University of Southern California Dissertations and Theses

Role of college program administrators in addressing gender gap in computer science

(USC Thesis Other)

Role of college program administrators in addressing gender gap in computer science

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Running head: GENDER GAP IN COMPUTER SCIENCE 1

Role of College Program Administrators in Addressing Gender Gap in Computer Science
by
Eugeny Levinzon

A Dissertation Presented to the
FACULTY OF THE USC ROSSIER SCHOOL OF EDUCATION
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF EDUCATION

August 2018

Copyright 2018 Eugeny Levinzon

GENDER GAP IN COMPUTER SCIENCE 2
DEDICATION

This dissertation is dedicated to my wife Victoria and to my children, Abby, Glen, and Nate.
I could never get this far without your encouragement, support, and love!
We are truly one.

GENDER GAP IN COMPUTER SCIENCE 3
ACKNOWLEDGEMENTS

First and foremost, I would like to thank my dissertation committee chair, Dr. Mora-
Flores. Her belief in my ability to succeed was a constant source of motivation. I will forever be
grateful for her support. I would like to thank my dissertation committee members, Dr. Freking
and Dr. Maddox. Their subject matter expertise guided my journey from start to finish.
I would also like to acknowledge the entire faculty team at Rossier. Because of their
dedication to student learning, I always felt I was on the 'right bus'.
Finally, I would like to thank my friend Dr. C.F. whose life-long dedication to achieving
diversity in computer science has served as an inspiration for this study.

GENDER GAP IN COMPUTER SCIENCE 4
ABSTRACT
A computer science program leadership team at Technical College (TC) realized that to
build a community of future computer science leaders, a primary mission of the program, they
need to achieve a more diverse and inclusive environment at TC. For over a decade several
initiatives focused on enhancing the organizational structure, work processes, and program
culture were implemented resulting in TC becoming one of the more gender balanced top
ranking computer science (CS) programs in the country. The purpose of this case study is to
focus on the role that CS program administrators play in addressing the gender gap in computer
science program at TC.

GENDER GAP IN COMPUTER SCIENCE 5
TABLE OF CONTENTS

Dedication 2
Acknowledgements 3
Abstract 4
List of Tables 7
List of Figures 8
Introduction of Problem of Practice 9
Importance of Addressing Problem 9
Purpose of Project and Questions 11
Organizational Context and Mission 11
Organizational Performance Goal 12
Description of Stakeholder Groups and Participating Stakeholders 13
Literature Review Sections Overview 14
Literature Review 15
The Clark and Estes Gap Analysis Conceptual Framework 18
Knowledge, Motivation, and Organizational Influencers 20
Conceptual Framework 30
Qualitative Data Collection and Instrumentation 34
Participating Stakeholders 38
Findings 39
Solutions and Recommendations 61
Limitations 77
Conclusion 77
References 79
Appendices
Appendix A: Participating Stakeholders with Sampling Criteria for Interview 90
Appendix B: Interview Protocol 92
GENDER GAP IN COMPUTER SCIENCE 6
Appendix C: Credibility and Trustworthiness 94
Appendix D: Validity and Reliability 96
Appendix E: Ethics 97
Appendix F: Implementation and Evaluation Plan 99

GENDER GAP IN COMPUTER SCIENCE 7
LIST OF TABLES

Table 1. Assumed Influencers on Performance at TC 19
Table 2. Summary of Knowledge Influencers and Recommendations 64
Table 3. Summary of Motivational Influencers and Recommendations 68
Table 4. Summary of Organization Influencers and Recommendations 72
Table F1. Outcomes, Metrics, and Methods for External and Internal Outcomes 100
Table F2. Critical Behaviors, Metrics, Methods, and Timing for Evaluation 102
Table F3. Required Drivers to Support Critical Behaviors 104
Table F4. Evaluation of Components of Learning for Program 108
Table F5. Components to Measure Reactions to Program 110

GENDER GAP IN COMPUTER SCIENCE 8
LIST OF FIGURES

Figure 1. Interactive Conceptual Framework 32
Figure F1. Diversity Training Assessment Survey 112
Figure F2. Diversity Training Evaluation Survey 113
Figure F3. Diversity Initiatives Survey 114
Figure F4. Diversity Scorecard at TC 117

GENDER GAP IN COMPUTER SCIENCE 9
Introduction of the Problem of Practice
The gender gap in the field of computer science (CS, computing) continues to distress the
industry that has gained a significant economic and social prominence in the U.S. over the last 30
years (Shapiro, 2014; Stahl, Timmermans, & Mittelstadt, 2016; Vitores & Gil-Juárez, 2016). An
increased demand for web developers, cloud computing experts, and information security
specialists has resulted in CS becoming one of the top ranking professions in the U.S. (Prey &
Weaver, 2013; Zweben & Bizot, 2014). Since 1985 the number of students pursuing CS degrees
has increased from 44,000 to over 55,000 in 2013 (NCES, 2015). During this period the
percentage of men selecting CS major has grown by more 75%, yet the percentage of women has
decreased by 30% (Gries, 1987; NCWIT, 2014; Zweben & Bizot, 2014). The resulting gender
gap in computing has led the higher education community to adjust college admission policies
and improve instructional practices in order to increase the number of female students pursuing
degrees in computer science (Stoilescu & McDougall, 2011). Despite their best efforts,
however, improving gender diversity in college computing programs continues to be an elusive
goal (Dasgupta & Stout, 2014; Vitores & Gil-Juárez, 2016). For the last decade, the average
number of female students entering U.S. computing programs has remained constant, at around
18% (Broad & McGee, 2014; NCWIT, 2014; Zweben & Bizot, 2014). The inability to address
the gender gap in computer science not only caps the economic opportunity for women at a time
of an increased demand for this skill set, but also limits the creative and innovative contribution
this group may make to the technology sector of the U.S. economy (Vitores & Gil-Juárez, 2016).
Importance of Addressing the Problem
Elimination of the gender gap in computer science would address existing social
inequality concerns and would provide economic benefits that extend well beyond the field of
GENDER GAP IN COMPUTER SCIENCE 10
computing (Prey & Weaver, 2013). Current social constructs portray the field of computer
science to be a male professional domain (Cheryan, Plaut, Handron, & Hudson, 2013). The
dominant role of men in computing is further legitimized by their hierarchical position and
control of societal capital (Charles, 2011; Stanton-Salazar, 1997; Varma, 2010). According to
Cohen (2013), any form of dominance associated with occupational status is not only unfair, but
is also a form of inequality that must be contested.
Inequitable representation of women in computing slows down the country’s economic
growth by preventing a large percentage of the population from entering high-in-demand
opportunities. According to the Department of Labor, in 2015 the median annual wage for
computer science skills was $81,000, significantly higher than the median annual wage for all
occupations at $36,000. Yet, CS graduates from U.S. programs can only fill about 55% of the
available computing opportunities (Prey & Weaver, 2013). Consequently, some social scientists
argue that the gender gap in high-in-demand, well compensated professions, such as computer
science, contributes to the labor shortage and restricts the country’s economic growth
(DiTomaso, Post, & Parks-Yancy, 2007; Weeden, 2002).
A lack of gender diversity has a direct impact on the computing industry’s ability to
innovate (Cohen, 2013). Innovation is a product of social interactions (Sawyer, 2012).
According to Horwitz (2005), a more diverse group utilizes its distinct cognitive abilities to
develop inventive solutions which are so critical in the computing industry. Other studies have
also shown that a more diverse team improves the organizational performance where new and
complex problems need to be addressed on a regular basis (Prieto, Phipps, & Osiri, 2009).
Subsequently, addressing the gender gap issue in computer science is of paramount importance
to the computing industry where the ability to innovate is a core competency.
GENDER GAP IN COMPUTER SCIENCE 11
Purpose of the Project and Questions
The purpose of this project is to study Technical College’s (TC) success in attracting female
applicants to its computer science program in relation to a larger problem of practice, a
disproportionately low rate of women being admitted to the computer science programs in U.S
colleges. The assets in the areas of knowledge and skill, motivation, and organizational
resources that enable TC to admit the above national average percentage of women into the
computer program without modifying curriculum and compromising academic rigor is being
examined. While a complete study would focus on all stakeholders, for practical purposes the
stakeholder group to be focused on in this analysis is the computing program leadership team at
TC. As such, the following two questions will guide this study:
1. What are the TC computing program leadership team’s knowledge and motivation in
relation to reducing the gender gap in computing?
2. How does the TC computing program leadership team’s knowledge and motivation
interact with the school’s context to reduce gender gap in computing?
Organizational Context and Mission
Technical College has one of the top ranked computing programs in the U.S. The school
awards computer science degrees in machine learning, robotics, software research, and several
others computing disciplines. Leading the school’s computing program is an internationally
recognized team of computer science faculty and administrators. Several current and past
members of the faculty have earned national and international prominence for their research
publications and scholarly awards.
TC is located in a metropolitan area, a home to several other higher education
institutions. Several professional sports teams, a number of museums and arts exhibits are
GENDER GAP IN COMPUTER SCIENCE 12
located within a twenty mile radius and are easily accessible. Location of the school, a well-
known faculty, and modern school facilities attract students not only from the U.S., but also from
a number of foreign countries. Almost 20% of the TC’s undergraduate population are
international students. The computer science program acceptance rate is typically between five
and seven percent, making it one of the most competitive CS programs in the country. The
mission of the school has not changed since its inception. For almost 30 years the school’s focus
has been to provide superior educational experience to its students and to attract the best
computer science faculty and staff to its program. The school has also identified student
diversity to be a top organizational priority and a critical contributor to the school’s future
success. For over a decade the percentage of female candidates being admitted to TC’s computer
science program has exceeded the national average by 100%. In the fall of 2017 almost 50% of
entering computer science students were women.
Organizational Performance Goal

During the early years of operation CS program leadership at TC has acknowledged the
fact that the goal of achieving a more diverse and inclusive culture, one of computing program’s
primary objectives, has not been met. Only a handful of female applicants were admitted to the
CS program each year. Having such a low admission rate of female candidates was unacceptable
to the school’s leaders. To achieve congruency with the program’s diversity objectives, the CS
administration has partnered with the National Science Foundation and several major corporate
stakeholders. A working group that included members of the dean’s office as well as experts in
gender equity in education was tasked to develop a set of recommendations that would lead to a
more gender diversity focused set of admission policies and program practices. The school set a
goal of annually exceeding the national average of women applying and being admitted to the
GENDER GAP IN COMPUTER SCIENCE 13
computer science program at TC while keeping the computing curriculum unchanged and
maintaining the program’s high academic standards.
By the mid-2000 the school administration had experienced a significant improvement in
their ability to attract talented female candidates into the program. For the first time the school
was able to consistently exceed the national average percentage rate of women being admitted to
computer science programs. Over the years the school leadership has experienced a number of
personnel changes, yet the administration’s success in reducing the gender gap in CS program
and still maintaining program’s national ranking has not been affected.
Description of Stakeholder Groups and Participating Stakeholders
All organizations, whether for-profit, nonprofit, government, or educational, have
stakeholders (Lewis, 2011). A stakeholder group is a group of individuals who directly
contribute to and benefit from the achievement of the organization’s performance goal (Wheeler
& Sillanpa'a, 1998). Organizations stratify stakeholders based on their ability to influence the
direction and timing of an organizational change (Lewis, 2011). As a result, some stakeholder
groups are recognized to be more important than others. At TC, the primary group of
stakeholders includes the CS program leadership team, CS program faculty, and CS students.
The higher education institution leadership teams are responsible for all aspects of the
administrative, academic, and social environment in the school (Kezar, 2001). At TC, the CS
program leadership team is responsible for designing and implementing the academic
curriculum, hiring the research and teaching faculty, and ensuring that organizational goals, such
as gender diversity, are achieved. The faculty’s primary goal is to deliver quality education and
to ensure their students learn (Tagg, 2012). At TC, the faculty focuses on both teaching and
research activities. Most faculty is expected to conduct and publish research, enhancing their
GENDER GAP IN COMPUTER SCIENCE 14
own and the school’s reputation amongst the peer institutions and corporate sponsors. The
undergraduate students are an important contributor to the successful reduction of the gender gap
in the program (Varma, 2010). Their active engagement and support for a more inclusive school
setting influences the academic and the social environment at the school. Student’s views and
feedback are also crucial in assessing success and shortcomings of the diversity focused
initiatives at TC. While the joint efforts of all stakeholder groups contribute to the school’s goal
of attaining the above national average percentage of women entering the computer science
program, it is important to focus on the CS program leadership team who is directly responsible
for the admission policies and organizational practices at TC. For the purposes of this study, the
TC leadership team comprised of the ranking members of the CS program administration and the
CS department chairs is the stakeholder group of focus.
Literature Review Sections Overview
This literature review examines factors that contribute to the gender gap in the U.S.
computing programs and discusses knowledge, motivation, and organizational (KMO)
influencers that facilitate a more gender balanced environment at undergraduate computing
programs. The review will begin with an overview of the computer science discipline and a
summary of the scholarly literature covering the topics related to the causes of the gender gap in
computer science. It will be followed with a discussion of a Gap Analysis Framework by Clarke
and Estes (2008), an approach that will be used to examine KMO factors leading to TC’s ability
to significantly reduce the gender gap in the computer science program. Next, based on the
literature review and the relevant theoretical models, the knowledge, motivation, and
organizational influencers will be individually discussed. Finally, the KMO influencers, along
GENDER GAP IN COMPUTER SCIENCE 15
with the research questions posed earlier, will be used to advance the Interactive Conceptual
Framework (Conceptual Framework) as a guiding mechanism for conducting this inquiry.
Literature Review
The computer science discipline is a body of knowledge that refers to the algorithmic
modeling and analysis of data underlying natural and human made processes (Denning, 2003). It
also includes the study of computer systems, communication technologies, and software design
and programming. During the early days of the modern computer era, the primary focus of
computer science was to develop mathematical models using digital computer platforms to solve
calculation intensive problems (Ceruzzi, 2012). Based on the theoretical foundation, related
abstractions, and corresponding designs, the field of computer science now includes more than a
dozen of subareas, such as programming languages, artificial intelligence and robotics, and
database and information retrieval, among others (Denning, 2003). In addition, computer
science is used as a tool to create and manage context specific models and processes in other
fields (Denning, 2003). Some researchers argue that other disciplines in the sciences and the
humanities are able to pursue new areas of research as a result of employing the ‘computational
thinking’ (Bundy, 2007). Computational thinking is an analytical tool, based on the fundamental
concepts of computer science, that allows researchers to analyze complex problems and create
multiple levels of abstraction used to solve the underlying problems (Wing, 2006). It
conceptualizes the way people think, learn, and interact. For example, models of cognition in
psychology, economic forecasting in business, and patient diagnosis in clinical medicine employ
techniques originally developed by computer scientists (Bundy, 2007; Denning, 2003).
In the last 30 years a topic of gender gap in computer science has been widely studied by
the social researchers. In their seminal work Cohoon and Aspray (2006) have reviewed an
GENDER GAP IN COMPUTER SCIENCE 16
existing body of research examining the underlying causes leading to the gender gap in computer
science. Several broad themes requiring further analysis have been suggested by the authors. A
first major theme discussed highlights a historical influence by the community of computing
experts from 1940’s on the present day male-dominated culture of college computer science
programs and the resulting gender gap. Modern field of computing has its roots in World War II
ballistic computation programs (Ceruzzi, 2012). Viewed as a war-time effort, most personnel
recruited to oversee a digital computation program, a precursor to the modern computer science
programs, were military men (Ceruzzi, 2012). In 1950’s and 1960’s, driven by the Cold War and
the Space Race related efforts, most university computer science programs were established and
led by the same group of men, portraying computing skills as masculine activity (Ceruzzi, 2012).
The resulting culture and image of computer science has become closely aligned to the
masculine culture deterring many qualified women from entering the field of computer science
and therefore contributing to the gender gap in CS college programs (Pechtelidis, Kosma, &
Chronaki, 2015). Ironically, the contribution by the female computer programmers and
mathematicians, who were employed to manually perform ballistic calculations for the War
Department during the World War II has been all but forgotten (Gurer, 2002; Light, 1999).
As a second major theme, the socio-cultural influences were also identified as negatively
impacting female college applicants considering a major in computer science (Stoilescu &
McDougall, 2011). From the early age young women are influenced by the social structures that
dictate which professional fields are more suitable for men and which ones are more appropriate
for women (Broad & McGee, 2014). The resulting gender-based professional divide steers many
highly qualified women away from considering the field of computer science towards a more
‘traditional’ female dominated professions such as nursing and teaching (Dasgupta & Stout,
GENDER GAP IN COMPUTER SCIENCE 17
2014). Depiction of computer professionals as technology-centric nerds by the popular media in
the U.S. was also singled out as a significant deterrent for many college-bound female students
considering a CS major (Cheryan et al., 2013). Research focusing on the socio-cultural factors
contributing to the gender gap in computer science suggests that changing cultural beliefs about
the field of computing is a prerequisite for achieving a more gender-balanced academic
environment (Varma, 2010).
A third major theme leading to the gender gap in computer science focused on the innate
differences between men and women (Trauth, 2002). According to these studies, as a result of
the physiological differences men and women possess different computer science abilities and
skills (Venkatesh, Morris, & Ackerman, 2000; Vitores & Gil-Juárez, 2016).). In order to reduce
the gender gap in computer science these studies recommend changing the computing program
curricula, a critical step in attracting and retaining more female students in the CS programs
(Trauth, 2002). These studies argue that changing the CS program curricula and creating an
alternative CS pathways for female students would enhance their computing skills and abilities
making the computing field more attractive to them (Cohoon & Aspray, 2006).
Although the scholarly literature focusing on the gender gap in computer science reviews
a broad spectrum of the underlying causes and recommends corresponding interventions, very
few studies examine the role of school administrators in achieving a more gender-balanced
environment in college computer science programs. Not surprisingly some social researchers
highlight an urgent need to examine the role of college administrators in being able to effectively
address current academic and social issues facing the institutions of higher education (Bray,
2010). As a key stakeholder group focused on improving organizational performance, the
college program administrators are in the position to develop, implement, and evaluate programs
GENDER GAP IN COMPUTER SCIENCE 18
that facilitate diversity focused initiatives, such as a reduction of the gender gap in computing.
Examining the knowledge and motivation of computer science program leadership as well as the
organizational factors with respect to TC’s ability to successfully reduce the gender gap in CS
program is the primary objective of this study.
The Clark and Estes Gap Analysis Conceptual Framework
Improving organizational performance in a practical and cost-effective manner is a
priority for all organizations (Clark & Estes, 2008). Performance research has provided ample
evidence that a systemic, researched-based approach to addressing performance issues can
deliver a timely and effective set of solutions at the same time avoiding the wastefulness of
human and capital resources (Clark & Estes, 2008). The Gap Analysis Framework by Clark and
Estes (2008) is a researched-based and practically validated approach that addresses two
common problems related to the organizational performance. First, this framework helps to set
and measure organizational goals that align with desired objectives. For organizations, whether
looking to improve their current performance or to maintain their already high level of
performance, it is mission critical to establish and support well-defined and time-bound set of
goals (Clark & Estes, 2008). Second, this framework helps to analyze the causes of the gaps
between current and espoused performance or to assist in identifying factors that facilitate
superior performance for the market leaders. Clark and Estes (2008) have identified three
primary factors: knowledge, motivation, and organizational barriers or facilitators that influence
performance gaps in the organizations. Under the Gap Analysis Framework, each factor is
examined to identify the root cause of the problem or the source of superior performance. In the
following sections the Gap Analysis Framework will be used to examine the knowledge,
motivation, and organizational influencers, shown in Table 1, that have contributed to the best
GENDER GAP IN COMPUTER SCIENCE 19
practices implemented by TC in attracting and enrolling female candidates to one of the more
academically challenging computing programs in the country. In this context the role of the TC
leadership, their knowledge, skills, and motivation, as well as their ability to implement
computing program changes to significantly reduce the gender gap will be examined through the
qualitative case study approach.
Table 1
Assumed Influencers on Performance at TC
Knowledge Motivation Organization
Conceptual Knowledge –
CS administrators need to
know social and cultural
factors influencing
women’s decision to avoid
pursuing college degrees in
computing.

Expectancy Value – CS
administrators pursue
utility value in reducing
gender gap in computing
program.

Cultural Influencers – CS
administrators need to
balance academic and
vocational influencers to
reduce gender gap in
computing programs.
Procedural Knowledge –
CS administrators need to
know how to enhance
school’s academic and
social environment in order
to attract and retain more
women with little or no
computing background.

Self-efficacy – CS
administrators believe in
their ability (driven by the
successful past
experiences) to facilitate
organizational changes that
reduce gender gap in
computing program.

Cultural Settings – CS
administrators need to
modify computing program
cultural settings by
sponsoring equity-focused
initiatives.
Metacognitive Knowledge
– CS administrators need to
be aware of their own
attitudes and biases towards
women pursuing degrees in
computing.

GENDER GAP IN COMPUTER SCIENCE 20
Knowledge, Motivation, and Organizational Influencers
Knowledge and Skills
Knowledge and skills are at the core of the innovative strategies organizations employ to
compete in the global economy (Dyer, Gregersen, & Christensen, 2011). Several studies have
highlighted a strong positive relationship between an organization’s ability to create and manage
knowledge and the corresponding improvement in the organizational performance (Clark &
Estes, 2008; Lee & Choi, 2003). According to cognitive psychologists, knowledge is created by
the individual’s cognitive system while observing and interpreting the environment (Mayer,
2011). In this theoretical framework the four types of knowledge are (a) factual knowledge, (b)
conceptual knowledge, (c) procedural knowledge, and (d) metacognitive knowledge (Krathwohl,
2002; Rueda, 2011). The factual knowledge refers to the specific facts, dates, and data needed to
understand and analyze a specific situation or content (Krathwohl, 2002; Rueda, 2011).
Knowing the specifics, such as a percentage of women graduating with CS degrees in U.S.
colleges, is an example of the factual knowledge that could be useful to the CS administrators.
Conceptual knowledge describes the models, theories, and ideologies associated with an area of
inquiry (Krathwohl, 2002; Rueda, 2011). Analysis of the social and cultural factors that preclude
college women from selecting CS as their major is an example of the conceptual knowledge.
Procedural knowledge, commonly referred to as the ‘how-to’ knowledge, identifies steps needed
to achieve a specific goal (Krathwohl, 2002; Rueda, 2011). A list of specific steps needed to
improve a graduation rate of women with CS degrees is a type of procedural knowledge.
Metacognitive knowledge is a knowledge of one’s own way of thinking and learning, as well as
the awareness of one’s own attitudes (Krathwohl, 2002; Rueda, 2011). For example, a
professor’s comprehension of his own stereotypical thoughts on whether women are able to
GENDER GAP IN COMPUTER SCIENCE 21
succeed in the field of computing is a type of metacognitive knowledge. Review of the academic
literature on the pervasiveness of the CS gender gap and its negative social and economic impact
on the society has identified a number of knowledge influencers. The following three knowledge
influencers: (a) social and cultural factors, (b) academic and school environment, and (c) school
administrators’ attitudes and biases, however, are at the forefront of the CS gender gap
discussions and are reviewed next.
Social and Cultural Factors. Educational leaders rely on their conceptual frameworks
to construct knowledge and resolve conflicts (Heck & Hallinger, 2005). Consequently,
computing administrators’ ability to reduce the gender gap in the field of computer science
depends on their conceptual understanding of the social and cultural factors that influence
college women’s decision not to pursue degrees in CS (Charles & Bradley, 2006). To reduce
the gender gap in computing programs, CS administrators need to be aware of such social and
cultural factors as stereotypes and social biases, among others (Charles, 2011; Marini, 1990).
Many of these factors are based on the beliefs that men and women have different neurological
and cognitive abilities (Trauth, 2002). Despite studies showing that a human brain is an
assortment of characteristics that vary between individuals, not between the genders, there is still
a perception that men are more competent in computer science than women (Charles & Bradley,
2006). The resulting computing stereotype views CS professionals as males solely interested in
computer related activities, e.g., writing software, playing video games, etc., with no interest in
the social benefits of their efforts (Cheryan et al., 2013). Not surprisingly the masculine
characteristic of the computing field is less appealing to female college students and may prevent
them from majoring in CS (Eckel & Grossman, 2005). As most administrators of the top U.S.
computer science programs are male, they may lack a conceptual understanding of social and
GENDER GAP IN COMPUTER SCIENCE 22
cultural barriers impeding women’s entry into the computing field and may also not have any
transferrable knowledge or procedural expertise essential for closing the gender gap in
computing (Mayer, 2011).
Academic and social environment in computing programs. Lacking a procedural
framework needed to identify and address performance gaps in education may lead to ineffective
interventions and further compound the underlying problem (Rueda, 2011). Addressing the
gender gap in computing creates a unique set of procedural challenges for CS administrators
(Stoilescu & McDougall, 2011). This is evident in the case of establishing procedures related to
the introduction of programming to women with no or little previous experience (Frieze &
Quesenberry, 2015; Stoilescu & McDougall, 2011). Based on the assumptions that women
typically display lower computer aptitudes and higher levels of computer anxiety compared to
men, one procedural approach used by some CS administrators was to develop a set of
introductory classes especially for women (Venkatesh et al., 2000). Creating a specialized
computing track, the so called ‘pink curriculum’, was an attempt to compensate these students
for their poor performance in regular computer programming classes (Frieze & Quesenberry,
2015). After two decades, a gender focused effort by CS administrators to implement procedures
improving the retention rate of women in the CS programs leveraging this approach has
remained largely unsuccessful (Frieze, Quesenberry, Kemp, & Velazquez, 2012). In fact, some
researchers made an argument that explicitly identifying gender differences and creating separate
tracks for men and women further enhanced the CS stereotypes discussed earlier and contributed
to the widening of the gender gap in computing (Frieze at al., 2012). As a result, a list of
procedural enhancements to the academic and social environment in CS programs needed to
GENDER GAP IN COMPUTER SCIENCE 23
attract and retain more women with no programming experience remains a challenge for many
U.S. colleges.
CS administrator’s attitudes and biases. Every person carries biases shaped by one’s
cultural environment (Banaji & Greenwald, 2013). Most people are unaware of these ‘hidden’
biases which can influence one’s behavior and actions (Banaji & Greenwald, 2013). Dasgupta
and Stout (2014) support this assessment by referring to studies where gender bias in academic
hiring was demonstrated by both men and women with neither group being aware of their bias.
As the number of male CS administrators across U.S. colleges outnumbers their female
counterparts, it is not unreasonable to see how more male students would be admitted to
computer science programs, further exacerbating the issue of gender inequality in CS. Nkomo
and Al Ariss (2014) have reviewed the phenomenon of homosocial reproduction. Their analysis
has shown that managers and administrators select candidates similar to their own sociocultural
identity. Knowing the effects associated with homosocial reproduction allows a decision making
group, whose membership is fairly homogeneous, to evaluate and compensate for their hidden
biases. Gaining awareness of one’s own attitudes and biases is an important step in ensuring that
one’s impartial views and equitable actions are maintained (Baker, 2006; Banaji & Greenwald,
2013). This higher level cognitive process is referred to by social scientists as metacognition
(Baker, 2006). Metacognitive processes allow individuals to evaluate their strengths and
weaknesses and, as in the case of CS school administrators, to assist with developing strategies
to compensate for any biased-driven decisions related to the program admissions and curriculum.
Motivational Influencers
Motivation is an internal process that commences and retains a goal-oriented activity
(Mayer, 2011; Rueda, 2011). As the definition implies, motivation is what drives individuals to
GENDER GAP IN COMPUTER SCIENCE 24
successfully perform a task. Three critical motivational factors influence one’s ability to
complete the task: (a) one must choose to work on the task, (b) one must persist and stay with the
task to completion, and (c) one must invest mental effort to achieve a desired level of
performance on the task (Clark & Estes, 2008). The impact of motivation on the performance
has been a topic of research for several decades (Bolman & Deal, 2013). In the early 1950’s a
simple formula, Performance = Ability x Motivation, was widely used to explain the relationship
between performance and motivation (Bolman & Deal, 2013). Clark and Estes (2008) have
emphasized the role of motivation as one of the ‘big three’ factors responsible for the
performance gap discussed earlier. Over the years research by social scientists has provided
further insight into the influence of motivation on academic performance with new models
examining motivational principles in educational environments (Rueda, 2011).
The following sections review two motivational models that influence the organizational
objective of improving diversity and reducing the gender gap in college computer science
programs. The expectancy value theory and the self-efficacy theory, as well as their
motivational influencers on the CS administration are detailed below.
Expectancy value motivational theory. The expectancy value motivational theory
(EVT) provides a framework for understanding the motivational drives for choosing and
performing a task (Eccles, 2006; Rueda, 2011). The EVT states that in order for the individual to
start the task, she needs to see the value in performing and completing the task (Eccles, 2006;
Rueda, 2011). Eccles (2006) has identified four types of perceived values that motivate an
individual to perform the task. They are: (a) intrinsic value, (b) attainment value, (c) utility
value, and (d) cost value (Eccles, 2006). The intrinsic value construct refers to the enjoyment
one expects to receive while performing the task. The attainment value construct refers to the
GENDER GAP IN COMPUTER SCIENCE 25
significance of achievement one perceives in performing the task. The utility value is derived
when one sees a task contributing to her future personal or professional goals. The cost value
refers to the cost of effort that is necessary to perform a task including the opportunity cost of not
performing other tasks. The four types of value play a vital role in the academic environment
(Eccles, 2006). For the CS administrators, who are focusing on improving student diversity and
reducing the gender gap in computing programs, in order to succeed they need to pursue a utility
value in their efforts.
CS administrators’ expectations and value. College administrators are responsible for
ensuring the high academic performance of the students and for meeting accountability
requirements of the stakeholders (Kezar, 2001). To succeed in their role as stewards of the
program, computing science administrators need to recognize the value in being able to achieve
both objectives. Research has shown that increased student diversity in CS programs improves
academic performance and attracts more female students to computing (Frieze at al., 2012).
Hoogendoorn, Oosterbeek, & van Praag (2013) also highlight positive effects that a more diverse
group of college students has on the overall organizational and academic performance. By
focusing on student diversity and on reducing the gender gap in computing, CS administrators
can improve the academic performance of students and while pursuing their professional goals.
Meeting accountability requirements is a top priorities for college programs. For CS
administrators this is especially important as many computing programs are funded by state
sponsored initiatives which mandate a diversity of students. A lack of focus on diversity or the
inability to reduce the gender gap can lead to a loss of funds at the time when the sources of
funding are becoming scarce. As a result, the CS administrators recognize utility value in
seeking to attract more female students and reduce the gender gap in computing programs.
GENDER GAP IN COMPUTER SCIENCE 26
Self-efficacy Theory. Self-efficacy theory describes motivational dynamics that
reinforce beliefs in own competence (Bandura, 1993). Strong efficacy beliefs increase one’s
self-confidence resulting in improved outcomes (Pajares, 2006; Rueda, 2011). Low efficacy
beliefs, on the other hand, may prevent an individual from engaging in tasks or reduce the level
of effort needed to successfully complete a task. According to self-efficacy theory there are four
primary factors that either facilitate or inhibit the formation of self-efficacy beliefs (Pajares,
2006). First, the self-efficacy beliefs are shaped by successes or failures associated with
mastering a given task (Bandura, 1993, Rueda, 2011). Achieving a successful outcome increases
self-efficacy, a failure to reach a goal lowers it (Pajares, 2006). Second, viewing success or
failure of others affects the formation of self-efficacy beliefs (Pajares, 2006). Research shows a
strong influence of vicarious experiences in educational setting (Rueda, 2011; Zimmerman,
2000). Third, social influences exerted by others can either positively or negatively influence the
self-efficacy of an individual (Pajares, 2006). Several studies have found that CS faculty in
many U.S. colleges expect male students to perform better in computer science classes than
female students (Varma, 2010). As a result, female CS students may feel less confident in their
own computing abilities. Finally, the physiological condition of an individual can increase or
decrease the self-efficacy (Pajares, 2006). According to studies, attending computer science
classes where faculty, teacher assistants, and a significant majority of students are male, can lead
to high levels of anxiety in female students and decrease their self-efficacy (Varma, 2010).
CS administrator’s self-efficacy. Self-efficacy beliefs determine the choice of actions
undertaken by the individuals (Bandura, 1993; Pajares, 2006). As such, self-efficacy beliefs of
CS program administrators play a role in their choice of priorities and allocation of resources
needed to reduce the gender gap. One individual factor influencing leaders’ efficacy for
GENDER GAP IN COMPUTER SCIENCE 27
implementing organizational changes have been frequently cited in the literature (Paglis &
Green, 2002). Self-confidence is a concept that allows individuals to be certain about their
abilities to succeed (Northouse, 2016). A proven record of overcoming difficult tasks increases
leaders’ level of self-confidence and therefore increases their self-efficacy in dealing with
challenges (Paglis & Green, 2002). Having a CS leadership team with a track record of
supporting and succeeding in diversity focused initiatives is likely to increase their efficacy in
reducing the gender gap in the computer science program.
Organizational Influencers
Organizational influencers play an essential role in causing organizations to
underperform or in facilitating organizational changes that lead to performance improvements
(Clark & Estes, 2008). The level of employee engagement, communication strategy, policies
and procedures, and executive leadership are just some of the organizational influencers that
either support or inhibit institutional performance (Berger, 2014; Burke, 2005; Miller & Lee,
2014). Organizational culture, however, affects all of these influencers and ultimately
determines whether organizational change approaches focused on improving performance are
effective and sustainable (Clark & Estes, 2008). Given a wide variety of the industries, the types
of firms, and the organizational structures that exist, it is not surprising to find many categories
of the organizational change models (Kezar, 2001). Each category employs a different set of
assumptions to explain how and why organizations change (Kezar, 2001). Despite their different
approaches, however, all models share the same objective of developing a framework for
improvement that includes the rationale for change, the process of change, and the evaluation of
the effectiveness of change (Kezar, 2001; Langley et al., 2009; McEwan & McEwan, 2003).
GENDER GAP IN COMPUTER SCIENCE 28
Organizational models of change in higher education. The cultural models and
settings uniquely define and affect each organization (Clark & Estes, 2008; Schein 2010).
Understanding a cultural identity is an important step in developing a road map for improving
organizational performance (Bolman & Deal, 2013). Identifying the cultural profile of higher
education institutions with complex decision making structures, different institutional histories
and goals, and many competing priorities is an especially challenging endeavor (Kezar, 2001).
Failure to accurately assess the organization’s cultural models and settings may lead to
ineffective performance-enhancing strategies and may cause further deterioration in the
institution’s performance (Clark & Estes, 2008; Eckel, 2001). Adapting the appropriate change
approach for improving performance or maintaining a market leading performance for a higher
education institution warrants a special attention (Kezar, 2001). The type of institution and its
cultural profile determines which model of change is more suitable for achieving the
organizational goals (Kezar, 2001). Review of recent research has identified the cultural models
of change as some of the more effective approaches to improving organizational performance in
higher education institutions (Kezar, 2001). The cultural models of change emphasize the
symbolic characteristic of higher education institutions. These models purport that the
organizational change ensues only when the cultural models and settings of the institution are
altered (Schein, 2010). Understanding the culture of a higher education institution along with its
organizational symbols, history, and rituals are requisite to enhancing and facilitating the change
process (Bolman & Deal, 2013; Kezar, 2001). The multitude of organizational goals that various
academic disciplines within higher education pursue highlights a need to understand the cultural
values and beliefs of each one individually (Kezar, 2001).
GENDER GAP IN COMPUTER SCIENCE 29
Cultural influences in computing programs. It is not unusual for a higher education
institution to inherit cultural influences from more than one group with a joint set of influences
emerging and ultimately defining the culture of that organization (Kezar, 2001). College
computer science programs have characteristics of both the academic culture of the higher
education institutions as well as the computing culture closely aligned with strong professional
traditions. The academic culture values the history and traditions of the higher education
institutions (Kezar, 2001). Academic accountability, school’s independence, and safeguarding
educational and social ideals are some of the characteristics of the academic culture (Burke,
2005; Ek, Ideland, Jonsson, & Malmberg, 2013; Kezar, 2001). The measure of school’s
successful performance includes students achieving mastery in the academic disciplines as well
as the presence of the diverse student body (Ek et al., 2013). The computing culture, on the other
hand, is professionally oriented and inherits cultural characteristics from the vocational side of
computing (Ek et al., 2013). Social researchers have observed a strong parallel between
computing professional culture and the masculine culture, a link sometimes used to explain the
presumed aversion of women towards the field of computing (Cheryan et al., 2013; Cohoon &
Aspray, 2006). Studies have also shown that the computing professional culture tends to
emphasize the technical side of the field, such as programming and computer architecture, rather
than focusing on the social priorities, such as diversity and inclusion (Cheryan et al., 2013;
Cohoon & Aspray, 2006; Morelock, 2017). The dual influence of the academic and computing
cultures on the cultural profile of the computing undergraduate programs guides the actions of
the program’s leadership.
Computing program culture and leadership. Changing values and beliefs ingrained in
the organizational culture is a long-term effort that depends on the leader’s ability to alter the
GENDER GAP IN COMPUTER SCIENCE 30
prevailing cultural models and settings (Kezar, 2001; Schein, 2010). In undergraduate
computing programs, the program administrators oversee the academic and the cultural
environment of the school ensuring that the organizational goals, such as academic excellence,
diversity, and inclusion are being attained (Broad & McGee, 2014; Cohoon & Aspray, 2006).
As organizational leaders, they are responsible for implementing change processes that modify
the institutional culture to ensure the program objectives are being pursued. Consequently, to
successfully implement equity-focused actions, such as incorporating diversity benchmarking in
performance measuring framework, instituting more inclusive admission policies, and securing
necessary resources in support of new initiatives, CS administrators need to alter and sustain the
symbolic aspect of their program’s cultural settings.
Conceptual Framework
The Role of the Conceptual Framework
Qualitative studies are characterized by their attention to the philosophical, scientific,
historical, and social perspectives allowing the authors to pursue a unique form of inquiry
(Merriam & Tisdell, 2016). Yet, despite a limitless way of structuring qualitative studies, there
is a need to organize and synthesize empirical, theoretical, and experiential knowledge into
logically arranged configuration that guides each study from its stated goal to the selection and
implementation of the research design approach (Maxwell, 2013). The conceptual framework
provides that foundational structure. It helps to narrow the focus of the study and connects the
concepts, theories, and ideas to the problem being studied (Rocco & Plakhotnik, 2009). Without
a conceptual framework underlying a qualitative study, there is a strong possibility of the study
design, research questions, and the supporting literature being misaligned negating the scholarly
value of the inquiry (Merriam & Tisdell, 2016). Maxwell (2013) identifies four factors that
GENDER GAP IN COMPUTER SCIENCE 31
contribute to the development of the conceptual framework: experiential knowledge, prior theory
and research, exploratory studies, and thought experiments. According to Maxwell (2013),
experiential knowledge plays an important role in helping the researchers to develop the vision
and understanding of the problem being examined. To avoid the unduly influence of any
personal experience on the study, researchers are advised to employ a concept of critical
subjectivity, specifically developed to avoid a bias risk by leveraging individual experiences as
only one of several contributing sources (Maxwell, 2013, Reason, 1994). Using prior research is
also an important contributor to the development of the conceptual framework as it provides the
background for the problem being studied, introduces theoretical concepts utilized in the earlier
research, and assists the study authors in determining whether previous work has shortcomings
and warrants another look (Maxwell, 2013; Merriam & Tisdell, 2016; Rocco & Plakhotnik,
2009). Pilot studies are sometimes conducted to further validate researchers’ views and
understanding of the problem being examined (Creswell, 2014; Maxwell, 2013). Finally, the
thought experiments, which are popular in physical sciences, are also used by the social scientists
to enumerate possible explanations for the phenomenon being studied and contribute to the
development of the conceptual framework (Maxwell, 2013). The conceptual framework is
uniquely constructed for each research study to encompass key concepts ascertained from the
literature review and to capture the interaction between the knowledge and motivation of the
stakeholder group and the organizational context. The conceptual framework for studying the
best practices in attracting and enrolling female candidates at TC is presented below.
Interactive Conceptual Framework
To explain the best practices implemented by the TC administrators to attract female
candidates to the computing program the following interactive conceptual framework, Figure 1,
GENDER GAP IN COMPUTER SCIENCE 32
is being advanced. It draws from the literature on the institutional culture, the social identity
formation, and organizational models of change in higher education institutions that were
previously discussed. Also, more than 30 years of experience in the computing industry by the
study author as a software developer, program manager, and the industry advisor is being utilized
in developing the conceptual framework.
Figure 1. Interactive Conceptual Framework

Figure 1. Conceptual framework depicts relationship between computing program culture and
administrators’ beliefs, behaviors, and actions resulting in an increased enrollment rate of female
students at TC.
Conceptual Framework Components
The earlier sections of this literature review focused on the individual theories, concepts,
and research. The conceptual framework leverages them as foundational components used to
establish a relationship between the computing program culture at TC and CS program
administrators’ skills and beliefs that guide their actions towards increasing number of women
being admitted to the program. The focal point of the framework is the Venn diagram depicting
GENDER GAP IN COMPUTER SCIENCE 33
the overlap between the vocation-focused computing culture and the higher education academic
culture. As an undergraduate computing program at a large research university, TC culture
inherits cultural characteristics of both the computing culture and the academic culture. In
Figure 1, the intersection of these two cultures is shown as the shaded area and signifies the
resultant culture of the computing program at TC. Based on the institutional history, university’s
mission and vision, and the accountability pressures imposed by various stakeholder groups, the
cultural profile of TC may closely resemble either the computing culture or the academic culture,
a claim embedded in one of the guiding questions for this study. In addition to knowledge and
motivational influencers of TC leaders, the conceptual framework also highlights the importance
of choosing the appropriate organizational model of change needed to implement the diversity
focused agenda at TC. This is depicted as the downward pointing arrow at the bottom of Figure
1. Implementing diversity focused policies in the male-dominated environment of the computer
science programs could be a challenging effort. Research on social identity suggests that
homogeneous teams with respect to social groupings tend to share the same norms and values
and therefore resist any changes to their membership (Eckel & Grossman, 2005). To achieve a
more diverse student body without sacrificing the academic standards of the CS program, the
appropriate organizational change approach needs to be selected. The cultural models of change
have shown promise in facilitating organizational changes in higher education institutions
(Kezar, 2001). By challenging prevailing attitudes and addressing the symbolic characteristic of
the higher education institution, policies and procedures that facilitate gender diversity in the
program could be implemented (Kezar, 2001). The conceptual framework presented here brings
several theoretical studies and scholarly discussions together to form a coherent view on the
primary KMO influences affecting the computing program leadership at TC.
GENDER GAP IN COMPUTER SCIENCE 34
Qualitative Data Collection and Instrumentation
Data Collection
In qualitative studies, data is being collected in order to capture the description of the
phenomenon experienced by the study participants as well as the meaning they assign to these
experiences (Bogdan & Biklen, 2007; Maxwell, 2013; Merriam & Tisdell, 2016). Data
collection is therefore a critical component of the researcher’s ability to inductively develop a
theory about the phenomenon being studied (Creswell, 2014). There are several methods of data
collection, including direct observations, questionnaires, document analysis, and interviews
(Johnson & Christensen, 2014; Merriam & Tisdell, 2016). For the purposes of the TC study,
document analysis and interviewing was the primary method of data collection.
Document Analysis
Document analysis refers to the collection of existing data and materials created by others
that are useful for the purposes of the new research (Bogdan & Biklen, 2007; Merriam & Tisdell,
2016). Documents such as previous research, memos, and public records are just some of the
examples of the documents that could be included in qualitative studies (Bogdan & Biklen, 2007;
Johnson & Christensen, 2014; Merriam & Tisdell, 2016). For the purpose of this inquiry, several
scholarly articles published by the TC leaders that focused on the program’s efforts to reduce the
gender gap in CS were examined. Publically available information on TC’s website discussing
the on-going diversity focused initiatives at the school were also reviewed.
In addition to collecting information useful in answering research questions, there were
other benefits of performing a document analysis. First, review of the previously published
research provided a historical perspective on TC’s efforts to reduce the gender gap in CS and
assisted in preparing for the on-campus interviews with program leaders. Understanding the
GENDER GAP IN COMPUTER SCIENCE 35
organization’s culture from the existing documentation is a prerequisite for any effective
communication between the organizational leaders and the researcher (Schein, 2010). Second,
by examining existing documents and reviewing both successes and challenges related to
achieving a more gender-balanced environment at TC, the study author could learn more about
the decision-making practices of the CS program leaders. This knowledge assisted in examining
the knowledge and motivational influences guiding the actions of the CS program leaders.
Finally, reviewing available documentation facilitated researcher’s understanding of the
characteristics of the inclusive culture of the computer science program that resulted in a more
gender-balanced environment at TC.
Interviewing
Interviews can be conducted in a number of different ways, e.g., in person, via the phone,
or on-line. In this study the two initial interviews were conducted on the phone. The rest of the
interviews were conducted on TC campus in person over a five-day period. A typical purposeful
sampling approach, detailed in Appendix A, was employed to select the interview participants.
Once a list of TC participants has been established, the study investigator contacted a diversity
program director at TC who has agreed to coordinate and schedule the interviews. Based on the
availability of all the participants, on-campus interviews were scheduled and the researcher
traveled to TC to initiate interviewing process. The interviews were conducted in a semi-
structured manner and the questions, see Appendix B, guided the flow of the interview. The
semi-structured type of interviewing style was chosen to allow the interviewer to maintain a
maximum level of flexibility in wording the questions and in ordering of the questions asked
while keeping the interviewees focused on the topics being discussed (Johnson & Christensen,
2014; Merriam & Tisdell, 2016). Each interview was scheduled for an hour and conducted at the
GENDER GAP IN COMPUTER SCIENCE 36
participant’s office. With participant’s permission, as discussed in Appendix E, each interview
was recorded using a digital audio recording device and transcribed by the researcher within 24
hours of the interview taking place.
Role of Interviewer
An interview is a data-collection method where the interviewer asks questions related to
the research being conducted and the interviewee provides answers (Johnson & Christensen,
2014; Merriam & Tisdell, 2016). To make interviewing process more effective it is critical to
establish a research relationship between the interviewer and the interviewee (Bogdan & Biklen,
2007; Glesne, 2011; Maxwell, 2013; Merriam & Tisdell, 2016). Research relationships are both
complex and dynamic and can either facilitate or impede data collection (Maxwell, 2013). There
are three primary dimensions of the research relationship that affect the role of the interviewer in
collecting data (Creswell, 2014; Merriam & Tisdell, 2016). First dimension relates to whether or
not the interviewer is a member of the community being studied. As a former technology
executive and advisory board member at several college computer science programs, it is
possible that the study investigator could have been viewed as an insider. Being an insider has
both advantages and disadvantages. One major advantage of being an insider is that it provides
accessibility to the participants of the study (Maxwell, 2013). However, it is possible for the
interviewees not to share personal experiences and beliefs for the fear of being embarrassed
(Maxwell, 2013; Rubin & Rubin, 2012). The role of investigator is to anticipate such situations
and periodically remind the participants of the interview purpose and the importance of the study
(Creswell, 2014; Glesne, 2011). Second dimension of the research relationship relates to the
positionality of the investigator with respect to the participants (Maxwell, 2013). The perceived
power difference between an interviewer and the participant due to researcher’s professional
GENDER GAP IN COMPUTER SCIENCE 37
experience, social status, or demographics may influence the researcher’s ability to collect data
(Glesne, 2011; Merriam & Tisdell, 2016). Since all interview participants at TC were senior
members of the CS program leadership team, it is not likely that positionality was a factor in the
interviewing process. Lastly, during the data collection phase of the research, an investigator
needs to be cognizant of reflexivity, i.e., how the role and personal biases of the interviewer
influence the participants and how participants’ knowledge and views influence the interviewer
(Maxwell, 2013; Merriam & Tisdell, 2016). During the interview, for example, a study
participant may provide information that contradicts interviewer’s own views and beliefs. In
such situations, however, it is critical for the interviewer to remain focused on collecting the data
and to continue treating participants with utmost respect (Merriam & Tisdell, 2016; Rubin &
Rubin, 2012). At all times, the role of an interviewer is to learn about beliefs and experiences of
the interviewees without the influence of the interviewer’s own interpretation of the issues being
discussed (Creswell, 2014). Appendix C, Credibility and Trustworthiness, and Appendix D,
Validity and Reliability, document critical concepts adopted by the interviewer throughout the
study.
Interview Instrument
Conducting interviews provides an opportunity for researchers to learn about the
experiences and perspectives of the people being interviewed (Maxwell, 2013; Merriam &
Tisdell, 2016; Seidman, 2013). To keep the TC interview participants focused on the topic of the
inquiry a list of questions, sometimes referred to as the interview guide or instrument, has been
prepared, see a full list in Appendix B. The semi-structured type of interviewing style allowed
for the flow of the TC interviews to be attuned to the information being provided by the
GENDER GAP IN COMPUTER SCIENCE 38
participants. When appropriate, probing questions were utilized to further explore comments
made by the TC leaders being interviewed.
The interview question guide included questions focused on three related areas posed by
the research questions and the conceptual framework of the study. First, the TC leaders’
knowledge and personal views on the role of women in computer science and the impact of the
gender gap on the field will be examined. Second, interview questions that explore motivational
forces behind TC leader’s on-going efforts to reduce the gender gap in the program were asked.
Finally, as highlighted by the conceptual model, the influence of the computing and academic
social group identities on the actions and behaviors of the TC program leaders were explored.
Participating Stakeholders
To conduct this case study nine interviews were conducted with members of the CS
administration team. Members of the dean’s office, office of admissions, student advisors, and
diversity program managers were interviewed for the purposes of this study. Two of the nine
participants were women and everyone had over five years of experience in their current role.
Two interviews were conducted on the phone. The seven remaining interviews were conducted
during the campus visit several weeks later. In addition to the interviews, a number of artifacts
were also examined. Internal documents, including the faculty hiring guidelines, meeting
memos, management reports highlighting student application and admission rates, as well as
publicly available information, such as TC published studies and school communication bulletins
to prospective students were used as a source of information for analyzing the CS leaders’ role in
reducing the gender gap in the program. The Findings section below provides the results of the
qualitative data analysis and discusses themes that answer the research questions.

GENDER GAP IN COMPUTER SCIENCE 39
Findings
Modern organizations are multifaceted and ever-changing (Bolman & Deal, 2013). No
longer a mastery in one functional area is enough for the management teams to successfully lead
organizational changes needed to improve performance (Northouse, 2016). To implement
organizational agenda, managers need to possess knowledge, skills, and motivation to perform
multiple roles (Dyer, Gregersen, & Christensen, 2011). Not surprisingly, the CS administration
team at TC plays a number of roles in order to advance the diversity agenda in the program. A
list of significant findings, described below, highlights a number of roles that CS administrators
play in order to reduce the gender gap in computer science program. Although these findings are
interrelated and may not precisely fit into any single analytical group, the organizational
framework developed by Bolman and Deal (2013) is used to categorize them. This four-frame
model views all organizations through the symbolic lens, the human resources lens, the structural
lens, and the political lens. Using the symbolic lens, the first set of findings describes the role of
CS administrators as diversity champions. In this role they focus on changing organizational
mental models and attitudes that contribute to the gender gap in computer science. The role of
CS administrators as adaptive leaders is analyzed using the human resource lens. Creating a
supportive environment enables female students to overcome challenges associated with the CS
stereotype. Next, the role of CS administrators as organizational managers is examined using the
structural lens. Removing barriers and enhancing existing policies allows CS administrators to
encourage more female students to select computer science as their major and prevents others
from leaving the program. Finally, the role of administrators in setting an organizational agenda
to prioritize diversity initiatives is viewed through the political lens. Leveraging their
knowledge, experience, and beliefs enables CS administrators to work as a team addressing one
GENDER GAP IN COMPUTER SCIENCE 40
of the more challenging issues facing computer science programs in the U.S., i.e., how to
improve the gender balance in CS.
Finding 1: Role of CS Administrators as Diversity Champions
The role of CS administrators as diversity champions can be examined using the
symbolic frame of the organizational model developed by Bolman and Deal (2013). The
symbolic frame examines the basic elements of the culture, such as shared beliefs, daily routines,
and common traditions that bind organizations together (Bolman & Deal, 2013). Over the years,
the computer science culture has acquired a distinctive identity of being male-dominated and
unwelcoming to women (Bartol & Aspray, 2006). As Schein (2010) points out, however, it is
not uncommon for a small group of people within a large organization to develop their own
culture, sometimes called a microculture, that relies on a different set of beliefs and motivational
forces needed to achieve organizational objectives. At TC, the CS administration team has
developed a different perspective on the role of women in computer science. Every member of
the CS administration team interviewed has expressed the same shared view that women do
belong in computer science and that the CS stereotype is a by-product of the socio-cultural
beliefs. The newly evolved microculture supported by the CS administration team not only
recognizes the negative impact the gender gap in computer science has on academic programs,
but also recognizes that gender diversity is a major contributor to creativity and innovation in the
field of computer science. The following three assertions highlight common beliefs that enabled
CS administrators to become diversity champions focused on improving the gender balance at
TC.

GENDER GAP IN COMPUTER SCIENCE 41
Gender Bias in Computer Science is a Cultural Phenomenon
Stereotypical CS culture deters women from pursuing college degrees in computer
science (Cheryan et al., 2013). A number of studies have shown that career choices for young
women are influenced by social structures that define which professions are appropriate for men
and women (Broad & McGee, 2014; Stoilescu & McDougall, 2011). Reinforced by cultural
stereotypes that depict computer scientists as geeky men further detracts female students from
taking CS courses in college (Pechtelidis, Kosma, & Chronaki, 2015; Cohoon & Aspray, 2006).
The CS administration team at TC believes the gender gap in computer science is a cultural
phenomenon that can be addressed through a coordinated effort of diversity focused initiatives.
By leading the organization as diversity champions, they recognize and accept the challenge of
creating a more inclusive CS program at TC. While discussing the effects of the CS stereotype
in the field of the computer science, Alex who has been a part of the TC administration team
since the program inception said,
Most of cultural norms are established from very early on… like the notion that women
are caregivers. Well, women are caregivers because women were in the home taking care
of the kids all the God damn time. My mom was a caregiver, but she was also a really
good secretary…I believe just from personal experience that gender bias in CS and these
other things are typically gendered stereotypes. I think they are more a cultural norm than
anything else.
David and Albert, both members of the administration team who came to the U.S. after
receiving undergraduate degrees in their home countries, agree that socio-cultural factors deter
many women from considering a career in computer science in the U.S. and in many Western
countries. During the interview David stated, “I was born in a different society…[where] gender
GENDER GAP IN COMPUTER SCIENCE 42
[in CS] was never an issue. Coming here and doing what I do now, I never thought that I should
treat this person differently because they have a different gender.” Albert concurs, “I think it
[the gender gap in CS] pertains to the general role of women in these [Western] countries and
how they're being viewed. I think we have a general, sort of gender imbalance in many different
ways and I think computer science may be simply a reflection of that.”
Henry, a long tenured member of the leadership team, adds that “research shows that
middle school teachers get used to the boys answering the technical questions and start to
overlook the girls…Kids have already been brought up to think that these technical engineering
type things are more male.”
Mary, who oversees several diversity focused programs at TC, also expresses her
frustration at how pervasive the negative attitudes and biases are towards women entering the
field of computer science,
I was just looking at one of the judges on the Supreme Court were saying things like we
shouldn't have black students going to the top schools because they can't cope. That kind
of attitude, I think we have to stop thinking that you need to somehow lower the bar for
women and minorities. I think there is a cultural assumption that they're not there, women
[in computer science] are not there because they're no good at it or because they're just
not up to it. I think it's part of our culture, really. Broad culture. We need to address that
as a nation and I don't think given the current political climate we're doing a good job on
that.
The CS administration team at TC believes that the gender gap in computer science is a
result of the socio-cultural attitudes in the U.S. This view is supported by a number of studies
that highlight the CS stereotype and social structures as significant factors in women choosing to
GENDER GAP IN COMPUTER SCIENCE 43
avoid entering the field of computer science (Broad & McGee, 2014; Stoilescu & McDougall,
2011). As a result, the CS administration team is focused on creating a more gender-balanced
program environment by supporting the diversity focused initiatives.
Diversity is a Source of Creativity and Innovation in Computer Science
Team diversity improves organizational performance and enhances organizational ability
to innovate (Cohen, 2013; Prieto, Phipps, & Osiri, 2009). In their research, Bolman and Deal
(2013) have observed situations where team diversity has provided technology focused
organizations with a competitive advantage in attaining organizational goals. All interviewees
support this view. Alex, a long tenured CS administrator at TC agrees by stating,
Diversity matters because differences of experience, differences of opinion, all can drive
better products, better outcomes, because if everybody's quacking like a duck, then all
you make is something that works for ducks…You really need different ideas and
different experiences at the table. And if you exclude women, you're excluding half the
population and half the good ideas. So we shouldn't be doing that, and thankfully we're
not.
Bill, a CS administrator with over ten years of experience at TC says, “when they
[students] come in, they’ll often tell us they want to change the world somehow, and so I think
having that [gender] diversity here does help that process because they get a lot of different ideas
about what they can do.” He further states that “the benefit [of gender diversity] is more about
what they [students] create as they start to use what they’ve learned.”
Mary highlights the importance of gender diversity in computer science by saying,
GENDER GAP IN COMPUTER SCIENCE 44
There is more and more research to show that diversity pays off in business, for
example…People are paying more attention to diversity in general because it's been
found that we need to have diverse people on teams to make products that are inclusive.
Anthony, a CS administrator who is involved with program admission process at TC,
comments that “diversity is important to the field of computer science, diversity of experience,
ideas. People diversity in general enriches computer science, makes a better field.”
These interviewees summarize a common view shared by everyone on the TC leadership
team that gender diversity in computer science leads to creativity and innovation. This view is
also congruent with studies showing that a more diverse organization uses the distinct cognitive
abilities of its members to develop innovative solutions (Cohen, 2013; Horwitz, 2005; Prieto,
Phipps, & Osiri, 2009). Recognizing the importance of gender diversity in computer science is
an important driver of CS administrators’ efforts to reduce the gender gap in the program.
Women Do Not Display Lower Computer Science Aptitude than Men
It is not uncommon for many professionals in CS to accept the essentialist perspective
that the gender gap in computer science can at least be partially attributed to physiological
differences between women and men (Charles & Bradley, 2006). According to this perspective
women are incapable of developing the same computer science knowledge, skills, and
motivation needed to succeed in the field (Trauth, 2002). As a result, the claim is that the gender
gap in computer science is a direct outcome of the human physiological differences between
sexes. The entire CS administration team firmly rejects this perspective. Based on their
professional and academic experience in reducing the gender gap in computer science program at
GENDER GAP IN COMPUTER SCIENCE 45
TC, every member of the CS administration team interviewed found this traditional view to be
unrealistic, unfair, and offensive.
Responding to the question on whether physiological differences result in women staying
away from the field of computer science Henry said,
At the moment the [essentialist-based] hypothesis that because of our reproductive
systems [there is a gender gap in CS] …I genuinely think that seems unlikely. Because
we as a whole society would have noticed this somehow before, if there was that much of
a difference in how we think. It would've come up somewhere else.
Rachel, who is a world-renowned computer scientist and a member of the CS
administration team, has described her experiences when faced with questions regarding girls’
abilities to succeed in computer science,
Because I am computer scientist, when people said we need to change computer science
curriculum to appeal to girls, I knew this was wrong. I knew that girls can do CS as well
as the boys. Nobody could argue with me because they didn’t have the same
qualifications I did.
In reference to people who still think that women are incapable of becoming computer
scientists, Alex’ response was blunt and to the point, “Yes, some folks say it’s human biology [a
reason why girls unable to succeed in CS]. Well, these people are full of shit. I don’t think it
[aptitude in CS] is gendered because the X chromosome dictates that you’re going to be like this.
Jon agrees by commenting,
There are still dinosaurs there that will tell you that, ‘Hey, computer science is for men,
and you’re a woman so you can’t do it. You should be studying in an easier program,
like a nursing school or something.’ But those are fortunately fewer and fewer.
GENDER GAP IN COMPUTER SCIENCE 46
Why…because they are seeing that, that’s not the right way. There is no foundation,
there is no scientific explanation for these things.
According to Mary, there are still many educators around the country who feel that
women are incapable of succeeding in rigorous CS academic programs. She says,
We challenge the findings of the research that went on in the 90’s. That is so frustrating
when people say to me, ‘What did you do to change the curriculum to make it female
friendly.’ We have to keep saying that we did not change curriculum. It’s crazy to think
that our faculty would ever agree to anything like that. Women can do the work.
The CS administrators share a common belief that gender does not influence the aptitude
in computer science. Years of teaching and observing the academic performance of men and
women in CS programs have led this group to conclude that men and women are equally capable
of succeeding in the field of computer science. The conclusion is supported by recent
neurological studies where a human brain was shown to have more pronounced differences
between individuals rather than between genders (Joel et al., 2015).
Understanding the underlying socio-cultural causes of the gender gap in CS and
recognizing the importance of diversity in the field of computer science enables CS
administrators to play an active role in creating a microculture that promotes a more inclusive
environment at the school. As diversity champions, the CS administrators show their unyielding
commitment to initiating and sponsoring initiatives that reduce the gender gap in computer
science. They are effectively countering the socio-cultural forces that prevent many talented
female college applicants from considering computer science as their major.

GENDER GAP IN COMPUTER SCIENCE 47
Finding 2: Role of CS Administrators as Adaptive Leaders
An adaptive leadership style employs strategy that assist followers in overcoming
challenges they face (Northouse, 2016). Employing human resource strategies, such as creating
a nurturing environment, encouraging participation, and promoting diversity, is an effective way
for adaptive leaders to provide followers with the support and motivation they need to
accomplish their goals (Bolman & Deal, 2013). In higher education, it is the responsibility of the
school leaders to develop an academic and social environment that promotes student learning
and supports student needs (Kezar, 2001). At TC, computer science program administrators
have created a supportive environment that not only attracts female candidates to the program,
but also encourages them to succeed in a field that has traditionally been dominated by male
students. By frequently interacting with students and keeping them motivated, CS administrators
have succeeded in creating one of the top computing programs in the country that is also gender-
balanced.
Creating a Supportive Environment Improves Gender Balance in CS Programs
Creating an environment where people know they are being supported and feel they are
part of the community is one of the critical roles played by adaptive leaders (Northouse, 2016).
At TC, every member of the CS administration team is either teaching, advising, or directing
student-run programs. This high level of student interaction allows CS administrators to ensure
that both men and women in the computer science program feel a sense of belonging. It also
enables administrators to identify specific issues that may contribute to the gender gap in CS.
During the early days of the program members of the CS administration team realized that a
significant number of female students admitted to the program succumbed to impostor syndrome
and questioned whether they should actually be in the program. As Jack explains,
GENDER GAP IN COMPUTER SCIENCE 48
We have primarily female students who come and think, “Wow, they accepted me. That
probably was a mistake, I'm going to keep it quiet because if they find out, they're going to send
me back.” So there is a lot of work that we need to do for that.
Highlighting the negative effects of impostor syndrome on female students in the
program, Jon says, “female students who are suffering from the impostor syndrome think that it
was a mistake I got admitted, I don't belong here. Maybe I should change my major, go
somewhere else, yada, yada, yada.”
During his interview Henry also mentions the impostor syndrome as a possible cause for
the gender gap in computer science programs,
Lots of people do have this impostor syndrome thing, where [they] think, “We don't
really belong here,” and I can see how that has this non-liner effect, that if it happens a
little bit, and it happens to other people like you, then it's going to just grow in on itself,
so I think that might be one of the issues [contributing to the gender gap in CS].
The notion of fit has been extensively examined through several school administered
studies. Student surveys and interviews are periodically conducted to assess how female
students socially fit in the CS program. In describing the study findings, a recent report
published by the school states that “we have consistently found that women were feeling
comfortable in the CS community and fitting in with peers.”
A frequent level of interaction with students provides CS administrators with
opportunities to develop a nurturing environment insuring that female students feel they do
belong in the program. Describing a leadership role that CS administrators play in reducing the
negative effects of the impostor syndrome, Jon explains,
GENDER GAP IN COMPUTER SCIENCE 49
[We] tell them, "Well, you need to be a little bit more proactive, you are in the right
place. We chose you, you got admitted because you have what it takes. You chose us and
then we chose you. So this is a coalition that we have, and we're going to give you the
support so that you don't feel totally isolated socially… you’re here to shine and this is
the right place.”
Rachel further details the level of support provided by the CS administration team to
female students,
We make them feel as part of the school. The extra support by the way is not in terms of
academic, like asking them to take remedial classes, or asking them to take extra classes,
or having one-on-one special tutors for them. No. What I mean is a support in terms of
making them feel comfortable about being in a male-dominated place.
In describing CS administration’s effort to develop a supportive environment for female
students Mary adds,
Over the years, we've built more and more programs, more and more activities so that we
have a very rich source in which the women can get visibility, leadership, the kind of
opportunities that made them very central to the culture, to changing the culture…There
is some research that shows women are less likely to ask for help. I think we've turned
that around by telling the women this is okay. We all need help at times. We should all
be asking for help when we need it.
Frequent and direct interaction with students has allowed CS administrators to identify a
significant issue that prevents many potential female applicants and newly admitted students
from considering a career in computer science. Believing that women belong in computer
GENDER GAP IN COMPUTER SCIENCE 50
science and can be as successful as their male counterparts motivates CS administrators to create
a supportive school environment where female students are encouraged to succeed. In a school
published report describing how to succeed building an effective CS program, the organizational
support involving administrators, diversity program directors, and faculty members was cited as
one of the more critical factors in improving the gender diversity in the program. All members
of the CS administration team participating in this study believe that the gender gap in computer
science can be significantly reduced, as Mary professed, if only program leaders “stop paying lip
service and become directly involved in creating a nurturing environment for all students.”
Maintaining Gender Balance in CS Programs is an On-going Effort
Implementing organizational changes, especially in tradition-bound institutions of higher
education, is a lengthy process typically measured in years (Kezar, 2001; Schein, 2010). Despite
achieving impressive results in admitting and graduating female students majoring in the
computer science program, CS administrators recognize a need for continuing to invest in
diversity focused initiatives. Several members of the CS administrators team have highlighted a
possibility of losing momentum in their current efforts if diversity programs are not maintained.
During the interview Jon commented,
So we still offer the support [to female students]. Somebody may say, "Okay, well, you
met your goal…why don't you get rid of all the support?" Well, because there are still all
these issues, and still this is a male-dominated field even with female students [in our
program].
Alex agrees with a need to continue investing in diversity initiatives that focus on
reducing the gender gap in CS,
GENDER GAP IN COMPUTER SCIENCE 51
I think you still need to create an environment in which women feel empowered and feel
successful. If all they did was walk into class, and in class it was the same old stupid
sexist jokes that were coming out of professors' mouths, I don't think that would work
well. I don't think we'd retain all the women that we admitted…I think you can lose
something as easily as you created it if you're not careful to adhere to the principles that
got you to where you are…I do think we need to make sure that we do what has made us
be successful and continue that as best as we can.
Mary also highlights the role program administrators play in sustaining the gender
diversity focused efforts,
I think you actually do have to keep your eye on the ball. I really do…I mean if
leadership came in that had no interest, or didn't see this as a value, didn't see diversity as
a value, things could change for the worse…I think the tone whether we’re inclusive or
not is said by many people throughout the school...Something that goes on in the
classroom can send a [wrong] message. If there's somebody who can step in and say
that's not acceptable, then you can sustain [current efforts] …I think we can still do more.
These interviewees highlight an important point that creating a nurturing environment for
female students who may not feel they belong or can succeed in the program is just the first step
in a protracted process of reducing the gender gap in CS. Investment in diversity initiatives
needs to continue until the cultural environment of the organization is changed to sustain them.
Albert has eloquently summarized this point, “When it comes to diversity [initiatives], you just
can’t take the foot off the pedal.”
GENDER GAP IN COMPUTER SCIENCE 52
Program administrators and faculty’s participation in social activities and professional
conferences with students is also cited as an important component of the on-going efforts to
support the gender diversity in the program. A school published report highlights the value
associated with faculty-student lunch and dinner sessions sponsored by the school. This
document states,
These dinners...provide a chance for students to meet faculty in a relaxed, non-
judgmental atmosphere, and to increase the visibility of successful women computer scientists.
We [program leaders] have found that a core group of senior faculty…are very supportive.
Published plans to jointly attend and present at The Grace Hopper Celebration of Women
in Computing, ACM’s Special Interest Group for Computer Science Education, and other
conferences also highlight the on-going commitment by the CS program leaders to support the
gender diversity efforts.
As adaptive leaders, the CS administration team at TC is focused on students’ success. A
frequent interaction between students and administration facilitates a nurturing program
environment that provides all students with a support network. Many female students especially
benefit from studying in such an environment as they feel empowered to succeed in a field still
dominated by men.
Finding 3: Role of CS Administrators as Organizational Managers
Managers are responsible for developing organizational structures that enhance
operational efficiency and improves organizational performance (Bolman & Deal, 2013).
Whereas leadership is focused on developing and implementing adaptive organizational changes,
the role of managers is to ensure that organizations function in an orderly and predictable manner
(Northouse, 2016). One of the key roles of the organizational manager is to remove any barriers
GENDER GAP IN COMPUTER SCIENCE 53
and implement new procedures that help organization to achieve their goals (Clark & Estes,
2008; Rueda, 2011). Over the years, program administrators at TC have enhanced a number of
policies and procedures focused on reducing the gender gap in CS. In particular, student
admission policies and faculty hiring procedures were revised to accommodate a more inclusive
environment at school.
Changing Student Admission Criteria to Focus on Conceptual and Analytical Abilities
Reduces the Gender Gap in CS
Many computer science programs in the U.S. favor applicants with coding experience.
As the curriculum of these schools is focused on computer science languages and programming
techniques, prior exposure to coding is viewed as a major advantage for student academic
success in computer science. Due to the CS stereotype, however, many young women avoid
learning computer programming and may not be admitted to CS programs even when they apply
(Cheryan et al., 2013). Although programming plays a prominent role in the field of computer
science, CS administrators reject the view that computer science is mostly about programming.
They view CS field to be more intellectual with a theoretical core analogue to the ones in
mathematics and sciences. As a result, at TC the focus is on identifying CS program applicants
with strong analytical and conceptual skills. Changing admission polices to accept students with
broader abilities, and not just programming experience, also widens the pool of qualified
applicants resulting in more women being admitted. Bill explains the current admission
philosophy at TC which contributes to a more gender-balanced school environment,
[We] basically told admissions, programming and especially programming in high school
is not absolutely necessary. What is necessary is mathematical and scientific
thinking…[We] want to see students that are strong on the STEM side, specifically
GENDER GAP IN COMPUTER SCIENCE 54
science and math, because we know that that's in high schools, of course. A lot of places
don't have computer science, even today…When we ask for things like that, more women
start to pop up towards the top because they are very strong in math and science…They
can pick up programming, they can pick up the computer science ideas because we were
so connected to those fields.
Alex also comments on how changing the admissions policies at TC facilitates an
increase in women being admitted to the computer science program,
We explicitly instructed the admissions office not to look for prior programming
experience as a key admissions decision maker because again, women were so
underrepresented in high school computer science classes. If you're going to hold that
against them just because of whatever is operating at the high school level, that's a
problem. So we were looking more at mathematical aptitude…We believed that we
could take smart kids and teach them what we needed to teach them. And that turned out
to be true.
Rachel adds, “We knew that computer science courses at the high school level were
strongly mis-gendered in terms of representation. So if we were going to focus on that as a pre-
req, if you will, or a hidden pre-req, that was not going to serve this [reducing the gender gap in
CS] cause well at all.”
Anthony strongly agrees with Rachel, Alex, and Bill that de-emphasizing programming
experience for CS program admissions enables more women to be admitted,
[We] look for things at a different level than how much computer science has this person
actually had. And it allowed us to put students perhaps on a more level playing field
because there are some school districts and some independent schools that offer a great
GENDER GAP IN COMPUTER SCIENCE 55
deal in computer science, and then there are others that just don't offer much at all. And
so if you're really looking for a representative group of students, whether or not there's a
lot of computer science or a deep dive in computer science coursework available, [it]
limits who you can ultimately be interested in. And so we found this [revised admission
policies] to be one of the freeing factors for us.
While reviewing a list of requirements in order apply to the computer science program at
TC, one common requirement or a recommendation was noticeably missing. None of the
examined documents and web pages mention the importance of or even a preference for
candidates with programming background. Instead, a strong emphasis is given to candidates
who are interested in leveraging the CS education to pursue broader interests. The Office of
Admissions web pages, describing the CS program to prospective students state, “Computing is a
discipline with strong links to many fields, and our program gives you unparalleled flexibility to
pursue these fields.”
According to the Information Processing Theory, any new information that is connected
with prior pertinent knowledge is acquired and assimilated more quickly (Schraw & McCrudden,
2006). CS administrators at TC believe that conceptual understanding of mathematics and
sciences is more relevant to the students’ future success in computer science than knowing how
to program. In addition, the CS administration team recognizes that requiring a programming
experience for CS program admission is reducing the pool of qualified students and excludes
many qualified female applicants from being considered. Changing the criteria for student
admission to focus more on conceptual and analytical abilities attracts stronger candidates and
reduces the gender gap in CS.

GENDER GAP IN COMPUTER SCIENCE 56
Diversity Focused Education for Faculty Improves Gender Balance in CS
Education is a critical, research-based approach to acquiring conceptual knowledge to
deal with new situations (Clark & Estes, 2008). As most faculty in the U.S. computer science
programs are male, they may not be aware of their unconscious biases that can potentially lead to
discriminatory behavior (Jackson, Hillard, & Schneider, 2014). Implementing diversity focused
training is an effective intervention to reduce gender biases in organizational settings (Carnes et
al., 2015).
To negate effects of the homosocial reproduction and to ensure a more inclusive hiring
process, the CS administration team has implemented mandatory diversity training for all
members of the faculty hiring committee. As part of this process one of the member of the
faculty hiring committee is nominated as a diversity champion, advocating for female and other
diversity candidates. Rachel describes the role of the faculty diversity champion as a “diversity
advocate, who is on the faculty search committee, who is designed to make sure that we're
paying attention and not just selecting people who walk and look and talk like us.” While
discussing the faculty hiring process, Bill adds, “we have a representative in the school, in the
[faculty hiring] committee that purposely makes sure that people don't forget about diversity and
diversity is not just male female and so there's all dimensions that way.”
Additionally, diversity and unconscious bias training is offered to all CS faculty at school
to ensure they gain awareness of any hidden biases that may influence their own attitudes
towards female students in CS major. These sessions are either presented or attended by
members of the CS administration team. Albert, who attended one of the diversity sessions for
CS faculty, describes his experience,
GENDER GAP IN COMPUTER SCIENCE 57
We had a session on gender issues in the classroom [focusing on how] to sort of be equal.
If you had one female in your class and you had ten guys, well, you call on the woman
ten times because then it's all equal. Right? So we had some videos that were
produced…about sort of, again, stereotypical comments that instructors might make. We
talked about how to actually give women a voice, not by calling on them ten times
because we had a lot of women in the major who said, "You know, I can always tell
when an instructor thinks that he's got to do something because he calls on me all the
bloody time."
Members of the CS administration team believe that diversity focused education has
already contributed to a more inclusive culture in the program. Mary says,
Faculty are thinking about diversity…It comes into play when they're making decisions.
Major decisions, for sure. Like hiring, promotion, these kinds of things. I think that they
are aware that we can be very biased towards women, especially in a field like computer
science. I think there is a general awareness.
Jack feels that outcomes of the diversity focused education can also be seen in the
classroom. As evidence, Jack comments on how the class assignments have become more topic
relevant and less gender specific,
We get very few cases where a [female] student, I can't think of anything recently where
a student came and said, "Wow, that particular assignment, that was like so male biased.”
I've never seen anything like that here. I have seen that in my previous stints where part
of your assignment is to help with some sport or something like that, like football. And
it's like, the female students got disengaged almost immediately. Even some of the male
students did too, they're like, "I don't even follow this, why is this here?"
GENDER GAP IN COMPUTER SCIENCE 58
Alex supports Jack’s comments and adds, “[our faculty is] thinking about creating
assignments that are not all first-person shooters that appeal to one stripe of personality, not
strongly gendered, but also exhibits traits on both sides.”
Education plays a central role in implementing organizational changes that require new
knowledge and skills (Schein, 2010). To reduce the gender gap in the computer science
program, CS administrators promote diversity and unconscious bias educational sessions for all
CS faculty. A 2017-2018 Faculty Hiring Plan memo from the dean’s office outlines a 5-year
diversity plan. Highlighting the importance of diversity education, one line of the memo states,
“You must read chapter 4 of Joann Moody’s excellent book on faculty diversity or I will kill you.
The key observation is that we’re not helpless when it comes to ensuring diverse faculty.” The
focus on diversity education has resulted in improved faculty hiring practices and enhanced
teaching methods employed in the classroom, which CS administrators believe, further reduces
the gender gap in the computer science program.
Organizational managers possess authoritative powers to revise policies and procedures
in order to achieve organizational goals (Bolman & Deal, 2013). As part of their management
responsibilities, CS administrators focus on implementing organizational changes that reduce the
gender gap in the program. Enhancing student admission policies and introducing diversity
focused training for the faculty were some of the more prominent changes implemented by CS
administrators. In their role as managers, they continue to review existing program policies and
introduce new procedures that can further improve student performance and diversity in the
program.

GENDER GAP IN COMPUTER SCIENCE 59
Finding 4: Role of CS Administrators in Organizational Agenda Setting
Organizations are comprised of different internal groups, each having different priorities,
expectations, and resource requirements (Bolman & Deal, 2013). Managing competing priorities
of each stakeholder group in higher education institutions is a complex endeavor that can distract
leaders from maintaining focus on organizational goals (Kezar, 2001). Achieving organizational
congruency often requires political prowess by the leadership team (Schein, 2010). To ensure
that attainment of the gender-balanced program environment is being viewed by internal groups
as a top organizational priority, CS administrators set an organizational agenda to include
diversity topics.
Including Diversity on the List of Organizational Priorities Improves Gender Balance in
the Program
Agenda setting is a political skill that allows leaders to promote organizational changes
and define a roadmap to achieve them (Bolman & Deal, 2013). Information dissemination and
collecting feedback is an effective method to set and communicate the organizational agenda to
the stakeholders (Lewis, 2011). This is especially important where professional priorities, such
as research, and organizational priorities set by the administrators may diverge (Lewis, 2011).
At TC, CS administrators regularly meet with faculty to ensure that they appreciate the
priority of improving the gender balance in the computer science program and maintain vigilance
in their efforts. According to Mary, it is critical for the school faculty to understand that
reducing the gender gap in computer science is a top priority and requires their support. She
explains,
GENDER GAP IN COMPUTER SCIENCE 60
The big message [to faculty] is you don't do this alone. Diversity is not a one-person
thing. It really is a community effort. From the top, from the leadership to the students,
it's a community you build.
Henry agrees that involving faculty members in diversity discussions not only creates a
more participatory program environment, but also emphasizes diversity as a major focus for the
administration team,
I'm very comfortable with smallish meetings, where everyone is around the table like
this, and gets a chance to talk. So for example, it's very important we have these
[diversity] conversations with faculty…I bring them in and say, "What do you think? Tell
me what you really think, and let's see if we can make it better." In a recent faculty
survey, we asked people about their experience, one of the things were the gender related
issues.
Albert explains how collecting feedback from faculty is an important part of the diversity
agenda in the program,
[While meeting faculty] you do get reasonable concerns, but you actually get them out
on the table. A typical concern might be something like, "Well, if we look like we're
working harder to attract women to apply here, wouldn't that mean that the women who
are here…feel like they got in on some special play there?” It's not crazy. These are
actually real concerns…So I might go to some of those faculty lunches and talk at that
level.
Another example of how the diversity-focused agenda is being communicated to the
faculty can be seen in the management meeting minutes. One such memo discusses a proposal
to include a new language into the faculty recruiting job ads, “We particularly encourage
GENDER GAP IN COMPUTER SCIENCE 61
applications from candidates who have a demonstrated track record in mentoring and nurturing
female and under-represented minority students.” This example clearly demonstrates the
importance that the CS leadership team places on achieving the gender diversity in their
program.
A frequent dialogue with faculty is an effective way to sustain focus on such diversity
initiatives as improving the gender balance in CS program. It sets diversity as an integral part of
the organizational agenda and allows CS administrators to emphasize diversity as one of the
organizational priorities. Discussions with faculty also provide CS administrators with an
opportunity to address any diversity related concerns that faculty may have.
Agenda setting is a political tool that enables leadership teams to specify organizational
priorities and areas of focus (Bolman & Deal, 2013). Having clarity with respect to
organizational goals allows internal groups to become more cohesive in implementing
organizational agenda (Schein, 2010). CS administration team continues to emphasize diversity
initiatives to be an area of focus ensuring that a reduction of the gender gap in computer science
remains a top organizational priority at TC.
Solutions and Recommendations
Discussion
Very few schools in the U.S. have achieved the same level of success in addressing the
gender gap in their computer science programs as the CS program at TC. While preparing for
the on-campus interviews, the study author fully expected CS administrators to take the
opportunity and discuss how their unique leadership style or innovative management practices
led to the program becoming one of the more gender balanced in the country. Yet, this is not
what happened. As the interviews progressed, it became obvious that improving the gender
GENDER GAP IN COMPUTER SCIENCE 62
balance in CS program was not viewed as the ultimate objective, but rather as a first step in the
process of achieving the overall diversity in the program. The most interesting and somewhat
surprising aspect observed during the interviews was how CS administrators internalized the
importance of diversity to the field of computer science. They did not see diversity as an add-on
function in their daily practice. To them, achieving diversity in the program was a foundational
element in becoming successful CS program leaders. Accepting the gender gap in computer
science as a professional challenge, CS administrators applied their knowledge, skills, and
motivation to solve it as they would solve any other challenging problem. Once solved, their
expectation was to move on to face the next set of challenges in the diversity space. Every single
member of the CS administration team spent time describing diversity challenges facing the
organization. For example, they talked about difficulties in attracting and admitting Latino and
African American students to the undergraduate computer science program and hiring female
candidates for the open faculty positions. At times it felt like this group of leaders was more
interested in discussing their diversity challenges rather than their successes in reducing the
gender gap in the program. As a team, they were quite pleased with progress made in the gender
space, but were disappointed in slow progress of attracting other underrepresented groups to the
computer science program. To a computer scientist with over 30 years of experience in leading
technology organizations and recruiting graduates from dozens of CS programs in the U.S., this
was a new and a very welcomed experience.
Recommendations for Practice to Address KMO Influences
The CS administration team at TC has made a significant progress in reducing the gender
gap in the computer science program. Over the years several interventions have been
successfully implemented resulting in the computer science program at TC to become one of the
GENDER GAP IN COMPUTER SCIENCE 63
most gender balanced programs in the U.S. Despite recent success in changing the cultural
aspects of the program that has significantly reduced the gender gap in the program, several
recommendations described below, as well as associated implementation and evaluation plan
included in Appendix F, are being proposed in order to sustain the momentum of the diversity
focused initiatives. Many of the recommendations support the already existing practices at TC.
However, as mentioned by most administrators, there are several factors that necessitate the CS
administration team to continue investing in gender and other diversity focused initiatives. The
length of time it takes for the cultural environment of the professional practice to change,
personnel changes on the leadership team, as well as the ever-evolving field of computer science
are just some of the factors that can negatively impact the progress of the organizational changes
(Bolman & Deal, 2013; Denning, 2005; Schein, 2010).
Recommendations section is based on the KMO model that permeates all components of
this study. It highlights a role that declarative, procedural, and metacognitive knowledge factors
play in CS administrators’ ability to reduce the gender gap in computer science program at TC.
Motivation influencers show how self-efficacy and utility value enable CS administration team
to see significance in reducing the gender gap in computer science program and gain confidence
in their ability to successfully implement aspired changes. Finally, the organizational influences
review how the organizational vision, policies and procedures, as well as the allocation of
resources provide support to the CS administrators’ efforts in designing and implementing
diversity focused initiatives that reduce the gender gap in the program.
Knowledge Recommendations
CS program administrators’ ability to reduce the gender gap in computer science at TC
has been recognized by many educators and practitioners as a significant milestone. Despite the
GENDER GAP IN COMPUTER SCIENCE 64
accomplishment, most members of the CS program leadership team have highlighted a risk of
losing the gains unless the school maintains its current level of knowledge-based interventions.
Hall and Theriot (2016) have identified three types of knowledge that support a more diverse
environment in college programs. First type refers to the conceptual knowledge that influences
stakeholders’ views and actions. At TC, CS administrators need to know different social and
cultural factors that influence female students not to choose majoring in computer science.
Second type is knowing how to structure interventions in order to achieve a more inclusive
environment. CS administrators at TC need to know how to enhance social setting in the
program to attract and retain more women with little or no programming experience. Third type
of knowledge is a knowledge of one’s own awareness as it plays a significant role in achieving
diverse focused efforts. At TC, CS administrators need to reflect on their own attitudes and
biases towards women pursuing degrees in computer science. Jointly, conceptual, procedural,
and metacognitive knowledge of the CS program leaders at TC plays a significant role in
sustaining gender parity in computer science program. Table 2 provides a summary of
knowledge influences and associated recommendations.
Table 2
Summary of Knowledge Influences and Recommendations
Assumed Knowledge
Influence: Cause, Need,
or Asset*
Validated
(H, HP,
N)
Priority
(Y, N)
Principle and Citation
Context-Specific
Recommendation
GENDER GAP IN COMPUTER SCIENCE 65
CS administrators need
conceptual knowledge
of different social and
cultural factors
influencing female
students’ decision not to
pursue college degrees
in computer science.
(D)
HP Y
Information
Processing Theory –
How individuals
organize knowledge
influences how they
learn and apply what
they know (Schraw &
McCrudden, 2006).
Provide CS
administrators with
diversity education to
further strengthen their
conceptual understanding
of the social and cultural
influences that affect the
college major choice of
female students.
CS administrators need
to know how to
structure program
environment in order to
attract and retain more
women with little or no
programming
experience. (P)
HP Y
Information learned
meaningfully and
connected with prior
knowledge is stored
more quickly and
remembered more
accurately because it is
elaborated with prior
learning (Schraw &
McCrudden, 2006).
Provide CS
administrators with
feedback data from
students new to
programming to
determine how to
enhance computer science
social environment at TC.
CS administrators need
to be cognizant of their
own attitudes and biases
towards female students
pursuing degrees in
computer science. (M)
HP Y
Information
Processing Theory --
The use of
metacognitive
strategies facilitates
learning (Baker,
2006).
Continue to provide CS
administrators with
implicit bias training to
reflect on their own
attitudes and biases
towards female students
seeking degrees in
computer science.
*Indicate knowledge type for each influence listed using these abbreviations: (D)eclarative =
(C)onceptual & (F)actual; (P)rocedural; (M)etacognitive
Declarative: (Conceptual) knowledge solutions, or description of needs or assets. To
reduce the gender gap in computer science, program administrators need conceptual knowledge
of various social and cultural factors that influence female college students not to pursue degrees
in CS. Conceptual knowledge describes theories and models underlying the problem of practice
being studied (Krathwohl, 2002). According to Information Processing Theory, which describes
the role of cognition, memory, and thinking, an individual’s ability to learn and to apply
GENDER GAP IN COMPUTER SCIENCE 66
knowledge is highly dependent on how the knowledge is organized (Schraw & McCrudden,
2006). According to Clark and Estes (2008), education is an effective approach to organizing
and delivering conceptual knowledge necessary to address new, complex, and unanticipated
challenges. Providing computer science program leaders with diversity education based on the
latest research findings will further strengthen their conceptual understanding of the social and
cultural influences that affect the college major choice of female students (see Table 2, for
conceptual knowledge influences and recommendations).
Many educational leaders have conceptual knowledge gaps in understanding social and
cultural factors contributing to a gender disparity in computer science programs (Cohoon &
Aspray, 2006). Studies show that a diversity-focused education promotes a more inclusive
academic and social environment by further enhancing the conceptual knowledge of the trainees
(Hall & Theriot, 2016). These studies also indicate that the same educational approach can
provide information that facilitates development of the strategies focused on how to reduce the
gender gap in CS.
Procedural knowledge solutions, or description of needs or assets. Procedural
knowledge provides guidelines for designing and managing tasks needed to achieve a goal
(Krathwohl, 2002). Many of the female CS program applicants do not have any previous
experience in computer programming and may feel they do not belong in the program. As a
result, CS administrators need procedural knowledge of how to enhance school’s social
environment to ensure that students with little or no programming experience can succeed.
Information Processing System Theory posits that information learned meaningfully and
connected with prior knowledge is stored more quickly and remembered more accurately
(Schraw & McCrudden, 2006). Information also guides development and implementation of
GENDER GAP IN COMPUTER SCIENCE 67
procedures in a familiar environment (Clark & Estes, 2008). Providing CS administrators with
academic performance data collected from female students with no or little programming
experience can enhance their procedural knowledge on how to develop a more inclusive
academic curriculum (see Table 2, for procedural knowledge influences and recommendations).
Improving academic and social environment of the educational programs in higher
education institution is a primary responsibility of its leadership (Kezar, 2001). According to
leadership theories, information plays a powerful role in guiding leaders’ actions (Northouse,
2016). Missing or incorrect information may result in the implementation of the ineffective
procedures, further compounding the problem of practice (Rueda, 2011). As a result, developing
and utilizing procedural knowledge based on elucidated information is a critical factor in
enhancing the social environment of the computer science program.
Metacognitive knowledge solutions, or description of needs or assets. To develop
effective strategies for reducing the gender gap in computer science, programs administrators
need to be cognizant of their own attitudes and biases towards female students pursuing degrees
in computer science. Metacognitive knowledge facilitates awareness of one’s own views and
allows individuals to assess their strengths and weaknesses (Krathwohl, 2002). According to
Information Processing Theory, this type of knowledge also enhances learning (Baker, 2006).
Studies have shown that a lack of metacognitive knowledge negatively influences decision-
making and corresponding actions of the educational leaders (Dasgupta & Stout, 2014). To
inhibit the influence of subconscious preconceptions, program leaders may need to continue to
participate in unconscious bias training. Training allows individuals to acquire knowledge and
practice skills while receiving corrective feedback (Clark & Estes, 2008). Providing bias
training to CS administrators may be an effective strategy in acquiring metacognitive knowledge
GENDER GAP IN COMPUTER SCIENCE 68
required for developing and implementing the gender diversity focused initiatives (see Table 2,
for metacognitive knowledge influences and recommendations).
All people have attitudes and biases shaped by their cultural environment that they may
not be aware of (Banaji & Greenwald, 2013). As most of CS administrators are men, such
implicit biases can lead to inadvertent discriminatory behaviors and can negate the efforts of
achieving the gender parity in computer science programs (Jackson, Hillard, & Schneider, 2014).
Recent research has shown that bias training focused on enhancing the metacognitive skills is an
effective intervention in reducing gender bias across all academic environments, even those
where such training has been previously conducted (Carnes et al., 2015; Jackson, Hillard, &
Schneider, 2014).
Motivation Recommendations
Motivation plays a critical role in leaders’ ability to develop and implement initiatives
that advance organizational goals (Clark & Estes, 2008). To be effective in their roles, leaders
need to choose and persist in addressing a complex organizational problem as well as to take
responsibility for advancing the social good of the organization (Northouse, 2016). Driven by
the broader sociocultural, economic, and demographic changes, higher education leaders are
challenged to implement diversity focused initiatives in programs they oversee (Kezar, 2001). In
order to champion diversity initiatives, such as a reduction of the gender gap in computer science
program, school administrators need to perceive a significance in achieving the task as well as to
believe in their own abilities to succeed (Eccles, 2006, Pajares, 2006). Expanding interactions
and establishing a participative dialogue with various stakeholder groups may increase
educational leaders’ motivation to manage complexity of successfully implementing diversity
GENDER GAP IN COMPUTER SCIENCE 69
initiatives (Pucciarelli & Kaplan, 2016). Table 3 below shows motivation influences and the
corresponding recommendations for CS administrators at TC.
Table 3
Summary of Motivation Influences and Recommendations
Assumed Motivation
Influence
Validated
(H, HP, N)
Priority
(Y, N)
Principle and
Citation
Context-Specific
Recommendation
CS administrators
need to believe in
their own
competences to
engage in initiatives
that reduce the gender
gap in computer
science programs.
HP Y Feedback and
modeling increases
self-efficacy
(Pajares, 2006).
Provide CS
administrators with
opportunities to engage
in diversity focused
discussions with
educators in other
college STEM
disciplines where
gender parity has been
achieved.
CS administrators
need to perceive
utility value in their
efforts to reduce the
gender gap in
computer science
program.
HP Y Learning and
motivation are
enhanced if the
learner values the
task (Eccles, 2006)
Provide CS
administrators with
opportunities to engage
in diversity focused
discussions with
external stakeholders
who advocate for a
greater diversity in
computer science
programs.

Self-Efficacy. To engage in challenging tasks, such as achieving gender parity in
computer science program, CS administrators need to believe in their own abilities to succeed.
Self-efficacy beliefs determine how individuals assess their own competencies (Bandura, 1993).
GENDER GAP IN COMPUTER SCIENCE 70
Strong self-efficacy beliefs motivate individuals to select and perform challenging tasks and to
prioritize task completion over other competing alternatives (Pajares, 2006). According to self-
efficacy theory, a number of factors facilitate formation of self-efficacy beliefs. Vicarious
experiences play a major role in affecting individual’s self-efficacy (Bandura, 1993). Observing
success or failure of others working in the same domain and facing similar challenges can either
raise or lower beliefs of one’s own capabilities (Pajares, 2006; Rueda, 2011). Providing CS
administrators with opportunities to engage in diversity focused discussions and receive
feedback from peers in other STEM disciplines, such as mathematics and sciences where gender
parity has been largely achieved, may assist in implementing initiatives that reduce the gender
gap in computer science programs.
In the last two decades several college STEM disciplines have made a significant
progress toward achieving gender parity in their respective fields. More than half of
undergraduate degrees in mathematics, biology, and chemistry are now awarded to women
(Cheryan, Ziegler, Montoya, & Jiang, 2016). Although several factors have facilitated this
increase, program changes that emphasized hard work and not gender-based innate abilities of
students may have also contributed to a greater number of female students selecting mathematics
and sciences as their majors (Cheryan et al., 2016). A field of computer science shares many
educational principles with mathematics and sciences (Denning, 2005). Examining diversity
focused initiatives successfully implemented by other college STEM educators may increase CS
administrators’ self-efficacy in achieving gender parity in computer science program.
Value. To prioritize and sponsor initiatives that aim to reduce the gender gap in
computer science programs CS administrators need to perceive utility value in accomplishing
diversity focused goals. According to the Expectancy Value Motivational Theory, utility value
GENDER GAP IN COMPUTER SCIENCE 71
is derived when achievement of specific tasks is congruent with individual’s personal or
professional goals (Rueda, 2011). Eccles (2006) states that recognizing utility value associated
with a given task facilitates learning and motivates individuals to accomplish that task. Many
external stakeholder groups, including federal and state legislators, industry sponsors, etc.,
support diversity focused efforts in higher education and hold school administrators accountable
for their implementation (Firestone & Shipps, 2005). Providing CS administrators with
opportunities to directly engage in diversity focused discussions with external stakeholders may
achieve accountability goals and further increase utility value associated with efforts to reduce
the gender gap in college computer science programs.
College leaders are responsible not only for the high academic performance of students
and for hiring the best faculty, but also for meeting accountability requirements imposed by
many stakeholder groups (Kezar, 2001). State legislatures, among other external stakeholders,
are looking to increase a number of diverse students majoring in engineering and computer
science as part of state higher education appropriations (State General Session, 2001)
1
.
Inability to improve program performance related to diversity initiatives may cause both
financial and reputational risks to the higher education institutions failing to comply with
stakeholder mandates (Conner & Rabovsky, 2011). By interacting directly with stakeholder
groups that advocate for a greater gender diversity in CS programs, school administrators can
further ascertain importance of reducing the gender gap and prioritize diversity focused
initiatives in programs they lead.

1
To ensure anonymity of the source, the link to the state website is not provided.
GENDER GAP IN COMPUTER SCIENCE 72
Organization Recommendations
Organization influencers play a critical role in determining whether organizations
underperform and fail or are able to implement organizational changes needed to achieve
organizational goals (Clark & Estes, 2008). Different types of organizations have different
organizational structures, goals, and philosophies (Kezar, 2001). As a result, each organization
has a unique combination of influencers that facilitate implementation of organizational
objectives (Bolman & Deal, 2013). Higher education institutions, despite their primary focus on
delivering quality education and supporting research, are also influenced by different
organizational factors (Kezar, 2001). Each institution adapts a different approach to
implementing initiatives, driven by organization’s own identity. Computer science program at
TC has its own set of organization influencers driven by the school’s priority to achieve a more
gender diverse environment. Communication processes, policies and procedures, as well as
resources allocated by CS administrators to diversity initiatives are just some of the key
organization influencers that support organizational goals at TC. Table 4 below highlights
organization influences and corresponding recommendations supporting the ongoing efforts to
reduce the gender gap in computer science program at TC.
Table 4
Summary of Organization Influences and Recommendations
Assumed Organization
Influence: Cause, Need,
or Asset
Validated
(H, HP,
N)
Priority
(Y, N)
Principle and Citation
Context-Specific
Recommendation
GENDER GAP IN COMPUTER SCIENCE 73
CS administrators need
to communicate
organizational vision
and their commitment
to achieving the gender
balance in the computer
science program.
HP Y Effective leaders are
knowledgeable of
communication
processes and how to
use them for effective
organizational change
(Denning, 2011;
Lewis, 2011).
Enable CS administrators
to conduct internal
information campaign
about the academic and
social benefits of student
diversity in the computer
science program.
CS administrators need
to develop student
admission policies to
foster a culture of
inclusion in the
computer science
program.

HP Y Effective leaders
demonstrate a
commitment to
valuing diversity
through inclusive
action. They promote
an organizational
culture that promotes
equity and inclusion
and cultivate an
atmosphere where
diversity is viewed as
an asset to the
organization and its
stakeholders (Prieto,
Phipps & Osiri, 2009).
Allow CS administrators
to enhance admission
policies that extend the
pool of candidates being
considered for admission.

CS administrators need
to align the allocation of
resources with the
gender-diversity
focused goals of the
computer science
program.

HP

Y

Effective change
efforts ensure that
everyone has the
resources (personnel,
time, etc.) needed to
do their job, and that if
there are resource
shortages, then
resources are aligned
with organizational
priorities (Clark and
Estes, 2008)

CS administrators to
ensure that diversity
focused initiatives are
prioritized, and have
adequate funding and
human capital resources
allocated to reducing the
gender gap in computer
science programs.

GENDER GAP IN COMPUTER SCIENCE 74
Communication Processes. To lead organizational changes focused on improving the
gender diversity in computer science program CS administrators need to effectively
communicate the organizational vision as well as specific initiatives related to this goal.
Communication is a social interaction tool by which the timing and the direction of the
organizational change is negotiated between the organizational leaders and other stakeholders
(Hattaway, 2012). It plays a critical role in introducing, implementing, and evaluating
organizational changes (Lewis, 2011). For higher education institutions communication
processes help to transmit important organizational values such as diversity and inclusion that go
beyond the traditional focus on research and teaching (Denning, 2011). Developing and
conducting an internal information campaign that highlights academic and social advantages of
reducing the gender gap in computer science program may be an effective approach to
successfully implementing diversity focused initiatives.
A communication between the organizational leadership and its stakeholders enhances
the decision-making process and improves the results of the organizational changes (Clark &
Estes, 2008; Denning, 2011; Wheeler & Sillanpa'a, 1998). Research has shown that poor
communication of organizational goals can act as a barrier to implementing a change (Gilley,
Gilley, & McMillan, 2009; Waters, Marzano, & McNulty, 2003). An internal information
campaign designed to socialize how organization can benefit from the espoused change is an
integral component of the communication process (Berger, 2014; Lewis, 2011). At TC, a
communication process that focuses not only on disseminating information and soliciting
feedback on specific diversity focused initiatives, but also on creating shared values among all
stakeholders with respect to achieving a more inclusive school environment can further enhance
administrators’ ability to reduce the gender gap in computer science program.
GENDER GAP IN COMPUTER SCIENCE 75
Policies. To reduce the gender gap in computer science program, CS administrators need
to develop new and enhance existing policies that would foster a culture of diversity and
inclusion. Organizational policies provide guidance for identifying and implementing tasks that
advance the organizational mission (Bolman & Deal, 2013). In higher educational institutions,
policies are established to support learning and working environment as well as to promote
diversity and inclusion (USC Policy, 2011). The amount of diversity in college programs is
directly dependent on the student admission policies (Galinsky et al., 2015). By enhancing
existing policies that broaden the criteria for student admission, CS administrators may promote
a culture of inclusion and reduce the gender gap in computer science program.
Alignment of policies with organizational goals, such as creation of an inclusive
academic environment, narrows the gap between organizational culture and organization’s
ability to achieve espoused outcomes (Clark & Estes, 2008; Prieto, Phipps & Osiri, 2009).
Research has shown that diversity focused policies are readily accepted by the organization when
the entire leadership team is involved in developing and implementing specific actions that
promote the culture of inclusion (Hunt, Layton, & Prince, 2015). By enacting admission policies
that allow the non-traditional candidates, such as those without any prior programming
experience to be admitted to TC, CS administrators can improve the gender balance in the
computer science program.
Resources. Allocating adequate resources to diversity focused initiatives is a critical
component of CS administrators’ ability to reduce the gender gap in computer science programs.
Resource allocation is a process which ensures that necessary resources are shifted to initiatives
that support organizational goals of the higher education institution (Rueda, 2011). It is also one
of the core functions performed by the management teams (Northouse, 2016). At a time when
GENDER GAP IN COMPUTER SCIENCE 76
availability of public and private funds accessible to universities continues to be under pressure
resource allocation to diversity focused initiatives, such as a reduction of the gender gap in
computer science, is of paramount importance (Burke, 2005). To ensure the advances in
achieving the gender balance in computer science program are not hindered, CS administration at
TC needs to continue prioritizing and deploying sufficient level of resources to the diversity
focused initiatives.
A critical review of the research on the gender gap in computer science has identified
several factors, including a lack of familiarity with the field and a lack of peer interaction, as
significant barriers to entry for many female college applicants (Cohoon & Aspray, 2006). Not
surprisingly, the same studies have highlighted the importance of allocating various types of
resources to ensure that such barriers were removed. Capital resources used for organizing
recruiting events and sponsoring guest speakers are needed in order to market computer science
programs to female college applicants (Robertson, Newell, Swan, Mathiassen, & Bjerknes,
2001). Human resources, including senior CS leaders overseeing the female-run student
programs, are also needed to be available on a full-time basis (Robertson et al., 2001).
Allocation of resources to the diversity focused initiatives is a critical element in reducing the
gender gap in computer science programs and needs to remain a primary focus for the CS
administrators at TC.
Based on the New World Kirkpatrick Model, the implementation and evaluation plan
discussed in Appendix F can provide a practical approach to implementing the KMO-based
recommendations. Using this approach would further support CS program administrators’
efforts in reducing the gender gap in computer science and achieving one of the primary
organizational objectives.
GENDER GAP IN COMPUTER SCIENCE 77
Limitations
This research is a case study of a single U.S. higher education institution that has made a
significant progress in reducing the gender gap in its computer science program. As such, a
number of limiting factors bound this study. First, TC is a high-ranking school that attracts a
large number of applicants to its CS program. This allows the CS program administrators to
review and potentially select more female candidates. Second, TC is in a geographical area with
a high concentration of technology companies. Proximity to these companies may potentially
attract more female applicants seeking employment in the industry. Lastly, as a high-ranking
school TC attracts more funding from the private donors and industry sponsors. Having access
to more funding may allow CS administrators to provide additional support to the diversity
focused initiatives on the on-going basis.
TC is just one of several schools in the U.S. that was able to significantly reduce the
gender gap in CS. Other higher education institutions that have also succeeded in improving the
gender balance in their CS programs have different school characteristics, such as the size of the
program, funding sources, and geographic location, among others. Additional research focused
on identifying common factors that enable all these schools to reduce the gender gap in CS
programs is needed.
Conclusion
The field of computer science is a primary driver behind major technological innovations.
Yet, almost 50% of the population is missing from the field. For decades, both educators and CS
practitioners have been looking for ways to increase the number of women entering college
computer science programs, in most cases only to be disappointed with the results. One school,
however, has risen to the challenge and has been able to succeed where many others have failed.
GENDER GAP IN COMPUTER SCIENCE 78
For over a decade the CS leadership team at TC has been examining various reasons for female
students’ avoidance of the computer science major. Focusing on the socio-cultural factors and
organizational influencers that deter many talented young women from entering a CS major, the
program administrators implemented several interventions resulting in the number of women
being admitted to the computing program to significantly exceed a national average of about
16%. As a result, TC has now become one of the more gender balanced top ranking computer
science programs in the country.
This case study examines the role of CS college administrators in reducing the gender
gap in the computer science program at TC. Literature Review section of the manuscript
provides a background and serves as a foundation for the analysis conducted in this study. Using
the Analytic Conceptual Framework, developed by Clarke and Estes (2008), knowledge,
motivation, and organizational factors that contribute to the gender gap in CS are reviewed. To
organize and synthesize empirical, theoretical, and experiential knowledge into logically
arranged configuration guiding this qualitative study, the original conceptual framework is
advanced. Data Collection and Instrumentation section provides details on how the data
collection and analysis were performed. A detailed discussion outlining the study findings based
on the document analysis and interviews describes the multiple roles that CS administrators play
to achieve a gender balanced environment in the program. Finally, a list of recommendations to
sustain CS administrators’ efforts to reduce the gender gap is provided.
The CS administrator team at TC has demonstrated that with proper knowledge,
motivation, and organizational support it is possible to reduce the gender gap in the computer
science program. It is a hope of this investigator that this study encourages additional research
needed to achieve a diverse and inclusive environment in the field of computer science.
GENDER GAP IN COMPUTER SCIENCE 79
References
Baker, L. (2006). Metacognition. Retrieved from
http://www.education.com/reference/article/ metacognition/.
Banaji, M., & Greenwald, A. G. (2013). Blindspot: Hidden biases of good people. New
York: Delacorte Press.
Bandura, A. (1993). Perceived self-efficacy in cognitive development and functioning.
Educational Psychologist, 28(2), 117-148.
Bartol, K. M., & Aspray, W. (2006). The transition of women from the academic world to
the IT workplaces: a review of the relevant research. In J. Cohoon & W. Aspray (Eds.),
Women and Information Technology: Research on Underrepresentation. Cambridge,
MA: The MIT Press.
Berger, B. (2014). Read my lips: leaders, supervisors, and culture are the foundations of
strategic employee communications. The Research Journal of the Institute for Public
Relations, 1(1).
Bogdan, R. C., & Biklen, S. K. (2007). Qualitative research for education: An introduction
to theories and methods (5the ed.). Boston, MA: Allyn and Bacon.
Bolman, L., & Deal, T. (2013). Reframing organizations: Artistry, choice and leadership
(5th ed.). San Francisco, CA: Jossey-Bass.
Bray, N. J. (2010). The deanship and its faculty interpreters: do mertonian norms of science
translate into norms for administration? The Journal of Higher Education, 81(3), 284-
316.
Broad, S., & McGee, M. (2014). Recruiting women into computer science and information
systems. Association Supporting Computer Users in Education “Our Second Quarter
Century of Resource Sharing”, 29-40.
GENDER GAP IN COMPUTER SCIENCE 80
Bundy, A. (2007). Computational thinking is pervasive. Journal of Scientific and Practical
Computing, 1(2), 67-69.
Burke, J. C. (2005). Achieving accountability in higher education: Balancing public,
academic, and market demands. Jossey-Bass. San Francisco, CA.
Carnes, M., Devine, P. G., Manwell, B. L., Byars-Winston, A., Fine, E., Ford, C.E.,
Forscher, P., Isaac, C., Kaatz, A., Magua, W., & Sheridan, J. (2015). Effect of an
intervention to break the gender bias habit for faculty at one institution: a cluster
randomized, controlled trial. Academic Medicine: Journal of the Association of
American Medical Colleges, 90(2), 221-230. doi: 10.1097/ACM.0000000000000552.
Ceruzzi, P. E. (2012). Computing: a concise history. Cambridge, MA: The MIT Press.
Charles, M. (2011). What gender is science? Contexts, 10(2), 22-28.
Charles, M., & Bradley, K. (2006). A matter of degrees: female underrepresentation in
computer science programs cross-nationally. In J. Cohoon & W. Aspray (Eds.), Women
and Information Technology: Research on Underrepresentation. Cambridge, MA: The
MIT Press.
Cheryan, S., Plaut, V. C., Handron, C., & Hudson, L. (2013). The stereotypical computer
scientist: Gendered media representations as a barrier to inclusion for women. Sex
Roles, 69(1-2), 58-71.
Cheryan, S., Ziegler, S. A., Montoya, A. K., & Jiang, L. (2016). Why are some STEM fields
more gender balanced than others? Psychological Bulletin, 10, 1-36. Retrieved from
http://dx.doi.org/10.1037/bul0000052.
Clark, R. E., & Estes, F. (2008). Turning research into results: A guide to selecting the right
performance solutions. Charlotte, NC:IAP.
GENDER GAP IN COMPUTER SCIENCE 81
Cohen, P. N. (2013). The persistence of workplace gender segregation in the US. Sociology
Compass, 7(11), 889-899.
Cohoon, M. J., & Aspray, W. (2006). A critical review of the research on women’s
participation in postsecondary education. Women and Information Technology:
Research on Underrepresentation, 137-182. MIT Press. Cambridge, MA.
Conner, T. W., & Rabovsky, T. M. (2011). Accountability, affordability, access: A review
of the recent trends in higher education policy research. Policy Studies Journal, 39(s1),
93–112.
Creswell, J. W. (2014). Research design: Qualitative, quantitative, and mixed methods
approaches. Sage publications.
Dasgupta, N., & Stout, J. G. (2014). Girls and women in science, technology, engineering,
and mathematics STEMing the tide and broadening participation in STEM careers.
Policy Insights from the Behavioral and Brain Sciences, 1(1), 21-29.
Denning, P.J. (2003). Computer Science. In A. Ralston, E. D. Reilly, & D. Hemmendinger
(Eds.), Encyclopedia of Computer Science (pp. 405–419). Chichester, UK: John Wiley
and Sons Ltd.
Denning, P. J. (2005). Is computer science science? Communications of the ACM, 48(4), 27-
31.
Denning, S. (2011). The leader’s guide to storytelling: Mastering the art and discipline of
business narrative. San Francisco, CA: Jossey-Bass.
Department of Labor. (2015). Computer and Information Technology Occupations.
Occupational Outlook Handbook. Retrieved from https://www.bls.gov/ooh/computer-
and-information-technology/home.htm
GENDER GAP IN COMPUTER SCIENCE 82
DiTomaso, N., Post, C., & Parks-Yancy, R. (2007). Workforce diversity and inequality:
power, status and numbers. Annual Review of Sociology, 33, 473-501.
Dowd, A. C. (2005). Data don’t drive: Building a practitioner-driven culture of inquiry to
assess community college performance. Indianapolis, IN: Lumina Foundation.
Retrieved from: https://www.luminafoundation.org/files/publications/
datadontdrive2005.pdf
Dowd, A. C., & Bensimon, E. M. (2014). Engaging the "race question": Accountability and
equity in US higher education. New York: Teachers College Press.
Dyer, J. H., Gregersen, H. B., & Christensen, C. M. (2011). The innovator's DNA:
Mastering the five skills of disruptive innovators. Boston, MA: Harvard Business Press.
Eccles, J. (2006). Expectancy value motivational theory. Retrieved from http://www.
education.com/reference/article/expectancy-value-motivational-theory/.
Eckel, C. C., & Grossman, P. J. (2005). Managing diversity by creating team identity.
Journal of Economic Behavior & Organization, 58(3), 371-392.
Eckel, P. D. (2001). [Foreword to report Understanding and facilitating organizational
change in the 21st century by A. Kezar]. ASHE-ERIC Higher Education Report, 28(4),
xiii-xiv.
Ek, A., Ideland, M., Jönsson, S., & Malmberg, C. (2013). The tension between
marketisation and academisation in higher education. Studies in Higher Education,
38(9), 1305-1318.
Firestone, W. A., & Shipps, D. (2005). How do leaders interpret conflicting accountabilities
to improve student learning? In W. A. Firestone & C. Riehl (Eds.), A new agenda for
research in educational leadership (pp. 81–91). New York: Teachers College Press.
GENDER GAP IN COMPUTER SCIENCE 83
Frieze, C., & Quesenberry, J. (2015). Kicking butt in computer science. Indianapolis, IN:
Dog Ear Publishing.
Frieze, C., Quesenberry, J., Kemp, E., & Velazquez, A. (2012). Diversity or difference? new
research supports the case for a cultural perspective on women in computing. Journal
of Science Education and Technology, 21(4), 423-439.
Galinsky, A. D., Todd, A. R., Homan, A. C., Phillips, K. W., Apfelbaum, E. P., Sasaki, S. J.
Richeson, J. A., Olayon, J. B., & Maddux, W. W. (2015). Perspectives on
Psychological Science, 10(6), 742-748.
Gilley, A., Gilley, J. W., & McMillan, H. S. (2009). Organizational change: Motivation,
communication, and leadership effectiveness. Performance Improvement Quarterly,
21(4), 75-94.
Glesne, C. (2011). Becoming qualitative researchers: An introduction (4th ed.). Boston,
MA: Pearson.
Gries, D. (1987). The 1985-1986 Taulbee survey. Communications of the ACM, 30(8), 688-
694.
Gurer, D. (1995). Pioneering Women in Computer Science. Communications of the ACM,
38(1), 45-54.
Hall, J.C. & Theriot, M. T. (2016) Developing Multicultural Awareness, Knowledge, and
Skills: Diversity Training Makes a Difference? Multicultural Perspectives, 18(1), 35-
41, DOI: 10.1080/15210960.2016.1125742.
Hattaway Communications (2012). The Art and Science of Storytelling [Webinar].
Available from https://vimeo.com/55898045.
GENDER GAP IN COMPUTER SCIENCE 84
Heck, R. H. & Hallinger, P. (2005). The study of educational leadership and management:
where does the field stand today? Educational Management Administration and
Leadership, 33(2), 229-244.
Hoogendoorn, S., Oosterbeek, H., & van Praag, M. (2013). The impact of gender diversity
on the performance of business team: evidence from a field experiment. Management
Science, 59(7), 1514-1528.
Horwitz, S. K. (2005). The compositional impact of team diversity on performance:
Theoretical considerations. Human Resource Development Review, 4(2), 219-245.
Hunt, V., Layton, D., & Prince, S. (2015). Why diversity matters. McKinsey & Company.
Retrieved from: https://www.richlandcountyfoundation.org/upload/documents/
diversity_matters.pdf
Jackson, S. M., Hillard, A. L., & Schneider, T. R. (2014). Using implicit bias training to
improve attitudes toward women in STEM. Social Psychology of Education, 17, 419-
438. DOI 10.1007/s11218-014-9259-5.
Joel, D., Berman, Z., Tavor, I., Wexler, N., Gaber, O., Stein, Y., Shefi, N., Pool, J., Urchs,
S., Margulies, D., Liem, F., Janggi, J., Jancke, L., & Assaf, Y. (2015). Sex beyond the
genitalia: The human brain mosaic. Proceedings of the National Academy of Sciences,
112(50), 15468-15473. doi:10.1073/pnas.1509654112
Johnson, R. B., & Christensen, L. (2014). Educational research: Quantitative, qualitative,
and mixed approaches (5th ed.). Thousand Oaks, CA: Sage Publications.
Kezar, A. (2001). Understanding and facilitating organizational change in the 21st century.
ASHE-ERIC Higher Education Report, 28(4), 1-147.
GENDER GAP IN COMPUTER SCIENCE 85
Kirkpatrick, J. D., & Kirkpatrick, W. K. (2016a). Kirkpatrick’s four levels of training
evaluation. Alexandria, VA: ATD Press.
Kirkpatrick, J. D., & Kirkpatrick, W. (2016b) Evaluation blunders and missteps to avoid.
Training and Development, November, 36-40.
Krathwohl, D. R. (2002). A revision of bloom's taxonomy: An overview. Theory into
Practice, 41(4), 212-218.
Krueger, R. A., & Casey, M. A. (2009). Focus groups: A practical guide for applied
research (4th ed.). Thousand Oaks, CA: SAGE Publications.
Langley, G. J., Moen, R. D., Nolan, K. M., Nolan, T. W., Norman, C. L., & Provost, L. P.
(2009). The improvement guide: A practical approach to enhancing organizational
performance. John Wiley & Sons.
Lee, H., & Choi, B. (2003). Knowledge management enablers, processes, and organizational
performance: An integrative view and empirical examination. Journal of Management
Information Systems, 20(1), 179-228.
Lewis, L. K. (2011). Organizational change: creating change through strategic
communication. Wiley-Blackwell.
Light, J. S. (1999). When computers were women. Technology and Culture, 40(3), 455-483.
Marini, M. M. (1990). Sex and gender: What do we know? Sociological Forum, 5(1), 95-
120.
Marsh, J. A. (2012). Interventions promoting educators’ use of data: Research insights and
gaps. Teachers College Record, 114(11), 1–48.
Maxwell, J. A. (2013). Qualitative research design: An interactive approach (3rd ed.). Sage
publications.
GENDER GAP IN COMPUTER SCIENCE 86
Mayer, R. E. (2011). Applying the science of learning. Boston, MA: Pearson/Allyn &
Bacon.
McEwan, E. K., & McEwan, P. J. (2003). Making sense of research: What's good, what's
not, and how to tell the difference Corwin Press.
Merriam, S. B., & Tisdell, E. J. (2016). Qualitative research: A guide to design and
implementation (4th ed.). San Francisco: John Wiley & Sons.
Miller, L. J., & Lee, J. S. (2014). Policy barriers to school improvement: What's real and
what's imagined? Center on Reinventing Public Education, 1-16.
Morelock, J. R. (2017). A systematic literature review of engineering identity: Definitions,
factors, and interventions affecting development, and means of measurement. European
Journal of Engineering Education, 1-23. DOI: 10.1080/03043797.2017.1287664.
NCES. (2015). Digest of education statistics. National Center for Education Statistics.
Retrieved from https://nces.ed.gov/programs/digest/d16/tables/dt16_322.10.asp.
NCWIT. (2014). By the numbers. National Center for Women and Information Technology.
Retrieved from
https://www.ncwit.org/sites/default/files/resources/btn_02282014web.pdf.
Nkomo, S. M., & Al Ariss, A. (2014). The historical origins of ethnic (white) privilege in
US organizations. Journal of Managerial Psychology, 29(4), 389-404.
Northouse, P. (2016). Leadership: Theory and practice. (7th ed.). Thousand Oaks, CA: Sage
Publishers.
Paglis, L. L., & Green, S. G. (2002). Leadership self-efficacy and managers' motivation for
leading change. Journal of Organizational Behavior, 23(2), 215-235.
doi:10.1002/job.137
GENDER GAP IN COMPUTER SCIENCE 87
Pajares, F. (2006). Self-efficacy theory. Retrieved from
http://www.education.com/reference/article/self-efficacy-theory/
Pechtelidis, Y., Kosma, Y, & Chronaki, A. (2015). Between a rock and a hard place: women
and computer technology. Gender and Education, 27(2), 164-182.
Prey, J. C. & Weaver, A. C. (2013). Fostering gender diversity in computing. Computer,
46(3), 22-23. DOI: 10.1109/MC.2013.97 .
Prieto, L. C., Phipps, S. T., & Osiri, J. K. (2009). Linking workplace diversity to
organizational performance: A conceptual framework. Journal of Diversity
Management, 4(4), 13-21.
Pucciarelli, F. & Kaplan, A. (2016). Competition and strategy in higher education:
managing complexity and uncertainty. Business Horizons, 59, 311-320.
Reason, P. (1994). Three approaches to participative inquiry. Handbook of Qualitative
Research, 324-339.
Robertson, M., Newell, S., Swan, J., Mathiassen, L., & Bjerknes, G. (2001). The issue of
gender within computing: reflections from the UK and Scandinavia. Information
Systems Journal 11, 111–126.
Rocco, T. S., & Plakhotnik, M. S. (2009). Literature reviews, conceptual frameworks, and
theoretical frameworks: Terms, functions, and distinctions. Human Resource
Development Review, 8(1), 120-130.
Rubin, H. J., & Rubin, I. S. (2012). Qualitative interviewing: The art of hearing data (3rd
ed.). Thousand Oaks, CA: SAGE Publications.
Rueda, R. (2011). The 3 dimensions of improving student performance: Finding the right
solutions to the right problems. New York, NY: Teachers College Press.
GENDER GAP IN COMPUTER SCIENCE 88
Sawyer, R. (2012). How to be more creative. In Explaining creativity: The science of human
innovation (2nd ed., chap. 22). London: Oxford University Press.
Schein, E. H. (2010). Organizational culture and leadership. John Wiley & Sons.
Schraw, G., & McCrudden, M. (2006). Information processing theory. Retrieved from
http:// www.education.com/reference/article/information-processing-theory/.
Seidman, I. (2013). Interviewing as qualitative research: A guide for researchers in
education and the social sciences. New York, NY: Teachers College Press.
Shapiro, R. J. (2014). The U.S. Software Industry as an Engine for Economic Growth and
Employment. Georgetown McDonough School of Business Research Paper No.
2541673. Retrieved from: https://ssrn-com.libproxy1.usc.edu/abstract=2541673
Stahl, B. C., Timmermans, J., & Mittelstadt, B. D. (2016). The ethics of computing: A
survey of the computing-oriented literature. ACM Computing Surveys, 48(4), DOI:
http://dx.doi.org/10.1145/2871196
Stanton-Salazar, R. D. (1997) A Social capital framework for understanding the
socialization of racial minority children and youths. Harvard Educational Review,
67(1), 1-41.
Stoilescu, D., & McDougall, D. (2011). Gender digital divide and challenges in
undergraduate computer science programs. Canadian Journal of Education, 34(1), 308-
333.
Tagg, J. (2012). Why does the faculty resist change? Change: The Magazine of Higher
Learning, 44(1), 6-15.
Trauth, E. M. (2002). Odd girl out: an individual differences perspective on women in the
IT profession. Information Technology & People, 15(2), 98-118.
GENDER GAP IN COMPUTER SCIENCE 89
USC Policy. (2011). Equal Opportunity, Affirmative Action and Non-Discrimination.
Retrieved from: https://policy.usc.edu/equal-opportunities/.
Varma, R. (2010). Why so few women enroll in computing? gender and ethnic differences
in students' perception. Computer Science Education, 20(4), 301-316.
Venkatesh, V., Morris, M. G., & Ackerman, P. L. (2000). A longitudinal field investigation
of gender differences in individual technology adoption decision-making processes.
Organizational Behavior and Human Decision Processes, 83(1), 33-60.
Vitores, A. & Gil-Juárez, A. (2016). The trouble with ‘women in computing’: a critical
examination of the deployment of research on the gender gap in computer science.
Journal of Gender Studies, 25(6), 666-680, DOI:10.1080/09589236.2015.1087309.
Waters, T., Marzano, R. J., & McNulty, B. (2003). What 30 years of research tells us about
the effect of leadership on student achievement. Mid-Continent Regional Educational
Lab., Aurora, CO. Retrieved from: https://files-eric-ed-gov.libproxy1.usc.edu/fulltext/
ED481972.pdf
Weeden, K. A. (2002). Why do some occupations pay more than others? Social closure and
earnings inequality in the United States. American Journal of Sociology, 108(1), 55-
101.
Wheeler, D. & Sillanpa’a, M. (1998). Including the stakeholders: the business case. Long
Range Planning, 31(2), 201-210.
Wing, J. M. (2006). Computational Thinking. Communications of the ACM, 49(3), 33-35.
Zimmerman, B. J. (2000). Self-efficacy: an essential motive to learn. Contemporary
educational psychology, 25, 82-91.
Zweben, S. & Bizot, B. (2014). 2013 Taulbee Survey. Computing Research News, 26(5).
GENDER GAP IN COMPUTER SCIENCE 90
Appendix A: Participating Stakeholders with Sampling Criteria for Interviews
Deciding on how to select interview participants is one of the critical decisions facing a
qualitative study investigator (Maxwell, 2013). A purposeful selection or a purposive sampling
of interviewees is the most common approach and provides researchers with a strategy to gain an
in-depth knowledge and to learn a meaning that participants assign to a set of experiences
(Maxwell, 2013; Merriam & Tisdell, 2016). The richness of the data gathered from the
purposefully selected participants assists the researcher in learning about issues that are principal
to the study’s focus (Merriam & Tisdell, 2016; Seidman, 2013). To examine the KMO
influencers that facilitate TC’s efforts in reducing the gender gap in computing, nine members of
the TC computing program leadership team were purposefully selected for the interviews. Two
primary limitations dictated the size of the sample. First, there were practical considerations.
Interviewing all members of the leadership team was likely to be a challenging and unrealistic
endeavor considering how busy these individuals are and the amount of time it takes to prepare
for and conduct each interview. Hence, interviewing nine participants seemed a more reasonable
objective. Second, the purpose of the interviews was to learn about the knowledge, motivation,
and the organizational influencers that impact CS leader’s actions with respect to reducing the
gender gap. As such, the effectiveness of the interviews, i.e. the focus on the specific actions and
experiences of the interviewees was by far more important than the number of the interviews
conducted (Johnson & Christensen, 2014; Maxwell, 2013; Merriam & Tisdell, 2016).
To ensure that data collected during the interviews captures as many diverse responses of
interviewees as possible, a maximum variation sampling strategy was adopted (Merriam &
Tisdell, 2016; Johnson & Christensen, 2014). A maximum variation sampling technique allows
the investigator to capture the widest possible range of views and experiences that participants
GENDER GAP IN COMPUTER SCIENCE 91
may share (Merriam & Tisdell, 2016; Seidman, 2013). In this study the sampling was stratified
along two dimensions, the gender and the role within administration. As the total number of
women leading the computing program at TC stood at three, the goal was to interview at least
two female leaders in order to capture their views and experiences in addressing the gender gap
in CS complementing data collected for their male counterparts. Also, two of the interviewees
selected were members of the dean’s office; the others were either the heads of the departments
or program directors. Since the goal of the interviews was to learn about the TC leaders’ views,
experiences, and actions in reducing the gender gap in computer science program, two selection
criteria were observed.
Criterion 1. Each interview participant must be a member of the computing leadership
team at TC. The focus of the study is to examine the KMO influencers of the TC leadership
team that facilitate increased admission rates of female candidates to the computing program. As
discussed earlier, the computing program leaders at TC are accountable for reducing the gender
gap in the program and therefore are a key stakeholder group of focus.
Criterion 2. Each interview participant has been a member of the TC’s leadership team
for at least two years. This criterion is used to ensure the TC leaders being interviewed
understand both the organizational culture and the program specific interventions implemented
to reduce the gender gap in CS.

GENDER GAP IN COMPUTER SCIENCE 92
Appendix B: Interview Protocol
Thank you for your time. If that’s ok with you, first I would like to ask you questions about you.
Q: Tell me when and why you decided to pursue a career in computer science?
Q: What is your favorite programming language?
Q: When was the last time you compiled your own code? Was it work related or for fun?
Q: In your role as ___________, what part of your job do you find to be most enjoyable? Why?
Q: Tell me about your typical day.
Q: When you were an undergraduate student, were there many female students in your computer
science classes?
--If yes, in your opinion what are the reasons for gender gap in computing to widen?
--If no, in your opinion what causes the gender gap in computing to persist?
Q: Do you think the existence of the reducing gender gap in computing is an issue? If yes, in
what sense?
Q: Has your personal perspective on women in computing been influenced in any way by your
professional/academic experiences?

For the last decade the admission rate of female students at TC has exceeded the national
averages. This is an incredibly impressive achievement. The next set of questions will focus on
the TC’s efforts in addressing the gender gap in computing:
Q: At TC, reducing the gender gap in computing has been a priority for many years. Why is it
such a priority?
Q: How much effort is needed to maintain this high rate and do you think there is room for
further improvement?
Q: Do you think all faculty share your view on the importance of gender diversity in computing?
Q: How does the school’s cultural environment (e.g., university policies, allocation of resources,
support from other TC departments) enhance or impede your ability to implement policies that
reduce the gender gap in the program?
Q: In addition to admission policies that focus on the candidates’ overall competencies, and not
necessarily on the extent of the previous programming experience, what other factors contribute
to the school’s success in attracting female students?
(If necessary, can you tell me about your Dr. Matlock (pseudonym) role as a diversity program
director?)
GENDER GAP IN COMPUTER SCIENCE 93

Q: How often does the topic of reducing gender gap in computing being discussed amongst
senior leaders at TC?
Q: What role, if any, do program administrators play in reviewing and modifying the admission
policies to the program?
(If necessary - Can you give me an example of a recent meeting or a discussion on the topic?)

There is an on-going debate on whether senior leaders in computer science programs consider
themselves to be computing professionals first and then educators or vice versa.
Q: Can you tell me how you feel about this argument and how you would describe yourself on
this spectrum?
Q: What is your level of interaction with computing industry leaders? With CS leaders at other
higher education institutions?
Q: Do you feel leaders in computing industry/education community share/support your views on
school’s goal of achieving a more student gender-balanced program?

Q: Is there anything that I have missed and should have asked you about the school’s efforts in
addressing the gender gap but did not?

Thank you so much for your time again!
GENDER GAP IN COMPUTER SCIENCE 94
Appendix C: Credibility and Trustworthiness
The credibility and trustworthiness of the qualitative studies ultimately depends on the
credibility and trustworthiness of the study authors (Merriam & Tisdell, 2016). Only when a
researcher collects and analyzes data following the best ethical practices, meticulously reviews
results, and relies on the evidence to explain any data discrepancies could the research study be
considered credible and trustworthy (Maxwell, 2013). As discussed earlier, employing ethical
practices during the study design, data collection, and analysis phases of the project is a critical
step in ensuring that the study is both credible and trustworthy (Creswell, 2014; Maxwell, 2013;
Merriam & Tisdell, 2016).
Researcher bias. Researcher is a primary instrument of data collection in qualitative studies
(Merriam & Tisdell, 2016). Eliminating one’s beliefs and perspective to achieve objectivity in a
research study is extremely difficult, if not impossible (Maxwell, 2013). It is therefore critical
for the researcher to reflect on his own feelings and document any biases that she or he may have
during the entire course of the study (Creswell, 2014; Merriam & Tisdell, 2016). An earlier
section of the manuscript, titled Role of Interviewer, describes potential concerns with how the
researcher’s extensive experience in the field of computing may influence data collection
process. To further reduce the effects of the research bias, the study author has also reviewed
each study finding while reflecting on his own views and beliefs with respect to the role of the
program administrators in CS. As part of this mental exercise, the key question: “Is this finding
skewed in any way by my perceptual lens?” was always considered.
Triangulation. Cross-checking data for consistency using various methods of data collection is
a sound strategy for enhancing credibility of the research (Creswell, 2014; Merriam & Tisdell,
2016). For the purposes of this study, data collected from the interviews was compared to the
GENDER GAP IN COMPUTER SCIENCE 95
internal documents provided by the CS administration team and publicly available information,
such as studies published by the school. Uniformity of data collected further contributed to the
credibility of the study.
Peer review. A peer review is a process that allows a person with significant knowledge of the
topic or with deep experience in research methodology to independently review the research as it
is being conducted (Creswell, 2014; Merriam & Tisdell, 2016). To ensure that study findings
accurately reflect data being collected, the author of this study has asked a researcher, not
connected to this study and with significant experience conducting investigations, to provide an
independent review. Coded data, which preserves the anonymity of participants and all other
aspects of the study, was reviewed to confirm the study findings.
GENDER GAP IN COMPUTER SCIENCE 96
Appendix D: Validity and Reliability
In qualitative studies, researchers focus on understanding how study participants interpret
their experiences and assign meaning to them (Merriam & Tisdell, 2016). As a result, some
researchers believe that concepts of validity and reliability are not relevant in qualitative studies
and are more attuned to the quantitative studies (Creswell, 2014). Other researchers, however,
disagree. Unlike the concepts of credibility and trustworthiness, which ensure the researchers’
findings are supported by the underlying data, this group of researchers view validity and
reliability in qualitative studies primarily focusing on whether study participants consistently
interpret their experiences (Seidman, 2013). In interview-based studies, validity and reliability
play a primary role in ensuring that participants reconstruct their experiences and assigns
meaning to them independently of who the interviewer is, a context of the interview protocol, or
the interview settings (Seidman, 2013). Checking consistency of interview responses to specific
events at different times is a common approach to increasing validity and reliability in qualitative
studies. During the TC campus visit, two of the study participants were met more than once. In
addition to formal interviewing sessions, they were met in a more casual setting. Although not
part of the formal interviewing process, discussions focused on the factors contributing to the
gender gap in CS and the role of the leaders at school to achieve a more gender-balanced
environment. Not surprisingly, both study participants were consistent in describing their
experiences in addressing the gender gap in the computer science program at TC. Due to
schedule conflicts no other follow-up discussions were set up. Despite the limitation, even these
additional interactions with two interviewees increased the level of validity and reliability
associated with the study.

GENDER GAP IN COMPUTER SCIENCE 97
Appendix E: Ethics
Ethical considerations and integrity of the researchers play a central role in qualitative
studies (Merriam & Tisdell, 2016; Maxwell, 2013). As the researchers are the primary
instrument used in the qualitative studies, it is their trustworthiness and ethical approach to study
design and data collection and analysis that determine the credibility of the study (Creswell,
2014; Merriam & Tisdell, 2016). Ethical considerations need to be part of every component of
the research design (Creswell, 2014). However, given that most of the data collection in
qualitative studies is performed through interaction with study participants, it is vital that
vigilance to the ethical standards pertaining to the researcher-participant relationship is sustained
(Glesne, 2011; Merriam & Tisdell, 2016).
There are several guidelines that are considered universal in defining ethicality in
qualitative studies (Glesne, 2011; Merriam & Tisdell, 2016). The ‘no harm’ provision mandates
that interviewees should not feel pressured or exploited during the interviewing process, and
subsequent to the study being published, be exposed to legal or financial hardship (Krueger &
Casey, 2009; Rubin & Rubin, 2012). The study participant’s anonymity needs to be preserved at
all times, and any data collected during the interviews must be kept confidential and secured
(Glesne, 2011). A consent form, outlining the participant’s voluntary participation, the right to
withdraw, and the right to privacy, needs to be written in a jargon-free language, and be
reviewed and signed by the interviewees prior to any interviews taking place. Study
investigators also need to reciprocate participants for their contribution to the study (Creswell,
2014; Glesne, 2011). As Maxwell (2013) points out, it is important to show to participants that
their contribution to the study has been a worthwhile effort.
GENDER GAP IN COMPUTER SCIENCE 98
To ensure that ethical standards are adhered to, the study investigator developed a written
plan of study, as suggested by Krueger and Casey (2009), and submitted it to the Institutional
Review Board (IRB) at the University of Southern California (USC). The IRB’s feedback was
incorporated prior to the on-site interaction with program administrators at TC. Also, at the
inaugural meeting the purposefully selected members of the TC administration team were
informed of their right to privacy assuring them that any data collected would be kept
confidential. Permission to audio record was solicited prior to any interviews being conducted.
Finally, the TC program administrators selected for the study were asked to sign an informed
consent form. Throughout the study the program administrators were reminded that their
participation is voluntary and they have an option to withdraw from the study at any time. Once
the study has been concluded, as a token of appreciation for their time and contribution, all study
participants at TC were sent a photo book capturing the natural beauty of the five national parks
in Utah.

GENDER GAP IN COMPUTER SCIENCE 99
Appendix F: Integrated Implementation and Evaluation Plan
Implementation and evaluations section suggests a practical approach to designing and
measuring the efficacy of a program used to implement recommendations. As many components
of this program are already in place at TC, the value of this section is to develop a
comprehensive approach that is effective, clear, and measurable. The New World Kirkpatrick
Model (New Model) is used to implement and evaluate the effectiveness of program in support
of the organizational goals (Kirkpatrick & Kirkpatrick, 2016a). The four levels of the New
Model aid in clarifying the organizational goals (Level 4), support the monitoring and
reinforcement of the critical behaviors needed to achieve these goals (Level 3), and identify the
appropriate KMO based learning practices (Level 2) based on the levels of engagement and
perceived relevance expressed by the CS administrators (Level 1).
Implementation and Evaluation Framework
Based on the original work by Dr. Don Kirkpatrick, the New Model still retains the four
levels of the original implementation and evaluation framework: Reaction (Level 1), Learning
(Level 2), Behavior (Level 3), and Results (Level 4). Yet, the primary purpose of the New Model
is to emphasize the business values that implementation and evaluation plan delivers as well as
to highlight the importance of the Behavior and Results levels of the framework (Kirkpatrick &
Kirkpatrick, 2016a). As a result, the New Model presents the four levels in reverse, i.e., initially
focusing on the results the organization is looking to attain in Level 4, and then identifying the
behaviors stakeholders need to demonstrate for achieving these results in Level 3. Learning and
Reaction, Levels 2 and 1 respectively, are then presented in support of critical behaviors in Level
3. By following this approach, the New Model improves the effectiveness of the framework and
further clarifies the purpose of each level (Kirkpatrick & Kirkpatrick, 2016b).
GENDER GAP IN COMPUTER SCIENCE 100
Level 4: Results and Leading Indicators
A variety of benchmarks and outcomes are used by educational leaders to measure
organizational performance and to identify areas needing further investment (Dowd, 2005).
Performance enhancing initiatives are then implemented and evaluated to ensure the aspired
organizational goals are being achieved (Kirkpatrick & Kirkpatrick, 2016a). Table F1 identifies
internal and external outcomes, used as key performance indicators by CS administration team at
TC, to assess the progress in reducing the gender gap in computer science program.
Table F1

Outcomes, Metrics, and Methods for External and Internal Outcomes

Outcome Metric(s) Method(s)
Internal Outcomes
Increase number of female
applicants being selected for
admission in computer
science program at TC.
KPI – Annual college
admission report breakdown
by gender
Annual review of student
admission criteria for
computer science program
with Office of Admissions
personnel.
Increase number of female
undergraduate students
graduating from computer
science program at TC.
KPI – Annual college
graduation report breakdown
by gender
Annual assessment of
diversity focused initiatives
that provide academic and
social support to female
students.
Improve organizational
sentiment regarding the
importance of reducing the
gender gap in computer
science program at TC.
Annual assessment of
computer science faculty
engagement in diversity
focused initiatives.
Monthly meetings with small
group of faculty to reinforce
and encourage support for the
gender diversity focused
initiatives at TC.
GENDER GAP IN COMPUTER SCIENCE 101
Increase number of female
faculty being hired in
computer science program at
TC.
KPI – Annual college hiring
report breakdown by gender.
Assign diversity champions to
the computer science program
Faculty Hiring Committee at
TC.
External Outcomes
Increase awareness external
stakeholders have of the
gender focused initiatives
being supported at TC.
Annual level of corporate
support for diversity focused
initiatives.
Monthly meetings with small
group of industry stakeholders
to communicate school’s
diversity focused priorities
and to provide progress
updates.

The number of female applicants to CS program at TC indicates both the school’s ability
to attract a broad spectrum of students as well as its ability to market the success of diversity
focused initiatives in the program. The number of female students graduating from CS program
measures whether school’s cultural environment is equally conducive to female and male
students dealing with a rigor of the academic curriculum. The number of faculty involved with
diversity focused initiatives, especially those focusing on the gender diversity, indicates the
organizational sentiment towards achieving a more diverse computer science program at school.
Finally, an interaction with and support from the external stakeholders allows CS administrators
to account for the program’s achievements as well as to solicit additional feedback and
recommendations.
Level 3: Behavior
Critical behaviors. As stewards of the program, the CS leadership team is responsible
for achieving organizational goals, such as diversity and inclusion. According to Kirkpatrick and
GENDER GAP IN COMPUTER SCIENCE 102
Kirkpatrick (2016a), critical behaviors are specific actions that when performed consistently
ensure the organizational goals defined in Level 4 are achieved. Schein (2010) further adds that
individuals who are unable to consistently exhibit critical behaviors needed to support
organizational goals will negatively impact organizational performance. Table F2 shows critical
behaviors that CS leadership team needs to demonstrate to achieve the espoused outcomes
described in Level 4.
Table F2

Critical Behaviors, Metrics, Methods, and Timing for Evaluation
Critical Behavior Metric(s) Method(s) Timing
1). CS administration
team at TC to promote
gender diversity
focused initiatives and
collect feedback from
internal and external
stakeholders on
progress being made.
Number of internal
and external
meetings where
topics related to
improving the gender
diversity are included
on the agenda and
discussed.
Group discussions to
monitor and reinforce
importance of
achieving a gender
balanced environment
in computer science
program at TC.
Monthly
2). Members of the CS
administration team at
TC to sponsor or co-
sponsor a gender
diversity focused
initiative.
All top priority
diversity focused
initiatives to have a
diversity champion
who is a member of
CS administration
team.
Reinforce the
importance of gender
diversity in computer
science program to the
rest of the
organization by
becoming a diversity
champion.
Annual process for
CS administrators to
champion a gender
diversity focused
initiative.
GENDER GAP IN COMPUTER SCIENCE 103
3). CS administration
team to allocate
financial and human
capital resources to the
initiatives supporting
the gender diversity in
computer science
program at TC.
Itemized dollar
investment by TC in
gender diversity
focused initiatives.
Reinforce the
importance of gender
diversity in computer
science program to the
rest of the
organization by
providing budget
allocation to
prioritized diversity
focused initiatives.
Annual budgeting
process

The CS administration team needs to promote gender diversity as a critical element
leading to the program’s academic excellence. Also, by personally leading initiatives that
promote gender diversity and by allocating resources to such initiatives, CS administrators signal
to the rest of the organization the value being placed on achieving diversity in the computer
science program.
Required drivers. Support of critical behaviors in Level 3 is facilitated through the
processes or drivers that reinforce, encourage, and monitor espoused performance (Kirkpatrick &
Kirkpatrick, 2016a). Table F3 identifies key drivers, their relationship to critical behaviors, and
the timing of each driver that facilitate the gender diversity focused initiatives in computer
science program.

GENDER GAP IN COMPUTER SCIENCE 104
Table F3
Required Drivers to Support Critical Behaviors

Method(s) Timing
Critical
Behaviors
Supported
Reinforcing
Enable CS administrators to conduct internal
information campaign about the academic and social
benefits of gender diversity in the computer science
program.
Monthly 1, 2
Allow CS administrators to enhance admission policies
that extend the pool of candidates being considered for
admission
Annually 1, 2
Encouraging
Support CS administrators’ efforts to sponsor and
allocate resources to initiatives that reduce the gender
gap in computer science programs.
Annually 1, 2, 3
Provide CS administrators with opportunities to engage
in diversity focused discussions with educators in other
college STEM disciplines where gender parity has been
achieved.
Quarterly 1, 3
Monitoring
GENDER GAP IN COMPUTER SCIENCE 105
Evaluate impact of gender diversity focused initiatives
with respect to achieving organizational performance
goals

Monitor CS administrator’s engagement with internal
and external stakeholders who advocate for a greater
gender diversity in computer science program.
Semiannually

Quarterly
1, 2, 3

1

Critical behaviors are reinforced by the on-going diversity efforts to promote and support
initiatives that reduce the gender gap. Enhancing admission policies to include non-traditional
candidates, such as those with little or no programming experience, can also reinforce the
importance of having student diversity in the program. Sponsoring and personally leading
initiatives to reduce the gender gap in CS could encourage other internal stakeholders to follow.
Encouraging meetings with educators from other STEM fields where gender parity has been
achieved can incentivize CS faculty and staff to adapt best practices in their professional work.
Finally, actively monitoring faculty’s involvement and support of the diversity focused initiatives
and regularly assessing impact of such initiatives can be a significant contributor to attaining
critical behaviors needed to reduce the gender gap in CS program.
Organizational support. Organizational goals of higher education institutions are based
on the cultural values and beliefs of that institution (Kezar, 2001). Since diversity is one of the
core institutional pillars at TC, college leaders need to provide guidance, support, and resources
to the CS program administration team’s efforts to reduce the gender gap in computer science.
First, institutional leaders need to ensure that diversity and inclusion remain a priority across all
disciplines and all programs taught at the school. Gender diversity focused initiatives in the CS
program and their success need not only to be acknowledged by the institutional leaders at TC,
but also exemplified to other programs in the school. Second, TC leaders need to permit CS
GENDER GAP IN COMPUTER SCIENCE 106
administrators to implement policies and procedures the team feels are important to achieving
student diversity in the program. Having confidence in members of the CS administration team
empowers them to implement novel and innovative solutions without fear of being censured if
any specific gender diversity initiatives fail to produce expected results. Institutional leaders at
TC also need to support CS administration’s diversity focused efforts by providing resources
needed to develop, implement, and evaluate initiatives that reduce the gender gap in CS.
Consistently allocating budget to diversity focused initiatives is imperative for such initiatives to
succeed.
Level 2: Learning
Learning goals. All learning goals need to be clear, specific, and relevant to the
organizational objectives (Clark & Estes, 2008). To be effective these goals also need to
facilitate development of the critical behaviors of Level 3 (Kirkpatrick & Kirkpatrick, 2016a).
By implementing the recommended solutions discussed earlier, CS administrators at TC will
continue to pursue the following learning goals that have contributed to the program’s success in
reducing the gender gap in computer science:
1. Recognize the social and cultural factors that contribute to the gender gap in computer
science programs (Declarative)
2. Learn how to develop a plan for implementing diversity focused initiatives that improve
gender parity in computer science program (Procedural)
3. Recognize own attitudes and potential biases towards women pursuing degrees in
computing science (Metacognitive)
4. Develop confidence in own abilities to develop and implement diversity focused
initiatives to reduce the gender gap in computer science program (Self-efficacy)
GENDER GAP IN COMPUTER SCIENCE 107
5. Value the achievement of reducing the gender gap in computer science program (Value)
Program. Computer science administrators at TC need to participate in diversity
focused learning program to continue supporting initiatives that reduce the gender gap in
computer science program. According to Kezar (2001), continual learning is often needed to
sustain organizational changes in higher education institutions. Advancing diversity focused
knowledge, skills, and motivational factors of the CS administrators can be accomplished
through a combination of formal training and informal learning methods. Formal training
includes both online and live sessions focused on improving gender diversity and inclusion in the
computer science program. Training, such as effects of implicit biases, already offered at TC is
an example of formal training. Informal learning refers to the training where individuals
determine the content of interest, resources used for learning, and their time commitment
(Kirkpatrick & Kirkpatrick, 2016a). Reading scholarly articles, watching educational videos,
and having diversity focused discussions with colleagues and external program stakeholders are
just some of the examples of the informal learning. It is expected that informal learning will
constitute a significant portion of the diversity focused training incurred by the members of the
CS administration team at TC.
Evaluation of the Components of learning. Kirkpatrick and Kirkpatrick (2016a) have
identified five elements needed to develop critical behaviors supporting the attainment of the
organizational goals. These are knowledge, skills, attitude, confidence, and commitment. Table
F4 identifies recommended evaluation method for each of these components of learning.

GENDER GAP IN COMPUTER SCIENCE 108
Table F4
Evaluation of the Components of Learning for the Program.

Method(s) or Activity(ies) Timing
Declarative Knowledge “I know it.”
Group discussions focusing on social and cultural factors
contributing to the gender gap in computer science programs.
During diversity training
(Formative)
During management meetings
(Summative)
Procedural Knowledge “I can do it right now.”
Structure social environment to provide support for students
with no prior programming experience.
During Undergraduate
Review Committee
During management meetings
Attitude “I believe this is worthwhile.”
Group discussions on the importance the external stakeholders
place on achieving the gender parity in computer science
education
During diversity training
(Formative)
During management meetings
(Summative)
GENDER GAP IN COMPUTER SCIENCE 109
Confidence “I think I can do it on the job.”
Group discussions highlighting school’s diversity focused
initiatives contributing to recent success in attracting and
graduating female students without sacrificing academic rigor
in the program.

During management meetings
Discussions with educators from other STEM disciplines
where gender parity has been achieved.
Individual ad hoc meetings
Commitment “I will do it on the job.”
Individually developed action plan to champion diversity
focused initiatives.
Annual performance review

Level 1: Reaction
Reaction of the participants to the performance improvement program plays a critical role
in measuring program effectiveness (Clark & Estes, 2008). According to Kirkpatrick and
Kirkpatrick (2016a), there are three components of Level 1 that need to be evaluated. They are
the level of engagement by participants, relevance of the material presented to the one’s
functional role, and the overall level of satisfaction derived by the participants from the
performance improvement program. Table F5 below identifies methods used to assess the
reaction of CS administrators as well as their level of engagement during the group sessions and
individual discussions emphasizing the importance of diversity in computer science education.

GENDER GAP IN COMPUTER SCIENCE 110
Table F5
Components to Measure Reactions to the Program.
Method(s) or Tool(s) Timing
Engagement
Training instructor’s observations On-going
Meeting leader’s observations On-going
Relevance
Diversity training assessment survey Post training
School’s diversity initiatives rating survey Annual
Customer Satisfaction
Diversity training assessment survey Post training
School’s diversity initiatives rating survey Annual

Evaluation Tools
GENDER GAP IN COMPUTER SCIENCE 111
Evaluation tools are a key component of assessing performance of the learning programs.
The type of the learning program and the timing of evaluation determine what components of the
program can be evaluated and which evaluation tools are appropriate to use (Kirkpatrick &
Kirkpatrick, 2016a). For in-class training, surveys focusing on Level 1 (Reaction) and Level 2
(Learning) are typically done immediately following the learning program since impact of Level
3 (Behavior) and Level 4 (Results) takes time to take effect. For informal learning, evaluation
tools are typically based on the measurement of the key performance indicators of the
organizational effectiveness and are typically reviewed on the semi-annual or annual basis
(Kirkpatrick & Kirkpatrick, 2016a).
Immediately following the program implementation. Formal diversity training at TC is
strongly advised to all new and current members of the CS faculty and administration. It has
been cited by all members of the CS administration participating in this study as an important
resource for positively influencing their diversity knowledge and motivational factors. Diversity
Training Assessment Survey, similar to the one shown in Figure F1, can be used to evaluate
Level 1 as well as Level 2 components of the New Model immediately following the diversity
training session.

GENDER GAP IN COMPUTER SCIENCE 112
Figure F1. Diversity Training Assessment Survey
Instructions: Please indicate your level of agreement with the statements listed below.

Strongly Strongly
Disagree Agree
1. Topics included in this diversity training
program were interesting to me.

2. Topics focused on the gender gap in
computer science were easy for me to follow.

3. Topics focused on the gender gap in
computer science will help me do my job better.

4. After each diversity topic in the program we
discussed how to apply what was learned.

5. I am committed to applying newly acquired
diversity focused knowledge and skills in my
current role at TC.

6. I feel more confident in being able to lead the
diversity focused initiatives at TC.

7. I believe this diversity training program will
be useful in reducing the gender gap in computer
science at TC.

8. I would recommend this diversity training
program to my colleagues at TC.
1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

1 2 3 4 5 6 7

Delayed for a period after the program implementation. Delayed evaluation is comprised of
two separate components, depending on whether the diversity focused training was done in-class
GENDER GAP IN COMPUTER SCIENCE 113
or performed individually on the informal basis. Approximately six weeks after the formal, in-
class training is completed, an evaluation survey similar to the one shown in Figure F2 can be
distributed to the class participants.
Figure F2. Diversity Training Evaluation Survey
• What did you like most about this diversity training? ________________________
• What aspects of the diversity training could be improved? ________________________
• Were topics covered in this training relevant to your work? ________________________
• How do you hope to change your practice because of this training?
________________________
• What additional diversity training to help you with your practice would you like to have
in the future? ________________________

Evaluation of Levels 3 and 4 of the New Model, which measures whether the transfer of
learning to the participants’ professional practice is taking place and organizational outcomes are
being attained, is the primary focus of the survey (Kirkpatrick & Kirkpatrick, 2016a).
Informal learning typically lacks formal learning objectives. It takes place over a period
of time, and usually occurs unconsciously (Kirkpatrick & Kirkpatrick, 2016a). As such,
assessment of the informal learning is to be performed annually, as part of the employee review
process. Diversity Initiatives Survey, Figure F3, can be used to evaluate four levels of the
informal diversity learning process.

GENDER GAP IN COMPUTER SCIENCE 114
Figure F3. Diversity Initiatives Survey
Do you feel informal learning is an appropriate way to advance one’s skills and knowledge
needed to lead gender diversity focused initiatives at TC? (Y/N). Please explain.

In the last 12 months, have you used informal learning resources to study topics on gender
diversity? (Y/N)
***If no, please select all that apply from the list below:
Not interested in informal learning methods
Informal learning does not work for me
I have no need to study diversity related topics
I am too busy during work hours to focus on informal learning
I have not found any diversity focused topics of interest
Other. Please specify _____________________________

***If yes, from the list below, in order of priority please identify resources that you use to
further your diversity and inclusion focused knowledge
Reading a scholarly article/book
Watching an educational program/video
Interacting with other CS administrators at TC
Interacting with faculty at TC
Interacting with students at TC
Interaction with external (non-TC affiliated) stakeholders
Other. Please specify _____________________________
GENDER GAP IN COMPUTER SCIENCE 115

If applicable, what were some of the major gender diversity related concepts you learned from
the informal learning sources in the last 12 months? Please specify.

Based on your informal learning experience, do you feel that you are better prepared to lead
initiatives that further reduce the gender gap in computer science at TC? Please explain.

Please list gender diversity focused initiatives at TC or outside activities you have personally
been involved with in the last 12 months.

Please specify diversity related topics you intend to learn more about in the upcoming period.

Data Analysis and Reporting
Data analysis and reporting is an important step in guiding educational decision-making
(Marsh, 2012). It monitors progress of the performance improvement training programs and
supports leadership efforts in attaining organizational goals (Kirkpatrick & Kirkpatrick, 2016a).
Internal and external outcomes discussed in Level 4 of the New Model can only be achieved
when there is evidence that newly acquired knowledge and motivational factors are contributing
to behavioral changes that improve organizational results (Clark & Estes, 2008; Kirkpatrick &
Kirkpatrick, 2016a). To measure progress in reducing the gender gap in the computer science
program and to assess the effectiveness of the training interventions at TC, a data analysis and
reporting approach similar to the Equity Scorecard discussed by Down and Bensimon (2014) can
be implemented. The scorecard at TC can be compiled and reviewed on the annual basis during
GENDER GAP IN COMPUTER SCIENCE 116
the budgeting and strategic planning discussions. It would provide a comparison of the number
of undergraduate applications to the program, students’ progress for each year, and the
graduation rates for female and male students. It would also the diversity focused faculty
recruiting efforts in the program as well as the summarizes student feedback from the individual
and group discussions held throughout the year. This approach would allow CS administrators
to focus on the areas that need additional investments. For example, if the number of
applications from female candidates is low in comparison to male applicants, additional
information session to highlight the program’s student diversity could be set up. If there is an
increase in female student dropout rate, a comprehensive analysis of the academic and social
environment by the members of the TC administration team could be initiated. An example of
the management report that would enable TC administrators to measure progress in reducing the
gender gap in CS shown in Figure F4.

GENDER GAP IN COMPUTER SCIENCE 117
Figure F4. Diversity Scorecard at TC

GENDER GAP IN COMPUTER SCIENCE 118

Summary
New World Kirkpatrick Model guides implementation and evaluation planning of the
study’s recommendations. It provides a framework that (a) enables leadership teams to develop
programs to address any knowledge and motivational gaps within their organization, (b) ensures
critical behaviors needed to achieve organizational goals are sustained, and (c) assesses the
effectiveness of the program (Kirkpatrick & Kirkpatrick, 2016a). Level 1 of the New Model
evaluates the level of engagement, relevance, and satisfaction and is helpful in developing and
enhancing the diversity focused learning programs at TC. Level 2 relies on the KMO type of
analysis to identify the knowledge and motivational gaps that need to be addressed to improve
the gender balance in computer science program. Critical behaviors of Level 3, when perform
consistently, ensure that the gender diversity focused outcomes of the program are achieved.
Level 4 lists aspired outcomes and indicates whether the expected results of the diversity focused
learning programs have been achieved. Evaluating the return on expectations (ROE) is an
important measure of success supported by the New Model. ROE indicates how well the
program results match the set expectations and ultimately determine the effectiveness of the
learning program. Only those diversity focused initiatives that meet or exceed expected
outcomes receive further investment and support from the CS program leaders.

Abstract (if available)

Abstract A computer science program leadership team at Technical College (TC) realized that to build a community of future computer science leaders, a primary mission of the program, they need to achieve a more diverse and inclusive environment at TC. For over a decade several initiatives focused on enhancing the organizational structure, work processes, and program culture were implemented resulting in TC becoming one of the more gender balanced top ranking computer science (CS) programs in the country. The purpose of this case study is to focus on the role that CS program administrators play in addressing the gender gap in computer science program at TC.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Girls in cybersecurity: an evaluation study

PDF

Developing a computer science education program: an innovation study

PDF

Low teacher retention rates in private schools

PDF

Evaluation study: building teacher efficacy in K8 computer science integration

PDF

The cyclic lack of female engineers in computer science: an evaluation study

PDF

Wicked problems for high poverty schools: an improvement study in Texas

PDF

Labor displacement: a gap analysis an evaluation study addressing professional development in a small business environment

PDF

K-12 dress codes and gender equity: an examination of policy and practice of dress code implementation

PDF

Latinas’ self-efficacy in computer science and potential factors undergirding their perspective

PDF

Social emotional learning curriculum implementation: an evaluation study of fidelity

PDF

Child welfare turnover: an evaluation study of work-life balance practices

PDF

Developing a culture of teacher collaboration in middle school

PDF

Explicit instruction’s impact on the student achievement gap in K-12 English language learners

PDF

First-year retention: Community college students -- a gap analysis

PDF

Factors impacting the effectiveness of mentor teachers in a national teacher residency

PDF

Bridging the gap of gender equity in entrepreneurship: an evaluation study

PDF

The mentorship of instructors and its impact on computer science interest among middle school girls: an evaluation study

PDF

High school single-gender science classrooms, minority females, and perceptions of self-efficacy

PDF

A case study of promising practices mentoring K-12 chief technology officers

PDF

Leadership and the impact on organizational citizenship behaviors: an evaluation study

Asset Metadata

Creator Levinzon, Eugeny (author)

Core Title Role of college program administrators in addressing gender gap in computer science

School Rossier School of Education

Degree Doctor of Education

Degree Program Organizational Change and Leadership (On Line)

Publication Date 07/26/2018

Defense Date 07/26/2018

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

Tag Computer Science,gender gap,OAI-PMH Harvest,stereotypes

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Mora-Flores, Eugenia (committee chair), Freking, Frederick (committee member), Maddox, Anthony (committee member)

Creator Email gene.levinzon@gmail.com,levinzon@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-27357

Unique identifier UC11670845

Identifier etd-LevinzonEu-6504.pdf (filename),usctheses-c89-27357 (legacy record id)

Legacy Identifier etd-LevinzonEu-6504.pdf

Dmrecord 27357

Document Type Dissertation

Format application/pdf (imt)

Rights Levinzon, Eugeny

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