University of Southern California Dissertations and Theses

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

(USC Thesis Other)

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

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Running head: THE CYLIC LACK OF FEMALE ENGINEERS IN COMPUTER
SCIENCE
1
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE:
AN EVALUATION STUDY
by
Justin A. Cox
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
December 2019
Copyright 2019 Justin A. Cox
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 2
DEDICATION
With all my heart and strength for my daughters - Alexa and Addison. I hope this serves to be
another stone in the path that will enable you to achieve your dreams in any field.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 3
ACKNOWLEDGMENTS
First, to God the glory!
My sincerest gratitude for my dissertation chair, Dr. Leila Hasan. Your guidance and
motivation have been invaluable. Dr. Susanne Foulk, your skills were vital in this entire process,
and I do not know how I could have made it to the finish line without your dedication to student
success.
To Dr. John Burk, I certainly did not expect to end this journey with a retired Army
General on my chair! Your intelligence, professionalism, and guidance helped focus my studies
while inspiring to keep everything to the highest standards. Thank you so much for the time and
resources you have put into this research, it has truly been an honor.
Thank you to Dr. Eugenia Mora-Flores. Your unique perspective and ideas helped this
complex study find its footing and space in a complex global issue. Your thorough reviews and
critical questions helped sharpen and solidify this study.
To my wife, Ashley. Thank you for your patience and support throughout this entire
process. Thank you for never complaining about sleeping on the couch during late-night class
sessions, sleeping alone during late-night writing sessions, dealing with interrupted vacations,
and doubling up on household duties. You are my rock!
Lastly, thanks to all my colleagues in Cohort 8. It has been a truly fun and memorable
experience I will always cherish. Everyone’s altruism and passion for their studies have inspired
me, and I am humbled to be part of this cohort. Fight on!
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 4
TABLE OF CONTENTS
DEDICATION ................................................................................................................................ 2
ACKNOWLEDGMENTS .............................................................................................................. 3
LIST OF FIGURES ...................................................................................................................... 10
LIST OF TABLES ........................................................................................................................ 11
CHAPTER 1: INTRODUCTION TO THE PROBLEM OF PRACTICE .................................... 12
ORGANIZATIONAL CONTEXT AND MISSION .............................................................. 13
ORGANIZATIONAL GOAL ................................................................................................. 13
RELATED LITERATURE ..................................................................................................... 13
IMPORTANCE OF THE EVALUATION ............................................................................. 14
DESCRIPTION OF STAKEHOLDER GROUPS.................................................................. 14
STAKEHOLDER GROUPS’ PERFORMANCE GOALS .................................................... 15
ORGANIZATIONAL MISSION ........................................................................................... 15
STAKEHOLDER GROUP FOR THIS STUDY .................................................................... 16
PURPOSE OF THE PROJECT AND QUESTIONS ............................................................. 16
METHODOLOGICAL FRAMEWORK ................................................................................ 17
ORGANIZATION OF THE PROJECT ................................................................................. 17
CHAPTER 2: REVIEW OF THE LITERATURE ....................................................................... 18
ROLE OF THE STAKEHOLDER GROUP OF FOCUS ...................................................... 19
STAKEHOLDER KNOWLEDGE, MOTIVATION AND ORGANIZATIONAL
INFLUENCES .............................................................................................................................. 19
KNOWLEDGE AND SKILLS ......................................................................................... 19
KNOWLEDGE INFLUENCES.................................................................................. 20
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 5
STEREOTYPE THREAT AND AMBIENT BELONGING. .................................... 21
MOTIVATION INFLUENCES........................................................................................ 23
GOAL-ORIENTATION THEORY. ........................................................................... 24
SELF-EFFICACY THEORY ..................................................................................... 25
ORGANIZATIONAL INFLUENCES ............................................................................. 27
MANAGERS MUST CREATE AN ATMOSPHERE THAT EMPOWERS AND
EMBRACES WOMEN. ............................................................................................................... 28
SHORTAGE OF FEMALE CANDIDATES WITH COMPUTER SCIENCE
DEGREES..................................................................................................................................... 29
IMPLICIT BIAS IN THE RECRUITING AND HIRING PROCESS. ...................... 29
EXECUTIVE LEADERSHIP AT QUICKCHIP HAS NOT PLACED ENOUGH
EMPHASIS ON RECRUITING AND RETAINING FEMALE SOFTWARE ENGINEERS.... 32
CONCLUSION ....................................................................................................................... 34
CHAPTER 3: METHODS............................................................................................................ 35
PURPOSE OF THE PROJECT AND QUESTIONS ............................................................. 35
CONCEPTUAL FRAMEWORK: THE INTERACTION OF STAKEHOLDERS’
KNOWLEDGE AND MOTIVATION AND THE ORGANIZATIONAL CONTEXT .............. 35
DATA COLLECTION ........................................................................................................... 36
EXPLANATION FOR CHOICES ................................................................................... 37
PARTICIPATING STAKEHOLDERS .................................................................................. 37
SURVEY SAMPLING CRITERIA AND RATIONALE ................................................ 38
CRITERION 1. ........................................................................................................... 38
CRITERION 2. ........................................................................................................... 38
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 6
SURVEY SAMPLING STRATEGY AND RATIONALE .............................................. 38
INTERVIEW SAMPLING CRITERIA AND RATIONALE .......................................... 39
CRITERION 2. ........................................................................................................... 39
INTERVIEW SAMPLING STRATEGY AND RATIONALE ....................................... 39
SAMPLING STRATEGY AND TIMELINE ......................................................................... 39
QUANTITATIVE DATA COLLECTION AND INSTRUMENTATION ............................ 40
SURVEYS ........................................................................................................................ 40
SURVEY INSTRUMENT. ......................................................................................... 40
SURVEY PROCEDURES. ........................................................................................ 42
QUALITATIVE DATA COLLECTION AND INSTRUMENTATION............................... 43
INTERVIEWS .................................................................................................................. 43
INTERVIEWEES. ...................................................................................................... 43
INTERVIEW PROTOCOL ........................................................................................ 43
INTERVIEW PROCEDURES ................................................................................... 44
DATA ANALYSIS ................................................................................................................. 45
CREDIBILITY AND TRUSTWORTHINESS ...................................................................... 45
VALIDITY AND RELIABILITY .......................................................................................... 46
ETHICS................................................................................................................................... 47
CONCLUSION ....................................................................................................................... 48
CHAPTER 4: RESULTS ............................................................................................................. 49
RESULTS AND FINDINGS FOR KNOWLEDGE GAPS ................................................... 50
VALIDATED KNOWLEDGE GAPS .............................................................................. 50
FACTUAL KNOWLEDGE........................................................................................ 50
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 7
METACOGNITIVE KNOWLEDGE. ........................................................................ 53
KNOWLEDGE FINDINGS ............................................................................................. 55
RESULTS AND FINDINGS FOR MOTIVATION GAPS ................................................... 55
VALIDATED MOTIVATION GAPS .............................................................................. 55
GOAL-ORIENTATION. ............................................................................................ 55
SELF-EFFICACY....................................................................................................... 57
MOTIVATION FINDINGS ............................................................................................. 59
RESULTS AND FINDINGS FOR ORGANIZATION GAPS .............................................. 59
VALIDATED ORGANIZATION GAPS ......................................................................... 59
CULTURAL MODEL INFLUENCES....................................................................... 59
CULTURAL SETTING INFLUENCES. ................................................................... 63
SCORING AND CORRELATION ........................................................................................ 67
SCORING ......................................................................................................................... 67
EDUCATION LEVEL................................................................................................ 68
ROLE WITHIN THE COMPANY. ............................................................................ 69
AGE. ........................................................................................................................... 70
RACE. ......................................................................................................................... 71
CONCLUSION ....................................................................................................................... 72
CHAPTER 5: SOLUTIONS, IMPLEMENTATION, AND EVALUATION .............................. 73
INTRODUCTION AND OVERVIEW .................................................................................. 73
RECOMMENDATIONS FOR PRACTICE TO ADDRESS KMO INFLUENCES ............. 73
KNOWLEDGE RECOMMENDATIONS ....................................................................... 73
INTRODUCTION ...................................................................................................... 73
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 8
REDUCING IMPLICIT BIAS IN COMPUTER SCIENCE ENGINEERS .............. 76
INCREASING PERFORMANCE THROUGH REDUCTION OF STEREOTYPE
THREAT. ...................................................................................................................................... 77
MOTIVATION RECOMMENDATIONS ....................................................................... 78
INTRODUCTION ...................................................................................................... 78
CHALLENGES IN RETAINING FEMALE ENGINEERING TALENT CAUSED
BY UNSUPPORTIVE CULTURE............................................................................................... 79
EMPOWERING AND TRAINING MANAGERS TO INSTILL A CULTURE THAT
ATTRACTS WOMEN. ................................................................................................................ 80
ORGANIZATION RECOMMENDATION ..................................................................... 81
INTRODUCTION ...................................................................................................... 81
INTEGRATED IMPLEMENTATION AND EVALUATION PLAN .................................. 84
IMPLEMENTATION AND EVALUATION FRAMEWORK ....................................... 84
ORGANIZATIONAL PURPOSE, NEED AND EXPECTATIONS ............................... 85
LEVEL 4: RESULTS AND LEADING INDICATORS .................................................. 86
LEVEL 3: BEHAVIOR .................................................................................................... 87
CRITICAL BEHAVIORS .......................................................................................... 87
REQUIRED DRIVERS .............................................................................................. 90
ORGANIZATIONAL SUPPORT .............................................................................. 91
LEVEL 2: LEARNING .................................................................................................... 92
LEARNING GOALS .................................................................................................. 92
TRAINING PROGRAM ............................................................................................ 93
EVALUATION OF THE COMPONENTS OF LEARNING .................................... 93
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 9
LEVEL 1: REACTION..................................................................................................... 94
EVALUATION TOOLS ................................................................................................... 95
IMMEDIATELY FOLLOWING THE PROGRAM IMPLEMENTATION ............. 95
DELAYED FOR A PERIOD AFTER THE PROGRAM IMPLEMENTATION...... 96
DATA ANALYSIS AND REPORTING.......................................................................... 96
SUMMARY ............................................................................................................................ 96
FUTURE RESEARCH ........................................................................................................... 97
CONCLUSION ....................................................................................................................... 98
REFERENCES ............................................................................................................................. 99
APPENDIX A: COPY OF THE EMAIL SENT TO THE SOUTHWESTERN US QCHEW
BOARDS .................................................................................................................................... 106
APPENDIX B: COPY OF THE INFORMED CONSENT ........................................................ 107
APPENDIX C: COPY OF SURVEY QUESTIONS .................................................................. 109
APPENDIX D: PROPOSED IMMEDIATE FEEDBACK SURVEY QUESTIONS ................ 115
APPENDIX E: DELAYED FEEDBACK SURVEY QUESTIONS .......................................... 119
APPENDIX F: PROPOSED DATA ANALYSIS CHART ....................................................... 121
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 10
LIST OF FIGURES
Figure 1 - Results of Survey Question 9 ....................................................................................... 51
Figure 2 - Results of Survey Question 10 ..................................................................................... 52
Figure 3 - Results of Survey Question 11 ..................................................................................... 52
Figure 4 - Results of Survey Question 2 ....................................................................................... 54
Figure 5 - Results of Survey Question 8 ....................................................................................... 56
Figure 6 - Results of Survey Question 3 ....................................................................................... 57
Figure 7 - Results of Survey Question 5 ....................................................................................... 61
Figure 8 - Results of Survey Question 6 ....................................................................................... 61
Figure 9 - Results of Survey Question 1 ....................................................................................... 62
Figure 10 - Comparison by category of the percentage of women in Computer Science ............ 63
Figure 11 - Results of Survey Question 7 ..................................................................................... 64
Figure 12 - Results of Survey Question 4 ..................................................................................... 66
Figure 13 - Overall Scores ............................................................................................................ 68
Figure 14 - Score by Highest Education Level Completed .......................................................... 69
Figure 15 - Score by Position........................................................................................................ 70
Figure 16 - Score by Age .............................................................................................................. 71
Figure 17 - Score by Race ............................................................................................................. 72
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 11
LIST OF TABLES
Table 1 - Knowledge Influence, Types, and Assessment ............................................................. 23
Table 2 - Motivation Influence, Types, and Assessment .............................................................. 27
Table 3 - Organizational Influence, Types, and Assessment ........................................................ 32
Table 4 - Sampling Strategy and Timeline ................................................................................... 40
Table 5 - Assumed Knowledge Influences with Recommendations ............................................ 74
Table 6 - Summary of Motivation Influences and Recommendations ......................................... 79
Table 7 - Summary of Organization Influences and Recommendations ...................................... 81
Table 8 - Outcomes, Metrics, and Methods for External and Internal Outcomes ........................ 86
Table 9 - Critical Behaviors, Metrics, Methods, and Timing for Evaluation ............................... 88
Table 10 - Required Drivers to Support Critical Behaviors ......................................................... 90
Table 11 - Evaluation of the Components of Learning for the Program ...................................... 94
Table 12 - Components to Measure Reactions to the Program .................................................... 95
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 12
CHAPTER ONE
INTRODUCTION TO THE PROBLEM OF PRACTICE
This evaluation research project addresses the problem of underrepresentation of females
in engineering positions in technology companies. This is a problem because according to the
Equal Employment Opportunity Commission (EEOC), in the technology industry, females make
up only 18% of the engineering workforce, while they make up 48% of the entire workforce
(2016). According to the National Science Foundation, females receive only 20% of the
Computer Science and Engineering degrees, which makes it more difficult for women to enter
the career field. However, the ratio of women in both the career field and Computer Science
degree programs has steadily increased over the past decade (2016). This problem is important
to address because according to the Bureau of Labor and Statistics there are currently 750,000
unfilled computer science and engineering jobs, with the number projected to rise 1,150,000 by
2024 (2016).
The lack of female engineers is systemic and cyclic. There is an underrepresentation of
women in the career field because of the underrepresentation in the academic arena. There is an
underrepresentation of women in computer science in higher education because of the perceived
and actual barriers caused by social threat against women in the courses (Murphy, Steele, &
Gross, 2007). These both lead to young girls who become disinterested in computer science and
engineering at a young age because American culture portrays the field as masculine and hostile
to women (Sadler, Sonnert, Hazari, & Tai, 2012). Until the elimination of barriers for women in
universities and workplaces, biases and stereotype threat will continue to exist and dissuade girls
from showing interest in Computer Science and Engineering and the gender gap will still exist.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 13

Organizational Context and Mission
The organization for this study is a semiconductor manufacturing corporation, which the
researcher gave the pseudonym, “QuickChip.” QuickChip is based in the Southwestern United
States, and relies heavily on STEM engineering for the fabrication of its computer chips.
QuickChip’s focus on computer chip manufacturing requires engineering efforts from thousands
of people to produce transistors less than 20nm, with software and electrical engineering to
program these chips.
QuickChip has struggled both with its own reputation working with female engineers,
recently settling a large class action lawsuit for over $10 million which alleged gender
discrimination, and recruiting female engineers in a talent pool facing a tremendous shortage of
female engineers. With the current talent shortage and ever-evolving requirements to fit smaller
and smaller transistors onto microchips, the need for engineers is greater than ever.
Organizational Goal
QuickChip is currently around 16% female engineers, with the goal to reach 50% by the
year 2030, increasing at least 1.5% each year, until it achieves equitable representation. This is
in addition to its goals of increasing its workforce capacity, meaning that QuickChip must both
recruit new female talent in addition to the male talent it is already bringing in, and retain its
existing talent.
Related Literature
The researcher broke down the review of literature into knowledge, motivation, and
organizational influences.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 14

Importance of the Evaluation
The shortages in STEM shown by the Bureau of Labor and Statistics (2016) in what the
EEOC (Equal Employment Opportunity Commission, 2016) report describes as America’s most
important career field for global competitiveness means the United States need to get as many
girls and young women excited about STEM as possible. According to the Microsoft (2017)
study of over 11,500 girls and women ages 11-30, the United States could regain the interest in
STEM of 60% of girls aged 11-17 if the field were gender balanced. Addressing the issue of
female underrepresentation helps against the STEM talent shortage in the United States and
increasing the nation’s global competitive. Most importantly, it allows for even greater
technological advances and innovations with the inputs and minds representative of the society
they benefit.
Description of Stakeholder Groups
QuickChip has three primary stakeholders for the purpose this study. The senior
management who are responsible for setting the culture and retaining talent in the organization,
the recruiters and hiring managers responsible for bringing in new talent, and the female
engineers and the center of the research.
The senior management of QuickChip are responsible to the shareholders of the
company. To ensure long term growth and viability of QuickChip, senior management are
interested in obtaining talented engineers from all demographics to ensure it can keep up with the
demands of the sector’s rapid growth. Less practical, but still important, is the focus of the tech
industry’s diversity initiatives. QuickChip must also show it is keeping pace with the industry,
less it should appear unwelcoming to female engineers. Lastly, to meet its goal of a
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 15

representative engineering workforce, QuickChip must bolster its retention efforts to stem the
losses while the recruiting team bring in new talent.
The recruiters and hiring managers are responsible for bringing in new talent into the
organization. QuickChip’s recruiters receive training on implicit bias and techniques to actively
recruit women to its engineering team. The recruiters struggle to recruit from a non-diverse pool
of candidates, and must also work on growing their own talent pipelines to meet the growing
demand and increasing targets for female engineers.
The female engineers at QuickChip are the core of the study. Most of the engineers have
a Bachelor’s degree and come from Southeast Asian countries. Despite the prestigious
reputations from working at QuickChip, they are in a struggle to change the landscape and
normalize women in computer science and engineering positions while maintain a comfortable
working life most professionals seek.
Stakeholder Groups’ Performance Goals
QuickChip’s stated performance goal is to increase female representation by 1.5% per
year, until it achieves equity. This goal represents modest annual increases, which acknowledges
the long-term strategy required to achieve its goal by filling the talent pool, which means
engaging girls from elementary into higher education (Hewlett & Luce, 2005). This does not
account for a “tipping point,” or exponential change, but this could just be meant to provide
simple targets which are easy to communicate.
Organizational Mission
QuickChip’s mission is to help make the future available to all by bringing the future to
all. This lofty, long-term mission demonstrates QuickChip’s desire to benefit society in the long-
term by being a positive change and catalyst. The performance goal coincides well, since they
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 16

are also serving as a catalyst to bring about positive change in the STEM field through its
commitment to increase the talent pool.
Stakeholder Group for this Study
There are two stakeholder groups in this study. The first is the current female engineers
at QuickChip, the second are senior leaders at Qualcomm. The primary focus on the research
will be on the engineers. Identifying the engineers as the key stakeholder is crucial, because
their experiences and perceptions of the onboarding process and culture are the reality that reflect
the metrics that QuickChip is trying to achieve. The senior leadership are secondary in this
study. The research will not interview them people directly, but will extrapolate performance
metrics based on feedback from the female engineers.
Purpose of the Project and Questions
The purpose of this project is to analyze and understand how the culture and policies of
QuickChip affect the female engineers who work there. The goal is to help QuickChip identify
issues that keep women from joining and staying in the organization, thus continuing the cycle,
by preventing the culture that excludes those who are in the minority of the population. The
questions that this project will ask are:
1. Do employees at the organization understand how the culture within the organization
impacts performance?
2. Do employees at QuickChip believe that a more diverse and welcoming environment
is better for them personally and the organization?
3. Is QuickChip’s senior management taking effective steps to create a culture where
women feel welcome and equal?
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 17
Methodological Framework
Clark and Estes’s (2008) Gap Analysis framework provides an analytical framework to
understand the problem and how it relates to the organizational goals and mission. The
researcher adapted this framework to suit the unique needs of this problem of practice where
QuickChip’s solutions rely on a slow-moving “domino effect” before it sees immediate results.
The study looks at the experience of current female engineers from the period of recruitment to
understand how recruiters affect QuickChip’s ability to bring in new female talent. The study
also examines female engineers’ experience with, implicit bias, stereotype threat, and ambient
belonging, since they are issues reported by many women across the industry, to see how it
affects QuickChip’s ability to retain its female engineering staff to meet its goals.
Organization of the Project
This dissertation consists of five chapters. Chapter 1 provided the reader with a synopsis
of the problem and how the researcher structured the study. It frames QuickChip’s struggles
within the industry’s struggle which contribute to societal and educational issues, created a cyclic
problem. Chapter 2 contains a literature review examining causes and rippling effects of the
gender gap organized into how the related to QuickChip’s gaps caused by knowledge,
motivation, and influence. Chapter 3 explains the data collection methodology and the
instruments used to collect that data. Chapter 4 presented the data the researcher collected and
analysis. Chapter 5 will discuss research-based solutions, implementation guidance, and
instruction on how to evaluate the effectiveness of those solutions.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 18

CHAPTER TWO
REVIEW OF THE LITERATURE
The computer science and engineering profession has a troubled history with women.
While other engineering fields and similar professions have seen marked increases in
representation of women, there has been a slow and steady decline in of women in Computer
Science (Equal Employment Opportunity Commission, 2016). A large reason for this decline is
the cyclic nature of how career fields evolve. Young girls lose interest in computer science
because there are so few women in the field (Microsoft Corporation, 2017). This leads to young
women who do not pursue computer science and engineering degrees in college (Sadler, Sonnert,
Hazari, & Tai, 2012). This leads to a continual input of talent where men outnumber women
four-to-one (Bureau of Labor and Statistics, 2016), causing the cycle of underrepresentation to
continue.
QuickChip also shares in this disparity, where only 20% of its engineering workforce is
comprised of women (2017). According to Murphy, Steele, and Gross (2007), this gender
imbalance leads to a culture afflicting women with stereotype and signaling threat when left
unaddressed, which leads to a toxic work environment where women do not feel like they
belong. QuickChip is not properly addressing the gender gap which is costing it in terms of
recruiting expenses, reputation, and a recent $19.5 million gender discrimination lawsuit
(Freeman, 2016). QuickChip’s failures to combat the negative culture make it more difficult to
recruit and retain female software engineering talent. The two biggest problems QuickChip’s
failure create are women who experience implicit bias and stereotype threat (Master, Cheryan, &
Meltzoff, 2016). This toxic environment affects QuickChip’s bottom line because it affects
female engineer performance through increased mental energy and hypervigilance due to
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 19

stereotype threat and increases recruiting costs through poor reputation while perpetuating the
cyclic shortage of female talent (Murphy, Steele, & Gross, 2007).
Role of the Stakeholder Group of Focus
QuickChip’s organizational goal is to achieve equitable representation of engineers,
which means it needs to recruit new talent while retaining its existing talent. Therefore, the
primary stakeholder group of focus are the female engineers at QuickChip. The secondary
groups are the senior leaders who drives the culture and policy which largely determine whether
the engineers join and stay in the organization.
Stakeholder Knowledge, Motivation and Organizational Influences
Knowledge and Skills
While organizations cannot fix the chronic shortage of female talent coming from the
universities to the talent pipeline, they can take steps inside their own organizations to create a
hospitable and inclusive atmosphere which encourages them to pursue the career during their
education (Master, Cheryan, & Meltzoff, 2016). Currently, the tech industry is male dominated,
with males outnumbering females more than two to one (Bureau of Labor and Statistics, 2016).
In large software engineering firms, this figure climbs to nearly four to one (Equal Employment
Opportunity Commission, 2016).
Inside large tech firms, senior leadership develops policy and drives the culture which
recruiters and hiring managers form a team to bring in talent. This impact the entire career
lifecycle from entrance into the talent pool, to recruitment, and ultimately until exit from the
organization. The culture within the organization can inspire girls and young women the enter
the career field with aspirations to work within a welcoming organization (Microsoft
Corporation, 2017). Typically, recruiters work full-time interfacing with either recruits or hiring
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 20

manager. The hiring manager ultimately has the authority on who to hire, but the recruiter is the
one responsible for sourcing candidates (McCuller, 2012). Senior leadership must ensure both
have the appropriate skills and motivation to effectively recruit female talent. Lastly, senior
leadership create the culture and policy that impact the lives of the female engineers who work
there. Senior leadership must have the knowledge and motivation to create this culture in the
workplace to achieve long-term success.
According to Clark and Estes (2008), a researcher can frame organizational issues into
performance gaps of knowledge, motivation, and the organization. This paper will examine the
impacts of knowledge and motivational influences on recruiters’ ability to attract female
candidates into a male dominated computer engineering workforce.
Knowledge influences. According Krathwohl (2002) and Rueda (2011), there are four
types of knowledge to consider when analyzing knowledge influence: factual, conceptual,
procedural, and metacognitive. According to Krathwohl (2002) and his recharacterization of
Bloom’s Taxonomy, factual knowledge is the basic element that one must be familiar with in
learning. In the context of recruiting, this means the recruiters must understand the facts of why
women avoid tech. According to Anderson (2013), conceptual knowledge is the how the facts
work together to create the environment. For senior leadership, it is important for them to
understand how large the issue of the lack of women in tech - both the causes and effects.
Procedural knowledge is knowing how to do something, not just the why (Krathwohl, 2002). For
senior leaders, it is important to understand how to attract, engage, grow, and retain women, to
market itself as female-friendly, and ensure it has an inclusive environment. Lastly, Krathwohl
(2002) describes metacognition as the awareness and knowledge of one’s own cognition. For
senior management, this means being aware of the potential future talent, the current talent pool
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 21

and pipeline, and the women who currently work at QuickChip. They must understand how
those women perceive the company, their potential, and the environment, to ensure a tailored
approach to that all feel welcomed as potential or current employees.
Stereotype threat and ambient belonging. One of the greatest factors affecting women
in the tech industry is stereotype threat. According to Murphy, Steele, and Gross (2007),
stereotype threat occurs when a class feels inflicted by a negative stereotype. It causes a hyper-
awareness of one’s surrounding causing the individual to put extra mental effort towards
questioning whether they are fitting in with those without a negative stereotype. This is a cyclic
issue, because the internal distractions decrease performance, raising the risk of reinforcing
negative stereotypes to both observers and the individual experiencing the stereotype threat
(Schmader, Johns, & Forbes, 2008).
Another factor for women seemingly avoiding careers in technology is that men have an
implicit bias towards other men, creating a hostile environment towards women causing them to
experience a lack of the phenomenon referred as “Ambient Belonging” (Cheryan, Plaut, Davies,
& Steele, 2009). Simply stated, it is human nature to seek out a safe and welcoming atmosphere.
When a person is notably different through either race, gender, culture, or other easily
identifiable characteristic, they are likely to experience uncomfortable situations, and therefore
avoid that atmosphere, especially one as intimate as the workplace.
How negative culture propagates. Stereotype threat and the lack of ambient belonging
are a factor in the lack of female applicants to computer engineering positions. It is important for
leadership to know that the stereotype threat and ambient belonging in computer science are not
just exclusive to the workplace, in fact, it starts much earlier. According to a study by Microsoft
Corp. (2017) of over 11,500 girls and women, most girls lose interest in computer science at the
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 22

age of 11, citing direct pressure to follow traditionally female professions like education and
healthcare and citing indirect social pressure of viewing computer science as a job for males.
Also, the threat lasts through college, with women only earning 20% of Computer Science
degrees (Equal Employment Opportunity Commission, 2016). Women in college who dropped
out of the degrees programs most often cited difficulties in fitting in and learning the material
and nearly all admitted to experience stereotype threat (Rydell, Rydell, & Boucher, 2010).
Counteracting stereotype threat and ambient belonging. While intuitively, it may seem
out of grasp for an organization to combat stereotype threat and encourage ambient belonging.
There are many proven ways for organizations to counteract the toxic culture to attract and retain
female computer scientists and engineers to their organizations:
1. Ensure the workplace is general neutral and free of cues that suggest coworkers or
management are unwelcoming to female candidates (Cheryan, Plaut, Davies, & Steele,
2009).
2. Create a culture of inclusion among the existing staff, including training on sexism and
inclusion (Logel, et al., 2009).
3. Communicate a message of diversity and inclusion, ensuring that coworkers’ value
everyone’s ideas regardless of background or identity (Purdie-Vaughns, Steele, Davies,
Ditlmann, & Crosby, 2008).
4. Increase visibility and empower current female minorities to promote unity and have a shared
voice until the company achieves equitable representation (Murphy, Steele, & Gross, 2007).
Organizations that can demonstrate these characters are less likely to have an environment where
women do not feel welcome. Table 1 provides the organizational mission and goal, and
discusses the knowledge influences, knowledge types, and knowledge influence assessments.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 23

Table 1 - Knowledge Influence, Types, and Assessment
Knowledge Influence, Types, and Assessment
Organizational Mission
QuickChip creates all types of mobile technologies
Organizational Global Goal
Build a demographic profile at all levels of our company that is a more direct reflection of the
gender and ethnic diversity available in the areas where we do business.
Stakeholder Goal
Improve the percentage of females in the engineering workforce by 1.5% each year, for the next
20 years.
Assumed Knowledge
Influences
Knowledge Type Knowledge Influence Assessment

Goal-Orientation
Senior Leaders should set
mastery goals that mirror the
values they would like to see in
the organization and
communicate those goals to the
whole organization

Declarative

Survey: Ask women if senior leaders
within the organization have
communicated the goals within the
organization.
Self-Efficacy
Managers should feel
empowered and confident in
their ability to recruit and retain
female talent.

Metacognition

Survey: Find out if women’s
experiences during the recruiting
process was affected positively
negatively, or neutrally by their gender.

Survey: Are the policies and culture at
QuickChip supportive of the gender
representation goals?

Motivation Influences
According to Mayer (2011) meaningful learning requires motivation of the learner.
Three key components of motivation are active choice, persistence, and mental effort (Clark &
Estes, 2008). Clark and Estes (2008) describe active choice is the conscious decision to engage
in a task over other priorities, persistence is sticking with the task despite challenges, and mental
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 24

effort is the investment of knowledge and skills to complete a task. High value of a task and
self-efficacy increase active choice, persistence, and mental effort, and therefore motivation
(Bandura, 3, 2000).
This research reviews literature discussing goal orientation theory and Bandura’s self-
efficacy theory as they are two pertinent motivation theories to consider for changing
QuickChip’s culture.
Goal-Orientation Theory. VandeWalle (1997) describes goal-orientation as a person’s
outlook toward developing or confirming their own aptitude in an achievement setting. A study
by Allen and O’ Brien (2006) showed that goal-orientation is important for both the relationship
between the talent-pool and the organization, and between the leadership and employees. The
study showed that the talent pool was more likely to show interest in job openings at a company
with which they felt their values and goals aligned. The study also showed that managers were
more effective when given specific and diverse goals, rather than goals based on a single metric.
Research by Degeest and Brown (2011) shows how effective goal-orientation theory can
be as a management tool. Specifically, they look at how managerial goals affect learning
outcomes of performers. This shows promise as a tool to motivate first-line managers to close
the knowledge gaps through learning. According to the research, managers can state the
organizational goal in altered ways to change how they assign tasks, give rewards, and delegate
authority to motivate subordinates more effectively to learn and close gaps in knowledge through
motivation.
Using achievement motivation to change culture. The senior leaders at QuickChip must
set goals that will motivate the entire organization to achieve its goals. Although each
employees’ orientation to the goals might be different, the end-state is largely the same.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 25

QuickChip’s broke down its long-term goal of achieving equitable representation into a linear
goal of a 1.5% per year, however it must continually optimize its short-term (annual) goals to
ensure that it is attainable, yet difficult. Tipping points, market changes, and long-term
investments in elementary and middle school girls means the path to equity might be a parabolic
curve, rather than a straight line.
An additional benefit of all employees collectively trying to reach a goal is a change in
culture that actively looks for ways to attract and retain female talent. It lets the entire
organization and potential candidates know that it values all engineering talent, helping to
counteract stereotype threat and encourage ambient belonging without calling out negative
stereotypes. The non-recruiting workforce will also become more conscious of QuickChip’s
efforts as they adopt the new standard of diversity and belonging.
Self-Efficacy Theory. Self-efficacy theory is a central motivational tenant of Bandura’s
social-cognitive theory. According to Bandura (2000), one’s belief in their own ability to
accomplish the task is an important factor in their motivation to smart and persist with the task.
Bandura (2000) describes self-efficacy as the one’s belief in their ability to succeed. Someone
with high-self efficacy approaches tasks for readily and with greater confidence. When someone
approaches a task that they believe is valuable and that they can accomplish, there will be a high
level of motivation (Bandura, 3, 2000).Additionally, the individual’s self-efficacy guides the
choices that person makes and how they identify with the given task. It is also important to
distinguish between personal and collective self-efficacy. According to Bandura (2000),
personal self-efficacy derives from just the individual’s own agency, whereas collective self-
efficacy derives from the synergized agency of all within the group. While both are important in
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 26

recruiting because it is such a personalized task, organizational results will be the result of
collective self-efficacy, thus this paper will study the effects of collective self-efficacy.
Maintaining high collective self-efficacy through improbable goals. It is important in
recruiting that senior leaders demonstrate the need for need for female talent by creating
performance goals which reflect the organization’s mission. Because of the shortage of female
computer science engineers in the workforce, QuickChip’s goals might seem unachievable,
which could have a negative impact on managers who feel the task is too difficult or out of their
control. Difficulty is inversely correlated with task acceptance (Erez & Zidon, 1984) meaning
there is a threat that even those with high self-efficacy will not persist with this task or question
their own efficacy (Bandura & Locke, Negative self-efficacy and goal effects revisited, 2008).
The alarming shortage in tech will continue to make the managers’ jobs harder if not addressed
in the short-term, and jeopardize the company’s ability to sustain itself in the increasingly
competitive environment. Thus, senior leadership must continue to monitor the goal and ensure
the goals given to first line managers responsible for recruitment and retention are difficult, yet
attainable, based on the available talent pool. While the task of recruiting female talent from a
non-diverse talent pool is daunting, through goal-alignment and increasing collective self-
efficacy through training and management techniques, QuickChip can become competitive in the
battle for female engineering talent. Additionally, by incentivizing managers to implement new
programs designed to improve the environment for existing female engineers, QuickChip can
bolster its number of female engineers.
Table 2 describes the organizational mission, global goal, stakeholder goal, motivational
influences, and motivational influence assessments identified in this literature review. I will use
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 27

the influences to describe further analyze how motivation contributes to the struggles of creating
an environment where women are able to feel welcome and able to work to their highest level.
Table 2 - Motivation Influence, Types, and Assessment
Motivation Influence, Types, and Assessment
Organizational Mission
QuickChip creates all types of mobile technologies
Organizational Global Goal
Build a demographic profile at all levels of our company that is a more direct reflection of the
gender and ethnic diversity available in the areas where we do business.
Stakeholder Goal
Improve the percentage of females in the engineering workforce by 1.5% each year, for the
next 20 years.
Assumed Motivation Influences Motivation Influence Assessment

Goal-Orientation
Senior Leaders should set mastery goals that
mirror the values they would like to see in the
organization and communicate those goals to
the whole organization

Survey: Ask women if senior leaders within
the organization have communicated the
goals within the organization.
Self-Efficacy
Managers should feel empowered and
confident in their ability to recruit and retain
female talent.

Survey: Find out if women’s experiences
during the recruiting process was affected
positively negatively, or neutrally by their
gender.

Survey: Are the policies and culture at
QuickChip supportive of the gender
representation goals?

Organizational Influences
This section focuses on the influences contributing to the problem attributable to the
organization. Influences which affect an organization’s ability to adapt through change, develop
efficient processes, or a healthy organizational culture which supports the organizational mission
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 28

are factors to consider when identifying organizational influences (Clark & Estes, 2008). While
previously discussed knowledge influences could also be attributable as organizational
influences, the below influences are gaps caused be other than a gap in knowledge.
Managers must create an atmosphere that empowers and embraces women. Since
the workplace and teams have been historically male, the policies of an organization typically
favor male employees. QuickChip offers six weeks of maternity leave, leaving a mother with a
difficult decision to make. With no ultimately positive outcome, mothers must choose to either
take the maternity leave, appearing uncommitted, or forego the leave and appear cold
(Morgenroth & Heilman, 2017). This is just one example of many where traditional household
structures have created a work environment disadvantageous to women.
These difficult considerations lead to qualified women deciding against having children
(Mason & Goulden, 2008) and other qualified female engineers to give up careers to raise
children (Hewlett & Luce, 2005) at a far greater rate than their male counterparts. Policies like
mandatory bonding time, vice the “offer” of bonding time, remove the stigma, and ensure
parents of both genders retain equal footing while maintaining a positive relationship with a new
child.
Simply making available programs fails to adequately address the concerns unique to
female software engineers. These failures directly lower the number of female engineers, but
also negatively impact the number of female candidates in the talent pipeline, and in senior
management positions. Having fewer female engineers makes the company and profession less
desirable to young women (Microsoft Corporation, 2017). According to Morgenroth and
Heilman (2017), forcing women to choose between work and home responsibilities degrades
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 29

both perception and actual performance in the workplace, putting women at a sharp
disadvantage.
Shortage of female candidates with computer science degrees. Women only hold one
out of every five computer science degrees across all levels (Equal Employment Opportunity
Commission, 2016). This reflects the percentage of female software engineers at QuickChip,
and the bachelors in computer science has typically been the requirement for entry-level
positions at QuickChip.
As previously discussed, the number of women receiving computer science degrees is not
likely to improve without for raising the number of female software engineers currently working
in companies. Rather than competing for competing for an extremely limited and valuable
resource, QuickChip’s should focus on filling entry-level positions from alternative sources.
According to an informal survey of over 1,000 tech employers by Indeed.com, the world’s
largest tech recruiting site, 84% of employers feel code graduates are just as prepared or more to
be high-performing programmers (2017). These camps provide more classroom contact hours
and have a greater focus on specialized areas and real-world scenarios than a traditional 4-year
degree (Craig, 2016).
Non-traditional hires can bolster the ranks of female engineers in the near-term, lessening
the stereotype threat, and increasing ambient belonging, all while providing additional engineers
to bolster QuickChip’s existing workforce.
Implicit bias in the recruiting and hiring process. The recruiting and hiring process is
generally the same throughout all organizations. A manager identifies a need and works with a
recruiter to create a job description, identify qualifications, salary, and ideal traits. Next, the
recruiter either actively or passively fields candidates. In passive fielding, recruiters review
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 30

applications from candidates, while in active recruiting, the recruiter scours sources and engages
candidates directly. The recruiter screens the candidates, with the agreed upon candidates
coming on-site for interviewing, typically with the hiring manager at a minimum. Next, the
manager selects a candidate, and the company extends an offer of employment. The last step is
on-boarding the employee which includes relocation assistance, answering questions, and
general support until the candidate begins working. Each step presents its own challenges in
ensuring implicit gender bias does not stifle recruitment of qualified female candidates, but the
two areas where companies lose the most female candidates from the talent pipeline are during
the advertisement of position and the screening and interview stages.
Masculine bias in job descriptions. In a study by Gaucher, Friesen, & Kay (2011)
identified lists of words used in job descriptions of male-dominated, female-dominated, and
gender-neutral professions to identify whether there was bias in the job descriptions. The study
shows that while not responsible for create gender dominance in their respective fields, they
reinforce and reflect the norms of the hiring managers and recruiters who write the
advertisements. While the study showed that feminine descriptions had a slightly negative
impact on male’s perception of the job, the words had a much larger impact on the female study
participants’ perception, signaling that these job descriptions are discouraging potential female
applicant. By contrast, the gender-specific nouns had no impact on either gender’s support for
their respective roles. In other words, descriptions using male-dominated terms were not more
likely to attract male candidates, but only detract female candidates. By writing job descriptions
that avoid masculine-associated words like, “Dominant,” “Superior,” or “Aggressive,” and
feminine-associated words like, “Gentle,” “Committed,” and “Interpersonal,” recruiters can more
effectively attract candidates regardless of gender. QuickChip has made tremendous strides over
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 31

the last 24 months in removing gender-specific pronouns from their job descriptions.
QuickChip’s job descriptions are strictly matter-of-fact, using only descriptive adjectives to
describe the exact nature of the roles.
Male dominated job interviews. In a study conducted by Latu, Mast, & Stewart (2015),
researchers conducted mock job between male interviewers and female interviewees. Prior the
interview, researchers conducted a survey to measure each participant’s implicit stereotypes and
explicit bias. The study showed that many men had implicit stereotypes of women as less
competent than men. This bias reflected during the interview process, the male interviewer made
the female applicant feel incompetent during the interview, thus performing at a lower level,
creating a self-fulfilling prophecy. The study suggests, but did not prove, that the reverse would
hold true, that when a male conducts an interview with another male, the interviewer’s bias
towards the competence of another male would bolster the confidence and thus performance of
the male interviewee.
Even overcompensation by a well-meaning interviewer or recruiter can have a negative
impact. Danaher and Crandall (2008) conducted a study which showed that simply discussing
negative stereotypes by a person not included in that stereotype, reflect an implicit bias that
changes the perception of the candidate by the interviewer, and the actual performance of the
interviewee. This includes when the discussion is meant to be positive, e.g., “most women
struggle in this kind of environment, but you are not having any issues at all.”
By ensuring adequate representation when conducting candidate screening and
interviews, recruiters can get more female engineering talent into companies. QuickChip has not
publicly made interview data available, but interviewing candidates after interviews could help
understand whether interviewers at QuickChip have a gender bias.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 32

The two biggest areas where recruiters can have an impact on attracting and retaining
female talent are in the elimination of stereotype threat through encouragement of ambient
belonging and eliminating the effects of implicit bias during the recruiting and hiring process.
Executive Leadership at QuickChip has not placed enough emphasis on recruiting
and retaining female software engineers. Recruiters and hiring managers at QuickChip have
a relatively low bar to hit. This feeds a complacency within the organization that maintaining the
status quo is acceptable. Executive leadership have a long-term goal in place, but little to no
accountability. Because the problem does not affect the bottom-line, executives relegate the
problem to publicly released reports and statistics. The goals, while technically “SMART,” have
such a long timeline for success (over 20 years) that the likelihood of any current executives
being held responsible for the success or failure is low enough to not make accomplishing the
goal a priority.
By placing a larger emphasis on the topic, and holding recruiters and hiring managers
responsible for diversity, QuickChip could see a cultural shift in fostering a more positive
climate for women.
Table 3 provides the organizational mission and goal, and discusses the knowledge
influences, knowledge types, and knowledge influence assessments.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 33

Table 3 - Organizational Influence, Types, and Assessment
Organizational Influence, Types, and Assessment
Organizational Mission
QuickChip creates all types of mobile technologies
Organizational Global Goal
Build a demographic profile at all levels of our company that is a more direct reflection of the
gender and ethnic diversity available in the areas where we do business.
Stakeholder Goal
Improve the percentage of females in the engineering workforce by 1.5% each year, for the
next 20 years.
Assumed Organizational Influences Organization Influence Assessment

Cultural Model Influence 1
Software engineering is a male dominated
field, and not seen by many as a female-
friendly profession.

Survey: Do females believe there is a bias
against them in the workplace?

Survey: Do females believe there is a
negative stereotype against them in the
workplace?
Cultural Model Influence 2
There is a shortage of female candidates with
computer science degrees.

Survey: Does QuickChip struggle to identify
qualified female applicants?
Cultural Setting Influence 1
QuickChip is male dominated, and therefore
likely to a show a bias against women and
therefore less likely to hire a female candidate.

Interview: Do recruiters or hiring managers
show bias during any step of the recruiting
and hiring process?
Cultural Setting Influence 2
QuickChip’s senior management has not
placed enough emphasis on recruiting female
talent outside the recruitment team.

Survey: Do QuickChip employees feel that
the company is actively engaged in achieving
equity and creating a gender-neutral, positive
environment?

THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 34

Conclusion
QuickChip’s “noxious culture” contributes to its low retention rates, which makes it
difficult for it to recruit new talent. By addressing these gaps in the knowledge and motivation
of its key leaders, recruiters, and hiring managers, QuickChip could improve the awareness of
the recruiters and hiring managers, and bolster its number of new female recruits. By identifying
its organizational issues, it can work toward a goal to improve the social climate, reducing
turnover while also increasing performance for its female staff.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 35
CHAPTER 3
METHODS
Purpose of the Project and Questions
The purpose of this evaluation study is to analyze and understand how the culture and
policies of QuickChip affect the women who work there. The goal is to help QuickChip identify
issues that keep women from joining and staying in the organization, thus continuing the cycle,
by preventing a culture that excludes those who are in the minority of the population. The
questions that this project will ask are:
1. Do employees at the organization understand how the culture within the organization
impacts performance?
2. Do employees at QuickChip believe that a more diverse and welcoming environment
is better for them personally and the organization?
3. Is QuickChip’s senior management taking effective steps to create a culture where
women feel welcome and equal?
Conceptual Framework: The Interaction of Stakeholders’ Knowledge and Motivation and
The Organizational Context
A conceptual framework provides an outline in which to contextualize a proposed theory.
The conceptual framework combines relevant elements from various other theories which
support and inform research and incorporates research-based information coupled with the
researcher’s experience and assumptions. When all the elements combine, they provide a
scaffolding that helps guide and contextualize research (Maxwell, 2013; Merriam & Tisdell,
2015). This evaluation research addresses how QuickChip’s atmosphere impacts the women
who work there, and analyze how it affects the retention and attraction of female employee in the
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 36

context of research surrounding stereotype threat, implicit bias, and ambient belonging. The
stakeholder groups this project addresses are the recruiters and hiring managers responsible for
attracting and brining in female talent and senior leadership who establish the culture at
QuickChip. Recruiters and hiring managers encompass are how QuickChip bolsters its numbers,
making them the primary stakeholder. The senior leadership are ultimately responsible for those
numbers but have the most influence over the retention of female engineers. All stakeholders
must be aware of how current and future female engineers perceive the company, and any gaps
that exist in knowledge, motivation, and organization which are causing the shortage.
Data Collection
This project used mixed methods to collect data. According to Hurmerinta-Peltomakl
and Nummeia (2006) mixed methods are best for researching complex business issues like
QuickChip’s shortage of female engineers. After the researcher administered the surveys, he
conducted follow-up interviews with randomly selected participants who indicated they were
willing to participate further. According to Ivankova, Creswell, and Stick (2006), the sequential
explanatory strategy allows the researcher to collect quantitative data and use it to guide and
create more meaningful qualitative data collection.
Surveys were the most effective way for the researcher to collect information from the
participating stakeholders (Fink, 2015). The researcher used surveys to gather quantitative data
that indicated they worked in a culture where they experienced stereotype threat, implicit bias,
and a lack of ambient belonging. In addition to the survey, this research incorporated data
compiled by government agencies such as the Bureau of Labor and Statistics and other
Department of Labor agencies. These data points help and contextualize the data gathered
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 37

through the surveys (Merriam & Tisdell, 2015). The contextualized results of the surveys guided
the survey protocols and questions.
The researcher conducted interviews further analyze and gain context into the
quantitative results of the survey (Ivankova, Creswell, & Stick, 2006). Based on the results of
the survey, the researcher developed interview questions for qualitative analysis.
Explanation for Choices
The Sequential explanatory strategy is perfect for this complex problem. The problem is
complex because there are so many factors and variables, and it is easy to shift blame and
responsibility. By gathering data through surveys targeted at the entire stakeholder population, it
is possible to get valuable quantitative data to guide the interview process to specific areas of
strength and opportunity. Combined, these help the researcher understand the biases and culture
problem that keep women from getting hired and cause them to have shorter careers than their
male counterparts.
Participating Stakeholders
The women who work in computers science and engineering positions within QuickChip
are the participating stakeholders in the study. QuickChip has an employee resource group, with
a pseudonym called QuickChip Computer Hardware Engineering Women (QCHEW). QCHEW
is an official resource group of women in computer science and engineering positions at
QuickChip which serves a platform for communication and professional development. While
these women are the not ones in control of the culture and actions to increase the numbers, they
are reflective of the culture at QuickChip and are representative of the effectiveness of its
programs and policies. The researcher developed a survey for QCHEW members to better
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 38

understand the culture and stereotype threat towards women who already work in science and
engineering at QuickChip and followed it up with interviews to
Survey Sampling Criteria and Rationale
Criterion 1. An engineer or computer scientist at QuickChip who identifies as a
cisgender female and at least 18 years old. While the experience of transgender women is just as
valuable, the experiences are likely to be significantly different and are not part of the focus of
this study.
Criterion 2. An engineer or computer scientist at QuickChip for at least 12 months.
Most employees begin to feel comfortable in their working environment after several months.
Employees with fewer than 12 months are still on a probationary period, and the lack of
experience makes their input unreliable.
Criterion 3. Works in an on-site department with at least one other coworker. There are
many engineers who work remotely, or on a special project as a department of one. While their
role in the organization is important, their views will not reflect the overall culture of the
organization.
Survey Sampling Strategy and Rationale
The researcher worked with a point of contact at QuickChip to distribute the surveys.
The point of contact distributed the surveys via email to team leaders within the organization for
further dissemination. The stakeholder population the survey provided valuable quantitative data
to guide further exploratory research. QCHEW is a ready-made group exclusively of women
engineers and computer scientists in QuickChip. Respondents who do not meet the qualification
will not be able to proceed beyond the qualifying question portion of the survey.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 39
The survey was sent to a point of contact who works in the Human Resources department
to engage stakeholder. The point-of-contact distributed the survey to all first line supervisors
who had the option to forward the survey onward. Out of roughly 3,000 eligible respondents,
153 qualified women responded.
Interview Sampling Criteria and Rationale
Criterion 1. Engineer who completed the entire survey, and indicated she would be
willing to participate in a follow-up interview, and provided valid contact information.
Criterion 2. Potential interviewees will receive additional screening to control for
preconceptions and different departments. This will allow the researcher to gain perspective
from a variety of sources and perspectives about the atmosphere and QuickChip to gain a more
complete view.
Interview Sampling Strategy and Rationale
The sampling strategy for the interviews is quota sampling. Quota sampling couples the
randomness of the participants with non-random controls to ensure interviews comes from a
broad background when controlled for things like department and views (Creswell, 2014).
Because individual experience in an organization is largely dependent on a person’s direct
supervisor or department (Danaher & Crandall, 2008), results could be misconstrued without
these types of controls. Quotas also offer the most effective sampling method for this study
because it eliminates the bias of selection from the researcher, while still obtaining diverse
viewpoints.
Sampling Strategy and Timeline
The project will use exploratory sequential mixed methods for its research. Table 4
shows the sampling strategy and timeline used to collect data for this project.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 40

Table 4 - Sampling Strategy and Timeline
Sampling Strategy and Timeline
Assumed
Knowledge
Influence
Sampling
Strategy
Number in
Stakeholder
Population
Number of
participants
from
stakeholder
population
Start and End
Date for Data
Collection
Surveys Convenience
Sampling
2,000 female
engineers at
QuickChip in
the United
States
1,200 members
in the QCHEW
group
7/1/2019-
8/15/2019
Interviews Quota
Sampling
2,000 female
engineers at
QuickChip in
the United
States
3 qualified
engineers
based off
random
sampling
8/16/19 -
9/2/19
Document
Analysis
As Needed N/A N/A As Needed

Quantitative Data Collection and Instrumentation
This project collected quantitative data using surveys. The researcher worked with a
point of contact at QuickChip to distribute the surveys. The point of contact distributed the
surveys via email to team leaders within the organization for further dissemination. The
stakeholder population the survey provided valuable quantitative data to guide further
exploratory research.
Surveys
Survey Instrument. The survey instrument included 12 questions aimed at identifying
common causes and responses of stereotype threat and ambient belonging. The survey
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 41

predominantly used the Likert Scale, using agreement statements ranging from Strongly Agree to
Strongly Disagree, with options to skip the question or mark is as not applicable (Fink, 2015).
The project’s conceptual framework is structured around three principals, stereotype threat,
implicit, bias, and ambient belonging. The researcher designed the survey to identify different
ways in which these can occur and tie them back to gaps identified in Chapter 2. One sample
question with possible from the survey is, “Please rate how you agree with the following
statement, “My gender positively impacted my recruiting experience at QuickChip.” This
question can identify if knowledge, motivation, or organizational influences are causing issues at
the talent pool that are perpetuating stereotype threat and ambient belonging issues within the
organization.
The questions are categorized and follow a sequential order, starting with questions about
the recruiter, then the hiring manager, to current team, then questions about QuickChip’s culture.
This sequence of progression helped respondents follow the pattern, and the continuous use the
Likert scale will assist in familiarity and ease of responses (Fink, 2015). The researcher coded
and analyzed the data to identify trends and areas of concern for further research during the
exploratory phase.
Lastly, the survey asked additional questions about the respondent to identifying other
contributing factors. These questions will ask the respondents’ age range, race, education level,
position level within QuickChip, and general team. These questions mitigate the possibility of
misidentifying gender as the cause of potential issues when it could be related to another factor,
or isolated to specific ages or levels within the organization. The survey ended with asking the
respondent if she would be willing to participate in a one-hour interview.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 42

Survey Procedures. The researcher disseminated links to a point of contact to the
survey upon approval from the University of Southern California Internal Review Board.
Because the project is exploratory sequential, the surveys were administered first. The surveys
were used to gather general information to help narrow the focus to specific areas of interest
(Merriam & Tisdell, 2015). The survey was distributed via a resource group called QCHEW.
QCHEW stands for QuickChip Computer Hardware Engineering Women, it consists over 1,200
members. It is a private group that requires verified employment at QuickChip and to identify as
female. The point of contact and QCHEW director did not want the researcher to send the
surveys directly. Instead, the point-of-contact distributed the surveys to leaders within the
organization for further dissemination. However, not all leaders forwarded the survey, and the
point of contact did not want to push the issue. The QCHEW group profile that matches the
stakeholder group of this project. The group is highly active and its status as a sponsored official
QuickChip Resources Group lend itself to the credibility of the study. In total, there were 153
responses from a diverse set of individuals wide good variation in age, race, level of seniority,
and education.
The researcher conducted the survey using Google Forms, with responses coded and
stored onto a Google Sheet for easier manipulation. Google Forms is a cloud-based solution,
with all changes tracked for easy recovery should catastrophic conditions or malicious data
manipulation occur. Google Forms also integrates directly with Google Sheets, allowing for
streamlined data analysis, manipulation, and visualization using charts and graphs. Google
Forms also contains advanced features to help mitigate duplicate responses or “bots” from
completing multiple surveys to skew the results.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 43

Qualitative Data Collection and Instrumentation
The project gathered qualitative responses through interviews with survey respondents
who indicated willingness to participate. The researcher selected interviewees through pseudo-
quota sampling to ensure other factors such as race, age, or team were adequately represented in
the research. However, because only 9 respondents volunteered, most of whom fit the same
profile, the researcher only selected three for interview. Only three were selected, as most
volunteers fit the same profile, and were likely to skew the results the of qualitative research.
The researcher conducted the interviews telephonically. The researcher used a tablet computer
with a pen to take notes, and convert into digital notes. The researcher recorded the interviews
using a voice recorder for transcription.
Interviews
Interviewees. The researcher interviewed three female engineers. Engineer 1 was a
senior individual contributor, from East Asian, and lived in the United States for 12 years, right
after finishing her undergraduate degree in China. She is between 30 and 35 years old and
completed a Master’s Degree in Electrical Engineering in 2016 during a hiatus from working at
QuickChip. Engineer 2 is also a senior individual contributor between 25 and 30 years old, of
East Asian descent, but is a second generation-American and finished her undergrad in 2015.
Engineer 3 is a first-line manager of East Asian descent, but did not indicate how long she has
been in the United States. She has managed employees for 4 years, had an undergraduate degree
in Computer Science and an Masters in Business Administration. She did not indicate her age.
Interview Protocol. The interview used a semi-structured format, using a mixture a
formal and unstructured questions (Maxwell, 2013). The semi-structured approach is most
appropriate because the structured questions allow traversal across the breadth of the topic, with
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 44

the unstructured questions allowing for more depth where appropriate (Merriam & Tisdell,
2015). The interviews focused primarily on the organizational influences at QuickChip. The
interviews also helped gain a greater understanding of the knowledge and motivation gaps,
which are more difficult to articulate and understand though survey questions (Clark & Estes,
2008).
Interview Procedures. The researcher conducted the interviews after gathering and
analyzing the survey. The researcher finalized the survey questions based on the quantitative
results from the survey (Maxwell, 2013). The survey asked for willingness to participate in an
interview, basic demographics, and general information about level within QuickChip. The
selection ensured a wide representation with consideration given for age, race, and level at
QuickChip. The computer had alternates selected should a primary be unable or unwilling to
participate in the interview.
The researcher interviewed each interviewee one a time. The researcher scheduled each
for one hour, with a target for 45 minutes. This timeframe allowed the respondents to provide
focused responses while appreciating the limits of the interviewees’ energy level and time
(Merriam & Tisdell, 2015). The researcher did not conduct any follow-on interviews since they
would provide no additional benefit. The additional 15 minutes allowed for additional time
should the unstructured questions run long. Altogether, the researcher conducted three hours of
interviews.
The researcher conducted the interviews telephonically to protect the privacy and
anonymity of the respondents. Because QuickChip uses an open work plan, there are no private
offices. Therefore, the in-person interviews would be conducted in conference rooms or public
places where the anonymity of responses would be great limited given the low number of
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 45

interviews. The researcher recorded audio using a and tablet computer with the interviewee
gives permission. Additionally, the researcher took notes on a tablet computer using a stylus.
The notes will help capture non-textual cues such as voice inflection gestures which may not be
captured through transcription alone (Merriam & Tisdell, 2015).
Data Analysis
The researcher coded survey responses and interviews for analysis. Because the research
is sequential-explanatory, the researcher coded and analyzed the surveys before conducting the
interviews. The surveys identified areas of interest for further qualitative analysis using
interviews.
The researcher transcribed then coded the interviews. Once coding was complete, the
researcher created a codebook using open, then axial codes. After the codebook was finished,
the researcher cross-referenced data to the knowledge, motivation, and organizational influences
discussed in the literature review. From there, the researcher created assertions based on the
results of the analysis.
Credibility and Trustworthiness
Trustworthiness of researcher is directly correlated to the overall credibility and
trustworthiness of the qualitative data (Maxwell, 2013). First, the researcher must follow all
USC policy and procedures regarding interview design, conduct, and analysis. These policies
are for the protection of the respondents and to ensure the credibility of the data gathered.
Second, the quota selection process increased data credibility and trustworthiness by ensuring
responses were gathered from a representative population (Merriam & Tisdell, 2015). Lastly, the
researcher must ensure the responses are free from the manipulation or pressure of QuickChip
(Hunter, 2002). The researcher accomplished this through ensuring privacy and anonymity of all
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 46

responses. The researcher securely stored and archive all notes and audio recording, and never
used any personally identifying information in writings or findings. These protections gave
respondents the security to speak honestly about their experiences without fear of retribution.
Validity and Reliability
The survey provided the groundwork for the remainder of the explanatory sequential
research; therefore, it is crucial the data collection is valid and reliable. The researcher
specifically phrased the questions to test for stereotype threat, implicit bias, ambient belonging,
or a combination, in varying scenarios at QuickChip. Although the questions are not
psychometrically tested, they were useful to identify key areas of concern or praise to warrant
additional research.
To ensure the validity and reliability of the survey, the researcher took the following
steps. First, the high volume of data and wide variety of responses increase the data credibility
and trustworthiness due to its large sample size (Maxwell, 2013; Merriam & Tisdell, 2015; Fink,
2015). Second, the researcher validated data to ensure all responses were genuine (Fink, 2015).
This step used automation to ensure the target population could not manipulate the data by
ensuring the survey could only be completed by a single computer. Additionally, the researcher
used the Google Forms Captcha functionality. The Captcha uses technology to primarily prevent
any automated responses, but is also useful to prevent multiple responses. Combined with the
verified users in the QCHEW group, this ensured all responses were unique. Third, the
researcher used data triangulation to validate responses. Because of the large number of
respondents, the researcher could detect anomalies and outliers which would not have been
possible with a smaller sample (Fink, 2015). Because the survey responses were automatically
placed into a Google Sheet, the researcher could model and conduct regression that control for
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 47

other possible causes in the survey such as race, age, or level within QuickChip. By using
multiple people and conducting modeling from multiple perspectives, the researcher ensured data
validity and reliability (Denzin, 2006).
Ethics
This project seeks to understand a corporate culture and environment through
communication with tech employees and recruiters. Although the researcher neither works for
nor has any formal connections to the organization research, precautions were taken the protect
the anonymity and security of the responses, especially due to the legal ramifications of the
topic.
Informed consent is fundamental to legal and ethical research (Stahl, 2011). All survey
invitations included the purpose of the study to ensure respondents understood the reason the
researcher was seeking their input. The first question on the survey contained the full disclosure
of the survey. By checking the box acknowledging they read the disclose and decision to
continue with the survey after acknowledging the purpose indicates the informed consent of all
respondents. The researcher took a similar approach with interviews, where respondents
indicated through a signature before the recording began that the interviewee acknowledged the
purpose of the interview and confirmation of their willingness for the interview to be recorded.
All respondents had reached the age of majority and were current employees of
QuickChip with at least 12 months of employment. The researcher informed respondents that
their identities would be protected and that all data from surveys would be aggregated. Interview
recordings are kept confidential, and all use of the interview in the research will remove any
traceable information such as personal stories with names or specific examples that include small
departments. Personally identifiable information (PII) is stored in Amazon Web Services
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 48
GovCloud sector, providing the highest level of cloud security for the data at rest. Non
personally identifiable information is stored in Google Cloud Applications.
The researcher disclosed his former employment with Intel Corporation, QuickChip’s
largest competitor in its talent acquisition department, including completing a special project
surrounding diversity recruiting. Although the researcher does not have current connection with
either Intel or QuickChip, it was important to disclose this information, as failure to disclose
could crumble trust or call into question the integrity of the research.
Due to the researcher’s former employment with work in the field of diversity recruiting
in a large technology company, there was potential for preconceived ideas or notions.
Additionally, because the researcher is a heterosexual white male seeking to better understand
the culture which historically benefitted heterosexual white males, the researcher must be
cognizant of both the internal biases this could produce and the external perception, especially
amongst the respondents. To minimize the effect of the researcher’s demographic on the
outcome of the study, the researcher shared his own experience as a diversity recruit, a disabled
veteran; and the motivation for the study, the researcher’s daughters who face discrimination and
ridicule at a young age for their ambition in STEM. Lastly, Hunter (2002) states that conducting
a meta-analysis of one’s own research can identify bias in research findings. The researcher
conducted a meta-analysis of the findings to check for implicit bias.
Conclusion
Chapter 3 identified the methodology and procedures used to collect the data for this
study which included surveys and interviews. Chapter 4 will present the data the researcher
collected.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 49

CHAPTER 4
RESULTS AND FINDINGS
The purpose of this study was to evaluate the knowledge, motivation, and organization
gaps inside a large technology company which contribute to the cyclic lack of women in
computer science and engineering. This chapter presents the data collected to test the hypotheses
identified in Chapter 3. This study used quantitative methods to collect and analyze data. The
researcher worked with QuickChip to send out an email containing the purpose of the study and
a link to an online survey.
While the researcher had only one point of contact, the point of contact liaised with the
leaders of the women’s advocacy group in the organization. The leaders of the advocacy group
approved the survey, but not all sent out the survey to the women under their divisions. The
researcher received no feedback about who received a survey, just that not all leaders would send
it to women in their departments. The total number of responses indicated that only a small
portion of the population received the survey, although the data collected was still enough to
validate gaps identified in Chapter 3. Out of potential 2,000 female engineers at QuickChip, 153
completed valid survey.
Although the researcher originally planned to conduct quota random sampling for
interviews, the number who indicated willingness to be interviewed was too low (N=9) and not
sufficiently diverse. Out of the 11 respondents nine identified themselves as senior individual
contributors and 10 identified themselves as Asian. Thus, pseudo-random sampling was used to
conduct three interviews. Additionally, during analysis, the researcher had to modify the race of
Black (N=3) and Hawaiian/Pacific Islander responses (N=1) to “Other” to protect the identity of
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 50

those individuals as guaranteed to the respondent during the survey and as approved by the
Internal Review Board (UP-19-00285).
The researcher categorized the data by knowledge, motivation, and organizational
influence type. Further, each item of interest is show as validated or not based on the survey data
collected. In analyzing the data, the researcher considered and gape validated if more than 30%
of the respondents answered a survey questions related to the category in a way which confirmed
the gap. An identified as having no gap if fewer than 30% of respondents answered a survey
question relating to the category in a way which refutes the gap. The researcher categorized
items which had more than 50% of responses as “No response” or ambiguous responses as
unable to validate.
Results and Findings for Knowledge Gaps
Validated Knowledge Gaps
Factual Knowledge. The researcher collected and analyzed data on factual knowledge
influences. The results are below.
Stereotype Threat and Ambient Belonging. Gap Validated. Both the survey and
interiews identified a lack of factual knowledge of stereotype threat and ambient belonging, and
how they impact performance in the workplace.
Survey results. Three survey questions asked directly about stereotype threat and
ambient belonging. The first question asked how much women felt the stereotypical engineer at
QuickChip. This question did not specify gender, but rather, was asking how women felt they
belonged in the organization. Almost 60% of women could not say they felt they were
stereotypical of a QuickChip engineer, which indicates that they may not feel they belong. The
full results of this question are in Figure 1.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 51

Figure 1 - Results of Survey Question 9
The second question asked if women consider how their words or actions might by be
perceived differently in the workplace because of gender before speaking or acting. Of those
surveyed, 43.4% responded affirmatively that they consider their gender before speaking or
action, reflecting a stereotype threat within the organization. Figure 2 shows the full responses
to the survey question.

THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 52

Figure 2 - Results of Survey Question 10
The third and final question on this topic was asking women if they felt there was a
positive stereotype for female engineers at QuickChip. The responses to this question were
largely neutral, with a majority indicating neither agreement nor disagreement with the
statement. This is potentially positive for QuickChip since it also does not indicate a negative
stereotype at the organization, which is a major indicator of stereotype threat. The full responses
to the question are in Figure 3.

Figure 3 - Results of Survey Question 11
Interview results. The researcher conducted three interviews to further discuss the survey
results with female QuickChip engineers at different levels of the organization. None of the
employees were familiar with the concept of stereotype threat, but agreed that it was prevalent at
QuickChip after having the principle explained to them. When discussing ambient belonging,
one senior individual contributor mentioned:
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 53

I do feel like I belong at [QuickChip]. There are many women who look and sound like
me in the organization, and it feels like I belong. I don’t want management or to lead
teams, just focus on the engineering. I suppose it depends on which team your on to
know whether or not belong, and if you are the only women, then you might not.
A first-line manager expressed frustration with the programs and how she felt they reminded her
that women are treated different than men:
I think [QuickChip] works to make everyone feel included, but when they keep going out
of their way to make you feel included, you know that you aren’t. They don’t have flyers
up, celebrating what every man did, but when they make it so women are front and center
for every little thing, it’s a reminder of how everyone views us in the organization and
that they are trying, but we’re not there yet.
Both the survey results and interviews validate the factual knowledge gaps surrounding
stereotype threat and ambient belonging.
Metacognitive knowledge. The researcher collected and analyzed data on metacognitive
knowledge influences. The results are below.
Implicit bias. Gap Validated. Both the survey and interiews identified a lack of factual
surround stereotype threat and ambient belonging, and how they impact performance in the
workplace.
Survey results. There were many questions around implicit bias in the survey. However,
one was specifically designed to gauge the metacognitive knowledge gaps that exist within the
recruiting and hiring process that controls the inflow of new candidates into the organization.
The survey results were positive for QuickChip, showing over 70% of women did not feed
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 54

uncomfortable because of their gender during the recruiting and interview process. The full
results of the survey question are shown in Figure 4.

Figure 4 - Results of Survey Question 2
Document analysis results. There is an inherent selection bias in the survey, since the
woman survey made it through the selection process. Thus, the researched leveraged document
analysis for the validation of this gap.
Documents obtained by the researcher show that men conduct 81% of interviews at
QuickChip. While this is representative of the demographic currently at QuickChip, it also
allows implicit bias to affect both the performance of female candidates and the perception of
those candidates by male recruiters (Latu, Mast, & Stewart, 2015).
Interview results. Only one woman interviewed by the researcher said that she had an
interview with a female employee. In total, male interviewers conducted seven out of eight
interviews. No respondent indicated they felt any bias.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 55

While the surveys and interviews do not indicate any serious issues, 10% indicating
negative experiences during the recruiting process because of gender. The data also does not
show women who declined offers from QuickChip due to gender-related issues as a candidate.
The data collected shows that the interview protocol at QuickChip and the lack of implicit bias
training for interviewees is likely impacting the number of women coming into the organization.
Knowledge Findings
As discussed in Chapter 2, knowledge of one’s own bias is often a sufficient motivator
for people to strive toward correcting the behavior (Merriam & Tisdell, 2015). The surveys and
interviews conclude that while there does not appear to be intentional bias against and threats
against women, they do exist. The fact that women are experience stereotype threat and ambient
belonging at QuickChip, but are not aware of the terms describing the experience indicate that
QuickChip has gaps in knowledge of stereotype threat, ambient belonging, and implicit bias.
Results and Findings for Motivation Gaps
Validated Motivation Gaps
Goal-Orientation. The researcher collected and analyzed data on goal-orientation
motivation influences. The influence which the research validated is below.
Using achievement motivation to change the culture. Gap validated. Both survey and
interviews demonstrate QuickChip does not use goal-orientation throughout the organization to
motivate employees to achieve the goal of equitable representation.
Survey results. The survey asked the female engineers that if their manager was
assembling a team, what would the gender makeup of that group look like. This question seeks
to understand the goals QuickChip gives its managers when give a fundamental task such as
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 56

building a team. Nearly half (49.3%) stated that the manager would build a team with more
males than females. The results of the survey question are in Figure 5.

Figure 5 - Results of Survey Question 8
Interview results. The researcher asked the interviewees if they were familiar with
QuickChip’s specific goals. When questioned, the manager said she thought the goal was to
interview at least one candidate for every position while the two individual contributors both
thought it was just to increase the number of women, but did not to what extent or by when.
While all three made educated guesses based off the guidelines or what they see going around,
neither knew the specific goal laid out to QuickChip’s investors by executive leadership.
When the researcher explained the goal, none of the interviewees had heard either the
equitable representation or the 1.5% per year targets. When the manager was asked if she would
be motivated if she knew there was a goal, she responded with:
I’m not sure that just the goal alone would motivate me. I have a lot of goals, and if I
worry about the composition of my team, it might cloud my judgement. It might just be a
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 57

goal for the recruiters and senior management to worry about now since there’s so few
available women in the mark. …It doesn’t make sense to give that goal to me.
Self-Efficacy. The researcher collected and analyzed data on self-efficacy motivation
influences. The influence which the research validated is below.
Setting improbable goals to increase self-efficacy. Gap Validated. Both survey results
and interview questions that QuickChip is not setting goals to motivate employees to achieve
equitable representation.
Survey results. The researcher asked the respondents whether they felt they were given
the same opportunities as their male counterparts. Only 47.4% felt they had been given equal
opportunities at QuickChip. This indicates that QuickChip is not setting goals with the
management to develop and challenge female employees to achieve higher levels of success and
feel valued at the organization. The full results of the survey question are shown below in Figure
6.

Figure 6 - Results of Survey Question 3
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 58

Interview results. During the interviews, the researcher asked each interviewee to
describe the opportunities for growth they had been given, then followed that up with a question
on whether it was the same opportunities give to male employees in the organization. The
researcher left the interpretation of “opportunity” and the question of equality to the perception
of the interviewee.
The first individual contributor defended QuickChip’s model of growth by saying:
[QuickChip] has opportunities for everyone, but it’s not like they’re going to fall in your
lap. You have to go out and ask for it.
When asked if she went out and asked for opportunities, she responded with:
Yes, and I’ve always been given the opportunity I asked for. [QuickChip] has paid for
me to go to conferences, take training, and learn new things. Sometimes they put them
out to the whole company, sometimes they are things that I found and asked to attend.
It’s not like they said ‘no, you’re a girl, girls can’t do that.’
The other individual contributor shared a different sentiment:
There are opportunities, but there’s always a ‘golden child’ who they think is the next
hotshot and he will get all the important projects and chances to shine. Since every one
of these people I see is a guy, I can’t help but wonder how many women we have with
similar potential who don’t get recognized or are thought of differently and end up stuck
on a committee where their A-type personalities can lead small groups instead of whole
teams.
When asked about the goals her leadership had given her, she responded with:
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 59

It’s typical engineering tasks. They want me to improve processes and keep the status
quo basically. …Nothing specific or managerial, I like being an engineer though, so I’m
not sure I’d take a leadership opportunity if they ever offered.
The interviews shows mixed results, but reflect the existence of a bias within the
organization that favors men.
Motivation Findings
Goal-setting can be an effective tool for management to motivate employees to reach
goals (Degeest & Brown, 2011). Based on the survey and interview results, QuickChip is neither
setting effective goals with employees to meet the company-wide goal, nor is it setting goals for
female engineers that make them feel like valued staff with the same opportunities as their male
counterparts. These contribute to the cyclic lack of women in the field by reinforcing bias
against women and decreasing their ambient belonging.
Results and Findings for Organization Gaps
Validated Organization Gaps
Cultural Model Influences. The researcher collected and analyzed data on cultural
model organizational influences. The influences and whether the researcher validated the gaps
are below.
Male-Dominated field causes females to avoid the profession. Validated. While there
is a natural selection bias since the respondents did join the profession, the surveys indicate there
is a culture in the workplace that causes women to avoid and leave the profession. While the
researcher did discuss this topic during the interviews, all three women did not move to the
United States until after college. While all women acknowledged a gender gap, they did not feel
their culture views computer science as a “man’s job,” but did acknowledge a bias in promotion
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 60

and leadership of males. Given the complexity and nuances of cross-cultural norms, the
researcher excluded the interviews from the gap analysis.
Survey results. The number one reason women leave the career field is due the birth or
adoption of a child (Morgenroth & Heilman, 2017). The survey contained two related questions,
designed to help the organization understand if there is a culture supportive of women. The
same perceptions that prevent or attract women from to the career field are the same perceptions
on whether they have a desire to return after a significant life event. The researcher modeled the
two-part question after the “unwinnable dilemma” that is part of a culture unsupportive of
women (Morgenroth & Heilman, 2017). Women often face the perception that their colleagues
will view them as a bad team player if they take maternity or a bad mother if they do not. The
researcher designed these questions knowing the questions could be divisive, inciteful, or
uncomfortable, but draws an honest assessment of the number one reason women do not view
technology as female-friendly. The results of survey questions are in Figure 7, which asks if a
woman would be viewed as a poor team player if she took leave, and Figure 8, which asks if a
woman would be viewed as a poor mother if she did not. In both cases, the number of women
who felt their colleagues would not consider them a bad mother or bad team player was less than
50%.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 61

Figure 7 - Results of Survey Question 5

Figure 8 - Results of Survey Question 6

Shortage of female candidates in the talent pool. Gap Validated. While QuickChip does
not directly control the talent, at the core of this study, is the problem of the cycle of low
numbers. The survey and interviews reflect a positive experience, with over 75% not feeling any
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 62

negativity about their gender in the interview and hiring process. However, document analysis
of the environmental factors and QuickChip’s hiring data show there is a gap that QuickChip
must continue to contend with while continuing to contribute to the solution.
Survey results. The survey results show very strong numbers for QuickChip with
overwhelming majority, over 90%, having neutral or positive experiences because of their
gender. There is a selection bias in this survey, since women hired by the organization are likely
to have had favorable experiences. However, the high number of neutral responses is indicative
of highly trained recruiters who are presenting a hiring experience with low implicit bias. The
full results of response are listed in Figure 12.

Figure 9 - Results of Survey Question 1
Document analysis results. QuickChip’s percentage of female engineers at the end of
2018 was currently at 16%, up from 15% in 2017, but still well below the industry average for
large tech firms at 23.9%, and the percentage of employed in computer science and engineering
positions in the United States, at (National Center for Science and Engineering Statistics, 2016).
REDACTED
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 63

According the National Center for Education Statistics, in 2017, women received 18.7% of
computer science degrees (U.S. Department of Education, 2017). A chart showing a comparison
of the numbers is in Figure 10.

Figure 10 - Comparison by category of the percentage of women in Computer Science
While QuickChip is behind its competitors and the national average, it also must find a
way to increase the talent pool to meet its long-term goals.
Cultural Setting Influences. The researcher collected and analyzed data on self-efficacy
motivation influences. The influences which the researcher validated are below.
Natural bias due to the low number of female engineers. Gap Validated. The surveys
and interviews reflect an environment that has a natural implicit bias caused when a social group,
such as women in this case, is greatly outnumbered (Petriglieri, 2011).
Survey results. The researched asked the women directly if they felt QuickChip had any
bias towards male or female employees. More than half (54.1%) said they felt the organization
has a bias towards men. There are many reasons for this perception, but the most likely is that
0.00% 5.00% 10.00% 15.00% 20.00% 25.00% 30.00%
QuickChip
Large Tech Avg
Overall Industry
New Graduates
Women in Computer Science
QuickChip Large Tech Avg Overall Industry New Graduates
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 64

the large percentage of men in the organization have created a culture which favors other men
and where women feel separate from the core due to the obvious difference and lack of
inclusivity in the workplace. The results of the question in Goals Figure.

Figure 11 - Results of Survey Question 7
Interview results. The interviewed women had mixed perceptions on bias at QuickChip.
The manager and one senior individual contributor felt there was a moderate bias towards men,
while another senior individual contributor. When discussion bias caused by the high number of
males, the manager responded:
I have always felt like it was harder for me to get “in” because I am a woman. My bosses
and their bosses are all men, and so are most of my colleagues and most of the people on
my team. The interactions can vary, where some people look at me like a diversity hire
or some people change their behavior when I’m around because they don’t know how I
will take their jokes. There’s definitely a boy’s club culture here, but I don’t really feel
like it’s prevented me from moving up or being able to prove myself.
REDACTED
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 65

When asked if she felt the culture might be different if there were an equal number of women in
the organization, the manager followed up with:
I guess it would. It would be harder for those groups to form. Right now, seeing an
office full of guys chatting and laughing is normal because there are so few women,
where it would be far more obvious if there were just as many women. I don’t think the
men do it intentionally, it’s just how things go… so yes, I could see how having more
women could change it [the culture].
The individual contributor who did not feel like there was a bias caused by the number of men in
the organization said:
I’ve never experienced anything like that. Nobody really excludes me from conversation,
I get invited to lunch, my boss makes sure I know what’s going on. I’m sure it happens
here just like it could anywhere, but it’s not something I see, and I don’t think it’s a big
problem that must be dealt with. But still, I don’t see how having more women in the
organization would fix those kinds of people.
Both the interview responses and survey data show a bias, while the interview responses confirm
behavior within the organization likely to be resolved by having more women to prevent natural
social groups from forming.
Insufficient engagement from senior leadership. Gap Validated. Both surveys and
interviews showed that female engineers at QuickChip do not feel that senior leadership is truly
committed to addressing the issues facing women in the organization.
Survey results. The researcher asked women directly if they felt that the executive team
of QuickChip is taking enough action to ensure women are treated equitably. The responses
were evenly split, even amongst the neutral, unsure, or not applicable options. However, only
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 66

30% felt that senior leadership was taking enough action on the job, while 30% felt more could
be done. The survey validates the gap because 70% do not have confidence to say the executive
leadership makes equitable treatment of women a priority.

Figure 12 - Results of Survey Question 4
Interview results. During the interviews, the researcher asked the participants what steps
the executive leadership has taken at QuickChip, and whether they felt it was enough to ensure
equitable treatment of women. The manager who previously acknowledged bias said she felt the
executives should not have to do anything, and that she felt all the trying from senior
management just hurts women more than helps. She felt that programs in place make people
question whether she is an “Affirmative Action hire” or “actually qualified.” Both individual
contributors felt that executives could be doing more. One individual contributor said that not
just QuickChip, but all tech companies should be doing more:
REDACTED
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 67

These people have so much power and money that they should be solve all these
problems. If they really wanted to make it equitable, they would pay for women to go to
college to study computer science. Places like [QuickChip], Microsoft, and Google could
start their own college, for free, and give all the people they want to hire more of an
education directly. I see the posters and know about the programs, and while I think the
idea is nice, I feel like they’re just doing ‘something’ to say they are trying. But I know,
if they really wanted to solve not just this problem, but any problem, they have the money
and people to do anything.
Based off the results of the survey and the responses from the interviews, there is enough
evidence to consider the gap in engagement from senior leadership at QuickChip validated.
Scoring and Correlation
To better understand the data, further assist in recommendations, and take a
comprehensive approach to analysis, the researcher scored each survey and correlated the score
to race, age, role, and education level.
Scoring
The research came up with a simple score for the surveys. If a respondent had a
“positive” response that indicated no gap, they would receive a half point. The score reduced by
half a point If the response was “negative” in a way that indicated a gap. Any unanswered,
neutral, unsure, or not-applicable responses did not change the score. On a 12-question survey,
the potential range was between -6 and 6, however, to simplify scoring and lessen the effect of
respondents simply selecting all positive or negative, the actual scoring scale went from -5 to 5.
Because there is insufficient benchmarking data, it is difficult to make a claim on what is
good or bad. However, the researcher used the scores to identify groups which responded more
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 68

favorably or negatively. There is no causal implication in the data. The average overall score
was 0.336, the median and mode scores were both 1, which indicates that responses were slightly
favorable, indicating a workplace that is not unbearable, but does have multiple areas for
improvement. Figure 13 shows a histogram with scores and an exponential trendline reflecting a
bell curve with a positive score lean.

Figure 13 - Overall Scores

Education Level. In general, a higher the education level had more centric views of the
organization with scores closer to the center. Those with Bachelors and Masters degrees had
outliers on both sides, while those with only a High School diploma were generally favorable.
There were only two respondents with Associates degrees, so their responses were removed to
protect their anonymity. The chart showing scores by education level is in Figure 14.
0
0.2
0.4
0.6
0.8
1
1.2
Score
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 69

Figure 14 - Score by Highest Education Level Completed
Role within the company. The role within the company seems to play a significant
factor in individual’s perception of bias within the organization. The higher someone is within
QuickChip, the less likely they are to report perceived negativity from their gender. Senior
individual contributors are most likely to experience bias. This is highly susceptible to selection
bias since implicit bias and perception are heavily affected by promotions. In other words,
someone in senior management is more likely to not have experienced implicit bias while
someone in an individual contributor might have been passed up for promotion because of bias.
shows the score distribution by role within QuickChip.
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Doctorate
Masters
Bachelors
HS Diploma
Score by Education Level
-5 -4 -3 -2 -1 0 1 2 3 4 5
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 70

Figure 15 - Score by Position
Age. Age played a large factor in perceived bias within the organization. The 26-30 age
had largely positive experiences, while a sharp dropoff in the 31-45 age brackets. The 46 and
above age brackets see a decrease in perception of gender bias. There are several hypotheses for
why the 26-30 age bracket is so positive and is a promising area for future research (Jackson,
Hillard, & Schneider, 2014). shows the score distribution by age within QuickChip
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Entry-Level
Intermediate Individual Contributor
Senior Individual Contributor
First-Line Management
Middle-Management
Senior Management (VP, Director)
Score by Position
-5 -4 -3 -2 -1 0 1 2 3 4 5
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 71

Figure 16 - Score by Age
Race. Race was the largest factor in scoring. While the researcher used all Census
categories for race, low numbers within Hispanic, Black, and Native American necessitated
consolidation into the “Other” category to protect individual privacy. Asian women perceived
the least amount of bias while underrepresented minorities all experienced significant more bias.
This also reinforced the principles of stereotype threat, implicit bias, and ambient belonging by
demonstrating the races that are most represented are least likely to perceive bias, even when
underrepresented by gender. The score breakdown by race is in Figure 17.
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
18-25
26-30
31-35
36-40
41-45
46-50
51-55
56-60
Score by Age
-5 -4 -3 -2 -1 0 1 2 3 4 5
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 72

Figure 17 - Score by Race
Conclusion
This chapter discussed Gathering data from a large, external organization to gauge its
climate provides unique challenges, especially when gathering data solely from the class who
experiences the bias. However, the survey, interviews, and analysis of public documents show
gaps in the knowledge, motivation, and organization at QuickChip contributing to the cyclic lack
of women in computer science and engineering. Chapter 5 will discuss research-based solutions,
implementation guidance, and instruction on how to evaluate the effectiveness of those solutions.
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Asian
White
Two or More
Other
Score by Race
-5 -4 -3 -2 -1 0 1 2 3 4 5
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 73

CHAPTER 5
SOLUTIONS, IMPLEMENTATION, AND EVALUATION
Introduction and Overview
In Chapter 4, the researcher presented the results and findings in detail. Through
sequential explanatory research methodologies, combined with document analysis, the researcher
was able to validate the gaps in knowledge, motivation, and organization at QuickChip. This
chapter will make recommendations to close those gaps based on resources within the
organization.
The solutions discussed in this chapter are based on the New World Kirkpatrick Model
(Kirkpatrick & Kirkpatrick, 2016). The New World Kirkpatrick Model framework coincides
tightly with the Gap Analysis Framework (Clark & Estes, 2008). The New World Kirkpatrick
Model provides a training model where the training targets each validated gaps and each
component is designed and evaluated to close the gap.
Like previous chapters, the researcher categorized recommendations by knowledge,
motivation, and organization. The first section discusses recommendations for training,
mentoring, and job aides. The next section follows the implementation and evaluation using the
framework from the New World Kirkpatrick Model. The final part of this chapter will go over
the limitations of this study and recommend further research areas.
Recommendations for Practice to Address KMO Influences
Knowledge Recommendations
Introduction. According Krathwohl (2002) and Rueda (2011), there are four types of
knowledge to consider when analyzing knowledge influence: factual, conceptual, procedural,
and metacognitive. According to Krathwohl (2002) and his recharacterization of Bloom’s
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 74

Taxonomy, factual knowledge is the basic element that one must be familiar with in learning. In
the context of recruiting, this means the recruiters must understand the facts of why women
avoid tech. According to Anderson (2013), conceptual knowledge is the how the facts work
together to create the environment. For recruiters, it is important for them to understand how
large the issue of the lack of women in tech - both the causes and effects. Procedural knowledge
is knowing how to do something, not just the why (Krathwohl, 2002). For recruiters, it is
important to understand how to attract and engage female candidates and for the business to
market itself as female-friendly, and ensure it has an inclusive environment. Lastly, Krathwohl
(2002) describes metacognition as the awareness and knowledge of one’s own cognition. For
recruiters, this means being aware of the candidates and how those candidates perceive the
company, role, environment, and recruiting experience, and ensuring a tailored to approach to
ensure all feel welcomed as potential future employees.
Table 5 shows the assumed knowledge influences and context specific recommendations for
the assumed knowledge influence that impact QuickChip’s mission.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 75

Table 5 - Assumed Knowledge Influences with Recommendations

Assumed Knowledge Influences with Recommendations
Assumed Knowledge
Influence
Principle and Citation Context-Specific
Recommendation
All employees should know
about stereotype threat and
ambient belonging and how
it affects performance and
culture in the workplace (F)
Modeled behavior is
more likely to be
adopted if the model is
credible, similar (e.g.,
gender, culturally
appropriate), and the
behavior has functional
value (Denler, Wolters,
Benzon., 2009).
Provide training on implicit
bias and its influence in
interviews, using models of
positive techniques that can be
implemented to overcome
implicit bias.
Employees are aware of
implicit bias and understand
how everyone’s implicit
biases impact performance
and culture in the workplace
(MC)
Rationales that include a
discussion of the
importance and utility
value of the work or
learning can help
learners develop positive
values (Fredericks &
Eccles, 2006; Pintrich,
2003).
Provide recruiters and
hiring managers job aids
to help avoid gender-
specific verbiage. More
importantly, provide the
rationale and the research
that shows these words
deter four times as many
candidates it they attract.
QuickChip’s male dominated
workforce creates an implicit
bias against women in the
workplace and degrades
female performance through
stereotype threat. (M)
To develop mastery,
individuals must acquire
component skills,
practice integrating
them, and know when to
apply what they have
learned (Schraw &
McCrudden, 2006).
Provide training that
brings awareness of the
effects of stereotype
threat and how it affects
those in the workplace
and provides effective
solutions to reduce it.
Education programs on
how team performance
can be improved by
creating a diverse and
welcoming culture where
everyone’s ideas are
valued.

THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 76

Reducing implicit bias in computer science engineers. Men conduct 81% of the
computer engineering interviews at QuickChip. This procedural area is of higher priority than
the job descriptions because the resolution helps shape QuickChip’s culture. The researcher
recommended a solution based on Social Cognitive Theory as shown in Table 5. According to
Denler, Wolters, and Benzon (2009) modeled behavior that is credible and has functional value
is likely to be adopted in an organization. This means by demonstrating the positive impact that
reducing the effects of implicit bias could have in new female hires and retention, adoption of
that behavior by interviewers into practice would be likely. QuickChip should provide training
on implicit bias and its influence in interviews, using models of positive techniques that can be
implemented to overcome implicit bias. The reduced implicit bias by male interviewers would
decrease negative feelings against female interviewees, which gives the candidates increased
odds of performing their best, and therefore more likely to be hired.
Implicit association testing helps make an individual aware of their own biases. Simply
being aware of one has own implicit bias is often enough to motivate individuals to correct the
behavior (Green, et al, 2007). Implicit association tests are widely available with varying
options. While association testing provides motivation, QuickChip should provide educational
tools to guide learning and adoption. It is essential to avoid common pitfalls, such as discussing
negative stereotypes during interviews. According to Danaher and Crandall (2008), simply
discussing negative stereotypes by a person not included in that stereotype reflect an implicit bias
that changes the perception of the candidate by the interviewer as well as the actual performance
of the interviewee. This includes when the discussion is meant to be positive, e.g., “most women
struggle in this kind of environment, but you are not having any issues at all.” Awareness
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 77

training coupled with selected information to guide self-learning will alleviate the effects of
implicit bias and help bolster the number of female engineers in the organization.
Increasing performance through reduction of stereotype threat. QuickChip’s male
dominated workforce creates an implicit bias against women in the workplace and degrades
female performance through stereotype threat. The researcher recommended a solution based on
Information Processing Theory as shown in Table 5. To develop mastery, individuals must
acquire component skills, practice integrating them, and know when to apply what they have
learned (Schraw & McCrudden, 2006). This means that cultural issues within QuickChip are
preventing women on their teams from operating at their fullest capacity. QuickChip should
provide training that brings awareness of the effects of stereotype threat and how it affects those
in the workplace, providing effective solutions to reduce it. For example, the organization can
provide team-based training where the effects of Stereotype threat can be demonstrated, followed
by more positive examples where it is reduced.
QuickChip’s training must communicate a message of diversity and inclusion, ensuring
that coworkers’ value everyone’s ideas regardless of background or identity (Purdie-Vaughns,
Steele, Davies, Ditlmann, & Crosby, 2008). The job aids should be posters or digital displays
which provide reminders of how to identify and respond to incidents of stereotype threat and
implicit bias. These material should increase visibility and empower current female minorities to
promote unity and have a shared voice (Murphy, Steele, & Gross, 2007). These two proven key
strategies can reduce stereotype threat. The training should be given to functional teams to
increase its effectiveness and reception (Logel, et al., 2009). The reduction stereotype threat will
increase productivity at QuickChip while increasing retention, bringing it closer to its goal of
equitable representation.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 78

Motivation Recommendations
Introduction. According to Mayer (2011) meaningful learning requires motivation of
the learner. Three key components of motivation are active choice, persistence, and mental
effort (Clark & Estes, 2008). Clark and Estes (2008) described active choice as the conscious
decision to engage in a task over other priorities, persistence is sticking with the task despite
challenges, and mental effort is the investment of knowledge and skills to complete a task. High
value of a task and self-efficacy increase active choice, persistence, and mental effort, and
therefore motivation (Bandura, 3, 2000). Table 6 shows the assumed Motivation influences and
context specific recommendations for the assumed knowledge influence that impact QuickChip’s
mission.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 79

Table 6 - Summary of Motivation Influences and Recommendations
Summary of Motivation Influences with Recommendations
Assumed Motivation
Influence
Principle and Citation Context-Specific
Recommendation
Goal-Orientation -
Managers should set
mastery goals that mirror
the values they would like
to see in the organization.
Goals motivate and direct
students
(Pintrich, 2003).

Link rewards with progress
(Pintrich, 2003).

Goal-Orientation -
Managers should set
mastery goals that mirror
the values they would like
to see in the organization.
Self-Efficacy - Managers
should feel empowered
and confident in their
ability to recruit and retain
female talent.
High self-efficacy can
positively influence motivation
(Pajares, 2006).

Set close, concrete, and
challenging goals that allow the
learner to experience success at
the task (Pajares, 2006).

Make it clear that individuals
are capable of learning what is
being taught or can perform a
task (Pajares, 2006).
Self-Efficacy - Managers
should feel empowered and
confident in their ability to
recruit and retain female
talent.

Challenges in retaining female engineering talent caused by unsupportive culture.
At QuickChip, 80% of women reported they believe the organization has a bias toward male
employees. Half of those who felt the organization carried a bias toward male employees
considered changing careers or leaving the workforce, compared to the 10% of those who did not
believe there was a bias. The researcher’s recommendation presented in Table 7 is based on
goal-orientation theory. According to Pintrich (2003), a way to motivate managers to improve
culture is to link rewards with progress. For QuickChip, this is likely to take the form of
reallocation and restructuring of existing performance bonuses for managers. Recruiting and
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 80

retention targets should be set by senior leaders that are reflective of the organization they are
trying to emulate. Meeting female recruiting goals should factor into the regular compensation
bonuses manager receive.
According to Clark and Estes, (2008) an individual is more likely to start or persist in a
behavior if he or she believes they can accomplish the task and that it has an impact. Turner,
Lasserre, and Beauchet (2007) demonstrated that a properly aligned bonus structure can greatly
improve motivation for the beneficiaries. The authors provide further evidence that independent
metrics increase when tied to financial compensation, which means that if the recruiters can
recruit female employees without sacrificing other portions of their bonus a financial incentive is
likely to be effective. Bishop (2007) also supports that individuals show higher motivation when
there is a financial benefit. By tying financial compensation to the overall number of women in a
managers’ purview, QuickChip can motivate its managers to increase female recruitment and
retention.
Empowering and training managers to instill a culture that attracts women.
Currently, less than 20% of the new engineers at QuickChip are female. The researcher’s
recommendation presented in Table 7 is based on Self-Efficacy theory. According to Pajares
(2006), setting short-term, achievable goals can reinforce managers’ confidence in their ability to
achieve the larger goal. For QuickChip, this means senior leaders must take an active role in
setting these short-term goals. Senior leaders must empower and enable managers to recruit and
retain female talent while providing tools and training to reinforce their skills. Senior leaders can
empower managers through goal-setting with short achievable targets to reinforce managers’
ability to meet the long-term goal.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 81

Clark and Estes (2008) state that believing in one’s ability to meet a goal is a large factor
in determining the desire pursue or meet that goal. By setting short-term, achievable goals that
drive toward the larger goal, senior leaders can establish “little wins,” which increase self-
efficacy by shifting the locus of control from external to internal (Anderman & Anderman,
2009). This is reinforced by Hadsell (2010), who shows that goal achievement leads to
decreased fear of failure and in increase in perception of an internal locus of control. By
achieving short-term goals set by senior leaders, managers can increase their own self-efficacy,
helping achieve long-term success.
Organization Recommendation
Introduction. Influences which affect an organization’s ability to adapt through change,
develop efficient processes, or a healthy organizational culture which supports the organizational
mission are factors to consider when identifying organizational influences (Clark & Estes, 2008).
Clark and Estes (2008) also state that the goals of the organization must be aligned to address the
gaps in the organization. DiTomaso, Post, and Parks-Yancy (2007) further show that senior
leaders’ actions towards meeting aligned goal can shift the culture for the entire organization.
Table 7 shows the assumed organizational influences and context specific recommendations for
the assumed influence that impact QuickChip’s mission.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 82

Table 7 - Summary of Organization Influences and Recommendations
Summary of Organization Influences and Recommendations
Assumed Motivation
Influence
Principle and Citation Context-Specific Recommendation
Software
engineering is a
male dominated
field, and not seen
by many as a
female-friendly
profession.
Effective leaders demonstrate
a commitment to valuing
diversity through inclusive
action. They promote an
organizational culture that
promotes equity and inclusion
(Angeline, 2011).
QuickChip should enhance its gender
diversity programs to include
educational materials that are
applicable for the entire organization
(male and female). QuickChip
should reinforce this educational
information through affirmative
action.
There is a shortage
of female candidates
with computer
science degrees.
Leaders are more accountable
when accountability is framed
both internally and externally,
and takes many forms.
Accountability is contextually
defined (Hentschke &
Wohlstetter, 2004).

Teach recruiters and leaders that the
number of female candidates with
computer science degrees is tied to the
number of women already in the
career field. While there is a five to
nine-year lag, non-traditional
recruitment and aggressive retention
are critical for breaking the cycle.
QuickChip is male
dominated, and
therefore likely to a
show a bias against
women and less
likely to hire a
female candidate.
Effective leaders are aware of
biases that occur in the
organization at the individual
and structural levels. They
acknowledge their own biases
and protect the organization
from their negative impact.
(Bensimon, 2005).
Provided Implicit association testing
for all employees to make them aware
of their own implicit biases, followed
by educational material that discusses
the impact of implicit biases and how
to counteract it.
QuickChip’s senior
management has not
placed enough
emphasis on
recruiting and
retaining female
talent outside the
recruitment team.
Effective leaders promote
diversity at the highest levels
of the organization.
(DiTomaso, Post & Parks-
Yancy, 2007)
Senior Management should take an
active role in attracting and retaining
female talent. Currently, management
acknowledges the issue, but pushes
the responsibility to recruiters. Active
involvement by leadership signals a
cultural significance to the rest of the
organization.

THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 83

Influencing cultural change from the top. QuickChip’s engineering workforce is 20%
female, while the workforce on a national level is 48% female (Equal Employment Opportunity
Commission, 2016). According to Bensimon (2005), leaders must be aware of both explicit and
implicit biases within their organizations and take steps to remediate negative behaviors. This
means that organizational change must include tangible action from senior leadership.
QuickChip should enhance its gender diversity programs to include educational materials that are
applicable for the entire organization (all genders) and the short-term and long-term benefits of
having a diverse workforce. QuickChip should reinforce this educational informative through
affirmative action to reduce stereotype threat and implicit bias within the organization.
Women face stereotype threat and implicit bias because of the gender imbalance at
QuickChip (Murphy, Steele, & Gross, 2007). Through the expenditure of time, staff, and
money, senior leaders can demonstrate their commitment to diversity and reinforce the
importance of diversity and promote a culture of awareness and acceptance. According to
Petriglieri (2011), an effective way to address social threat is to demonstrate how important the
afflicted identity is to the organization. Furthermore, establishing affirmative action steps that
allow women to make a visible impact will also positively influence the effects of identity threat
within the organization (Holmes et al., 2016). By educating all employees of the benefits of a
diverse resource and how women positively impact the organization, coupled with affirmative
action to increase the presence of women throughout the organization, leading to the reduction of
stereotype threat and implicit bias.
Shifting to long-term, internal locus of control. According to the National Science
Foundation (2016), only 20% of computer science graduates are female, confirming that
QuickChip’s workforce closely resembles the talent pool. Table 7 shows principles and
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 84

recommendations based on attribution theory. Accountability should be framed both internally
and external, and should be measured differently depending on the context (Hentschke &
Wohlstetter, 2004). This confirms that QuickChip cannot simply use the external talent pool as
its metric for success, but must also have accountability for internal metrics. QuickChip must
teach recruiters and leaders that the number of female candidates with a computer science degree
is tied to the number of women already in the career field. Currently, there is a five to nine-year
lag, indicating that non-traditional recruitment such as code academies and aggressive retention
strategies like mandatory new-parent leave are critical for breaking the cycle.
While QuickChip’s workforce is representative of the external talent pool, it does not
meet the organization’s internal demands. According to Yough and Anderman (2006), by
shifting the focus to improvement and avoiding external metrics, performance can be improved.
This also internally shifts the locus of control from basing goals on the talent pools to internal
goals set solely on achieving a truly equitable workforce. Schraw and Lehman (2009), show that
explicitly demonstrating the value of accomplishing the task can increase the motivation to meet
the goal. Using alternative benchmarking that use internal goals with less focus on external
factors will help QuickChip achieve its goals (Bogue & Hall, 2003). By showing this is a cyclic
issue, where meeting goals today contribute to what appears to be an impossible long-term goal,
shifts the locus of control internally, and supports QuickChip achieving its mission.
Integrated Implementation and Evaluation Plan
Implementation and Evaluation Framework
The researcher will use the New World Kirkpatrick Model (Kirkpatrick & Kirkpatrick,
2016) as the framework for creating the implementation and evaluation plan for this study. The
original Kirkpatrick model had four levels: reaction, learning, behavior, and results.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 85

Organizations used these levels based on Kirkpatrick’s original publication of the Evaluation
Training Programs (Kirkpatrick, 1994). Kirkpatrick’s son and daughter-in-law published the
New World Order in 2016 which offers a flipped model in which “the end is the beginning”
(Kirkpatrick & Kirkpatrick, 2016).
The New World Model approaches the levels starting from the level four, looking at the
results of programs first and initial reaction last. In the Kirkpatrick model, level four looks at the
impact the program had on the organization. Level three examines the behavior of individuals to
see if the desired changes occurred. Level two checks to see if participants learned the material
presented. Finally, level one checks the respondent’s reaction and engagement to the training
(Kirkpatrick, 1994).
The New World Model prescribes development of the evaluation tools in conjunction
with the program. This allows for greater alignment of the program to the desired end-state by
ensuring the designer keeps the end in mind. As the program continues developments and
improves, the designer can focus on the outcomes at level three and below (Kirkpatrick &
Kirkpatrick, 2016).
Organizational Purpose, Need and Expectations
QuickChip’s mission is to help make the future available to all by bringing the future to
all. This lofty, long-term mission demonstrates QuickChip’s desire to benefit society in the long-
term by being a positive change and catalyst. However, QuickChip is unable to fill its current
engineering needs and the gap is projected to get wider.
QuickChip is currently around 20% female engineers, with the goal to reach 50% by the
year 2030, increasing 1.5% each year until it achieves equitable representation. This is in
addition to its goals of increasing its workforce capacity, meaning that QuickChip must both
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 86

recruit new female talent in addition to the male talent it is already bringing in, and retain its
existing talent to meet future demand.
QuickChip could create a surplus of engineers if it had the same number of female
engineers as men. However, it is difficult to recruit women when the organization does not have
many women. The desired outcome of the recommendations is to increase recruitment and
retention of female talent to bolster its current number, and put it on a path for long-term success
Level 4: Results and Leading Indicators
There are several short-term metrics that would indicate the recommendations are
achieving the desired results. Internally, the metrics are measured from the female engineers
currently working at QuickChip to see if there in an increase in retention and performance due to
the removal of negative environmental factors. Externally, QuickChip can determine if the
programs are having the desired effect by measuring female engagement and new hires. Table 8
lists the desired level four outcomes, metrics, and the methods used to collect the data.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 87

Table 8 - Outcomes, Metrics, and Methods for External and Internal Outcomes
Outcomes, Metrics, and Methods for External and Internal Outcomes
Outcome Metric(s) Method(s)
External Outcomes
Increased engagement of
female engineers in job
postings.
Number of female applicants to
engineering positions.
Senior management will data-
mine the human resource
information system to determine
the increase or decrease in the
total number of female
applicants to engineering
positions.
Increased recruitment of
female new hires.
Number of female new hires. Senior management will data-
mine the human resource
information system to determine
the increase or decrease in the
number of female new hires.
Internal Outcomes
Increased retention of
female engineers.
Annual turnover percentage of
female engineers.
Senior management will data-
mine the human resource
information system to determine
the increase or decrease of
annual turnover of female
engineers.
Increased performance by
female engineers
Performance scores based on
annual review of female
engineers.
Senior management will data-
mine the human resource
information system to determine
the increase or decrease of
performance ratings of female
engineers.

Level 3: Behavior
Critical behaviors. QuickChip has several metrics against which it can measure the
level three impact of the recommendations. Most are measured from the female engineers
currently working at QuickChip to see if the desired shift in their feelings of implicit bias,
stereotype threat, identity threat, and ambient belonging. There are also metrics to determine
whether the recruiters’ training had the desired impact on their behavior. Finally, measuring the
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 88

implicit bias of those in the organization again would demonstrate if the program had the desired
effect on QuickChip’s internal beliefs. Externally, QuickChip can determine if the programs are
having the desired effect by measuring increases of female engagement. Table 9 lists the desired
level three outcomes, metrics, and the methods used to collect the data.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 89

Table 9 - Critical Behaviors, Metrics, Methods, and Timing for Evaluation
Critical Behaviors, Metrics, Methods, and Timing for Evaluation
Critical Behavior Metric(s) Method(s) Timing
1. Reduction in
perceived stereotype
threat among female
engineers
Reduced stereotype
threat score based on
perception of female
engineers at
QuickChip
Survey - re-conduct the
survey and measure
change in the stereotype
threat score.
Recurring every 6
months
2. Reduction in
perceived identity
threat among female
engineers
Reduced identity
threat score based on
perception of female
engineers at
QuickChip
Survey - re-conduct the
survey and measure
change in the identity
threat score
Recurring every 6
months
3. Increase in ambient
belonging among
female engineers
Reduced stereotype
threat score based on
perception of female
engineers at
QuickChip
Survey - re-conduct the
survey and measure
change in the ambient
belonging score
Recurring every 6
months
4. Reduction in
implicit bias
Improved
performance scores
for female engineers,
approaching equity in
distribution between
male and female
engineers
Data-mine - Analyze
the human resource
information system for
variances between male
and female engineers.
Recurring every 6
months
5. Increase in the
number of females
considered for
position at QuickChip
5a. Recruiter
engagement data
identifies an increase
in the number of
females recruited for
engineering positions

5b. Increase in
number of females
interviewed for
positions at
QuickChip
5a. Data-mine -
Analyze the human
resource information
system to identify
variances in recruiters’
interaction with female
engineers

5b. Data-mine -
Analyze the human
resource information
system to identify
variances in the
percentage of females
selected for interviews
Recurring every 6
months
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 90

Required drivers. Kirkpatrick and Kirkpatrick (2016) added required drivers as a new
dimension in the New World Model. Required drivers add a layer of accountability and
reinforce the program to ensure implementation through organizational factors such as
monitoring and encouragement. Because this is such a complex and long-term initiative,
meeting the goal requires substantial organizational support. Table 10 describes the
recommended drivers and timing to support the required behaviors.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 91

Table 10 - Required Drivers to Support Critical Behaviors
Required Drivers to Support Critical Behaviors
Method(s) Timing
Critical
Behaviors
Supported
Reinforcing
Senior Management will enhance the Affirmative Action
program to place female leaders into higher visibility roles
As needed 2, 4
Team-level managers will conduct periodic re-education
based on curriculum provided by Human Resources
Annual 2, 4
Senior management will send out inclusion-awareness
resources
Annual 1-3
Team-level managers will lead discussion of implicit bias Annual 4
Senior management will conduct outreach to women’s
engineering groups and degree programs
Semi-annual 5
Encouraging
Senior management will set short-term, achievable hiring and
retention goals for recruiters and team-managers to boost
confidence
Quarterly 4-5
Senior Management will remind women of their importance
to the organizations through company-wide communication
Quarterly 2-3
Rewarding
Senior managers factor diversity into regular compensation
bonuses for team-level managers
Annual 1-3, 5
Senior management will send out company-wide recognition
of women’s accomplishments
Quarterly 1-4
Monitoring
Team-level managers will distribute surveys to check for
climate
Semi-annual 1-3
Senior management will review recruiter and hiring manager
interview data
Semi-annual 5

Organizational support. QuickChips’s goal requires long-term commitment from all
facets of the organization. Through application of the required drivers in Table 10, QuickChip
can position itself for long-term success while supporting those accountable to meeting its
mission. Leaders can reinforce the behaviors that contribute to a healthier environment through
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 92

affirmative action, periodic re-education, inclusion awareness, implicit bias training, and
outreach to non-profit technology organizations. QuickChip can encourage the behaviors
through short-term goal setting to boost confidence and communication to all employees
discussing the importance of women to everyone’s success. QuickChip can reward behaviors
that contribute to a healthier work environment by factoring in female hiring and retention goals
into annual performance bonuses. Lastly, QuickChip will be able to monitor progress by re-
conducting climate surveys and through analysis of recruitment and retention data to see if the
programs are having the desired effects.
Level 2: Learning
Learning goals. There are two distinct portions of the program. The first is training for
leaders that will be given to all QuickChip supervisors that teaches about biases and threats
within the recruitment process and workplace climate. The second is training for all employees
to increase awareness of implicit bias and understand how these biases collectively create a toxic
environment for coworkers. Following the leaders’ training, Senior Leadership will be able to:
1. Identify biases and threats within QuickChip that are causing female talent to leave.
2. Identify and implement policies that will help QuickChip become a more female friendly
organization.
3. Communicate the successes of female employees and the need for diversity without
patronization.
4. Create partnerships with female-focused engineering organizations.
Following the leaders’ training, team-level managers will be able to:
1. Identify those who are most affected and most at risk of leaving QuickChip because of a
negative culture and provide intervention.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 93

2. Conduct job interviews free of cues that may trigger identity or stereotype threat.
3. Assess candidates with reduced implicit bias.
4. Identify and correct behaviors of employees who contribute to a negative work climate.
Following the company-wide awareness training, all employees will be able to:
1. Identify implicit bias and describe how it affects everyone.
2. Identify Stereotype and Identity threat and describe how those affected are negatively
impacted.
3. Although not further assessed or shared outside of the individual tested, the implicit
association test will tell the respondent the level of implicit bias based on associations.
Training Program. The leaders’ training, which will be for both senior leaders and
managers, will be a 4-hour block of classroom-based discussion in small groups. This training
will cover implicit bias, stereotype threat, and identity threat. While these topics are related, they
are distinct and each has nuances in how to counteract them. The classroom sizes will be no
more than 20 people to facilitate candid discussion while still allowing the instructor to maintain
control while facilitating.
The implicit bias training that will be given to the entire organization will be an informal,
web-based training. The training will begin with information about implicit bias and how it
affects the company, followed by an implicit association test, then ends with how implicit bias
impacts individuals. Placing the implicit association test in the middle allows for the training to
establish the seriousness of the topic before taking the test, then following it up with a more
personal message to appeal to trainees’ humanity.
Evaluation of the components of learning. According to the New World Model, there
are five components of learning: knowledge, skills, attitude, confidence, and commitment
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 94

(Kirkpatrick & Kirkpatrick, 2016). Successful implementation of a training program will deliver
these components to all members of the organization. QuickChip must be able to demonstrate
Table 11 lists the recommended evaluation methods for both the leader and organization-wide
trainings as well as timing.
Table 11 - Evaluation of the Components of Learning for the Program
Evaluation of the Components of Learning for the Program
Method(s) or Activity(ies) Timing
Declarative Knowledge - “I know it.”
Leaders - Knowledge checks with training
facilitators during peer discussion
During live training
Organization - Multiple choice questions to
confirm understanding of stereotype threat,
identity threat, and implicit bias
During online training
Procedural Skills - “I can do it right now.”
Leaders - Demonstrate response to different
types of threats present in the
During live training
Organization - Identify different types of
threats and bias
During online training
Attitude “I believe this is worthwhile.”
All - Discussion on the impacts of threats and
bias in the organization
Leaders - During classroom sessions
Organization - During team-level meetings
Confidence “I think I can do it on the job.”
All - Follow on implicit association tests Six months after initial training, then annually
thereafter.
Commitment “I will do it on the job.”
Leaders - Create a team action plan and goals
for improved climate.
During classroom training sessions
Organization - Create individual goals During online training

Level 1: Reaction
It is important to measure the reaction of participants to understand initial
reactions to training. The reaction shows participants’ engagement, belief in the material’s
relevance to their job, and perceived usefulness. Kirkpatrick and Kirkpatrick (2016) show that
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 95

reactions can be measured by “pulse checks,” surveys, and observations by a leader or third-
party observer. Table 12 shows the methods and timing of evaluations that will be used to
measure the trainees’ reaction to the two different training sessions.
Table 12 - Components to Measure Reactions to the Program
Components to Measure Reactions to the Program.
Method(s) or Tool(s) Timing
Engagement
Leaders - Instructor observations Ongoing during classroom training
Organization - Course module completion based on
analytics of learning platform data
Ongoing
Relevance
Leaders - Brief survey Conclusion of classroom training
Organization - Brief survey Conclusion of online training
Customer Satisfaction
Leaders - Brief survey Conclusion of classroom training
Organization - Brief survey Conclusion of online training

Evaluation Tools
Immediately following the program implementation. Following the classroom
training for managers, the facilitator will conduct a brief questionnaire via paper handouts at the
end of the session to gauge whether participants understood the material and how they felt about
the training. The questionnaire will also include questions about the participant’s believe in the
program’s quality and relevance. These two sections will capture both the level 1 (reaction) and
level 2 (learning) feedback of the Kirkpatrick model. Likewise, the organization-wide training
with a similar questionnaire delivered on the training platform to assess the participants’ reaction
and learning. APPENDIX D contains the questionnaires to be used for both programs.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 96

Delayed for a period after the program implementation. The training facilitators will
send out a follow-on survey via email six weeks after the training period closes. The email will
contain a link to a survey conducted anonymously online. The survey will be the same for all
training recipients and will focus on levels three and four of the Kirkpatrick model (behavior and
results, respectively) while still soliciting feedback on levels one and two, albeit with less
emphasis. By checking for an improvement in behaviors, QuickChip can reinforce the training
principles while also demonstrating its commitment to the program. APPENDIX E contains the
questions to be used on the survey.
Data Analysis and Reporting
The results of both immediate instruments will be collected through Google Forms and
placed in a spreadsheet, where they will be analyzed with the results viewable to senior leaders
and the training facilitators to ensure the program is having the desired level 1 and 2 results.
Likewise, the results of the delayed instrument will be collected through Google Forms and
placed in a spreadsheet to gauge feedback on the expected behaviors and allow senior leadership
the ability to refine and improve the program. Level 4 results will still be gathered through
traditional measures, already gathered using corporate information systems to measure
recruitment and retention and the percentage of female engineers within the organization.
APPENDIX F shows several charts and graphs which convey the results of the Level 1 and 2
analysis.
Summary
The Kirkpatrick and Kirkpatrick (2016) New World Model was used to design the
recommended training. The model provides a framework on which to implement and evaluate
the training. By designing training with the end-goal in mind, and evaluating the training based
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 97

on how those targets are met, QuickChip can create a feedback loop to refine, improve, and
continually measure the effectiveness of the program. The program will also capture elements
important to training, but of lesser focus as to not detract from the overall goal. By designing,
implementing, and evaluating the training with the end-goal in mind, QuickChip will achieve the
strongest possible results from the training program.
Future Research
There are many areas of future research stemming from this project. Most importantly,
research should see if other organizations produce similar correlations from the demographics to
a score. Second, further research is warranted how to effectively hold executive leadership
accountable for change. Lastly, additional research would be beneficial in conducting similar
surveys and scores on different demographics to see if the results are repeatable.
The correlations suggest many things, but the issues are complex. By conducting the
same survey and checking for similar correlations, it could open up avenues for causal research
to see why there is such a sharp decline between 26 and 35 for perceive bias. It could also lead to
a more complex “threat” model demonstrating how likely an organization’s employees are to
experience implicit bias and stereotype threat based on a combination of factors.
Long-term goals are not real goals for short-term executive leadership. Further research
is warranted on how to properly motivate executive leadership through regulations, penalties, or
incentives to make their commitments to diversity more concrete and tangible.
Lastly, implicit bias, stereotype threat, and ambient belonging are not unique issues to
women working in technology companies. These phenomena occur in wherever there are groups
segregated for any social reason such as race, religion, physical ability, or nationality. Therefore,
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 98

it stands to reason the survey and scoring mechanism could also be applied to other organizations
to enhance the reliability and usability of what is currently an arbitrary point system.
Conclusion
Girls lose interest in Computer Science because they think it is something for boys.
Those girls go to college and do not study Computer Science because they are not interested in it.
Those women graduate and work in other fields, continuing the cycle. This leaves women
working in the field underrepresented and working in a noxious environment caused by being
outnumbered four-to-one by men.
This research demonstrates the effects of being heavily outnumbered, and how it
contributes to keeping people from entering the field and causes them to want to leave. A
combination of factors makes women feel out of place where they stand out for being different,
where they are questioned if they are qualified, and must work harder under greater scrutiny by
virtue of being different in the workplace. Meanwhile, these impacts reduce actual performance
and therefore the bottom line of the organization, meaning there is a potential return on
investment for fixing these issues.
There are solutions to this problem that QuickChip can implement. They were successful
in law and medicine, and can work in computer science too. By proving these methods
successful, other companies can apply these recommendations as well and help break the cyclic
lack of female engineers in computer science.
(Denler, Wolters, & Benzon, 2009) (Pintrich, 2003) (Fredricks & Eccles, 2006) (Schraw & McCrudden, 2006) (Kirkpatrick & Kirkpatrick, 2016) (Kirkpatrick D. L., 1994) (Anderson, 2013) (Bandura, 3, 2000) (Clark & Estes, 2008) (Denler, Wolters, & Benzon, 2009) (Fredricks & Eccles, 2006) (Hewlett & Luce, 2005) (Kirkpatrick D. L., 1994) (Kirkpatrick
& Kirkpatrick, 2016) (Krathwohl, 2002) (Logel, et al., 2009) (Mayer, 2011) (Murphy, Steele, & Gross, 2007) (Pajares & Urdan, 2006) (Purdie-Vaughns, Steele, Davies, Ditlmann, & Crosby, 2008) (Schraw & McCrudden, 2006) (Pintrich, 2003) (Anderman & Anderman, 2009) (DiTomaso, Post, & Parks-Yancy, 2007) (Hadsell, 2010) (Angeline, 2011) (Hentschke & Wohlstetter, 2004)
(Bensimon, 2004) (Equal Employment Opportunity Commission, 2016) (Rueda, 2011) (Petriglieri, 2011) (Holmes IV, Whitman, Campell, & Johnson, 2016) (Schraw & Lehman, 2009)

THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 99

REFERENCES
Allen, T. D., & O' Brien, K. E. (2006). Formal mentoring programs and organizational attraction.
Human Resource Development Quarterly, 17(1), 43-58. doi:10.1002/hrdq.1160
Anderman, L. H., & Anderman, E. M. (2009). Oriented toward mastery: Promoting postivie
motivational goals for students. In M. J. Furlong, R. E. Gilman, & E. S. Huebner,
Handbook of Positive Psychology in Schools (pp. 161-173). Abingdon-on-Thames,
United Kingdom: Routledge.
Anderson, L. (2013). A taxonomy for learning, teaching, and assessing: A revision of bloom's
taxonomy of educational objectives. London: Pearson Higher Education.
Angeline, T. (2011, January 18). Managing generational diversity at the workplace: Expectations
and perceptions of different generations. African Journal of Business Management, 5(2),
249-255. doi:10.5897/AJBM10.335
Bandura, A. (2000). Excercise of human agency through collective efficacy. Current Directions
in Psychological Science(9), 75-78. doi:10.1111/1467-8721.00064
Bandura, A., & Locke, E. A. (2008). Negative self-efficacy and goal effects revisited. Journal of
Applied Psychology, 88(1), 87-99.
Bensimon, E. M. (2004). The diversity scorecard: A learning approach to institutional change.
Change: The magazine of higher learning, 36(1), 44-52.
Bureau of Labor and Statistics. (2016). Employed persons by detailed occupation, sex, race, and
Hispanic or Latino ethnicity. U.S. Department of Labor. Washington, DC: U.S.
Government Printing Office. Retrieved July 29, 2017
Bureau of Labor and Statistics. (2016). Employment projections. U.S. Department of Labor.
Washington, DC: U.S. Government Printing Office. Retrieved July 29, 2017
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 100

Cheryan, S., Plaut, V. C., Davies, P. G., & Steele, C. M. (2009). Ambient belonging: How
stereotypical cues impact gender participation in computer science. Journal of
Personality and Social Psychology, 97(6), 1045-1060.
Clark, R. E., & Estes, F. (2008). Turning research into results: A guide to selecting the right
performance solutions. Charlotte: Information Age Pub Inc.
Craig, R. (2016). Providing students with pathways to high-value careers. Change: The
Magazine of Higher Learning, 58-63.
Creswell, J. W. (2014). Research design: Qualitative, quantitative, and mixed methods
approaches (4th ed.). Thousand Oaks, CA: SAGE Publications.
Danaher, K., & Crandall, C. S. (2008). Stereotype threat in applied settings re‐examined. Journal
of Applied Social Psychology, 38(6), 1639-1655. doi:10.1111/j.1559-1816.2008.00362.x
Degeest, D., & Brown, K. G. (2011). The role of goal orientation in leadership development.
Human Resource Development Quarterly, 22(2), 157-175. doi:10.1002/hrdq.20072
Denler, H., Wolters, C., & Benzon, M. (2009, December 23). Social cognitive theory. Retrieved
from Education.com:
https://project542.weebly.com/uploads/1/7/1/0/17108470/social_cognitive_theory__educ
ation.com.pdf
Denzin, N. (2006). Sociological methods: A sourcebook. Piscataway, NJ: Transaction Publishers.
DiTomaso, N., Post, C., & Parks-Yancy, R. (2007). Workforce diversity and inequality: Power,
status, and numbers. Annual Renview of Sociology, 33, 473-501. doi:33.040406.131805
Equal Employment Opportunity Commission. (2016). Diversity in High Tech. Washington:
Equal Employment Opportunity Commission.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 101

Erez, M., & Zidon, I. (1984). Effect of goal acceptance on the relationship of goal difficulty to
performance. Journal of Applied Psychology, 69(1), 69-78.
Fink, A. (2015). How to conduct surveys: A step-by-step guide. Thousand Oaks, CA: Sage.
Fredricks, J. A., & Eccles, J. S. (2006). Is extracurricular participation associated with beneficial
outcomes? Concurrent and longitudinal relations.(Author abstract). Developmental
Psychology,, 42(4), 698-714.
Freeman, M. (2016, July 26). Qualcomm enters $19.5 million gender bias settlement. The San
Diego Union-Tribune.
Gaucher, D., Friesen, J., & Kay, A. C. (2011). Evidence that gendered wording in job
advertisements exists. Journal of Personality and Social Psychology, 101(1), 109-128.
doi:10.1037/a0022530
Hadsell, L. (2010, May). Achievement goals, locus of control, and academic success in
economics. American Economic Review, 100(2), 272-276. doi:10.1257/2.272
Hentschke, G. C., & Wohlstetter, P. (2004). Cracking the code of accountability. University of
Southern California Urban Education, 17-19.
Hewlett, S. A., & Luce, C. B. (2005). Off-ramp and on-ramps: Keeping talented women on the
road to success. Harvard Business Review, 83, 43-54.
Holmes IV, O., Whitman, M. V., Campell, K. S., & Johnson, D. E. (2016). Exploring the social
identity threat response framework. Equality, Diversity and Inclusion: An International
Journal, 35(3), 205-220.
Hunter, J. E. (2002). Methods of meta-analysis: Correcting error and bias in research findings.
Thousand Oaks, CA: SAGE.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 102

Hurmerinta-Peltomaki, L., & Nummela, N. (2006). Mixed methods in international business
research: A value-added perspective. Management International Review, 46(4), 439-459.
Indeed. (2017, May 2). What do employers really think about coding bootcamps? Retrieved from
Indeed Blog: http://blog.indeed.com/2017/05/02/what-employers-think-about-coding-
bootcamp/
Ivankova, N. V., Creswell, J. W., & Stick, S. L. (2006). Using mixed-methods sequential
explanatory design: From theory to practice. Field Methods, 18(1), 3-20.
Jackson, S. M., Hillard, A. L., & Schneider, T. R. (2014, September). Using implicit bias training
to improve attitudes toward women in STEM. Social Psychology of Education : An
International Journal, 7(3), 419-438. doi:10.1007/s11218-014-9259-5
Kirkpatrick, D. L. (1994). Evaluating training programs: The four levels. San Francisco: Berrett-
Koehler Publishers.
Kirkpatrick, W. K., & Kirkpatrick, J. D. (2016). Kirkpatrick’s four levels of training evaluation.
Alexandria, VA: Association for Talent Development.
Krathwohl, D. R. (2002). A revision of Bloom's taxonomy: An overview. Theory Into Practice,
41(4), 212-218. doi:10.1207/s15430421tip4104_2
Latu, I. M., Mast, M. S., & Stewart, T. L. (2015). Gender biases in (inter) action: The role of
interviewers’ and applicants’ implicit and explicit stereotypes in predicting women’s job
interview outcomes. Psychology of Women Quarterly, 39(4), 539-552.
doi:10.1177/0361684315577383
Logel, C., Walton, G. M., Spencer, S. J., Iserman, E. C., Von Hippel, W., & Bell, A. E. (2009).
Interacting with sexist men triggers social identity threat among female engineers.
Journal of Personality and Social Psychology, 96(6), 1089-1103. doi:10.1037/a0015703
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 103

Mason, M. A., & Goulden, M. (2008). Do babies matter? The effect of family formation on the
lifelong careers of academic men and women. Academe, 21-17.
Master, A., Cheryan, S., & Meltzoff, A. N. (2016). Computing whether she belongs: Stereotypes
undermine girls' interest and sense of belonging in computer science. Journal of
Educational Psychology, 108(3), 424-437.
Maxwell, J. A. (2013). Qualitative research design: An interactive approach, 3rd ed. Thousand
Oaks, CA: Sage.
Mayer, R. E. (2011). Applying the science of learning. Boston: Pearson/Allyn & Bacon.
McCuller, P. (2012). How to recruit and hire great software engineers: Building a crack
development team. New York City: Apress.
Merriam, S. B., & Tisdell, E. J. (2015). Qualitative research: A guide to design and. San
Francisco: Jossey-Bass.
Microsoft Corporation. (2017). Why Europe's girls aren't studying STEM. Redmond: Microsoft.
Morgenroth, T., & Heilman, M. E. (2017, September). Should I stay or should I go? Implications
of maternity leave choice for perceptions of working mothers. Journal of Experimental
Social Psychology, 72, 53-56. doi:10.1016/j.jesp.2017.04.008
Murphy, M. C., Steele, C. M., & Gross, J. J. (2007). Signaling threat: How situational cues affect
women in math, science, and engineering settings. Psychological Science, 18(10).
doi:10.1111/j.1467-9280.2007.01995.x
National Center for Science and Engineering Statistics. (2016). Employed scientists and
engineers, by ethnicity, race, occupation, highest degree level, and sex: 2015.
Washington: National Science Foundation.
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 104

Pajares, F., & Urdan, T. (2006). Self-efficacy beliefs of adolescents. Greenwich, CT: Information
Age Publishng Inc.
Payne, B. K., & Vuletich, H. A. (2017, December 21). Policy insights from advances in implicit
bias research. Policy Insights from the Behavioral and Brain Sciences, 5(1), 49-56.
doi:10.1177/2372732217746190
Petriglieri, J. L. (2011). Under threat: Responses to and the consequences of threats to
individuals' identities. Academy of Management Review, 36(4), 641-662.
Pintrich, P. (2003). A motivational science perspective on the role of student motivation in
learning and teaching contexts. Journal of Educational Psychology, 95(4), 667-686.
doi:10.1037/0022-0663.95.4.667
Purdie-Vaughns, V., Steele, C., Davies, P., Ditlmann, R., & Crosby, J. (2008). Social identity
contingencies: how diversity cues signal threat or safety for African Americans in
mainstream institutions. Journal of personality and social psychology, 94(4), 615-630.
doi:10.1037/0022-3514.94.4.615
Qualcomm Incorporated. (2017). EEO-1 Report. San Diego.
Romzek, B. S., & Dubnick, M. J. (1987). Accountability in the public sector: Lessons from the
challenger tragedy. Public Administration Review, 43(3), 227-238.
Rueda, R. (2011). The 3 dimensions of improving student performance : finding the right
solutions to the right problems. New York City: Teachers College Press.
Rydell, R. J., Rydell, M. T., & Boucher, K. L. (2010). The effect of negative performance
stereotypes on learning. Journal of Personality and Social Psychology, 99(6), 883-896.
doi:10.1037/a0021139
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 105

Sadler, P. M., Sonnert, G., Hazari, Z., & Tai, R. (2012). Stability and volatility of STEM career
interest in high school: A gender study. Science Education, 96, 411-427.
doi:10.1002/sce.21007
Schmader, T., Johns, M., & Forbes, C. (2008). An integrated process model of stereotype threat
effects on performance. Psychological Review, 336-356. doi: 10.1037/0033-
295X.115.2.336
Schraw, G., & Lehman, S. (2009). The use of adjunct displays to facilitarte comprehemsion of
causal relationships in expository text. Instructional Science, 37(1), 65-86.
Schraw, G., & McCrudden, M. T. (2006). Relevance and goal-focusing in text processing.
Educational Psychology Revie, 19, 113-139. doi:10.1007/s10648-006-9010-7
Stahl, W. S. (2011). Informed consent. In H. R. Feldman, Nursing Leadership (2nd ed.). New
York City: Spring Publishing Company.
U.S. Department of Education. (2017, October). Integrated postsecondary education data system
(IPEDS). Retrieved from National Center for Education Statistics:
https://nces.ed.gov/programs/digest/d18/tables/dt18_322.50.asp?current=yes
VandeWalle, D. (1997). Development and validation of a work domain goal orientation
instrument. Educational and Psychological Measurement, 57(6), 995-1015.
doi:10.1177/0013164497057006009

THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 106

APPENDIX A: COPY OF THE EMAIL SENT TO THE SOUTHWESTERN US QCHEW BOARDS
COPY OF THE EMAIL SENT TO THE SOUTHWESTERN US QCHEW BOARDS
***********************
A doctoral candidate at the University of Southern California is conducting
research on the social factors impacting women in computer science and engineering
positions. Part of this study is looking at how individual perception of the company's
culture impacts performance, retention, and recruitment. [REDACTED] is going to take
part in the study, and USC is asking for the members of [REDACTED] to partake in the
study. The study is anonymous and voluntary, and should take about 10-15 minutes.

The survey is available here:
https://forms.gle/p2ZeyYm1AvahU9rd7

***********************
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 107

APPENDIX B: COPY OF THE INFORMED CONSENT
COPY OF THE INFORMED CONSENT

INFORMED CONSENT NOTIFICATION
This is a research project being conducted by Justin Cox, a Doctoral Candidate at the University
of Southern California.

PARTICIPATION
Your participation in this survey is voluntary. You may refuse to take part in the research
or exit the survey at any time without penalty. You are free to decline to answer any particular
question you do not wish to answer for any reason.

BENEFITS
You will receive no direct benefits from participating in this research study. However,
your responses may help us learn more about the culture at your organization and how it affects
the career field overall.

RISKS
There are no physical risks involved in participating in this study other than those
encountered in day-to-day life. There is the risk that you may find some of the questions to be
sensitive.

THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 108

CONFIDENTIALITY
Your survey answers will be stored in a password protected electronic format. The
survey does not collect identifying information such as your name, email address, or IP address.
Therefore, your responses will remain anonymous. No one will be able to identify you or your
answers, and no one will know whether or not you participated in the study.

At the end of the survey you will be asked if you are interested in participating in an
additional interview. If you choose to provide contact information such as your phone number or
email address, your survey responses may no longer be anonymous to the researcher. However,
no names or identifying information would be included in any publications or presentations
based on these data, and your responses to this survey will remain confidential.

CONTACT
If you have questions at any time about the study or the procedures, you may contact the
researcher, Justin Cox via email at coxjusti@usc.edu.

IF YOU DO NOT AGREE, PLEASE CLOSE THE SURVEY

THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 109

APPENDIX C: COPY OF SURVEY QUESTIONS
COPY OF SURVEY QUESTIONS
Qualifying Questions
1. Check here if you consent to the survey. You also acknowledge you may close the survey
at any time without penalty. *
Check all that apply.
I Consent
2. Are you at least 18 years of age?
Mark only one oval.
Yes
No
3. What kind of position(s) do you hold at QuickChip
Check all that apply.
Engineering
Administrative (Administrative, HR, Finance, procurement, etc.)
Information Technology
Project/Product Managemenet
Other:
4. Have you been employed at QuickChip for the last 12 months?
Mark only one oval.
Yes
No
5. Are you cisgender female? (I.e. Are you someone who identifies as a woman and was
assigned female at birth?)
Mark only one oval.
Yes
No

Research Questions
1. Please rate how you agree with the following statement, “My gender positively impacted
my recruiting experience at QuickChip.”
Mark only one oval.
Strongly Agree
Agree
Neither Agree nor Disagree
Disagree
Strongly Disagree
Not Applicable
Unsure

THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 110

2. Please rate how you agree with the following statement, “I did not feel uncomfortable
because of my gender during the interviews with my hiring manager.”
Mark only one oval.
Strongly Agree
Agree
Neither Agree nor Disagree
Disagree
Strongly Disagree
Not Applicable
Unsure
3. Please rate how you agree with the following statement, “I am given the same
opportunities as
my male counterparts.”
Mark only one oval.
Strongly Agree
Agree
Neither Agree nor Disagree
Disagree
Strongly Disagree
Not Applicable
Unsure
4. Please rate how you agree with the following statement, “The executive leadership at
QuickChip are taking enough action to ensure women are treated equitably.”
Mark only one oval.
Strongly Agree
Agree
Neither Agree nor Disagree
Disagree
Strongly Disagree
Not Applicable
Unsure
5. Please rate how you agree with the following statement, “If I were to have a newborn or
adopt a child, my colleagues would question my commitment to the team if I took off 12
weeks of unpaid leave under the Family Medical Leave Act (FMLA).”
Mark only one oval.
Strongly Agree
Agree
Neither Agree nor Disagree
Disagree
Strongly Disagree
Unsure

THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 111

6. Please rate how you agree with the following statement, “If I were to have a newborn or
adopt a child, my colleagues would question my role as a mother if I DECLINE 12 weeks of
unpaid leave under the Family Medical Leave Act (FMLA).”
Mark only one oval.
Strongly Agree
Agree
Neither Agree nor Disagree
Disagree
Strongly Disagree
Unsure
7. Do you feel QuickChip as a whole has any bias towards male or female employees?
Mark only one oval.
High bias toward male employees
Moderate bias toward male employees
About the same
Moderate bias toward female employees
High bias toward female employees
Unsure
8. If your manager was assembling a team, do you feel she or he would prefer to work with
male or female employees?
Mark only one oval.
Would build a team with all males
Would build a team with mostly males
Would build a team about balanced
Would build a team with mostly females
Would build a team with all females
Unsure
9. Please rate how you agree with the following statement, “I feel like I am the stereotypical
engineer at QuickChip.”
Mark only one oval.
Strongly Agree
Agree
Neither Agree nor Disagree
Disagree
Strongly Disagree
Not Applicable
Unsure

THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 112

10. Please rate how you agree with the following statement, “I consider how my words or
actions
might be perceived because of my gender before speaking or acting.”
Mark only one oval.
Strongly Agree
Agree
Neither Agree nor Disagree
Disagree
Strongly Disagree
Unsure

11. Please rate how you agree with the following statement, “There is a positive stereotype
for
female engineers at QuickChip.”
Mark only one oval.
Strongly Agree
Agree
Neither Agree nor Disagree
Disagree
Strongly Disagree
Unsure
12. Please rate how you agree with the following statement, “I have considered leaving my
career
in computer science for other professions or to stay at home.”
Mark only one oval.
Strongly Agree
Agree
Neither Agree nor Disagree
Disagree
Strongly Disagree
Unsure

Demographic Questions
1. What is your race?
Check all that apply.
Black/African American
White
Native American/Alaskan Native
Asian
Hawaiian Native/Pacific Islander
Other

THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 113

2. What is your Ethnicity?
Mark only one oval.
Hispanic or Latino (A person of Cuban, Mexican, Puerto Rican, South or Central
American or other Spanish culture or origin regardless of race)
Not Hispanic or Latino
3. What is your age?
Mark only one oval.
18-25
26-30
31-35
36-40
41-45
46-50
51-55
56-60
60+
4. What is the highest level degree you have completed?
Mark only one oval.
Less than HS Diploma
HS Diploma
Associates Degree
Bachelors Degree
Masters Degree
Research Doctorate (Ph.D., Ed.D, D. Eng, D.B.A, etc.)
Professional Doctorate (M.D., J.D., D. Min, etc.)
5. What level is your current role?
Mark only one oval.
Entry-Level
Intermediate Individual Contributor
Senior Individual Contributor
First-Line Management
Middle-Management
Senior Management (VP, Director)
Executive Leadership

Interview Questions
1. In order to better understand the culture and responses the survey, would be willing to
participate in either an in-person or telephone interview? The interview would last 30-45
minutes.
Mark only one oval.
Yes
No

THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 114

2. What is the best email address to contact you if you are selected for an interview? (This
email address will only be used to contact you if selected for an interview, it will not be used
for any additional purposes, or linked to your survey responses in any way).
Short Answer
THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 115

APPENDIX D: PROPOSED IMMEDIATE FEEDBACK SURVEY QUESTIONS
PROPOSED IMMEDIATE FEEDBACK SURVEY QUESTIONS
Leaders’ Training Feedback and Assessment
These questions are designed to assess whether the leaders who attended the training
learned about the material that was recommended.
1. Please rate how you agree with the following statement, “I found the training relevant to
my team’s work.”
a. Strongly Agree
b. Agree
c. Netural
d. Disagree
e. Strongly Disagree

2. Please rate how you agree with the following statement, “I believe the topic of the
training is important.”
a. Strongly Agree
b. Agree
c. Netural
d. Disagree
e. Strongly Disagree

3. Please rate how you agree with the following statement, “The training was engaging.”
a. Strongly Agree
b. Agree
c. Netural
d. Disagree
e. Strongly Disagree

4. Please rate how you agree with the following statement, “This training is worthwhile.”
a. Strongly Agree
b. Agree
c. Netural
d. Disagree
e. Strongly Disagree

THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 116

5. Match the following terms with the correct definition. (Correct answers in curly braces)
____: Implicit Bias {B}
____: Stereotype Threat {A}
____: Identify Threat {C}
____: Ambient Belonging {D}
A. The risk of confirming negative stereotypes about an individual's racial, ethnic,
gender, cultural group, or other collective identity status.
B. A situationally triggered concern caused by hyper-awareness that one is at risk of
stigmatizing a collective to which they belong.
C. Attitudes or stereotypes that affect our understanding, actions, and decisions in an
unconscious manner.
D. Feeling of fit in the environment, and similarity to the people imagined to occupy
it.

6. Jason is male daycare worker. One day, the principal tells Jason that generally speaking,
parents are less trusting of male employees because statistically, males are 10 times more
likely to abuse a child. As a result, Jason is less interactive with the children, is more
guarded while talking to parents, and is careful to not make contact with children. This is
an example of: (Correct answer in italics)
a. Stereotype Threat
b. Identity Threat
c. Implicit Bias
d. Ambient Belonging

7. Rebecca is a female forklift driver who works at a recycling center that is predominantly
male. On Fridays, the male employees go out for drinks, but don’t usually invite Rebecca
because it would change the group dynamic. What does this behavior deprive Rebecca of:
a. Stereotype Threat
b. Identity Threat
c. Implicit Bias
d. Ambient Belonging

8. Nancy is a 64-year old female triathlete who works as an accountant in an IT firm.
Elizabeth is coordinating runners for a relay, but does not invite Nancy, assuming the
event would be too strenuous. This is an example of:
a. Stereotype Threat
b. Identity Threat
c. Implicit Bias
d. Ambient Belonging

THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 117

Organization Training Feedback and Assessment
These questions are designed to assess whether the training delivered to the entire
organization was effective in teaching the material.
1. Please rate how you agree with the following statement, “I found the training relevant to
my work.”
a. Strongly Agree
b. Agree
c. Netural
d. Disagree
e. Strongly Disagree

2. Please rate how you agree with the following statement, “I believe the topic of the
training is important.”
a. Strongly Agree
b. Agree
c. Netural
d. Disagree
e. Strongly Disagree

3. Please rate how you agree with the following statement, “The training was engaging.”
a. Strongly Agree
b. Agree
c. Netural
d. Disagree
e. Strongly Disagree

4. Please rate how you agree with the following statement, “This training is worthwhile.”
a. Strongly Agree
b. Agree
c. Netural
d. Disagree
e. Strongly Disagree

5. According to most research, who holds implicit bias against people.
a. Only people who have not been trained on Implicit Bias
b. Only people in the majority group (I.e. those not likely negatively impacted by
implicit bias)
c. Only people who acknowledge they have a bias
d. Everyone

THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 118

6. Which of these is an example of Implicit Bias
a. Trent fought in the war against the Empire that ended over 50 years followed by
decades of peace and cooperation. He tells everyone not to trust anyone from the
Empire, even the children.
b. Klingons are twice as likely as everyone else to commit a crime, therefore the
police patrol Klingon territories vigilantly and are defensive when dealing with
any Klingon.
c. Denise just met Sharon on a flight from San Francisco to Denver. Sharon
proudly says that her child is actually the pilot of that flight. Denise is then
embarrassed after asking “what’s his name,” not realizing it was Sharon’s
daughter.
d. Coach is looking for a new Quarterback and is only considering candidates who
are taller than 6’3”.

THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 119

APPENDIX E: DELAYED FEEDBACK SURVEY QUESTIONS
DELAYED FEEDBACK SURVEY QUESTIONS
These questions are designed to gauge the level interest and retained knowledge within
two weeks after the training concluded.
1. Please rate how you agree with the following statement, “I found the training relevant to
my team’s work.”
a. Strongly Agree
b. Agree
c. Netural
d. Disagree
e. Strongly Disagree

2. Please rate how you agree with the following statement, “I believe the topic of the
training is important.”
a. Strongly Agree
b. Agree
c. Netural
d. Disagree
e. Strongly Disagree

3. Please rate how you agree with the following statement, “After the training, I find myself
more readily recognizing implicit bias in my own actions.”
a. Strongly Agree
b. Agree
c. Netural
d. Disagree
e. Strongly Disagree

4. Please rate how you agree with the following statement, “After the training, I can better
identify and address implicit bias at QuickChip.”
a. Strongly Agree
b. Agree
c. Netural
d. Disagree
e. Strongly Disagree

THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 120

5. Please rate how you agree with the following statement, “I believe QuickChip is moving
in the right direction towards an inclusive environment for all engineers.”
a. Strongly Agree
b. Agree
c. Netural
d. Disagree
e. Strongly Disagree

6. Do you feel this training has had any affect on the treatment of female engineers at
QuickChip?
a. Strongly Positive shift in treatment
b. Positive shift in treatment
c. No change
d. Negative shift in treatment
e. Strongly Negative shift in treatment

THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 121

APPENDIX F: PROPOSED DATA ANALYSIS CHART
PROPOSED DATA ANALYSIS CHART
These are sample dashboard-level charts which can convey the level 1 and 2 reactions
and knowledge checks.

THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 122

THE CYCLIC LACK OF FEMALE ENGINEERS IN COMPUTER SCIENCE 123

Abstract (if available)

Abstract Men outnumber women in the computer science and engineering field nearly 4-to-1. There is also a gap in the field of over 600,000 positions in the United States. While women enter the field at a far less rate than their male counterparts, they also leave the industry at nearly double the rate as men. This study investigates the culture within organizations and how it contributes to women avoiding and leaving the career field using the Clark and Estes gap analysis framework. This study surveyed 150 women and interviewed three women working at a single, large technology firm to better understand the culture that reflects public perception. The survey results indicated a culture where women are subjected to implicit biases and stereotype against them, and where they were deprived of ambient belonging. Recommendations for designing, implementing, and evaluating solutions to address these issues are presented based on the Kirkpatrick and Kirkpatrick New World Model.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Girls in cybersecurity: an evaluation study

PDF

STEM teacher education: An evaluation study

PDF

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

PDF

The academic implications of providing social emotional learning in K-12: an evaluation study

PDF

Increasing the number of minoritized teachers in the Apex public schools: an evaluation study

PDF

Social emotional learning curriculum implementation: an evaluation study of fidelity

PDF

Narrowing the English learner achievement gap through teacher professional learning and cultural proficiency: an evaluation study

PDF

Characteristics that create a quality early learning center: An evaluation study

PDF

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

PDF

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

PDF

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

PDF

Educator professional development for technology in the classroom: an evaluation study

PDF

Mitigating the low enrollment rates for women in engineering and computer science

PDF

Rethinking Hispanic attrition rates at U.S. post-secondary institutions: an evaluation study conducted at Latino private college

PDF

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

PDF

Discovering implicit academic science norms through the experience of historically underrepresented STEM faculty

PDF

Lack of African-American undergraduate male student retention: an evaluation study on perspectives from academic advisors

PDF

Retaining female field grade officers in the USAF: an evaluative study

PDF

The educational and professional choices of Iranian women in STEM doctoral programs

PDF

Closing the achievement gap for students with disabilities: a focus on instructional differentiation - an evaluation study

Asset Metadata

Creator Cox, Justin Allen (author)

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

School Rossier School of Education

Degree Doctor of Education

Degree Program Organizational Change and Leadership (On Line)

Publication Date 11/20/2019

Defense Date 11/19/2019

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

Tag ambient belonging,big tech,equality,implicit bias,OAI-PMH Harvest,STEM,stereotype threat,women in STEM,women in technology

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Hasan, Leila (committee chair), Burk, John (committee member), Mora-Flores, Eugenia (committee member)

Creator Email cox.justin.a@gmail.com,coxjusti@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-236165

Unique identifier UC11675261

Identifier etd-CoxJustinA-7941.pdf (filename),usctheses-c89-236165 (legacy record id)

Legacy Identifier etd-CoxJustinA-7941.pdf

Dmrecord 236165

Document Type Dissertation

Rights Cox, Justin Allen

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