University of Southern California Dissertations and Theses

Increasing underrepresented minority participation in the STEM PhD programs

(USC Thesis Other)

Increasing underrepresented minority participation in the STEM PhD programs

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Increasing Underrepresented Minority Participation in the STEM PhD Programs
by
Kevin Hsing Tzu Yeung

Rossier School of Education
University of Southern California
A dissertation submitted to the faculty
in partial fulfillment of the requirements for the degree of
Doctor of Education

August 2021

© Copyright by Kevin Hsing Tzu Yeung 2021
All Rights Reserved

The Committee for Kevin Hsing Tzu Yeung certifies the approval of this Dissertation

Raquel Torres-Retana
Tatiana Melguizo
Bryant Adibe, Committee Chair

Rossier School of Education
University of Southern California
2021

iv
Abstract
According to 2014 data, underrepresented minorities (URM) made up 36.8% of America’s
college age (18-24) population (NSF, 2017b), but only received 12.2% of PhD degrees (NSF,
2017a) and a scant 7.9% of all PhD degrees in the science and engineering fields (NSF, 2017a).
The figures mean that URM received only a quarter of the science and engineering PhD degrees
relative to their percentage makeup of America’s college age population. At UCSF though, the
percentage of URM enrolled in UCSF’s basic and biomedical sciences PhD programs is 22%
(UCSF Graduate Division, 2020a). UCSF was chosen for this evaluation study because the
percentage of URM students enrolled in its basic and biomedical sciences PhD programs is
nearly three times the national average and UCSF is classified as a “very high research activity”
or R1 institution by the Carnegie Classification (Carnegie Classification of Institutions of Higher
Education by Indiana University Center for Postsecondary Research, 2017). R1 institutions
award the most science and engineering doctorate degrees (London et al., 2014).
The study will utilize the Clark and Estes’ (2008) gap analysis framework to evaluate the
knowledge, motivation, and organizational (KMO) influences from the student perspective. The
research methodology uses a quantitative approach consisting of an online survey sent to URM
students enrolled in the basic and biomedical sciences PhD programs. The study concludes with
a series of recommendations to help UCSF ensure its continued success with a more diverse
student body population in the basic and biomedical sciences PhD programs.
Keywords: URM, PhD, STEM, KMO, UCSF

v
Dedication
To my parents, Jim Chung Hao Yeung, and Helen Hsin Erh Yeung, who emigrated to this
country without a cent to their name and spoke not a word of English. Through their hard work,
love, care, dedication, and huge personal sacrifices, they provided and nurtured us children with
a very warm and loving home so that we never, ever felt anything lacking or wanting.
To my beautiful and lovely daughter Olivia, my bundle of joy and pride, your laughs,
playfulness, love, and happiness inspired and motivated me to complete my dissertation.
To my wife Betty who made huge personal sacrifices and endured hardships to allow me to focus
and immerse myself with my work and my education during the past 3 years
To Cui Ping Su and Emily Tse for their huge help!
To my sister, Stacey, and her husband Henry, along with my nieces Chloe and Penny whose
warm presence, cheerful demeanor and great help made this day possible.
To my wonderful friends David, Pui Yee, Gordon, and Howey, I thank you for looking out for
me, for your help, and being there for me during these 3 years.
To my dad’s great friends Alice and Kevin Cooper, Shiu Yeung (David) Yau, and Hon Ying
Wan for their kindness and consideration of my parents as I pursued my degree.
To Robert Field, Thomas Field, Michael Turan M.D., Charlene Dallas Turan, Brian Johnson,
PhD, Mr. Gibson, and R. Romano for treating me with the greatest kindness and giving me the
opportunity, knowledge, and skills to gain a head start in life.

vi
Acknowledgements
“If I have seen further, it is by standing upon the shoulders of giants.”
- Sir Isaac Newton, February 1675
Thank you to my dissertation chair, Dr. Bryant Adibe, whose generous time, honest
encouragement, and invaluable guidance made this dissertation possible. I am very grateful for
your patience with my endless probing and questions and instilling in me valuable leadership and
organizational skills in the courses you taught.
To my committee members, Dr. Raquel Torres-Retana and Dr. Tatiana Melguizo, my
deepest gratitude for the guidance, suggestions, and feedback you provided me. To Dr. Melanie
Brady, Dr. Marcus Pritchard, and Dr. Christopher Mattson, my great appreciation and gratitude
for your help. You were so responsive and gave me such quick turnaround times despite your
very busy schedule. I am grateful for your generous time and patience.
Dr. Bryant Adibe, Dr. Raquel Torres-Retana, Dr. Tatiana Melguizo, Dr. Melanie Brady
Dr. Marcus Pritchard, and Dr. Christopher Mattson your suggestions have made my research so
much richer, and my dissertation something I can be very proud of having written thanks to you.
To my dear colleagues who helped me through this long and difficult journey with
advice, help, guidance, inspiration, motivation, and simply being there for me: Alece Alderson,
Alex Siu, Alvin Yue, Angela Fang, David Tse, Doug Carlson, J.D., Elizabeth Silva, PhD,
Elizabeth Watkins, PhD, Farah Lare-Masters, Garland Woo, Kathy Chew, LaMisha Hill, PhD,
Lawrence Roberts, Lisa Dong, Lisa Raskulinec, Maria Jaochico, Naika McDonald, Orlando
Leon, Robert Turbyfill, Veronica Nepveu, Wendong Wang, and Wendy Winkler. Another thank
you to Doug Carlson, JD, Elizabeth Silva, PhD, and Wendy Winkler for their in giving me the
necessary letters of recommendation to begin this journey.

vii
Table of Contents
Abstract .......................................................................................................................................... iv
Dedication ....................................................................................................................................... v
Acknowledgements ........................................................................................................................ vi
List of Tables ................................................................................................................................. ix
List of Figures ................................................................................................................................ xi
Chapter One: Overview of the Study .............................................................................................. 1
Context and Background of the Problem ............................................................................ 2
Purpose of the Project and Research Questions .................................................................. 4
Importance of the Study ...................................................................................................... 5
Overview of Theoretical Framework and Methodology .................................................... 6
Definition of Terms............................................................................................................. 7
Organization of the Dissertation ......................................................................................... 9
Chapter Two: Literature Review .................................................................................................. 10
Introduction and Historical Context .................................................................................. 10
Student Persistence ........................................................................................................... 13
Knowledge Acquisition .................................................................................................... 28
Academic Institutional Barriers ........................................................................................ 39
Conceptual Framework ..................................................................................................... 53
Summary ........................................................................................................................... 61
Chapter Three: Methodology ........................................................................................................ 62
Research Questions ........................................................................................................... 62
Overview of Design .......................................................................................................... 63
Research Setting................................................................................................................ 65
The Researcher .................................................................................................................. 72

viii
Data Sources ..................................................................................................................... 74
Validity and Reliability ..................................................................................................... 80
Ethics................................................................................................................................. 82
Chapter Four: Findings ................................................................................................................. 86
Coding and Scaling ........................................................................................................... 88
Results for Knowledge, Motivation, and Organizational Influences ................................ 89
Summary ......................................................................................................................... 135
Chapter Five: Recommendations ................................................................................................ 140
Discussion of Findings .................................................................................................... 140
Recommendations for Practice ....................................................................................... 159
Limitations and Delimitations......................................................................................... 169
Recommendations for Future Research .......................................................................... 175
Conclusion ...................................................................................................................... 177
References ................................................................................................................................... 179
Appendix A: Assessment Tools to Evaluate Assumed KMO Influences ................................... 215
Appendix B: Qualtrics Online Survey Questions ....................................................................... 216
Appendix C: Rating Scale for Values Used in Closed Questions of Survey .............................. 235
Appendix D: Data Analysis of Qualtrics Responses .................................................................. 237
Appendix E: Data Frequency of Qualtrics Responses ................................................................ 239

ix
List of Tables
Table 1: Knowledge Influences on URM Students in STEM PhD Programs 54
Table 2: Motivational Influences on URM Students in STEM PhD Programs 56
Table 3: Organizational Influences on URM Students in STEM PhD
Programs
57
Table 4: Data Sources for Research Questions 65
Table 5: PhD Programs Offered 66
Table 6: Student Respondents' Gender 87
Table 7: Student Respondents' Ethnicity 88
Table 8: First or Continuing-Generation Student 90
Table 9: Guidance for Preparing and Applying to College, First-Generation
Students
91
Table 10: Guidance for Preparing and Applying to College, Continuing-
Generation Students
92
Table 11: First Generation Student Attending R1 Universities 93
Table 12: R1 University Students’ Responses for Question 20 96
Table 13: Non-R1 University Students’ Responses for Question 20 97
Table 14: Students Completed Master’s Degree Prior to Applying to PhD
Program
98
Table 15: Reasons for Having Mentor that Looks Like Me 100
Table 16: Mentorship Questions 101
Table 17: Students Heard About Bridge Programs 102
Table 18: Prosocial Goals 104
Table 19: Self-Efficacy Questions 107
Table 20: Self-Efficacy Questions, Percentage 108
Table 21: Self-Efficacy Questions Set 2 109

x
Table 22: Self-Efficacy Questions Set 2, Percentage 109
Table 23: Self-Efficacy Questions Set 2 111
Table 24: Coping and Resiliency Skills and Learning Strategies 112
Table 25: Coping and Resiliency Skills and Learning Strategies, Percentage 113
Table 26: Student Services 114
Table 27: Belonging 116
Table 28: Belonging and Prosocial Goals 117
Table 29: Scientist Identity 118
Table 30: Research 120
Table 31: Seeking Prosocial Goals and Social Justice 121
Table 32: More URM Students in STEM PhD Programs 127
Table 33: Negative Stereotypes 128
Table 34: Transition to PhD Program 129
Table 35: Lack of URM Role Models 131
Table 36: Alienated from Instructors 132
Table 37: Alienated from Instructors, Percentage 132
Table 38: Factors Very Important for Persisting and Succeeding in PhD
Programs
133
Table 39: Sufficient Opportunities Not Given to URM PhD Students 134
Table 40: Summary of Influences Results Data Evaluated 139
Appendix A: Assessment Tools to Evaluate Assumed KMO Influences 215
Appendix B: Qualtrics Online Survey Questions 216
Appendix D: Data Analysis of Qualtrics Responses 237
Appendix E: Data Frequency of Qualtrics Responses 239

xi
List of Figures
Figure 1: KMO Influences on URM STEM PhD Students 60

1

Chapter One: Overview of the Study
In recent years, American society has experienced protests, anger and outrage against the
violence and systemic racism towards people of color. While these events have captured most of
the nation’s attention, research and data has shown that there are “more subtle and pervasive
forms of discrimination and bias in education” (Hurtado & Alvarado, 2015, para. 1). On June 11,
2020, 4,500 STEM faculty and students, under the hashtag, #ShutDownSTEM”, pledged to forgo
research and meetings to instead focus on a day of action dedicated to dismantling discriminatory
systems in academia and STEM (Burke, 2020) that hinders more representation of
underrepresented minorities (URM) students and faculty.
Although progress has been made in recent years to increase URM enrollment at the
undergraduate level as exemplified by the solid 72.3% increase in Hispanic undergraduate
enrollments from 2004 to 2014 (NSF, 2017c), there is much to be done at the doctoral level
where less attention is received and thus, the focus of this dissertation. Although URM students
make up more than one-third of America’s college age (18-24) population (NSF, 2017b), they
received less than one-twelfth of all PhD degrees in the science and engineering fields that were
awarded (NSF, 2017a). Furthermore, NSF data of doctorate recipients from U.S. universities
showed that there were a dozen fields within STEM where not a single PhD was awarded to an
African American in 2017 (NSF, 2017a).
This can be attributed to a “leaky pipeline”. This terminology is used to define women
and URM students who “leak” from the STEM pipeline as they progress in their STEM field of
study from high school to bachelor’s to master and to doctoral level programs. In other words,
there is “the loss of diversity that occurs as one moves through tiers of educational and/or career
advancement” (Mattingly, 2019, para. 2). The leaky pipeline is attributed to a host of factors
2

including hostile campus climate and biases, lack of role models, demands of family and
financial responsibilities, and lack of academic preparedness being first-generation to attend
college (New Mexico State University, 2006). For an institution to increase the URM student
population in their STEM PhD programs, it is important that these factors be addressed. One
such institution that has made progress in addressing these factors is University of California,
San Francisco (UCSF).
Context and Background of the Problem
Employing more than 30,000 employees with an economic impact of $9 billion, the
UCSF is made up of three related, but separate entities – patient care, research, and education
(UCSF, 2018). UCSF consistently ranks as one of the nation’s top medical schools, top hospitals,
and top medical research centers (Harder, 2019). UCSF sees about 45,000 hospital admissions
and 1.7 million outpatient visits annually (UCSF, 2018) and is the top public recipient of funding
from the National Institutes of Health (NIH) (Alvarez, 2020). The focus of this dissertation is on
the education component of UCSF. As of fall 2019, UCSF had 3,212 students, 1,659 residents
and fellows, and 1,180 postdoctoral scholars (UCSF Office of Institutional Research, 2020).
Students at UCSF are enrolled in either one of the four professional schools (consisting of the
schools of dentistry, medicine, pharmacy, and nursing) or in the graduate programs where PhD
degrees are awarded. About a quarter of the student population, or 821 students, are enrolled in
the PhD programs (UCSF Office of Institutional Research, 2020).
Of the current students enrolled in the basic and biomedical sciences PhD programs at
UCSF, 22% are URM (UCSF Graduate Division, 2020a). This percentage is nearly three times
the national average of 7.9% (NSF, 2017a). The situation was very different more than 50 years
ago. The big push for better representation of ethnicity minorities for UCSF's staff, faculty and
3

students came in 1968, at the time of the Civil Rights Movement. The UCSF Black Caucus was
founded as a "direct result of the inequities and injustices that existed at UCSF in the late 1960s"
(UCSF Department of Anesthesia and Perioperative Care, 2020, para. 2) when the campus was
referred to as the "plantation on the hill" due to the treatment of ethnic minorities (UCSF
Department of Anesthesia and Perioperative Care, 2020). Since then, UCSF has made efforts to
promote diversity, equity, and inclusion. In 2010, the Office of Diversity and Outreach (ODO)
was created with the goal of building diversity in all aspects of UC San Francisco’s mission of
advancing healthcare worldwide (UCSF Office of Diversity and Outreach, 2020a) and has
developed and executed a comprehensive strategic plan for diversity and outreach that includes
recruiting and retaining URM students to UCSF (UCSF Office of Diversity and Outreach,
2020b).
UCSF is involved in community outreach programs to increase the pool of K-12 and
undergraduate students from diverse backgrounds including the Regional Collaboration to
Strengthen and Expand STEM (RECESS) (UCSF Center for Community Engagement, 2020), SF
BUILD (Building Infrastructure Leading to Diversity), and the Post Baccalaureate Program
which is a joint effort between the schools of Dentistry, Medicine, and Pharmacy to better
prepare URM students for the healthcare and science fields (Mattingly, 2019). UCSF has set up
the Advancing Faculty Diversity Grant to provide recruitment resources for faculty. Faculty
advisors have been trained in the best practices for strategically recruiting a diverse faculty pool
(Mattingly, 2019). At the student level, students across all schools and programs have formed a
group called the Interprofessional Diversity and Equity Alliance (IDEA) to improve support and
inclusion of students of marginalized and underrepresented backgrounds (UCSF
Interprofessional Diversity and Equity Alliance, 2020). UCSF also has an intervention program,
4

known as the Summer Research Training Program, that aims to strengthen the STEM academic
pipeline by seeking undergraduates with a desire to pursue a PhD offered by the UCSF Graduate
Division in life/health sciences to participate in a summer of research at UCSF (UCSF Graduate
Division, 2020b) where they are provided with skills and the guided mentoring needed to
succeed in the competitive STEM PhD fields.
Purpose of the Project and Research Questions
The purpose of this project is to conduct an evaluation study of UCSF in its ability to
enroll nearly three times the national average of URM students in the basic and biomedical
sciences PhD programs. URM students make up 22% of the current student body in these
programs (UCSF Graduate Division, 2020a) while the national average for STEM PhD programs
is 7.9% (NSF, 2017a). A complete evaluation will focus on all stakeholders, but for practical
purposes, this evaluation will focus on URM students who are enrolled in the basic and
biomedical sciences PhD programs at UCSF. The evaluation will look at the students’
perspective using Clark and Estes’ (2008) gap analysis framework to understand their
knowledge, motivation, and organizational influences. Two research questions will guide the
study:
1. What are URM students’ knowledge, motivation, and organizational influences to
achieve acceptance and persist in their STEM PhD programs at UCSF?
2. What are the recommendations, for knowledge, motivation, and organizational
resources, for UCSF to increase and retain URM students in their STEM PhD
programs?
5

Importance of the Study
The United States is currently the world’s most innovative and scientifically advanced
economy due to its strength in advanced STEM education (Han et al., 2015). At the same, the
U.S. population is undergoing a vast demographic shift in which non-Hispanic Whites are
projected to compromise less than 50% of the U.S. population by mid-century (Craig &
Richeson, 2017). In the midst of this, American society is experiencing a rise in protests strongly
correlated with grim racial economic inequalities (Economist, 2020) stemming from gaps in
earnings and employment caused by racial disparities in higher education enrollment and
attainment, particularly in the STEM fields (U.S. Department of Education, 2016). As the U.S.
population becomes more diverse (Frey, 2014), making STEM doctoral programs equally
accessible to all Americans must be part of our national dialogue if the nation is to remain
globally competitive and equitable. Addressing this imperative and equitable goal requires a
closer examination of the learning outcomes knowledge acquisition, persistence, and academic
institutional barriers playing in relation to URM students pursuing and persisting in science
doctoral programs.
Knowledge acquisition is imperative to students’ academic success when navigating
through the STEM pipeline (McGee, 2016). For URM students in particularly, the most
important knowledge acquisition relates to knowledge to matriculate and complete
undergraduates degrees (Roksa et al., 2018; Rudolph et al., 2019), knowledge needed to applied
and establish pathways to doctoral degrees (open up with a graduate degree might hinter
applicants from pursuing a graduate degree (Moreira et al., 2019; Oguntebi et al., 2012) and
knowledge to successfully navigate through their doctoral programs at research universities
(McKee, 2016; Roksa et al., 2018). Although students may initially lack the knowledge, the
6

knowledge can be acquired if students have access to resources to fill this gap (Ovink & Veazey,
2010).
Student persistence is the quality of allowing a student to continue to persist to degree
completion even when challenges arise (Tinto, 2017). Simply put, it is another way of defining
motivation (Tinto, 2017). Student persistence and student retention have been used
interchangeably and the difference is that “retention as an institutional measure and persistence
as a student measure” (Seidman et al., 2018, p. 88). Student persistence is measured by how well
an individual student continues striving toward degree completion (Myers-Briggs Company,
2020). Major factors that shape student motivation, and thereby student persistence is self-
efficacy, sense of belonging and perceived value of the education (Tinto, 2017).
Institutional barriers “are policies, procedures, or situations that systematically
disadvantage certain groups of people.” (Ashcraft, 2019, para. 1). Academic institutional
barriers, in the context of this dissertation, are barriers that exist in academic institutions that
prevent STEM doctoral programs from achieving diversity, equity and inclusion for URM
(Whittaker & Montgomery, 2012).
Overview of Theoretical Framework and Methodology
Clark and Estes’ (2008) gap analysis conceptual framework uses knowledge, motivation,
and organizational (KMO) factors to analyze performance gaps. Clark and Estes (2008) believed
that all three factors must be in place and aligned with each other before any successful goal can
be achieved. The knowledge and motivation systems are the most important systems that people
have in an organization; these two systems must cooperate effectively to work well in any
organizational environment (Clark & Estes, 2008). The performance gap that is being analyzed is
that URM students received only a quarter of the science and engineering PhD degrees relative
7

to their percentage makeup of America’s college age population (NSF, 2017b). This framework
will be the basis for doing a quantitative method study through the use of a Qualtrics© survey of
URM students who are enrolled in the basic and biomedical sciences PhD programs at UCSF.
The study will examine the students’ knowledge, motivation, and organizational (KMO)
influences that factor in UCSF’s ability to enroll nearly three times the national average of URM
students in the basic and biomedical sciences PhD programs.
Definition of Terms
First-generation students: Students who are the first in their family to attend college.
(Roksa et al., 2018)
Bridges to the Doctoral program: A program that allows students in the biomedical
sciences to obtain a master's degree on the way to the PhD program through partnership between
an institution that offers the master's degree and a research-intensive college or university
granting the PhD degrees in the biomedical sciences. The goal is developing a diverse pool of
PhD scientists in the biomedical sciences. (NIH, 2020)
Continuing-generation students: Students who have at least one parent who had some
postsecondary education experience. (Roksa et al., 2018)
Historically Black College or University (HBCU): According to The Higher Education
Act of 1965, HBCU is defined as: “…any historically black college or university that was
established prior to 1964, whose principal mission was, and is, the education of black Americans,
and that is accredited by a nationally recognized accrediting agency or association determined by
the Secretary [of Education] to be a reliable authority as to the quality of training offered or is,
according to such an agency or association, making reasonable progress toward accreditation.”
(U.S. Department of Education, n.d., , para. 1)
8

Minority-serving institution (MSI): Institutions of higher education that serve minority
populations. (U.S. Department of the Interior, n.d.)
National Institutes of Health (NIH): An agency that is part of the U.S. Department of
Health and Human Services and is the nation’s medical research agency responsible for making
important discoveries that improve health and save lives. (NIH, n.d.)
National Science Foundation (NSF): An independent federal agency created by Congress
in 1950 to promote the progress of science. NSF is vital because we support basic research and
people to create knowledge that transforms the future. (NSF, n.d.)
Predominantly white institutions (PWI): Although PWI is not an official designation for
any institution in the United States, it is used as a term in higher education scholarship for
defining the racial composition between predominantly white institutions and minority-serving
institutions. (Bourke, 2016)
R1 University: Used to indicate universities in the United States that engage in the
highest levels of research activity. (Carnegie Classification of Institutions of Higher Education
by Indiana University Center for Postsecondary Research, 2017; University of Chicago, 2008)
R2 University: Used to indicate universities in the United States that engage in a very
high level of research activity. (Carnegie Classification of Institutions of Higher Education by
Indiana University Center for Postsecondary Research, 2017; University of Chicago, 2008)
Science, Technology, Engineering, and Math (STEM): Used to classify disciplines in
science, technology, engineering, and mathematics. (Granovskiy, 2018)
STEM Pipeline: The educational pathway in the science, technology, engineering, and
mathematics disciplines. The STEM pipeline starts in elementary school and extends through the
professoriate and the rest of the workforce. (Insider Higher Ed, 2015)
9

Underrepresented Minority (URM): NSF defines underrepresented minorities as
Hispanic, Native American, and African American and uses STEM to classify disciplines in
science, technology, engineering, and mathematics (SACNAS Harvard Chapter, n.d.).
Organization of the Dissertation
The traditional dissertation of the five chapters is used for this evaluation study. The first
chapter provides the reader with the introduction and the background of the problem of practice,
the low percentage of underrepresented minorities (URM) in STEM PhD programs relative to
their college-age population. Chapter One continues with the context and background of this
problem; the organization being evaluated and the purpose of the study. The first chapter
concludes with the importance of the study and an overview of the theoretical framework
guiding this dissertation. Chapter Two focuses on the review of the current literature surrounding
the scope of the literature guide by the two research questions from Chapter One. The review of
current literature consists of a synthesis of the learning outcomes knowledge acquisition,
persistence, and academic institutional barriers play in relation to URM students pursuing and
persisting in science doctoral programs. Chapter Two concludes with an introduction and
explanation of the conceptual framework for the dissertation. Chapter Three is about the
knowledge, motivation, and organizational influences that will be to be examined in the study
including the research questions, methodical design, selection of participants, data collection,
and analysis. Chapter Four is a discussion of the data collected through an assessment and
analysis of the data. Chapter Five, the last chapter, discusses the findings with the review of the
literature and concludes with a set of recommendations.
10

Chapter Two: Literature Review
The purpose of this literature review is to provide the reader with a closer examination of
the learning outcomes, knowledge acquisition, persistence, and academic institutional barriers
play in relation to URM students pursuing and persisting in science doctoral programs. The
topics discussed in the literature review will be used to identify the key knowledge, motivational,
and organizational influences from the perspective of URM applicants and graduates in STEM
PhD programs. This chapter begins with an introduction and historical context and then proceeds
with the literature review of the key topics. The chapter concludes with a discussion of the Clark
and Estes’ (2008) gap analysis framework which is the conceptual framework used in this study.
The conceptual framework discussion will include summarizing the core concepts in the
literature view and present a graphic representation of the framework at the end of the chapter.
Introduction and Historical Context
Black/African American, Latinx/Hispanic and Indigenous/Native American, defined by
the National Science Foundation (NSF) as underrepresented minority or URM, make up 36.8%
of America’s college age (18-24) population (NSF, 2017b), but were awarded a scant 7.9% of all
PhD degrees in the science and engineering fields (NSF, 2017a). This implies that URM students
only received a quarter of the science and engineering PhD degrees relative to their percentage
makeup of America’s college age population.
In all of 2017, only one Native American was a PhD recipient in mathematics and
computer science in all of 2017 (NSF, 2018c) despite a Native American population of nearly
2.4 million (NSF, 2017b). Or that a dozen fields within STEM where not a single PhD was
awarded to an African American in 2017 (NSF, 2017d). This higher education inequity manifests
into social problems (Allen & Garg, 2016; Shadding et al., 2016), healthcare disparities (Pérez-
11

Stable, 2018; Sierra-Mercado & Lázaro-Muñoz, 2018), impacts the nation’s economy (London et
al., 2014; Rodriguez & Lehman, 2017), and innovations (Antón et al., 2018; Okahana et al.,
2018).
The event of October 4, 1957 would forever change U.S. education. When the Soviet
Union was able to put a satellite (Sputnik) into space, it put the spotlight on a national problem,
science education. About 1 year later, the National Defense Education Act (NDEA) was signed
into law providing increased funding to education institutions at all levels and with a focus on
scientific and technical education (27). This act contributed to the large rise in the number of
earned doctorates from 8,773 in 1958 to 55,195 in 2018 or nearly an 8x increase in 60 years
(NSF, 2019c) even though the nation’s population only doubled at the same time (United States
Regional Economic Analysis Project, 2020). Women were able to reap this growth as the
percentage of earned doctorates awarded to females went from 12% to 50% (American Physical
Society Physics, 2020). Asian students captured 25% of these doctorates earned (NSF, 2019d)
even though Asians represent only 5% of the nation's population (NSF, 2017f).
Amidst this phenomenal growth in the number of doctorates with notable gains for both
females and Asians, URM were left out. Even though URM make up more than one third of the
college age population, only one twelfth of all PhD science and engineering doctorates were
awarded to them. Geosciences is a vivid example that illustrates this trend of little to no
improvements in granting STEM doctorates to URM students over the ensuing decades. Between
1973 and 2016, the number of PhDs granted in the geosciences increased 60%, yet the
percentage awarded to URM students has remained stagnant (Bernard & Cooperdock, 2018).
Only 6% of all geosciences were awarded to URM students (Oguntebi et al., 2012) despite URM
students making up more than one third of the college age population (NSF, 2017b). At the same
12

time, in 1973, men had vastly outnumbered women as doctorate recipients in the geosciences,
but since 2009 women have exceeded men (Bernard & Cooperdock, 2018).
The problem of underrepresentation of URMs in the STEM doctorates has not been
ignored. Efforts have been made in higher education institutions and at the local, state, and
federal level to address this inequality for well over several decades (Maton et al., 2016; Moreira
et al., 2019; Salto et al., 2014; Syed & Chemers, 2011). The National Institutes of Health (NIH)
has initiated research training programs targeted at minorities since 1972 (Syed & Chemers,
2011) and in the decade to 2000 alone, more than $650 million was spent to train an annual
average of 15,000 trainees (Syed & Chemers, 2011). The National Science Foundation in
conjunction with various private and public universities have established and funded bridges to
doctoral programs, such as the NSF-AGEP program at Texas A&M, to increase STEM URM
undergraduates to pursue graduate degrees (Moreira et al., 2019).
Despite huge efforts and investments, the gains by URM students in the STEM PhD
program has been minimal as exemplified by the geosciences. According to the growing body of
literature, the low URM representation in the STEM PhD programs is complex and multifaceted
as there are several major contributing factors to the “leaky” STEM pipeline including
institutional barriers and biases, scarcity of URM faculty, challenges of first-generation students
pursuing graduate degrees, lack of helpful knowledge, misalignment of students values in STEM
versus institutions’, biases in the doctorate admission process, and a hostile campus climate with
stereotype threats and loneliness.

13

Student Persistence
Self-Efficacy and Motivation
Bandura (1986) asserted that people’s thoughts and beliefs affect how they behave. These
behaviors are acquired and maintained through the result of interactions between personal,
behavioral, and environmental elements (Bandura, 1986). "Beliefs in one’s capabilities to
organize and execute the courses of action required to produce given attainments" is defined as
self-efficacy by Bandura (Bandura, 1977, p. 3). Self-efficacy predicts behavior outcomes such as
when individuals don’t perform to their expectations, they either exert more effort, lower their
aspirations, or withdraw entirely (Judge & Bono, 2011).
People with a high level of self-efficacy are more engaged in a task, expend more effort
on the task, and persist longer in its completion even when encountering challenges or
difficulties (Chemers et al., 2001). Furthermore, those with high self-efficacy are not only more
likely to succeed in their tasks but are more likely to recover and bounce back from any failures
(Chemers et al., 2001). On the other hand, people with a low level of self-efficacy will become
discouraged and withdrawn from their task when encountering difficulties or not engaged with
the task through delay and avoidance (Klassen et al., 2008). Therefore, self-efficacy becomes the
foundation upon which student success is built. Students need to believe they can succeed in
their studies, otherwise there is little reason for them to continue.
In STEM, students with higher self-efficacy in mathematics and science are more likely
to continue studying those subjects (Andersen & Ward, 2014). Eighth-grade students' self-
efficacy in mathematics and academic proficiency predicted who would persist to a STEM
degree and career (Mau, 2003). At the college level, a student’s persistence in a STEM program
is predicted by their expectation for success (Andersen & Ward, 2014). For graduate
14

post/doctoral students, their science, leadership, and teamwork self-efficacy each independently
predicted the students' persistence to a science career (Chemers et al., 2011). According to
the self-efficacy theory, it is these beliefs in one’s personal capabilities that become the primary,
explicit explanations for motivation (Lumen Learning, n.d.).
Numerous studies have shown the relationship between motivation and educational
outcomes (Katz et al., 2014). When dealing with academic work, motivation can serve as a
protective buffer for students to overcome negative consequences and various difficulties (Katz
et al., 2014). The three biggest motivations for students to pursue a PhD degree are: career
development, interest in a topic or research, and personal motives (London et al., 2014). As
motivated as a URM student may be at the start of the doctoral program, the attrition rates for
URM students in the science and engineering PhD programs can be more than 50% for certain
fields of study (Council of Graduate Schools, 2020) and two-thirds amongst Black students
(Joseph, 2012). There is a strong correlation between the high attrition rates and the lack of
motivation to complete the doctoral program (Rudolph et al., 2019). The four motivations theme
of the experiences of URM STEM graduate students are: mentoring, research, opportunities
(through funding and support), and the climate of their academic environment (Guy & Boards,
2019).
Prosocial Values and Identities
One type of motivation is the desire “to relate to and care for others” (Deci & Ryan,
1991, p. 243). For aspiring URM students in the STEM fields, there is a strong link between their
career and helping their communities (Guy & Boards, 2019), such as linking research to making
positive community changes (Guy & Boards, 2019). URM students place great value in helping
others and giving back to the community through one’s work (Jackson et al., 2016). This is due
15

in part that STEM URM students come from cultures with an emphasis on prosocial values
(Jackson et al., 2016) and seek jobs or educational opportunities with the altruistic goals of
helping others and giving back to the community (Jackson et al., 2016; Thoman et al., 2014).
A study of 6,000 STEM undergraduate students showed that working for social change is
more important for STEM URM students than their non-URM counterparts (McGee et al., 2016).
This is carried over to the doctoral level, where surveys and interviews of graduating doctoral
students highlight the importance of prosocial values for persisting in science education (Jackson
et al., 2016). URM PhD holders were significantly more likely to place the application of
research to address health problems specific to their communities (Jackson et al., 2016). In fact,
URM students were more involved in lab activities and put greater interest in their science career
when those opportunities fulfill prosocial goals of helping others and giving back to the
community (Thoman et al., 2015).
URM students often encounter in PWIs the lack of prosocial values from majority non-
URM peers (Joseph, 2012) that are firmly rooted in URM students’ cultural and ethnic
background (Thoman et al., 2015). When URM students experience disconnects between their
prosocial values and their science education and between relevance of science to social issues,
they will find a contradiction between their cultural and science identities (the importance of
science identities to URM students will be discussed shortly) (Jackson et al., 2016). For URM
students, there is great importance of finding prosocial meanings in their science education and
future science careers (Jackson et al., 2016).
Identifying with a science or scientist identity is another very strong motivator for URM
students. Students, in general, have better academic performance and persistence when
identifying with a role identity such as a scientist, than racial or ethnic identity (Chemers et al.,
16

2011). Science identity building is one of the main characteristics attributed to high-achieving
students whose goals are to pursue graduate education and a career in scientific research
(Hurtado et al., 2011). URM students have greater persistence to complete their degree when
identifying with a role identity (Eccles & Barber, 1999). A strong science identity could serve as
a buffer for URM graduates who face devaluation through racial biases and negative stereotypes
as well as a motivator for success (Kim‐Prieto et al., 2013). It is believed that URM graduate
students working with peers and scientists in research and attending conferences reinforce
experiences and recognition to support their identity as future scientists (Kim‐Prieto et al., 2013).
Female STEM URM doctoral students mention passion and research as the main driver
for their persistence in STEM and in doing so academics became a core part of their identity
(Guy & Boards, 2019). Research opportunities are closely associated with a science identity and
are shown to have positive effects on retaining students in STEM (Shadding et al., 2016). Interest
in the research is one of the major factors that motivate people to seek a PhD across almost all
disciplines (London et al., 2014). Research is an effective strategy to increase a student's self-
efficacy in science as experiences in search builds a better understanding of the sciences through
practice and active learning and sets an environment for students to propose and test out new
ideas (Burton et al., 2019).
URM seeking a science identity to increase persistence in completing their degree
(Chemers et al., 2011; Shadding et al., 2016) may find it difficult to integrate (Jackson et al.,
2016) with the prosocial values that are firmly rooted in their cultural and ethnic background
(Thoman et al., 2015). URM students are only a small minority in most STEM PhD programs
and their prosocial values are likely to conflict and out of the norm of the career and research
focused interests of the majority of their peers (Jackson et al., 2016). URM students are seeking
17

in their education whether in the classroom, conferences, or research the prosocial goals and
opportunities (such as helping their communities) that is heavily valued in their culture (Jackson
et al., 2016; Thoman et al., 2015).
When students can’t see this connection between their prosocial orientation and their
science education or career, they experience a disconnection between their cultural and science
identities (Jackson et al., 2016) that can lead to loss of interest and disengaging from science
(Greenhalgh et al., 2004). Working for social change was most important for URM students who
left the discipline (Tran et al., 2011). This lack of inclusiveness and relevance in STEM for URM
graduate students leads to a cultural conflict between the relevance of their science career and
education with the social issues they are concerned about (Tran et al., 2011). On the other hand,
when URM students are encouraged to realize their academic potential by embracing their ethnic
and gender identity, this led to more than one-fifth increase in obtaining STEM graduate degrees
(Jackson & Winfield, 2014).
Sense of Belonging
Students’ sense of belonging increases persistence toward degree completion (Hausmann
et al., 2017). The sense of belonging is the students’ psychological sense of seeing themselves as
a member of a community of other students (Hurtado, 1997). This sense of connection to the
student community is important as people typically have strong desires to belong to
communities. The lack of connection can have a harmful effect on students’ mental health and
behavior (Museus et al., 2018).
Belonging, whether to an academic, professional, or social environment is a fundamental
human need (Rattan et al., 2018). The lack of URM faculty role models along with lacking peers
with shared cultural and ethnic identities means URM students do not feel they belong to their
18

academic environment (Jackson et al., 2016). Another factor that reduces URM students’ sense
of belonging to the STEM discipline are the widespread negative stereotypes that URM don’t
have math or science abilities (Rattan et al., 2018). Belonging is an important factor for URM
students to pursue (Rattan et al., 2018) and persist (Maton et al., 2016) in STEM.
The lack of belonging leads to loneliness and isolation that can demotivate students in
persisting in their doctoral studies (Mantai, 2019). Many PhD students already suffer from
loneliness (Brown, 2013), being a URM PhD student makes it even lonelier and more
problematic (Harmon, 2019). The Advancement of Chicanos/Hispanics and Native Americans in
Science (SACNAS) chapter at Harvard University mentions that a typical STEM PhD program
in the U.S only awards a scant five PhDs to URM individuals every year (SACNAS Harvard
Chapter, n.d.). NSF data of doctorate recipients from U.S. universities showed that there were a
dozen fields within STEM where not a single PhD was awarded to an African American in 2017
(NSF, 2018c). These are very telling signs of how isolated and lonely it can be for the few URM
students that are enrolled in a STEM PhD program despite all odds of getting admitted into the
program (Okahana & Zhou, 2017).
Edray Goins has a PhD in mathematics. His lonely and isolated postsecondary
educational experience covered recently in a New York Times article titled, “For a Black
Mathematician, What It’s Like to be the ‘Only One’” (Harmon, 2019) is typical of URM
students in a STEM PhD program. Goins was one of only three African Americans to have ever
obtained a PhD from Stanford University’s math department (Harmon, 2019). Prior to that,
Goins was only one of 10 African American students in his incoming class at Caltech, and only
one of 3 that actually graduated 4 years later (Harmon, 2019). Goins completed his PhD
dissertation in 1999 (University of Buffalo, 2008). The dissertation was on elliptic curves
19

(University of Buffalo, 2008), an area of mathematics covering geometry and algebra where in
2017 only one African American received a PhD degree in (NSF, 2018c). In his blog, Goins
lamented his feeling of isolation “I am an African American male, I have been the only one in
most of the universities I’ve been to-the only student or faculty in the mathematics department.
To say that I feel isolated is an understatement” (Harmon, 2019, para. 19).
The doctoral education experience of Dr. Goins is similarly echoed by other URM
doctoral students. URM doctoral students report academic isolation and insufficient faculty
mentoring (Lundy-Wagner et al., 2013) along with findings to support receiving fewer teaching
and research assistantships (Salters, 1997) that further isolates them from their academic
environments. A 6-year study of African American males pursuing graduate degrees in the
engineering fields at research universities showed common themes of isolation, unwelcoming
environments, and lack of camaraderie (Burt et al., 2018). The transition into PhD programs by
Latino students has been described as “lonely,” “isolating,” and “alienating” (Arocho, 2017). A
2015 report from the Council of Graduate Schools identifies isolation as a particular problem for
minority students (Gardner, 2015). Pamela McCauley, the first African American female to earn
an engineering PhD in the state of Oklahoma and current engineer professor, mentions that the
bias, isolation, and marginalization experienced by female URM students while pursuing their
undergraduate and master’s degrees in the engineering fields heavily discourages further study
towards a PhD (Grimsley-Vaz, 2018). Given the loneliness, isolation, and alienation that URM
PhD students encounter, it is not surprising that few students wish to pursue the PhD program.
This translates to fewer doctoral degrees awarded to URM students.
Students’ sense of belonging “is most directly shaped by the broader campus climate and
students’ daily interactions with other students, academics, professional staff and administrators”
20

(Tinto, 2017, p. 4). A result arising from having a sense of belonging is a commitment that
serves to bind the student to the group or community even in the face of challenges; it leads to
increased motivation and involvement in the community that further promotes student
persistence (Tinto, 2017). Tinto’s Integration Model (of student retention) is based on the need
for a match between the institutional environment and the student’s individual commitment
(Seidman et al., 2018). A good match will lead to higher levels of student integration into the
academic and social aspects of college life resulting in increased persistence (Seidman et al.,
2018). On the other hand, a hostile campus environment is negatively associated with URM
students’ sense of belonging in college and a predictor of intent to persist to degree completion
(Museus et al., 2018).
Historically Black colleges and universities (HBCU) are examples of institutional
environments where URM students have a sense of belonging that allows them to thrive and
persist in their studies (Winkle-Wagner & McCoy, 2018). HBCUs are the top ten baccalaureate
institutions of awarding the doctoral degree to African Americans (NSF, 2019b). This success
can be attributed to students’ strong sense of belonging (Joseph, 2012). From the first-time
students enter the campus, HBCUs provide a developmentally powerful environment based on a
very supportive culture of both peers and faculty (Perna et al., 2009). There is a community of
faculty and staff that focuses on the student’s well-being and advancement (Joseph, 2012). This
environment makes students feel a deep sense of belonging, like at home with family members
(Fries-Britt & Turner, 2002). Belonging is an important factor for URM students to pursue
(Rattan et al., 2018) and persist (Maton et al., 2016) in STEM and the lack of it can demotivate
students to leave their doctoral studies without completing it (Williams et al., 2017).
21

The diversity and availability of faculty at HBCUs (Joseph, 2012) provides mentoring
opportunities that further increase minority students' participation and persistence in STEM
fields (Sánchez et al., 2018) and are keys to their academic success (Lisberg & Woods, 2018).
The availability of URM faculty role models allows students to successfully create a scientist
identity with role models like themselves that have ‘made it’ (McGee, 2016). A scientist identity
can be a very strong motivator for URM students. Students, in general, have better academic
performance and persistence when identifying with a role identity such as a scientist, than racial
or ethnic identity (Chemers et al., 2011). Science identity building is one of the main
characteristics attributed to high-achieving students pursuing graduate education and a career in
scientific research (Hurtado et al., 2011).
Having URM faculty role models along with peers with shared cultural and ethnic
identities mean URM students feel a sense of belonging to their academic environment (Jackson
et al., 2016). Belonging correlates to the acquisition of knowledge that is imperative to students’
academic success (Joseph, 2012). Finally, URM faculty mentor is strongly preferred by URM
students (Whitfield & Edwards, 2011) and can draw from their mentor’s own racial encounters
and experiences to counsel URM students who are questioned about their abilities due to racial
stereotyping or confronted with “fears of fulfilling stereotypes or addressing feelings about being
representatives of their race” (Mondisa, 2018, p. 296).
The deep sense of belonging in a HBCU (Fries-Britt & Turner, 2002) is shattered when
African American students go to PWI to continue with their graduate studies. At HBCUs,
students are accustomed to a supportive and nurturing environment where the student-faculty
ratio is lower and there are frequent interactions with faculty (Kim & Conrad, 2006). The
graduate academic culture in PWIs doesn’t have much personal involvement, is filled with
22

ambiguity and lack of advice and guidance (Jackson et al., 2016). This sets the expectation that
the students need to work independently or in isolation, and without close one-on-one teacher
relationships African American students have gotten accustomed to. Furthermore, doctoral
engineering program culture weans out weaker students while testing the mettle of students
(Morrell, 1996). The lack of African Americans or URMs in the doctoral programs (SACNAS
Harvard Chapter, n.d.) only re-enforces faculty members' low expectations and negative
stereotypical perceptions of URM students (Jackson et al., 2016). The effects of all this are a
cultural shock for African Americans students entering graduate studies outside of HBCUs.
Instead of feeling inspired and confident, attending graduate school made African American
students feel boxed in and their identities challenged (Jackson et al., 2016).
The experiences of African American graduate students attending PWI were similarly
experienced by other students of color (Museus et al., 2018). Students of color perception of a
hostile environment are negatively associated with the students’ sense of belonging (Nuñez,
2009). A study showed that even though the rate of harassment on campuses are the same for
white students as for students of color, the latter group perceived their campus environment to be
more racist and less accepting than their white peers (Rankin & Reason, 2005). In PWI, white
students generally reported higher levels of belonging than students of color (Johnson et al.,
2007). On the other hand, positive environmental factors, such as supportive residence halls,
positive cross-racial relationships and perceived faculty interest in students increases students of
color sense of belonging (Museus et al., 2018).
Building Persistence Through Mentorship and Research Opportunities
Research experience is integral to developing and sustaining interests in STEM (Estrada
et al., 2018). Research experience directly contributes to degree completion and persistence
23

amongst STEM students (Estrada et al., 2018). Studies have shown that research experience
leads to enhanced abilities to think critically (Cooper et al., 2019). Increased frequency and
longer research experiences help students develop the thought processes, skills, and relationships
that are critical to student academic performance needed to persist through STEM graduate
degree completion (Cooper et al., 2019). For URM students, participation in research
opportunities led to greater gains in thinking and working like a scientist along with confidence
in their research skills and ability to succeed in graduate school (Cooper et al., 2019).
When research opportunities are tied to engaging with their communities, such as
focusing on health disparities for people of color, it leads to increased performance and retention
in URM students (Malotky et al., 2020). These opportunities remove the common misconception
amongst URM students that science is an isolated discipline that has no relation or interest
outside the scientific community (Malotky et al., 2020). For URM students, there is great
importance of finding prosocial meanings in their science education and future science careers
(Jackson et al., 2016). The research opportunities allow URM students to form connections
between their prosocial values and their science education and between the relevance of science
to social issues. Surveys and interviews of graduating doctoral students highlight the importance
of prosocial values for persisting in science education (Jackson et al., 2016). There is a lack of
diversity in STEM doctoral programs for both faculty (Yadav & Seals, 2019) and students
(Harmon, 2019; SACNAS Harvard Chapter, n.d.). This leads to loneliness and isolation that can
demotivate URM students in persisting in their doctoral studies (Mantai, 2019). The research
opportunities to engage with diverse communities can help alleviate this barrier to URM student
retention in STEM (Malotky et al., 2020).
24

Quality student mentorship contributes to increased student efficacy, academic
achievement, productivity in scholarship and persistence (Estrada et al., 2018). Even informal
mentoring networks are essential for nurturing minority students (Carter-Sowell, 2016). Eby et
al. (2013) identified three mentorship factors in providing students with positive outcomes. The
first factor, known as instrumental support, is providing opportunities and resources needed for
students to reach their goals (Estrada, 2018). The second factor is psychological support which is
providing emotional and personal development to students to enhance their competence, identity,
and effectiveness (Estrada, 2018). The final factor is relationship quality which is building
feelings of trust, empathy, respect, and connection between mentor and student (Estrada, 2018).
Combined, these factors contribute positively to students’ performance, motivation, career
outcomes, and health (Estrada, 2018). Poor mentor relationship is a major factor in high attrition
rates for STEM PhD students (Moreira et al., 2019).
Preliminary assessments from various research institutions show that research
opportunities and mentorship are critical components for community building between students
and their institutions, programs, and faculty (Moreira et al., 2019). Community building leads to
a sense of belonging and contributes to student engagement for URM graduate students (Moreira
et al., 2019). Community building alleviates URM negative perceptions of campus racial
climates that has shown to link to diminished sense of belonging to campus and lack of
persistence (Maton et al., 2016). Community building is essential to the success of STEM URM
graduate students to complete their graduate degree (Moreira et al., 2019).
Faculty and Mentors Looking Like Me
One of the main factors for URM students’ persistence and academic success is having
access to URM faculty (Joseph, 2012; Whittaker & Montgomery, 2012). A study showed that
25

students, particularly males and non-whites, perform better when their teachers share the same
race and gender (Miller, 2018). There is a strong desire for URM students to prefer mentors that
are like them (Whitfield & Edwards, 2011) as it is difficult for students to create a career or
scientist identity when there is a lack of role models like themselves that have ‘made it’ (McGee,
2016). When URM students see a successful faculty with a shared ethnic and cultural
background, it signals to the student that they too can be successful (Yadav & Seals, 2019).
URM students are more likely to persist in completing their degree program when they have
access to a URM faculty role model (Yadav & Seals, 2019). Racial, cultural, and ethnic
differences can also lead to difficulties and challenges with communication (McGee, 2016).
HBCUs are far more successful in producing URM doctoral recipients than other higher
educational institutions (Whittaker & Montgomery, 2012) and the top ten institutions awarding
doctoral degrees to African Americans are all HBCUs (Joseph, 2012). This is due in large part to
the diversity of the faculty in the HBCUs (Joseph, 2012). URM students find it difficult to have a
positive identity when there is a lack of URM faculty role models to identify with (Brittian et al.,
2009) in a PWI (Joseph, 2012) or in a campus climate that is hostile to them (Sánchez et al.,
2018). In a hostile campus climate, URM faculty mentors may need to draw from their own
racial encounters and experiences to counsel URM students who are questioned about their
abilities due to racial stereotyping or confronted with “fears of fulfilling stereotypes or
addressing feelings about being representatives of their race” (Sánchez et al., 2018, p. 294).
As important as URM faculty to the mentorship of student protégés are, the faculty in
higher education doesn’t mirror the racial and ethnic makeup of the student body they serve
(Yadav & Seals, 2019). According to the National Research Council (2011), the lack of diversity
of faculty members is a barrier for URM students to complete degrees in the STEM fields. Given
26

the low number of doctoral degrees awarded to URM, the percentage of URM serving as full-
time faculty in the academic science and engineering workforce from the NSF is not surprising.
Despite URM making up 36.8% of America’s college age (18-24) population (NSF, 2017b), only
8.6% of all full-time faculty members were URM (NSF, 2018b).
The increase in PhD awarded to URM students in recent years don’t translate to increased
URM faculty members (NSF, 2019e; NSF, 2019f). In the field of biomedical sciences, from
1980 to 2013, there has been a 930% increase in PhDs awarded to URM students, yet it hasn’t
translated to any significant increase in the hiring of URM assistant professors (Johnson, 2019).
In fact, the biomedical sciences are a small microcosm of what is happening in most STEM
fields in academia. Despite almost 6,000 biomedical sciences PhD awarded to URMs between
2005 and 2013, there were fewer URM assistant professors in 2014 than in 2005, when during
the same period there was an 8.6% growth of professors overall (Johnson, 2019). At a very
selective and highly ranked college, African American and Hispanic faculty were 33 times and
13 times more likely to be denied tenure than their white colleagues, respectively (Chen, 2018).
If the positive effects of a teacher being a role model is distinct from a teacher being
effective (Gershenson et al., 2018), the scarcity of URM serving as full-time faculty members
puts URM STEM PhD students in a further disadvantage. With lesser URM faculty, URM
students in PWI are less likely than their non-URM peers to have the personal interactions with
faculty (Anneke et al., 2018) which is so crucial to student academic success (Joseph, 2012). A
study of female doctoral students showed that minority students had lesser mentoring and
apprentice experiences and fewer professional social opportunities than their non-minority peers
(Joseph, 2012). Finally, it has been found that faculty members create additional barriers to
URM in the form of sending mixed messages to URM students about their worth and belonging
27

in science and responding less often and more slowly to prospective URM applicants (Rudolph
et al., 2019).
The scarcity of URM faculty can have a great effect on student’s motivation. URM
students find it difficult to have a positive identity when there is a lack of URM faculty role
models to identify with (Brittian et al., 2009) in a predominately white institution (PWI) (Joseph,
2012) or in a campus climate that is hostile to them (Sánchez et al., 2018). URM STEM PhD
students already face significant psychological barriers with the mixed messages faculty
members send to them (Pyne & Grodsky, 2019), therefore the absence of URM faculty role
models make the situation worse. The lack of URM role models weighs heavily with URM
students pursuing and persisting in a STEM disciple (Sánchez et al., 2018). URM students feel
there isn’t a high utility value for taking science and mathematics coursework where there is a
lack of successful URM faculty or scientists (Andersen & Ward, 2014). A high utility value in
STEM means a URM student is more likely to pursue STEM (Lewis, 2005) and to persist in the
program (Andersen & Ward, 2014).
In the doctoral biomedical sciences, URM faculty mentors are essential for URM
students to persist in the programs due to the lack of role models and mentors in the biomedical
sciences (Wilson et al., 2018; Yadav & Seals, 2019). When URM students see faculty with a
shared identity in successful positions, it signals they too can be successful (Yadav & Seals,
2019). At the same time, the lack of URM faculty role models motivates URM PhD students to
persist in their program so that they can eventually become faculty members themselves and
combat the lack of role models that subsequent URM students face (Jackson et al., 2016).
28

Knowledge Acquisition
First-Generation Students Knowledge Deficiencies
Knowledge is imperative to students’ academic success when navigating through the
STEM pipeline. First-generation students might not have as much of the knowledge that
continuing-generation students have to get into and succeed in college (Roksa et al., 2018). This
problem is more apparent for URM students, as nearly half of Black and Latinx college students
are first-generation students (Rudolph et al., 2019). First-generation students rely more on
secondary education teachers and counselors to encourage them to pursue their field of STEM
study (Kong et al., 2013). The lack of parental knowledge and encouragement to provide useful
advice for career and educational planning for aspiring first-generation college students when
they are in high school may negatively impact their college education (Witkow & Fuligni, 2011).
Without a parent that has gone through the college admissions process, high school students
might not be aware or knowledgeable about the college eligibility requirements (Rudolph et al.,
2019). For example, if a high student starts taking required advanced math courses late in their
high school years, the student might not be able to meet all the required math courses for college
eligibility (such as for the University of California (UC) or California State University (CSU)
systems) or lack the essential math skills needed for successful careers in science and
engineering (Rudolph et al., 2019).
Once in college, first-generation STEM students might have lower retention rates due to
the unique requirements to succeed in STEM programs – mathematics and research skills and
seeing the relevancy of science to their careers (Andersen & Ward, 2014; Kezar & Holcombe,
2020). Only about 40% of students intending to major in a STEM field actually complete the
degree within 6 years (Kezar & Holcombe, 2020). Non-URM students enrolled in the University
29

of California system show that higher persistence rates in STEM programs were due in part to
better the academic preparations than their URM peers (Arcidiacono et al., 2016). To address the
deficit in STEM preparatory coursework for URM students that leads to high attrition, the
integration of culture into science should be considered. URM interest in science increases when
science courses incorporate culturally responsive teaching principles (Andersen & Ward, 2014).
Otherwise, URM students feel that pursuing science careers means losing their racial
identity (Andersen & Ward, 2014). The scarcity of URM faculty role models (Mendoza-Denton
et al., 2017) further strengthens this belief. Employing active learning methods, such as concept
maps, increases student grades and retention in STEM courses and results in more effective and
better learning outcomes (Lisberg & Woods, 2018).
Mentoring Students
Mentorship allows students to gain familiarity and knowledge of on-campus student
support programs (student organizations, faculty office hours, undergraduate research, and
academic support) that are known to increase student success (Lisberg & Woods, 2018). Access
and use of these services increase students’ sense of belonging on the campus. Belonging makes
students not feel alienated from faculty and peers and comfortable enough to ask questions,
attend seminars, or request help to acquire the knowledge to succeed in the programs (Ellis,
2001). Students’ knowledge gained from mentoring is mainly limited to a research relationship
(McGee, 2016). When applying for graduate programs, the lack of awareness of the career
opportunities that open up with a graduate degree might hinder applicants from pursuing a
graduate degree (Oguntebi et al., 2012). For competitive doctoral programs, such as MD-PhD
programs, limited or no information on the board range of test scores and GPA of successful
30

applications further deters applicants from applying to the programs (Christophers & Gotian,
2019).
Coaching has emerged as a tool to complement mentoring and address the knowledge
gaps with just mentorship alone (McGee, 2016). “Coaches can also bring unique expertise to
scientific development beyond what many mentors can, such as in teaching grant writing,
networking, and communications skills, and deciding among career options” (McGee, 2016, p.
235). Mentoring and coaching are needed to address another unique problem faced by first-
generation students. First-generation students enrolled in PhD programs are less likely than their
continuing-generation students to have obtained their undergraduate degree from a high research
doctoral-granting university (classified as R2 on the Carnegie Classification) or did graduate
studies from a very high research doctoral-granting university (classified as R1 university)
(Carnegie Classification of Institutions of Higher Education by Indiana University Center for
Postsecondary Research, 2017; University of Chicago, 2008). This implies first-generation
students feel a disconnection between the institution where they are completing their doctoral
program and the institution where they obtained their undergraduate degree. The disconnection
results in these students having a more difficult time navigating through their doctoral programs
as they are unfamiliar with social and cultural norms or “unspoken” rules (Roksa et al., 2018) in
high research-intensive universities where most of the STEM PhD degrees are granted (Buswell,
2017; Roksa et al., 2018).
Mentoring is a key source of knowledge acquisition for student academic success. When
students enroll in college, they are often faced with social pressures and issues transitioning to
college which mentoring addresses by promoting coping skills and resiliency (Lisberg & Woods,
2018). College students’ GPA and retention rate are positively correlated with the frequency of
31

mentor-protégé contact (Campbell & Campbell, 1997). At the heart of academic success is the
interpersonal relationships between student and faculty (Joseph, 2012). The interactions between
the two groups and the support faculty provided on both personal and academic levels led to
increased self-concept traits for students in the areas of psychosocial wellness (emotional and
physical health, social and intellectual self-confidence, understanding of others, and
cooperativeness), achievement orientation (drive to achieve, leadership ability and
competitiveness), and academic ability (artistic, writing, public speaking and academic abilities)
(Berger & Milem, 2000). Without the mentorship and interaction with faculty, students would
lag in the critical areas needed for student success—academic preparedness and strategies for
success (Joseph, 2012).
For STEM students, mentoring was also found to positively correlate to self-development
activities that led to satisfaction, effective communication and involvement in their programs of
study (Holland et al., 2012). For first-generation students, of whom half of URM students are
(Rudolph et al., 2019), mentorship was identified as a key to their academic success (Lisberg &
Woods, 2018). Furthermore, for URM students in general, mentoring is especially important as it
builds professional networks and enhances research skills that are needed in closing the gap that
exists between URM and non-URM students (Lisberg & Woods, 2018). URM students found
mentorship to be particularly effective when it is multi-dimensional and comprehensive to
include mentoring with not just faculty, but advisors, researchers, peers, and staff that
encompassed campus support services, research experience, learning strategies and study skills
(Wilson et al., 2011). Mentoring increases minority students’ participation and persistence in
STEM fields (Sánchez et al., 2018) while minority students who aren’t in any mentoring
networks are put at a distinct disadvantage against their peers (Moreira et al., 2019). Finally,
32

mentoring helps minority students cope and combat negative college experiences – chilly
campus climate, racial stereotypes, and negative racial experiences and (Sánchez et al., 2018).
Not addressing these negative experiences can lead to students feeling isolated and lonely and
negatively impact academic success (Arocho, 2017; Brown, 2013; Burt, Williams & Smith,
2018; Harmon, 2019).
The transition to graduate school can be difficult for students especially for African
Americans who felt a cultural shock when they attend graduate school that was different from
where the majority obtained their undergraduate degree in minority-serving institutions (MSI) or
historically Black colleges and universities (HBCU) (Brown & Davis, 2001; Joseph, 2012; NSF,
2019a). Similarly, at the graduate level as in the undergraduate level, doctoral students face
biases, stereotype threats and lack of belonging (Yadav & Seals, 2019). URM students find on
entering graduate school that they need to successfully navigate through new cultures,
expectations and environments and learn to build networks to avoid isolation and learn strategies
for being successful in graduate school (Moreira et al., 2019). Thus, the mentorship at the
graduate level is even more critical than at the undergraduate level and involves a personal
connection with mentors beyond just academic and career counseling (Moreira et al., 2019) as
“graduate education is based on mastering the techniques of the discipline with a focus on the
academic norms and behaviors” (Joseph, 2012, p. 135).
Master’s Degree: Acquiring the Necessary Skills
Having a prior master’s degree is an important path for URM students pursuing STEM
doctoral programs (Okahana et al., 2018). A master’s degree can help URM students acquire the
knowledge and preparation needed to be successful in STEM doctoral programs (Okahana et al.,
2018) such as access to courses and knowledge unavailable to them as undergraduates or gain
33

research experiences (Rudolph et al., 2019). There is a positive correlation between having a
master’s degree and earning a doctorate within 10 years for African American students and male
URM students in the STEM fields. URM students are 50% more likely to obtain a master’s
degree on their way to a PhD than non-URM peers (Stassun et al., 2011). Also, applicants who
are denied to PhD programs might pursue a master’s degree before reapplying to the PhD
programs (Rudolph et al., 2019).
In the past two decades, partnerships have been established and grown between STEM
master’s programs offered at regional universities with STEM doctoral programs at research
universities to leverage this positive correlation (Okahana et al., 2018; Rudolph et al., 2019;
Stassun et al., 2011). These partnerships are commonly referred to as Masters-to-PhD Bridge
programs (Stassun et al., 2011) such as the joint program between Vanderbilt University and
Fisk University (Whittaker & Montgomery, 2012). The retention rate for URM doctoral students
in Masters-to-PhD Bridge programs is as high as 92% (Rudolph et al., 2019) versus a national
average of 50% (Joseph, 2012). Part of this is attributed to offering students opportunities to
attend workshops for professional development and exposure to a variety of research
opportunities to build up research experience and skills (Rudolph et al., 2019).
Traditional admission criteria to the STEM doctoral program – GRE, undergraduate
GPA, the ranking of undergraduate institutions attended, and research experience weighs
negatively against URM applicants. On average, URM applicants have lower GRE scores
(Wilson et al., 2019) and undergraduate GPA (Miller et al., 2019) as compared to their white and
Asian counterparts. GRE scores are not good predictors of student productivity (first-author
student publications) or degree completion in the STEM PhD programs (Hall et al., 2017; Miller
et al., 2019) based on a growing body of research in recent years.
34

A study of STEM PhD programs found that male students who completed the STEM
doctoral degrees had significantly lower GRE scores than those students who left the program
(Petersen et al., 2018) or that there was less than 10% completion rate difference for Physics PhD
applicants who scored in the 10th percentile versus those scoring in the 90th percentile (Miller et
al., 2019). Most URM students earn their undergraduate degrees in public universities (NSF,
2017d) while white students make up 87% of all undergraduates at private universities (Brown,
2016). The grades given out at public universities are usually about 1/3 letter grade lower than
private universities (Miller et al., 2019). Furthermore, many URM students enrolling in
engineering programs face difficulties during their first 2 years, but eventually, do well
academically in the last 2 years if they persist with the program (Reichert, 2013). Their first 2
years of difficulties result in these students having a lower undergraduate GPA and potentially
eliminating them as strong candidates to the STEM PhD programs.
Faculty members at elite institutions tend to favor PhD applicants who come from similar
institutions (Posselt, 2018) and institutions classified as “very high research activity” (R1 in the
Carnegie Classification) award the most science and engineering doctorate degrees (London et
al., 2014). This implies that students wishing to pursue a STEM doctoral degree, enrolling in the
most selective institutions at the undergraduate level is very important. This can put URM PhD
applicants at a disadvantage. Only 2% of African Americans and Latino high school students
enroll in the most selective institutions as opposed to 16% for Asian-Americans and 7% for
white students (Posselt, 2018). Furthermore, completing an undergraduate degree or doing
graduate studies in an R1 university means the student is exposed to higher academic standards
and rigors, has more experience with research, and access to faculty coming from a strong
research background than students attending a lower-tier university (Hollman et al., 2018).
35

When a URM candidate overcomes the vast hurdles to be admitted into a STEM PhD
program, only 36% of African American and 40% of Hispanic/Latino students completed their
PhD programs in the life sciences, engineering, and physical and mathematical sciences fields 7
years into their program (Sowell et al., 2008). The low completion rate is due to many factors
including loneliness (Brown, 2013), not belonging to the academic environment (Jackson et al.,
2016), problems transitioning from undergraduate institutions to research universities (Joseph,
2012) and lack of opportunities to interact and receive mentorship with faculty (Jackson et al.,
2016; Joseph, 2012).
STEM PhD Bridges programs address these deficiencies both in the admission process
and completion rates that disproportionately affect URM applicants and students (Rudolph et al.,
2019; Stassun et al., 2011; Stassun et al., 2018). Bridges programs focus on talented students that
might be missed with the usual graduate admissions criteria of undergraduate grades, GRE,
scores, and research experience (Stassun et al., 2011; Stassun et al., 2018). For admission into
the bridge programs, a holistic, comprehensive approach that goes beyond academic preparation
is used (Rudolph et al., 2019; Stassun et al., 2011; Stassun et al., 2018). This includes accounting
for the applicants' socioemotional skills necessary to succeed in the PhD program such as
perseverance and ability to focus on long-term goals and the applicant’s diverse set of life
experiences that would enhance scientific work (Rudolph et al., 2019).
Students pursuing the master’s degree in these programs provide access to STEM courses
and knowledge unavailable as undergraduates or even their first opportunity to do research
work (Rudolph et al., 2019; Stassun et al., 2011; Stassun et al., 2018). Once in the program,
students are provided with extensive mentoring (Rudolph et al., 2019; Stassun et al., 2011;
Stassun et al., 2018). For STEM students, mentoring was also found to positively correlate to
36

self-development activities that led to satisfaction, effective communication and involvement in
their majors (Holland et al., 2012) and particularly for minority students, mentoring increases
participation and persistence in STEM fields (Sánchez et al., 2018). Finally, most of the bridge
programs provide full financial support (Rudolph et al., 2019; Stassun et al., 2018) which allows
students to focus on their program of study instead of looking for part-time jobs to support
themselves or deter them from applying to graduate programs due to rising school debt (Kreutzer
& Gotto, 2015; Mangan, 2019). The American Physical Society Master’s-to-PhD Bridge
Program and Fisk–Vanderbilt Master’s-to-PhD Program discussed in greater detail below
exemplifies the structure and success of these programs.
The American Physical Society Master’s-to-PhD Bridge Program sends applicants to
over 25 vetted PhD programs nationally for consideration (Rudolph et al., 2019). The program
recruits applicants by requesting undergraduate physics departments to nominate students or
graduate physics departments to encourage rejected applicants to apply to the bridge program
(Rudolph et al., 2019). This process affords the former the opportunity to apply to many PhD
programs all at once, while the latter is given second opportunities to apply to PhD programs.
The effort has been successful with the placement of more than 100 students in 4 short years and
with continuing expansion (Rudolph et al., 2019). The retention rate of PhD students is 92%
(Rudolph et al., 2019) which is much higher than the national average of 50% (Joseph, 2012).
This is due in part of the rigorous vetting and evaluation process of participating PhD programs
to ensure students will be fully supported once they have matriculated into their programs
(Rudolph et al., 2019).
The Fisk–Vanderbilt Master’s-to-PhD Program is another bridge program and focuses on
physics and astronomy and is a partnership between Fisk University, an HBCU, and Vanderbilt
37

University, an R1 research university two miles away (Stassun et al., 2011). The core of the
program’s purpose is to build a relationship between master’s degree seeking students attending
Fisk University and Vanderbilt University (Stassun et al., 2011). This includes Fisk students
being jointly mentored and conducting research with faculty at both institutions and taking at
least one core PhD course at Vanderbilt University each term (Stassun et al., 2011). The end
result is that by the time Fisk students complete their master’s degree and enter Vanderbilt’s PhD
program, they would already have the critical foundations needed to succeed in a PhD doctoral
program (Joseph, 2012; Lisberg & Woods, 2018; Rudolph et al., 2019): strong mentor
relationships with faculty, valuable research experience, and adjusting to the academic
environment of a research university that can be far different from their undergraduate
environment (Jackson et al., 2016).
The retention for the PhD in the Fisk–Vanderbilt Master’s-to-PhD Program is 90%
(Stassun et al., 2011) and the 8 year PhD completion rate is 89% (Rudolph et al., 2019), nearly
double the 48% completion rate of URM students from a survey funded by NSF (Okahana &
Allum, 2015). Finally, even with obtaining a master’s degree on the way to a PhD, the time to
complete the PhD degree is only 1 year longer than students entering the PhD program directly
(Stassun et al., 2018).
Knowing STEM Educational Pathways
According to Moreira et al. (2019), the four top reasons why URM students don’t pursue
doctoral degrees are related to lack of knowledge of the program:
• consider graduate study or academia as a possible goal
• understand the process by which a student can pursue such a goal
• know the availability of potential career options for those who pursue that path
38

• think that they could be successful (p. 3)
Universities should share information to prospective applicants about the PhD and beyond as
research has shown that what motivates an applicant to get a science or engineering PhD degree
in the first place may be different from their actual experience in obtaining the degree (London et
al., 2014). Applicants should be made aware of career trajectories for post-PhD careers such as
the advancement and opportunities that are available, to get them interested in pursuing a STEM
PhD This information is especially important for URM applicants who cite a lack of mentorship
and lack of needed information as barriers to applying for graduate schools (Christophers &
Gotian, 2019). The lack of these resources means URM applicants are heavily reliant on
information provided from the programs, such as their website to decide whether or not to apply
to the program (Christophers & Gotian, 2019). One common statement from first-generation
URM students was “For me, being from a really small town, they won’t understand the PhD and
what that really means” (Moreira et al., 2019, p. 3).
URM applicants may deter from applying to competitive doctoral programs based on the
test scores and GPA of successful applicants published on the program’s website or that this data
is not published at all. In the highly competitive MD-PhD programs, the mean GPA was 3.79 +/-
0.19 and a mean MCAT score of 515.6 +/- 5.6 for applicants who matriculated into the programs
(AAMC, 2020). These scores can look overwhelming to any applicant except that both the GPA
and MCAT scores broadly range from 2.68 to 4.00 and from 497 to 528 respectively (AAMC,
2020). Of the 116 of the 121 MD-PhD programs that published admissions-related details on
their websites, more than half did not publish MCAT or GPA information and the programs that
did include this information only provided the mean (Christophers & Gotian, 2019) and not the
fuller picture of the broad ranges.
39

URM applicants cite a lack of mentorship and lack of needed information as barriers to
applying for graduate schools (Christophers & Gotian, 2019). The lack of these resources means
URM applicants are reliant on information from the programs’ website to decide whether or not
to apply to the program (Christophers & Gotian, 2019). When competitive programs’ websites,
such as for the MD-PhD programs, don’t provide any data or not a full picture of the range of
test scores and GPA of successful applicants, it can lead to applicants having imposter syndrome
of not being as well qualified as their peers to apply for these programs (Christophers & Gotian,
2019). For URM applicants, this problem is further exacerbated by lower GRE scores and
undergraduate GPA than non-URM applicants (Miller et al., 2019; Wilson et al., 2019).
Publishing the full range of test scores and GPAs for successful applicants in doctoral programs
like the MD-PhD programs would encourage more URM applicants to apply to these
competitive programs as it would help applicants reshape the current image that students
enrolled in programs have near perfect GPA and test scores (Christophers & Gotian, 2019).
Academic Institutional Barriers
Barriers for First-Generation Students
A student with both parents who have never completed college is considered a first-
generation student (Choy et al., 2000). About 30% of doctorate recipients are first-generation
students (Roksa et al., 2018) and amongst URM this percentage is even higher. For URM, the
number of Black and Latinx students that were first-generation students was nearly half,
compared to about a quarter for white students (Rudolph et al., 2019). There were notable
differences in the experience of first-generation students’ experiences versus continuing-
generation students that negatively impacted student academic success.
40

The odds of first-generation students attending college after high school is already
stacked against them. Only 27% of first-generation students attend college compared to 42% and
71% for students with parents that had some college and parents who were college graduates,
respectively (Choy et al., 2000). Two important factors, at the high school level, that contribute
to this wide discrepancy are taking a rigorous mathematics curriculum and planning for and
applying to college (Choy et al., 2000). There is a strong correlation between participating in a
rigorous mathematics curriculum in high school and attending college (Choy et al., 2000).
Students whose parents never attended college are much less likely than students whose parents
graduated from college to participate in any rigorous mathematical curriculum (Choy et al.,
2000). In terms of college planning, there is a notable percentage decrease from parents with a
college degree to parents without a college degree in discussing college test preparation exams
(16% versus 27%) and postsecondary plans (42% versus 61%) with their high school-age
children (Choy et al., 2000). College preparation activities of visiting colleges, seeking financial
aid, and educational opportunity programs also decrease with parents' highest level of education
(Choy et al., 2000). Once in college, first-generation students continue to face further
disadvantages. Faculty encountered issues about the students’ ability to understand and follow
their expectations in three broad areas that contribute to student retention and academic
success—workload and priorities, the explicitness of expectations and assignments, and issues
related to communication and problem solving (Collier & Morgan, 2008). This explicit and
implicit knowledge is what contributes to students’ mastery of the college student role and
possessing it contributes to the greater systematic academic success of continuing-generation
students as compared to first-generation students (Collier & Morgan, 2008).
41

If a first-generation student obtains an undergraduate degree and is able to enter a PhD
program, further challenges and disadvantages await the student. First-generation students
enrolled in PhD programs are less likely than their continuing-generation students to have
obtained their undergraduate degree from a high research doctoral-granting university, classified
as R2 on the Carnegie Classification or did graduate studies from a very high research doctoral
granting university, classified as R1 (Carnegie Classification of Institutions of Higher Education
by Indiana University Center for Postsecondary Research, 2017; University of Chicago, 2008).
This puts first-generation PhD applicants at a disadvantage as faculty generally prefer applicants
that come from elite institutional affiliations (Rudolph et al., 2019) that are usual R1 classified
universities. Besides better admissions prospects, completing an undergraduate degree or doing
graduate studies in an R1 university means the student is exposed to higher academic standards
and rigors, has more experience with research, and access to faculty coming from strong research
background than students attending a lower-tier university (Hollman et al., 2018). These factors
are important in better preparing the student for the rigors of a PhD program. This lack of
preparedness presents challenges to first-generation students in navigating around their doctoral
program as the students lack the experience and knowledge of the “unspoken” rules in high
research-intensive universities (R1 and R2 institutions) (Roksa et al., 2018) where most of the
STEM PhD degrees are granted (Buswell, 2017). This is one of the factors that lead to high
attrition rates in science and engineering of approximately 50% (Joseph, 2012).
Biases in Admission Process for Doctoral Programs
Faculty members in elite institutions tend to favor PhD applicants who come from similar
institutions (Posselt, 2018) and institutions classified as “very high research activity” (R1 in the
Carnegie Classification) award the most science and engineering doctorate degrees (London et
42

al., 2014). This implies that for students wishing to pursue a STEM doctoral degree, enrolling in
the most selective institutions at the undergraduate level is very important. This can put URM
PhD applicants at a disadvantage. Only 2% of African Americans and Latino high school
students enroll in the most selective institutions as opposed to 16% for Asian-Americans and 7%
for white students (Posselt, 2018). Another disadvantage is that first-generation students enrolled
in PhD programs are less likely than their continuing-generation students to have obtained their
undergraduate from a high research activity university (R2 in the Carnegie Classification) or
very high research activity university (R1 in the Carnegie Classification) (Rudolph et al., 2019).
Half of the Black and Latinx students enrolling in the doctoral programs are first-generation
students (Phruksachart, 2017).
The Graduate Record Exam (GRE) score and undergraduate grade point average (GPA)
are the two important objective criteria used by most universities to evaluate applicants for
admission into their STEM programs (Hall et al., 2017; Miller et al., 2019; Wilson et al., 2019).
On average, URM applicants have lower GRE scores (Wilson et al., 2019) and undergraduate
GPA (Miller et al., 2019) as compared to their white and Asian counterparts and serves as major
barriers for more URM representation in the STEM PhD programs (Miller et al., 2019; Sealy et
al., 2019). The use of GRE scores for graduate admissions is already controversial (Wilson et al.,
2019). In recent years, a growing body of research is showing that GRE scores are not good
predictors of student productivity (first-author student publications) or degree completion in the
STEM PhD programs (Hall et al., 2017; Miller et al., 2019) and may have missed the applicants
that are most capable of completing the program (Rudolph et al., 2020). A study of STEM PhD
programs found that male students who completed the STEM doctoral degrees had significantly
lower GRE scores than those students who left the program (Petersen et al., 2018). A study of
43

Physics PhD students showed that their GRE score is not associated with PhD completion
(Miller et al., 2019). The attrition rates for science and engineering PhD programs are
approximately 50% (Joseph, 2012), yet there was less than 10% completion rate difference for
Physics PhD applicants who scored in the 10th percentile versus those scoring in the 90th
percentile (Miller et al., 2019). Similarly, the GRE scores of Biomedical PhD applicants in
another study found no association with PhD completion rates (Hall et al., 2017). Most URM
students earn their undergraduate degrees in public universities (NSF, 2017d) while white
students make up 87% of all undergraduates at private universities (Brown, 2016).
The grades given out at public universities are usually about 1/3 letter grade lower than
private universities (Miller et al., 2019). Furthermore, many URM students enrolling in
engineering programs face difficulties during their first 2 years, but eventually do well
academically in the last 2 years if they persist with the program (Reichert, 2013). Their first 2
years of difficulties means these students will have lower undergraduate GPA and potentially
eliminate them as strong candidates to the STEM PhD programs. It is important that admissions
committees weigh these facts into their admissions considerations, otherwise undergraduate
GPA, along with cutoff scores for GRE would significantly reduce the diversity of the applicant
pool to their PhD programs. Universities that use a holistic admissions review are able to
increase the diversity of their applicant pool (Miller et al., 2019; Wilson et al., 2019). A holistic
review process prevents a part(s) of the application, such as low GRE scores or low
undergraduate GPA from disproportionately impacting the admission chances for an applicant
(Wilson et al., 2019). The College of Nursing at the University of Illinois, Chicago was able to
increase the number of URM applicants that were offered admissions from 36.8% to 42.5%
when the school changed a holistic review process (Scott & Zerwic, 2015). The University of
44

Texas MD Anderson Cancer Center saw an increase in URM students in their biomedical
sciences PhD programs after taking similar actions (Wilson et al., 2018).
Hostile Campus Climate
Campus environments play a very important role in the academic success of URM
students in STEM fields (Lancaster & Xu, 2017). Furthermore, URM students find it difficult to
have a positive identity in a campus climate that is hostile to them (Sánchez et al., 2018). A
hostile campus environment leads to attrition in the STEM fields among URM students
(Lancaster & Xu, 2017). Macroaggressions, stereotype threat and unconscious bias coupled by
the lack of faculty mentor and the isolation and loneliness of being the single or handful of URM
students in a doctoral program (Harmon 2019; NSF, 2018c; SACNAS Harvard Chapter, n.d.)
contribute to a hostile campus climate (Jackson et al., 2016; Rattan et al., 2018).
Microaggressions are every day, subtle, intentional—and oftentimes unintentional efforts,
interactions, or behaviors by members of the dominant culture to communicate some sort of bias
towards people of color or marginalized group (Limbong, 2020). There are three forms of
microaggressions: microassaults, microinsults, and microinvalidations (Sue et al., 2007).
Microassaults are intentional efforts to hurt or harm an individual in a discriminatory way using
negative racial or ethnic epithets while not intending to be offensive (Nadal, 2014). Microinsults
are insensitive remarks, rude behavior, or demeaning comments that are unintentionally
discriminatory (Nadal, 2014). Microinvalidations are communications intended to invalidate or
undermine the experiences, feelings, and thoughts of historically marginalized groups (Sue et al.,
2007).
Although racial microaggressions are usually brief and commonplace indignities, the
impact on students is long-lasting and inflicts cumulative wounds to people of color (Sue et al.,
45

2007). A phenomenon known as racial battle fatigue in postsecondary education arises from the
constant physiological, psychological, and behavioral strains imposed on students of color
because of microaggressions (Smith, 2010). For URM students in the STEM disciplines,
common examples of microaggression are the widespread negative stereotypes that they don’t
have mathematical or scientific abilities (Rattan et al., 2018) or responding less often and more
slowly to prospective URM applicants (Rudolph et al., 2019). Microaggressions can be
detrimental on whether a URM student sleets and remains in science academics (Sue et al.,
2007). For example, documented microaggressions of academic advisors against URM STEM
graduate students led to difficulties in developing and building a STEM identity (Sue et al.,
2007), one of the main characteristics attributed to high-achieving students whose goals are to
pursue graduate education and a career in scientific research (Hurtado et al., 2011). A common
recommendation for addressing microaggression is to implement cultural competency training
for faculty, staff, and students (Sue et al., 2007).
According to Steele (1997), stereotype threat is
The social-psychological threat that occurs when one is in a situation or doing something
for which a negative stereotype about one’s group applies. This predicament threatens
one with being negatively stereotyped, with being judged or treated stereotypically, or
with the prospect of conforming to the stereotype. (p. 614)
Stereotype threat occurs when an individual from a marginalized group knowledge negative
stereotype exists in reference to their group and demonstrates apprehension about confirming the
negative stereotype by engaging in particular activities or feel pressure to perform (NIH, 2014).
Stereotype threats contribute negatively to students’ well-being and sense of belonging on
campus (Sánchez et al., 2018; Yadav & Seals, 2019). Stereotype threat is also related to students
46

building an identity (e.g., scientist identity) and self-efficacy (Callahan et al., 2018). Demeaning
stereotypes can lead to academic underperformance when the intelligence of students is being
evaluated (Steele, 1997). Students perform better academically when there is an absence of
stereotype threat (Callahan et al., 2018).
A typical stereotype threat STEM URM students encounter is being questioned about
their abilities due to racial stereotyping or confronted with “fears of fulfilling stereotypes or
addressing feelings about being representatives of their race” (Mondisa, 2018, p.
296). Stereotype threats are cumulative and corrosive to URM students’ sense of interest and
identity in STEM (Callahan et al., 2018). Stereotype threats have led to direct URM students’
attrition in STEM programs more than a lack of academic preparation (Beasley & Fischer, 2012).
Students' worries of encountering reinforce negative stereotypes about their group lead to
underperformance and their “evaluative concerns elicit a concomitant increase in physiological
arousal and self-monitoring, resulting in reduced working memory capacity and
underperformance” (Ben-Zeev et al., 2017, p. 2). There are two ways for URM students in
STEM to overcome stereotype-based concerns and threats. The first way is to provide tutorials to
students on stereotype threat, by highlighting that it is not a unique experience in order to
normalize and depersonalize it and to help locate the threat-induced anxiety in their
environments (Johns, 2005). The second way is to provide opportunities for URM students to
engage their knowledge or responses related to their life experience or community in order to
help overcome stereotype-based concerns (Ben-Zeev et al., 2017).
Unconscious or implicit biases are social stereotypes about certain groups of people that
humans unknowingly draw upon assumptions about individuals and groups to make decisions
about them usually in a way that’s considered to be unfair (UCSF Office of Diversity and
47

Outreach, 2020d). Unconscious bias is harmful in that it values some people more than others.
Unconscious bias occurs involuntarily, automatically, and beyond one’s awareness (Allen &
Garg, 2016). Unconscious bias is often incompatible with one's conscious values and positive
intentions (UCSF Office of Diversity and Outreach, 2020d). Unconscious bias affects formal and
informal decision-making processes (Allen & Garg, 2016) with repercussions for URM students
such as the recruitment of students (Moss-Racusin, 2012). Unconscious bias on the part of
faculty can lead to which students to mentor, work on a research project, invite to conferences,
recognize their contributions, or give credit to. Ways to reduce unconscious bias towards URM
students at the institution level are educate people about unconscious bias and how to lessen its
effects through conducting holistic reviews and interviews, agreeing on guidelines for decision
making, and reviewing student candidates and their application materials for research
opportunities, conferences, mentorship, or recruitment that have been passed over due to social
identity cues like gender, race, or name that invoke unconscious bias (Allen & Garg, 2016).
The scarcity of URM students in the doctoral programs (SACNAS Harvard Chapter, n.d.)
makes the students that are in these programs feel completely isolated (Harmon, 2019) and only
enforces faculty members' low expectations and negative stereotypical perceptions of URM
students (Jackson et al., 2016). The lack of prosocial values from majority non-URM peers and
faculty that are firmly rooted in URM students’ cultural and ethnic background (Thoman et al.,
2015) only adds to the incongruity and adds further to the campus hostility.
The majority of URM doctoral students obtain their undergraduate degree in MSIs or
HBCUs before going to a research university that is mainly PWIs (Brown & Davis, 2001;
Joseph, 2012; NSF, 2019a). Once in research universities, these students try to operate as they
did in their undergraduate environment such as communicating openly and honestly in hopes of
48

finding support (Whittaker & Montgomery, 2012). To faculty at research universities, this is a
sign of the student not adjusting or performing well or possibly may be incapable of adjusting or
performing (Whittaker & Montgomery, 2012). This cultural difference results in URM students
having difficulties navigating around their doctoral program as the students lack the experience
and knowledge of the “unspoken” rules in high research-intensive universities. Furthermore,
URM students often find a discontinuity between their communities of origin and their research
university campus environment (Whittaker & Montgomery, 2012) that majority of students do
not often encounter. All this, along with the microaggressions, stereotype threat and unconscious
bias from peers and faculty (Sánchez et al., 2018; Steele, 1997; Yadav & Seals, 2019) results in a
perception of a hostile campus environment that leads to attrition (Lancaster & Xu, 2017).
Instead of feeling inspired and confident, attending graduate school made URM students feel
boxed in and their identities challenged (Jackson et al., 2016).
Hostile campus environments for URM STEM students can be addressed with increased
URM faculty. Non-URM faculty members create additional hostilities to URM students in the
form of sending mixed messages to URM students about their worth and belonging in science
and responding less often and more slowly to prospective URM applicants (Rudolph et al.,
2019). Racial, cultural, and ethnic differences can also lead to difficulties and challenges with
communication (McGee, 2016). Therefore, there is a strong desire for URM students to prefer
mentors that are like them (Whitfield & Edwards, 2011). In a hostile campus climate, URM
faculty mentors may need to draw from their own racial encounters and experiences to counsel
URM students who are questioned about their abilities due to racial stereotyping or confronted
with “fears of fulfilling stereotypes or addressing feelings about being representatives of their
race” (Sánchez et al., 2018, p. 294). Unfortunately, the faculty in higher education doesn’t mirror
49

the racial and ethnic makeup of the student body they serve (Yadav & Seals, 2019). With lesser
URM faculty, URM students in PWI are less likely than their non-URM peers to have the
personal interactions with faculty (Anneke et al., 2018) needed to alleviate and address the long-
lasting and inflicts cumulative wounds from microaggressions, stereotype threat, and
unconscious bias (Sánchez et al., 2018; Steele, 1997; Yadav & Seals, 2019). Institutions,
particularly PWIs, must diversify their faculty if they are to address the perception of hostile
campus environments faced by URM students (Sánchez et al., 2018).
Lack of Faculty Diversity
One of the major reasons attributed to the low representation of URM in the STEM
programs, particularly at the doctorate level, is the lack of faculty diversity in the STEM fields
Gibbs et al., 2015; Hassouneh et al., 2014). There is a strong desire for URM students to prefer
mentors that are like them (Whitfield & Edwards, 2011). Despite several decades of efforts to
recruit and retain URM faculty in the STEM fields, there has only been a modest success (Allen-
Ramdial & Campbell, 2014).
The lack of URM faculty directly affects the recruiting, mentoring, and retaining minority
students. URM students have a greater need for social support as they have feelings of isolation
and loneliness (Harmon, 2019) and encounter negative stereotype threats (Mondisa, 2018) in
their course of studies. Social support has proven to cultivate the academic success of URM
students throughout the STEM pipeline, from high school to the doctoral level (Azmitia et al.,
2013; Hurtado et al., 2011; Syed et al., 2011; Williams et al., 2017; Yadav & Seals, 2019). Social
support foments the development of a scientist identity, finding prosocial values through
education and creating a sense of belonging to the campus (Azmitia et al., 2013; Gibau, 2015;
Syed et al., 2011; Williams et al., 2017; Yadav & Seals, 2019), all of which are important
50

motivators for recruiting and retaining URM students in the STEM PhD programs (Chemers et
al., 2011; Hurtado et al., 2011; Jackson et al., 2016).
The social support that comes from a URM faculty member is more effective and
impactful to URM students than from non-URM faculty (Anneke et al., 2018; Brittian et al.,
2009; Cole & Espinoza, 2008; Pyne & Grodsky, 2019). Non-URM faculty can lead to racial,
cultural, and ethnic differences that result in difficulties and challenges with communication
(McGee, 2016). It is also difficult for URM students to create a career or scientist identity when
there is a lack of role models like themselves that have made it (McGee, 2016). As a result, the
scarcity of URM faculty members in the STEM fields results in talented URM students opting
out of STEM education and careers despite the significant investment of time, resources, and
funds by government organizations and academic institutions (Maton et al., 2016; Moreira et al.,
2019; Salto et al., 2014; Syed & Chemers, 2011; Williams et al., 2017).
Limited Opportunities
Research, mentorship, presenting in conferences, and publishing papers are all critical to
student persistence and academic success in graduate STEM programs, yet these essential skill-
building opportunities are lacking for URM students (Cooper et al., 2019; Malotky et al., 2020;
Moreira et al., 2019; Whittaker & Montgomery, 2012).
Research experiences develop and sustain student interests in STEM (Estrada et al.,
2018) while contributing to student persistence till STEM graduate degree completion (Cooper et
al., 2019). Increasing research experiences positively correlates to increase science self-efficacy
and forming a science identity for URM students (Estrada et al., 2018). Yet research
opportunities for URM students are lacking and difficult to obtain (Toretsky et al., 2018). URM
students from socially or economically disadvantaged backgrounds are also unaware of the
51

importance of or how to obtain research opportunities (Toretsky et al., 2018). Even when URM
students are presented with research opportunities, they are more likely than their white peers to
leave their research opportunities due to working on less desirable tasks, working harder, or
weren't learning important skills or knowledge (Cooper et al., 2019). When research
opportunities focus on community-engaged learning that is relevant to the URM students’
background, experience, race, or community, it leads to increased lab participation (Thoman et
al., 2015) and academic performance in URM students (Malotky et al., 2020) as URM students
place great importance of finding prosocial meanings in their science education and future
science careers (Jackson et al., 2016).
Mentoring increases minority students’ participation and persistence in STEM fields
(Sánchez et al., 2018), but fruitful mentorships are not widely available to URM students (Allen
& Garg, 2016; Callahan et al., 2018; Moreira et al., 2019). NSF sponsored programs to increase
diversity in the STEM student body, such as the Opportunities for Enhancing Diversity in the
Geosciences (OEDG), recommends increasing mentorship opportunities in order to recruit and
retain URM students (Callahan et al., 2018). Unconscious bias amongst predominately non-
URM faculty members can play a role in selecting who to mentor and institutions have been
guiding faculty leaders to mentor diverse protégés (Allen & Garg, 2016). The lack of non-URM
faculty who can serve as mentors to URM students further reduces the effectiveness of a
mentorship (Anneke et al., 2018; Brittian et al., 2009; Cole & Espinoza, 2008; Pyne & Grodsky,
2019).
Attending and presenting at conferences increase student academic success and
persistence in the STEM PhD programs (Kim‐Prieto et al., 2013; Moreira et al., 2019). The
conferences allow students the opportunity to learn from peers, mentors, and collaborators. URM
52

STEM PhD students who are bridges to the doctoral programs (where PhD completion rates are
twice the national average) noted that presenting at conferences was extremely helpful and
provided the skills needed to succeed in their programs (Moreira et al., 2019). The same students
also indicated the importance of having more conferences and institutional financial support to
attend and present at the conferences (Moreira et al., 2019).
One area that is critical for student academic success in the doctoral program, but
infrequently discussed in the current literature is the number of STEM doctoral publications by
URM students. The publication is the gold standard by which entry into the professoriate is
judged (Mendoza-Denton et al., 2017). In the STEM fields, students feeling prepared during the
undergraduate to graduate transition was associated with the publication of papers in high impact
academic journals (Martinez et al., 2018). In recent years there has also been a shift in the STEM
fields of making first publications happen during the students' graduate years instead of during
the postdoc period (Martinez et al., 2018). Despite the growing importance and need to publish
papers in their graduate years, “URM students are under-encouraged to publish and are provided
fewer opportunities to present their research” (Martinez et al. 2018, p. 8) than non-URM peers.
Campus factors play an important part in graduate success in STEM fields that goes
beyond individual factors. The ability of institutions to provide more research, mentoring,
conference, and publication opportunities will go along towards helping URM students in
doctoral programs increase their persistence and rate of degree completion. Doctoral programs
that provide these opportunities and are well structured also have the added benefit of being able
to recruit more URM students into their programs (Callahan et al., 2018; Jackson et al., 2016;
Moreira et al., 2019; Sánchez et al., 2018).
53

Conceptual Framework
The conceptual framework that will be used for this dissertation is Clark and Estes'
(2008) gap analysis framework. This conceptual framework focuses on closing performance gaps
by looking at knowledge, motivation, and organizational (KMO) causes. The gap analysis looks
at the people involved to determine whether they have adequate knowledge, motivation, and
organizational support for the goals of closing a performance gap (Clark & Estes, 2008). The
purpose of this dissertation is to investigate KMO influences for URM students to achieve
acceptance to, choose to attend and complete their STEM PhD programs at a very high research
activity doctoral university (R1 in the Carnegie Classifications of Institutions of Higher
Education). The performance gap here is that URM received only a quarter of the science and
engineering PhD degrees relative to their percentage makeup of America’s college-age
population; URM make up 36.8% of America’s college-age (18-24) population (NSF, 2017b),
but only received a scant 7.9% of all PhD degrees in the science and engineering fields (NSF,
2017a). From the review of literature, the KMO influences for URM students that impact their
matriculation, retention, and graduation from STEM PhD programs are shown in Tables 1
through 3.

54

Table 1
Knowledge Influences on URM Students in STEM PhD Programs
Influences Sources
Knowledge deficiencies in first-generation students Kong et al., 2013; Roksa et
al., 2018; Witkow &
Fuligni, 2011
Benefits of student mentorship Lisberg & Woods, 2018;
McGee, 2016; Roksa et
al., 2018; Sánchez et al.,
2018
Master’s degree as stepping stone to doctorate degree Okahana et al., 2018;
Rudolph et al., 2019;
Stassun et al., 2011;
Stassun et al., 2018
Understanding STEM educational pathways Christophers & Gotian,
2019; Maton et al., 2016;
Moreira et al., 2019

Knowledge, or a lack of, may be a deterring factor for URM students not pursuing a
doctoral degree. Data collected from the Texas A&M University System (one of the nation’s
largest system of higher education), reveals that two of the main reasons URM students do not
pursue STEM doctoral degrees are not knowing the career options available for such a doctoral
degree and not understanding the process to reach such a goal (Moreira et al., 2019). One major
underlying reason for this is because nearly half of URM STEM PhD students are first-
generation students (Rudolph et al., 2019). First-generation students don’t have as much
knowledge that continuing-generation students have to get into and succeed in college (Roksa et
al., 2018). For example, research experience is critical for succeeding in STEM programs and
continuing-generation students were 1.9 times more likely than first-generation students to stay
in research knowing they will be gaining important skills or knowledge (Cooper et al, 2019). A
55

focus on addressing the skills and knowledge deficiencies of first-generation students is
necessary to increase URM participation and persistence in STEM PhD programs.
Mentorship allows students to gain familiarity and knowledge of on-campus student
support programs (student organizations, faculty office hours, undergraduate research, and
academic support) that are known to increase student success (Lisberg & Woods, 2018). College
students’ GPA and retention rate are positively correlated with the frequency of mentor-protégé
contact (Campbell & Campbell, 1997). Furthermore, for URM students in general, mentoring is
especially important as it builds professional networks and enhances research skills that are
needed in closing the gap that exists between URM and non-URM students (Lisberg & Woods,
2018).
Having a prior master’s degree is an important path for URM students pursuing STEM
doctoral programs (Okahana et al., 2018). A master’s degree can help URM students acquire the
knowledge and preparation needed to be successful in STEM doctoral programs (Okahana et al.,
2018) such as access to courses and knowledge unavailable to them as undergraduates or gain
research experiences (Rudolph et al., 2019). There is a positive correlation between having a
master’s degree and earning a doctorate within 10 years for African American students and male
URM students in the STEM fields (Rudolph et al., 2019). Masters-to-PhD Bridge Program are
partnership between STEM master’s programs offered at regional universities with STEM
doctoral programs at research universities to leverage this positive correlation (Okahana et al.,
2018; Rudolph et al., 2019; Stassun et al., 2011).

56

Table 2
Motivational Influences on URM Students in STEM PhD Programs
Influences Sources
Self-efficacy and motivation Bandura, 1977; Chemers
et al., 2001; Katz et al.,
2014
Finding prosocial values and identities Guy & Boards, 2019;
Jackson et al., 2016;
Thoman et al., 2014
Sense of belonging to campus Harmon, 2019; Hausmann
et al., 2017; Hurtado,
1997; Rattan et al., 2018
Building persistence through mentorship and research
opportunities
Cooper et al., 2019;
Estrada et al., 2018;
Malotky et al., 2020;
Mantai, 2019; Moreira
et al., 2019
URM faculty and mentors Joseph, 2012; Sánchez et
al., 2018; Whittaker &
Montgomery, 2012;
Yadav & Seals, 2019

Motivation is what determines the direction of a person’s behavior, level of effort and
level of persistence in an organization. In short, it explains why people behave the way they do
in an organization. This study is to determine the motivation influences for why URM students
pursue a STEM PhD program and why they either leave or complete the program. The review of
literature identifies several key areas of student motivation.
Students with high self-efficacy especially in mathematics and science are more likely to
persist in their program to completion even when encountering challenges or difficulties
(Chemers et al., 2001). Having a sense of belonging to an environment is a fundamental human
need (Rattan et al., 2018) and students’ sense of belonging to a campus increases their
motivation towards degree completion (Hausmann et al., 2017). When students are provided
57

with mentorship and research opportunities, it further enhances a sense of belonging and
increases interest and persistence in their program (Estrada et al., 2018; Malotky et al., 2020).
For URM students there are unique needs that if addressed, would further increase their
motivation. There is a strong desire for STEM URM students to prefer mentors that are like them
(Whitfield & Edwards, 2011) and when that occurs, it further strengthens the students’ academic
motivation (Yadav & Seals, 2019). STEM URM students place great value in helping others and
giving back to the community through one’s work (Jackson et al., 2016) and carried over to the
doctoral level where graduating URM doctoral students highlight the importance of prosocial
values for persisting in science education (Jackson et al., 2016).

Table 3
Organizational Influences on URM Students in STEM PhD Programs
Influences Sources
Barriers for first-generation students Choy et al., 2000; Rudolph
et al., 2019
Admission biases in doctoral programs Hall et al., 2017; Miller et
al., 2019; Posselt, 2018;
Wilson et al., 2019
Hostile campus environment Ben-Zeev et al., 2017;
Limbong, 2020; Nadal,
2014; Sánchez et al.,
2018; Steele, 1997
Lack of faculty diversity Gibbs et al., 2015;
Hassouneh et al., 2014;
Williams et al., 2017;
Yadav & Seals, 2019
Limited opportunities Mendoza-Denton et al.,
2017; Moreira et al.,
2019; Toretsky et al.,
2018;

58

Organizational influences are critical in explaining a performance gap (Clark & Estes,
2008). When students have top motivation and exceptional knowledge and skills, organizational
policies, processes, or resource levels that prevent closing a performance gap can be viewed as
organizational barriers (Clark & Estes, 2008). There exist several organizational barriers that
were identified in the review of literature that is preventing increased URM enrollment and
persistence in the STEM PhD programs.
Nearly half of URM STEM PhD students are first-generation students (Rudolph et al.,
2019). There are institutional barriers that first-generation students encounter throughout the
STEM pipeline, from high school to the doctoral programs. As these students progress in their
STEM field of study, these barriers reduce the number of students at each step. In other words,
there is “the loss of diversity that occurs as one moves through tiers of educational and/or career
advancement” (Mattingly, 2019, para. 2).
Additional institutional barriers that contribute to this “leaky” pipeline for URM students
in particular are the lack of URM faculty mentors (Yadav & Seals, 2019), hostile campus
environments (Sánchez et al., 2018), biases in the admission process (Posselt, 2018) and limited
opportunities to acquiring the skills and increasing persistence in STEM doctoral programs
(Mendoza-Denton et al., 2017; Moreira et al., 2019; Toretsky et al., 2018).
The interrelationships between the KMO influences are displayed in Figure 1. When only
two of the KMO influences align, URM students will lack all the necessary support to close the
performance gap. For instance, when only the organizational and knowledge influences align,
URM students will not have the motivation and will lead to the high attrition in the PhD
programs that exist today (Joseph, 2012). When only knowledge and motivational influences
align, the organizational barriers such as bias in the admission process or a chilly, or racial
59

campus climate will lead to less URM applicants matriculating in STEM PhD programs as well
as high attrition for those that do get into the program. When only organizational and
motivational influences align, URM students are not equipped with the research
skills/experience, learning strategies, networking knowledge, and study skills to persist and
successfully graduate from a doctoral program. According to Clark and Estes (2008), goal
achievement can only be reached when all three influences are in place and aligned. Thus, the
conceptual framework theorizes that under this scenario, URM students will be able to close the
performance gap being analyzed in this study and result in the percentage of science and
engineering PhD degrees awarded to URM students being closer to their percentage makeup of
America’s college-age population. The research methodology, discussed in detail in Chapter
Three, will use a quantitative method approach to test this assumption and Chapter Four will
discuss the findings.

60

Figure 1
KMO Influences on URM STEM PhD Students

61

Summary
The review of existing literature highlights major barriers and obstacles for URM
applicants and students whose educational pathway is to lead to the completion of a STEM PhD
degree. This educational pathway can be commonly referred to as a STEM pipeline. URM
STEM PhD degree seekers have “leaky pipeline”. This terminology is used to URM students
who “leak” from the STEM pipeline as they progress in their STEM field of study from high
school to bachelor’s to master and to doctoral level programs. In other words, there is “the loss
of diversity that occurs as one moves through tiers of educational and/or career advancement”
(Mattingly, 2019, para. 2). The existing literature suggests the leaky pipeline can be attributed to
several major factors including institutional barriers and biases, scarcity of URM faculty,
challenges of first-generation students pursuing graduate degrees, not informed, misalignment of
student values in STEM versus institutions’, biases in the doctorate admission process, and a
hostile campus climate with stereotype threats and loneliness. These major factors are grouped
into the knowledge, motivational and organizational influences that are summarized in Tables 1
through 3. Chapter Three will evaluate these barriers in the context of URM students enrolled in
UCSF’s basic and biomedical sciences PhD programs using Clark and Estes’ (2008) KMO gap
analysis.
62

Chapter Three: Methodology
This chapter will discuss the overview of the methodological design which will use a
quantitative method, the knowledge, motivation, and organizational influences to be examined,
as well as the methodology for selecting participants, the instrumentation, data collection, and
analysis. The chapter will also include discussions on the description of the researcher and
overall strategies used for maximizing the validity and reliability of this study.
Research Questions
The purpose of this project is to conduct an evaluation study of UCSF in its ability to
enroll nearly three times the national average of URM students in the basic and biomedical
sciences PhD programs. URM students make up 22% of the current student body in these
programs (UCSF Graduate Division, 2020a) while the national average for STEM PhD degrees
awarded to URM students is 7.9% (NSF, 2017a). While a complete evaluation will focus on all
stakeholders, for practical purposes, this evaluation will focus on URM students who are
enrolled in the basic and biomedical sciences PhD programs at UCSF and evaluate from their
perspective using Clark and Estes’ (2008) gap analysis framework to understand their
knowledge, motivation, and organizational influences. Two research questions will guide the
study:
1. What are URM students’ knowledge, motivation, and organizational influences to
achieve acceptance and persist in their STEM PhD programs at UCSF?
2. What are the recommendations, for knowledge, motivation, and organizational
resources, for UCSF to increase and retain URM students in the STEM PhD
programs?
63

Overview of Design
This is an evaluation study of the University of California, San Francisco (UCSF), an
urban-based, “very high research activity” doctoral-granting university (R1 in the Carnegie
Classifications of Institutions of Higher Education) The study is to evaluate the knowledge,
motivation, and organizational (KMO) influences of URM students enrolled in STEM PhD
programs at the university that led them to get accepted to and persist in these extremely
selective programs. These assumed KMO influences impacting URM students in STEM PhD
programs were presented in Chapter Two:
● declarative, procedural, and metacognitive knowledge
● theories of expectancy-value and self-efficacy for motivation
● organizational changes
The methodological design for this study will be a quantitative method. The data
collection will consist of an online survey. The researcher will work with the Graduate Division
and student registered campus organizations to identify student participants who meet the criteria
for the study (see Participants section below).
The data collection will start with an online survey that will be created from Qualtrics©
and sent to students. The survey contains several broad categories of questions:
● student demographic data including ethnicity, gender, and first-generation
● student program data including doctoral program information and level
● student background including their educational pathway and experience in their
STEM pipeline leading to up to the PhD program
● KMO influences on URM students in STEM PhD Programs covered in the review of
the literature and summarized at the end of Chapter Two
64

● provide comments and feedback that the survey questions might not have covered or
addressed in the questions specifically
The rationale for a quantitative methodology and specifically a Qualtrics© survey being
used is based on several reasons:
● Participants do not need to report to a specific location to collect the data as it can be
done online. This reduces the risk of contact during this Covid-19 Pandemic.
● Data can be collected quickly and with minimal effort and costs using online surveys.
● Data collected is reliable and repeatable. The repeatable factors can be used to
measure the outcomes or success of any decision or recommendation made.
Furthermore, the repeatable allows this study to be expanded to other R1 institutions
to increase validity.
● Statistics generated from the data is a reliable resource with the confidence interval
needed to do decision-making processes or suggest recommendations.
● Data collected can be generalized from the sample group to the overall demographics
looked at in this study.
● Data collection from a quantitative methodology can be randomized to prevent biases
from entering into the data.
● Participation information provided can remain anonymous in order to ensure student
confidentiality and a safe and secured environment to provide their responses. This
type of environment makes participants more willing to provide truthful responses.
This is important in a tighter knit doctoral program (versus an undergraduate
program) where students might have reservations about criticizing the programs and
faculty that they work closely with and grant them their doctoral degrees.
65

Table 4
Data Sources for Research Questions
Research questions
Online
survey
What are URM students’ knowledge, motivation, and
organizational influences to achieve acceptance and persist in
their STEM PhD programs at UCSF?
X
What are the recommendations, for knowledge, motivation, and
organizational resources, for UCSF to increase and retain URM
students in their STEM PhD programs?
X

Research Setting
UCSF employs more than 30,000 employees with an economic impact of $9 billion
(UCSF, 2018). The University of California, San Francisco (UCSF) is made up of three related,
but separate entities – patient care, research, and education (UCSF, 2018) that work closely
together to meet its vision of being “the best provider of healthcare services, the best place to
work, and the best environment for teaching and research.” (UCSF Health, 2020, para. 3). The
focus of this dissertation is on the education component of UCSF. As of fall 2019, UCSF had
3,212 students, 1,659 residents and fellows, and 1,180 postdoctoral scholars (UCSF Office of
Institutional Research, 2020). Students at UCSF are enrolled in either one of the four
professional schools (consisting of the schools of dentistry, medicine, pharmacy, and nursing) or
in the Graduate Division where PhD degrees are awarded. About a quarter of the student
population, or 821 students are enrolled in the PhD programs (UCSF Office of Institutional
Research, 2020). The focus of the study is on PhD programs.
66

The Graduate Division offers 19 PhD programs in the basic, translational, and
social/population sciences as shown in Table 5. The study is only interested in looking at the
basic and biomedical sciences PhD programs as these are considered STEM disciplines. The
Graduate Division also offers 10 master’s degrees and two professional doctorates. On top of
graduation education, the Graduate Division is the home of postdoctoral training at UCSF. There
are 1,000 postdoctoral scholars doing research across the campus.

Table 5
PhD Programs Offered at UCSF
Basic and Biomedical Sciences Social and Populational Sciences
Biochemistry and Molecular Biology (Tetrad) Global Health Sciences
Bioengineering (joint with UC Berkeley) History of Health Sciences
Biological and Medical Informatics Medical Anthropology
Biomedical Sciences Nursing
Biophysics Sociology
Cell Biology (Tetrad)
Chemistry and Chemical Biology
Developmental and Stem Cell Biology
Epidemiology and Translational Science
Genetics (Tetrad)
Neuroscience
Oral and Craniofacial Sciences
Pharmaceutical Sciences and Pharmacogenomics
Rehabilitation Science

67

UCSF is chosen for this evaluation study because the institution is both a representative
and ideal model (based on its metrics discussed below) of the institutions under this study. UCSF
is classified as a very high research activity doctoral-granting university or R1 institution by the
Carnegie Classification (Carnegie Classification of Institutions of Higher Education by Indiana
University Center for Postsecondary Research, 2017).
On several different metrics for basic and biomedical sciences PhD students, UCSF is a
model institution. Of the PhD students that are not foreign citizens, about 22% are URM (UCSF
Graduate Division, 2020a). This percentage is much higher than the national average of 7.9%
(NSF, 2017a). Female students made up nearly 56% of the student body population (UCSF
Graduate Division, 2020a) as compared to the national average of 47% (NSF, 2017d). In the all-
important PhD completion, the rate is 89.2% of all students and 84.6% of all URMs. This
compares very favorably against national attrition rates in science and engineering PhD programs
of more than 50% for certain fields of study (Council of Graduate Schools, 2020) and two-thirds
amongst Black students (Joseph, 2012). Another national study had 36% of African American
and 40% of Hispanic/Latino students completing their PhD programs in the life sciences,
engineering, and physical and mathematical sciences fields 7 years into their program (Sowell et
al., 2008). URM make up a little more than 9% of the faculty (UCSF Office of Diversity and
Outreach, 2020c) which is only slightly better than the national average of 8.6% (NSF, 2018b),
but URM make up more than one-fifth of the staff members (UCSF Office of Diversity and
Outreach, 2020c).
Nearly half of all URM doctoral students in the nation are first-generation students
(Rudolph et al., 2019). First-generation students are less likely than their continuing-generation
students to have obtained their undergraduate degree from a high (classified as R2) or very high
68

research activity doctoral-granting university (University of Chicago, 2008). R1 institutions
award the most science and engineering doctorate degrees (London et al., 2014). This implies
that when most first-generations arrive at R1 institutions to do their doctoral program, they lack
the experience and knowledge of the “unspoken” rules found in these institutions (Roksa et al.,
2018). The transition to a high or very high research activity doctoral university for graduate
school can be difficult experience for students in general and especially for African Americans
who feel a cultural shock when they attend graduate school where the academic environment and
campus climate is negatively different from the minority-serving institutions (MSI) or
historically Black colleges and universities where the majority obtained their undergraduate
degrees (Brown & Davis, 2001; Joseph, 2012; NSF, 2019a). As UCSF is an R1 university, we
can study and evaluate these transition problems URM students face when going from their
undergraduate to a doctoral program that was found in the review of literature in Chapter Two.
Also, UCSF is unique in that there are no undergraduate degrees offered at this institution, so all
incoming doctoral students would be arriving new from a different undergraduate institution,
which puts additional relevancy of using UCSF to evaluate the transitional experience of
students arriving from different institutions to pursue their STEM PhD degrees.
UCSF consistently ranks as one of the nation’s top medical schools, top hospitals, and top
medical research centers (Harder, 2019). UCSF sees about 45,000 hospital admissions and 1.7
million outpatient visits annually (UCSF, 2018) and is the top public recipient of funding from
the National Institutes of Health (NIH) (Alvarez, 2020). In 2019, UCSF was awarded nearly
1,300 NIH grants and contracts totaling over $684.4 million in funding. NIH grants add prestige
to any medical and research institution and UCSF received the top NIH funding for any public
institution and second overall (UCSF, 2020a). The aforementioned Graduate Division doctoral
69

programs are all housed within the four of the professional schools (medicine, nursing, pharmacy
and dentistry) and Global Health Sciences (UCSF Graduate Division, 2020c). According to the
2021 US News and World Report, the School of Medicine is in the top six nationally, the School
of Pharmacy's program is ranked second nationally, and five of the School of Nursing’s
specialties ranked in the top six (UCSF, 2020b). There is no ranking for dentistry in the 2021 US
News and World Report (UCSF, 2020b). The size and reputation of UCSF makes it an economic
powerhouse and a world-class research institution that draws many applicants from around the
world to its doctoral programs.
UCSF’s admission process is very selective as only 10.9% of all applicants are admitted
to the basic and biomedical sciences PhD programs. GRE and undergraduate GPA are the two
important objective criteria used by most universities to evaluate applicants for admission into
their STEM programs (Hall et al., 2017; Miller et al., 2019; Wilson et al., 2019). Given that the
average URM applicants have lower GRE scores (Wilson et al., 2019) and undergraduate GPA
(Miller et al., 2019) as compared to their white and Asian counterparts, and that UCSF has nearly
three times the national average of URM students in the basic and medical sciences PhD
programs, these programs are using an admissions review process that goes beyond the
traditional review of academics and relies on a more holistic, comprehensive review of
applicants; for example, including leadership, socioemotional skills, and life experience in the
review process. The basic and biomedical sciences PhD programs using a holistic review similar
to the pioneering and successful bridge to doctoral programs (Rudolph et al., 2019; Stassun et al.,
2011; Stassun et al., 2018) are another reason for choosing UCSF as the setting to address the
research questions.
70

When URM students decide to pursue educational pathways into STEM fields, there is a
strong link between their career and helping their communities (Guy & Boards, 2019). URM
students come from cultures that emphasizes prosocial values (Jackson et al., 2016) and seek
education or jobs that would lead to altruistic opportunities to help others and give back to their
communities (Jackson et al., 2016; Thoman et al., 2014). URM STEM undergraduates place
greater importance on working for social changes than non-URM STEM undergraduates (McGee
et al., 2016). This sense of helping the community is carried by URM students to the doctoral
levels. URM doctoral students are more likely to persist in their education when there is a link
with prosocial values (Jackson et al., 2016) and put greater interest in lab activities and science
career when it can lead to fulfilling prosocial goals of helping others and the community
(Thoman et al., 2015).
Zuckerberg San Francisco General Hospital and Trauma Center (ZSFG) is San
Francisco’s largest public hospital that focuses on providing care to all regardless of the ability to
pay including the most vulnerable population in the city (SFDPH, 2020). UCSF and ZSFG have
been partners in public health since 1872 (SFDPH, 2020). Faculty from all the professional
schools at UCSF provide patient care, conduct research, and teach at ZSFG (UCSF, 2020c). In
research, there are more than 280 UCSF researchers working in 20 UCSF research centers and
laboratories based at ZSFG with a budget exceeding $200 million every year (UCSF, 2020c).
UCSF researchers at ZSFG have been involved in many community public health breakthroughs
including HIV/AIDS, the effects of smoking and the treatment of tuberculosis (SFDPH, 2020).
San Francisco, the city that UCSF and ZSFG serves, is a minority-majority city (World
Population Review, 2020) meaning that non-Hispanic whites make up less than half of the city’s
population and 81% of ZSFG’s patient population are minorities (SFDPH, 2017). UCSF is
71

involved in community outreach programs to increase the pool of K-12 and undergraduate
students from diverse backgrounds including the Regional Collaboration to Strengthen and
Expand STEM (RECESS) (UCSF Center for Community Engagement, 2020). Given UCSF and
ZSFG’s huge monetary and resource commitment to increasing student diversity in STEM
programs and providing extensive public health services to a very diverse minority community
coupled with the importance of fulfilling prosocial goals for pursuing and persisting in a science
doctoral STEM, it is important to consider in the study if that is one of main factors for the high
number of URM students pursuing their STEM PhD degree at UCSF along with an extremely
high program completion rate.
Two other factors make UCSF an ideal setting for this study. One draw for URM students
to UCSF might be that it is a public university with a minority-majority student and staff
population (UCSF Office of Diversity and Outreach, 2020c). Most URM students earn their
undergraduate degrees in public universities (NSF, 2017d) while white students make up 87% of
all undergraduates at private universities (Brown, 2016). The majority of African American
students obtained their undergraduate degree in minority-serving institutions (MSI) or
historically Black colleges and universities (HBCU) (Brown & Davis, 2001; Joseph, 2012; NSF,
2019a) where the student body is more diverse. The graduate academic culture in PWIs for URM
students doesn’t have much personal involvement, is filled with ambiguity and lack of advice
and guidance (Jackson et al., 2016). African American males pursuing graduate degrees in the
engineering fields at research universities showed common themes of isolation, unwelcoming
environments, and lack of camaraderie (Burt et al., 2018) that is similarly echoed by Latino PhD
students describing environments that are lonely, isolating and alienating (Arocho, 2017). At
UCSF, whites make up only 40% of all postdoc, student, and trainee population and 37% of the
72

staff population. Similarly, UCSF is located in a dense, urban environment that is a minority-
majority city (World Population Review, 2020) which may be a further draw for URM students
seeking a diverse environment both on and off-campus.
The Researcher
The researcher works as the IT Director in Student Information Systems (SIS) at UCSF.
SIS is one of the twenty office and functional units within Student Academic Affairs (SAA). As
an IT unit, SIS staff doesn’t serve or work directly with any students in the basic or biomedical
sciences PhD programs, though other SAA units do serve and work directly with students
including Office of the Registrar, Student Health and Counseling Services, and Student Financial
Services (UCSF SAA, 2020b). The responsibility of SIS is to maintain student data and records
and the applications that interface between users (staff, faculty, and student) and the student data
and records. As such, the researcher doesn’t have any positionality and relationship with any of
the participants in the study. The researcher has worked at UCSF for 15 years and in SIS for the
entire duration of that time. The 15 years working with staff and managers in offices and units
that do support and serve students directly means the researcher is able to understand the services
that are provided to graduate students to help them persist and graduate from their programs. At
the same time, the years working at UCSF might provide the researcher with unconscious bias in
the study that views UCSF favorably that a researcher with no affiliation might not have.
The researcher has been in IT for more than 23 years and has an undergraduate degree in
computer science and graduate degrees in business and engineering. Furthermore, the researcher
is a candidate for the Doctor of Education degree (EdD). The researcher’s strong technical
experience and educational background provide a good foundational background for him to
better understand and relate to the difficulties and barriers that URM students experience as they
73

proceed through the STEM pipeline from high school students to undergraduates to doctoral
students and into post-graduate careers. As the researcher has faced similar challenges and
barriers in STEM education and yet has been able to obtain a solid STEM education and a
successful IT career afterward, the researcher can form biases as to why URM students are not
able to complete their STEM doctoral degrees, least of all completing their STEM undergraduate
degrees. At the same time, the challenges the researcher encountered as a first-generation college
student and immigrant background can also mean more sympathetic biases towards URM
students with similar backgrounds.
Another bias arising from the researcher’s strong technical experience and STEM
educational background is a heavy reliance on collecting metrics and using quantitative analysis
to look at problems and come up with recommendations when analytical analysis or non-metric
data is more appropriate. After having spent several months going through more than one
hundred articles as part of the review of literature in Chapter Two, the researcher has also formed
opinions and views of the challenges and barriers for increasing URM students in the STEM
PhD programs. This implies that the researcher can have biases when developing questions to
ask the participants for the survey.
Researchers’ assumptions and biases are important to address. The bias can impact the
validity and reliability of the data or lead to misinterpretations of the data (Smith & Noble,
2014). The researcher “is the primary instrument for data collection and analysis” (Merriam &
Tisdell, 2015, p. 16). The researcher’s own personal experience and training will influence their
choice of approach (Creswell, 2009). In extreme cases, a researcher’s bias can cause a perceived
association of the data collected in the direct opposite of the true association (Pannucci &
Wilkins, 2010).
74

Bias exists in all research and across all stages of the research process: design bias,
selection/participant bias, data collection and measurement bias, analysis bias, and public bias
(Pannucci & Wilkins, 2010). In order to mitigate potential assumptions and biases, it is important
for the researcher to identify and monitor their biases or subjectivities in the theoretical
framework and the research’s personal interests. The research needs to make clear how their
biases and subjectivities might shape the collection and interpretation of the data in the study
(Merriam & Tisdell, 2015). According to Creswell (2009), as quantitative studies have been
traditionally the mode of research, there are highly systematic procedures, rules, and guidance
that research using this methodology can follow. Finally, standardized protocols for data
collection can help further eliminate biases (Pannucci & Wilkins, 2010).
It is the role of the researcher to clearly articulate the rationale for choosing the research
design for the study to minimize any biases. The researcher also needs to have “a well-designed
research protocol explicitly outlining data collection and analysis” to reduce biases (Smith &
Noble, 2014, p. 100). The dissertation committee and ethics committees like the Office for the
Protection of Research Subjects need to consider whether both the research design and
methodological approaches are biased and if suitable to address the problem that is being studied
and explored (Smith & Noble, 2014). Random sampling of participants in the study, discussed in
the “Participants” section below, can further reduce biases.
Data Sources
Survey
An online survey will be created from Qualtrics© and sent out to the participants. The
survey will consist of several broad categories of questions: student demographics, student
program data, student educational background, KMO influences (covered in the review of the
75

literature and summarized at the end of Chapter Two) and comments and feedback that the
survey questions might not have covered. Questions will be both open-ended and closed-ended.
Participants
Participants in the study will need to meet the following three criteria:
1. A current student at UCSF at the time of the online survey. Current student is defined
as having a registration status of “registered” or “expected to registered status” for the
current term.
2. Student is enrolled in any of the fourteen basic and biomedical sciences PhD
programs listed in Table 5.
3. Student whose racial or ethnic makeup is from one or more of the following:
o African American/Black
o Hispanic/Latinx
o Native American/Alaskan Native
The recruitment approach will be to work with the Graduate Division and Office of
Diversity and Outreach along with URM focused student groups (e.g., Black Student Health
Alliance and Chicanx Latinx Campus Association) at UCSF in recruiting participants that meet
the above criteria to participate in the study.
Out of the 736 students currently enrolled in the fourteen basic and biomedical sciences
PhD programs, 672 are either U.S. citizens or permanent residents. Out of the latter case, 22% or
148 students are URM. In an ideal situation, all 148 URM students would be interested in
participating in this study, but that might not materialize. Depending on the number of actual
URM participants for the study, the sampling approach can end up being either random sampling
or census for the online survey.
76

Before doing an online survey for the quantitative part of this study, it is important to
account for the population size, confidence level and margin of error to determine the sample
size that would be needed. The population size is the total number of people in the target
population in this study. In our case, the population size is 148 (Rumsey 2019). The confidence
level, expressed in percentage means if the online survey is repeated over and over, this is the
percentage of results that one would get from the population (Rumsey 2019). A researcher can
set a confidence level of 0% to 100% though it is usually set to 90% or 95% (Sauro et al., 2016).
The margin of error is the percentage of the results of the survey that will be different from the
real population value (Sauro et al., 2016).
The total number of URM students currently in the basic and biomedical sciences PhD
programs at UCSF is 148 students. This is the population size in our study. Given how busy PhD
doctoral students are (Smolarek, 2019) and a recent student services survey sent out to all
enrolled students only yielded a 31% participation rate (UCSF SAA, 2020a), the probability of
getting a participation rate above 1/3 or 49 participants will be low. Given this expected low
participation rate, the target sample size for this study is N = 47 which is large enough to target
90% confidence level and +/- 10% margin of error. This implies there is a 10% chance the result
happened by accident and the level of certainty is 90%. The higher confidence or lower margin
of error will require higher sample sizes that are probably not feasible to meet.
In random sampling, each student in the URM student population has an equal
probability of being selected (Creswell, 2009). In an ideal environment where all 148 URM
students are willing to participate in the online survey, random sampling could be used to select
the 47 needed participants. In our study, the nonprobability sample (or convenience sample) of
77

choosing participants based on their convenience and availability will be used (Creswell, 2009) if
this study gets less than 47 participants for the online survey.
Instrumentation
This study will use a quantitative methodology, in the form of a Qualtrics© survey, to
address the two research questions. The survey is online and self-administered. The survey
consisted of 86 questions of which 74 were Likert-type questions and 12 open ended questions.
All responses were collected digitally and online. Participants were reached by e-mail or through
word of mouth about the survey. The participants were informed the survey was optional, could
skip any question in the survey, and could quit the survey anytime. In order to create a secure
and safe environment where participants could have a higher level of engagement with honest
opinions and personal thoughts knowing their responses were confidential, the participants were
informed the survey was completely anonymous, no online user tracking was used, no personally
identifiable information (PII) was collected, and the data collected would be permanently deleted
at the end of the study.
Surveys are used to capture information that might not be available through existing data
sources. The questions in a survey are consistently administered. This implies the data collected
from the survey is reliable and repeatable. Data reliability means statistics generated from the
data is also a reliable resource with the confidence interval needed to do decision-making
processes or suggest recommendations. The repeatable factors can be used to measure the
outcomes or success of any decision or recommendation made. Furthermore, the repeatability
allows this study to be expanded to other R1 institutions to increase validity. There are several
compiling reasons for using Qualtrics© as the survey tool in this study. Participants do not need
to report to a specific location to take the survey and can do it in the comfort of their home. Data
78

can be collected quickly and with minimal effort and cost. Participants can take comfort in
knowing that their data will remain anonymous to ensure confidentiality and a safe environment
to honestly answer the questions being asked of them.
The survey will contain both close-ended and open-ended questions (see Appendix A –
Survey Questions). Open-ended questions are included because it presents the researcher a
unique opportunity to discover new and unexpected responses that will not be found in rigid, and
structured closed-ended questions (Robinson & Leonard, 2018). However, the open-ended
questions will be limited in order to minimize participation fatigue that can result in no responses
or minimal or poor response to the questions. Furthermore, as much of the open-ended questions
as possible will be made optional. The survey will consist of the following broad categories of
questions: student demographics, student program data, student educational background, KMO
influences (covered in the review of the literature and summarized at the end of Chapter Two),
and comments and feedback that the questions might not have covered or asked.
Data Collection Procedures
The researcher will put the questions into a Qualtrics© survey. Before the survey can be
sent out, the questions will first need to get approved by the Office for the Protection of Research
Subjects at USC by having the researcher submit an IRB application. In parallel with the
submission and awaiting for the IRB approval, the researcher will work with the Graduate
Division and Office of Diversity and Outreach along with URM focused student groups (e.g.,
Black Student Health Alliance and Chicanx Latinx Campus Association) at UCSF in recruiting
participants that meet the above criteria to participate in the study.
Once the IRB application has been approved and e-mail information for the participants
is obtained, an individualized, unique link of the Qualtrics© survey for this study will be sent to
79

each participant’s e-mail address. The online survey is expected to take one hour to complete. As
this is an online survey, there is no specific location and participants can take the survey
anywhere. All respondent information from the online survey will be captured in Qualtrics©.
The researcher will have access to this survey data by logging into Qualtrics. The researcher will
allow six weeks for the participants to complete the hour-long survey. Every two weeks, the
research will review the survey submission data and send out reminders to those that haven’t
responded.
Data Analysis
A statistical analysis will be done on the submitted survey data using a tool such as IBM
SPSS Statistics or similar tools. These statistical tools help the researcher better understand and
gather insights into the data that was collected through advanced statistical procedures. This
results in high accuracy and quality decision making (IBM, 2020). The researcher will address
the response bias, should it arise in the survey results. Response bias is looking at the effect of
nonresponses from the participants on the survey estimates (Creswell, 2009). For example, if the
nonresponses were replaced with responses, how would the overall results change? The close-
ended, multiple-choice questions in the survey will use either the Likert 5-point or 6-point scale
(Wuensch, 2015). In order to do the statistical analysis, a zero value will be put in for all non-
responses to close-ended questions. The researcher will also identify all independent and
dependent variables in the study as part of the descriptive analysis and include means, standard
deviations, and score ranges for these variables (Creswell, 2009). The researcher will calculate
the average response for each close-ended question in the survey (Johnson & Christensen, 2015).
Any conflicting data, should this situation ever arise, will be excluded from the statistical
analysis (Salkind, 2017).
80

Validity and Reliability
In quantitative research, such as this one, there is a need to generalize the results of the
study to beyond the limits of the study itself. In order to do so, it is important that the results of
the study be reliable, and inferences made from the results to be valid (Johnson & Christensen,
2015).
Research reliability means that when the study is conducted again, the same results will
be obtained (Johnson & Christensen, 2015). The survey that will be used for this study is online
and self-administered using Qualtrics©. As this is an online survey, the questions are
consistently administered to every participant. This implies the data collected from the survey is
repeatable and yields the same results.
One thing that the research needs to be aware of when making inferences from the results
of the study are the possibility of a variable other than the independent variables; this is the
extraneous variable that might influence the dependent variables or hamper the ability of the
researcher to generalize the results (Johnson & Christensen, 2015). It is important for the
researcher designing the study to be aware of extraneous variables and to control or eliminate
their influences on dependent variables (Johnson & Christensen, 2015).
In our quantitative research, there are four major types of validity that the researcher
needs to use to validate the inference from the results of the study – internal, external, construct,
and statistical conclusion (Johnson & Christensen, 2015; Scandura & Williams, 2000; Taylor,
2013):
1. Internal Validity is establishing trustworthy evidence of cause and effect.
2. External Validity is the extent the results from the study from the sample size can be
generalized to the larger population.
81

3. Statistical conclusion validity is inferring the strength of the relationship between two
variables. It is the ability to draw conclusions on two variables based on statistical
evidence.
4. Construct validity refers to what extent does the study actually test the hypothesis or
theory the researcher is trying to study, in other words, how well does the measures
being used for the study fit the hypothesis or theory for which a study is designed.
The researcher will implement strategies recommended by Merriam and Tisdell (2016) to
further promote validity and reliability:
● Employ a triangulation strategy by existing sources of data, if available, to confirm
the emerging findings from the study.
● Awareness and critical self-reflection on the part of the researcher regarding his
assumptions, worldview, biases, and theoretical orientation.
● Create an audit with notes of a detailed account of the methods, procedures, and
decision points of the study.
● Purposely try to create a diversity and variation in the sample population used in the
survey.
● Rich description of the study to help readers of this study. The end goal of this is to
see if the findings for this study can be transferred. (p. 259)
The total number of URM students currently in the basic and biomedical sciences PhD
programs at UCSF is 148 which is our total population size. Using a confidence level of 90%
which is one of two frequently used confidence levels (Sauro et al., 2016) and a +/- 10% means
we need to have a sample size of 47 students or almost a third of the total population. Given a
previous important survey for enrolled students at UCSF only yielded a 31% participation
82

(UCSF SAA, 2020a), meeting the 1/3 participation rate would not be easy. Therefore, a
recruitment strategy needs to be in place to get a higher participation rate. The recruitment
strategy will include communicating the purpose of this study to the targeted population and
working closely with the campus offices that directly work with URM students and registered
campus student organizations focusing on URM students (e.g., Black Student Health Alliance
and Chicanx Latinx Campus Association). If the number of participants is above 47 (N > 47),
then random sampling can be applied to pick only 47 participants, otherwise, all participants will
be asked to do the survey. In order to attempt to increase the number of participants, the
researcher will frame this research study to potential participants on the use of this survey result,
opportunities to respond is limited and to use rewards for participating in the study (Robinson &
Leonard, 2018).
To maximize the response rate, it is important to keep the number of questions in the
survey minimal so that no more than one hour is needed to complete the survey. The open-ended
questions in the survey will be limited in order to minimize participation fatigue that can result in
no responses or minimal or poor response to the questions. Also, the open-ended questions will
be made optional as much as possible. Furthermore, the researcher will reduce the length and
complexity of the survey as well as avoid collecting sensitive or threatening information
(Robinson & Leonard, 2018).
Ethics
It is necessary and important for the researcher to address ethical issues as the research
study involves human subjects. The protection of human subjects from harm, right to privacy,
notion of informed consent and issue of deception needs to be considered before and during the
study (Merriam & Tisdell, 2016).
83

According to Creswell (2009), ethical issues in a research study will need to be
considered during five different phases – prior to conducting the study, beginning the study,
collecting data, analyzing data, and reporting/sharing/storing data.
Research efforts involving humans in a study are subject to legal and ethical guidelines
that are set by an institutional review board IRB0 (Robinson & Leonard, 2018). IRB guidelines
dictate how participants in a study are informed of any risks from their participation and be
provided a written consent or opportunity not to participate in the study (Robinson & Leonard,
2018). IRB committees exist because of federal laws and regulations for protecting human rights
(Creswell, 2009). Prior to conducting the study, a detailed proposal of the research study will be
submitted to USC IRB for review and approval before initiating the data collection part,
consisting of e-mailing out the Qualtrics© survey link to participants. This is to ensure that legal
and ethical considerations have been thoroughly addressed and vetted. One must also note that
even with an IRB approval, federal guidelines and regulations and national professional
associations published standards or codes of ethics (Creswell, 2009), the actual ethical practices
are based upon the researcher’s own values and ethics (Merriam & Tisdell, 2016). To ensure the
ethical treatment of human subjects in this study, the researcher will provide all participants with
the following information prior to the study:
• informed consent form for participants to sign
• maintaining their privacy and anonymity
• how Qualtrics© will securely store and keep their submitted responses from the
survey
The informed consent form would include detailed information from Creswell (2009):
• identification of researcher
84

• identification of the sponsoring institution
• identification of the purpose of the study
• identification of the benefits of participating
• identification of the level and type of participant involvement
• notation of risks to the participant
• guarantee of confidentiality to the participant
• assurance that the participant can withdraw at any time
• provision of names of persons to contact if questions arise (p. 96)
As previously described in the “Researcher” section, the researcher works in the same
institution as the student participants in this study. The researcher has an IT director role
responsible for maintaining student data and records and the applications that interface between
users (staff, faculty, and student) and the student data and records. As such, the researcher
doesn’t have any positionality and relationship with any of the participants in the study nor does
the research come into contact or work with any of the participants. The researcher does not have
a vested interest in the outcomes of this study. Prior to the recruitment of participants, the
researcher will contact the “gatekeeper” (Creswell, 2009) to gain permission to study the
participants. The gatekeepers for this study would include administrators and leaders of the
registered student campus organizations.
During the beginning of the study, the researcher needs to convey the purpose of the
study to the participants and specify the sponsorship of the study to establish trust and credibility
(Creswell, 2009). It is important for the researcher to provide participants a complete and honest
picture of the survey’s purpose and contents (Robinson & Leonard, 2018). The relationship a
researcher establishes with participants can facilitate or hinder aspects of the research such as
85

data collection (Maxwell, 2013). The researcher will never force any participants to sign the
informed consent form and explain that participation in the study is voluntary and does not need
to participate in the study. It is important for the researcher to respect and anticipate any cultural,
religious, gender, or other differences in the participants (Creswell, 2009).
During the data collecting phase, the researcher needs to let the participants know they
are actively participating in a research study by providing instructions in the survey to
participants about the purpose of the study (Creswell, 2009). As participants are spending time
filling out and submitting the survey, reciprocity to the participants will be established in the
form of a small reward for participating. The research will avoid any questions in the survey that
is harmful information as the “code of ethics for researchers is to protect the privacy of the
participants and to convey this protection to all individuals involved in a study” (Creswell, 2009,
p. 99).
In the data analysis part, Creswell (2009) warns that it is easy for the researcher to
support and embrace the perspectives and responses of participants. As the researcher is the one
who ultimately decides what is important to analyze, opportunities exist for excluding data that
contradicts the researcher’s views (Merriam & Tisdell, 2016). The researcher will take these into
consideration and carefully review all data points and not disregard any data whether it supports
or goes against his personal view or hypothesis about this study. Creswell (2009) also states that
the research needs to disclose all the findings and not just positive results. When data is
disclosed, the researcher will dissociate any participants’ personally identifiable information with
their responses. Data that is collected will be kept by the researcher for 5 years. The information
will be encrypted and stored on a secure Cloud with only authorization and authentication from
the researcher’s own account.
86

Chapter Four: Findings
In this chapter, the researcher presents a discussion of the findings organized by
knowledge, motivation, and organizational influences identified in the conceptual framework from
Chapter Three. The focus of this evaluation study is on UCSF and the data collection is done in
the form of a Qualtrics© survey. UCSF was chosen for this evaluation study because the
percentage of URM students enrolled in its basic and biomedical sciences PhD programs is nearly
three times the national average. UCSF still needs to bring more URM students into its PhD
programs, but if other research universities can increase their URM students to the same
percentage, it will go a long way towards bringing more diversity, equity, and inclusion into the
nation’s STEM PhD programs. Two research questions that align with the Clark and Estes’ (2008)
gap analysis framework will guide this study:
1. What are URM students’ knowledge, motivation, and organizational influences to
achieve acceptance and persist in their STEM PhD programs at UCSF?
2. What are the recommendations, for knowledge, motivation, and organizational
resources, for UCSF?
Participating Stakeholders
In order to participate in the survey, the participant needed to meet all three of the
following criteria:
1. A current student at UCSF at the time of the online survey. Current student is defined
as having a registration status of “registered” or “expected to registered status” for the
current term.
2. Student is enrolled in any of the fourteen basic and biomedical sciences PhD
programs listed in Table 5.
87

3. Student whose racial or ethnic makeup is from one or more of the following:
o African American/Black
o Hispanic/Latino
o Native American/Alaskan Native
One hundred forty-eight students met all three criteria and 33 of these students participated
in the survey. This translated to a participation rate of 22.3% for the one hundred forty-eight
students. The participation rate is not surprising given that a previous important survey, concerning
student services for all enrolled students at UCSF only yielded a participation rate that was a bit
more (UCSF SAA, 2020a) and PhD students are generally very busy (Smolarek, 2019).
The gender and ethnic break of the 33 student respondents are shown in Tables 6 and 7.

Table 6
Student Respondents' Gender
Gender % Count
Male 33.33% 11
Female 66.67% 22
X 0.00% 0
Total 100.00% 33

88

Table 7
Student Respondents' Ethnicity
Ethnicity % Count
Hispanic/Latino 64.7% 22
African American 35.3% 12
Total 34

Note. There are only 33 respondents, but one identified as both Hispanic/Latino and African
American

Coding and Scaling
Coding was necessary for the open-ended questions in order to better analyze the
responses. The coding was done using a flat coding frame. A flat coding means all codes that are
created for the responses are treated with the same level of importance and specificity. The
advantage of using flat coding is the ease and consistency of using it when manually coding the
responses. The researcher used indicative coding, which involved building a set of codes directly
from the responses and then re-coding as necessary to ensure flexibility and coverage of every
response.
A five-point scale was used for every multiple-choice value in all closed questions of the
survey. Appendix C shows the point that was assigned to each multiple-choice value for each
closed question. Using the rating scale, the frequency of the response was tabulated (see
Appendix E) and the mean, standard deviation, min, max and mode for the aggregated responses
of each closed question was calculated (see Appendix D).
89

Results for Knowledge, Motivation, and Organizational Influences
The analysis of the data is organized around the research questions and correlates to the
Clark and Estes' (2008) KMO gap analysis framework. The KMO influences impacting URM
students in STEM PhD programs are categorized and organized around three main themes:
● declarative, procedural, and metacognitive knowledge
● theories of expectancy-value and self-efficacy for motivation
● organizational changes
There is one section each for the knowledge, motivation, and organizational influences.
Each section contains detailed discussions of the survey data broken down by specific influences.
After the results are presented, the chapter will conclude with whether each KMO influence was
validated, invalidated, or partially validated.
Knowledge Influences Results
Knowledge Deficiencies in First-Generation Students
In this section, the researcher discusses the knowledge deficiencies in first-generation
students as compared to continuing-generation students. First-generation students are students
who are the first in their families to attend college (Roksa et al., 2018). The three knowledge
types being assessed in the quantitative survey are: factual knowledge, procedural knowledge,
and conceptual knowledge. Conceptual knowledge is knowing or understanding principles,
theories, models, concepts, and classifications (Anderson et al., 2001). Procedural knowledge is
the knowledge needed to perform a task and is goal-oriented (Anderson et al., 2001). Factual
knowledge is facts or basic information that one must know to solve a problem or task (Anderson
et al., 2001). This section will address the factual knowledge deficiencies of first-generation
90

students. A later section in this chapter, Barriers for First-Generation Students, will address the
other two knowledge deficiencies of first-generation students.
Students were asked to identify themselves as first-generation students or continuing-
generation students (see Appendix B, Question 15). Nineteen out of the thirty-three respondents
classified themselves as first-generation students as shown in Table 8. This is 57.5% of the total
respondents and corresponds closely to national figures for URM college students (Rudolph et
al., 2019).

Table 8
First or Continuing-Generation Student
Response Count Percent
First-generation 19 57.5%
Continuing-generation 14 42.5%

Total 33 100%

91

Next the students were asked who provided guidance and preparation for applying to
college (see Appendix B, Question 16). This was an open question. After applying a flat coding
to the responses, the possible sources mentioned were parents, myself, teachers/counselors,
relatives, and self-help programs to get students into college. Responses for first-generation and
continuing -generation students are shown in Table 9 and Table 10, respectively.

Table 9
Guidance for Preparing and Applying to College, First-Generation Students
Source of guidance
mentioned
Count Percent
Parents
Myself
Teachers/counselors
Relatives
Friends
Programs for students
2
10
10
2
2
3
10.5%
52.6%
52.6%
10.5%
10.5%
15.8%

92

Table 10

Guidance for Preparing and Applying to College, Continuing-Generation Students

Source of guidance
mentioned
Count Percent
Parents
Myself
Teachers/counselors
12
2
4
85.7%
14.3%
28.6%

Factual knowledge is imperative to students’ academic success when navigating through
the STEM PhD pipeline. The lack of parental experience with preparing, applying, and attending
a college means first-generation students will not have as much of the factual knowledge that
continuing-generation students have to get into and succeed in college (Roksa et al., 2018).
According to the survey data, a majority (52.6%) of first-generation students relied on
themselves or teachers/counselors for providing guidance when preparing and applying for
college. Only 10.5% of first-generation students mentioned their parents. First-generation
students rely more on secondary education teachers and counselors for preparing for STEM
educational pathways (Kong et al., 2013), assuming the students know what or where to seek that
guidance. Without a parent that has gone through the college admissions process, high school
students might not be aware or knowledgeable about the college eligibility requirements
(Rudolph et al., 2019) until it is too late. An example of this would be meeting all the required
math courses for college eligibility to the University of California (UC) or California State
University (CSU) systems (Rudolph et al., 2019). For the 42.5% of the respondents who are
continuing-generation students, an opposite picture of first-generation students’ responses
emerges. Continuing-generation students showed that the vast majority, 85.7%, had mentioned
93

their parents. Only a small percentage of continuing-generation students mentioned themselves
(14.3%) or their teachers/counselors (28.6%).
Knowledge, or a lack of it, may be a deterring factor for URM students not pursuing a
doctoral degree (Moreira et al., 2019). Respondents were asked the importance of being provided
a career, and academic guidance and advice on pursuing a graduate degree (master's, PhD) in
their undergraduate years (5-point scale; 5 = Extremely Important, 1 = Not at all important, see
Appendices B and C, Question 64). The respondents responded that this was extremely
important (N = 33, µ = 4.52, σ = 0.70, min = 3.00, max = 5.00, and mode = 5.00).
Respondents were asked to list the institution where they obtained their bachelor's degree
(see Appendix B, Question 6) and then the researcher used the Carnegie Classification to
determine which of these institutions were R1 universities. The results are shown in Table 11 and
shows that 53.6% of first-generation students didn’t attend an R1 university. This confirms with
the sizable percentage of first-generation students attend minority-serving institutions (MSI)
rather than very high research doctoral-granting (R1) university (Carnegie Classification of
Institutions of Higher Education by Indiana University Center for Postsecondary Research, 2017;
University of Chicago, 2008) found in the literature review.

Table 11
First Generation Student Attending R1 Universities
Response Count Percent
Yes 9 47.4%
No 10 53.6%
Total 19 100%

94

R1 universities award the most science and engineering doctorate degrees (London et al.,
2014). This results in first-generation students not being able to obtain the critical knowledge and
guidance for pursuing a STEM PhD degree due to the lack of resources at MSI institutions
compared to wealthier and more prestigious research-intensive universities. Data collected from
the Texas A&M University System (one of the nation’s largest system of higher education),
reveals that two of the main reasons URM students do not pursue STEM doctoral degrees are not
knowing the career options available for such a doctoral degree and not understanding the
process to reach such a goal (Moreira et al., 2019). Respondents were asked about how much
career and academic guidance and advice pursuing a graduate degree (master's, PhD) was
provided to them in their undergraduate institution (5-point scale; 5 = Very Sufficient, 1 = Not
sufficient, see Appendices B and C, Question 65). On average, the responses (N = 33, µ = 3.42, σ
= 0.85, min = 1.00, max = 4.00, and mode = 4.00) showed that students were getting some
guidance and advice, but not enough to be considered sufficient (4 = Sufficient). If we only look
at the responses for this question from the ten respondents who didn’t attend an R1 university for
their undergraduate degree, their responses (N = 10, µ = 3.30, σ = 0.86, min = 1.00, max = 4.00,
and mode = 3.50) to this question was even slightly lower and thereby the advice and guidance
received was even less sufficient. The analysis of the responses for Question 64 mentioned
earlier highlights the importance the respondents place on getting graduate degree advice and
guidance in their undergraduate years, and the analysis of the responses for Question 65 here
highlights the actual level of guidance first-generation students attending non-R1 universities
received.
95

Understanding STEM Educational Pathways
The previous section (see “Knowledge Deficiencies in First-Generation Students”)
analyzed data to show how the lack of procedural, factual, and conceptual knowledge related to
STEM career and academic guidance for first-generation students in their high school and
undergraduate years is a deterring factor for not pursuing a doctoral degree. This section does
further analysis of survey results for respondents in general (both first-generation and
continuing-generation students) showing from their educational perspective and experiences, the
lack of career and educational guidance contributes to an insufficient understanding of the STEM
educational pathway (to the PhD level). This in turn results in a leaky STEM pipeline for URM
students.
Respondents were asked if the lack of career and educational guidance is a major barrier
for URM students from applying for graduate schools (5-point scale; 5 = Strongly Agree, 1 =
Strongly Disagree, see Appendices B and C, Question 37). The responses showed students
strongly felt the lack of career and educational guidance is a major barrier for URM students
from applying for graduate schools (N = 33, µ = 4.76, σ = 0.43, min = 4.00, max = 5.00, and
mode = 5.00). This negative response could correlate to the respondents’ own experience as only
10.5% of first-generation respondents mentioned receiving parental guidance when applying for
college (see Table 9).
Furthermore, students who come from minority-serving (MSI) institutions rather than
research focused universities, might not have access to counseling and guidance needed to
establish a solid pathway to a STEM PhD degree.
Respondents were asked to list up to three top reasons why the transition into their PhD
program was difficult (see Appendix B, Question 20). This was an open question and 16 or
96

48.5% of the respondents responded with one or more reasons. After applying a flat coding, the
responses for students graduating from a R1 university (N = 6) and students graduating from a
non-R1 university (N = 10) are shown in Table 12 and Table 13, respectively.

Table 12
R1 University Students’ Responses for Question 20
Reason mentioned Count % of Responds
Lack of mentorship
1 16.7%
Less welcoming environment/lack of
community/discrimination
2 33.3%
High cost of living
2 33.3%
Lack of preparation from undergrad
to PhD
1 16.7%
Adjusting to new environment
2 33.3%
Time management
0 0.0%
More difficult/rigorous than
undergraduate
3
50.0%

Less focus on student
1 16.7%

97

Table 13
Non-R1 University Students’ Responses for Question 20
Reason mentioned Count % of Responds
Lack of mentorship 0 0%
Less welcoming environment/lack of
community/discrimination 5 50.0%
High cost of living 2 20.0%
Lack of preparation from undergrad
to PhD 7 70.0%
Adjusting to new environment 6 60.0%
Time management 1 10.0%
More difficult/rigorous than
undergraduate 7 70.0%
Less focus on student 4 40.0%

Seventy percent of non-R1 university graduates felt their undergraduate program
preparation was inadequate for their PhD program or found the PhD program was quite difficult
and rigorous as compared to their undergraduate degree. A master’s degree can be used as a
stepping stone to a STEM PhD as it can help students acquire the necessary knowledge and skills
(Okahana et al., 2018) needed to be successful in STEM doctoral programs such as access to
courses unavailable to them as undergraduates or gain research experiences (Rudolph et al.,
2019). Furthermore, a master’s degree gives students experience taking graduate courses and
provides the educational experience necessary to map out the educational pathway for a PhD
degree. Yet only 6% of the respondents completed a master’s degree prior to starting their PhD
program (see Appendix B, Question 11, and Table 14).

98

Table 14
Students Completed Master’s Degree Prior to Applying to PhD Program
Response Count Percent
Yes 2 6.0%
No 31 94.0%
Total 19 100%

Completing a master’s degree would likely provide a smoother transition for the 70% of
non-R1 university respondents who responded that their undergraduate program preparation was
inadequate for their PhD program or found the PhD program was quite difficult and rigorous as
compared to their undergraduate degree (see Table 13).
Benefits of Student Mentorship
Mentoring positively correlates to self-development activities that led to student
satisfaction, effective communication, and involvement in their programs of study (Holland et
al., 2012). Incoming graduate URM students need to learn to successfully navigate through new
cultures, expectations, and environments, build networks to avoid isolation and strategies for
being successful in graduate school (Moreira et al., 2019). Mentorship helps graduate students
build the skills to focus on academic norms and behaviors associated with their discipline
(Joseph, 2012). Productive and quality mentorship also contributes to increased student efficacy,
academic achievement, productivity in scholarship and persistence (Estrada et al., 2018). Thus,
the knowledge obtained from student mentorship is very important. Respondents were asked
how important mentorship in the undergraduate level is in motivating them to pursue their PhD
degree (5-point scale; 5 = Extremely Important, 1 = Not at all Important, see Appendices B and
C, Question 10). The quantitative data highlights that even at the undergraduate level,
99

respondents felt that mentorship was very important for motivating students to pursue STEM
PhD degrees (N = 33, µ = 3.45, σ = 0.56, min = 2.00, max = 4.00, and mode = 4.00).
Respondents were asked if the lack of mentorships is a barrier for URM students from
applying for graduate schools (5-point scale; 5 = Strongly Agree, 1 = Strongly Disagree, see
Appendices B and C, Question 36). The very strong agreement from respondents to this question
points to the lack of mentorships as indeed being a barrier (N = 33, µ = 4.64, σ = 0.48, min =
4.00, max = 5.00, and mode = 5.00). Mentors can help students gain familiarity and knowledge
of on-campus student support programs (student organizations, career guidance, undergraduate
research, and academic support) and builds professional networks and enhances research skills
(Lisberg & Woods, 2018). The lack of mentorship means that these foundational skills needed
for planning and pursuing graduate education are not available to URM students.
When asked if the lack of underrepresented minority role models is a major demotivation
for URM students from pursuing and persisting in a STEM degree, either bachelor, master's, or
PhD (5-point scale; 5 = Strongly Agree, 1 = Strongly Disagree, see Appendices B and C,
Question 38), respondents strongly agreed that the lack of underrepresented minority role models
is a major demotivation for URM students from pursuing and persisting in a STEM degree (N =
33, µ = 4.45, σ = 0.56, min = 3.00, max = 5.00, and mode = 5.00). Respondents were asked to
list up to three reasons why a mentor that looks like me (i.e., same ethnic and cultural identity) is
important (see Appendix B, Question 35). Twenty-six or 78.8% of all respondents provided at
least one reason. The responses are listed in Table 15.

100

Table 15
Reasons for Having Mentor that Looks Like Me
Reason Mentioned Count % of Responds
Never had a mentor
1 3.8%
Reduces imposter
syndrome/stereotype threat
8 30.8%
More relatable/sense of belonging
14 53.8%
Seen as role model
12 46.2%
More empathy/sympathy
11 42.3%
Think/care more about you
6 23.1%
Share cultural experience
12 46.2%
Use to seek advice
17 65.4%
Gives me confidence
9 34.6%
Think/care more about you
6 23.1%

Respondents felt that underrepresented minority role models in mentorship roles can help
reduce imposter syndrome and stereotype threat, are more relatable and create a sense of
belonging, and more comfortable going to seek advice than non-URM role-models.
The presence of mentorships has many added benefits for students (Estrada et al., 2018).
Students were asked several mentor questions (5-point scale; 5 = Strongly Agree, 1 = Strongly
Disagree, see Appendices B and C, Questions: 30-33). The responses are shown in Table 16.

101

Table 16
Mentorship Questions
Question
N Mean SD Min Max Mode
Mentorship is a key to my academic success.
(Q30) 33 4.85 0.36 4.00 5.00 5.00
Mentorship increases participation and
persistence in my doctoral program. (Q31) 33 4.82 0.39 4.00 5.00 5.00
Mentorship increases my sense of belonging
to the academic environment. (Q32) 33 4.76 0.60 2.00 5.00 5.00
I have sufficient mentorship opportunities in
my PhD program. (Q33) 33 3.61 0.55 3.00 5.00 4.00

The data points to the importance of mentorship to students (see Appendices B and E,
Question 30-31), yet students in general are not given enough mentorship opportunities in their
doctoral program (see Appendices B and E, Question 33).
Master’s Degree As Stepping Stone to Doctorate Degree
In Chapter Three, the review of literature mentioned the importance of completing a
master’s degree prior to pursuing a STEM PhD as it can help URM students acquire the
knowledge and preparation (Okahana et al., 2018) needed to be successful in STEM doctoral
programs. Furthermore, bridge programs allow students in the biomedical sciences to obtain a
master's degree on the way to the PhD program through a partnership between an institution that
offers the master's degree and a research-intensive college or university granting the PhD degrees
in the biomedical sciences. The program’s goal is to develop a diverse pool of PhD scientists in
the biomedical sciences (NIH, 2020). URM STEM PhD students who are in bridge programs
have completion rates that are twice the national average (Moreira et al., 2019).
102

Given the importance and necessity of obtaining a master’s degree prior to a doctorate
degree, the survey data was surprising in that only about 6% of the respondents completed a
master’s degree prior to starting their PhD program (see Table 14). Also given bridge programs’
goal to increase URM PhD students and its high completion rates, only about 27.3% of the
students have heard about the program (see Table 17).

Table 17
Students Heard About Bridge Programs
Response Count Percent
Yes 9 27.3%
No 24 72.7%
Total 33 100%

103

Of those students that heard about the bridge program, their responses of whether the
program can increase underrepresented minorities enrollment in the STEM PhD programs (see
Appendix B, Question 47) was neutral (3 = Neither agree or disagree and with responses N = 9,
µ = 3.44, σ = 0.50, min = 3.00, max = 4.00, and mode = 3.00). The data points to mixed feelings
on the part of the respondents on these bridge programs’ effectiveness in enrolling more
underrepresented minority students.
Motivation Influences Results
Finding Prosocial Values and Identities
One motivation addressed in the review of literature in Chapter Three stresses the desire
“to relate to and care for others” (Deci & Ryan, 1991, p. 243). Coming from cultures with an
emphasis on prosocial values (Jackson et al., 2016), URM students seek educational
opportunities with the altruistic goals of helping others and giving back to their communities
(Jackson et al., 2016; Thoman et al., 2014). URM doctoral students highlight the importance of
prosocial values as one of the main factors for persisting in science education (Jackson et al.,
2016) such as focusing on fields of study and research that address health problems in their
communities (Jackson et al., 2016).
URM students often encounter a lack of prosocial values from the majority of non-URM
peers (Joseph, 2012). Students were asked a series of questions related to prosocial goals within
their STEM PhD program and activities, and with their peers and departments (5-point scale; 5 =
Extremely Important, 1 = Not at all Important, see Appendices B and C, Questions: 46, 61-63).
The responses are shown in Table 18.

104

Table 18
Prosocial Goals
Question
N Mean SD Min Max Mode
How important is aligning or framing STEM
education and research with social justice and
helping the community? (Q46)
33 4.33 0.97 2.00 5.00 5.00
How important is seeking prosocial goals and
opportunities (such as helping communities)
in the classroom, conferences or research of
my STEM PhD program? (Q61)
33 4.24 0.89 1.00 5.00 5.00
I feel that peers do not have the same
enthusiasm for seeking prosocial goals and
opportunities in a STEM PhD program as I
do. (Q62)
33 3.48 0.56 2.00 4.00 4.00
I feel that my department does not have the
same enthusiasm for seeking prosocial goals
and opportunities in a STEM PhD program as
I do. (Q63)
33 3.36 0.92 2.00 5.00 3.00

Comparing the responses from Question 62 and Question 63 shows students feel their
peers more than their departments share their same level of enthusiasm in seeking prosocial goals
and opportunities in a STEM PhD program. Furthermore, the data shows a disconnect between
their prosocial values and their science education that in turn creates a contradiction between
their cultural and science identities (Jackson et al., 2016).
Identifying with a science or scientist identity is a very strong motivator for URM
students. Students have better academic performance and persistence when identifying with a
role identity such as a scientist, than a racial or ethnic identity (Chemers et al., 2011). Science
identity building is one of the characteristics attributed to high-achieving students whose goals
are to pursue graduate education and a career in scientific research (Hurtado et al., 2011). URM
105

students have greater persistence to complete their degrees when identifying with a role identity
(Eccles & Barber, 1999).
When asked if identifying with a science career or scientist identity is a very important
motivator to persist in the PhD program (5-point scale; 5 = Strongly Agree, 1 = Strongly
Disagree, see Appendices B and C, Question 41), students strongly agreed (N = 33, µ = 4.39, σ
= 0.60, min = 3.00, max = 5.00, and mode = 4.00). When asked if pursuing a science career or
scientist identity means losing their racial or cultural identity (5-point scale; 5 = Strongly Agree,
1 = Strongly Disagree, see Appendices B and C, Question 40), the responses disagreed though
there was a large range in the min and max that contributed to a high standard deviation (N = 33,
µ = 2.24, σ = 1.02, min = 1.00, max = 5.00, and mode = 3.00).
Despite respondents feeling a disconnection with their peers and departments regarding
the importance of connecting prosocial values with science education, respondents didn’t feel
that pursuing a science career or scientist identity means losing their racial or cultural identity.
When URM students see a successful faculty with shared ethnic and cultural
backgrounds, it signals to the student that they too can be successful (Yadav & Seals, 2019). At
the same time, it may be difficult for URM students to create a science career or scientist identity
when there is a lack of successful role models like themselves (McGee, 2016). Responses to
whether having underrepresented minority faculty role models allows URM students to
successfully create a scientist identity with role models like themselves that have made it (5-
point scale; 5 = Strongly Agree, 1 = Strongly Disagree, see Appendices B and C, Question 42)
indicated strong agreement (N = 33, µ = 3.94, σ = 0.60, min = 3.00, max = 5.00, and mode =
4.00). On the other hand, when asked if it is difficult to have a scientist identity when there is a
lack of role models that look like them (5-point scale; 5 = Strongly Agree, 1 = Strongly Disagree,
106

see Appendices B and C, Question 39), the responses showed a high standard deviation though
the mean value was very close to the responses for Question 42 (N = 33, µ = 3.97, σ = 1.11, min
= 2.00, max = 5.00, and mode = 5.00).
Self-Efficacy and Motivation
Bandura defines self-efficacy as "beliefs in one’s capabilities to organize and execute the
courses of action required to produce given attainments" (Bandura, 1977, p. 3). A low self-
efficacy person will disengage and withdraw from a task when it becomes difficult or not engage
with the task through delay and avoidance (Klassen et al., 2008). A high self-efficacy person, on
the other hand, is more engaged and expend more effort on a task (Chemers et al., 2001). When a
task becomes difficult or challenging, the high self-efficacy person persists longer and is more
likely to recover and bounce back from any failures (Chemers et al., 2001), thereby the person is
more likely to succeed in his or her task.
Students were asked to rate their self-efficacy through a series of questions (5-point scale;
5 = Strongly Agree, 1 = Strongly Disagree, see Appendices B and C, Questions: 73-78) with
responses shown in Table 19.

107

Table 19
Self-Efficacy Questions
Question
N Mean SD Min Max Mode
Able to achieve most of the goals that I have set
for myself (Q73) 33 4.00 0.60 3.00 5.00 4.00
Can obtain outcomes that are important to me
(Q75) 33 3.94 0.55 3.00 5.00 4.00
Confident that I can perform effectively on
many different tasks (Q78) 33 4.03 0.94 2.00 5.00 4.00
Able to successfully overcome many challenges
(Q77) 33 3.91 0.57 3.00 5.00 4.00
When faced with difficult tasks, was certain to
accomplish them (Q74) 33 3.45 1.21 1.00 5.00 4.00
Succeed at almost any endeavor to which I set
my mind (Q76) 33 3.15 1.35 1.00 5.00 4.00

UCSFs STEM PhD programs are extremely selective programs given that UCSF ranks as
one of the most renowned health universities in the world (U.S. News & World Report, 2021b)
and receives the top NIH funding for any public institution and second overall (UCSF, 2020a). It
is therefore not surprising that in the survey data, respondents rated their self-efficacy very
highly. Table 20 shows the percentage of respondents who rated a 4.0 or higher for questions 73-
78.

108

Table 20
Self-Efficacy Questions, Percentage
Question % rated 4.0 or higher to question
Able to achieve most of the goals that I have set
for myself (Q73) 81.2%
Can obtain outcomes that are important to me
(Q75) 81.2%
Confident that I can perform effectively on many
different tasks (Q78) 81.2%
Able to successfully overcome many challenges
(Q77) 81.2%
When faced with difficult tasks, was certain to
accomplish them (Q74) 72.7%
Succeed at almost any endeavor to which I set
my mind (Q76) 72.7%

Yet, when the respondents were asked how well they did their tasks as compared to
others and if able to perform well when things got tough, their responses shown in Tables 21 and
22 below, had lower scoring than the first set of self-efficacy questions. This is not surprising
given the high caliber of students attending UCSF and that a sizable number of respondents felt
that their PhD program was quite difficult and rigorous as compared to their undergraduate
degree (see Table 12 and Table 13).

109

Table 21
Self-Efficacy Questions Set 2
Question N Mean SD Min Max Mode
As it relates to your PhD program, compared
to other people, I can do most tasks very
well. (Q79)
33 3.39 0.92 2.00 5.00 3.00
As it relates to your PhD program, even when
things are tough, I can perform quite well.
(Q80)
33 3.45 0.92 2.00 5.00 3.00

Table 22
Self-Efficacy Questions Set 2, Percentage
Question % rated 4.0 or higher to question
As it relates to your PhD program, compared to
other people, I can do most tasks very well.
(Q79)
33.0%
As it relates to your PhD program, even when
things are tough, I can perform quite well.
(Q80)
45.5%

110

There is a strong correlation between motivation and educational outcomes (Katz et al.,
2014). Motivation can also serve as a protective buffer for URM students overcoming negative
consequences and difficulties (Katz et al., 2014). The lack of motivation leads to attrition
(Rudolph et al., 2019) which can be as high as two-thirds amongst Black students (Joseph,
2012). The three biggest motivations for students to pursue a PhD degree are career
development, interest in a topic or research, and personal motives (London et al., 2014). There
are four motivational themes related to the experiences of STEM graduate URM students:
mentoring, research, opportunities (through funding and support), and the climate of their
academic environment (Guy & Boards, 2019).
When asked for up to the top three reasons that motivated the respondent to enroll in their
PhD program, the following were the top three reasons (see Appendix B, Question 18):
● Interest in science/desire to learn more
● Career opportunities/increased skills
● Financial stability/job prospect
The top three reasons from the respondents closely match the published results from
London et al, 2014 mentioned above. One of the major motivating factors for pursuing and
persisting in science education and doctoral degrees by URM students were altruistic goals of
helping others and giving back to the community (Jackson et al., 2016; Thoman et al., 2014)
discussed in the previous section. Surprisingly in the survey data, helping their communities or
seeking prosocial goals from their doctoral degree was not ranked in the top three reasons.
One important motivator for PhD students, discussed in the previous section, is identifying with
a science career or scientist identity. Students were asked a series of questions (5-point scale; 5 =
Strongly Agree, 1 = Strongly Disagree, see Appendices B and C, Questions: 38-42) about their
111

science identity and how underrepresented minority role models, or lack of, play into this
identity. The responses to this series of questions are shown in Table 23.

Table 23
Self-Efficacy Questions Set 2
Question N Mean SD Min Max Mode
The lack of underrepresented minority role
models is a major de-motivation for URM
students from pursuing and persisting in a
STEM degree (bachelor, master’s, or PhD).
(Q38)
33 4.45 0.56 3.00 5.00 5.00
It is difficult to have a scientist identity when
there is a lack of role models that look like
me (i.e., same ethnic and cultural identity).
(Q39)
33 3.97 1.11 2.00 5.00 5.00
Pursuing a science career or scientist identity
means losing my racial or cultural identity.
(Q40|)
33 2.24 1.02 1.00 5.00 3.00
Identifying with a science career or scientist
identity is a very important motivator to
persist in the PhD program. (Q41)
33 4.39 0.60 3.00 5.00 4.00
The availability of underrepresented minority
faculty role models allows me to
successfully create a scientist identity with
role models like themselves that have made
it. (Q42)
33 3.94 0.60 3.00 5.00 4.00

112

An analysis of the responses establishes that identifying with a science career or scientist
identity is a very important motivator to persist in the PhD program and that the lack of
underrepresented minority role models is a major demotivation for URM students from pursuing
and persisting in a STEM degree (bachelor, master's, or PhD) or establishing a science identity.
Respondents were asked questions about the importance of learning coping and resilience skills
and learning strategies (5-point scale; 5 = Strongly Agree, 1 = Strongly Disagree, see Appendices
B and C, Questions: 17, 69-72). The responses are shown in Table 24.

Table 24
Coping and Resiliency Skills and Learning Strategies
Question
N Mean SD Min Max Mode
Learning coping skills and resiliency dealing
with social pressures and transition to
college is very important. (Q17)
33 4.36 0.98 2.00 5.00 5.00
How important is acquiring learning strategies
in relation to persisting and succeeding in
your PhD program? (Q69)
33 4.64 0.48 4.00 5.00 5.00
I have sufficient opportunities to acquire
learning strategies either in my current PhD
program or in my undergraduate or master’s
degree. (Q70)
33 3.45 0.74 2.00 4.00 4.00
How important is acquiring study skills in
relation to persisting and succeeding in your
PhD program? (Q71)
33 4.06 1.10 2.00 5.00 5.00
I have sufficient opportunities to acquire study
skills either in my current PhD program or in
my undergraduate or master’s degree. (Q72)
33 3.48 0.93 2.00 5.00 4.00

113

Respondents agreed that learning coping skills and resiliency dealing with social
pressures and transition to college are very important and that acquiring learning strategies in
relation to persisting and succeeding in their PhD program were very important, but a sizable
percentage of respondents lack sufficient opportunities to acquire these skills and strategies as
shown in Table 25.

Table 25
Coping and Resiliency Skills and Learning Strategies, Percentage
Question % rated 3.0 or higher to question
I have sufficient opportunities to acquire learning
strategies either in my current PhD program or
in my undergraduate or master’s degree. (Q70)
39.4%
I have sufficient opportunities to acquire study
skills either in my current PhD program or in
my undergraduate or master’s degree. (Q72)
54.5%

114

Sense of Belonging to Campus
Belonging increases students’ persistence towards degree completion (Hurtado, 1997)
and a lack of belonging can have a harmful effect on students’ mental health and behavior
(Museus et al., 2018). The lack of belonging leads to loneliness and isolation that can demotivate
students in persisting in their doctoral studies (Mantai, 2019). Many doctoral students already
suffer from loneliness (Brown, 2013), and being a URM doctoral student makes it even lonelier
and more problematic (Harmon, 2019) due to multiple factors including widespread negative
stereotypes (Rattan et al., 2018), the lack of URM faculty role models (Jackson et al., 2016), and
the scant number of URM students in STEM PhD programs (SACNAS Harvard Chapter, n.d.).
Having access to campus services is also another important factor for creating a sense of
belonging for students. Respondents were asked questions related to student services (5-point
scale; 5 = Strongly Agree, 1 = Strongly Disagree, see Appendices B and C, Questions: 49-50).
The analysis of the responses is shown in Table 26.

Table 26
Student Services
Question
N Mean SD Min Max Mode
Publicizing and providing on-campus student
support programs and services is very
important for student persistence and
success in the PhD programs. (Q49)
33 4.67 0.47 4.00 5.00 5.00
I have access to student services when I need
it. (Q50)
33 4.21 0.64 2.00 5.00 4.00

115

In the survey data, all respondents agree that publicizing and providing on-campus
student support programs and services is very important for student persistence and success in
the PhD programs. All respondents felt that in their doctoral program, they have access to
student services when they need them.
The lack of belonging leads doctoral students to feelings of loneliness and isolation that
in turn demotivates them from persisting in their doctoral studies (Mantai, 2019). PhD students,
in general, suffer from loneliness (Brown, 2013), but being a URM PhD student makes it even
lonelier and more problematic (Harmon, 2019) due to the only handful of URM students in most
STEM PhD programs (SACNAS Harvard Chapter, n.d.).
Respondents were asked questions related to belonging (5-point scale; 5 = Strongly
Agree, 1 = Strongly Disagree, see Appendices B and C, Questions: 51-52, 58-60). The analysis
of the responses to these questions are shown in Table 27.
116

Table 27
Belonging
Question
N M SD Min Max Mode
I have a strong sense of belonging to the
campus. (Q51)
33 3.27 0.79 2.00 4.00 4.00
I have a strong sense of belonging to my
program. (Q52)
33 3.79 0.41 3.00 4.00 4.00
Do you feel there is loneliness and isolation
being an underrepresented minority student
in a STEM PhD program? (Q58)
33 4.00 0.55 3.00 5.00 4.00
How important is learning to build peer and
faculty networks to avoid isolation and
loneliness in a STEM PhD program? (Q59)
33 4.61 0.49 4.00 5.00 5.00
I have sufficient opportunities to learn to
build peer and faculty networks either in
my current PhD program or in my
undergraduate or master’s degree. (Q60)
33 3.97 0.67 2.00 5.00 4.00

Respondents agreed that there is loneliness and isolation being an underrepresented
minority student in a STEM PhD program and it was important to build peer and faculty
networks and to have more URM students in the STEM PhD programs. Furthermore,
respondents felt they had sufficient opportunities to build peer and faculty networks in their
current PhD program or in their undergraduate or master's degree program. Finally, respondents
felt that was a strong sense of belonging to their program, but the same can’t be said of their
campus.
URM doctoral students stress the importance of prosocial values as one of the main
factors in pursuing and persisting in science education (Jackson et al., 2016). Often, URM
students encounter a lack of prosocial values from the majority of non-URM peers (Joseph,
2012). Respondents were asked questions related to this (5-point scale; 5 = Strongly Agree, 1 =
117

Strongly Disagree, see Appendices B and C, Questions: 62-63). The analysis of the responses to
these questions is shown in Table 28.

Table 28
Belonging and Prosocial Goals
Question
N Mean SD Min Max Mode
I feel that peers do not have the same
enthusiasm for seeking prosocial goals and
opportunities in a STEM PhD program as I
do. (Q62)
33 3.48 0.56 2.00 4.00 4.00
I feel that my department does not have the
same enthusiasm for seeking prosocial
goals and opportunities in a STEM PhD
program as I do. (Q63)
33 3.36 0.92 2.00 5.00 3.00

118

Respondents felt that their peers do not share the same enthusiasm as them in seeking
prosocial goals and opportunities in a STEM PhD program, and it even less so for their
departments. This disconnection between their peers and departments further adds to URM
students’ lack of belonging to their campus.
Belonging is a fundamental human need (Rattan et al., 2018). For URM STEM PhD
students, having a strong science identity creates a sense of belonging and buffers against racial
biases and negative stereotypes (Kim‐Prieto et al., 2013). Furthermore, a strong scientist identity
could positively offset URM students’ lack of belonging to their academic environment due to
the scarcity of URM faculty role models and the different cultural and ethnic identities of their
peers (Jackson et al., 2016). Questions about a scientist's identity were asked (5-point scale; 5 =
Strongly Agree, 1 = Strongly Disagree, see Appendices B and C, Questions: 39-41). The analysis
of the responses to these questions is shown in Table 29.

Table 29
Scientist Identity
Question
N Mean SD Min Max Mode
It is difficult to have a scientist identity when
there is a lack of role models that look like
me (i.e., same ethnic and cultural identity.
(Q39)
33 3.97 1.11 2.00 5.00 5.00
Pursuing a science career or scientist identity
means losing my racial or cultural identity.
(Q40)
33 2.24 1.02 1.00 5.00 3.00
Identifying with a science career or scientist
identity is a very important motivator to
persist in the PhD program. (Q41)
33 4.39 0.60 3.00 5.00 4.00

119

Respondents felt that identifying with a science career or scientist identity is a very
important motivator to persist in the PhD program (Question 41), yet it was difficult to have a
scientist identity when there is a lack of role models that look like them (i.e., same ethnic and
cultural identity) (Question 39). Respondents disagreed that pursuing a science career or scientist
identity means losing their racial or cultural identity (Question 40).
Building Persistence Through Mentorship and Research Opportunities
Providing students with mentorship, even informal mentoring networks (Carter-Sowell,
2016) increases their efficacy and persistence (Estrada et al., 2018). Even at the undergraduate
level, about half of the respondents felt that mentorship was very important for motivating
students to pursue their STEM PhD degree (see Appendix B, Question 10). Estrada (2018)
identified three factors that increase student efficacy and persistence through mentorship:
● Instrumental support which is providing opportunities and resources needed for
students to reach their goals.
● Psychological support which is providing emotional and personal development to
students to enhance their competence, identity, and effectiveness.
● Relationship quality is building feelings of trust, empathy, respect, and connection
between mentor and student. (p. 4)
The survey data correlates to this. Respondents felt that mentorship is a key to their
academic success and increases their participation and persistence (see Table 16, Question 30).
When asked though if they had sufficient mentorship opportunities in their doctoral program,
respondents did not agree (see Table 16, Question 31). This can be an issue as poor or lack of
mentorship are major factors in high attrition rates amongst doctoral students (Moreira et al.,
2019).
120

Research also increases student retention and persistence in STEM programs (Estrada et
al., 2018). The main reason is that research experiences heavily engage students in developing
the thought processes, skills, and relationships critical to successful student academic
performance needed to persist throughout their entire STEM graduate education (Cooper et al.,
2019). Furthermore, the research opportunities help students to think and work like a scientist
(Cooper et al., 2019). Respondents were asked questions related to research (5-point scale; 5 =
Strongly Agree, 1 = Strongly Disagree, see Appendices B and C, Question 23-25). The analysis
of the responses to these questions is shown in Table 30.

Table 30
Research
Question
N Mean SD Min Max Mode
Research opportunities are very important
for me to persist and to succeed in my PhD
program. (Q23)
33 4.79 0.41 4.00 5.00 5.00
There are sufficient research opportunities
given to me in the PhD program. (Q24)
33 4.76 0.43 4.00 5.00 5.00
From my knowledge, underrepresented
minorities PhD students, in general, are
not given enough research opportunities.
(Q25)
33 2.45 1.35 1.00 5.00 1.00

121

All the respondents felt that research opportunities are very important for them to persist
and to succeed in their PhD programs (Question 23). The respondents felt they were given
sufficient research opportunities in their PhD programs (Question 24). Finally, from their own
experience and knowledge, the respondents disagreed that underrepresented minorities PhD
students, in general, are not given enough research opportunities (Question 25).
URM students come from cultures that emphasize prosocial values (Jackson et al., 2016),
and therefore seek educational opportunities to help others and serve their communities (Jackson
et al., 2016; Thoman et al., 2014). When URM students can seek prosocial values from their
research opportunities, such as engaging with their communities, it leads to even more increased
retention and persistence (Malotky et al., 2020). Questions related to this topic were asked to the
students (5-point scale; 5 = Strongly Agree, 1 = Strongly Disagree, see Appendices B and C,
Questions: 46, 61). The analysis of the responses to these questions is shown in Table 31.

Table 31
Seeking Prosocial Goals and Social Justice
Question N Mean SD Min Max Mode
How important is aligning or framing STEM
education and research with social justice
and helping the community? (Q46)
33 4.33 0.97 2.00 5.00 5.00
How important is seeking prosocial goals and
opportunities (such as helping communities)
in the classroom, conferences or research of
my STEM PhD program? (Q61)
33 4.24 0.89 1.00 5.00 5.00

122

Respondents agreed that it is important to align or frame STEM education and research
with social justice and helping the community, and it is also important to seek prosocial goals
and opportunities (such as helping communities) in the classroom, conferences, or research of
their STEM PhD program.
URM Faculty and Mentors
A major factor in the persistence and academic success of URM students is having URM
faculty and mentors (Joseph, 2012; Whittaker & Montgomery, 2012). Studies have shown that
URM students have a preference and perform better when teachers are the same race and gender
(Miller, 2018; Whitfield & Edwards, 2011). One important motivator for PhD students
previously discussed (Chapter Four, Finding Prosocial Values and Identities) is identifying with
a science career or scientist identity. It is difficult for URM students to envision or create a
science career or scientist identity when there are no role models like themselves that have ‘made
it’ (McGee, 2016). Differences in racial, cultural, and ethnic backgrounds can also lead to
difficulties and challenges with communication (McGee, 2016). Mentorship contributes to
increased student efficacy, academic achievement, and persistence (Estrada et al., 2018) and
URM students have a strong desire for mentors that are like them (Whitfield & Edwards, 2011).
Furthermore, URM faculty who serve as mentors can draw from their own racial encounters and
experiences to counsel URM students who face microaggressions and racial stereotyping in
hostile campus environments (Sánchez et al., 2018).
When respondents were asked if having a mentor that looks like them was important (5-
point scale; 5 = Strongly Agree, 1 = Strongly Disagree, see Appendices B and C, Question 34),
the responses were generally in agreement (N = 33, µ = 4.03, σ = 0.80, min =3.00, max = 5.00,
and mode = 4.00). When asked why that was the case, the three most frequently cited reasons
123

were: more relatable/sense of belonging, more empathy/sympathy, and going to seek advice (see
Table 15). Less frequent reasons included alleviating imposter syndrome/stereotype threat,
thinking/caring more about them, shared cultural experience, and being seen as a role model.
Organizational Influences Results
Barriers for First-Generation Students
First-generation students’ lack of parental experience with preparing, applying, and
attending college (addressed in Chapter Four, Knowledge Deficiencies in First-Generation
Students), along with these students not knowing what or where to seek academic guidance in
their high schools or outside programs leads to major organizational barriers to getting accepted
into more prestigious and research-intensive institutions.
These factors result in a sizable percentage of first-generation students being accepted
into minority-serving institutions (MSI) rather than more prestigious universities. None of the
respondents who are first-generation students attended Ivy Leagues colleges and only 5.2%
attended a top ten ranked national university (U.S. News & World Report, 2021a) as compared
to 28.5% and 35.7%, respectively, for continuing-generation students. If the top ten ranked
national liberal arts colleges (U.S. News & World Report, 2021a) are included, then about half of
all continuing-generation students attended either a national liberal arts college or a national
university ranked in the top ten in the nation (U.S. News & World Report, 2021a). This
percentage is nearly ten times higher than first-generation students.
Only 42.1% of respondents who are first-generation students graduated from an R1
university (very high research doctoral-granting university) as compared to 85.7% for
continuing-generation students. Not obtaining an undergraduate degree from an R1 university
creates a disconnection between the institution where they are starting their doctoral program and
124

the institution where they obtained their undergraduate degree. This disconnection leads to a gap
in the students’ procedural knowledge as this knowledge is critical for the goal attainment of
completing a STEM PhD The disconnection results in students having a more difficult time
navigating through their doctoral programs as they are unfamiliar with social and cultural norms
or “unspoken” rules (Roksa et al., 2018) in high research-intensive universities where most of the
STEM PhD degrees are granted (Buswell, 2017; Roksa et al., 2018). Furthermore, completing an
undergraduate degree in an R1 university exposes the student to higher academic standards and
rigors, more experience with research, and access to faculty coming from a strong research
background than students attending a lower tier ranking university (Hollman et al., 2018).
Conceptual knowledge is essential for any successful STEM PhD student (Fan & Yu,
2017). Yet, from an organizational perspective, non-R1 universities do not adequately prepare
undergraduate students with the conceptual knowledge necessary for dealing with the academic
rigors of a STEM PhD program that R1 universities do. When asked if their undergraduate
degree institution provided them with the adequate skills and foundations needed to persist and
succeed in their STEM PhD program (see Appendix B, Question 8), 89.5% of the respondents
who attended an R1 university responded “Definitely” and “Very Probably”, compared to only
28.6% for those that attended a non-R1 university. Furthermore, of those that attended a non-R1
university, about 15% responded “Definitely Not” to the same question. The survey responses
indicated 70% of non-R1 university graduates felt that they came to the STEM PhD program
from their undergraduate program inadequately prepared or felt the PhD program was quite
difficult and rigorous compared to their undergraduate degree (Table 13). This contrasts to only
16.7% of those attending an R1 university feeling the same way (Table 12).
125

Admission Biases in Doctoral Programs
Faculty members at elite institutions tend to favor PhD applicants who come from similar
institutions (Posselt, 2018). This is not surprising given that institutions classified as “very high
research activity” (R1 in the Carnegie Classification) award the most science and engineering
doctorate degrees (London et al., 2014). Furthermore, Clauset et al. (2015) showed in their
analysis of 19,000 regular faculty in three disparate disciplines, about a quarter of the institutions
produced 71% to 86% of all tenure-track faculty. Furthermore, 70% to 90% of the faculty at
these elite institutions obtained their doctorates from other elite institutions. Another
disadvantage is that first-generation students, of whom half are URM students (Rudolph et al.,
2019), are much less likely than their continuing-generation students to have obtained their
undergraduate degree from a high research activity university (R2 in the Carnegie Classification)
or very high research activity university (R1 in the Carnegie Classification) (Rudolph et al.,
2019). At UCSF, the admission process is more equitable for applicants coming from non-elite
institutions. Although two-thirds of the applicants came from R1 (“very high research activity”
institutions in the Carnegie Classification) or R2 (“high research activity” institutions in the
Carnegie Classification), only one-third of the applicants came from the top thirty national
universities as ranked by US News and World Report (U.S. News & World Report, 2019).
URM applicants are more likely to apply to competitive doctoral programs when the
admissions process is transparent and a fuller picture of admissions data is available
(Christophers & Gotian, 2019). The lack of full admissions data creates a very misleading picture
for potential applicants especially first-generation students coming from MSIs or HBCUs
institutions who are already hindered with many disadvantages compared to their elite institution
peers (see Chapter Four, Knowledge Deficiencies in First-Generation Students and Chapter Four,
126

Barriers for First-Generation Students). In the very selective and competitive MD-PhD
programs, more than half did not publish MCAT or GPA information on their websites, and
programs that did include this information only provided the mean values (Christophers &
Gotian, 2019). This can be misleading as an analysis of the MD-PhD programs shows a mean
GPA of 3.79 +/- 0.19 and a mean MCAT score of 515.6 +/- 5.6 (AAMC, 2020) but the GPA and
MCAT scores broadly range from 2.68 to 4.00 and from 497 to 528 respectively (AAMC, 2020).
Respondents were asked if that transparency with admissions data for STEM PhD
programs, in general, will attract more underrepresented minority applicants to these programs
(5-point scale; 5 = Strongly Agree, 1 = Strongly Disagree, see Appendices B and C, Question
48). The responses indicate the respondents generally agreed (N = 33, µ = 4.36, σ = 1.04, min =
2.00, max = 5.00, and mode = 5.00).
Hostile Campus Environment
A very important factor in the academic success of URM students is the role of their
campus environment (Lancaster & Xu, 2017). Hostile campus environments lead to higher
attrition in the STEM fields amongst URM students (Lancaster & Xu, 2017). With only a
handful of URM students in most STEM PhD programs (SACNAS Harvard Chapter, n.d.), there
is a negative sense of belonging that creates a hostile environment (Nuñez, 2009). Of the basic
and biomedical sciences PhD students at UCSF, 22% are URM students (UCSF Graduate
Division, 2020a). This percentage is much higher than the national average of 7.9% (NSF,
2017a). Nevertheless, respondents felt it was important to have more URM students in the STEM
PhD programs (see Appendix B, Question 57, and Table 32).

127

Table 32
More URM Students in STEM PhD Programs
Question
N Mean SD Min Max Mode
How important is it to you to have more
URM students in the STEM PhD
programs? (Q57) 33 4.67 0.68 3.00 5.00 5.00

Another factor that creates a hostile environment for URM students in STEM disciplines
is the widespread negative stereotypes that URM don’t have math or science abilities (Rattan et
al., 2018). Stereotype threat is the “social-psychological threat that occurs when one is in a
situation or doing something for which a negative stereotype about one’s group applies.” (Steele,
1997, p. 614). Students perform much better academically when there is an absence of stereotype
threats (Callahan et al., 2018) and perform worse when students’ intelligence is being questioned
(Steele, 1997).
Respondents were asked if negative stereotypes of underrepresented minorities not
having abilities in math or science is a major factor that reduces URM students’ sense of
belonging to the STEM discipline (5-point scale; 5 = Strongly Agree, 1 = Strongly Disagree, see
Appendices B and C, Question 44). The analysis of the responses to these questions are shown in
Table 33.
128

Table 33
Negative Stereotypes
Question
N Mean SD Min Max Mode
Negative stereotypes that underrepresented
minorities don’t have math or science
ability is a major factor that reduces URM
students’ sense of belonging to the STEM
discipline. (Q44)
33 4.30 0.76 3.00 5.00 5.00

Furthermore, in a question asking if the respondent personally faced such negative types
(see Appendix B, Question 45), 55% said yes.
Microaggressions are brief and common every day, subtle, intentional and
unintentional biases or indignities communicated by members of a domain group or culture
towards members of a marginalized group (Limbong, 2020). Racial microaggressions toward
students of color are long-lasting and inflict cumulative wounds (Sue et al., 2007). Students of
color experience racial battle fatigue in higher education from the constant physiological,
psychological, and behavioral strains of microaggressions (Smith, 2010). In STEM disciplines,
URM students' most common experience of microaggression are the widespread negative
stereotypes that they don’t have math or science abilities (Rattan et al., 2018). Negative
stereotypes of math and science abilities can also be considered microinvalidations, which are
comments that invalidates or undermines the psychological thoughts, feeling or experiential
reality of a certain group of people (Lee et al., 2020; Sue et al., 2007).
Respondents felt it was difficult to transition to the PhD program from the institution
where they obtained their most recent degree (5-point scale; 5 = Extremely Difficult, 1 = Not at
129

all Difficult, see Appendices B and C, Question 19). The analysis of the responses to these
questions is shown in Table 34.

Table 34
Transition to PhD Program
Question
N Mean SD Min Max Mode
Adjusting and transiting to the PhD program
from the institution where you obtained
your most recent degree (bachelor or
master’s) was… (Q19)
33 2.85 0.93 1.00 4.00 2.00

130

Respondents felt adjusting and transiting to the PhD program from the institution where
they obtained their most recent degree was difficult. The top reasons included a less welcoming
environment, lack of community, and discrimination mainly in the form of micro and macro
aggressive comments and remarks (see Appendix B, Question 20).
Lack of Faculty Diversity
The importance and preference for URM students to have faculty and mentors that
looked like them were discussed earlier (see Chapter Four, URM Faculty and Mentor). The racial
and ethnic makeup of higher education faculty don’t mirror the racial and ethnic makeup of the
student body they serve (Yadav & Seals, 2019). At the institutional level, it is important to
address this deficiency in order to increase URM student enrollment and retention in the STEM
fields. The lack of diversity of faculty members is a barrier for URM students to complete
degrees in the STEM fields (National Research Council, 2011). HBCUs are extremely successful
in producing URM doctoral recipients (Whittaker & Montgomery, 2012). The top ten institutions
awarding doctoral degrees to African Americans are all HBCUs (Joseph, 2012) and this can be
largely attributed to a diverse faculty that reflects the ethnic makeup of the student body (Joseph,
2012).
Table 35 shows the data analysis of the responses for questions related to the impact of
the lack of URM role models for the respondents (5-point scale; 5 = Strongly Agree, 1 = Strongly
Disagree, see Appendices B and C, Questions: 38, 42).

131

Table 35
Lack of URM Role Models
Question
N Mean SD Min Max Mode
The lack of underrepresented minority role
models is a major de-motivation for URM
students from pursuing and persisting in a
STEM degree (bachelor, master’s, or
PhD). (Q38)
33 4.45 0.56 3.00 5.00 5.00
The availability of underrepresented minority
faculty role models allows me to
successfully create a scientist identity with
role models like themselves that have
made it. (Q42)
33 3.94 0.60 3.00 5.00 4.00

Almost all of the respondents agreed that the lack of underrepresented minority role
models is a major de-motivation for URM students from pursuing and persisting in a STEM
degree (whether bachelor, master's, or PhD) and that the availability of underrepresented
minority faculty role models allows them to successfully create a scientist identity with role
models like themselves that have made it. When asked if respondents felt alienated from their
instructor, only one student, or 3% of the respondents felt that way (see Tables 36 and 37).

132

Table 36
Alienated from Instructors
Question
N Mean SD Min Max Mode
I feel alienated with my instructors.
(Q53)
33 2.12 0.88 1.00 4.00 3.00

Table 37
Alienated from Instructors, Percentage
Question % rated 4.0 or higher to question
I feel alienated with my instructors. (Q53) 3.00%

133

The reason provided by this one student was the lack of connection with the student’s
instructor and difficulty being the student’s true self in front of instructors (see Appendix B,
Question 54).
Limited Opportunities
URM students find that the essential building skills necessary to be successful in their
STEM doctoral programs and postdoctoral careers – research opportunities, mentorship,
presenting in conferences, and publishing papers in scientific journals – are lacking (Cooper et
al., 2019; Malotky et al., 2020; Moreira et al., 2019; Whittaker & Montgomery, 2012). From an
organizational perspective, it is important that institutions consider these opportunities as part of
overall graduate success and beyond individual factors. When these opportunities are provided to
URM students, they go a long way towards helping them increase their academic persistence and
rate of degree completion (Callahan et al., 2018; Jackson et al., 2016; Moreira et al., 2019;
Sánchez et al., 2018).
When asked if these four areas – research opportunities, mentorship, presenting in
conferences and publishing papers in scientific journals – are very important for persisting and
succeeding in their PhD programs, the survey data shown in Table 38, agreed with the literature
(Callahan et al., 2018; Jackson et al., 2016; Moreira et al., 2019; Sánchez et al., 2018).

Table 38
Factors Very Important for Persisting and Succeeding in PhD Programs
Factors % agree % Strongly agree
Research opportunities 21.2% 78.8%
Mentorships 18.2% 81.8%
Presenting and attending conferences 48.5% 33.3%
Publishing papers in academic
journals
30.3% 57.6%
134

When the respondents were then asked from their own knowledge and experiences
whether underrepresented minorities PhD students, in general, were not given enough of these
opportunities in their PhD programs, their responses, shown in Table 39, confirms to the
literature (Cooper et al., 2019; Malotky et al., 2020; Moreira et al., 2019; Whittaker &
Montgomery, 2012) except with respect to the research opportunities.

Table 39
Sufficient Opportunities Not Given to URM PhD Students
Factors % Agree % Strongly Agree
Research opportunities 18.2% 6.1%
Mentorships 54.5% 3.0%
Presenting and attending conferences 39.4% 0.0%
Publishing papers in academic
journals
51.5% 6.0%

135

UCSF is a huge institution with more than $7 billion in revenues (UCSF, 2018) and more
than 26,000 full-time employees (UC, 2020). UCSF received the top NIH funding for any public
institution and second overall (UCSF, 2020a). Given these factors, it is not surprising that all the
respondents agree that there were sufficient research opportunities given to them in their PhD
programs. When it came to mentorship though, only about 57.6% of the respondents felt that
they had sufficient mentorship opportunities in their PhD programs (see Appendix B, Question
33).
Summary
This study evaluated the KMO influences that the researcher assumed were the factors
that allow STEM URM students to achieve acceptance and persist in their STEM PhD programs
at UCSF. The study identified opportunities for improving students’ acceptance and persistence
in competitive STEM PhD programs.
First-generation students might not have as much of the knowledge that continuing-
generation students have to get into and succeed in college (Roksa et al., 2018). Respondents
classifying themselves as first-generation students represented 57.5% of the total respondents.
This percentage corresponds closely to national figures for URM college students (Rudolph et
al., 2019). Only 10.5% of the first-generation students in the survey mentioned any parental
guidance for preparing and applying to college versus 85.7% for continuing generation students.
Furthermore, although students were getting some guidance and advice (see Appendices
B and C, Question 64), they were not considered sufficient. Participants strongly felt the lack of
career and educational guidance is a major barrier for URM students from applying for graduate
schools (see Appendices B and C, Question 37). Improving students’ knowledge about STEM
educational pathways, and college in general, particularly with first-generation students starting
136

at the high school level will increase URM students’ enrollment in both the undergraduate and
graduate STEM degrees. UCSF can expand its educational and informational outreach to high
school students and undergraduate institutions to provide the critical educational and career
advice early on in the students’ STEM pipeline.
Knowledge obtained with a prior master’s degree can provide students with acquiring the
skills and knowledge needed to be successful in STEM doctoral programs. Masters-to-PhD
Bridge Programs are partnerships between STEM master’s programs offered at regional
universities with STEM doctoral programs at research universities. The retention rate for URM
doctoral students in Masters-to-PhD Bridge programs are as high as 92% (Rudolph et al., 2019)
versus a national average of 50% (Joseph, 2012). Part of this is attributed to offering students
opportunities to attend workshops for professional development and exposure to a variety of
research opportunities to build up research experience and skills (Rudolph et al., 2019). Only
27.3% of the survey participants have heard about the Masters-to-PhD Bridge program. UCSF
can increase both the retention rate and recruitment of URM students by developing similar
programs with regional state colleges (See Chapter Five, Recommendations for Practice).
Respondents strongly felt that the lack of mentorships is a barrier for URM students from
applying for graduate schools (see Appendices B and C, Question 36). Mentorship can further
help incoming graduate URM students acquire the knowledge to successfully navigate through
new cultures, expectations, and environments, build networks to avoid isolation and strategies
for being successful in graduate school. Responses confirmed to the literature that mentorship is
most impactful for URM students when the mentors “look like me” (i.e., same ethnic and
cultural identity). At UCSF, the percentage of URM faculty is 12.4% (UCSF, 2020c) while the
percentage of URM enrolled in UCSF’s basic and biomedical sciences PhD programs is 22%
137

(UCSF Graduate Division, 2020a). Recruiting more URM faculty members, at the minimum, to
mirror the racial and ethnic makeup of the student body they serve will improve student
mentoring experiences and a more inclusive environment.
The study showed how student persistence in the long and rigorous PhD program can be
improved by first focusing on improving a student’s self-efficacy and motivation through finding
prosocial values and identities, having a sense of belonging to campus, engaging with URM
faculty members and mentors, and participation in mentorship and research opportunities.
Increasing persistence correlates to increased retention and graduation rates amongst URM
STEM PhD students. However, simply providing students with these opportunities isn’t enough.
To be impactful and inclusive, these opportunities need the full engagement of the organization.
For example, although two-thirds of the respondents felt it was “Extremely Important” in
aligning and framing STEM education and research with social justice and helping the
community (see Appendices B and C, Question 46), respondents felt peers and especially their
departments did not share the same level of enthusiasm (see Appendices B and C, Question 62).
The study validated and provided opportunities for addressing organizational barriers for
URM students in STEM PhD programs. The barriers that exist for URM students start at their
initial college application, through the admissions process and extend to the entire duration of
their undergraduate and graduate programs. URM students face admissions biases, and for the
few who are accepted into the doctoral program, they face a hostile campus environment, sharing
little cultural commonality with non-diverse faculty and limited opportunities to persist and
increase their skills and knowledge in both their programs and post-doctoral careers. Institutions
need to address these barriers if they are to make any real progress addressing this problem of
practice. UCSF in particular can address these barriers with outreach and educational programs
138

with high schools and undergraduate institutions to provide URM students with the critical
educational guidance and advice early in the STEM pipeline. UCSF can ensure that admissions
into its PhD program use a holistic, comprehensive approach that goes beyond academic
preparation that currently focuses on GRE, undergraduate GPA, the ranking of undergraduate
institutions attended, and research experience that weigh negatively against URM applicants.
Furthermore, these criteria may not be good predictors of student productivity (first-author
student publications) or degree completion in the STEM PhD programs (Hall et al., 2017; Miller
et al., 2019) and may have missed the applicants that are most capable of completing the
program (Rudolph et al., 2020).
It is interesting to note that in the traditionally white and Asian, male-dominated STEM
PhD programs, two-thirds of the participants in the survey were URM and female. Although this
is not surprising given that 56% of the students in the programs are female (UCSF Graduate
Division, 2020a) as compared to the national average of 47% (NSF, 2017d), the information
collected in the survey could be of use to UCSF to improve the experiences and outcomes of
female, URM students enrolled in the basic and biomedical sciences PhD programs.
To conclude, all fourteen KMO influences (four knowledge influences, five motivational
influences, five organizational influences) were validated in the study and summarized in Table
40. Clark and Estes’ (2008) gap analysis conceptual framework indicates that these types of
barriers result in the gaps between desired performance goals and actual performance. Clark and
Estes (2008) believed that knowledge, motivation, and organizational factors must be in place and
aligned with each other before any successful goal can be achieved.

139

Table 40
Summary of Influences Results Data Evaluated
Influences Sources Evidence
Knowledge
Knowledge deficiencies in first-
generation students
Kong et al., 2013; Roksa et al.,
2018; Witkow & Fuligni, 2011
Validated
Benefits of student mentorship Lisberg & Woods, 2018; McGee,
2016; Roksa et al., 2018;
Sánchez et al., 2018
Validated
Master’s degree as stepping stone to
doctorate degree
Okahana et al., 2018; Rudolph et
al., 2019; Stassun et al., 2011;
Stassun et al., 2018
Validated
Understanding STEM educational
pathways
Christophers & Gotian, 2019;
Maton et al., 2016; Moreira et
al., 2019
Validated
Motivational
Self-efficacy and motivation Bandura, 1977; Chemers et al.,
2001; Katz et al., 2014
Validated
Finding prosocial values and
identities
Guy & Boards, 2019; Jackson et
al., 2016; Thoman et al., 2014
Validated
Sense of belonging to campus Hausmann et al., 2017; Hurtado,
1997; Rattan et al., 2018
Validated
Building persistence through
mentorship and research
opportunities
Cooper et al., 2019; Estrada et al.,
2018; Malotky et al., 2020;
Mantai, 2019; Moreira et al.,
2019
Validated
URM faculty and mentors Joseph, 2012; Sánchez et al., 2018;
Whittaker & Montgomery,
2012; Yadav & Seals, 2019
Validated
Organizational
Barriers for first-generation students Choy et al., 2000; Rudolph et al.,
2019
Validated
Admission biases in doctoral
programs
Hall et al., 2017; Miller et al.,
2019; Posselt, 2018; Wilson et
al., 2019
Validated
Hostile campus environment Ben-Zeev et al., 2017; Limbong,
2020; Nadal, 2014; Sánchez et
al., 2018; Steele, 1997
Validated
Lack of faculty diversity Gibbs et al., 2015; Hassouneh et
al., 2014; Williams et al., 2017;
Yadav & Seals, 2019
Validated
Limited opportunities Mendoza-Denton et al., 2017;
Moreira et al., 2019; Toretsky et
al., 2018
Validated
140

Chapter Five: Recommendations
Chapter Four focused on discussing the study’s finding of KMO influences pertinent to
this study’s first research question: What are URM students’ knowledge, motivation, and
organizational influences to achieve acceptance and persist in their STEM PhD programs at
UCSF?
This chapter discusses the findings from the previous chapter to address the study’s
second research question: What are the recommendations, for knowledge, motivation, and
organizational resources, for UCSF to increase and retain URM students in their STEM PhD
programs? Following the discussion of the findings, the researcher will come up with a set of
recommendations based on the KMO barriers identified in the previous chapter. Finally, the
researcher will finish the dissertation with a conclusion.
Discussion of Findings
The analysis of the data is organized around the research questions and correlates to the
Clark and Estes' (2008) KMO gap analysis framework. The KMO influences impacting URM
students in STEM PhD programs are categorized and organized around three major themes:
● Declarative, procedural, and metacognitive knowledge
● Theories of expectancy-value and self-efficacy for motivation
● Organizational changes
There is one section each for the knowledge, motivation, and organizational influences. Each
section contains detailed discussions of the survey data as it relates to the influences.
Discussion of Findings for Knowledge Influences
First-generation students are students who are the first in their families to attend college
(Roksa et al., 2018). The percentage of respondents who are first-generation students is 57.5%.
141

This figure corresponds closely to the percentage of Black and Latinx college students who are
first-generation students (Rudolph et al., 2019). The survey data showed that first-generation
students have deficiencies with three knowledge types: factual knowledge, procedural
knowledge, and conceptual knowledge. According to Roksa et al. (2018), the lack of parental
experience with preparing, applying, and attending a college means first-generation students will
not have as much of the factual knowledge that continuing-generation students have to get into
and succeed in college. Without a parent that has gone through the college admissions process,
high school students might not be aware or knowledgeable about college eligibility requirements
(Rudolph et al., 2019) until it is too late. An example of this would be meeting all the required
math courses for college eligibility to the University of California (UC) or California State
University (CSU) systems (Rudolph et al., 2019). The study confirms Kong et al. (2013) that
first-generation students rely more on secondary education teachers and counselors for preparing
for STEM educational pathways. When asked who provided the guidance and preparation for
applying to college (see Appendix B, Question 15), only 10.5% of first-generation respondents
mentioned that their parents played a part in providing guidance for preparing and applying for
college (see Table 10). First-generation respondents relied on themselves (52.6%) or teachers
and counselors (also 52.6%) for providing the guidance. For the 42.5% of the respondents who
are continuing-generation students and responded to the same question, 87.5% mentioned
that their parents played a role and 14.3% mentioned relying on themselves.
A deterring factor for URM students not pursuing a doctoral degree is the lack of
knowledge of how to pursue such a degree. The majority of URM, first-generation students
obtained their undergraduate degree in minority-serving institutions (MSI) or historically Black
colleges and universities (HBCU) (Brown & Davis, 2001; Joseph, 2012; NSF, 2019a) rather than
142

R1 universities. R1 universities award the most science and engineering doctorate degrees
(London et al., 2014). Most first-generation URM students are not able to obtain the critical
knowledge and guidance for pursuing a STEM PhD degree due to the lack of resources at MSIs
or HBCUs compared to wealthier and more prestigious R1 universities. The respondent data
showed that a sizable majority (57.9%) of first-generation students graduated from non-R1
university as compared to 14.3% for continuing-generation students (see Appendix B, Question
6), corresponding to the literature (Brown & Davis, 2001; Joseph, 2012; NSF, 2019a). Texas
A&M University System (one of the nation’s largest systems of higher education) reveals two of
the main reasons URM students do not pursue STEM doctoral degrees are not knowing the
career options available for such a doctoral degree and not understanding the process to reach
such a goal (Moreira et al., 2019). The survey data confirms to the literature about the
importance of being provided career, and academic guidance and advice on pursuing a graduate
degree (master's, PhD) in their undergraduate years (see Appendix B, Question 64) as 87.8%
responded that this was “Very Important” (Xi = 4) or “Extremely Important” (Xi = 5). Yet only
20% of all respondents who didn’t attend an R1 university for their undergraduate degree felt
that the guidance and advice they received was “Sufficient” (see Appendix B, Question 65).
According to Moreira et al. (2019), four of the top reasons that URM students don’t
pursue doctoral degrees are all related to lack of knowledge of the program. Confirming this,
75.8% of the respondents strongly agree (Xi = 5) the lack of career and educational guidance is a
major barrier for URM students from applying for graduate schools (see Appendix B, Question
37). Universities must share program information to prospective applicants about the PhD
program and beyond as research has shown that what motivates an applicant to get a science or
engineering PhD degree in the first place may be different from their actual experience in
143

obtaining the degree (London et al., 2014). One common statement from first-generation URM
students was “For me, being from a really small town, they won’t understand the PhD and what
that really means” (Moreira et al., 2019, p. 3).
To get them interested in pursuing a STEM PhD, applicants must be made aware of
career trajectories for post-PhD careers such as the advancement and opportunities that are
available. This information is especially important for URM applicants who cite a lack of
mentorship and lack of needed information as barriers to applying for graduate schools
(Christophers & Gotian, 2019). The survey responses indicated 70% of the respondents who
were non-R1 university graduates felt their undergraduate program preparation was inadequate
for their PhD program or found the PhD program was quite difficult and rigorous as compared to
their undergraduate degree (see Table 13).
Productive and quality mentorship contributes to increased student efficacy, academic
achievement, productivity in scholarship, and persistence (Estrada et al., 2018) as it helps
graduate students build the skills to focus on academic norms and behaviors associated with their
discipline (Joseph, 2012). Furthermore, mentorship provides incoming URM graduate students
opportunities to successfully navigate through new cultures, expectations, and environments,
build networks to avoid isolation, and strategies for being successful in graduate school (Moreira
et al., 2019). The respondents agree with the importance of the mentorship stated in the
literature. Even at the undergraduate level, 48.5% of the respondents felt that mentorship was
very important (Xi = 5) for motivating students to pursue STEM PhD degrees (see Appendices B
and E, Question 10). All the respondents felt (Xi >= 4) that the lack of mentorship is a barrier for
URM students from applying for graduate schools (see Appendices B and E, Question 36).
144

The presence of mentorship has many added benefits for students (Estrada et al., 2018).
Mentors can help students gain familiarity and knowledge of on-campus student support
programs (student organizations, career guidance, undergraduate research, and academic
support), build professional networks, and enhance\ research skill (Lisberg & Woods, 2018). The
lack of mentorship means that these foundational skills needed for planning and pursuing
graduate education are not available to URM students. This belief is also reflected in the survey
data. The survey responses indicated 85% of the respondents “Strongly Agree” (Xi =5) that
mentorship is a key to their academic success (see Appendices B and E, Question 30). All the
respondents felt (Xi >= 4) that mentorship increases participation and persistence (see
Appendices B and E, Question 31) and 97% also felt (Xi >= 4) mentorship increases a sense of
belonging (see Appendices B and E, Question 32) in their doctoral program. However, given the
importance of mentorship to students especially for graduate students, it was surprising that
42.4% of respondents were neutral (Xi = 3) when asked if they had sufficient mentorship
opportunities in their doctoral program (see Appendices B and E, Question 33).
Furthermore, 97% of the respondents felt (Xi >= 4) that the lack of underrepresented
minority role models is a major de-motivation for URM students from pursuing and persisting in
a STEM degree (bachelor, master's, or PhD) (see Appendices B and E, Question 38). This is
because respondents felt that underrepresented minority role models in mentorship roles can help
reduce imposter syndrome and stereotype threat, are more relatable, can create a sense of
belonging, and are more comfortable going to seek advice than non-URM role models (see Table
15).
Having a prior master’s degree helps students acquire the knowledge and preparation
needed to be successful in STEM doctoral programs (Okahana et al., 2018). Students with
145

master’s degrees can have access to courses and knowledge unavailable to them as
undergraduates or gain strong research experiences (Rudolph et al., 2019). URM students are
50% more likely than their non-URM peers to obtain a master’s degree on their way to a PhD
(Stassun et al., 2011). Also, applicants who are denied to PhD programs might pursue a master’s
degree before reapplying to the PhD programs (Rudolph et al., 2019). Surprisingly, only about
6% of the respondents completed a master’s degree prior to starting their PhD program (see
Table 14) even though 68.8% of respondents felt that their PhD program was quite difficult and
rigorous as compared to their undergraduate degree (see Table 12 and Table 13).
In the past two decades, partnerships have emerged to leverage the benefits of obtaining a
master’s degree prior to entering a STEM doctoral program. These are commonly referred to as
Masters-to-PhD Bridge Program (Stassun et al., 2011) and consists of partnerships between
STEM master’s programs offered at regional universities with STEM doctoral programs at
research universities to leverage this positive correlation (Okahana et al., 2018; Rudolph et al.,
2019; Stassun et al., 2011). Part of the program’s goal is to develop a diverse pool of PhD
scientists in the biomedical sciences (NIH, 2020) and URM STEM PhD students participating in
bridge programs have completion rates that are twice the national average (Moreira et al., 2019).
The bridge programs address the deficiencies in the current admission process of mainly looking
at GRE, undergraduate GPA, the ranking of undergraduate institutions attended, and research
experience that weigh negatively against URM applicants (Rudolph et al., 2019; Stassun et al.,
2011). Instead, a holistic, comprehensive approach that goes beyond academic preparation is
used (Rudolph et al., 2019; Stassun et al., 2011; Stassun et al., 2018) including accounting for the
applicants' socioemotional skills necessary to succeed in the PhD program such as perseverance
and ability to focus on long-term goals (Rudolph et al., 2019). Given bridge programs’ goal to
146

increase URM students (NIH, 2020) and its high completion rates (Rudolph et al., 2019; Stassun
et al., 2011), only 27.2% of respondents have heard about the program and of those that heard
about it, their responses were aligned between agreeing and neutral (“Agree” or “Neither agree
nor disagree”) about whether the programs can help increase URM enrollment in the STEM PhD
program (see Table 17 and Appendices B and E, Question 47).
Discussion of Findings for Motivation Influences
A strong motivational influence for URM students is the desire to help others (Deci &
Ryan, 1991). URM students find a strong link between their education and their desire to help
their communities (Guy & Boards, 2019). Coming from a culture with a strong emphasis on
prosocial values (Jackson et al., 2016), URM students place great value in helping others and
giving back to the community through their work (Jackson et al., 2016). Surveys and interviews
of URM graduating doctoral students highlight the importance of prosocial values for persisting
in science education (Jackson et al., 2016) and more likely to place the application of research to
address health problems specific to their communities (Jackson et al., 2016). The survey
responses indicated 85% of the respondents felt that seeking prosocial goals and opportunities
(such as helping communities) in the classroom, conferences, or research of their STEM PhD
program was either “Very Important” (Xi = 4) or “Extremely Important” (Xi =5) to them (see
Appendices B and E, Question 61). Furthermore, when asked how important it is to align or
frame STEM education and research with social justice and helping the community (see
Appendices B and E, Question 46), two-thirds of the respondents felt it was “Extremely
Important” (Xi =5).
URM students attending PWIs often encounter the lack of prosocial values from majority
non-URM peers (Joseph, 2012) that are firmly rooted in URM students’ cultural and ethnic
147

background (Thoman et al., 2015). Even at the undergraduate level, the desire of working for
social change was more important for URM students than their non-URM counterparts (McGee
et al., 2016). The survey responses indicated 51.5% of the respondents felt (Xi >= 4) that their
peers do not share the same enthusiasm as them in seeking prosocial goals and opportunities in a
STEM PhD program (see Appendices B and E, Question 62) while 45.4% of the respondents
“Neither agree nor disagree” (Xi = 3). Finally, when asked the same question for their
department instead of their peers (see Appendices B and E, Question 63), about 42.4% of the
respondents either “Agree” (Xi = 4) or “Strongly Agree” (Xi = 5). This disconnection between
their prosocial values and their science education creates a contradiction between their cultural
and science identities (Jackson et al., 2016).
Students’ academic performance and persistence increases when they identify with a role
identity such as a scientist (Chemers et al., 2011). URM students have greater persistence to
complete their degrees when identifying with a role identity (Eccles & Barber, 1999). When
asked if identifying with a science career or scientist identity is a very important motivator to
persist in the PhD program (see Appendices B and E, Question 41), 93.9% of the respondents
answered either “Agree” (Xi = 4) or “Strongly Agree” (Xi = 5). Even though respondents felt a
disconnection with their peers and departments regarding the importance of connecting prosocial
values with science education (see previous paragraph), 100% of the respondents either
“Disagree” (Xi = 2), “Strongly Disagree” (Xi = 1) or “Neither agree nor disagree” (Xi = 3) that
pursuing a science careers or scientist identity means losing their racial or cultural identity (see
Appendices B and E, Question 40). When URM students are encouraged to realize their
academic potential by embracing their ethnic and gender identity, it led to a more than one-fifth
increase in obtaining STEM graduate degrees (Jackson & Winfield, 2014).
148

A strong science identity could serve as a buffer in the face of devaluation through racial
biases and negative stereotypes (Kim‐Prieto et al., 2013) that many URM students face. Of the
54.5% of the respondents who felt (Xi >= 4) adjusting and transiting to the PhD program from
the institution where they obtained their most recent degree was difficult (see Appendices B and
E, Question 19), 62.5% mentioned a less welcoming environment, lack of community, and
discrimination mainly in the form of micro and macro aggressive comments and remarks (see
Table 12 and Table 13).
URM students seeing a successful faculty with shared ethnic and cultural backgrounds
signals to the student that they too can be successful (Yadav & Seals, 2019) while the lack of
successful URM role models makes it difficult to create a science career or scientist identity
(McGee, 2016). This is confirmed by 78.8% of the respondents who “Agree” (Xi = 4) or
“Strongly Agree” (Xi = 5) that having underrepresented minority faculty role models allows
them to successfully create a scientist identity with role models like themselves that have made it
(see Appendices B and E, Question 42). Two-thirds of the respondents “Agree” (Xi = 4) or
“Strongly Agree” (Xi = 5) that it is difficult to have a scientist identity when there is a lack of
role models that look like them (see Appendices B and E, Question 39).
In the rigorous and challenging STEM PhD programs, it is very important for students to
have a high level of self-efficacy. A student’s persistence in a STEM college program is
predicted by their expectation for success (Andersen & Ward, 2014). At the graduate and
postdoctoral levels, science, leadership, and teamwork self-efficacy each independently predicted
the students' persistence all the way to a science career (Chemers et al., 2011). A high level of
self-efficacy results in a person to not just persist and succeed in their tasks but also are more
likely to recover and bounce back from any failures (Chemers et al., 2001). When tasks are
149

challenging and difficult, people with a low level of self-efficacy will become discouraged and
withdrawn (Klassen et al., 2008). As UCSF’s STEM PhD programs is extremely selective, it is
not surprising that on most the self-efficacy questions, 81.2% of the respondents rated
themselves very highly (Xi >= 4) on the self-efficacy survey questions and 72.7% responded
favorably (Xi >= 4) to dealing with difficult tasks (see Table 20).
The review of literature in Chapter Two stressed the strong linkage of STEM education
with prosocial values for URM students (Guy & Boards, 2019; Jackson et al., 2016). STEM
URM students felt that working for social causes and changes were more important than their
non-URM counterparts (McGee et al., 2016). URM students place great value in helping others
and giving back to the community through one’s work (Jackson et al., 2016) such as linking
research to making positive community changes (Guy & Boards, 2019). URM PhD holders were
significantly more likely to place the application of research to address health problems specific
to their communities (Jackson et al., 2016) and placed the importance of prosocial values for
persisting in science education (Jackson et al., 2016). Surprisingly, none of the top three reasons
that motivated respondents to pursue a PhD degree had to do with prosocial values and social
justice causes (see Appendix B, Question 18).
Students enrolling in college are often faced with social pressures and issues transitioning
to college (Lisberg & Woods, 2018). The survey responses indicated 88.9% of the respondents
agreed (Xi >=4) that learning coping skills and resiliency dealing with social pressures and
transition to college is very important (see Appendices B and E, Question 17). On entering
graduate school, URM students find that they need to successfully navigate through new
cultures, expectations and environments and learn to build networks to avoid isolation and learn
strategies for being successful in graduate school (Moreira et al., 2019). All respondents felt that
150

acquiring learning strategies in relation to persisting and succeeding in their PhD program was
very important (see Appendices B and E, Question 69), but when asked if they had sufficient
opportunities to acquire learning strategies in their PhD program or in their undergraduate or
master's degree only 60.6% agreed (Xi >=4) (see Appendices B and E, Question 70). The survey
responses indicated 72% of the respondents felt (Xi >=4) that acquiring study skills in relation to
persisting and succeeding in their PhD program was important, but similarly to Question 70, only
48.5% of the respondents felt (Xi >=4) they had sufficient opportunities to acquire such study
skills (see Appendices B and E, Question 72).
Most of these “soft skills” are obtained through mentors that go beyond just academic
and career counseling (Moreira et al., 2019) to build the skills needed to focus on academic
norms and behaviors associated with their discipline (Joseph, 2012). All respondents felt (Xi
>=4) that mentorship is a key to their academic success (see Appendices B and E, Question 30)
and all respondents felt (Xi >=4) that mentorship increases their participation and persistence
(see Appendices B and E, Question 31). When asked if they had sufficient mentorship
opportunities in their doctoral program (see Appendices B and E, Question 33), 42.4% of the
respondents were neutral (Xi =3). This can be an issue as poor or lack of mentorship is a major
factor in high attrition rates amongst doctoral students (Moreira et al., 2019).
Research opportunities increase student retention and persistence in STEM programs
(Estrada et al., 2018) and are an effective strategy to increase student self-efficacy in science
(Burton et al., 2019). Shadding et al. (2016) showed that research opportunities have positive
effects on retaining students in STEM. For URM students in general, participating in research
opportunities led to greater gains in thinking and working like a scientist along with confidence
in their research skills and ability to succeed in graduate school (Cooper et al., 2019). Female
151

STEM URM doctoral students mention research as one of the two main drivers for their
persistence in STEM (Guy & Boards, 2019). All the respondents conferred (Xi >=4) with the
review of literature that research opportunities are very important for them to persist and to
succeed in their PhD programs (see Appendices B and E, Question 23). All respondents felt (Xi
>=4) they are given sufficient research opportunities in their PhD programs (see Appendices B
and E, Question 24). Finally, when asked from their own experience and knowledge, only 24.2%
of the respondents felt underrepresented minority PhD students, in general, are not given enough
research opportunities (Xi >=4) (see Appendices B and E, Question 25).
Students’ sense of belonging increases persistence toward degree completion (Hausmann
et al., 2017) while lack of belonging demotivates students, decreases academic persistence
(Mantai, 2019), and causes mental health and behavior issues (Museus et al., 2018). For URM
students, belonging is even more critical given the loneliness and isolation that arises from the
scant number of URM students in STEM PhD programs (Brown, 2013; Harmon, 2019;
SACNAS Harvard Chapter, n.d.) and lack of URM faculty members to turn to (Yadav & Seals,
2019). The survey responses indicated 87.9% of the respondents felt (Xi >=4) it was important to
have more URM students in the STEM PhD programs (see Appendices B and E, Question 57),
and 90% felt it was either “Very Important” (Xi = 4) or “Extremely Important” (Xi = 5). The
isolation and alienation experienced by Edray Goins as a PhD doctoral student in mathematics
covered in a New York Times article titled, “For a Black Mathematician, What It’s Like to be
the ‘Only One’” (Harmon, 2019) typifies the experience of URM students in a STEM PhD
program. The survey responses indicated 85% of the respondents agreed (Xi >= 4) that there is
loneliness and isolation being an underrepresented minority student in a STEM PhD program
(see Appendices B and E, Question 58). All respondents agreed (Xi >= 4) that it is important to
152

build peer and faculty networks to avoid isolation and loneliness in a STEM PhD program (see
Appendices B and E, Question 59). Yet 18.2% of the respondents felt (Xi <=3) that they never
had sufficient opportunities to build peer and faculty networks in their current PhD program or in
their undergraduate or master's degree programs (see Appendices B and E, Question 60). The
disconnect between their peers and departments on prosocial values discussed earlier in this
section further adds to URM students’ lack of belonging to their campus.
On whether they have a strong sense of belonging to the campus, only 36.4% of the
respondents agreed (Xi = 4), 27.2% Neither agree nor disagree (Xi = 3), and 36.4% disagreed (Xi
= 2) (see Appendices B and E, Question 51). When asked if they have a strong sense of
belonging to their program, 63.6% of the respondents agreed (Xi > 3) and 36.4% disagreed (Xi
<=3) (see Appendices B and E, Question 52).
Access and use of student services can also increase a student’s sense of belonging on the
campus (Lisberg & Woods, 2018). All respondents agree (Xi >= 4) that publicizing and
providing on-campus student support programs and services is very important for student
persistence and success in the PhD programs (see Appendices B and E, Question 49). All
respondents felt (Xi >= 4) that in their doctoral program, they have access to student services
when they need it (see Appendices B and E, Question 50).
There is a lack of faculty diversity in STEM doctoral programs (Yadav & Seals, 2019).
URM students’ persistence and academic success hinge on having access to URM faculty
(Joseph, 2012; Whittaker & Montgomery, 2012). URM students strongly desire mentors that are
like them (Whitfield & Edwards, 2011) as it is difficult for students to create a career or scientist
identity when there is a lack of role models like themselves (McGee, 2016). When URM
students see a successful faculty with a shared ethnic and cultural background, it signals to the
153

student that they too can be successful (Yadav & Seals, 2019). The survey responses indicated
69.7% of the respondents agreed (Xi >= 4) that having a mentor that looks like them (i.e., same
ethnic and cultural identity) was important (see Appendices B and E, Question 34). URM
students find it difficult to have a positive identity in a campus climate that is hostile to them
(Sánchez et al., 2018). In a hostile campus climate, URM faculty mentors may need to draw from
their own racial encounters and experiences to counsel URM students who are questioned about
their abilities due to racial stereotyping (Sánchez et al., 2018). The respondents’ top three
reasons to have a mentor that looks like them were feeling more relatable/sense of belonging,
more empathy/sympathy, and more likely to go seek advice from (see Table 15).
Discussion of Findings for Organizational Influences
First-generation students make up 30% of doctorate recipients (Roksa et al., 2018) and
amongst URM this percentage is nearly half (Rudolph et al., 2019). First-generation students
face major organizational barriers throughout the STEM pipeline. Statistically speaking, first-
generation students' chance of attending college is already stacked against them. Only 27% of
first-generation students attend college compared to 42% and 71% for students with parents that
had some college and parents who were college graduates, respectively (Choy et al., 2000). First-
generation students’ lack of parental experience means they need to rely on high school teachers,
counselors, and college prep programs to guide them through the college admissions process.
Only 10.5% of first-generation respondents mentioned that their parents played a part when
applying for college versus 42.5% of the respondents who are continuing-generation students
(See Table 9 and Table 10).
First-generation students experience a decrease in parental guidance (versus parents of
continuing-generation students) to discuss college test preparation exams (16% versus 27%) and
154

postsecondary plans (42% versus 61%) with their high school-age children (Choy et al., 2000).
Critical and even mandatory college preparation activities like visiting colleges, ensuring all
course requirements met, seeking financial aid, and educational opportunity programs also
decrease with parents' highest level of education (Choy et al., 2000). This parental disadvantage
results in a sizable percentage of first-generation students being accepted into minority-serving
institutions (MSI) rather than more prestigious or high research universities. Of the respondents
who are first-generation, none attended Ivy Leagues colleges and only 5.2% attended a top ten
ranked national university (U.S. News & World Report, 2021a) as compared to 28.5% and
35.7% respectively for continuing-generation students. If the top ten ranked national liberal arts
colleges (U.S. News & World Report, 2021a) are included, then about half of all continuing-
generation students attended either a top ten ranked national liberal arts college or national
university. Furthermore, only 42.1% of respondents who are first-generation students graduated
from an R1 university (very high research doctoral-granting university) as compared to 85.7%
for continuing-generation students.
Not obtaining an undergraduate degree from an R1 university can be a severe
disadvantage for students wanting to get into a competitive and highly selective STEM PhD
program. In elite institutions, faculty members tend to favor PhD applicants who come from
similar institutions (Posselt, 2018). This is not surprising given that institutions classified as
“very high research activity” (R1 in the Carnegie Classification) award the most science and
engineering doctorate degrees (London et al., 2014). Clauset et al. (2015) showed in their
analysis of 19,000 regular faculty in three disparate disciplines that about a quarter of the
institutions produced 71% to 86% of all tenure-track faculty. Furthermore, 70% to 90% of the
faculty at these elite institutions obtained their doctorates from other elite institutions (Clauset et
155

al., 2015). At UCSF, the admission process is more equitable for applicants coming from non-
elite institutions. Although two-thirds of the applicants came from R1 (“very high research
activity” institutions in the Carnegie Classification) or R2 (“high research activity” institutions in
the Carnegie Classification) universities, only one-third of the applicants came from the top
thirty national universities as ranked by US News and World Report (U.S. News & World
Report, 2019).
Even if a non-R1 graduate is accepted into a STEM PhD program, the student will not be
adequately prepared with the conceptual knowledge necessary for dealing with the academic
rigors of a STEM PhD program that R1 universities prepare their students for. For respondents
who attended a non-R1 university, only 28.6% felt their undergraduate degree institution
provided them with the adequate skills and foundations needed to persist and succeed in their
STEM PhD program (see Appendices B and E, Question 8) versus 89.5% of respondents who
obtained their undergraduate degree from an R1 university. Furthermore, 70% of non-R1
graduates felt that they came to the STEM PhD program from their undergraduate program
inadequately prepared or felt the PhD program was quite difficult and rigorous (compared to the
undergraduate degree) (See Table 13). This contrasts to only 16.7% of those attending an R1
university feeling the same way (See Table 12).
Transparency in admission data can lead to an increase in URM applicants to STEM PhD
programs. The majority of URM students obtained their undergraduate degrees in minority-
serving institutions (MSI) or historically Black colleges and universities (HBCU) (Brown &
Davis, 2001; Joseph, 2012; NSF, 2019a). MSIs and HBCUs do not have the resources of their
wealthier R1 universities to help their students obtain the critical knowledge and guidance for
pursuing a STEM PhD degree. URM applicants cite a lack of mentorship and lack of needed
156

information as barriers to applying for graduate schools (Christophers & Gotian, 2019). The lack
of these resources means URM applicants are heavily reliant on possibly non-complete
admissions data provided from program websites of R1 universities to decide whether or not to
apply to graduate programs (Christophers & Gotian, 2019). As a telling example, in the very
selective and competitive MD-PhD programs, more than half did not publish MCAT or GPA
information on their websites, and programs that did include this information only provided the
mean values (Christophers & Gotian, 2019). This can be misleading as an analysis of the MD-
PhD programs shows a mean GPA of 3.79 +/- 0.19 and a mean MCAT score of 515.6 +/- 5.6
(AAMC, 2020) but the GPA and MCAT scores broadly range from 2.68 to 4.00 and from 497 to
528 respectively (AAMC, 2020). The survey responses indicated 81.8% of the respondents agree
(Xi >=4) that more transparency with admissions data for STEM PhD programs, in general, will
attract more underrepresented minority applicants to these programs (see Appendices B and E,
Question 48).
A very important factor to increase both URM applicants and URM student retention is to
make campus environments less hostile. The campus environment plays a very important part in
the academic success of URM students (Lancaster & Xu, 2017) as a hostile campus environment
contributes to higher attrition for URM students in the STEM fields (Lancaster & Xu, 2017). The
handful of URM students found in most STEM PhD programs (SACNAS Harvard Chapter,
n.d.), means there is already a negative sense of belonging that creates a hostile environment
(Nuñez, 2009) along with a sense of loneliness and isolation (Harmon, 2009). All respondents
definitely felt it was important (Xi >=4) to have more URM students in the STEM PhD programs
(see Appendices B and E, Question 57).
157

Besides the loneliness and isolation is the widespread negative stereotypes that URM
don’t have math or science abilities (Rattan et al., 2018) that further contributes to a hostile
campus environment. Students perform much better academically when there is an absence of
stereotype threats (Callahan et al., 2018) and perform worse when students’ intelligence is being
questioned (Steele, 1997). The survey responses indicated 81.8% of the respondents felt (Xi >=4)
that negative stereotypes of underrepresented minorities not having math or science abilities is a
major factor that reduces URM students’ sense of belonging to the STEM discipline (see
Appendices B and E, Question 44). A majority (55%) of the respondents personally encountered
these negative stereotypes (see Appendices B and E, Question 45). Another factor contributing to
a hostile campus is microaggressions. Racial microaggressions toward students of color are long-
lasting and inflict cumulative wounds (Sue et al., 2007). Of the 54.5% of the respondents who
felt (Xi >= 4) adjusting and transiting to the PhD program from the institution where they
obtained their most recent degree was “Moderately Difficult” or “Very Difficult” (see
Appendices B and E, Question 19), top reasons included a less welcoming environment, lack of
community, and discrimination mainly in the form of micro and macro aggressive comments and
remarks (see Table 12 and Table 13).
A major barrier for URM students to complete degrees in the STEM fields is the lack of
diversity of faculty members (National Research Council, 2011). The survey responses indicated
97% of the respondents agreed (Xi >=4) that the lack of underrepresented minority role models is
a major de-motivation for URM students from pursuing and persisting in a STEM degree
(whether bachelor, master's, or PhD) (see Appendices B and E, Question 38). There is a strong
desire for URM students to prefer mentors that are like them (Whitfield & Edwards, 2011). The
survey responses indicated 78.8% of the respondents agreed (Xi >=4) that the availability of
158

underrepresented minority faculty role models allows them to successfully create a scientist
identity with role models like themselves that have made it (see Appendices B and E, Question
42). The presence of a diverse faculty results in better academic performance for URM students
(Miller, 2018). The diversity of the faculty in the HBCUs has resulted in HBCUs producing
more URM doctoral recipients than other higher educational institutions (Joseph, 2012;
Whittaker & Montgomery, 2012) and the top ten institutions awarding doctoral degrees to
African Americans are all HBCUs (Joseph, 2012).
Institutions need to diversify their faculty members as faculty in higher education doesn’t
mirror the racial and ethnic makeup of the student body they serve (Yadav & Seals, 2019). URM
made up 36.8% of America’s college-age (18-24) population (NSF, 2017b), but only 8.6% of all
full-time faculty members were URM (NSF, 2018b). Racial, cultural, and ethnic differences
between students and faculty can also lead to difficulties and challenges with communication
(McGee, 2016), however, this is not the case at UCSF. When asked if any of the respondents felt
alienated from their instructor, only 3% felt (Xi >=4) that way (see Appendices B and E,
Question 53), and for those that provided a reason, it was either the lack of connection with their
instructor or not able to be their true self in front of their instructor (see Appendices B and E,
Question 54).
The literature has identified research opportunities, mentorship, presenting in
conferences, and publishing papers in scientific journals as the four essentials for being
successful in STEM doctoral programs and postdoctoral STEM careers, yet in many instances
these opportunities are lacking for URM students (Cooper et al., 2019; Malotky et al., 2020;
Moreira et al., 2019; Whittaker & Montgomery, 2012). When asked if these four areas – research
opportunities, mentorship, presenting in conferences, and publishing papers in scientific journals
159

– are very important for persisting and succeeding in their PhD programs, the respondent data
(see Table 37) showed that 100% of the respondents agreed for research opportunities and
mentorships, and 81.8% and 87.9% agreed that presenting in conferences and publishing papers
in scientific journals, respectively are very important (Xi >= 4) for persisting and succeeding in
their PhD programs. The results agreed with the literature (Callahan et al., 2018; Jackson et al.,
2016; Moreira et al., 2019; Sánchez et al., 2018). When the respondents were then asked from
their own knowledge and experiences whether underrepresented minorities PhD students, in
general, were not given enough of these opportunities in their PhD programs, the responses (see
Table 38), with the exception of research opportunities, generally agreed with the literature
(Cooper et al., 2019; Malotky et al., 2020; Moreira et al., 2019; Whittaker & Montgomery,
2012). URM graduate students working with peers and scientists in research and attending
conferences reinforce experiences and recognition to support their identity as future scientists
(Kim‐Prieto et al., 2013) and address the loneliness and isolation these students face (Harmon,
2019). Therefore, from an organizational perspective, institutions need to ensure these
opportunities are available for URM students and be considered part of overall graduate success.
Recommendations for Practice
The three recommendations suggested by the researcher are tied back to the second
research question: What are the recommendations (for knowledge, motivation, and
organizational resources) for UCSF to increase and retain URM students in their STEM PhD
programs?
These recommendations are aligned and connected to a set of factors and considerations:
● Clark and Estes' (2008) gap analysis framework.
160

● Actual gap analysis to close (i.e., increasing the number of URM students in STEM
PhD programs relative to their college-age (18-24) population).
● Review of the literature in Chapter Two.
● Assertions supported by the data collection in Chapter Four.
● Evaluation and discussion of the finding in Chapter Five.
Even though many recommendations are possible, the three criteria used to select the
recommendations were based on the practicality of the recommendation, how well the
recommendation fitted in within the context and culture of UCSF’s organization, and delivering
the most benefits for the students (based on the review of literature and findings).
Recommendation 1: Holistic and Transparent Admissions Process
The current admission criteria of evaluating applicants’ undergraduate GPA, GRE scores,
ranking of the institution (where the undergraduate degree was obtained), and research
experience in competitive and selective STEM doctoral programs are heavily weighted against
URM applicants (Hall et al., 2017; Miller et al., 2019; Rudolph et al., 2019; Wilson et al., 2019).
Of the four criteria, GRE and undergraduate GPA are the two important objective criteria
used by most universities to evaluate applicants for admission into their STEM programs (Hall et
al., 2017; Miller et al., 2019; Wilson et al., 2019). On average, URM applicants have lower GRE
scores (Wilson et al., 2019) and undergraduate GPA (Miller et al., 2019) as compared to their
white and Asian counterparts. GRE scores are not a good predictor of degree completion in the
STEM PhD programs (Hall et al., 2017; Miller et al., 2019). Male students who completed the
STEM doctoral degrees had significantly lower GRE scores than those students who left the
program (Petersen et al., 2018). Most URM students earn their undergraduate degrees in public
universities (NSF, 2017d) while white students make up 87% of all undergraduates at private
161

universities (Brown, 2016). The grades given out at public universities are usually about 1/3
letter grade lower than private universities (Miller et al., 2019).
The admissions evaluation of ranking the institution where the applicant received an
undergraduate degree is also weighed against URM applicants. Faculty members in elite
institutions tend to favor applicants who come from similar institutions (Posselt, 2018) and
institutions classified as “very high research activity” (R1 in the Carnegie Classification) award
the most science and engineering doctorate degrees (London et al., 2014). Half of first-
generation students are URM students (Rudolph et al., 2019). First-generation students’ lack of
parental experience with preparing, applying, and attending college along with the students not
knowing what or where to seek outside academic help and guidance creates major organizational
barriers to getting accepted into more prestigious and research-intensive institutions. Only 10.5%
of first-generation students in the survey mentioned their parents providing guidance when
preparing to apply for college versus 85.7% for continuing-generation students (see Table 9 and
Table 10). These factors result in a sizable percentage of first-generation students being accepted
into minority-serving institutions (MSI) rather than more prestigious and research-intensive
institutions (Brown & Davis, 2001; Joseph, 2012; NSF, 2019a). None of the respondents who are
first-generation students attended Ivy Leagues colleges and only 5.2% attended a top ten ranked
national liberal arts college or national universities (U.S. News & World Report, 2021a) as
compared to 28.5% and 50.0% respectively for continuing-generation students. The percentage
of continuing-generation students attending either a national liberal arts college or a national
university ranked in the top ten (U.S. News & World Report, 2021a) is nearly ten times higher
than first-generation students. Only 42.1% of first-generation respondents obtained their
162

undergraduate degree from an R1 university (very high research doctoral-granting university) as
compared to 85.7% for continuing-generation students.
Students attending a non-R1 university are put at a disadvantage for the fourth critical
criteria used to evaluate admissions, which is research experience. Completing an undergraduate
degree in an R1 university exposes students to more experience with research and access to
faculty coming from strong research background than students attending a lower tier ranking
university (Hollman et al., 2018). As most URM students attend MSIs (Brown & Davis, 2001;
Joseph, 2012; NSF, 2019a), it isn’t surprising that research opportunities for URM students are
lacking and difficult to obtain (Toretsky et al., 2018). URM students from socially disadvantaged
or poor backgrounds are also unaware of the importance of or how to obtain research
opportunities (Toretsky et al., 2018). Even when URM students are presented with research
opportunities, they are more likely than their white peers to leave their research opportunities due
to working on less desirable tasks, working harder, or weren't learning important skills or
knowledge (Cooper et al., 2019). The survey responses indicated 44% of respondents who
attended a non-R1 university did not agree (Xi <= 3) that they were given opportunities to do lab
or research work in their undergraduate years (see Appendices B and E, Question 9).
Given that the current admission criteria are heavily weighed against URM applicants,
UCSF’s basic and medical sciences PhD programs should adopt a holistic, comprehensive
approach that goes beyond academic preparation for admissions to the programs. The evaluation
should consider the applicants' socioemotional skills necessary to succeed in the PhD program
such as perseverance, leadership traits, the ability to focus on long-term goals and the applicant’s
diverse set of life experiences that would enhance scientific work. Universities that use a holistic
admissions review are able to increase the diversity of their applicant pool (Miller et al., 2019;
163

Wilson et al., 2019). For example, the College of Nursing at the University of Illinois, Chicago
was able to increase the number of URM applicants that were offered admissions from 36.8% to
42.5% (Scott & Zerwic, 2015) and the University of Texas MD Anderson Cancer Center saw an
increase in URM students in their biomedical sciences PhD programs after taking similar actions
(Wilson et al., 2018).
UCSF has nearly three times the national average of URM students in the basic and
medical sciences PhD programs, and it won’t be surprising that some of these programs are
already using some elements of the recommended holistic approach towards admissions.
Furthermore, based solely on the respondents in the survey, the admission process at UCSF is
more equitable for applicants coming from non-elite institutions. Although two-thirds of the
applicants came from R1 (“very high research activity” institutions in the Carnegie
Classification) or R2 (“high research activity” institutions in the Carnegie Classification)
universities, only one-third of the applicants came from the top thirty national universities as
ranked by US News and World Report (U.S. News & World Report, 2019).
URM applicants are also more likely to apply to competitive doctoral programs when
applicants have access to complete and detailed admissions data (Christophers & Gotian, 2019).
The survey responses indicated 81.8% of the respondents agree (Xi >=4) that transparency with
admissions data for STEM PhD programs, in general, will attract more underrepresented
minority applicants to these programs (see Appendices B and E, Question 48). Along these lines,
UCSF making more admissions data available as part of this recommendation would encourage
more URM applicants to its programs.
164

Recommendation 2: Establish Masters-to-PhD Bridge Program
Most URM students attending MSIs (Brown & Davis, 2001; Joseph, 2012; NSF, 2019a)
aren’t exposed to the higher academic standards and rigors, gaining the research experience and
access to faculty coming from strong research background found in most research-intensive
universities (Hollman et al., 2018). This, in turn, reduces the applicant’s admissions prospect and
is not adequately prepared for the rigors of the STEM PhD programs. This lack of preparedness
presents challenges to first-generation students trying to navigate around their doctoral program
due to the lack of experience and knowledge of the “unspoken” rules in high research-intensive
universities (R1 and R2 institutions) (Roksa et al., 2018) where most of the STEM PhD degrees
are granted (Buswell, 2017). Seventy percent of non-R1 university graduates felt their
undergraduate program preparation was inadequate for their PhD program or found the PhD
program was quite difficult and rigorous as compared to their undergraduate degree (see Table
13). This is one of the factors that can lead to high graduate attrition rates in certain science and
engineering fields of more than 50% for URM students (Council of Graduate Schools, 2020;
Joseph, 2012).
Obtaining a prior master’s degree is important for URM students (Okahana et al., 2018)
as it allows them opportunities to acquire knowledge and preparation needed to be successful in
STEM doctoral programs (Okahana et al., 2018) or access courses, knowledge skills, and
research experiences unavailable to them as undergraduates (Rudolph et al., 2019). When asked
the importance of being provided career and academic guidance and advice on pursuing a
graduate degree (master's, PhD) in their undergraduate years (see Appendix B, Question 64)
87.9% responded that this was important (Xi >=4). Yet, only a fifth of all respondents who didn’t
attend an R1 university for their undergraduate degree felt that the guidance and advice they
165

received were “Sufficient” or higher (Xi >=4) (see Appendices B and E, Question 65). Also, all
the respondents agree (Xi >=4) that if there were more career and educational guidance provided
at the high school or undergraduate years, more URM students would pursue graduate studies in
the STEM disciplines (see Appendices B and E, Question 68).
There is a positive correlation between having a master’s degree and earning a doctorate
within 10 years for African American students and male URM students in the STEM fields
(Stassun et al., 2011). The Masters-to-PhD Bridge programs are a partnership between STEM
master’s programs offered at regional universities with STEM doctoral programs at research
universities to leverage this positive correlation (Okahana et al., 2018; Rudolph et al., 2019;
Stassun et al., 2011). These bridge programs also address the deficiencies and biases found in the
admissions evaluation that disproportionately affects URM applicants and students (Rudolph et
al., 2019; Stassun et al., 2011; Stassun et al., 2018) discussed in Recommendation 1.
Research universities that have established the bridge programs have been very
successful. The retention rate for URM doctoral students in Masters-to-PhD Bridge programs is
as high as 92% (Rudolph et al., 2019) which is almost double the national average (Joseph,
2012). The Fisk–Vanderbilt Master’s-to-PhD Program is a bridge program focusing on physics
and astronomy (Stassun et al., 2011). The partnership is between Fisk University (an HBCU) and
Vanderbilt University, an R1 research university two miles away from Fisk (Stassun et al.,
2011). The program’s purpose is to build a relationship between master’s degree-seeking
students attending Fisk University with Vanderbilt University (Stassun et al., 2011). This
includes Fisk students being jointly mentored and conducting research with faculty at both
institutions and taking at least one core PhD course at Vanderbilt University each term (Stassun
et al., 2011). The end result is that by the time Fisk students complete their master’s degree and
166

enter Vanderbilt’s PhD program, they would already have the critical foundations needed to
succeed in a PhD doctoral program (Joseph, 2012; Lisberg & Woods, 2018; Rudolph et al.,
2019): strong mentor relationships with faculty, valuable research experience, and adjusting to
the academic environment of a research university that can be far different from their
undergraduate environment (Jackson et al., 2016).
Despite these programs effectiveness in recurring and retaining URM students in the
STEM PhD programs, only 27.3% of the survey respondents have heard about such program.
UCSF should tap this potential and establish Masters-to-PhD Bridge programs similar to the
ones described above for their basic and biomedical sciences PhD programs. In such a program,
the most obvious candidates for regional institutions that offer master's degrees that UCSF can
partner with are California State Universities (CSU) campuses. The CSU system is made up of
23 campuses across the state of California, educates nearly half a million students every year,
and has a very diverse student body (CSU, 2021a). Both the CSU system and the University of
California (UC) system to which UCSF belongs to, are public university systems in California
and have many similarities (CSU, 2021a; UC, 2021). UCSF is very close in proximity to five
CSU campuses, and within drivable distance to six more campuses, making a total of eleven
CSU campuses (CSU, 2021a). These CSU campuses are Chico, East Bay, Fresno, Humboldt,
Maritime Academy, Monterey Bay, Sacramento, San Francisco, San Jose, Sonoma, and
Stanislaus. Within these eleven campuses, more than 180,920 students are enrolled (CSU,
2021b). Furthermore, the CSU campuses are MSIs reflected in the 61.6% makeup of the student
body population who are minority students (CSU, 2021c). Therefore, within a drivable distance
is a huge and diverse pool of potential URM STEM PhD applicants. Furthermore, 9% of the
respondents in the study came from a CSU (see Appendix B, Question 6) and there is already a
167

partnership program that exists between UCSF and a CSU campus for a doctorate in Physical
Therapy (Byl, 2007).
Recommendation 3: Provide More Mentoring Opportunities
Providing students with quality mentorship contributes to increased student efficacy,
academic achievement, productivity in scholarship, and academic persistence (Estrada et al.,
2018). Even informal mentoring networks are essential for nurturing minority students (Carter-
Sowell, 2016). On the other hand, a poor mentor relationship is a major factor in high attrition
rates for STEM PhD students (Moreira et al., 2019). The survey responses indicated 85% of the
respondents felt (Xi >=4) that mentorship is a key to their academic success (see Appendix B,
Question 30) and all respondents felt (Xi >=4) that mentorship increases their participation and
persistence (see Appendix B, Question 31). When asked if they had sufficient mentorship
opportunities in their doctoral program (see Appendix B, Question 33), 42.4% of the respondents
were neutral (Xi >=3). This can be an issue as poor or lack of mentorship is a major factor in
high attrition rates amongst doctoral students (Moreira et al., 2019).
The recommendation is for UCSF to create more mentorships for URM students. These
mentorships shouldn’t just be purely academic mentorships though. When students enroll in a
new college environment, they are often faced with social pressures and issues transitioning to
college (Lisberg & Woods, 2018). On entering graduate school, URM students find that they
need to successfully learn about and navigate through new cultures, expectations, environments,
and networks to avoid isolation and be successful in graduate school (Moreira et al., 2019). All
respondents felt (Xi >=4) that acquiring learning strategies in relation to persisting and
succeeding in their PhD program was very important (see Appendices B and E, Question 69), but
when asked if they had sufficient opportunities to acquire learning strategies in their PhD
168

program or in their undergraduate or master's degree only 60.6% agreed (Xi >=4) (see Appendix
B, Question 70). The survey responses indicated 72% of the respondents felt (Xi >=4) that
acquiring study skills in relation to persisting and succeeding in their PhD program was
important, but similarly to Question 70, only 51.5% of the respondents felt (Xi >=4) they had
sufficient opportunities to acquire such study skills (see Appendix B, Question 72). Thus, UCSF
needs to have comprehensive mentorships that help students acquire these “soft skills” that goes
beyond just academic and career counseling (Moreira et al., 2019) and build the skills to focus on
academic norms and behaviors associated with their discipline (Joseph, 2012).
Furthermore, these mentorships with URM students need to involve more URM faculty.
A major factor for URM students’ persistence and academic success is having access to URM
faculty (Joseph, 2012; Whittaker & Montgomery, 2012) and students perform better when their
teachers share the same race and gender (Miller, 2018). URM students have a strong desire to
prefer mentors that are like them (Whitfield & Edwards, 2011) as it is difficult for students to
create a career or scientist identity when there is a lack of role models like themselves that have
‘made it’ (McGee, 2016). When URM students see a successful faculty with a shared ethnic and
cultural background, it signals to the student that they too can be successful (Yadav & Seals,
2019). The survey responses indicated 70% of the respondents agreed that having a mentor that
looks like them (i.e., same ethnic and cultural identity) was important (see Appendix B, Question
34). When asked why that was the case, the three most frequently cited reasons were more
relatable/sense of belonging, more empathy/sympathy, and comfortable going to seek advice (see
Table 16). Less frequently mentioned reasons included alleviating imposter syndrome/stereotype
threat, thinking/caring more about you, shared cultural experience, and being seen as a role
169

model. Racial, cultural, and ethnic differences can also lead to difficulties and challenges with
communication (McGee, 2016).
In a hostile campus climate, URM faculty mentors may need to draw from their own
racial encounters and experiences to counsel URM students who are questioned about their
abilities due to racial stereotyping, being only a one of a handful of URM students in a program,
or feelings about being representatives of their race (Sánchez et al., 2018). The survey responses
indicated 81.8% of the respondents felt (Xi >=4) that negative stereotypes of underrepresented
minorities not having math or science abilities is a major factor that reduces URM students’
sense of belonging to the STEM discipline (see Appendix B, Question 44). The survey responses
indicated that that majority (55%) of the respondents personally encountered these negative
stereotypes (see Appendix B, Question 45). UCSF is similar to other higher education
institutions, where the faculty doesn’t mirror the racial and ethnic makeup of the student body
they serve (Yadav & Seals, 2019). URM make up a little more than 9% of the faculty (UCSF
Office of Diversity and Outreach, 2020c) which is only slightly higher than the national average
of 8.6% (NSF, 2018b). Thus, in order to have successful mentorship, UCSF will also need to
recruit more URM faculty members to serve as valuable mentors to URM students.
Limitations and Delimitations
Limitations
The actual number of participants in the survey was not big enough to reach the targeted
90% confidence level with +/- 10% margin of error (this implies there is a 10% chance the result
happened by accident and the level of certainty is 90%). In order to reach this target, the sample
size needed to be almost 1/3 of the general population, while the actual participation rate was
about 22%. This is not surprising at all, given that in a recent past survey, and on topics and
170

issues that concerned students directly, the participation rate for the target population was also
below 1/3. As the targeted sample size went down, the confidence level also went down and the
margin of error, statistically speaking, increased. This means there is a higher chance
(probability) that the results in the survey happen by accident. Hurley (2000) claims a lower
confidence level can make the results deceptive because this scenario would result in a low
likelihood of covering any actual discrepancy between the sample and the general population.
Furthermore, the external validity may not be met as the results of the study from the sample size
can’t be generalized to the larger population (Johnson & Christensen, 2015; Scandura &
Williams, 2000; Taylor, 2013) especially when the participant size was small (N = 33). Thus,
statements made here shouldn’t be generalized by the reader without further research and
validations. Furthermore, the small participant size meant that the data set was not large enough
to do statistical analysis which would use large amounts of data to discover significance,
correlations, relationship, predictive, and underlying patterns and trends.
The sample population included variables that may skew the results. For example, the
male to female gender ratio of the participants is 1:2 (see Table 6), and the Hispanic/Latino to
African American ethnicity ratio of the participants is close to 2:1 (see Table 7). These gender
and ethnicity ratios might be far from the actual ratio of the URM student population in the basic
and biomedical sciences PhD programs at UCSF or other R1 (very high research activity)
institutions across the nation. Then, there are other variables that could produce different
outcomes such as the year in the program and which STEM PhD program the student is enrolled
in is another factor. A student in their 1st year versus a student in their 8th year will have far
different life experiences and will answer the survey questions differently. Each basic and
biomedical sciences PhD program is run independently resulting in varying levels of support
171

provided to the students and the students would respond to the choices in the questions
accordingly. Also, the student participants in the study resulted in certain programs being either
over or underrepresented by their students relative to the number of actual URM PhD students in
their programs.
Surveys are self-reports from participants about themselves when addressing the survey
questions such as their behaviors, attitudes, perceptions, opinions, and knowledge. How
participants feel others see them or their own perception can be a barrier to accurate responses to
the survey questions (Robinson & Leonard, 2018). This is known as social desirability bias
where participants are reluctant to respond truthfully to anything that might look negative to
others or overly estimate positive behaviors and underestimate negative ones in their responses
(Robinson & Leonard, 2018). Response variance is another problem that arises with participant
responses. This has to do with the inability of human memory to remember things accurately or
exactly as it happens and results in responses to the same question varying over time (Robinson
& Leonard, 2018).
The specific, unique makeup of the student body at UCSF can contribute to survey results
that could be different if the survey was conducted at another R1 university. For example, of the
PhD students that are not foreign citizens, about 22% are URM (UCSF Graduate Division,
2020a) and much higher than the national average of 7.9% (NSF, 2017a). The higher percentage
of URM students in the programs could be a reason why participants felt a strong sense of
belonging in their program (see Appendices B and C, Question 52), yet feel that in general, there
is loneliness and isolation being an underrepresented minority student in a STEM PhD program
(see Appendices B and C, Question 58). Another example is that 56% of the students in the basic
172

and biomedical sciences PhD student body at UCSF (UCSF Graduate Division, 2020a) are
females as compared to the national average of 47% (NSF, 2017d).
Delimitations
In quantitative research, there is a need to generalize the results of the study to go beyond
the limits of the study itself. In order to do so, it is important that the results of the study are
reliable, and inferences made from the results are valid (Johnson & Christensen, 2015). Research
reliability means that when the study is conducted again, the same results will be obtained
(Johnson & Christensen, 2015). In an ideal environment where the researcher had unlimited
time, resources, and the necessary contacts, the researcher would administer the same Qualtrics©
survey to URM students in the same PhD programs at different institutions across the nation with
the same “very high research activity” doctoral-granting university or R1 classification by the
Carnegie Classification (Carnegie Classification of Institutions of Higher Education by Indiana
University Center for Postsecondary Research, 2017). Obtaining similar results would give the
researcher greater confidence in the elimination of extraneous variables that might influence the
dependent variables or hamper the ability of the researcher to generalize the results (Johnson &
Christensen, 2015). For example, what if the Likert 5-point or 6-point scale (Wuensch, 2015)
closed-end questions related to campus environment scored higher for STEM PhD students at
UCSF than R1 campuses that are in a less diverse, and non-urban setting? Could the extraneous
variable be that UCSF is located in a dense and diverse, minority-majority city (World
Population Review, 2020)?
When formulating questions in the survey, the researcher can be vague or ambiguous
with the word or group of words in the question sentence(s). One of the reasons this can arise is
the researcher brings his own assumptions, worldview, biases, and theoretical orientation into the
173

design of the survey questions. For example, a researcher with a long and solid experience in IT
might use the term “systems analysts” to mean someone who analyzes, designs, and implements
IT systems while a survey participant with an aerospace background might take it as meaning
any systems such as an airplane. Research has shown that vagueness and ambiguity in the
questions correlate with lower response quality and increased measurement error in the collected
data (Johnson & Christensen, 2015).
The researcher was advised by his committee from their experience and from the
literature (Marcus et al., 2007) to limit the length of survey due to survey exhaustion by the
participants. Therefore, the researcher reduced the set of questions in the survey to eighty-four
questions which was still a fairly large number even though seven-eighths of the questions were
closed end questions. The researcher felt the overall length of the survey might have led to a
degradation of the data quality of the responses (i.e., simply answering the question without
given it too much thought) for the seventy-four closed-ended questions that were all answered
and abandonment and no responses to some of the twelve open-ended questions. Also, the
decision of the researcher to use a 6-point rather than a 5-point scale for the Likert type closed
end questions (Crewsell, 2009) would yield different survey results.
Creswell (2009) touched upon the importance of reciprocity to the participants who will
need to spend time filling out the survey. The researcher believes reciprocity is necessary and
that in order to really get 1/3 of the general population to participate in the survey (to get to the
90% confidence level with +/- 10% margin of error), a small reward might need to be part of the
incentive. A small reward can increase participation but at the same time, the participants might
just be simply filling out the survey as quickly as possible in order to obtain the reward and are
not, in fact, committed to filling out the survey as accurately as possible.
174

The participants in this study are URM students in STEM PhD programs. NSF defines
URM (underrepresented minorities) as Hispanic, Native American, and African American
(SACNAS Harvard Chapter, n.d.). Most of the literature out there including the ones the
researcher has reviewed for Chapter Two focus on URM students at large rather than trying to
analyze the potentially different life experiences of Hispanic versus African American versus
Native American PhD students in their STEM pipeline pathway. The researcher has concerns
that there might be subtle, or even major differences in life experience, culture, or perspectives
that differ between the three groups that are statistically significant but not identified or
accounted for in the data collection or analysis. Furthermore, compared to their Hispanic and
African American counterparts, there is not as much research done for Native American STEM
PhD students. The researcher does feel there is some oversimplifying and generalizing (i.e.,
using a one size fits all) of treating the three groups similarly in this research topic, especially as
each of these three groups encompass many different cultures, national origins, history, and
heritage. Oversimplifying and generalizing can have negative consequences and incorrect
generalizations, taken as fact, being made to the general population being targeted. This is
already happening in clinical drug trials. Several FDA drug trials that were proven to be safe and
effective for a majority of participants who were white, were later found not to be as effective
with minority populations (Pérez-Stable, 2018). The FDA (PCORnet, 2018), NIH, and federal
law (LII, n.d.) requires that women and minorities be included in all clinical research studies as
“insufficient data from minority populations can result in medicine that is increasingly precise
for certain segments of the United States population” (Sierra-Mercado & Lázaro-Muñoz, 2018,
p. 44). The same can be said of this study.
175

Recommendations for Future Research
Further research should focus on whether the results and findings from this study are
unique only to UCSF or would be similar at other very high research doctoral-granting, R1
universities as well. UCSF is one of ten campuses in the University of California (UC) system
(UC, 2021). There are nine other UC campuses and eight of them, like UCSF are classified as a
very high research doctoral-granting university so it would be a good start to apply the same
study on these eight UC campuses and compare the findings and results. For example, validating
the responses from participants based on their experience and knowledge on whether URM PhD
students, in general, were not given enough opportunities for mentorship, presenting and
attending at conferences, and publishing papers (see Table 38). An even broader study would be
to apply the same study for private R1 universities to see if the outcome is the same given that
private and public research universities have a different operating model, mission, and
accountability (Feeney & Welch, 2012; Sav, 2020). The broader study will allow the ability to
increase the participant size significantly and create a large enough data set to do statistical
analysis which would allow large amounts of data to discover significance, correlations,
relationship, predictive, and underlying patterns and trends.
Several questions in the survey were asked if the respondents acquire sufficient skills,
experience, and knowledge, along with adequate career and academic guidance at their
undergraduate institutions (see Appendix B, Questions 8, 9, 60, 64, 65, 70 and 72). Completing
an undergraduate degree in an R1 university means the student is exposed to higher academic
standards and rigors, has more experience with research, and access to faculty coming from a
strong research background than students attending a lower tier ranking university (Hollman et
al., 2018). The survey data conferred with the literature as 70% of the respondents who obtained
176

their undergraduate degree from non-R1 universities thought their undergraduate program
preparation was inadequate for their PhD program or found the PhD program was quite difficult
and rigorous as compared to their undergraduate degree (see Table 13). This contrasts with only
16.7% of those attending an R1 university feeling the same way (see Table 12). Therefore, the
study should expand and include surveying current URM undergraduate students attending non-
R1 universities versus those attending R1 universities and using Clark and Estes' (2008) gap
analysis framework to focus on closing this performance gap by looking at knowledge,
motivation, and organizational (KMO) causes.
The research methodology used a quantitative approach in the form of a Qualtrics©
survey. Future research should follow up with a qualitative approach by evaluating the responses
from this survey to develop the approach and content for in-depth interviews with participants to
gain greater depth and richness of information not possible with just the quantitative data alone.
The problem with designing any study is the researcher bases the design on the knowledge he
has at hand at the time of the study. It is only after the data has been collected and the data
analysis is done, is when the researcher can see areas that warrant deeper and further
investigation through a qualitative approach. For example, in question 23 of the survey,
responses were asked to rate (5-point scale; 5 = Strongly Agree, 1 = Strongly Disagree) if
research opportunities were very important for them to persist and to succeed in their PhD
program. The respondents strongly agreed to the statement (N = 33, µ = 4.79, σ = 0.41, min =
4.00, max = 5.00, and mode = 5.00), but participants did not have any opportunities to say why
that was case and if their reasons, if they responded, would be any different from what is
mentioned in the literature. The detailed reasons that participants provided would have further
the research in this area greatly.
177

In the traditionally white and Asian, male-dominated STEM PhD programs, the survey’s
participants were two-thirds female and URM. Their survey responses provide a rare insight into
the intersectionality of being female and URM. Future research should focus on this often-
neglected segment of the student population in order to garner additional insights useful in
further increasing their representation in the STEM PhD programs.
Most studies, including this one, tend to lump Hispanic, Native American, and African
American students into one large group (i.e., URM students) and without much regard to
differences in social, historical, economic, cultural, gender, and ethnic backgrounds. Future
research in this area needs to segment the URM STEM PhD students to garner insights and
experiences that might be unique to a specific group. For example, the experiences of native-
born versus foreign-born Hispanic STEM PhD students or between African American and Native
Americans STEM PhD students could be very different. Similarly, along this line, Asians are
overrepresented in STEM relative to their population (Chen & Buell, 2018), but there are Asian
subgroups that aren’t faring well and in fact are underrepresented (Choi, 2008); future research
should also be done on these long-neglected Asian subgroups.
Conclusion
Grim racial economic inequalities in earnings and employment caused by decades of
racial disparities in higher education (Economist, 2020; U.S. Department of Education, 2016) has
led to an explosive rise in social justice protests in recent years. The time has come for higher
education institutions to seriously address these educational disparities especially in STEM PhD
where the disparity is the greatest. URM makes up 36.8% of America’s college-age (18-24)
population (NSF, 2017b), but only received a scant 7.9% of all PhD degrees in the science and
engineering fields (NSF, 2017a).
178

From Clark and Estes' (2008) gap analysis framework and the review of literature,
fourteen KMO influences impacting URM students in STEM PhD programs were formulated.
These influences were categorized and organized around organizational changes, theories of
expectancy-value and self-efficacy for motivation, and declarative, procedural, and
metacognitive knowledge. The findings validated the fourteen assumed KMO influences. From
the discussion of the findings, the researcher devised a handful of recommendations, and three
were chosen based on the following criteria of the practicality of the recommendation,
recommendation fits well within the context and culture of UCSF’s organization, and the
recommendation delivers the most benefits for URM STEM PhD students (based on the review
of literature and survey data).
At this time, the United States is still the world’s most innovative and scientifically
advanced economy due to its strength in advanced STEM education (Han et al., 2015). At the
same time, the U.S. population is undergoing a vast demographic shift where non-Hispanic
Whites are projected to compromise less than 50% of the U.S. population by mid-century (Craig
& Richeson, 2017). As the U.S. population becomes more diverse (Frey, 2014), making STEM
doctoral programs equally accessible to all Americans, particularly URM, must be part of our
national dialogue if the nation is to remain globally competitive and equitable.

179

References
Alba, R. (2018). What Majority-minority Society? A Critical Analysis of the Census Bureau’s
Projections of America’s Demographic Future.
Socius. https://doi.org/10.1177/2378023118796932
Allen, B., & Garg, K. (2016). Diversity Matters in Academic Radiology: Acknowledging and
Addressing Unconscious Bias. Journal of the American College of Radiology, 13(12),
1426–1432. https://doi.org/10.1016/j.jacr.2016.08.016
Allen-Ramdial, S.-A. A., & Campbell, A. G. (2014). Reimagining the Pipeline: Advancing
STEM Diversity, Persistence, and Success. Bioscience, 64(7), 612–618.
https://doi.org/10.1093/biosci/biu076
Alvarez, J. (2020, April 1). UCSF Remains Top Public Recipient of NIH Funding for 13th
Straight Year. University of California, San Francisco.
https://www.ucsf.edu/news/2020/04/417061/ucsf-remains-top-public-recipient-nih-
funding-13th-straight-year
American Physical Society Physics. (2020). Doctoral Degrees Earned by Women, by Major.
https://www.aps.org/programs/education/statistics/fraction-phd.cfm
Andersen, L., & Ward, T. (2014). Expectancy-Value Models for the STEM Persistence
Plans of Ninth-Grade, High-Ability Students: A Comparison between Black, Hispanic,
and White Students. Science Education (Salem, Mass.), 98(2), 216–242.
https://doi.org/10.1002/sce.21092
Anderson, L., Krathwohl, D., & Bloom, B. (2001). A taxonomy for learning, teaching, and
180

assessing: a revision of Bloom’s taxonomy of educational objectives / editors, Lorin W.
Anderson, David Krathwohl; contributors, Peter W. Airasian ... [et al.]. (Complete ed.).
Longman.
Anneke Lisberg, & Brett Woods. (2018). Mentorship, Mindset and Learning Strategies: An
Integrative Approach to Increasing Underrepresented Minority Student Retention in a
Stem Undergraduate Program. Journal of STEM Education, 19(3), 14–20.
Antón, S., Malhi, R., & Fuentes, A. (2018). Race and diversity in U.S. Biological Anthropology:
A decade of AAPA initiatives. American Journal of Physical Anthropology, 165, 158–
180. https://doi.org/10.1002/ajpa.23382
Arcidiacono, P., Aucejo, E., & Hotz, V. (2016). University Differences in the Graduation of
Minorities in STEM Fields: Evidence from California. The American Economic
Review, 106(3), 525–562. https://doi.org/10.1257/aer.20130626
Arnab, R. (2017). Survey Sampling Theory and Applications. In Survey Sampling Theory
and Applications. Elsevier Science & Technology.
Arocho, O. (2017). Understanding Latina Doctoral Student Experiences: Negotiating Ethnic
Identity and Academic Success. Seton Hall University.
https://scholarship.shu.edu/cgi/viewcontent.cgi?article=3346&context=dissertations/
Ashcraft, C. (2009). The 'Leaky Pipe' of Diversity. National Center for Women & Information
Technology. https://synapse.ucsf.edu/articles/2019/05/04/leaky-pipe-diversity
Association of American Medical Colleges (AAMC). (2017). Proceedings of the Diversity and
Inclusion Innovation Forum: Unconscious Bias in Academic Medicine.
https://store.aamc.org/downloadable/download/sample/sample_id/168/
Association of American Medical Colleges (AAMC). (2020). MCAT Scores and GPAs for
181

Applicants and Matriculants to U.S. Medical Schools, 2018-2019 through 2019-2020.
https://www.aamc.org/data-reports/students-residents/interactive-data/2019-facts-
applicants-and-matriculants-data
Azmitia, M., Syed, M., & Radmacher, K. (2013). Finding Your Niche: Identity and Emotional
Support in Emerging Adults’ Adjustment to the Transition to College. Journal of
Research on Adolescence, 23(4), 744–761. https://doi.org/10.1111/jora.12037
Baber, L. (2015). Considering the Interest-Convergence Dilemma in STEM
Education. Review of Higher Education, 38(2), 251–270.
https://doi.org/10.1353/rhe.2015.0004
Bandura, A. (1977). Self-efficacy: toward a unifying theory of behavioral change. Psychological
Review, 84(2), 191–215. https://doi.org/10.1037//0033-295X.84.2.191
Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory.
Prentice-Hall.
Beasley MA, Fischer MJ. Why they leave: The impact of stereotype threat on the attrition of
women and minorities from science, math and engineering majors. Soc Psychol
Educ. 2012; 15:427–448.
Ben-Zeev, A., Paluy, Y., Milless, K. L., Goldstein, E. J., Wallace, L., Márquez-Magaña, L.,
Bibbins-Domingo, K., & Estrada, M. (2017). 'Speaking Truth' Protects Underrepresented
Minorities' Intellectual Performance and Safety in STEM. Education sciences, 7(2), 65.
https://doi.org/10.3390/educsci7020065
Benderly, B. (2015). How Not to Attract Minorities to STEM. Science.
https://www.sciencemag.org/careers/2012/12/how-not-attract-minorities-stem
Berger, J., & Milem, J. (2000). Exploring the Impact of Historically Black Colleges in Promoting
182

the Development of Undergraduates’ Self-Concept. Journal of College Student
Development, 41(4), 381–.
Bernard, R., & Cooperdock, E. (2018). No progress on diversity in 40 years. Nature
Geoscience, 11(5), 292–295. https://doi.org/10.1038/s41561-018-0116-6
Bourke, B.N. (2016). Meaning and Implications of Being Labelled a Predominantly White
Institution. College and University, 91, 12.
Brittian, A., Sy, S., & Stokes, J. (2009). Mentoring: Implications for African American College
Students. The Western Journal of Black Studies, 33(2), 87–97.
Brown, E. (2016, March 29). The overwhelming whiteness of U.S. private schools, in six
maps and charts (Posted 2016-03-29 11:05:11). The Washington Post.
Brown, P. (2013). Loneliness at the bench. Is the PhD experience as emotionally taxing as
it is mentally challenging? EMBO Reports, 14(5), 405–409.
https://doi.org/10.1038/embor.2013.35
Brown, M., & Davis, J. (2001). The Historically Black College as Social Contract, Social
Capital, and Social Equalizer. Peabody Journal of Education, 76(1), 31–49.
https://doi.org/10.1207/s15327930pje7601_03
Burke, L. (2020, June 11). Scientists strike for black lives, a more inclusive academia. Inside
Higher Ed. https://www.insidehighered.com/news/2020/06/11/scientists-strike-black-
lives-more-inclusive-academia
Burt, B., Williams, K., & Smith, W. (2018). Into the Storm: Ecological and Sociological
Impediments to Black Males’ Persistence in Engineering Graduate Programs. American
Educational Research Journal, 55(5), 965–1006.
https://doi.org/10.3102/0002831218763587
183

Burton, G., Kennedy, E., & Vicente, M. (2019). Results from a 14-Year Intervention Program
Designed to Impact Pursuit of a PhD in Research among Underrepresented Students in
STEM Disciplines. Journal for STEM Education Research, 2(2), 128–153.
https://doi.org/10.1007/s41979-019-00019-6
Buswell, N. (2017). Swimming Upstream: Pathways of New Engineering Faculty at Non-R1
Institutions. ProQuest Dissertations Publishing.
Byl, Nancy (2007). Graduate Programs in Physical Therapy.
http://calstate.edu/app/jgb/jgb-restricted/FresnoToWASC_Byl.pdf
California State University Office of the Chancellor (CSU). (2021a). About the CSU.
https://www2.calstate.edu/csu-system/about-the-csu
California State University Office of the Chancellor (CSU). (2021b). Enrollment.
https://www2.calstate.edu/csu-system/about-the-csu/facts-about-the-csu/enrollment
California State University Office of the Chancellor (CSU). (2021c). Fall 2018 Enrollment
Demographics.
https://www2.calstate.edu/csu-system/about-the-csu/facts-about-the-
csu/enrollment/Pages/fall-2018-enrollment.aspx
Callahan CN, LaDue ND, Baber LD, Sexton J, van der Hoeven Kraft KJ, Zamani-Gallaher EM.
Theoretical Perspectives on Increasing Recruitment and Retention of underrepresented
Students in the Geosciences. Journal of geoscience education. 2018;65(4):563-576.
doi:10.5408/16-238.1
Carnegie Classification of Institutions of Higher Education by Indiana University Center for
Postsecondary Research. (2017). Basic Classification Description.
https://carnegieclassifications.iu.edu/classification_descriptions/basic.php
184

Carter-Sowell, A. R., Dickens, D. D., Miller, G., & Zimmerman, C. A. (2016). Present but not
accounted for: examining how marginalized intersectional identities create a double bind
for women of color in the academy. In J. Ballenger, B. Polnick, & B. Irby (Eds.), Women
of Color in STEM: Navigating the Workforce in Research on Women and Education
series (pp. 181–200). Charlotte: Information Age Publishing.
Chemers, M., Hu, L., & Garcia, B. (2001). Academic self-efficacy and first year college student
performance and adjustment. Journal of Educational Psychology, 93(1), 55–64.
https://doi.org/10.1037//0022-0663.93.1.55
Chemers, M., Zurbriggen, E., Syed, M., Goza, B., & Bearman, S. (2011). The Role of Efficacy
and Identity in Science Career Commitment Among Underrepresented Minority
Students. Journal of Social Issues, 67(3), 469–491. https://doi.org/10.1111/j.1540-
4560.2011.01710.x
Chen, G. A., & Buell, J. Y. (2018). Of models and myths: Asian (Americans) in STEM and the
neoliberal racial project. Race, Ethnicity and Education, 21(5), 607–625.
https://doi.org/10.1080/13613324.2017.1377170
Chen, S. (2018). A Flawed System: The Burden of Service Among Faculty of Color. The Amherst
Student. https://amherststudent.com/article/a-flawed-system-the-burden-of-service-
among-faculty-of-color
Cherng, H.-Y. S., & Halpin, P. F. (2016). The Importance of Minority Teachers: Student
Perceptions of Minority Versus White Teachers. Educational Researcher, 45(7), 407–
420. https://doi.org/10.3102/0013189X16671718
Chloe Lancaster, Yonghong Jade Xu. Challenges and Supports for African American STEM
185

Student Persistence: A Case Study at a Racially Diverse Four-Year Institution. The
Journal of Negro education. 2017;86(2):176-189. doi:10.7709/jnegroeducation.86.2.0176
Choi, Y. (2008). Diversity within: subgroup differences of youth problem behaviors among
Asian Pacific Islander American adolescents. Journal of Community Psychology, 36(3),
352–370. https://doi.org/10.1002/jcop.20196
Choy, S., Horn, L., Nuñez, A., & Chen, X. (2000). Transition to College: What Helps At-Risk
Students and Students Whose Parents Did Not Attend College. New Directions for
Institutional Research, 2000(107), 45–63. https://doi.org/10.1002/ir.10704
Christophers, B., & Gotian, R. (2019). Using admission statistics to encourage diverse applicants
to MD-PhD programs. The Journal of Clinical Investigation, 130(1), 17–19.
https://doi.org/10.1172/jci134941
Clark, R. E., & Estes, F. (2008). Turning research into results: A guide to selecting the right
performance solutions. Charlotte, NC: Information Age Publishing, Inc.
Cole, D., & Espinoza, A. (2008). Examining the academic success of Latino students in
science technology engineering and mathematics (STEM) majors. Journal of College
Student Development, 49(4), 285–300.
Collier, P. J., & Morgan, D. L. (2008). “Is That Paper Really Due Today?”: Differences in First-
Generation and Traditional College Students’ Understandings of Faculty
Expectations. Higher Education, 55(4), 425–446. https://doi.org/10.1007/s10734-007-
9065-5
Collins, K. H. (2018). Confronting Color-Blind STEM Talent Development: Toward a
Contextual Model for Black Student STEM Identity. Journal of Advanced Academics,
29(2), 143–168. https://doi.org/10.1177/1932202X18757958
186

Columbia University. (2018). Policy and Planning Committee Equity Reports. Columbia
University. https://fas.columbia.edu/files/fas/content/Columbia-ArtsandSciences-PPC-
Equity-Reports-2018.pdf
Comeaux, E. (2013). Faculty Perceptions of High-Achieving Male Collegians: A Critical
Race Theory Analysis. Journal of College Student Development, 54(5), 453–465.
https://doi.org/10.1353/csd.2013.0071
Cooper KM, Gin LE, Akeeh B, et al. Factors that predict life sciences student persistence in
undergraduate research experiences. PloS one. 2019; 14(8):e0220186-e0220186.
doi:10.1371/journal.pone.0220186
Council of Graduate Schools. (2020). Seven-year URM Doctoral Completion and Attrition Rates
by Academic-year Groups and Selected Student Characteristics, DIMAC Project.
https://cgsnet.org/seven-year-urm-doctoral-completion-and-attrition-rates-academic-year-
groups-and-selected-student
Craig, M., & Richeson, J. (2017). Information about the US racial demographic shift triggers
concerns about anti-White discrimination among the prospective White “minority.” PloS
One, 12(9), e0185389–e0185389. https://doi.org/10.1371/journal.pone.0185389
Crawford, L.M. (2019). Conceptual and Theoretical Frameworks in Research. In Burkholder,
G.J., Cox, K.A., Crawford, L.M., Hitchcock, J.H. (Eds.), Research Design
and Methods: An Applied Guide for the Scholar-Practitioner. (pp. 35-48). SAGE
Publications.
Creswell, J. (2009). Research design: qualitative, quantitative, and mixed methods approaches
(3rd ed.). Sage.
Deci, E.L. and Ryan, R.M. (1991), “A motivational approach to self: integration in personality”,
187

in Dienstbier, R. (Ed.), Nebraska Symposium on Motivation: Perspectives on Motivation,
University of Nebraska Press, Lincoln, NE, Vol. 38, pp. 237-288.
Granovskiy, B. (2018, June 12). Science, Technology, Engineering, and
Mathematics (STEM) Education: An Overview. Congressional Research Service.
https://fas.org/sgp/crs/misc/R45223.pdf
Eby, L. T., Allen, T. D., Hoffman, B. J., Baranik, L. E., Sauer, J. B., Baldwin, S., ... Evans, S. C.
(2013). An interdisciplinary meta-analysis of the potential antecedents, correlates, and
consequences of protégé perceptions of mentoring. Psychological Bulletin, 139(2), 441–
476. doi: 10.1037/a0029279
Eccles, J., & Barber, B. (1999). Student Council, Volunteering, Basketball, or Marching Band:
What Kind of Extracurricular Involvement Matters? Journal of Adolescent
Research, 14(1), 10–43. https://doi.org/10.1177/0743558499141003
Economist. (2020, June 3). The grim racial inequalities behind America’s protests.
https://www.economist.com/united-states/2020/06/03/the-grim-racial-inequalities-
behind-americas-protests
Ellis, E. (2001). The impact of race and gender on graduate school socialization, satisfaction with
doctoral study, and commitment to degree completion. The Western Journal of Black
Studies, 25(1), 30–45.
Estrada M, Hernandez PR, Schultz PW. A Longitudinal Study of How Quality Mentorship and
Research Experience Integrate Underrepresented Minorities into STEM Careers. CBE life
sciences education. 2018;17(1):ar9-. doi:10.1187/cbe.17-04-0066
Fan, SC., Yu, KC. How an integrative STEM curriculum can benefit students in engineering
188

design practices. Int J Technol Des Educ 27, 107–129 (2017).
https://doi.org/10.1007/s10798-015-9328-x
Fayer, S., Lacey, A., & Watson, A. (2017). STEM Occupations: Past, Present, and Future.
U.S. BUREAU OF LABOR STATISTICS. https://www.bls.gov/spotlight/2017/science-
technology-engineering-and-mathematics-stem-occupations-past-present-and-
future/pdf/science-technology-engineering-and-mathematics-stem-occupations-past-
present-and-future.pdf
Feeney, M. K., Welch, E. W. (2012). Realized Publicness at Public and Private Research
Universities. Public Administration Review, 72(2), 272–284.
https://doi.org/10.1111/j.1540-6210.2011.02521.x
Frey, W. (2014). Diversity Explosion: How New Racial Demographics are Remaking America.
In Diversity Explosion. Brookings Institution Press. https://doi.org/10.7864/j.ctt1vjqnns
Fries-Britt, S., & Turner, B. (2002). Uneven Stories: Successful Black Collegians at a Black and
a White Campus. Review of Higher Education, 25(3), 315–330.
https://doi.org/10.1353/rhe.2002.0012
Gardner, L. (2015). Helping Minority Ph.D.'s in STEM: Something's Working. The Chronicle of
Higher Education. https://www.chronicle.com/article/Helping-Minority-PhDs-in/229323
Gibau, G. S. (2015). Considering student voices: Examining the experiences of underrepresented
students in intervention programs. CBE Life Sciences Education, 14(3), ar28–.
https://doi.org/10.1187/cbe.14-06-0103
Gibbs, K. D., McGready, J., & Griffin, K. (2015). Career development among American
biomedical postdocs. CBE Life Sciences Education, 14(4), ar44–ar44.
https://doi.org/10.1187/cbe.15-03-0075
189

Grant, C., & Osanloo, A. (2014). Understanding, Selecting, and Integrating a Theoretical
Framework in Dissertation Research: Creating the Blueprint for Your “House.”
Administrative Issues Journal: Education, Practice, and Research, 4(2).
https://doi.org/10.5929/2014.4.2.9
Greenhalgh, T., Seyan, K., & Boynton, P. (2004). “Not a university type”: focus group study of
social class, ethnic, and sex differences in school pupils’ perceptions about medical
school. BMJ, 328(7455), 1541–1544. https://doi.org/10.1136/bmj.328.7455.1541
Grimsley-Vaz, E. (2018). What I Learned From the 1st Black Woman to Earn an Engineering
PhD in Oklahoma. The Moguldom Nation. https://moguldom.com/143508/what-i-
learned-from-the-1st-black-woman-engineering-phd-in-oklahoma/
Guy, B., & Boards, A. (2019). A seat at the table: Exploring the experiences of underrepresented
minority women in STEM graduate programs. Journal of Prevention & Intervention in
the Community: Social Inequality: Impact on Schools, Youth and Families, 47(4), 354–
365. https://doi.org/10.1080/10852352.2019.1617383
Hall, J. D., O’Connell, A. B., & Cook, J. G. (2017). Predictors of student productivity in
biomedical graduate school applications. PloS One, 12(1), e0169121–e0169121.
https://doi.org/10.1371/journal.pone.0169121
Han, X., Stocking, G., Gebbie, M., & Appelbaum, R. (2015). Will they stay or will they go?
International graduate students and their decisions to stay or leave the U.S. upon
graduation. PloS One, 10(3), e0118183–e0118183.
https://doi.org/10.1371/journal.pone.0118183
Harmon, A. (2019). For a Black Mathematician, What It’s Like to Be the ‘Only One’. New York
Times. https://www.nytimes.com/2019/02/18/us/edray-goins-black-mathematicians.html
190

Hassouneh, D., Lutz, K. F., Beckett, A. K., Junkins, E. P., & Horton, L. L. (2014). The
experiences of underrepresented minority faculty in schools of medicine. Medical
Education Online, 19(1), 24768–24768. https://doi.org/10.3402/meo.v19.24768
Hausmann, L., Schofield, J., & Woods, R. (2007). Sense of Belonging as a Predictor of
Intentions to Persist Among African American and White First-Year College Students.
Research in Higher Education, 48(7), 803–839. https://doi.org/10.1007/s11162-007-
9052-9
Holland, J., Major, D., & Orvis, K. (2012). Understanding How Peer Mentoring and
Capitalization Link STEM Students to Their Majors. The Career Development
Quarterly, 60(4), 343–354. https://doi.org/10.1002/j.2161-0045.2012.00026.x
Hollman, A., Bice, M., Ball, J., Bickford, N., Shafer, A., & Bickford, S. (2018). A comparison of
scholarly productivity among current professors who obtained terminal
degrees. American Journal of Distance Education, 32(4), 267–282.
https://doi.org/10.1080/08923647.2018.1511514
Hurley, P. (2000). A concise introduction to logic (7th ed.). Wadsworth/Thomson Learning.
Hurtado, S., & Alvarado, A. R. (2015). Discrimination and Bias, Underrepresentation, and
Sense of Belonging on Campus. University of California, Los Angeles.
https://www.heri.ucla.edu/PDFs/Discriminination-and-Bias-Underrepresentation-and-
Sense-of-Belonging-on-Campus.pdf
Hurtado, S., & Carter, D. (1997). Effects of College Transition and Perceptions of the Campus
Racial Climate on Latino College Students’ Sense of Belonging. Sociology of Education,
70(4), 324–345. https://doi.org/10.2307/2673270
Hurtado, S., Eagan, M., Tran, M., Newman, C., Chang, M., & Velasco, P. (2011). “We Do
191

Science Here”: Underrepresented Students’ Interactions with Faculty in Different College
Contexts. Journal of Social Issues, 67(3), 553–579. https://doi.org/10.1111/j.1540-
4560.2011.01714.x
IBM. (2020). IBM SPSS Statistics. https://www.ibm.com/products/spss-statistics
Insider Higher Ed. (2015). The STEM Pipeline. https://www.bu.edu/stem/files/2015/02/The-
STEM-Pipeline-booklet1.3-36.pdf
Jackson KM, Winfield LL. Realigning the Crooked Room: Spelman Claims a Space for African
American Women in STEM. Peer review: emerging trends and key debates in
undergraduate education. 2014;16(2):9-12.
Jackson, M., Galvez, G., Landa, I., Buonora, P., & Thoman, D. (2016). Science That Matters:
The Importance of a Cultural Connection in Underrepresented Students’ Science
Pursuit. CBE Life Sciences Education, 15(3), ar42–. https://doi.org/10.1187/cbe.16-01-
0067
Johns M, Schmader T, Martens A. Knowing is half the battle: teaching stereotype threat as a
means of improving women's math performance. Psychol Sci. 2005;16(3):175-179.
doi:10.1111/j.0956-7976.2005.00799.x
Johnson, D. R., Soldner, M., Leonard, J. B., Alvarez, P., Inkelas, K. K., Rowan-Kenyon, H., &
Longerbeam, S. (2007). Examining sense of belonging among first-year undergraduates
from different racial/ethnic groups. Journal of College Student Development, 48, 525–
542. http://dx.doi.org/10.1353/csd.2007.0054
Johnson, M. (2019). mSphere of Influence: Hiring of underrepresented Minority Assistant
Professors in Medical School Basic and biomedical sciences Departments Has a Long
Way To Go. mSphere, 4(5), e00599–19–. https://doi.org/10.1128/msphere.00599-19
192

Johnson, R. B. and Christensen, L. B. (2015). Educational Research: Quantitative, qualitative,
and mixed approaches. (5th ed.). Thousand Oaks: SAGE.
Joseph, J. (2012). From One Culture to Another: Years One and Two of Graduate School for
African American Women in the STEM Fields. International Journal of Doctoral
Studies, 7, 125–142. https://doi.org/10.28945/1571
Judge, T., & Bono, J. (2001). Relationship of Core Self-Evaluations Traits-Self-Esteem,
Generalized Self-Efficacy, Locus of Control, and Emotional Stability-With Job
Satisfaction and Job Performance: A Meta-Analysis. Journal of Applied
Psychology, 86(1), 80–92. https://doi.org/10.1037/0021-9010.86.1.80
Katz, I., Eilot, K., & Nevo, N. (2014). “I’ll do it later”: Type of motivation, self-efficacy and
homework procrastination. Motivation and Emotion, 38(1), 111–119.
https://doi.org/10.1007/s11031-013-9366-1
Kezar, A., & Holcombe, E. (2020). The Role of Collaboration in Integrated Programs Aimed at
Supporting Underrepresented Student Success in STEM. American Behavioral
Scientist, 64(3), 325–348. https://doi.org/10.1177/0002764219869421
Kim, M.M., & Conrad C.F. (2006). The Impact of Historically Black Colleges and Universities
on the Academic Success of African American Students. Research in Higher
Education, 47(4), 399–427. https://doi.org/10.1007/s11162-005-9001-4
Kim‐Prieto, C., Copeland, H., Hopson, R., Simmons, T., & Leibowitz, M. (2013). The role of
professional identity in graduate school success for under‐represented minority
students. Biochemistry and Molecular Biology Education, 41(2), 70–75.
https://doi.org/10.1002/bmb.20673
Klassen, R., Krawchuk, L., & Rajani, S. (2008). Academic procrastination of undergraduates:
193

Low self-efficacy to self-regulate predicts higher levels of procrastination. Contemporary
Educational Psychology, 33(4), 915–931. https://doi.org/10.1016/j.cedpsych.2007.07.001
Kong, X., Chakraverty, D., Jeffe, D., Andriole, D., Wathington, H., & Tai, R. (2013). How Do
Interaction Experiences Influence Doctoral Students’ Academic Pursuits in Biomedical
Research? Bulletin of Science, Technology & Society, 33(3-4), 76–84.
https://doi.org/10.1177/0270467613516754
Kraft, S. A., Cho, M. K., Gillespie, K., Halley, M., Varsava, N., Ormond, K. E., Luft, H. S.,
Wilfond, B. S., & Soo-Jin Lee, S. (2018). Beyond Consent: Building Trusting
Relationships With Diverse Populations in Precision Medicine Research. The American
journal of bioethics: AJOB, 18(4), 3–20. https://doi.org/10.1080/15265161.2018.1431322
Kreutzer, C., & Gotto, M. (2015). A Qualitative Study of Influences on Graduate Students’
Decision-Making When Incurring Graduate School Debt. Journal of the Academy of
Nutrition and Dietetics, 115(9), A12–A12. https://doi.org/10.1016/j.jand.2015.06.029
Laura Perna, Valerie Lundy-Wagner, Noah D. Drezner, Marybeth Gasman, Susan Yoon,
Enakshi Bose, & Shannon Gary. (2009). The Contribution of HBCUS to the Preparation
of African American Women for Stem Careers: A Case Study. Research in Higher
Education, 50(1), 1–23. https://doi.org/10.1007/s11162-008-9110-y
Laveist, T. A., & Nuru-Jeter, A. (2002). Is Doctor-Patient Race Concordance Associated with
Greater Satisfaction with Care? Journal of Health and Social Behavior, 43(3), 296–306.
https://doi.org/10.2307/3090205
Lee, M. J., Collins, J. D., Harwood, S. A., Mendenhall, R., & Huntt, M. B. (2020). “If you aren’t
194

White, Asian or Indian, you aren’t an engineer”: racial microaggressions in STEM
education. International Journal of STEM Education, 7(1), 1–16.
https://doi.org/10.1186/s40594-020-00241-4
Legal Information Institute (LII). (n.d.). 42 U.S. Code § 289a–2 - Inclusion of women and
minorities in clinical research. https://www.law.cornell.edu/uscode/text/42/289a-2
Lewis, B. (2005). African American Students’ Career Considerations and Reasons for Enrolling
in Advanced Science Courses. The Negro Educational Review, 56(2-3), 221–231.
Limbong, A. (2020). Microaggressions Are A Big Deal: How To Talk Them Out And When To
Walk Away. NPR. https://www.npr.org/2020/06/08/872371063/microaggressions-are-a-
big-deal-how-to-talk-them-out-and-when-to-walk-away
London, J., Farmer Cox, M., Ahn, B., Branch, S., Zephirin, T., Torres-Ayala, A., & Zhu, J.
(2014). Motivations for Pursuing an Engineering PhD and Perceptions of its Added
Value: A U.S.-based Study. International Journal of Doctoral Studies, 9, 205–227.
https://doi.org/10.28945/2050
Lumen Learning. (n.d.). Motivation as self-efficacy.
https://courses.lumenlearning.com/educationalpsychology/chapter/motivation-as-self-
efficacy/#:~:text=In%20self%2Defficacy%20theory%20the,action%20or%20goal%20are
%20specific.
Lundy-Wagner, V., Vultaggio, J., & Gasman, M. (2013). Preparing Underrepresented Students
of Color for Doctoral Success: The Role of Undergraduate Institutions. International
Journal of Doctoral Studies, 8, 151–172. https://doi.org/10.28945/1901
Ma, P., & Shea, M. (2019). First-Generation College Students’ Perceived Barriers and Career
195

Outcome Expectations: Exploring Contextual and Cognitive Factors. Journal of Career
Development, 89484531982765–. https://doi.org/10.1177/0894845319827650
Ma, Y., & Liu, Y. (2015). Race and STEM Degree Attainment. Sociology Compass, 9(7), 609–
618. https://doi.org/10.1111/soc4.12274
Malotky, M., Mayes, K. M., Price, K. M., Smith, G., Mann, S. N., Guinyard, M. W., Veale, S.,
Ksor, V., Siu, L., Mlo, H., Young, A. J., Nsonwu, M. B., Morrison, S. D., Sudha, S., &
Bernot, K. M. (2020). Fostering Inclusion through an Interinstitutional, Community-
Engaged, Course-Based Undergraduate Research Experience. Journal of microbiology &
biology education, 21(1), 21.1.31. https://doi.org/10.1128/jmbe.v21i1.1939
Mangan, K. (2019). New Data in College Scorecard Promises to Lift Veil on “Debt-Fueled”
Graduate-School Market. The Chronicle of Higher Education, 65(34), A18–.
Mantai, L. (2019). “A Source of Sanity”: The Role of Social Support for Doctoral Candidates’
Belonging and Becoming. International Journal of Doctoral Studies, 14, 367–382.
https://doi.org/10.28945/4275
Marcus, B., Bosnjak, M., Lindner, S., Pilischenko, S., & Schütz, A. (2007). Compensating for
Low Topic Interest and Long Surveys: A Field Experiment on Nonresponse in Web
Surveys. Social Science Computer Review, 25(3), 372–383.
https://doi.org/10.1177/0894439307297606
Martinez LR, Boucaud DW, Casadevall A, August A. Factors Contributing to the Success of
NIH-Designated Underrepresented Minorities in Academic and Nonacademic Research
Positions. CBE life sciences education. 2018;17(2):ar32-ar32. doi:10.1187/cbe.16-09-
0287
Maton, K., Beason, T., Godsay, S., Sto. Domingo, M., Bailey, T., Sun, S., & Hrabowski, F.
196

(2016). Outcomes and Processes in the Meyerhoff Scholars Program: STEM PhD
Completion, Sense of Community, Perceived Program Benefit, Science Identity, and
Research Self-Efficacy. CBE Life Sciences Education, 15(3), ar48–.
https://doi.org/10.1187/cbe.16-01-0062
Mattingly, A. (2019, May 4). The 'Leaky Pipe' of Diversity. University of California, San
Francisco Synapse. https://synapse.ucsf.edu/articles/2019/05/04/leaky-pipe-diversity
Mau, W. (2003). Factors That Influence Persistence in Science and Engineering Career
Aspirations. The Career Development Quarterly, 51(3), 234–243.
https://doi.org/10.1002/j.2161-0045.2003.tb00604.x
Maxwell, J. (2013). Qualitative Research Design: An Interactive Approach (3
rd
ed.). Thousand
Oaks, CA: SAGE
McGee, E., & Stovall, D. (2015). Reimagining Critical Race Theory in Education: Mental
Health, Healing, and the Pathway to Liberatory Praxis. Educational Theory, 65(5), 491–
511. https://doi.org/10.1111/edth.12129
McGee, E., White, D., Jenkins, A., Houston, S., Bentley, L., Smith, W., & Robinson, W. (2016).
Black engineering students’ motivation for PhD attainment: passion plus
purpose. Journal for Multicultural Education, 10(2), 167–193.
https://doi.org/10.1108/jme-01-2016-0007
McGee, R. (2016). Biomedical Workforce Diversity: The Context for Mentoring to Develop
Talents and Foster Success Within the “Pipeline.” AIDS and Behavior, 20(S2), 231–237.
https://doi.org/10.1007/s10461-016-1486-7
Melissa Phruksachart. (2017). On Mentoring Future Faculty of Color. Feminist Teacher, 27(2-3),
117–132. https://doi.org/10.5406/femteacher.27.2-3.0117
197

Mendoza-Denton, R., Patt, C., Fisher, A., Eppig, A., Young, I., Smith, A., & Richards, M.
(2017). Differences in STEM doctoral publication by ethnicity, gender and academic
field at a large public research university. PloS One, 12(4), e0174296–.
https://doi.org/10.1371/journal.pone.0174296
Merriam, S. B., & Tisdell, E. J. (2015). Qualitative research: A guide to design and
implementation. Malden, MA: John Wiley and Sons.
Miller, C. C. (2018). Does Teacher Diversity Matter in Student Learning?. New York Times.
https://www.nytimes.com/2018/09/10/upshot/teacher-diversity-effect-students-
learning.html?module=inline
Miller, C., Zwickl, B., Posselt, J., Silvestrini, R., & Hodapp, T. (2019). Typical physics Ph.D.
admissions criteria limit access to underrepresented groups but fail to predict doctoral
completion. Science Advances, 5(1), eaat7550–. https://doi.org/10.1126/sciadv.aat7550
Mondisa, J. (2018). Examining the Mentoring Approaches of African American
Mentors. Journal of African American Studies (New Brunswick, N.J.), 22(4), 293–308.
https://doi.org/10.1007/s12111-018-9411-y
Moreira, R., Butler-Purry, K., Carter-Sowell, A., Walton, S., Juranek, I., Challoo, L., Regisford,
G., Coffin, R., & Spaulding, A. (2019). Innovative Professional Development and
Community Building Activity Program Improves STEM URM Graduate Student
Experiences. International Journal of STEM Education, 6(1), 1–16.
https://doi.org/10.1186/s40594-019-0188-x
Morell, V. (1996). Computer Culture Deflects Women and Minorities. Science (American
Association for the Advancement of Science), 271(5257), 1915–1916.
Moss-Racusin, C. A., Dovidio, J. F., Brescoll, V. L., Graham, M. J., & Handelsman, J. (2012).
198

Science faculty’s subtle gender biases favor male students. Proceedings of the National
Academy of Sciences - PNAS, 109(41), 16474–16479.
https://doi.org/10.1073/pnas.1211286109
Museus, S., Yi, V., & Saelua, N. (2018). How culturally engaging campus environments
influence sense of belonging in college: An examination of differences between White
students and students of color. Journal of Diversity in Higher Education, 11(4), 467–483.
https://doi.org/10.1037/dhe0000069
Myers-Briggs Company. (2020). Why “Retention” and “Persistence” Aren’t Synonyms.
https://blog.vitanavis.com/why-retention-and-persistence-arent-synonyms/
Nadal, K. (2014). A Guide to Responding to Microaggressions. Asian Americans Advancing
Justice Los Angeles. https://advancingjustice-
la.org/sites/default/files/ELAMICRO%20A_Guide_to_Responding_to_Microaggressions
.pdf
National Institutes of Health (NIH). Who We Are. (n.d.).
https://www.nih.gov/about-nih/who-we-are
National Institutes of Health (NIH). Bridges to the Doctoral program (T32). (2020, June 10).
https://www.nigms.nih.gov/research/mechanisms/pages/bridgesdoctoral.aspx#:~:text=Go
al%20of%20the%20Program,in%20the%20biomedical%20research%20workforce
National Patient-Centered Clinical Research Network (PCORnet). (2018, May 11). Importance
of Diversity in Clinical Research. https://theaspirinstudy.org/2018/05/importance-of-
diversity-in-clinical-research/
National Research Council. (2011). Successful K-12 STEM Education: Identifying
199

Effective Approaches in Science, Technology, Engineering, and Mathematics.
Washington, DC: The National Academies Press. https://doi.org/10.17226/13158.
National Science Foundation (NSF). (n.d.). About the National Science Foundation.
https://www.nsf.gov/about/
National Science Foundation (NSF). (2017a). Doctoral degrees awarded, by field, citizenship,
ethnicity, and race: 2004–14.
https://www.nsf.gov/statistics/2017/nsf17310/static/data/tab7-4.pdf
National Science Foundation (NSF). (2017b). Resident population of the United States, by sex,
race or ethnicity, and age: 2014.
https://www.nsf.gov/statistics/2017/nsf17310/static/data/tab1-2.pdf
National Science Foundation (NSF). (2017c). Undergraduate enrollment at all institutions, by
citizenship, ethnicity, race, sex, and enrollment status: 2004–14.
https://www.nsf.gov/statistics/2017/nsf17310/static/data/tab2-1.pdf
National Science Foundation (NSF). (2017d). S&E doctorates awarded to U.S. citizens and
permanent residents, by field, sex, ethnicity, and race: 2004–14.
https://www.nsf.gov/statistics/2017/nsf17310/static/data/tab7-7.pdf
National Science Foundation (NSF). (2017e). Special Report NSF 17-310, National Science
Foundation, 2017. Women, minorities, and persons with disabilities in science and
engineering, and race. https://www.nsf.gov/statistics/2017/nsf17310/
National Science Foundation (NSF). (2017f). Resident population of the United States, by sex,
race or ethnicity, and age: 2014.
https://www.nsf.gov/statistics/2017/nsf17310/static/data/tab1-2.pdf
National Science Foundation (NSF). (2018a). NSF Scholarships in Science, Technology,
200

Engineering, and Mathematics Program (S-STEM).
https://www.nsf.gov/funding/pgm_summ.jsp?pims_id=5257
National Science Foundation (NSF). (2018b). Science & Engineering Indicators 2018, Chapter
5 Academic Research and Development.
https://www.nsf.gov/statistics/2018/nsb20181/report/sections/academic-research-and-
development/doctoral-scientists-and-engineers-in-academia
National Science Foundation (NSF). (2018c). 2017 Doctorate Recipients from U.S.
Universities. https://ncses.nsf.gov/pubs/nsf19301/data
National Science Foundation (NSF). (2019a). Top academic institutions awarding S&E
bachelor’s degrees, by race or ethnicity of minority graduates: 2012–16. Women,
minorities, and persons with disabilities in science and engineering, and race.
https://ncses.nsf.gov/pubs/nsf19304/data/
National Science Foundation (NSF). (2019b). Top baccalaureate institutions of black or African
American S&E doctorate recipients: 2013–17. Women, minorities, and persons with
disabilities in science and engineering, and race.
https://ncses.nsf.gov/pubs/nsf19304/data/
National Science Foundation (NSF). (2019c). Survey of Earned Doctorates (SED).
https://ncsesdata.nsf.gov/ids/sed
National Science Foundation (NSF). (2019d). Survey of Earned Doctorates.
https://ncses.nsf.gov/pubs/nsf20301/data-tables/
National Science Foundation (NSF). (2019e). S&E doctorates awarded to U.S. citizens and
permanent residents, by field, sex, ethnicity, and race: 2006–16.
https://ncses.nsf.gov/pubs/nsf19304/data/
201

National Science Foundation (NSF). (2019f). Science, engineering, and health doctorate
holders employed in universities and 4-year colleges, by broad occupation, race,
ethnicity, and faculty rank: 2017. https://ncses.nsf.gov/pubs/nsf19304/data/
Neumann, R., & Tan, K. (2011). From PhD to initial employment: the doctorate in a
knowledge economy. Studies in Higher Education (Dorchester-on-Thames), 36(5), 601–
614. https://doi.org/10.1080/03075079.2011.594596
New Mexico State University. (2006). Effective Strategies to Diversify STEM Faculty.
https://www.brown.edu/research/projects/advance/sites/brown.edu.research.projects.adva
nce/files/uploads/Frehill%20-
%20Effective%20Strategies%20to%20Diversify%20STEM%20Faculty.pdf
Nuñez, A. (2009). A critical paradox? predictors of Latino students’ sense of belonging in
college. Journal of Diversity in Higher Education, 2, 46–61.
http://dx.doi.org/10.1037/a0014099
Oguntebi, J., Shcherbakova, M., & Wooten, L. (2012). Applying Diversity Management
Concepts to Improve the Minority Educational Pipeline. Decision Sciences Journal of
Innovative Education, 10(4), 473–494. https://doi.org/10.1111/j.1540-4609.2012.00361.x
Okahana, H. & Allum, J. (2015). Doctoral Experience and Outcomes of URM students in STEM
Fields: Implications for Future Assessment and Practice. Council of Graduate Schools.
https://cgsnet.org/ckfinder/userfiles/files/2015-05-28-CGS_research_forum.pdf
Okahana, H., Klein, C., Allum, J., & Sowell, R. (2018). STEM Doctoral Completion of
Underrepresented Minority Students: Challenges and Opportunities for Improving
Participation in the Doctoral Workforce. Innovative Higher Education, 43(4), 237–255.
https://doi.org/10.1007/s10755-018-9425-3
202

Okahana, H. & Zhou, E. (2017). Graduate Enrollment and Degrees: 2006 to 2016. Council of
Graduate Schools.
https://cgsnet.org/ckfinder/userfiles/files/CGS_GED16_Report_Final.pdf
Ovink, S., & Veazey, B. (2010). More Than “Getting Us Through:” A Case Study in Cultural
Capital Enrichment of underrepresented Minority Undergraduates. Research in Higher
Education, 52(4), 370–394. https://doi.org/10.1007/s11162-010-9198-8
Pannucci, C. J., & Wilkins, E. G. (2010). Identifying and avoiding bias in research. Plastic and
reconstructive surgery, 126(2), 619–625. https://doi.org/10.1097/PRS.0b013e3181de24bc
Papageorge, N., Lindsay, C., Hyman, J., Gershenson, S., & Hart, C. (2018). The Long-Run
Impacts of Same-Race Teachers. https://doi.org/10.3386/w25254
Pérez-Stable, E.J. (2018, June 27). Communicating the Value of Race and Ethnicity in Research.
National Institutes of Health. https://www.nih.gov/about-nih/what-we-do/science-health-
public-trust/perspectives/science-health-public-trust/communicating-value-race-ethnicity-
research
Petersen, S., Erenrich, E., Levine, D., Vigoreaux, J., & Gile, K. (2018). Multi-institutional study
of GRE scores as predictors of STEM PhD degree completion: GRE gets a low
mark. PloS One, 13(10), e0206570–. https://doi.org/10.1371/journal.pone.0206570
Posselt, J. (2018). Trust Networks: A New Perspective on Pedigree and the Ambiguities of
Admissions. Review of Higher Education, 41(4), 497–521.
https://doi.org/10.1353/rhe.2018.0023
Pyne, J., & Grodsky, E. (2019). Inequality and Opportunity in a Perfect Storm of Graduate
Student Debt. Sociology of Education, 93(1), 003804071987624–003804071987639.
https://doi.org/10.1177/0038040719876245
203

Quinn, S. C., Butler, J., 3rd, Fryer, C. S., Garza, M. A., Kim, K. H., Ryan, C., & Thomas, S. B.
(2012). Attributes of researchers and their strategies to recruit minority populations:
results of a national survey. Contemporary clinical trials, 33(6), 1231–1237.
https://doi.org/10.1016/j.cct.2012.06.011
Rankin, S. R., & Reason, R. D. (2005). Differing perceptions: How students of color and White
students perceive campus climate for underrepresented groups. Journal of College
Student Development, 46, 43–61. http://dx.doi.org/10.1353/csd.2005.000
Rattan, A., Savani, K., Komarraju, M., Morrison, M., Boggs, C., & Ambady, N. (2018). Meta-
lay theories of scientific potential drive underrepresented students’ sense of belonging to
science, technology, engineering, and mathematics (STEM). Journal of Personality and
Social Psychology, 115(1), 54–75. https://doi.org/10.1037/pspi0000130
Reichert, W. (2013). Diversity and the Duke BME PhD Program: Then, Now and Moving
Forward. Annals of Biomedical Engineering, 41(10), 2019–2026.
https://doi.org/10.1007/s10439-013-0835-1
Roberts B. & Michael W. (2018). High-tech industries: an analysis of employment, wages, and
output. U.S. BUREAU OF LABOR STATISTICS.
https://www.bls.gov/opub/btn/volume-7/pdf/high-tech-industries-an-analysis-of-
employment-wages-and-output.pdf
Robinson, S. and Leonard, K., 2018. Designing Quality Survey Questions. 1st ed. Los Angeles:
Sage.
Rodriguez, S., & Lehman, K. (2017). Developing the next generation of diverse computer
scientists: the need for enhanced, intersectional computing identity theory. Computer
Science Education, 27(3-4), 229–247. https://doi.org/10.1080/08993408.2018.1457899
204

Roksa, J., Feldon, D., & Maher, M. (2018). First-Generation Students in Pursuit of the PhD:
Comparing Socialization Experiences and Outcomes to Continuing-Generation
Peers. The Journal of Higher Education (Columbus), 89(5), 728–752.
https://doi.org/10.1080/00221546.2018.1435134
Rudolph, A., Basri, G., Agüeros, M., Bertschinger, E., Coble, K., Donahue, M., … Posselt, J.
(2020). Final Report of the 2018 AAS Task Force on Diversity and Inclusion in
Astronomy Graduate Education. Bulletin of the AAS, 51(1). Retrieved from
https://baas.aas.org/pub/2019i0101
Rudolph, A., Holley-Bockelmann, K., & Posselt, J. (2019). PhD bridge programmes as engines
for access, diversity and inclusion. Nature Astronomy, 3(12), 1080–1085.
https://doi.org/10.1038/s41550-019-0962-1
Rumsey, D. (2019). Statistics Workbook for Dummies with Online Practice. In Statistics
Workbook for Dummies with Online Practice. John Wiley & Sons, Incorporated.
Salkind, N.J. (2014). Statistics for people who (think they) hate statistics. (5th ed.). Los Angeles:
SAGE.
Salters, R. (1997). Pursuing the Ph.D. in the Sciences and Engineering: Trends and
Observations. New Directions for Higher Education, 1997(99), 91–97.
https://doi.org/10.1002/he.9908
Salto LM, Riggs ML, Delgado De Leon D, Casiano CA, De Leon M (2014) Underrepresented
Minority High School and College Students Report STEM-Pipeline Sustaining Gains
After Participating in the Loma Linda University Summer Health Disparities Research
Program. PLoS ONE 9(9): e108497. https://doi.org/10.1371/journal.pone.0108497
San Francisco Department of Public Health (SFDPH). (2017).
205

Hospital Plan for Provision of Patient Care.
https://www.sfdph.org/dph/hc/HCAgen/HCAgen2017/Nov%2021/10%20Provision%20o
f%20Care%20Policy.pdf
San Francisco Department of Public Health (SFDPH). (2020).
Zuckerberg San Francisco General Hospital and Trauma Center.
https://www.sfdph.org/dph/comupg/oservices/medSvs/SFGH/MoreAboutUs.asp
Sánchez, N., Poll-Hunter, N., Spencer, D., Lee-Rey, E., Alexander, A., Holaday, L., Soto-
Greene, M., & Sánchez, J. (2018). Attracting Diverse Talent to Academia: Perspectives
of Medical Students and Residents. Journal of Career Development, 45(5), 440–457.
https://doi.org/10.1177/0894845317709997
Sander, R., & Taylor, S. (2012). Mismatch: How Affirmative Action Hurts Students It’s
Intended to Help, and Why Universities Won’t Admit It. New York: Basic Books.
Sauro, J., & Lewis, J. (2016). Quantifying the User Experience: Practical Statistics for User
Research. In Quantifying the User Experience (2nd edition.). Elsevier Science &
Technology.
Sav, G. (2020). Productivity catches up game on: can public research universities catch private
non-profit counterparts? Applied Economics Letters, 27(3), 228–232.
https://doi.org/10.1080/13504851.2019.1613487
Scandura T. A., & Williams E. A. (2000). Research Methodology in Management:
Current Practices, Trends, and Implications for Future Research. Academy of
Management Journal, 43(6), 1248–1264. https://doi.org/10.2307/1556348
Scott, L., & Zerwic, J. (2015). Holistic review in admissions: A strategy to diversify the nursing
206

workforce. Nursing Outlook, 63(4), 488–495.
https://doi.org/10.1016/j.outlook.2015.01.001
Sealy, L., Saunders, C., Blume, J., & Chalkley, R. (2019). The GRE over the entire range of
scores lacks predictive ability for PhD outcomes in the biomedical sciences. PloS
One, 14(3), e0201634–e0201634. https://doi.org/10.1371/journal.pone.0201634
Seidman, A., Astin, A., Berger, J., Bibo, E., Burkum, K., Cabrera, A., Crisp, G., Gansemer-Topf,
A., Hagedorn, L., & LaNasa, S. (2012). College Student Retention: Formula for Student
Success. In College Student Retention. Rowman & Littlefield Publishers.
Shadding, C., Whittington, D., Wallace, L., Wandu, W., & Wilson, R. (2016). Cost-Effective
Recruitment Strategies That Attract Underrepresented Minority Undergraduates Who
Persist to STEM Doctorates. SAGE Open, 6(3), 215824401665714–.
https://doi.org/10.1177/2158244016657143
Sierra-Mercado, D., & Lázaro-Muñoz, G. (2018). Enhance Diversity Among Researchers to
Promote Participant Trust in Precision Medicine Research. The American journal of
bioethics: AJOB, 18(4), 44–46. https://doi.org/10.1080/15265161.2018.1431323
Smith, J., & Noble, H. (2014). Bias in research. Evidence Based Nursing, 17(4), 100–101.
https://doi.org/10.1136/eb-2014-101946
Smith, W.A. 2010. Toward an understanding of Black misandric microaggressions and racial
battle fatigue in historically White institutions. In Zamani-Gallaher, E.S., and Polite,
V.C., eds., The state of the African American male: A courageous conversation. East
Lansing, MI: Michigan State University Press, 265–277.
Smolarek, B. B. (2019). The Hidden Challenges for Successful First-Generation Ph.D.s. Seton
207

Hall University.
https://scholarship.shu.edu/cgi/viewcontent.cgi?article=3346&context=dissertations/
Sowell, R. S., Zhang, T., Bell, N., & Redd, K. (2008). Ph.D. completion and attrition: Analysis
of baseline demographic data from the Ph.D. Completion Project. Washington, DC:
Council of Graduate Schools.
Sowell R., Allum J., & Okahana H. (2015). Doctoral Initiative on Minority Attrition and
Completion. Council of Graduate Schools.
https://cgsnet.org/ckfinder/userfiles/files/Doctoral_Initiative_on_Minority_Attrition_and
Completion_2015.pdf
Stassun, K., Burger, A., & Lange, S. (2018). The Fisk-Vanderbilt Masters-to-PhD Bridge
Program: A Model for Broadening Participation of underrepresented Groups in the
Physical Sciences through Effective Partnerships with Minority-Serving
Institutions. Journal of Geoscience Education, 58(3), 135–144.
https://doi.org/10.5408/1.3559648
Stassun, K., Sturm, S., Holley-Bockelmann, K., Burger, A., Ernst, D., & Webb, D. (2011). The
Fisk-Vanderbilt Master’s-to-PhD Bridge Program: Recognizing, enlisting, and cultivating
unrealized or unrecognized potential in URM students. American Journal of
Physics, 79(4), 374–379. https://doi.org/10.1119/1.3546069
Steele, C.M. 1997. A threat in the air: How stereotypes shape intellectual identity and
performance. American Psychologist, 52:613–629. National Institutes of Health (NIH). Who
We Are. (2014). https://diversity.nih.gov/sociocultural-factors/stereotype-threat
Sue, D., Capodilupo, C., Torino, G., Bucceri, J., Holder, A., Nadal, K., & Esquilin, M. (2007).
208

Racial Microaggressions in Everyday Life: Implications for Clinical Practice. The
American Psychologist, 62(4), 271–286. https://doi.org/10.1037/0003-066X.62.4.271
Syed, M., & Chemers, M. (2011). Ethnic Minorities and Women in STEM: Casting a Wide Net
to Address a Persistent Social Problem: Ethnic Minorities and Women in STEM. Journal
of Social Issues, 67(3), 435–441. https://doi.org/10.1111/j.1540-4560.2011.01708.x
Taylor, C. (2013). Validity and validation. Oxford University Press.
Thoman DB, Brown ER, Mason AZ, Harmsen AG, Smith JL (2015). The role of
prosocial values in motivating URM students for
biomedicine. BioScience 65, 183–188.
Thoman, D.B., Brown, E.R., Mason, A.Z., Harmsen, A.G. and Smith, J.L. (2014), “The role of
altruistic values in motivating URM students for biomedicine”,
BioScience, Vol. 65 No. 2, pp. 183-188.
Tinto, V. (2017). Reflections on student persistence.(Feature Article). Student Success, 8(2), 1–.
https://doi.org/10.5204/ssj.v8i2.376z
Toni A. Campbell, & David E. Campbell. (1997). Faculty/Student Mentor Program: Effects on
Academic Performance and Retention. Research in Higher Education, 38(6), 727–742.
https://doi.org/10.1023/A:1024911904627
Toretsky C., Mutha S., Coffman, J. (2018, July). Breaking Barriers for Underrepresented
Minorities in the Health Professions. Healthforce Center at University of California, San
Francisco. https://healthforce.ucsf.edu/sites/healthforce.ucsf.edu/files/publication-
pdf/Breaking%20Barriers%20for%20Underrepresented%20Minorities%20in%20the%20
Health%20Professions%20.pdf
Torpey, E. (2018). Measuring the value of education. U.S. BUREAU OF LABOR STATISTICS.
209

https://www.bls.gov/careeroutlook/2018/data-on-display/education-pays.htm
Tran M, Herrera F, Garibay J (2011). When science lacks diversity and social relevance can
students be objective scientists and still be themselves? Paper presented at the annual
meeting of the National Conference on Race and Ethnicity in American Higher
Education, held May 31–June 4 in San Francisco, CA.
University of Buffalo. (2008). Mathematicians of the African Diaspora.
http://www.math.buffalo.edu/mad/PEEPS/goins_edrayh.html
University of California (UC). (2020). UC employees, full-time equivalent (FTE).
https://www.universityofcalifornia.edu/infocenter/employee-fte
University of California (UC). (2021). The UC System.
https://www.universityofcalifornia.edu/uc-system
University of California, San Francisco (UCSF). (2018). 2018 Annual Financial Report.
https://controller.ucsf.edu/financial-statements/campus/2018/2018annual-financial-report
University of California, San Francisco (UCSF). (2020a). UCSF Remains Top Public Recipient
of NIH Funding for 13
th
Straight Year. https://www.ucsf.edu/news/2020/04/417061/ucsf-
remains-top-public-recipient-nih-funding-13th-straight-year
University of California, San Francisco (UCSF). (2020b). UCSF Schools Shine in 2021 U.S.
News Best Graduate Schools Rankings.
https://www.ucsf.edu/news/2020/03/416926/ucsf-schools-shine-2021-us-news-best-
graduate-schools-rankings
University of California, San Francisco (UCSF). (2020c). About UCSF at ZSFG.
https://zsfg.ucsf.edu/about-ucsf-zsfg
University of California, San Francisco, Center for Community Engagement
210

(UCSF Center for Community Engagement). (2020). Education & Training
https://partnerships.ucsf.edu/education-training
University of California, San Francisco, Department of Anesthesia and Perioperative Care
(UCSF Department of Anesthesia and Perioperative Care). (2020, February 24).
Anesthesia Department Members Honored to Attend UCSF Black Caucus Gala Dinner.
https://anesthesia.ucsf.edu/news/anesthesia-department-members-honored-attend-ucsf-
black-caucus-gala-dinner
University of California, San Francisco, Graduate Division (UCSF Graduate Division). (2020a).
Program Statistics - Demographics, Admissions, Time to Degree, and Completion Rates.
https://graduate.ucsf.edu/program-statistics
University of California, San Francisco, Graduate Division (UCSF Graduate Division). (2020b).
Summer Research Training Program. https://graduate.ucsf.edu/srtp
University of California, San Francisco, Graduate Division (UCSF Graduate Division). (2020c).
The Graduate Division. https://graduate.ucsf.edu/
University of California, San Francisco, Interprofessional Diversity and Equity Alliance (UCSF
Interprofessional Diversity and Equity Alliance). (2020). About Us.
https://ucsf.campusgroups.com/idea/home/
University of California, San Francisco, Office of Diversity and Outreach
(UCSF Office of Diversity and Outreach). (2020a). Office of Diversity and Outreach
Welcomes New Students to UCSF. https://diversity.ucsf.edu/content/office-diversity-and-
outreach-welcomes-new-students-ucsf
University of California, San Francisco, Office of Diversity and Outreach
211

(UCSF Office of Diversity and Outreach). (2020b). Who We Are.
https://diversity.ucsf.edu/about/who-we-are
University of California, San Francisco, Office of Diversity and Outreach
(UCSF Office of Diversity and Outreach). (2020c). Data Sharing Initiatives.
https://diversity.ucsf.edu/reports-data/diversity-data
University of California, San Francisco, Office of Diversity and Outreach
(UCSF Office of Diversity and Outreach). (2020d). Unconscious Bias.
https://diversity.ucsf.edu/reports-data/diversity-data
University of California, San Francisco, Office of Institutional Research
(UCSF Office of Institutional Research). (2020). UCSF at a Glance.
https://diversity.ucsf.edu/about/who-we-are
University of California, San Francisco, Student Academic Affairs (UCSF SAA). (2020a).
2020 Student Services Survey Results.
https://studentaffairs.ucsf.edu/2020-student-services-survey-results
University of California, San Francisco, Student Academic Affairs (UCSF SAA). (2020b).
About Student Academic Affairs
https://studentaffairs.ucsf.edu/about#Administrative
University of California, San Francisco, UCSF Health
(UCSF Health). (2020). Our Mission. https://www.ucsfhealth.org/about/our-mission
University of Chicago: National Opinion Research Center. (2008). Doctorate Recipients from
U.S. Universities: Summary Report 2006;2008 SRI U6395-5.
United States Regional Economic Analysis Project. (2020). United States Comparative Trends
212

Analysis: Population Growth and Change, 1958-2019. https://united-
states.reaproject.org/analysis/comparative-trends-analysis/population/tools/0/0/
U.S. Department of Education. (n.d.). What is an HBCU?. https://sites.ed.gov/whhbcu/one-
hundred-and-five-historically-black-colleges-and-universities/
U.S. Department of Education. (2016, November). Advancing Diversity and Inclusion in Higher
Education. https://www2.ed.gov/rschstat/research/pubs/advancing-diversity-inclusion.pdf
U.S. Department of the Interior. (n.d.). Minority Serving Institutions Program.
https://www.doi.gov/pmb/eeo/doi-minority-serving-institutions-program
University of Michigan. (2016). Exit Interview Study: Executive Summary University of
Michigan ADVANCE Program May 2016. https://advance.umich.edu/wp-
content/uploads/2018/09/UM-Exit-Interview-Study-2014-Report-executive-final.pdf
U.S. News & World Report. (2019). Best Colleges Rankings - Amherst College 2019 Rankings.
https://www.usnews.com/best-colleges/amherst-college-2115/overall-rankings
U.S. News & World Report. (2021a). U.S. News Best Colleges Rankings
https://www.usnews.com/best-colleges
U.S. News & World Report. (2021b). University of California--San Francisco
https://www.usnews.com/best-graduate-schools/university-of-california-san-francisco-
110699/overall-rankings
Whitfield, K., & Edwards, C. (2011). Mentoring Special Populations. Educational Gerontology:
Mentoring in Clinical Geropsychology, 37(5), 422–439.
https://doi.org/10.1080/03601277.2011.553563
Whittaker, J., & Montgomery, B. (2012). Cultivating Diversity and Competency in STEM:
213

Challenges and Remedies for Removing Virtual Barriers to Constructing Diverse Higher
Education Communities of Success. Journal of Undergraduate Neuroscience
Education, 11(1), A44–A51.
Williams, S., Thakore, B., & McGee, R. (2017). Providing Social Support for Underrepresented
Racial and Ethnic Minority PhD Students in the Biomedical Sciences: A Career Coaching
Model. CBE Life Sciences Education, 16(4), ar64–. https://doi.org/10.1187/cbe.17-01-
0021
Wilson, M., DePass, A., & Bean, A. (2018). Institutional Interventions That Remove Barriers to
Recruit and Retain Diverse Biomedical PhD Students. CBE Life Sciences
Education, 17(2), ar27–ar27. https://doi.org/10.1187/cbe.17-09-0210
Wilson, M., Odem, M., Walters, T., DePass, A., & Bean, A. (2019). A Model for Holistic
Review in Graduate Admissions That Decouples the GRE from Race, Ethnicity, and
Gender. CBE Life Sciences Education, 18(1), ar7–. https://doi.org/10.1187/cbe.18-06-
0103
Wilson, Z., Holmes, L., deGravelles, K., Sylvain, M., Batiste, L., Johnson, M., McGuire, S.,
Pang, S., & Warner, I. (2011). Hierarchical Mentoring: A Transformative Strategy for
Improving Diversity and Retention in Undergraduate STEM Disciplines. Journal of
Science Education and Technology, 21(1), 148–156. https://doi.org/10.1007/s10956-011-
9292-5
Winkle-Wagner, R., & McCoy, D. (2018). Feeling like an “Alien” or “Family”? Comparing
students and faculty experiences of diversity in STEM disciplines at a PWI and an
HBCU. Race Ethnicity and Education, 21(5), 593–606.
https://doi.org/10.1080/13613324.2016.1248835
214

Witkow, M., & Fuligni, A. (2011). Ethnic and Generational Differences in the Relations between
Social Support and Academic Achievement across the High School Years. Journal of
Social Issues, 67(3), 531–552. https://doi.org/10.1111/j.1540-4560.2011.01713.x
World Population Review. (2020). San Francisco, California Population 2020
https://worldpopulationreview.com/us-cities/san-francisco-ca-population
Wuensch, K. L. (2015). How Do You Pronounce "Likert?" What is a Likert Scale? East
Carolina University, Department of Psychology.
http://core.ecu.edu/psyc/wuenschk/StatHelp/Likert.htm
Yadav, A., & Seals, C. (2019). Taking the next step: supporting postdocs to develop an
independent path in academia. International Journal of STEM Education, 6(1), 1–11.
https://doi.org/10.1186/s40594-019-0168-1

215

Appendix A: Assessment Tools to Evaluate Assumed KMO Influences
Category Construct Assessment tool
Knowledge influences
Declarative Knowledge deficiencies in first-generation
students
Survey
Procedural Benefits of student mentorship Survey
Conditional Master’s degree as stepping stone to
doctorate degree
Survey
Declarative Understanding STEM educational
pathways
Survey
Motivation influences
Self-efficacy Student self-efficacy and motivation Survey
Utility value Finding prosocial values Survey
Attainment value Forging identities Survey
Maslow's hierarchy of
needs
Sense of belonging to campus Survey
Intrinsic value/utility
value
Mentorship and research opportunities Survey
Attainment value URM faculty and mentors Survey
Organization influences
Alignment Barriers for first-generation students Survey
Alignment Admission biases in doctoral programs Survey
Culture Hostile campus environment Survey
Culture Lack of faculty diversity Survey
Resource allocation Limited opportunities Survey

216

Appendix B: Qualtrics Online Survey Questions
Question

Open
or
closed
?

Type

Response options (if
close-ended)

Concept being
measured

Q1: Are you a member of any
of these ethnic groups?
(Check all that applies)
@Qualtrics logic: stop
survey and thank them if
participant does not select
one or more of these
race/ethnic groups.
Closed Nominal African American
Hispanic/Latino
American Indian or
Alaska Native
Demographic
information
Q2: Are you a student
currently enrolled in a basic
or biomedical sciences PhD
program at UCSF?
@Qualtrics logic: stop
survey and thank them if
participant selects ‘No’
Closed Nominal Yes
No
Demographic
information
Q3: Which basic or biomedical
sciences PhD program are
you enrolled in?
Closed Nominal Biochemistry and
Molecular Biology
(Tetrad)
Bioengineering
(joint with UC
Berkeley)
Biological and
Medical
Informatics
Biomedical
Sciences
Biophysics
Cell Biology
(Tetrad)
Chemistry and
Chemical Biology
Developmental and
Stem Cell Biology
Epidemiology and
Translational
Science
Demographic
information
217

Question

Open
or
closed
?

Type

Response options (if
close-ended)

Concept being
measured

Genetics (Tetrad)
Neuroscience
Oral and
Craniofacial
Sciences
Pharmaceutical
Sciences and
Pharmacogenomics
Rehabilitation
Science
Q4: What is your current
gender identity?

Closed Nominal Male
Female
X
Demographic
information
Q5: What is your current year
in the PhD program?
Closed Nominal 1st Year
2nd Year
3rd Year
4th Year
5th Year
6th Year
7
th
Year
8
th
Year
9
th
Year
10 or more Years
Demographic
information
Q6: At what institution did you
obtain your bachelor’s
degree?
Open Demographic
information
Q7: What was the major for
your bachelor’s degree?
Open Demographic
information
Q8: Do you feel your
undergraduate degree and
undergraduate institution
provided you with the
adequate skills and
foundations needed to persist
Closed Ordinal

Definitely
Very Probably
Probably
Probably Not
Definitely Not
Knowledge
influence
(declarative)

218

Question

Open
or
closed
?

Type

Response options (if
close-ended)

Concept being
measured

and succeed in the STEM
PhD program?
Q9: There were opportunities
for me to do lab or research
work in my undergraduate
years.

Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Knowledge
influence
(declarative),
motivation

influence
(attainment
value)
Q10: How important is
mentorship in the
undergraduate level in
motivating you to your PhD
degree?
Closed Ordinal

Extremely
Important
Very Important
Moderately
Important
Slightly Important
Not Important
Motivation
influence
(intrinsic value)
Q11: Did you complete a
master’s degree prior to
attending your PhD
program?
@Qualtrics logic: answering
‘Yes’ would then branch to
the following three questions.
Closed Nominal Yes
No
Demographic
information
Q12: At what institution did
you obtain your master’s
degree?
Open Demographic
information
Q13: What was the major for
your master’s degree?
Open Demographic
information
Q14: Having a master's degree
was helpful to acquire the
knowledge and preparation
needed to be successful in
STEM doctoral programs.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Knowledge
influence
(conditional)
219

Question

Open
or
closed
?

Type

Response options (if
close-ended)

Concept being
measured

Strongly
Disagree
Q15: Are you a first-
generation college student?
@Qualtrics logic: answering
‘Yes’ would then branch to
the following question.
Closed Nominal Yes
No
Demographic
information
Q16: Who provided you with
the guidance for preparing
and applying for college
(e.g., parents, high school
teachers)
Open Knowledge
influence
(declarative)
Q17: Learning coping skills
and resiliency dealing with
social pressures and
transition to college is very
important.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Knowledge
influence
(procedural),
motivation
influence (self-
efficacy)
Q18: List the top reasons (up
to three) that motivated you
to enroll in the PhD program
in order of importance (i.e.,
the top one being the most
important).
Open Motivation
influence
(attainment
value/utility
value)
Q19: Adjusting and transiting
to the PhD program from the
institution where you
obtained your most recent
degree (bachelor or master’s)
was…
@Qualtrics logic: selecting
“Extremely Difficult”, “Very
Difficult” and “Moderately
Difficult” would branch to
the following question.
Closed Ordinal

Extremely Difficult
Very Difficult
Moderately
Difficult
Slightly Difficult
Not at all Difficult
Knowledge
influence
(conditional),
motivation
influence
(Maslow’s
hierarchy of
needs)
220

Question

Open
or
closed
?

Type

Response options (if
close-ended)

Concept being
measured

Q20: List the top reasons (up
to three) that made you feel
transition into your PhD
program was difficult in
order of importance (i.e., the
top one being the most
important).
Open Knowledge
influence
(conditional),
motivation
influence
(Maslow’s
hierarchy of
needs)
Q21: How different was your
academic environment in the
institution where you
obtained your most recent
degree (bachelor or master’s)
from UCSF?
@Qualtrics logic: selecting
any choice, but
Not at all Different” would
branch to the following
question.
Closed Ordinal

Extremely Different
Very Different
Moderately
Different
Slightly Different
Not at all Different
Organizational
influence
(culture)
Q22: List the top differences
(up to three) in order of
importance (i.e., the top one
being the most important
difference).
Open Organizational
influence
(culture)
Q23: Research opportunities
are very important for me to
persist and to succeed in my
PhD program.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Knowledge
influence
(declarative),
motivation
influence
(attainment
value/utility
value)
Q24: There are sufficient
research opportunities given
to me in the PhD program.

Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Organizational
influence
(resource
allocation),
221

Question

Open
or
closed
?

Type

Response options (if
close-ended)

Concept being
measured

Disagree
Strongly Disagree
knowledge
influence
(declarative),
motivation
influence
(attainment
value/utility
value)
Q25: From my knowledge,
underrepresented minorities
PhD students, in general, are
not given enough research
opportunities.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Organizational
influence
(resource
allocation),
motivation
influence
(attainment
value/utility
value)
Q26: Attending and presenting
at conferences are very
important for me to persist
and to succeed in my PhD
program.

Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Organizational
influence
(resource
allocation),
motivation
influence
(attainment
value/utility
value)
Q27: From my knowledge,
underrepresented minorities
PhD students, in general, are
not given enough
opportunities to attend and
present at conferences.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Organizational
influence
(resource
allocation),
motivation
influence
(attainment
value/utility
value)
222

Question

Open
or
closed
?

Type

Response options (if
close-ended)

Concept being
measured

Q28: Publication of papers in
academic journals is very
important for me to persist
and to succeed in my PhD
program.

Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Organizational
influence
(resource
allocation),
motivation
influence
(attainment
value/utility
value)
Q29: From my knowledge,
underrepresented minority
PhD students, in general, are
not given enough
opportunities to publish
papers in academic journals.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Organizational
influence
(resource
allocation),
motivation
influence
(attainment
value/utility
value)
Q30: Mentorship is a key to
my academic success.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Knowledge
influence
(procedural),
motivation
influence
(intrinsic
value/utility
value)
Q31: Mentorship increases
participation and persistence
in my doctoral program.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Motivation
influence
(intrinsic
value/utility
value)
Q32: Mentorship increases my
sense of belonging to the
academic environment.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Organizational
influence
(culture),
223

Question

Open
or
closed
?

Type

Response options (if
close-ended)

Concept being
measured

Disagree
Strongly Disagree
motivation
influence
(Maslow’s
hierarchy of
needs/intrinsic
value)
Q33: I have sufficient
mentorship opportunities in
my PhD program
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Organizational
influence
(culture/
resource
allocation),
motivation
influence
(intrinsic
value/utility
value)
Q34: A mentor that looks like
me (i.e., same ethnic and
cultural identity) is important
to me.
@Qualtrics logic: whatever
selected automatically
branches to the question
below.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Organizational
influence
(culture),
motivation
influence
(attainment
value)
Q35: List the top reasons (up
to three) for the way you did
in answering the above
question in order of
importance (i.e., the top one
being the most important).
Open Motivation
influence
(attainment
value)
Q36: A lack of mentorship is a
barrier for URM students
from applying for graduate
schools.

Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Motivation
influence
(intrinsic
value/utility
value)
224

Question

Open
or
closed
?

Type

Response options (if
close-ended)

Concept being
measured

Q37: A lack of
career/education information
or guidance is a major barrier
for URM students from
applying for graduate
schools.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Knowledge
influence
(declarative)
Q38: The lack of
underrepresented minority
role models is a major de-
motivation for URM students
from pursuing and persisting
in a STEM degree (bachelor,
master’s, or PhD)

Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Motivation
influence
(attainment
value),
organizational
influence
(alignment/
culture)
Q39: It is difficult to have a
scientist identity when there
is a lack of role models that
look like me (i.e., same
ethnic and cultural identity.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Motivation
influence
(attainment
value)

Q40: Pursuing a science career
or scientist identity means
losing my racial or cultural
identity.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Knowledge
influence
(declarative),
motivation
influence
(attainment
value)

Q41: Identifying with a
science career or scientist
identity is a very important
motivator to persist in the
PhD program

Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Motivation
influence
(attainment
value)
Q42: The availability of
underrepresented minority
Closed Ordinal

Strongly Agree
Agree
Motivation
influence
225

Question

Open
or
closed
?

Type

Response options (if
close-ended)

Concept being
measured

faculty role models allows
me to successfully create a
scientist identity with role
models like themselves that
have made it.
Neither agree nor
disagree
Disagree
Strongly Disagree
(attainment
value)
Q43: The integration of culture
into science such as
incorporating culturally
responsive teaching
principles to science courses
are important to me.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Motivation
influence
(attainment
value/utility
value)
Q44: Negative stereotypes that
underrepresented minorities
don’t have math or science
ability is a major factor that
reduces URM students’ sense
of belonging to the STEM
discipline
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Motivation
influence
(Maslow’s
hierarchy of
needs),
organizational
influence
(culture)
Q45: I have personally faced
the negative stereotypes
mentioned above before.
Closed Nominal Yes
No
Motivation
influence
(Maslow’s
hierarchy of
needs),
organizational
influence
(culture)
Q46: How important is
aligning or framing STEM
education and research with
social justice and helping the
community?
Closed Ordinal

Extremely
Important
Very Important
Moderately
Important
Slightly Important
Not at all Important
Motivation
influence
(attainment
value/utility
value),
organizational
influence
(alignment/
culture)
226

Question

Open
or
closed
?

Type

Response options (if
close-ended)

Concept being
measured

Q47: Master’s-to-PhD bridges
programs can help increase
underrepresented minorities
enrollment in the STEM PhD
programs.
(If you are not familiar with
what a Master’s-to-PhD
bridges program is, you can
skip this question).
@Qualtrics logic: make this
question optional.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Knowledge
influence
(conditional),
motivation
influence (self-
efficacy),
organizational
influence
(alignment)
Q48: More transparency with
admissions data for STEM
PhD programs in general will
attract more
underrepresented minority
applicants to these programs.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Knowledge
influence
(declarative),
motivation
influence (self-
efficacy),
organizational
influence
(alignment/
culture)
Q49: Publicizing and
providing on-campus student
support programs and
services is very important for
student persistence and
success in the PhD programs.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Knowledge
influence
(procedural),
motivation
influence (self-
efficacy),
organizational
influence
(alignment/
culture)
Q50: I have access to student
services when I need it
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Organizational
influence
(alignment/
resource
allocation)
227

Question

Open
or
closed
?

Type

Response options (if
close-ended)

Concept being
measured

Q51: I have a strong sense of
belonging to the campus.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Motivation
influence
(Maslow’s
hierarchy of
needs),
organizational
influence
(culture)
Q52: I have a strong sense of
belonging to my program.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Motivation
influence
(Maslow’s
hierarchy of
needs),
organizational
influence
(culture)
Q53: I feel alienated with my
instructors.
@Qualtrics logic: selecting
“Strongly Agree” or “Agree”
would branch to the
following question.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Motivation
influence
(Maslow’s
hierarchy of
needs),
organizational
influence
(culture)
Q54: The main reason(s) for
alienation from instructors
Open Motivation
influence
(Maslow’s
hierarchy of
needs),
organizational
influence
(culture)
Q55: I feel alienated from my
peers.
@Qualtrics logic: selecting
“Strongly Agree” or “Agree”
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Motivation
influence
(Maslow’s
228

Question

Open
or
closed
?

Type

Response options (if
close-ended)

Concept being
measured

would branch to the
following question.
Disagree
Strongly Disagree
hierarchy of
needs),
organizational
influence
(culture)
Q56: The main reason(s) for
alienation from peers
Open Motivation
influence
(Maslow’s
hierarchy of
needs),
organizational
influence
(culture)
Q57: How important is it to
you to have more URM
students in the STEM PhD
programs?
Closed Ordinal

Extremely
Important
Very Important
Moderately
Important
Slightly Important
Not at all Important
Motivation
influence
(Maslow’s
hierarchy of
needs/
attainment value),
organizational
influence
(alignment/
culture)
Q58: Do you feel there is
loneliness and isolation being
an underrepresented minority
student in a STEM PhD
program?
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Motivation
influence
(Maslow’s
hierarchy of
needs/
attainment value),
organizational
influence
(alignment/
culture)
Q59: How important is
learning to build peer and
faculty networks to avoid
Closed Ordinal

Strongly Agree
Agree
Motivation
influence
(Maslow’s
229

Question

Open
or
closed
?

Type

Response options (if
close-ended)

Concept being
measured

isolation and loneliness in a
STEM PhD program?
Neither agree nor
disagree
Disagree
Strongly Disagree
hierarchy of
needs/self-
efficacy),
organizational
influence
(resource
allocation)
Q60: I have sufficient
opportunities to learn to build
peer and faculty networks
either in my current PhD
program or in my
undergraduate or master’s
degree.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Motivation
influence
(Maslow’s
hierarchy of
needs/self-
efficacy),
organizational
influence
(resource
allocation)
Q61: How important is seeking
prosocial goals and
opportunities (such as
helping communities) in the
classroom, conferences or
research of my STEM PhD
program?
@Qualtrics logic: selecting
“Extremely Important” or
“Very Important” would
branch to the following two
questions.
Closed Ordinal

Extremely
Important
Very Important
Moderately
Important
Slightly Important
Not at all Important
Motivation
influence
(utility value),
organizational
influence
(culture/
alignment)
Q62: I feel that peers do not
have the same enthusiasm for
seeking prosocial goals and
opportunities in a STEM
PhD program as I do.
Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Motivation
influence
(Maslow's
hierarchy of
needs/utility
value),
230

Question

Open
or
closed
?

Type

Response options (if
close-ended)

Concept being
measured

organizational
influence
(culture/
alignment)
Q63: I feel that my department
does not have the same
enthusiasm for seeking
prosocial goals and
opportunities in a STEM
PhD program as I do.
Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Motivation
Influence
(Maslow's
hierarchy of
needs/utility
value),
organizational
influence
(culture/
alignment)
Q64: How important is being
provided career and
academic guidance and
advice pursuing a graduate
degree (master’s, PhD) in
your undergraduate years?
Closed Ordinal

Extremely
Important
Very Important
Moderately
Important
Slightly Important
Not at all Important
Knowledge
influence
(procedural),
motivation
influence (self-
efficacy),
organizational
influence
(alignment/
resource
allocation)
Q65: How much career and
academic guidance and
advice pursuing a graduate
degree (master’s, PhD) was
provided to you in your
undergraduate institution?
Closed Ordinal

Very sufficient
Sufficient
Barely Sufficient
Not Sufficient
Not at all
Knowledge
influence
(declarative),
organizational
influence
(alignment/
culture)
Q66: What do you feel are the
biggest obstacles to
graduating from your STEM
PhD program?
Open Knowledge
influence
(conditional),
231

Question

Open
or
closed
?

Type

Response options (if
close-ended)

Concept being
measured

Motivation
influence (self-
efficacy),
organizational
influence
(alignment/
culture/resource
allocation)
Q67: From my knowledge,
underrepresented minorities
PhD students, in general, are
not given enough mentorship
opportunities.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Organizational
influence
(resource
allocation),
motivation
influence
(attainment/
utility value)
Q68: Do you feel if there were
more career and educational
guidance provided at the high
school or undergraduate
years, more URM students
would pursue graduate
studies in the STEM
disciplines?
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Knowledge
influence
(procedural),
organizational
influence
(resource
allocation)
Q69: How important is
acquiring learning strategies
in relation to persisting and
succeeding in your PhD
program?
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Knowledge
influence
(conditional),
motivation
influence (self-
efficacy)
Q70: I have sufficient
opportunities to acquire
learning strategies either in
my current PhD program or
in my undergraduate or
master’s degree.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Knowledge
influence
(conditional),
motivation
influence (self-
efficacy)
232

Question

Open
or
closed
?

Type

Response options (if
close-ended)

Concept being
measured

Q71: How important is
acquiring study skills in
relation to persisting and
succeeding in your PhD
program?
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Knowledge
influence
(conditional),
motivation
influence (self-
efficacy)
Q72: I have sufficient
opportunities to acquire
study skills either in my
current PhD program or in
my undergraduate or
master’s degree.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Knowledge
influence
(conditional),
motivation
influence (self-
efficacy)
Q73: As it relates to your PhD
Program, I will be able to
achieve most of the goals
that I have set for myself.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Motivation
influence (self-
efficacy)
Q74: As it relates to your PhD
Program, when facing
difficult tasks, I am certain
that I will accomplish them.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Motivation
influence (self-
efficacy)
Q75: As it relates to your PhD
Program, in general, I think
that I can obtain outcomes
that are important to me.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Motivation
influence (self-
efficacy)
Q76: As it relates to your PhD
Program, I believe I can
succeed at almost any
endeavor to which I set my
mind.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Motivation
influence (self-
efficacy)
233

Question

Open
or
closed
?

Type

Response options (if
close-ended)

Concept being
measured

Q77: As it relates to your PhD
Program, I will be able to
successfully overcome many
challenges.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Motivation
influence (self-
efficacy)
Q78: As it relates to your PhD
Program, I am confident that
I can perform effectively on
many different tasks.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly
Disagree
Motivation
influence (self-
efficacy)
Q79: As it relates to your PhD
Program, compared to other
people, I can do most tasks
very well.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Motivation
influence (self-
efficacy)
Q80: As it relates to your PhD
Program, even when things
are tough, I can perform
quite well.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Motivation
influence (self-
efficacy)
Q81: I feel a sense of
belonging at UCSF.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Organizational
influence
(culture)
Q82: I feel a sense of
belonging in my current
program/department.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Organizational
influence
(culture)
234

Question

Open
or
closed
?

Type

Response options (if
close-ended)

Concept being
measured

Strongly Disagree
Q83: Do you feel UCSF is
promoting diversity and
inclusion?
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Organizational
influence
(alignment/
culture)
Q84: List the top reasons (up
to four) that made you decide
to pursue your PhD program
at UCSF versus other
institutions in order of
importance (i.e., the top one
being the most important).
Open Organizational
influence
(alignment/
culture)
Q85: UCSF being located in
San Francisco, a diverse,
minority-majority city plays
a significant factor in coming
to UCSF.
Closed Ordinal

Strongly Agree
Agree
Neither agree nor
disagree
Disagree
Strongly Disagree
Organizational
influence
(culture)
Q86: Anything else you want
to add to the topic of the
knowledge, motivational and
organizational influences that
will help increase
underrepresented minorities
representation STEM PhD
programs?
Open

Note. The ‘Type’ column indicates the level of measurement, if any, used for each question. The
response options listed for closed-end questions are the actual values presented to students in the
Qualtrics© survey.
235

Appendix C: Rating Scale for Values Used in Closed Questions of Survey
Rating scale for every multiple-choice value in all closed questions of the survey.
● Scaling Used:
o Very sufficient 5
o Sufficient 4
o Barely Sufficient 3
o Not Sufficient 2
o Not at all 1
● Survey Question Scaling Applied to: 65

● Scaling Used:
o Extremely Important 5
o Very Important 4
o Moderately Important 3
o Slightly Important 2
o Not at all Important 1
● Survey Questions Scaling Applied to: 10, 46, 57, 61, 64

● Scaling Used:
o Strongly Agree 5
o Agree 4
o Neither agree nor disagree 3
o Disagree 2
o Strongly Disagree 1
● Survey Questions Scaling Applied to: 9, 14, 17, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32,
33, 34, 36, 37, 38, 39, 40, 41, 42, 43, 44, 47, 48, 49, 50, 51, 52, 53, 55, 58, 59, 60, 62,
63, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 85

● Scaling Used:
o Definitely 5
o Very Probably 4
o Probably 3
o Probably Not 2
o Definitely Not 1
● Survey Question Scaling Applied to: 8

● Scaling Used:
o Extremely Difficult 5
o Very Difficult 4
o Moderately Difficult 3
o Slightly Difficult 2
o Not at all Difficult 1
● Survey Question Scaling Applied to: 19

236

● Scaling Used:
o Extremely Different 5
o Very Different 4
o Moderately Different 3
o Slightly Different 2
o Not at all Different 1
● Survey Question Scaling Applied to: 21

237

Appendix D: Data Analysis of Qualtrics Responses
Question number N Mean SD Min Max Mode
8 33 3.73 0.96 1.00 5.00 3.00
9 33 4.00 0.74 1.00 5.00 4.00
10 33 3.45 0.56 2.00 4.00 4.00
17 33 4.36 0.98 2.00 4.00 4.00
19 33 2.85 0.93 1.00 4.00 2.00
21 33 2.94 0.92 2.00 5.00 2.00
23 33 4.79 0.41 4.00 5.00 5.00
24 33 4.76 0.43 4.00 5.00 5.00
25 33 2.45 1.35 1.00 5.00 1.00
26 33 4.15 0.70 3.00 5.00 4.00
27 33 2.97 1.06 1.00 4.00 4.00
28 33 4.45 0.70 3.00 5.00 5.00
29 33 3.18 1.17 1.00 5.00 4.00
30 33 4.85 0.36 4.00 5.00 5.00
31 33 4.82 0.39 4.00 5.00 5.00
32 33 4.76 0.60 2.00 5.00 5.00
33 33 3.61 0.55 3.00 5.00 4.00
34 33 4.03 0.80 3.00 5.00 4.00
36 33 4.64 0.48 4.00 5.00 5.00
37 33 4.76 0.43 4.00 5.00 5.00
38 33 4.45 0.56 3.00 5.00 5.00
39 33 3.97 1.11 2.00 5.00 5.00
40 33 2.24 1.02 1.00 5.00 3.00
41 33 4.39 0.60 3.00 5.00 4.00
42 33 3.94 0.60 3.00 5.00 4.00
43 33 4.33 0.68 3.00 5.00 4.00
44 33 4.30 0.76 3.00 5.00 5.00
46 33 4.33 0.97 2.00 5.00 5.00
47 9 3.44 0.50 3.00 4.00 3.00
48 33 4.36 1.04 2.00 5.00 5.00
49 33 4.67 0.47 4.00 5.00 5.00
50 33 4.21 0.64 2.00 5.00 4.00
51 33 3.27 0.79 2.00 4.00 4.00
52 33 3.79 0.41 3.00 4.00 4.00
53 33 2.12 0.88 1.00 4.00 3.00
55 33 2.21 0.88 1.00 4.00 2.00
238

Question number N Mean SD Min Max Mode
57 33 4.67 0.68 3.00 5.00 5.00
58 33 4.00 0.55 3.00 5.00 4.00
59 33 4.61 0.49 4.00 5.00 5.00
60 33 3.97 0.67 2.00 5.00 4.00
61 33 4.24 0.89 1.00 5.00 5.00
62 33 3.48 0.56 2.00 4.00 4.00
63 33 3.36 0.92 2.00 5.00 3.00
64 33 4.52 0.70 3.00 5.00 5.00
65 33 3.42 0.85 1.00 4.00 4.00
67 33 4.61 0.55 3.00 5.00 5.00
68 33 4.82 0.39 4.00 5.00 5.00
69 33 4.64 0.48 4.00 5.00 5.00
70 33 3.45 0.74 2.00 4.00 4.00
71 33 4.06 1.10 2.00 5.00 5.00
72 33 3.48 0.93 2.00 5.00 4.00
73 33 4.00 0.60 3.00 5.00 4.00
74 33 3.45 1.21 1.00 5.00 4.00
75 33 3.94 0.55 3.00 5.00 4.00
76 33 3.15 1.35 1.00 5.00 4.00
77 33 3.91 0.57 3.00 5.00 4.00
78 33 4.03 0.94 2.00 5.00 4.00
79 33 3.39 0.92 2.00 5.00 3.00
80 33 3.45 0.92 2.00 5.00 3.00
81 33 3.27 1.16 1.00 5.00 4.00
82 33 3.79 0.48 2.00 4.00 4.00
83 33 3.24 0.82 2.00 4.00 4.00
85 33 4.12 0.91 3.00 5.00 5.00

239

Appendix E: Data Frequency of Qualtrics Responses
Question
number
# of times
rating of
"1" chosen
# of times
rating of
"2" chosen
# of times
rating of
"3" chosen
# of times
rating of
"4" chosen
# of times
rating of
"5" chosen
Total
number of
responses
8 1 1 12 11 8 33
9 1 0 3 23 6 33
10 0 1 16 0 16 33
17 0 4 0 9 20 33
19 1 14 7 11 0 33
21 0 13 11 7 2 33
23 0 0 0 7 26 33
24 0 0 0 8 25 33
25 14 0 11 6 2 33
26 0 0 6 16 11 33
27 5 4 11 13 0 33
28 0 0 4 10 19 33
29 5 4 5 18 1 33
30 0 0 0 5 28 33
31 0 0 0 6 27 33
32 0 1 0 5 27 33
33 0 0 14 18 1 33
34 0 0 10 12 11 33
36 0 0 0 12 21 33
37 0 0 0 8 25 33
38 0 0 1 16 16 33
39 0 5 6 7 15 33
40 11 5 16 0 1 33
41 0 0 2 16 15 33
42 0 0 7 21 5 33
43 0 0 4 14 15 33
44 0 0 6 11 16 33
46 0 1 9 1 22 33
47 0 0 5 4 0 9
48 0 4 2 5 22 33
49 0 0 0 11 22 33
50 0 1 1 21 10 33
51 0 7 10 16 0 33
52 0 0 7 26 0 33
53 10 10 12 1 0 33
55 5 21 2 5 0 33
240

Question
number
# of times
rating of
"1" chosen
# of times
rating of
"2" chosen
# of times
rating of
"3" chosen
# of times
rating of
"4" chosen
# of times
rating of
"5" chosen
Total
number of
responses
57 0 0 4 3 26 33
58 0 0 5 23 5 33
59 0 0 0 13 20 33
60 0 1 5 21 6 33
61 1 0 4 13 15 33
62 0 1 15 17 0 33
63 0 6 13 10 4 33
64 0 0 4 8 21 33
65 1 5 6 21 0 33
67 0 0 1 11 21 33
68 0 0 0 6 27 33
69 0 0 0 12 21 33
70 0 5 8 20 0 33
71 0 5 4 8 16 33
72 0 5 12 11 5 33
73 0 0 6 21 6 33
74 5 0 8 15 5 33
75 0 0 6 23 4 33
76 9 0 2 21 1 33
77 0 0 7 22 4 33
78 0 4 2 16 11 33
79 0 4 18 5 6 33
80 0 5 13 10 5 33
81 4 6 1 21 1 33
82 0 1 5 27 0 33
83 0 8 9 16 0 33
85 0 0 12 5 16 33

Abstract (if available)

Abstract According to 2014 data, underrepresented minorities (URM) made up 36.8% of America’s college age (18-24) population (NSF, 2017b), but only received 12.2% of PhD degrees (NSF, 2017a) and a scant 7.9% of all PhD degrees in the science and engineering fields (NSF, 2017a). The figures mean that URM received only a quarter of the science and engineering PhD degrees relative to their percentage makeup of America’s college age population. At UCSF though, the percentage of URM enrolled in UCSF’s basic and biomedical sciences PhD programs is 22% (UCSF Graduate Division, 2020a). UCSF was chosen for this evaluation study because the percentage of URM students enrolled in its basic and biomedical sciences PhD programs is nearly three times the national average and UCSF is classified as a “very high research activity” or R1 institution by the Carnegie Classification (Carnegie Classification of Institutions of Higher Education by Indiana University Center for Postsecondary Research, 2017). R1 institutions award the most science and engineering doctorate degrees (London et al., 2014). The study will utilize the Clark and Estes’ (2008) gap analysis framework to evaluate the knowledge, motivation, and organizational (KMO) influences from the student perspective. The research methodology uses a quantitative approach consisting of an online survey sent to URM students enrolled in the basic and biomedical sciences PhD programs. The study concludes with a series of recommendations to help UCSF ensure its continued success with a more diverse student body population in the basic and biomedical sciences PhD programs.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Big dreams lost: the influence of mentoring to reduce PhD attrition

PDF

Independent research internship experiences of underrepresented minorities in biological sciences: an interview case study

PDF

Enhancing a culture of research support for first-generation PhD students in social science fields

PDF

Influence of mentorship on career development for underrepresented minoritized students in STEM undergraduate research experiences

PDF

Underrepresentation of African Americans in master’s engineering programs

PDF

The relationship among gender, race/ethnicity, sense of validation, science identity, science self-efficacy, persistence, and academic performance of biomedical undergraduates

PDF

Underrepresented minorities in medicine in the United States: an innovation study to develop an effective holistic admissions process for the New School of Medicine

PDF

Factors influencing admissions counselors’ ability to increase the number of African American males in a private college: a gap analysis

PDF

The consequences of stereotype threat on Black and Latinx students in science and engineering

Asset Metadata

Creator Yeung, Kevin Hsing Tzu (author)

Core Title Increasing underrepresented minority participation in the STEM PhD programs

School Rossier School of Education

Degree Doctor of Education

Degree Program Organizational Change and Leadership (On Line)

Degree Conferral Date 2021-08

Publication Date 07/26/2021

Defense Date 06/02/2021

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

Tag KMO,OAI-PMH Harvest,PhD,San Francisco (UCSF),STEM,underrepresented minorities (URM),University of California

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Adibe, Bryant (committee chair), Melguizo, Tatiana (committee member), Torres-Retana, Raquel (committee member)

Creator Email kevinyang@gmail.com,khyeung@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC15622993

Unique identifier UC15622993

Legacy Identifier etd-YeungKevin-9890

Document Type Dissertation

Format application/pdf (imt)

Rights Yeung, Kevin Hsing Tzu

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 author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

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

Repository Email cisadmin@lib.usc.edu