everyday life. It is no longer effective to stay within one’s functional unit and
expect to meet the demands and challenges that are being presented” (p. 22). Collaborative
relationships are thus an appropriate lens for identifying SA professionals who are engaging in
effective leadership. A useful theoretical and conceptual framework for operationalizing,
identifying, and examining this leadership behavior is social network theory. Specifically, social
network theory provides the tools to identify central individuals who play influential leadership
roles in overall structures of relationships by mapping collaboration networks of SA
professionals within higher education institutions (Daly, 2010a; Valente, 2010)
The purpose of this paper is to identify and explore behaviors and strategies associated
with systemic leadership
6
in collaborative social networks in three student affairs divisions. This
inquiry fills gaps in the literature by examining how specific collaborative behaviors of SA
professionals are associated with indicators of leadership in their institutions’ collaboration
networks and identifying behaviors and strategies employed by these individuals as they form
collaborative relationships. The specific research questions guiding this study are:
1. What collaborative network behaviors are associated with systemic leadership in SA
collaboration networks?

6
Defined in next section.

75
2. What do systemic leaders and their collaborators identify as being important to their
abilities to effectively collaborate in their institutions? How do these perceptions differ
from those for non-central individuals in the network?
3.2 Theoretical Frameworks
Two theoretical frameworks inform this study of leadership in collaborative networks:
systemic leadership theory and social network theory. Systemic leadership theory (Allen &
Cherry, 2000) is informed by complexity theory and suggests that organizations are not
predictable entities that can be controlled through hierarchical, bureaucratic practices; rather,
they are systems of interconnected networks that enable organizational effectiveness through
mutual influence and engagement of individuals across organizational boundaries (Heifetz, 1994;
Marion & Uhl-Bien, 2001; Uhl-Bien, Marion, & McKelvey, 2007; Wheatley, 1992). The second
is social network theory and analysis, which moves beyond examining the traits and behaviors of
individuals to focus on the nature of relationships between and among them (Borgatti & Ofem,
2010). These two frameworks are relevant as they contextualize and conceptualize leadership as
embedded in systems of relationships that cut across organizational hierarchies and boundaries
and provide strategies for operationalizing these behaviors. I briefly review the key tenets of
these theories before reviewing the literature on leadership in student affairs.
Systemic Leadership
In the past few decades, researchers and theorists have characterized leadership as
process-oriented, non-hierarchical, and collaborative, with the all members of an organization
holding the potential to practice leadership (Kezar et al., 2006; Komives & Dugan, 2010;
Komives et al., 2007; Rost, 1991), rather than an activity practiced by individuals in formal
leadership positions exerting power and influence through hierarchical structures (Kezar et al.,
2006; Komives & Dugan, 2010; Komives et al., 2007; Rost, 1991). These approaches to
leadership coincide with emerging views of the nature of organizations informed by complexity

76
theory (Axelrod & Cohen, 2000; Handy, 1999; Heifetz, 1994; Marion, 1999; Phillips & Hunt,
1992; Wheatley, 1992). Complexity theorists acknowledge the dynamic and unpredictable nature
of organizational systems and the role of leadership in trying to influence, rather than control,
those systems (Allen & Cherrey, 2000; Komives et al., 2006; Marion & Uhl-Bien, 2001; Uhl-
Bien et al., 2007). This reflects a networked perspective on organizations as constantly changing
and influenced by systems of relationships across them (Allen & Cherrey, 2000).
Leadership scholars utilizing complexity theory tend to be situated in the organizational
literature and challenge traditional notions of hierarchy in organizations, arguing for leadership
that is adaptive and collaborative (Heifetz, 1994; Kezar et al., 2006; Marion & Uhl-Bien, 2001;
Uhl-Bien et al., 2007; Wheatley, 1992). The literature explicitly connecting complexity theory
and student affairs leadership is limited, though several scholars have advanced notions of this
approach to leadership for student affairs. Systemic leadership theory
7
situates leadership in the
broader system of relationships within organizations (Allen & Cherrey, 2000; Love & Estanek,
2004; Wielkiewicz, 2000). A systems perspective on leadership suggests that:
One cannot make sense of relationships and connections by looking at a small part of the
system. Relationships and connections need to be viewed holistically…systems thinking
helps us to understand the interconnections between the individual, other human beings,
the organization, and the environment (Allen & Cherry, 2000, p .84).
Systems thinking views organizations as much more than the formal hierarchy represented by the
organizational chart; rather, organizations are “webs of relationships” (Love & Estanek, 2004, p.
21) in which individuals collaborate around core values and common goals. Central tenets of
systemic leadership include 1) that higher education institutions must capitalize on the potential
expertise of all organizational members and facilitate the flow of information among them; 2)
anyone within the organization can practice effective leadership and exert influence; and 3)

7
I utilize systemic leadership theory to encapsulate ideas in the student affairs literature included in pervasive
leadership (Love & Estanek, 2004) and ecological leadership (Wielkiewicz, 2000).

77
effective leadership involves making and facilitating connections across the institutional
boundaries, utilizing the systems of relationships to influence positive change, and facilitating
learning with others through dialogue, communication, and collaboration (Allen & Cherrey,
2000; Love & Estanek, 2004). These tenets of systemic leadership informs this study by
acknowledging leadership can be practiced by anyone in the organizational hierarchy, that it
entails individuals making connections across boundaries, and that one must understand the
broader systems of relationships in an organization to understand systemic leadership.
The chief criticisms of systemic leadership theory are that it is largely conceptual in
nature and that it is a difficulty theory to empirically test (Kezar et al., 2006; Komives et al.,
2007; Satish & Streufert, 1997; Streufert, 1997). A theoretical perspective that can contribute to
overcoming these criticisms and advance research on systemic leadership is social network
theory; its tenets provide a rich complement to informing and operationalizing research on
systemic leadership (Lichtenstein et al., 2006).
Social Network Theory
Identifying and measuring the extent to which individuals hold central positions is a
focus of network research (Valente, 2010), and it is an appropriate lens for this inquiry given the
similarities between network centrality measures and systemic leadership practice. Examining
the associations between other network behaviors described below to these leadership measures
will further help identify their relationship to systemic leadership practice. However, critiques of
social network theory suggest just examining structure may not be enough, which I address at the
end of this section.
Social network theory and research is predicated on four central principles that
distinguish it from rival theoretical and empirical approaches (Balkundi & Kilduff, 2006;

78
Kilduff, Tsai, & Hanke, 2006). These are 1) the importance of relationships among
organizational members, 2) the embeddedness of organizational networks in the larger social
world, 3) the utility of social network connections, and 4) the structured patterning of social life.
These four principles resonate with tenets of systemic leadership theory for student affairs. First,
social network analysis emphasizes relationships connecting individuals over the individuals’
attributes (Balkundi & Kilduff, 2006). This resembles the emphasis in systemic leadership on the
importance of relationships to the process of leadership (Allen & Cherrey, 2000; Love &
Estanek). Second, “human behavior is seen as embedded in networks of interpersonal
relationships” (Balkundi & Kilduff, p. 420) of friendship and acquaintance. Prominent student
affairs scholars call attention to the fact that genuine relationships in leadership that go beyond
the utility of the organizational exchange are central to effective systemic leadership (Roberts,
2001). Third, the utility of social network connections is to build and use social capital. Systemic
leadership theorists acknowledges the importance of utilizing connections to influence the larger
system within organizations (Allen & Cherrey, 2000; Marion & Uhl-Bien, 2001). Finally, social
network analysis emphasizes the structured patterning of social relations and the importance of
understanding the interplay between individuals’ positions in the larger structure and the overall
structure of individuals in organizations (Balkundi & Kilduff, 2006; Monge & Contractor, 2003).
This emphasis on structure is reflected in systemic leadership’s focus on understanding both the
micro and macro-level of relationships (Hannah & Lester, 2009; Marion & Uhl-Bien, 2001).
Social network theory utilizes many theoretically and empirically-tested concepts to
understand and operationalize human behavior, and the measures that are most pertinent to this
study of systemic leadership are centrality measures, which reveal a great deal about the role
played by individuals in their informal organization networks (Daly, 2010a; Kadushin, 2012;

79
Valente, 2010); Brass and Krackhardt (1999) suggest that “centrality is the key component to
social capital and leadership in organizations” (p. 183). In social network research, more central
individuals in informal networks are seen as leaders who have more power and influence in
organizations (Balkundi & Kilduff, 2006) and play important roles in diffusing innovation
through them (Valente & Davis, 1999; Valente & Pumpuang, 2007). Three centrality measures
in particular – degree, betweenness, and integration – provide valuable information for the roles
that individuals play in their organizational networks related to systemic leadership (summarized
in Table 3.1).
Degree centrality is a measure of how many connections someone has in a network and
is measured on both out-degree (the number of individuals a person nominates in a given
relationship) and in-degree (the number of individuals who nominate someone in a network;
Valente, 2010). Individuals who are nominated by more people (in-degree) hold an influential
role in the network because more people view them as someone to turn to for information and
advice. Someone with high in-degree indicates an individual who is connected to more members
of the network compared to someone with lower in-degree, exhibiting more systemic leadership
by being connected to more members of the organization. Betweenness centrality is a measure of
“the probability that a path from any two actors takes a particular path” that includes the
individual (Spillane, Healey, & Kim, 2010). Network members high on betweenness lie on more
paths connecting others across the organization and can often connect disparate parts of the
network (Burt, 1992). Betweenness can serve as a proxy for one’s strategic position in a network
based on the ability to connect others in a broader organizational context (Valente, 2010). High
betweenness values indicate someone who is practicing systemic leadership by serving as a
bridge or connector to others in the organization. Integration measures the degree to which

80
someone “…is connected to many and diverse others in a network. An integrated individual can
be reached by many others [in the organization] rapidly” (Valente & Foreman, 1998). Also
described as a measure of closeness, integration indicates the social proximity of an individual to
all others in the network, and higher integration suggests that someone has more access to the
information and resources of others in an organization. A high integration score is indicative of
an individual practicing systemic leadership due to their social proximity and potential for
influencing others in the network through their relationship structures.
Other concepts important to understanding the nature of collaborative leadership in
organizations include density, constraint, tie strength, and homophily (also summarized in Table
3.1). Density refers to the extent to which ego networks
8
are comprised of people who are also
connected to one other (Balkundi & Kilduff, 2006; Valente, 2010), while constraint uses more
information from each alter’s ego network to measure the extent to which each alter’s alters are
connected to one another. Constraint, therefore, measures the extent to which an individual is
closed off to the rest of the network based on the constraining relationships of the individual’s
connections (Valente, 2010). From a systemic leadership perspective, individuals should find
value in ego networks that are less dense and constrained, because they will indicate access to a
wider array of information and opinions.
Tie strength describes the quality (based on frequency of interaction and perception of the
relationship) of connections between individuals and their network connections (Brass &
Krackhardt, 1999; Gillis, 2008; Krackhardt, 1992; Valente, 2010). Information is transmitted
across organizations more freely through strong ties, in which individuals have more frequent

8
Networks surrounding an individual are referred to as ego networks in the network literature, and each individual
connected to ego is referred to as an alter.

81
interaction and trust with their network connections (Granovetter, 1973, 1982; Hoppe & Reinelt,
2010), while weak ties can serve as bridges of communication across organizations, connecting
disparate groups and facilitating diffusion of information, practice, and innovation (Burt, 1992).
From a systemic leadership perspective, ego networks should have both strong and weak ties in
order to facilitate broader access to information and connection to others across the organization
while still developing trust.
Homophily refers to the extent to which individuals form relationships with others who
are similar to themselves (Lazarsfeld & Merton, 1954; McPherson, Smith-Lovin, & Cook, 2001).
Homophily can be measured by personal demographics (i.e., gender, race) and professional
characteristics (i.e., hierarchical rank, functional unit, physical geographic location in an
institution). Systemic leadership is likely exhibited in ego networks with less homophily,
indicating that an individual is making connections with a broad cross-section of individuals in
the organization, both relating to functional areas as well as personal characteristics like gender.
Table 3.1: Social Network Theory and Systemic Leadership
Social Network Measure Connection to Systemic Leadership
Network Centrality: Degree High in-degree in the organizational network indicates that others
see the individual as influential in the organization
Network Centrality: Betweenness High betweenness in the organizational network indicates that this
individual serves as a connector or bridge between others in the
organization
Network Centrality: Integration High integration score in the organizational network indicates close
social proximity and potential for influencing others in the network
through an individual’s relationship structures
Ego Network: Density Less density in ego network indicates a network that is not closed
off to others in the organization
Ego Network: Constraint Less constraint in ego network indicates a network that has access
to rest of the organization because one’s connections are not
inhibiting access to others based on their relationships
Ego Network: Strength of Ties Presence of some weak ties in ego network indicates access to
information and potential to influence others; presence of some
strong ties indicates trust developing among individuals in network
Ego Network: Homophily Less homophily based on departmental affiliation and physical
location on campus indicates boundary-spanning relationships;
less homophily based on personal demographics indicates ability
to connect with diverse range of individuals

82
Social network theory’s emphasis on structure, while not focusing on underlying
dynamics of these structures, has generally limited its application in organizational research
(Kadushin, 2012; Uhl-Bien, 2006; Valente, 2010):
Until now, network theory has appeared to be concerned with description (e.g., who talks
to whom, who is friends with whom) and taxonomy (e.g., friendship network, advice
network, ego network) or relational links, focusing primarily on “mapping” network
interconnections (e.g., identifying the number and types of links that occur among
individual actors), rather than on how relational processes emerge and evolve – e.g., how
these interpersonal relationships develop, unfold, maintain, or dissolve in the context of
broader relational realities (including other social constructions) (Uhl-Bien, 2006, p.
660).
Given these critiques, social network research in organizations should take into account not only
individual attributes in conjunction with relationships but should also consider the underlying
processes involved in relationship forming in organizations. One of the aims of this study is to
account for this critique by assessing network members’ perceptions of strategies for successful
collaboration in order to explore the underlying processes involved in this phenomenon. I now
turn to literature on leadership in SA to highlight the current state of the research and identify its
gaps pertinent to this study.
3.3 Leadership in Student Affairs
The “first comprehensive study of leadership [using] a broad cross section of leaders
working in a variety of student services roles and functions” (Clement & Rickard, 1992, p. 9)
utilized a peer nomination process to identify professionals who had exhibited effective
leadership in the field. Clement and Rickard identified four topics that emerged related to
effective leadership: 1) attributes, skills, and institutional conditions necessary to lead
effectively; 2) essential relationships with staff, faculty, and students that lead to success; 3)
challenges faced by leaders; and 4) the importance of an ethic of care in leading. This study was
crucial in advancing the research on leadership in student affairs by filling some of the gaps

83
described by the authors by contextualizing leadership within the challenges of leaders’
institutions, seeking participants who represented a broader swath of leaders in higher education,
and relating leadership abilities to the roles these leaders held. However, it was still limited in
recruiting participants who held a relatively high position within the organizational hierarchy
(described in current literature as senior student affairs officers), essentially equating leadership
with position. Additionally, despite contextualizing participants’ leadership experience in their
institutions and departments, the leaders’ perspectives highlighted are theirs alone, with no
attempt to examine the perspectives of others who work with the participants to assess the true
effectiveness of these practices.
Most studies of SA leadership since Clement and Rickard’s (1992) study utilized either
behavioral/skills or functionalist/position-based approaches to their research, with most utilizing
a behavioral approach to studying leadership (e.g., Coffey, 2010; Daniel, 2011; Fey & Carpenter,
1996; Garcon, 2012; Kane, 2001; Katherine, 2011; Oh, 2013; Smith, 2013; Taylor, 2001; Tull &
Freeman, 2011). These studies tend to classify participants based on their preferred abstract
leadership styles or competencies. As examples, Taylor (2001), Daniel (2011), Oh (2013), and
Smith (2013) all identified that SA professionals hold affinity toward leadership that empowers
other to act above other strategies. Others take a functionalist approach by focusing on senior
student affairs professionals (Elkins, 2006; Macchio, 2012; Oh, 2013; Sandeen, 1991; 2000;
Schuh, 2002; Smith, 2013; Taylor, 2001), limiting the purview of leadership to the upper levels
of the organizational hierarchy. These authors argue for the importance of studying senior
leaders, as they set the direction for the larger organization and must know how to motivate
followers (Oh, 2013), yet they are limited by not acknowledging that leadership practice does not

84
need to be tied to organizational structure (Allen & Cherrey, 2000; Amey, 2006; Kuk et al.,
2010; Love & Estanek, 2004).
In general, these studies of leadership are limited from a systemic leadership perspective
because they rely on abstract statements and concepts when classifying leadership behavior, are
not ground in the context of the leadership settings, rely predominantly on self-report and do not
take into account perspectives of others with whom these individuals work (with some
exceptions – Kane, 2001; Kezar & Lester, 2011; Taylor, 2001), and focus predominantly on
hierarchical leaders. Several studies do allude to leadership practices and preferences indicative
of a systemic approach with some studies identifying that SA professionals view leadership from
a systemic approach (Antonnen & Chaskes, 2002; Clayborne, 2006), while others contend that
SA professionals tend to rely more on hierarchical notions of leadership (Eddy & VanDerLinden,
2006; Katherine, 2011). However, these studies focus more on participants’ preferences in
leadership style and do not seek to understand leadership in the broad context of institutions with
multiple perspectives and behaviors in institutional networks. This study fills the gaps in the
research literature by examining leadership in the context of three comparable institutions and
measures leadership based on both individual self-reports and the nominations and responses of
others in the organization from across the formal hierarchy.
3.4 Methods
This study employs a mixed model approach to studying systemic leadership (Creswell,
2003; Mertens, 2005). Mixed model research is distinct from a mixed methods approach in that
mixed methods research utilizes both quantitative and qualitative research to answer the same
research questions, whereas mixed model research answers several research questions, each
using a different methodological approach (Mertens, 2005). In the case of this study, I utilize

85
quantitative methods through network analysis to identify collaborative network behaviors
associated with systemic leadership (research question 1) and qualitative coding and analysis of
open-ended survey responses to identify strategies utilized by systemic leaders (research
question 2).
Data Collection & Sample
The sample for this study is comprised of SA professionals from three mid-size (total
enrollments between 8,100 and 14,300), public Masters universities in the Midwestern region of
the United States. These higher education institutions were chosen from a sample of institutions
identified through a previous survey administered to a national sample of SA professionals
examining collaboration. Participants on the earlier survey were asked if they would be willing
to be contacted for follow-up research to examine the collaborative social networks of all the SA
professionals within their institutions. I chose these institutions due to their enrollment and
Carnegie similarities; I was seeking institutions similar in size and mission in order to be able to
compare and aggregate data in similar contexts. These three universities all ended up being
located in the Midwest, which allowed for even more similar comparisons to be made. I worked
with each campus’ chief student affairs office to identify SA administrators to include in the
study who met the criteria of being professional staff for the institution (often classified as
exempt staff). Rosters for each campus contained information about each individual including
name, department, and position.
Data for this study were collected through an online survey administered to the SA
professionals of these three institutions between October and December, 2014. The purpose of
this survey was to gather data pertaining to these professionals’ collaborative relationships and
strategies utilized by them when engaging in collaborative relationships. Each campus received a

86
survey specifically designed to capture information about collaboration within their institution.
Participants were first asked to identify up to seven individuals from among the SA staff with
whom they most frequently collaborate in their work. They were asked to identify these
individuals from a roster of potential SA professionals at the institution identified by the chief
student affairs office staff as described above. Participants then answered questions about each
collaborator relating to the strength of their relationship and professional and personal
characteristics of each collaborator.
In addition to the quantitative network questions, participants also responded to open-
ended questions about each collaborator they identified, asking them to describe a) why they
collaborate with each individual in their work, and b) in what ways each individual is a
successful collaborator. Participants were also asked questions regarding their own work to
describe a) reasons they collaborate in their work, b) what characterizes successful collaboration
to them, and c) strategies they utilize when they collaborate with others in their work. These
items were designed to elicit responses to uncover the processes and strategies involved in
successful collaboration. Additional questions on the survey pertained to personal demographics
of participants (e.g., racial/ethnic identity, gender identity); professional demographics (e.g.,
position rank, functional area, years of experience at the institutions).Participation was
incentivized with the option for participants to be entered into a drawing to win one of two gift
cards to an online retailer per campus. Table 3.2 lists the response rates for the samples in this
study. The individual campus response rates are within the range to make accurate network
measurements specifically pertaining to centrality and other network measures (Valente, 2010).
Descriptive statistics pertaining to individual demographics and professional characteristics can
be found in Appendix 3A.

87
Table 3.2: Response Rates for Study of Three Collaborative Social Networks in Student Affairs
Campus 1 Campus 2 Campus 3 TOTAL
Population of SA Administrators 29 41 59 129
Completed Responses 20 23 44 87
Completed Response Rate 69.0% 56.1% 74.6% 67.4%

Variables for Quantitative Analysis
Dependent variables. The three dependent variables for this study are three measures of
centrality (in-degree, betweenness, and integration), which serve as proxies of systemic
leadership in SA collaborative networks. In-degree centrality is the proportion of instances an
individual is selected as a collaborator by others in the network compared to the total number of
possible others who could select the individual as a collaborator and can range from 0 (no
nominations) to 1 (nominated by every other individual in the network). Betweenness centrality
is a measure of the extent to which individuals lie on the shortest paths connecting all other
members of the network and can range from 0 (individual lies on no paths connecting
individuals) to 100 (the individual lies on every shortest path between all network members).
Finally, integration is a reverse measure of the distance individuals are from other individuals in
the network and can range from 0 (not connected and, therefore, not close to any others in the
network) to 1 (connected to all members and, therefore, close to all members of the network).
Focal variables. The focal variables for this study are those pertaining to social network
measures of participants’ collaborative networks aside from centrality. These include ego
network density, constraint, average tie strength, homophily based on propinquity (i.e., working
in the same building), homophily based on department, and homophily based on gender identity.
Density is the proportion of actual ties in an individuals’ ego network compared to the potential
ties in the network (not including ties to the participant), ranging from 0 (least dense) to 1 (most
dense). Constraint is similar to density, but rather than just being a proportion of possible
connections in ego’s network, constraint uses more information from each alter’s ego network to

88
measure the extent to which each alter’s alters are connected to one another. Constraint,
therefore, measures the extent to which an individual is closed off to the rest of the network
based on the constraining relationships of the individual’s connections. Scores for constraint
range from 0 (not constrained) to 1 (most constrained).
I calculated tie strength using a composite scale developed by Gillis (2008). Participants
responded to three items for each collaborative relationship they identified assessing their
perceived closeness to that individual (on a five-point Likert-like scale, with 1 = “Very Distant”
and 5 = “Very Close) and the frequency with which they went to that individual for both work-
related and other types of advice (on a five-point Likert-like scale, with 1 = “Rarely” and 5 =
“Often”). I calculated scale scores by calculating the mean of the three items, which could range
from 1 to 5. The scale was internally reliable for this sample (α=0.81). I calculated average tie
strength for each individual by taking the mean of all of the tie strength measures in each
individual’s network.
I calculated homophily as the proportion of ties in participants’ ego networks who shared
a given trait with the participant compared to the total number of ties in the network for
propinquity, department, and gender. When participants nominated someone who also completed
the survey, I utilized the information provided by the other individuals to determine if two
participants shared the same trait. When participants nominated someone who did not complete
the survey, I relied on their own perceptions about whether or not they worked in the same
building, worked in the same department, and shared the same gender identity as the other
person. Scores for homophily range from 0 (none of the participants’ ego network alters share
the trait with the participant) to 1 (all of the participants’ ego network alters share the trait with

89

the participant). Descriptive statistics for the focal and dependent variables are listed in Table
3.3.
Control variables. In addition to the focal variables listed above, other independent
variables were included in random effects models as controls. These include variables pertaining
to professional characteristics and gender. Individuals’ network behavior varies by personal and
professional characteristics (Borgatti & Foster, 2003; Bowler & Brass, 2006; Brass, 1985; Brass
et al., 2004; Ibarra, 1992; Kadushin, 2012; McPherson et al., 2001; Mehra et al., 2001;Mehra,
Kilduff, & Brass, 2006; Perry-Smith & Shalley, 2003). Therefore, I included several control
variables to account for these characteristics. Professional control variables include participants’
Table 3.3: Descriptive Statistics for Student Affairs Professionals’ Collaborative Social Networks
from Three Student Affairs Divisions
M (SD) Description
Centrality: In-Degree 0.17 (0.12)
The proportion of the number of links in which ego is nominated
as a collaborator compared to the total possible nominations
ego could receive (i.e., the entire network); ranges from 0 (no
nominations) to 1 (nominated by every member of the network)
Centrality: Betweenness 5.65 (5.43)
The frequency that ego lies on the shortest paths connecting all
other actors in the network; ranges from 0 (lies on no paths) to
100 (lies on every path)
Centrality: Integration 0.85 (0.16)
The reverse distance of ego to all other nodes in the network
measuring the extent to which ego is close to all others in the
network (therefore having an easier ability to exert influence);
ranges from 0 (not close to others in the network) to 1 (close to
all others in the network)
Ego Network Density 0.28 (0.14)
Proportion of number of connections among ego’s alters to the
total possible number of connections; ranges from 0 (least
dense) to 1 (most dense)
Ego Network Constraint 0.44 (0.16)
The extent to which ego’s alters constrain their access to the
network by being already linked to one another; ranges from 0
(least constrained) to 1 (most constrained)
Average Tie Strength 2.75 (1.00)
The average strength of relationships ego has to alters in
network; 5-point scale, with 1 = weak tie and 5 = strong tie
Ego Network Homophily:
Department
0.54 (0.33)
The proportion of ego’s alters from ego’s department; ranges
from 0 (no alters from ego’s department) to 1 (all alters from
ego’s department)
Ego Network Homophily:
Propinquity
0.48 (0.36)
The proportion of ego’s alters from ego’s building; ranges from 0
(no alters from ego’s building) to 1 (all alters from ego’s
building)
Ego Network Homophily:
Gender
0.62 (0.36)
The proportion of ego’s alters who share ego’s gender identity;
ranges from 0 (no alters share gender identity with ego) to 1 (all
alters share gender identity with ego)

90
rank in their institution (e.g., entry-level, mid-level, senior-level), functional area, and years of
experience at their current institution. Participants were asked to identify the functional area in
which they worked from a list of 22 potential functional areas. For the sake of parsimony in the
analyses, these functional areas were collapsed into seven umbrella functional areas which
perform similar functions on these campuses: activities/student union/recreational activities;
advising/career services/academic support; administrative leadership/generalist;
multicultural/diversity programs; wellness/health programs; residence life/student conduct; and
support/auxiliary student affairs areas. The personal characteristic included as control variables
in the analyses below is gender identity.
9
Descriptive statistics for these variables are listed in
Appendix 3A.
Analyses
Quantitative analyses. Prior to running the models for this paper, I explored the data to
identify any patterns in missing data. Fewer than 10% of participants in the study exhibited
missing data (eight out of 87 participants), which is well within the suggested range to be
included in the analyses (Newton & Rudestam, 1999). In all eight instances, participants did not
identify their gender. Given the categorical nature of this variable, these were accounted for by
listwise deletion in the random effects models below rather than utilizing imputations to replace
the missing data.
In order to answer the first research question pertaining to variables associated with
leadership in SA collaborative social networks, I began by calculating individual network
variables based on the network information for each of the three campuses utilizing UCINET, a
social network analysis software package (Borgatti, Everett, & Freeman, 2002). I then combined

9
I had planned to also include racial/ethnic identity given the body of research that points to its role in influencing
network behavior, but the sample exhibited very little racial/ethnic diversity (only 10% of the sample identified as
non-White) to allow for meaningful interpretations comparing multiple racial/ethnic groups.

91
the three separate institutional datasets into one dataset containing information for all the
participants in the study across the three campuses. The sample exhibits clustering, as
participants in the dataset work across three different institutions. Network behavior is
influenced by this kind of clustering, violating the assumption of independence (Valente, 2010),
which can result in biased standard errors if calculated using ordinary least-square regression
(Rabe-Hesketh & Skrondal, 2012). Therefore, network analysts recommend using multilevel
random effects models to account for this nonindependence. I utilized three multilevel models to
examine the associations of collaborative network behaviors and personal and professional
characteristics with three dependent variables: in-degree centrality, betweenness centrality, and
integration centrality. I entered the variables into the models in two blocks in order to examine
the variance accounted for by each block. The first block pertained to personal and professional
characteristics, followed by the block pertaining to collaborative network behaviors. I calculated
the change in variance (i.e., R
2
) from the control block of individual characteristics to the
collaborative network block. This allowed me to identify the amount of unique variance
accounted for by these variables after controlling for the individual characteristics. I also
standardized all continuous variables prior to including them in the models so that their
coefficients could be interpreted as effects sizes.
Qualitative analyses. In order to answer the second research question pertaining to traits
and strategies important for effective collaboration, I began first by identifying four to five
participants from each campus who exhibited the highest centrality scores across the three
centrality measures indicating they were systemic leaders in their collaborative networks (n=15),
as well as identifying approximately five participants from each campus network who scored the
lowest on these indicators (n=14). I refer to these two groups as systemic leaders and non-central

92
individuals. For each participant, I cumulated their open-ended responses pertaining to what
characterizes successful collaboration and strategies to ensure successful collaboration. In
addition, I also cumulated the responses made by each person who nominated these individuals
about why they collaborate with them, and what made them successful collaborators. My coding
strategy involved identifying both commonalities and differences between the two groups in
order to primarily identify strategies that were unique to systemic leaders, which could provide
clues as to strategies they employed that were not utilized by non-central individuals; I also
identified common strategies for collaboration in which the groups did not differ. I coded
individual strategies as they came up for each participant, utilizing Boyatzis’ (1998) inductive
coding techniques in HyperRESEARCH, a qualitative data analysis software package
(ResearchWare, Inc., 2013). I then examined patterns in these strategies across participants in
order to identify trends. Themes identified below as pertaining to systemic leaders are often
those where the corresponding codes for those themes were significantly less prevalent (i.e.,
mentioned only once) or entirely missing from responses for non-central individuals. I present
trends in the data that resulted in different themes with corresponding quotes that represent each
theme in the findings section.
Limitations
As with all research studies, this study exhibits several limitations. The first limitation
pertains to the way in which I operationalize systemic leadership. Centrality measures from
network theory certainly apply to many of the ways in which systemic leaders operate in their
organizations, but these encapsulate a specific strategy for operationalizing these behaviors.
While this approach does take into account and situate these practices in broader structures, by
identifying systemic leaders utilizing these measures, I may be limited in identifying others who

93
practice systemic leadership through other strategies not identified through this strategy. Other
methodological and theoretical frameworks could contribute to different or more nuanced ways
of operationalizing systemic leadership behaviors. The second limitation pertains to the cross-
sectional nature of the data, which precludes me from making causal inferences about network
behavior. Utilizing longitudinal and/or experimental study designs could have allowed me to
make these claims. However, given the focus of studying collaboration among SA administrators
in their natural institutional settings, it was not possible to design an experimental condition to
control for all of the various individual and institutional factors that could influence this
behavior. Third, while the response rates were within acceptable ranges to measure network
behavior, between 25% and 44% of individuals who work in these institutions are not part of the
samples. Without their information and 100% participation, the network indicators calculated
based on these samples may be biased. Finally, the qualitative data for this study were collected
through open-ended response items on a survey rather than through follow-up interviews, which
would have allowed me to dig deeper in participants’ responses and tease out more nuance in
these responses. I highlight future research utilizing additional qualitative data collection
techniques below.
3.5 Findings
Figure 3.1 exhibits the three collaborative networks of SA professionals utilized to
perform the data analyses described above. For each network, the systemic leaders are
highlighted.

94
Figure 3.1: Collaborative Social Networks among Student Affairs Professionals at Three
Mid-Size Masters Universities

NOTE: Red nodes indicate systemic leaders

95
Predictors of Systemic Leadership in SA Collaborative Networks
In-degree centrality. Results of the three random effects models identify several
collaborative network behaviors quantitatively associated with systemic leadership in SA
collaborative networks, as well as other associations with control variables (see Table 3.4). I
begin first with the model for in-degree centrality, which is an indicator of the extent to which an
individual is nominated by others in the network as a collaborator. The block containing
collaborative network behaviors accounts for 27% of variance in this dependent variable. Two of
the variables in this block are significantly associated with in-degree centrality. Specifically,
constraint in one’s ego network is negatively associated with in-degree centrality, suggesting that
the more individuals are constrained by their network connections, the less likely they will be
connected to others in the organization as collaborators. Additionally, density in one’s ego
network is positively associated with in-degree centrality after controlling for the other variables
in the model. The block for the control variables in the model account for 17% of the variance in
in-degree centrality, with several significant associations as well. Specifically, mid-level
professionals are expected to be less central than senior-level professionals on in-degree
centrality after controlling for the other variables in the model. Additionally, functional unit is
significantly associated with in-degree centrality after controlling for the other variables.
Specifically, participants who work in activities/union/recreation programs and auxiliary services
are expected to be less central on in-degree compared to participants in advising/career services.
This model accounts for a total of 44% of variance in in-degree centrality.
Betweenness centrality. The next model is for betweenness centrality, which measures
the extent to which an individual lies on paths connecting others in the network and serves as a
connector or bridge to others in the organization. The block containing collaborative network

96

behavior accounts for 34% of variance in the dependent variable. Similar to the model above,
constraint in one’s ego network is negatively associated with betweenness centrality after
controlling for other variables. Additionally, propinquity homophily is also negatively associated
with betweenness centrality. Turning to the control variables, which account for 10% of the
variance in this model, functional unit is significantly associated with betweenness centrality.
Specifically, participants who work in activities/union/recreation programs; residence life and
student conduct; and auxiliary services are expected to score lower on betweenness centrality
compared to participants in advising/career services. Years of experience in the institution is also
negatively associated with betweenness centrality. This model accounts for a total of 44% of the
variance in betweenness centrality.
Table 3.4: Random Effects Models Predicting Leadership Measures in Three Collaborative Student
Affairs Networks
Centrality: In-Degree
Centrality:
Betweenness
Centrality: Integration
R
2
β SE R
2
β SE R
2
β SE
Individual Characteristics .17 .10 .16
Female -.29 .33 -.13 .32 -.82
***
.29
Entry-Level Professional
a
-.15 .19 -.04 .19 .11 .17
Mid-Level Professional
a
-.58
**
.24 -.31 .24 .00 .21
Functional Area: Activities/Union/Rec
b
-.56
*
.31 -.93
***
.30 .15 .27
Functional Area: Wellness/Health
b
-.58 .40 -.15 .38 .49 .35
Functional Area: Administrative
Leadership/Generalist
b

-.69 .56

-.62 .55

.55 .49
Functional Area: Residence Life/Conduct
b
-.36 .37 -.92
**
.36 .12 .32
Functional Area: Multicultural/ Diversity
Programs
b

.19 .95

-.58 .93

-.34 .83
Functional Area: Auxiliary/Support
b
-1.17
**
.48 -1.21
**
.47 .50 .42
Years at Institution -.06 .12 -.30
**
.12 -.19
*
.11
Collaborative Network Variables .27 .34 .40
Ego Network Density .41
**
.19 .04 .18 -.10 .16
Ego Network Constraint -.78
***
.20 -.57
***
.20 -.39
**
.18
Ego Network Mean Tie Strength .07 .01 .13 .12 .42
***
.10
Ego Network Homophily: Department -.18 .13 -.08 .13 -.18 .12
Ego Network Homophily: Propinquity .03 .14 -.23
*
.14 .08 .12
Ego Network Homophily: Gender .07 .16 .02 .16 .24
*
.14
R
2
.44 .44 .56
NOTES:
*
p < .10;
**
p < .05;
***
p < .01; Reference Categories for Categorical Variables:
a
Senior-level;
b
Advising/Career
Services

97
Integration centrality. The final model is for integration centrality, which indicates the
extent to which an individual is close to all others in the network and can therefore exert
influence throughout the network. The block for collaborative network behaviors accounts for
40% of variance in the dependent variable for this model. As with the previous two models,
constraint is significantly and negatively associated with integration centrality after controlling
for other variables in the model. Two network behaviors are significantly and positively
associated with integration – mean tie strength in one’s ego network and gender homophily in
one’s ego network; individuals who have stronger relationships with their collaborators on
average and who collaborate more with others who share their gender identity are more likely to
be integrated in the larger network. Turning to the control variables, which account for 16% of
the variance in the dependent variable, participants who identify as female are expected to be less
integrated in their collaborative networks compared to those who identify as male, after
controlling for other variables. Additionally, years of experience in the institution is negatively
associated with integration centrality. This model accounts for a total of 56% of variance in the
dependent variable.
Strategies of Systemic Leaders
I now turn to the open-ended responses from the survey to identify patterns that emerge
for systemic leaders compared to individuals who do not hold central roles in the organization
(and are therefore not systemic leaders – described in methods section above). I refer to these
two types of individuals below as systemic leaders and non-central individuals. I begin first by
presenting themes from participants’ own perceptions of their behavior, followed by the
perceptions of their collaborators related to what makes them successful collaborators.

98
Participants’ own perceptions. To begin, systemic leaders (those who scored highest
across three centrality measures) and non-central individuals (those who scored lowest across
three centrality measures and essentially do not practice systemic leadership) generally exhibit
similar characterizations of successful collaboration, but some key differences emerge. The most
frequently cited characteristics of successful collaboration among systemic leaders included
having a common goal, open communication and listening, openness to new ideas, shared
responsibility, and being mutually beneficial to all parties. As for non-central individuals, the
most commonly cited characteristics of successful collaboration included openness to new ideas,
focus on benefits to students, cooperation, mutual benefit, follow-through, having a common
goal, and open communication/listening. These characterizations of successful collaboration by-
and-large are very similar, and these two groups of participants also exhibit similar descriptions
of these characteristics. For example, one systemic leader described the important of being open
to new ideas as a “willingness to discuss new ideas, willing to make change,” and a non-central
participant described it in a similar way as “willing to be open-minded to how others view your
department as well as being open to suggestions.” In general, both systemic leaders and non-
central participants identify successful collaboration as involving open and strong
communication, being open to ideas, and mutual or common goals.
Big picture focus of collaboration. However, these groups also differed in several ways.
First, non-central individuals focused more on specifics to the process of collaboration when
characterizing successful collaboration, as evidenced by their mentioning dividing up tasks and
follow-through, such as this participant: “It helps if you can divide up the work-load and can
depend on everyone to do their part.” Systemic leaders, on the other hand, tend to focus less on
process details and are more big-picture focused, such as this participant’s description of

99
successful collaboration: “Where both parties share a common vision of success and want to
have it succeed.” They also differ specifically in their student focus, as non-central participants
cited a focus on how students can benefit from collaborations several times, whereas only one
systemic participant mentioned the benefits to students. However, the extent to which these
groups actually focus on students in their collaborative work (as perceived by their fellow
collaborators) reveal departures from their self-perceived focus (described below).
Multiple strategies for collaboration. Turning to reported strategies for ensuring
successful collaborations, more difference emerge between the systemic leaders and non-central
participants. Open communication and listening once again emerged as strategies for both central
and non-central participants. However, differences emerge in the sheer number and diversity of
strategies described by these participants. In addition to communication and listening, systemic
leaders highlighted a variety of strategies to ensure successful collaboration, including
distributing tasks/action items, focusing on students, detailed notes for meetings, relationship
building, willingness to cooperate, and follow-through as strategies. All-in-all, systemic leaders
listed a total of 13 different strategies in their responses. Non-central individuals listed only a
total of five potential strategies; in addition to listening and communication, the only strategies
mentioned more than once include follow-through and holding regular meetings. While both
groups perceive communication to be the most important aspect of ensuring successful
collaboration, it appears that systemic leaders acknowledge more possible strategies to ensure
successful collaboration, which could contribute to their being central in their collaborative
networks.
Perceptions of collaborators. In addition to perceptions of systemic leaders and non-
central individuals relating to collaboration, I also analyzed the perceptions of these individuals’

100
collaborators (referred to here as alters, in-line with network theory) in order to understand both
why they choose to collaborate with these individuals and what makes them successful as
collaborators.
Function, knowledge, and communication common across participants. Three common
themes emerged for both systemic leaders and non-central individuals. These are functional
collaboration, collaboration due to knowledge/expertise, and effective communication/listening
skills, which I describe below. First, the most common reasons given for collaborating with both
systemic leaders and non-central participants related to their function (i.e., two departments share
a program so collaboration occurs), supervisory relationships (i.e., individuals collaborate with
their direct supervisors or supervisees), or committee work (i.e., individuals collaborate when
placed on the same committee). This suggests that structural aspects of higher education
institutions are large drivers of collaboration, regardless of one exhibiting systemic leadership as
both systemic leaders and non-central participants were identified by these functions.
The next most common theme across these individuals was knowledge/expertise, as alters
identified these individuals’ specific expertise as reasons for their effective collaboration.
Examples of knowledge and expertise include these examples from participants for both central
and non-central participants: “He also has vast experience and expertise in student development
and psychology;” “[Participant] is extremely knowledgeable and highly dependable. She is a
campus expert in assessment and serves as a great resource;”
10
and “She has expertise in an area
I do not have much experience in.” Finally, strong communication/listening skills also emerged
as common across these two groups as important for collaboration. Some examples of its
importance include: “She is easy to talk to and with;” “[Participant] listens to my concerns and
relays advice on how to address those concerns;” and “She takes my opinion into consideration.

10
When necessary, I replace a participant’s name with the term [Participant] to ensure anonymity.

101
She actively listens and works to understand my perspective in any situation or concern that is
brought up.”
These three themes reveal that, regardless of one’s place in the larger collaborative
network, SA professionals share common reasons and strategies for collaborating; they are
brought together to collaborate based on functions in their institutions and their collaborations
are successful due to knowledge/expertise to contribute to the collaboration and effective
listening and communication skills. However, differences also emerged that set the systemic
leaders apart from the non-central individuals.
Collaborative skills/relationship building. Four themes emerged for systemic leaders that
were not evident for non-central participants, which suggest several strategies important for this
type of leadership in SA. These themes were general collaborative skills; having a student focus;
openness to new ideas; and being well-connected and politically savvy. First, systemic leaders
were identified as exhibiting general collaborative skills that contributed to their successful
collaborations. These skills include a desire to collaborate – “[Participant] is always willing to
collaborate;” actively seeking others on campus to collaborate with – “He does seek
collaboration in many areas of his work. I think he looks for people to help promote and plan
events on campus;” and general relationship building within an institution – “He takes time to
meet with every single full time employee within the [building] once they begin working.”
Additionally, systemic leaders were observed as being effective collaborators, which drew
people to want to collaborate with them. Comments such as “[Participant] collaborates with
student affairs, faculty and other university departments,” and “He works across campus very
well,” suggest that the visibility of their collaborative relationships on campus contributed to
systemic leaders’ effectiveness.

102
Student focus. Second, despite the fact that systemic leaders did not mention a student
focus for their collaborations as much as non-central individuals, their alters identified their
emphasis on students as one of the most important reasons for their success as collaborators;
alters for non-central individuals did not address their student-focus. This example from a
participant’s alter concisely sums up the sentiments across these responses: “He has student
success at the heart of his mission.” So despite systemic leaders not directly acknowledging the
importance of students in their collaborative work, their emphasis on students stands out to their
collaborators as setting them apart from their peers and contributes to their effectiveness. Also,
while non-central individuals identified a student focus as being important for their collaborative
efforts (see above), others who collaborated with them did not observe this in their work, as
evidenced by their alters not reporting a student-focus when describing their work.
Openness to new ideas. A third characteristic that set systemic leaders apart from their
non-central peers is their openness to new ideas. Alters for these leaders indicated that these
leaders were successful as collaborators because they were willing to listen and were open to
others’ ideas in order to ensure a successful endeavor. This is evidenced by statements such as
“[Participant] is open to ideas in order to benefit both of our programs/departments;” “She is
supportive of new ideas and is open to feedback;” “He is open to any idea and any possibility;”
and “Always willing to see different points of view and change mind depending on information
presented.” Again, this openness was unique to the systemic leaders in this study and did not
emerge for the non-central participants.
Political savviness and connections to others. Finally, systemic leaders were seen as
politically savvy individuals whose connections with others made them desirable as collaborative
partners. Alters were appreciative of these leaders’ abilities to navigate campus politics, as

103
evidenced by comments such as these: “[Participant] has wisdom, navigates campus politics;”
and “She is good at working in a political framework.” Related to being able to navigate these
political environments, alters indicated that systemic leaders’ abilities to connect to multiple
constituencies across campus also contributed to their effectiveness. This was due at times to the
knowledge that comes from these connections, as with this participant: “[Participant] works with
many people on campus and is able to see a big picture and who would be a good fit for a project
or committee.” However, beyond being knowledgeable about others on campus due to their
connections, simply being seen as being able to connect with multiple constituencies on campus
also contributed to systemic leaders perceived effectiveness: “She’s very successful within both
academic and student affairs departments;” “Her ability to connect with others is something that
is both personally and professionally beneficial for her;” and “She engages in collaborative
relationships that allow for her to connect with many other people which makes others want to
work with her.” These four themes suggest certain skills and behaviors for success in systemic
leadership.
A Profile of a Systemic Leader
Several participants in this study exhibit systemic leadership based on their centrality
scores and the qualitative responses about them. I highlight one such individual here to illustrate
an example of a systemic leader in student affairs. Andrea (a pseudonym) works as a mid-level
professional in housing and residence life at a mid-size Masters university in the Midwest. Her
centrality scores reveal that she is central in her network; she has the highest or second highest
scores in all three centrality measures at her institution. Nearly one-third of the 44 SA
professionals at her institution indicate she is someone with whom they collaborate (the highest
in-degree of all at her institution), which reveals that she is someone working across the campus

104
to form collaborative connections. Additionally, she serves as a bridge or connector between
professionals at her institution more than all but one of her peers. She also exhibits the lowest
constraint score in her network, suggesting that she is not being blocked from the rest of the
network by her fellow collaborators relative to others at her institution; 92% of her collaborative
relationships are being formed with others outside of her physical location on campus (the
highest at her institutions); and her ties are, on average, stronger than those of her peers. In other
words, Andrea is a quintessential systemic leader based on her network behavior.
Beyond her collaborative network behaviors, Andrea’s collaborators also identify that she
engages in many of the systemic leadership strategies described earlier in the findings with her
collaborators. One collaborator sums up her general relationship-building skills and openness to
new ideas and feedback as follows: “She is very engaging and personable; people feel
comfortable opening up to her and sharing ideas and feedback. She is open to feedback but will
also challenge us to look at issues from multiple perspectives.” This openness and ability to
engage is important for collaborative systemic leaders to be able to form relationships with many
others across campus, and Andrea’s use of these strategies allows her be successful.
Andrea’s use of multiple strategies is another characteristic of her systemic leadership, as
she utilizes many behaviors to ensure successful collaboration. One of her collaborators
describes how: “Andrea gives clear direction and guidance in projects. She relies on feedback,
research and expertise of others. She shares responsibilities while carrying a great responsibility
herself within projects. Andrea is extremely knowledgeable and highly dependable.” This quote
highlights the ways in which Andrea ensures successful collaboration through four concrete
strategies – offering guidance, giving and receiving feedback, dependability, and taking

105
responsibility. Having the ability to utilize multiple strategies with others was another key
strategy employed by the central leaders in this study, and Andrea is no exception.
Across the data of the systemic leaders in this study, exuding a student-focus in
collaborative work was a cornerstone of effective collaboration. Andrea’s student-focus is
evident in comments by her collaborators and is another characteristic that sets her apart from
her peers as she places students’ concerns at the forefront, as in this this case for students who
are also parents: “[She] shares concerns when student parents show signs of stress, works to
connect families to resources and to stay in school.” This collaborator suggests this student-focus
was integral to Andrea’s ability to successfully collaborate with these specific programs.
Finally, several individuals highlight how Andrea’s knowledge of campus and its politics
make her a successful collaborator, which is summed up most succinctly by this individual:
“Andrea provides structure, has wisdom, [and] navigates campus politics.” Having the ability to
understand and work with campus politics is an important aspect of systemic leadership, and
Andrea ensures that collaborative efforts are effective and meet the demands of a complex,
political system like her university with her ability to navigate those politics.
In sum, Andrea’s centrality scores, as evidenced through network analysis, indicate that
she is a central systemic leader on her campus, who exhibits the behaviors (e.g., less constrained,
less propinquity homophily, and stronger ties) that are associated with this style of leadership.
Additionally, her collaborators provide evidence of the ways in which she is able to be a
systemic leader on her campus, through relationship building, openness, multiple collaborative
strategies, political savvy, and student focus, which encapsulate most of the central themes
describe above pertaining to successful collaborative strategies of systemic leadership.

106
3.6 Discussion
Leading voices in the student affairs profession continue to call for systemic leadership
that cuts across institutional boundaries (ACPA, 1994; Amey, 2006; Caple & Newton, 1991;
Joint Task Force, 1998; Keeling, 2004; Kuk et al., 2010; Manning, 1996; Manning et al., 2006;
Roberts, 2001; Rogers, 2003; Roper, 2002; Task Force, 2010; Winston et al., 2001),
necessitating research to identify behaviors and strategies to inform practice. In this study, I
identify several collaborative behaviors associated with higher levels of systemic leadership,
represented by network centrality measures, as well as strategies and characteristics of effective
systemic leaders. The findings from this study can inform practice for SA professionals and
leaders who seek to practice and foster systemic leadership in their institutions, as well as
advance the scholarship on systemic leadership in SA and higher education.
Behaviors and Strategies for Systemic Leadership
One of the key contributions of this research is identifying network behaviors and
strategies that emerged from both quantitative and qualitative analyses that are important for
systemic leadership. The nature of SA professionals’ collaborative relationships matters more in
predicting systemic leadership than personal and professional characteristics. The variables
measuring aspects of SA professionals’ collaborative networks accounted for the majority of
variance in the three dependent variables pertaining to systemic leadership, accounting for
between 27 and 40% of variance, compared to the 10-17% of variance accounted for by personal
and professional characteristics. This suggests that systemic leadership is more behavioral than
trait-based or functional, further highlighting the limitations of the extant literature on SA
leadership in accounting for leadership that cuts across organizational boundaries.

107
Based on the findings from the random effects models and open-ended analyses, systemic
leaders collaborate in unconstrained networks, take a big picture view of collaboration, utilize
multiple strategies to ensure effective collaboration, focus on students in their collaboration, are
open to new ideas, and are politically savvy and well-connected to others in their organizations.
These behaviors and strategies (summarized in Figure 3.1), when examined together, provide
important insights into engaging in systemic leadership.
Collaborative strategies in unconstrained networks. Systemic leaders are effective at
reaching beyond social and physical boundaries to form connections. The random effects models
indicate that systemic leadership is negatively associated with constrained collaborative networks
(for all three measures) and homophily based on propinquity (for betweenness), suggesting that
SA professionals who practice systemic leadership are successful at forming connections with
others across their institutions who are not all collaborating with one another to limit leaders’
exposure to the rest of the institution, as well as reaching beyond physical barriers to form
relationships with others located across campus. These are two key behaviors that relate to the
central tenets of systemic leadership to form relationships across boundaries dictated by formal
hierarchy and functional unit, and these analyses reveal their importance for SA professionals
who seek to practice systemic leadership in their work.

108
The most common reason for SA professionals to collaborate, whether or not they were
central leaders, was for functional reasons relating to hierarchy, functional unit, and formal
committee work. One’s functional unit was also associated with systemic leadership in this
study. This reinforces and complements other research which shows that one’s ability to
collaborate across organizational boundaries (Gehrke, 2015), as well as one’s physical location
on a campus (Borgatti & Cross, 2003; Brass et al., 2004; Festinger, Schachter, & Back, 1950;
Kadushin, 2012; Monge & Contractor, 2003), are inherently influenced by formal organizational
structures. Formal structures can inhibit SA professionals’ abilities to reach beyond them to
collaborate across their institutions. Despite the inhibitive aspects of these structures, systemic
leaders still find ways to reach beyond these boundaries to engage in effective collaboration; they
Network Behaviors for Systemic Leadership
Unconstrained - Less Homophily by Propinquity - Stronger Ties
Collaborative Strategies
for Systemic Leadership

Big Picture Orientation

Multiple Collaboration
Strategies

Collaborative &
Relationship Building Skills

Openness to New Ideas

Student-Focus

Political Savviness &
Connection to Others
NOTES: Thick lines denote stronger ties; house structures represent building individuals work in on campus.
Figure 3.2: Model of Network Behaviors and Strategies for Systemic Leadership in
Student Affairs Collaborative Networks

109
are connected to more individuals in their institutions and serve as bridges with access to others
across organizational boundaries. Their open-ended responses and those of their collaborators
suggest reasons why they may be able to do so.
First, systemic leaders’ collaborators indicated that these individuals are good at forming
relationships and seeking out collaborative partnerships on their campuses. This is further
supported by the finding of strong ties being positively associated with integration, a centrality
measure indicating one’s closeness to others in the organization. These findings point to the
importance of relationship building and seeking out partnerships across campus as a way to reach
beyond the confines of one’s local networks and physical location on campus. Forming strong
relationships is seen as an effective strategy for these leaders to successfully collaborate with
others in their work across their campuses, as well as actively seeking out others to partner with.
As the example with Andrea above indicated, she was someone who many people turned to and
who formed relationships with many others. Additionally, the fact that rank in the hierarchy was
not generally associated with systemic leadership in the random effects models suggests that
people from all ranks can do this in their work.
Related to relationship-building is being well-connected and the ability to navigate
politics on campuses, another set of skills identified as being possessed by systemic leaders. The
dual focus on building relationships and navigating campus politics is reminiscent of Bolman
and Deal’s (2008) organizational frames, which is a popular framework utilized by SA
professionals for understanding their work in organizational contexts (Kezar, 2010). Bolman and
Deal present a framework for understanding organizations from four different perspectives or
frames: structural, human resources, political, and symbolic. These frames represent different
mindsets and approaches to understanding and working within organizations. Systemic leaders’

110
abilities to form relationships and the importance of strong ties, along with their political
savviness, suggests that SA professionals should hone their skills related to both the human
resources and political frames in order to focus on relationships but also form coalitions with
others in order to effectively collaborate across organizational boundaries.
Finally, systemic leaders in this study were identified as utilizing multiple strategies to
ensure successful collaboration, were focused on the well-being of students, were open to new
ideas, and took a big-picture approach in their thinking about collaboration. These behaviors
seem central to a systemic approach to leadership, one in which professionals are driven by the
larger goals of their work, not bogged down with minutia of what it means to successfully
collaborate, and utilize a wide range of strategies and ideas with their collaborators to ensure that
work gets done and benefits one of the main stakeholders of SA work, the students. Such big-
picture and goal-oriented mindsets likely suggest an ability to think about the larger system of an
institution, which connections need to be made, and what kind of work needs to be done in order
to accomplish one’s goals. It is not difficult to understand why these behaviors might help
systemic leaders reach beyond social and physical barriers to ensure effective collaboration. One
must understand the larger systems of relationships and connections in their institutions in order
to know how and with whom to collaborate, and acting on such knowledge is central to systemic
leadership (Allen & Cherrey, 2000).
Implications
In identifying behaviors and strategies employed be systemic leaders, this study provides
a roadmap of sorts for SA professionals who desire to engage in this type of leadership in their
practice. Systemic leadership is much more accounted for by behaviors than innate
characteristics or structural factors of one’s work, which means that systemic leadership can be

111
learned and practiced by all members of an organization. SA professionals who are seeking to
develop these skills should consider ways to practice them. One way is to begin by observing or
forming relationships with those who already practice systemic leadership in their institutions.
The open-ended responses in this study suggest that SA professionals are able to perceive the
types of behaviors that set systemic leaders apart from their other colleagues. SA professionals,
especially entry-level professionals who identify leadership as an area for further development
(Cilente et al., 2007; Renn &Hodges, 2007), should seek out members of their institutions who
are actively collaborating across their institutions, who appear to be well-connected, and who
exhibit big-picture thinking in their work as mentors of sorts for engaging in systemic leadership.
Finding opportunities to serve on committees with these individuals or creating partnership
opportunities with them can provide SA professionals with exposure to these types of systemic
leadership behaviors while contributing to their development. Another strategy for SA
professionals looking to hone their systemic leadership abilities is to seek out professional
development opportunities that not only put them in contact with professionals from other
institutions who may be effective systemic leaders but to also find institutes and workshops that
address skill development in areas such as collaboration, systems thinking, and strategies for
effectively partnering and coordinating programs across institutions.
This research also provides a framework for upper-level campus administrators to
identify and foster systemic leadership among their staff. As suggested above, SA professionals
in this study identified several behavioral characteristics that set systemic leaders apart from
others on their campuses. Chief student affairs officers (CSAO) can observe the work of their
staff members in order to identify those individuals who collaborate frequently across
boundaries, show an ability to navigate political structures, are well-connected and adept at

112
forming relationships across campus, and utilize multiple strategies in their collaborative work.
Based on this research, these individuals will hold central positions in their collaborative
networks, indicating they are influential to many members of the organization (Valente, 2010).
CSAOs can capitalize on these leaders and their influence to ensure buy-in of strategic initiatives
that can help break down structural barriers to foster more cross-campus collaboration and
systems thinking among the staff. These individuals can also be identified to serve as mentors to
others in the organization, especially entry-level professionals, through formal mentoring
programs; this strategy could remove some of the pressure on new professionals to seek out this
mentoring by creating formal structures to link them with these influential leaders on their
campus.
This study also suggests future directions for research. First, as mentioned in the
limitations section, this study is cross-sectional in nature. One of the dynamics of social networks
in organizations is that they shift over time as members join and leave organizations and
interventions are utilized to alter network structures (Monge & Contractor, 2003). Future
research could utilize an intervention of sorts, based on the type of data gathered in this study, to
identify systemic leaders in student affairs divisions and utilize them in the ways described above
to encourage buy-in of strategic initiatives to encourage collaboration and mentor others in the
organization to improve systemic thinking and approaches to leadership. These divisions could
then be surveyed at different points in time to see if these methods do alter structures of
relationships and foster more systemic leadership among the members of the divisions. Second,
this study only examined the collaborative relationships among SA professionals, but many
scholars point to the necessity of collaboration between and among SA administrators and
academic administrators to contribute to institutional effectiveness (ACPA, 1994; Joint Task

113
Force, 1998; Keeling, 2004; Task Force, 2010). Future research should examine larger
collaborative networks in order to identify systemic leaders who not only cross departmental
boundaries but also broader structural boundaries in the institution and seek to understand the
behaviors associated with this broader boundary-spanning leadership. Such inquiry could also
identify if there are certain behaviors that are applicable across higher education institutions or if
different stakeholders employ different strategies when collaborating in their work and engaging
in this type of leadership. The patterns of systemic leadership in this study were also identified
on three comparable campuses, but more research should seek to replicate this study in different
types and sizes of institutions in order to identify dynamics of systemic leadership for these
different institutional contexts. Finally, as mentioned in the limitations section of this paper,
additional research should utilize additional qualitative data collection techniques, such as
interviews or observations, to identify more details and nuances in strategies employed by
systemic leaders.
3.7 Conclusion
Higher education is facing turbulent times, and SA professionals continue to be called to
engage in systemic leadership that cuts across organizational boundaries in order to help guide
their institutions through the permanent white water of challenges facing them today (Manning et
al., 2006). Research on these systemic leadership practices can provide professionals with
concrete behaviors and strategies to assist them in their navigation. This study utilizes social
network theory and analysis to identify individuals who are practicing this form of leadership
and uncover behaviors and strategies they employ in order to effectively collaborate across their
campuses and play an influential role in their collaborative networks. Increasing the number of
SA professionals who are engaging in effective systemic leadership can foster SA divisions and

114
institutions that will be able to successfully respond to the challenges facing U.S. higher
education, and this study identifies the kinds of skills that SA professionals can develop in order
to successfully guide their institutions through turbulent whitewater that awaits in the years to
come.

115
3.8 Appendix 3A: Descriptive Statistics for Other Variables in this Study
Mean S.D. Min Max Description
Individual Characteristics
Female 0.74 0.00 1.00 1 = Female; 0 = All others
Male 0.26 0.00 1.00 1 = Male; 0 = All others
Entry-Level Professional
a
0.18 0.00 1.00 1 = Entry-Level; 0 = All others
Mid-Level Professional
a
0.46 0.00 1.00 1 = Mid-Level; 0 = All others
Senior-Level Professional 0.36
Functional Area:
Activities/Union/Rec
b
0.27 0.00 1.00 1 = Senior-Level; 0 = All others
Functional Area: Advising/ Career
Services
0.28 1 = Advising/Career Services; 0 = All
others
Functional Area: Wellness/Health
b
0.08 0.00 1.00 1 = Wellness/Health; 0 = All others
Functional Area: Administrative
Leadership/Generalist
b

0.04 0.00 1.00 1 = Administrative
Leadership/Generalist; 0 = All others
Functional Area: Residence
Life/Conduct
b

0.24 0.00 1.00 1 = Residence Life/Student Conduct; 0 =
All others
Functional Area: Multicultural/
Diversity Programs
b
0.04 0.00 1.00 1 = Multicultural/Diversity Programs; 0
= All others
Functional Area: Auxiliary/Support
b

0.06 0.00 1.00 1 = Auxiliary/Support; 0 = All others
Years at Institution

7.98 7.06 1.00 31.00 Units in years

116
REFERENCES
ACPA – College Student Educators International. (1994). The student learning imperative:
Implications for student affairs. Retrieved from http://www.myacpa.org.sli_delet/sli.htm
Allen, K. E., & Cherrey, C. (2000). Systemic leadership. Lanham, MD: University Press of
America, Inc.
American Council on Education. (1949). The student personnel point of view, 1949. Retrieved
from http://www.myacpa.org/sites/default/files/student-personnel-point-of-view-1949.pdf
Amey, M. J. (2006). Leadership in higher education. Change, 38(6), 55-58.
Antonnen, R. G., & Chaskes, J. (2002). Advocating for first-year students: A study of the
micropolitics of leadership and organizational change. Journal of the First-Year
Experience, 14(1), 81-98.
Axelrod, R., & Cohen, M. D. (2000). Harnessing complexity: Organizational implications of a
scientific frontier. New York: Basic Books.
Baldwin, T. T., Bedell, M. D., & Johnson, J. L. (1997). The social fabric of a team-based M.B.A.
program: Network effects on student satisfaction and performance. The Academy of
Management Journal, 40, 1369-1397.
Balkundi, P., & Kilduff, M. (2006). The ties that lead: A social network approach to leadership.
The Leadership Quarterly, 17, 419-439.
Bensimon, E. M., & Neumann, A. (1993). Redesigning collegiate leadership: Teams and
teamwork in higher education. Baltimore, MD: Johns Hopkins University Press.
Biddix, J. P. (2011). “Stepping stones”: Career paths to the SSAO for men and women at four-
year institutions. Journal of Student Affairs Research and Practice, 48, 443-461.
Boardman, C. P., & Ponomariov, B. L. (2007). Reward systems and NSF university research
centers: The impact of tenure on university scientists’ valuation of applied and
commercially relevant research. The Journal of Higher Education, 78, 51-70.
Bolman, L. G., & Deal, T. E. (2008). Reframing organizations: Artistry, choice, and leadership,
4
th
edition. San Francisco: Jossey-Bass.
Borgatti, S. P., & Foster, P. C. (2003). The network paradigm in organizational research: A
review and typology. Journal of Management, 29, 991-1013.
Borgatti, S. P., & Halgin, D. S. (2011). On network theory. Organization Science, 22, 1168-
1181.

117
Borgatti, S. P., & Ofem, B. (2010). Overview: Social network theory and analysis. In A. J. Daly
(Ed.), Social network theory and educational change (pp. 259-274). Cambridge, MA:
Harvard Education Press.
Borgatti, S.P., Everett, M.G. & Freeman, L.C. (2002). UCINET for Windows: Software for social
network analysis [Computer software]. Harvard, MA: Analytic Technologies.
Bourassa, D. M., & Kruger, K. (2001). The national dialogue on academic and student affairs
collaboration.” In A. J. Kezar, D. J. Hirsch, & C. Burack (Eds.), Understanding the role
of academic and student affairs collaboration in creating a successful learning
environment (pp. 9-38). New Directions for Higher Education, no. 116. San Francisco:
Jossey-Bass.
Bowler, W. M., & Brass, D. J. (2006). Relational correlates of interpersonal citizenship behavior:
A social network perspective. Journal of Applied Psychology, 91, 70-82.
Boyatzis, R. E. (1998). Transforming qualitative information: Thematic analysis and code
development. Thousand Oaks, CA: SAGE.
Brass, D. J. (1985). Men's and women's networks: A study of interaction patterns and influence
in an organization. Academy of Management Journal,28, 327-343.
Brass, D. J. (1995). Creativity: It’s all in your social network. In C. M. Ford, D. A. Gioia (Eds.),
Creative action in organizations (pp. 94-99). Thousand Oaks, CA: SAGE.
Brass, D. J., & Krackhardt, D. (1999). Social capital for twenty-first century leaders. In J. G.
Hunt, & R. L. Phillips (Eds.), Out-of-the-box leadership challenges for the 21
st
century
army (pp. 179-194). Amsterdam: Elsevier B.V.
Brass, D. J., Galaskiewicz, J., Greve, H. R., & Tsai, W. (2004). Taking stock of networks and
organizations: A multilevel perspective. The Academy of Management Journal, 47, 795-
817.
Bresciani, M. J., Todd, D. K., & Associates. (2010). Professional competency areas for student
affairs practitioners. Washington, DC: ACPA/NASPA. Retrieved online from
http://www.naspa.org/programs/prodev/Professional_Competencies.pdf
Burkard, A. W., Cole, D. C., Ott, M., & Stoflet, T. (2005). Entry-level competencies of new
student affairs professionals: A Delphi study. Journal of Student Affairs Research and
Practice, 42(3), 545-571.
Burt, R. S. (1992). Structural holes: The social structure of competition. Boston, MA: Harvard
University Press.
Caple, R. B., & Newton, F. B. (1991). Leadership in student affairs. In T. K. Miller, R. B.
Winston, Jr., & Associates (Eds.), Administration and leadership in student affairs:
Actualizing student development in higher education, 2
nd
edition (pp. 111-133). Muncie,
IN: Accelerated Development.

118
Carboni, I., & Ehrlich, K. (2013). The effect of relational and team characteristics on individual
performance: A social network perspective. Human Resource Management, 52, 511-535.
Castellanos, J., Gloria, A. M., Mayorga, M., & Salas, C. (2007). Student affairs professionals’
self-report of multicultural competence: Understanding awareness, knowledge, and skills.
NASPA Journal, 44(4), 643–663.
Cilente, K., Henning, G., Skinner Jackson, J., Kennedy, D., & Sloane, T. (2007). Report on the
new professional needs study. Washington, DC: ACPA-College Student Educators
International.
Clayborne, H. L. (2006). Leadership experiences of African American women who are mid-level
student affairs administrators (Doctoral dissertation). Available from ProQuest
Dissertations and Theses database. (UMI No. 3217263).
Clement, L. M., & Rickard, S. T. (1992). Effective leadership in student services: Voices from
the field. San Francisco: Jossey-Bass.
Coburn, C. E., Choi, L, & Mata, W. (2010). “I would go to her because her mind is math”:
Network formation in the context of a district-based mathematics reform. In A. J. Daly
(Ed.), Social network theory and educational change (pp. 33-50). Cambridge, MA:
Harvard Education Press.
Coburn, C. E., Russell, J. L., Kaufman, J. H., & Stein, M. K. (2012). American Journal of
Education, 119, 137-182.
Coffey, C. M. (2010). A study of competencies perceived to be important by professionals in
entry-level positions within college student affairs (Unpublished doctoral dissertation).
University of Central Florida, Orlando.
Coleman, J. S. (1988). Social capital in the creation of human capital. The American Journal of
Sociology, 94, S95-S120.
Creswell, J. W. (2003). Research design: Qualitative, quantitative, and mixed methods
approaches, 2
nd
edition. Thousand Oaks, CA: SAGE.
Cross, R., & Cummings, J. N. (2004). Tie and network correlates of individual performance in
knowledge-intensive work. The Academy of Management Journal, 47, 928-937.
Cuyjet, M. J. (2009). Perceptions of new student affairs professionals and their supervisors
regarding the application of competencies learned in preparation programs. Journal of
College Student Development, 50, 104-119.
Daly, A. J. (2010a). Mapping the terrain: Social network theory and educational change. In A. J.
Daly (Ed.), Social network theory and educational change (pp. 1-16). Cambridge, MA:
Harvard Education Press.

119
Daly, A. J. (Ed.) (2010b). Social network theory and educational change. Cambridge, MA:
Harvard Education Press.
Daniel, E. D. (2011). Perceived leadership practices of student affairs professionals: An analysis
of demographic factors (Doctoral dissertation). Available from ProQuest Dissertations
and Theses database. (UMI No. 3482370).
DeMaio, T. J. (1984). Social desirability and survey measurement: A review. Surveying
subjective phenomena, 2, 257-281.
Denison, D. R., Hart, S. L., & Kahn, J. A. (1996). From chimneys to cross-functional teams:
Developing and validating a diagnostic model. Academy of Management Journal, 39,
1005-1023.
DiMaggio, P., & Garip, F. (2012). Network effects and social inequality. Annual Review of
Sociology, 38, 93-118.
Dose, J. J. (2012). Organizational effectiveness in higher education: Faculty informal structures
as social capital. Paper presented at the Association for the Study of Higher Education
National Conference, Las Vegas, NV.
Dungy, G. J. (2003). Organization and functions of student affairs. In S. R. Komives, D. B. &
Woodard, Jr. (Eds.), Student services: A handbook for the profession (pp. 339-357). San
Francisco: Jossey-Bass.
Eddy, P. L., & VanDerLinden, K. E. (2006). Emerging definitions of leadership in higher
education: New visions of leadership or same old “hero” leader? Community College
Review, 34, 5-26.
Elkins, G. G. (2006). Chief student affairs officers of institutions belonging to the Council on
Christian Colleges and Universities: Educator, leader, or manager (Unpublished
doctoral dissertation). Texas Tech University, Lubbock.
Evans, N. J., & Reason, R. D. (2001). Guiding principles: A review and analysis of student
affairs philosophical statements. Journal of College Student Development, 42, 359-377.
Festinger, L., Schachter, S., & Back, K. (1950). Social pressures in informal groups: A study of
human factors in housing. New York: Harper.
Fey, C. J., & Carpenter, D. S. (1996). Mid-level student affairs administrators: Management
skills and professional development needs. NASPA Journal, 33, 218-231.
Fried, J. (2007). Higher education’s new playbook: Learning reconsidered. About Campus,
12(1), 2-7.
Fritsch, M., & Kauffeld-Monz, M. (2010). The impact of network structure on knowledge
transfer: an application of social network analysis in the context of regional innovation
networks. The Annals of Regional Science, 44(1), 21-38.

120
Garcon. R. (2012). Identifying quantitative relationships among student affairs administrators
pertaining to their leadership frames and practices (Doctoral dissertation). Available
from ProQuest Dissertations and Theses database. (UMI No. 3570766).
Gehrke, S. (2015, March). A social network approach to student affairs: The power of
relationships. Program presented at the ACPA-College Student Educators International
Annual Convention, Tampa, FL.
Gillis, L. R. (2008). A network perspective of multiple social exchange relationships (Doctoral
dissertation). Available from ProQuest Dissertations and Theses database. (UMI No.
3311476).
Golde, C.M. & Pribbenbow, D.A. (2000). Understanding faculty involvement in residential
learning communities. Journal of College Student Development, 41, 27-40.
Googins, B. K., & Rochlin, S. A. (2000). Creating the partnership society: Understanding the
rhetoric and reality of cross-sectoral partnerships. Business and Society Review, 105(1),
127-144.
Gordon, S. E., Strode, C. B., & Mann, B. A. (1993). The mid-manager in student affairs: What
are CSAOs looking for? NASPA Journal, 30(4), 290–297.
Granovetter , M. S. (1982). The strength of weak ties: A network theory revisited. In P. Marsden,
& N. Linn (Eds.), Social structure and network analysis (pp. 201-233). Beverly Hills,
CA: SAGE.
Granovetter, M. S. (1973). The strength of weak ties. American journal of sociology, 1360-1380.
Hagadoorn, J. (1993). Understanding the rationale of strategic partnering: Interorganizational
modes of cooperation and sectoral differences. Strategic Management Journal, 14, 371-
385.
Handy, C. (1999). Beyond certainty: The changing worlds of organizations. Boston, MA:
Harvard Business School Press.
Hannah, S. T., & Lester, P. B. (2009). A multilevel approach to building and leading learning
organizations. The Leadership Quarterly, 20, 34-48.
Heifetz, R. A. (1994). Leadership without easy answers. Cambridge, MA: Harvard University
Press.
Herdlein, R., Riefler, L., & Mrowka, K. (2013). An integrative literature review of student affairs
competencies: A meta-analysis. Journal of Student Affairs Research and Practice, 50,
250-269.
Hoppe, B., & Reinelt, C. (2010). Social network analysis and the evaluation of leadership
networks. The Leadership Quarterly, 21, 600-619.

121
Ibarra, H. (1992). Homophily and differential returns: Sex differences in network structure and
access in an advertising firm. Administrative Science Quarterly, 37, 422-447.
Ibarra, H., & Andrews, S. B. (1993). Power, social influence, and sense making: Effects of
network centrality and proximity on employee perceptions. Administrative Science
Quarterly, 38, 277-303.
Joint Taskforce on Student Learning. (1998). Powerful partnerships: A shared responsibility for
learning. Washington, D.C.: AAHE, ACPA, and NASPA.
Kadushin, C. (2012). Understanding social networks: Theories, concepts, and findings. New
York: Oxford University Press.
Kane, S. D. (2001). Mid-level student affairs administrators’ organizational “frame”
orientations and perceptions of their effectiveness by senior student affairs officers
(Doctoral dissertation). Available from ProQuest Dissertations and Theses database.
(UMI No. 3035847).
Katherine, L. (2011). Leadership competency preferences of student affairs administrators: Does
generation matter? (Doctoral dissertation). Available from ProQuest Dissertations and
Theses database. (UMI No. 3477242).
Keeling, R. P. (Ed.). (2004). Learning reconsidered: A campus-wide focus on the student
experience. Washington, DC: NASPA and ACPA.
Kerr, K., Porterfield, K., Davenport, Z., & Roberts, G. (2014). Higher education affordability
and concerns about the rising costs: Blaming student services is not the answer.
Retrieved from http://www.myacpa.org/article/higher-education-affordability-and-
concerns-about-rising-costs-blaming-student-services-not
Kezar, A. (2005). Redesigning for collaboration within higher education institutions: An
exploration into the developmental process. Research in Higher Education, 46, 831-860.
Kezar, A. (2011). Organizational theory. In J. Schuh, S. Jones, and S. Harper (eds). Student
services: A handbook for the profession (5th edition) (pp. 226-241). San Francisco:
Jossey Bass.
Kezar, A. (2014). Higher education and social networks: A review of research. The Journal of
Higher Education, 85, 91-125.
Kezar, A. J. (2006). Redesigning for collaboration in learning initiatives: An examination of four
highly collaborative campuses. The Journal of Higher Education, 77, 804-838.
Kezar, A. J., & Lester, J. (2009). Organizing higher education for collaboration: A guide for
campus leaders. San Francisco: Jossey-Bass.
Kezar, A. J., & Lester, J. (2011). Enhancing campus capacity for leadership: An examination of
grassroots leaders in higher education. Stanford, CA: Stanford University Press.

122
Kezar, A. J., Carducci, R., & Contreras-McGavin, M. (2006). Rethinking the "L" word in higher
education: The revolution in research on leadership. ASHE Higher Education Report,
Vol. 31, No. 6. San Francisco: Jossey-Bass.
Kezar, A., & Gehrke, S. J. (in press). Supporting and enhancing student learning through
partnerships with academic colleagues. In G. S. McClellan & J. Stringer (Eds.), The
handbook of student affairs administration, 4
th
edition. San Francisco: Jossey-Bass.
Kilduff, M., & Brass, D. J. (2010). Organizational social network research: Core ideas and key
debates. The Academy of Management Annals, 4(1), 317-357.
Kilduff, M., & Tsai, W. (2003). Social networks and organizations. Thousand Oaks, CA: SAGE.
Kilduff, M., Tsai, W., & Hanke, R. (2006). A paradigm too far? A dynamic stability
reconsideration of the social network research program. The Academy of Management
Review, 31, 1031-1048.
Komives, S. R., & Dugan, J. P. (2010). Contemporary leadership theories. In R. Couto (Ed.),
Political and civic leadership, Vol. 1 (pp. 111-120). Thousand Oaks, CA: SAGE.
Komives, S. R., Lucas, N. & McMahon, T. R. (2007). Exploring leadership: For college
students who want to make a difference, 2
nd
edition. San Francisco: Jossey-Bass.
Krackhardt, D. (1990). Assessing the political landscape: Structure, cognition, and power in
organizations. Administrative Science Quarterly, 35, 342-369.
Krackhardt, D. (1992). The strength of strong ties: The importance of philos in organizations. In
R. Eccles, & N. Nohria Cambridge (Eds.), Networks and organizations: Structure, form,
and action (pp. 216-239). Cambridge, MA: Harvard Business School.
Krackhardt, D., & Stern, R. N. (1988). Informal networks and organizational crises: An
experimental simulation. Social Psychology Quarterly, 51, 123-140.
Kretovics, M. (2002). Entry-level competencies: What student affairs administrators consider
when screening candidates. Journal of College Student Development, 43(6), 912-920.
Kreuter, F., Presser, S., & Tourangeau, R. (2008). Social desirability bias in CATI, IVR, and
Web surveys the effects of mode and question sensitivity. Public Opinion
Quarterly, 72(5), 847-865.
Kuh, G. D., Kinzie, J., Schuh, J. H., & Whitt, E. J. (2011). Student success in college: Creating
conditions that matter. San Francisco: Jossey-Bass.
Kuk, L., Banning, J. H., & Amey, M. J. (2010). Positioning student affairs for sustainable
change: Achieving organizational effectiveness through multiple perspectives. Sterling,
VA: Stylus.

123
Kuk, L., Cobb, B., & Forrest, C. (2007). Perceptions of competencies of entry-level practitioners
in student affairs. NASPA Journal, 44(4), 664-691.
Lazarsfeld, P. F., & Merton, R. K. (1954). Friendship as social process: A substantive and
methodological analysis. In M. Berger, T. Abel, & C. H. Page (Eds.), Freedom and
control in modern society (pp. 18-66). New York: Van Nostrand.
Leana, C., & Pil, F. .(2006). Social capital and organizational performance: Evidence from urban
public schools. Organization Science, 17, 353-366.
Leslie, L. L., & Rhoades, G. (1995). Rising administrative costs: Seeking explanations. The
Journal of Higher Education, 66(2), 187-212.
Lichtenstein, B. B., Uhl-Bien, M., Marion, R., Seers, A., Orton, J. D., & Schreiber, C. (2006).
Complexity leadership theory: An interactive perspective on leading in complex adaptive
systems. E:CO, 8(4), 2-12.
Love, P., & Estanek, S. M. (2004). Rethinking student affairs practice. San Francisco: Jossey-
Bass.
Lovell, C. D., & Kosten, L. A. (2000). Skills, knowledge, and personal traits necessary for
success as a student affairs administrator: A meta-analysis of thirty years of research.
NASPA Journal, 37, 553-572.
Macchio, A. H. (2012). Organizational learning in student affairs: The role of leadership and
culture in fostering learning (Doctoral dissertation). Available from ProQuest
Dissertations and Theses database. (UMI No. 3510973).
Manning, K. (1996). Contemplating the myths of student affairs. NASPA Journal, 34, 36-46.
Manning, K., Kinzie, J., & Schuh, J. (2006). One size does not fit all: Traditional and innovative
models of student affairs practice. New York: Routledge.
Marcus, J. (2014). New analysis shows problematic boom in higher ed administrators. Center for
New England Investigative Reporting. Retrieved from http://necir.org/2014/02/06/new-
analysis-shows-problematic-boom-in-higher-ed-administrators/
Marion, R. (1999). The edge of organization: Chaos and complexity theories of formal social
systems. Thousand Oaks, CA: SAGE.
Marion, R., & Uhl-Bien, M. (2001). Leadership in complex organizations. The Leadership
Quarterly, 12, 389-418.
McPherson, M., Smith-Lovin, L., & Cook, J. M. (2001). Birds of a feather: Homophily in social
networks. Annual review of sociology, 27, 415-444.

124
Mehra, A., Dixon, A. L., Brass, D. J., & Robertson, B. (2006). The social network ties of group
leaders: Implications for group performance and leader reputation. Organization Science,
17, 64-79.
Mehra, A., Kilduff, M., & Brass, D. J. (1998). At the margins: A distinctiveness approach to the
social identity and social networks of underrepresented groups. The Academy of
Management Journal, 41, 441-452.
Mehra, A., Kilduff, M., & Brass, D. J. (2001). The social networks of high and low self-
monitors: Implications for workplace performance. Administrative Science Quarterly, 46,
121-146.
Mertens, D. M. (2005). Research and evaluation in education and psychology: Integrating
diversity with quantitative, qualitative, and mixed methods, 2
nd
edition. Thousand Oaks,
CA: SAGE.
Monge, P. R., & Contractor, N. S. (2003). Theories of communication networks. New York:
Oxford University Press.
Moolenar, N. M. (2012). A social network perspective on teacher collaboration in schools:
Theory, methodology, and applications. American Journal of Education, 119, 7-39.
Moolenar, N. M., & Sleegers, P. J. C. (2010). Social networks, trust, and innovation: The role of
relationships in supporting an innovative climate in Dutch schools. In A. J. Daly (Ed.),
Social network theory and educational change (pp. 97-114). Cambridge, MA: Harvard
Education Press.
Nelson, R. E. (1989). The strength of strong ties: Social networks and intergroup conflict in
organizations. Academy of Management Journal, 32, 377-401.
Newton, R. R., & Rudestam, E. K. (1999). Your statistical consultant: Answers to your statistical
questions. Thousand Oaks: SAGE.
Nidiffer, J. (2001). New leadership for a new century: Women’s contribution to leadership in
higher education. In J. Nidiffer & C. T. Brashaw (Eds.), Women administrators in higher
education: Historical and contemporary perspectives (pp. 101-131). Albany, NY: State
University of New York Press.
Oh, J. (2013). Leadership strategies, skills, and professional approaches utilized by effective
senior-level student affairs administrators at urban universities (Doctoral dissertation).
Available from ProQuest Dissertations and Theses database. (UMI No. 3563965).
Ostroff, C., Kinicki, A. J., & Tamkins, M. M. (2003). Organizational culture and climate. In W.
C. Borman, D. R. Ilgen, & R. J. Klimoski (Eds.), Handbook of Psychology: Volume 12,
Industrial and Organizational Psychology (pp. 565-594). San Francisco: Wiley & Sons.
Papa, M. J. (1990). Communication network patterns and employee performance with new
technology. Communication Research, 17, 344-368.

125
Pearce, J. A., & David, F. R. (1983). A social network approach to organizational design-
performance. Academy of Management Review, 8, 436-444.
Penuel, W. R., Frank, K. A., & Krause, A. (2010). Between leaders and teachers: Using social
network analysis to examine the effects of distributed leadership. In A. J. Daly (Ed.),
Social network theory and educational change (pp. 159-178). Cambridge, MA: Harvard
Education Press.
Penuel, W. R., Sun, M., Frank, K. A., & Gallagher, H. A. (2012). Using social network analysis
to study how collegial interactions can augment teacher learning from external
professional development. American Journal of Education, 119, 103-136.
Perry-Smith, J. E., & Shalley, C. E. (2003). The social side of creativity: A static and dynamic
social network perspective. Academy of Management Review, 28, 89-106.
Phillips, R. L., & Hunt, J. G. (Eds.) (1992). Strategic leadership: A multiorganizational-level
perspective. Westport, CT: Quorum Books.
Podsakoff, P. M., & Organ, D. W. (1986). Self-reports in organizational research: Problems and
prospects. Journal of Management, 12, 531-544.
Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common method
biases in behavioral research: a critical review of the literature and recommended
remedies. Journal of Applied Psychology, 88, 879-903.
Pope, R. L., & Reynolds, A. L. (1997). Student affairs core competencies: Integrating
multicultural awareness, knowledge, and skills. Journal of College Student Development,
38, 266–277.
Porterfield, K. T., Roper, L. D., & Whitt, E. J. (2011). Redefining our mission: What does higher
education need from student affairs? Journal of College and Character, 12(4), 1-7.
Rabe-Hesketh, S., & Skrondal, A. (2012). Multilevel and longitudinal modeling using stata,
volume 1: Continuous responses (3
rd
edition). College Station, TX: Stata Press.
Randall, K., & Globetti, E. (1992). Desired competencies of the chief student affairs officer as
perceived by college presidents. College Student Affairs Journal, 11(3), 54–61.
Reagans, R., & Zuckerman, E. W. (2001). Networks, diversity, and productivity: The social
capital of R&D teams. Organization Science, 12, 502-517.
Reagans, R., Zuckerman, E. W., & McEvily, B. (2004). How to make the team: Social network
versus demography as criteria for designing effective teams. Administrative Science
Quarterly, 49, 101-133.
Renn, K. A., & Hodges, J. P. (2007). The first year on the job: Experiences of new professionals
in student affairs. NASPA Journal, 44, 367-391.

126
ResearchWare, Inc. (2013). HyperRESEARCH computer software (Version 3.5.2) [Software].
Available from http://www.researchware.com
Reynolds, A. L. (2011). Helping competencies of student affairs professionals: A Delphi study.
Journal of College Student Development, 52, 362–369.
Roberts, D. C. (2001). Leading. In R.B. Winston, Jr., D. G. Creamer, & T. K. Miller (Eds.), The
professional student affairs administrator: Educator, leader, and manager (pp.379-398).
New York: Brunner-Routledge.
Rodems, M. R. (2011). Collaborative relationships between faculty and student affairs
professionals: A case study (Doctoral dissertation). Available from ProQuest
Dissertations and Theses database. (UMI No. 3493142).
Rogers, J. L. (2003). Leadership. In S. R. Komives & D. B. Woodard, Jr. (Eds.), Student
services: A handbook for the profession, 4
th
edition (pp. 447-465). San Francisco: Jossey-
Bass.
Roper, L. (2002). Relationships: The critical ties that bind professionals. In J. C. Dalton, & M.
McClinton (Eds.), The Art and Practical Wisdom of Student Affairs Leadership (pp. 11-
26). (New Directions for Student Services, No. 98). San Francisco: Jossey-Bass.
Rost, J. C. (1991). Leadership for the twenty-first century. New York: Praeger.
Sandeen, A. (1991). The chief student affairs officer: Leader, manager, mediator, educator. San
Francisco: Jossey-Bass.
Satish, U., & Streufert, S. (1997). The measurement of behavioral complexity. Journal of applied
Social Psychology, 27, 2117-2121.
Saunders, S. A., & Cooper, D. L. (1999). The doctorate in student affairs: essential skills and
competencies for midmanagement. Journal of College Student Development, 40(2), 185–
191.
Schroeder, C. C. (1999a). Collaboration and partnerships. In C. S. Johnson, & H. E. Cheatham
(Eds.), Higher education trends for the next century: A research agenda for student
success. Washington, D.C.: American College Personnel Association.
Schroeder, C. C. (1999b). Forging educational partnerships that advance student learning. In G.
S. Blimling, & E. J. Whitt (Eds.), Good practice in student affairs. San Francisco: Jossey-
Bass.
Schuh, J. H. (2002). Lessons from leaders. NASPA Journal, 39, 204-216.
Settoon, R. P., & Mossholder, K. W. (2002). Relationship quality and relationship context as
antecedents of person- and task-focused interpersonal citizenship behavior. Journal of
Applied Psychology, 87, 255-267.

127
Smith, B. L., & Hughey, A. W. (2006). Leadership in higher education – its evolution and
potential: A unique role facing critical challenges. Industry and Higher Education, 20(3),
157-163.
Smith, M. A. (2013). Senior-level student affairs administrators self-reported leadership
practices, behaviors, and strategies (Doctoral dissertation). Available from ProQuest
Dissertations and Theses database. (UMI No. 3564020).
Smith, R. A. (2011). Learning community transitions in the first year: A case study of academic
and social network change. Journal of the First-Year Experience & Students in
Transition, 23(2), 13-31.
Soda, G., & Zaheer, A. (2012). A network perspective on organizational architecture:
Performance effects of the interplay of formal and informal organization. Strategic
Management Journal, 33, 751-771.
Sparrowe, R. T., Liden, R. C., Wayne, S. J., & Kraimer, M. L. (2001). Social networks and the
performance of individuals and groups. The Academy of Management Journal, 44, 316-
325.
Spector, P. E. (1994). Using self‐ report questionnaires in OB research: A comment on the use of
a controversial method. Journal of Organizational Behavior, 15, 385-392.
Spillane, J. P., Healey, K., & Kim, C. M. (2010). Leading and managing instruction: Formal and
informal aspects of the elementary school organization. In A. J. Daly (Ed.), Social
network theory and educational change (pp. 129-158). Cambridge, MA: Harvard
Education Press.
Sriram, R. (2014). The development, validity, and reliability of a psychometric instrument
measuring competencies in student affairs. Journal of Student Affairs Research and
Practice, 51, 349-363.
Stein, C. H., Vickio, C. J., Fogo, W. R., & Abraham, K. M. (2007). Making connections: A
network approach to university disaster preparedness. Journal of College Student
Development, 48, 331-343.
Streufert, S. (1997). Complexity: An integration of theories. Journal of Applied Psychology, 27,
2068-2095.
Task Force on the Future of Student Affairs. (2010). Envisioning the future of student affairs.
Washington, DC: ACPA and NASPA.
Taylor, J. D. (2001). Finding the right stuff in chief student affairs officers (Doctoral
dissertation). Available from ProQuest Dissertations and Theses database. (UMI No.
3013033).
Thomas, S. L. (2000). Ties that bind: A social network approach to understanding student
integration and persistence. The Journal of Higher Education, 71, 591-615.

128
Tull, A., & Freeman, J. P. (2011). Reframing student affairs leadership: An analysis of
organizational frames of reference and locus of control. Research in the Schools, 18, 33-
43.
Uhl-Bien, M. (2006). Relational leadership theory: Exploring the social processes of leadership
and organizing. The Leadership Quarterly, 17, 654-676.
Uhl-Bien, M., Marion, R., & McKelvey, B.. (2007). Complexity leadership theory: Shifting
leadership from the industrial age to the knowledge era. The Leadership Quarterly, 18,
298-318.
Valente, T. W. (2010). Social networks and health: Models, methods, and applications. New
York: Oxford University Press.
Valente, T. W., & Davis, R. L. (1999). Accelerating the diffusion of innovations using opinion
leaders. Annals of the American Academy of Political and Social Sciences, 566, 55-67.
Valente, T. W., & Foreman, R. K. (1998). Integration and radiality: Measuring the extent of an
individual’s connectedness and reachability in a network. Social Networks, 20, 89-105.
Valente, T. W., & Pumpuang, P. (2007). Identifying opinion leaders to promote behavior change.
Health Education & Behavior, 34, 881-896.
Waple, J. N. (2006). An assessment of skills and competencies necessary for entry-level student
affairs work. NASPA journal, 43(1).
Wheatley, M J. (1992). Leadership and the new science: Learning about organization from an
orderly universe. San Francisco: Berrett-Koehler Publishers.
Whitt, E. J. (2011). Academic and student affairs partnerships. In J. H. Schuh, S. R. Jones, S. R.
Harper (Eds.), Student services: A handbook for the profession (pp. 482-496). San
Francisco: Jossey-Bass.
Wielkiewicz, R. M. (2000). The leadership attitudes and beliefs scale: An instrument for
evaluating college students’ thinking about leadership and organizations. Journal of
College Student Development, 41, 337-349.
Wilson, M. E., Comes, M. D., & Dannells, M. (2006). Chronology of foundational documents of
the student affairs profession. Retrieved from http://www2.bgsu.edu/colleges/library/cac/
sahp/word/Chronology%20of%20Found%20Docs11.28.pdf
Winston, R .B., Creamer, D. G., & Miller, T. K. (Eds.) (2011). The professional student affairs
administrator: Educator, leader, and manager. New York: Brunner-Routledge
Woodard, D. B., Love, P., & Komives, S. R. (2000). Leadership and management issues for a
new century. (New Directions for Student Services, No. 92). San Francisco: Jossey-Bass.

Abstract (if available)

Abstract The research in this dissertation examines cross‐institutional collaboration among student affairs administrators utilizing social network analysis as a theoretical and methodological framework. Student affairs professionals are increasingly being called to collaborate across organizational boundaries, yet the extent to which they engage in these practices and the outcomes and strategies associated with these practices is under‐studied in the student affairs and higher education literature. Social network analysis provides a useful theoretical and methodological framework for examining collaboration, as well as other practices, in student affairs. ❧ The three chapters in this dissertation are each written as standalone studies utilizing data from one of two surveys. I designed these surveys to examine collaborative social networks among student affairs administrators. The first survey was administered to a large, representative sample of student affairs administrators and focused on these professionals’ collaborative ego networks. The second survey was administered to professionals in three divisions of student affairs at comparable higher education institutions and was designed to capture data about the collaboration network among the professionals in each of these institutions. ❧ In Chapter 1, I explore the collaborative social networks of student affairs professionals and identify network behaviors associated with cross‐boundary collaboration, both across student affairs departments and with others outside of student affairs. The student affairs professionals in this study engage in collaboration with other student affairs departments more than with others outside of their student affairs divisions. Additionally, they tend to collaborate in closed networks and rarely collaborate with faculty. Findings from the random effects models reveal that collaborating in closed networks and with those who are physically nearby on campus is negatively associated with cross‐boundary collaboration, and that student affairs professionals are drawn to collaborate more with others who share their gender and racial/ethnic identity. ❧ In Chapter 2, I examine the extent to which student affairs professionals’ collaborative network behaviors are associated with organizationally‐related student affairs competencies, as well as trends in student affairs professionals’ perceived proficiencies in these areas. The findings from this study reveal that most of the variance in perceived proficiency in these competencies is due to individual characteristics, but collaborative behaviors, specifically developing strong ties to one’s collaborators, is significantly associated with proficiency in these competency areas. ❧ In Chapter 3, I utilize network analysis to measure systemic leadership among student affairs professionals from three institutions and examine collaborative behaviors and strategies employed by systemic leaders on these campuses. The resulting model informed by the analyses in this study highlights several collaborative network behaviors—including unconstrained personal networks, stronger ties, and less homophily based on propinquity—as well as strategies—including a big‐picture orientation, openness to new ideas, political savvy, and collaborating with a student‐focus—which provide a roadmap of sorts for student affairs professionals who seek to practice and foster systemic leadership. ❧ Together, these three studies highlight the important contributions that can be made to the student affairs and higher education literature by utilizing social network analysis as a theoretical and methodological framework. A social network paradigm holds the potential to allow student affairs professionals to think about their work in new ways and to examine student affairs practice from a new perspective, allowing the field to reframe important issues and conceptualize new approaches to understanding and engaging in its work.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

How college students use online social networks to gather information

PDF

An examination of collaboration amongst student affairs and academic affairs professionals in an action research project

PDF

Evaluating the effects of diversity courses and student diversity experiences on undergraduate students' democratic values at a private urban research institution

PDF

Leadership strategies, skills, and professional approaches utilized by effective senior-level student affairs administrators at urban universities

PDF

Whiteness: a narrative analysis on student affairs professionals, race, identity, and multicultural competency

PDF

How do non-tenure track faculty interact with Latino and Latina students in gatekeeper math courses at an urban community college?

PDF

Senior-level student affairs' administrators' self-reported leadership practices, behaviors, and strategies

PDF

Coloring the pipeline: an analysis of the NASPA Undergraduate Fellows program as a path for underrepresented students into student affairs

PDF

The intersection of grit and social capital: a mixed methods examination of successful first-generation college students

PDF

A network analysis of online and offline social influence processes in relation to adolescent smoking and alcohol use

PDF

Preparing for the field of student affairs: White graduate students and the social justice and inclusion competency

PDF

Exploring faculty-student interactions in a comprehensive college transition program for low-income and first-generation college students

PDF

Knowledge sharing among student affairs professionals at a small faith-based higher education institution

PDF

The gamification of judicial affairs: addressing learning for 'digital students'

PDF

Departure from the student affairs profession: a study of professionals who left the field

PDF

Building social Legoland through collaborative crowdsourcing: marginality, functional diversity, and team success

PDF

From coexistence to collaboration: understanding interschool interactions at co‐located sites

PDF

The social media dilemma in education: policy design, implementation and effects

PDF

Exploring three outcomes of online teacher preparation: teaching for social justice, critical reflection, and voluntary collaboration

PDF

The development and implementation of global partnerships in student affairs

Asset Metadata

Creator Gehrke, Sean (author)

Core Title Collaborative social networks in student affairs: an exploration of the outcomes and strategies associated with cross‐institutional collaboration

School Rossier School of Education

Degree Doctor of Philosophy

Degree Program Urban Education Policy

Publication Date 07/09/2015

Defense Date 04/23/2015

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Administration,collaboration,Higher education,leadership,OAI-PMH Harvest,social network analysis,social networks,student affairs,survey research

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Kezar, Adrianna J. (committee chair), Cole, Darnell G. (committee member), Tobey, Patricia Elaine (committee member), Valente, Thomas W. (committee member)

Creator Email sean.gehrke@gmail.com,sgehrke@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-590313

Unique identifier UC11299835

Identifier etd-GehrkeSean-3573.pdf (filename),usctheses-c3-590313 (legacy record id)

Legacy Identifier etd-GehrkeSean-3573.pdf

Dmrecord 590313

Document Type Dissertation

Format application/pdf (imt)

Rights Gehrke, Sean

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