University of Southern California Dissertations and Theses

Examining the factors leading to the continued underrepresentation of Latinas in STEM

(USC Thesis Other)

Examining the factors leading to the continued underrepresentation of Latinas in STEM

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Running head: UNDERREPRESENTATION OF LATINAS IN STEM 1

Examining the Factors Leading to the Continued Underrepresentation of Latinas in STEM

by

Jedidiah Izael Afable Lobos

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

August 2019

Copyright 2019 Jedidiah Izael Afable Lobos

UNDERREPRESENTATION OF LATINAS IN STEM 2

Acknowledgements
I am eternally grateful for the instruction that I received at the University of Southern
California. I would like to recognize the excellent professors at the Rossier School of Education
who helped to shape the foundation of my dissertation, specifically Dr. Kimberly Hirabayashi
and Dr. Shaun Harper. My time at the University of Southern California will be one that I will
cherish for the rest of my life.
Thank you to all of my close friends, old and new, who have offered emotional support
throughout this process. I would also like to thank my family. I thank Honey for her
encouragement and support to climb the educational ladder and strive for the best. I thank Papa
for giving me the strong will and determination to get things done. I thank Jezi for being a great
brother and confidant when I needed it the most. I thank Elizabeth for her words of
encouragement and for also offering support of my studies. I thank Sharon (MIL) for her
continued encouragement. I also thank Jannah and Adrian for both giving me the much-needed
breaks and support when I needed it. I would like to thank the people who are here in spirit,
namely Johoney, Tatay, Papy, and Papaw (FIL). I wish you could have been here to help me
celebrate this great accomplishment, but I know you are always there looking on.
This dissertation would not be possible if it were not for the love and encouragement of
my loving wife and kids. Genevieve has been supportive, patient, and devoted throughout this
process. She has been my rock during the hardest moments of this process. Josiah, Jorah, and
John- my hope for you is that one day you understand that you were always my inspiration for
everything I have done in my life and by far, you three are the best works I have ever co-created.

UNDERREPRESENTATION OF LATINAS IN STEM 3

Table of Contents
List of Tables 5
Abstract 6
Chapter One: Introduction 7
Background of the Problem 7
Statement of the Problem 9
Purpose of the Study 11
Importance of the Study 12
Limitations of the Study 13
Definition of Terms 14
Organization of the Study 14
Chapter Two: Literature Review 15
Introduction 15
Factors Influencing STEM Major Choice 15
Factors Influencing STEM Persistence Among 20
Hispanics and Women
Experiential Learning and Multiple Dimensions 23
of Identity
Critical Race Theory and LatCrit 26
Origins 26
Critical Race Theory 28
LatCrit 28
Critique of the Literature 31
Implications 32
Chapter Three: Methodology 34
Introduction 34
Research Questions 35
Sample and Population 36
Access/Entry 36
Instrumentation 37
Why Mixed Methodology? 37
Quantitative Method 38
Qualitative Method 38
Survey 39
Participants 39
Survey Protocol 39
Interview 40
Participants 40
Interview Protocol 40
Data Collection 41
Survey Distribution 41
Interview Approach 41
Data Analysis 42
Limitations 42

UNDERREPRESENTATION OF LATINAS IN STEM 4

Chapter Four: Results 44
Demographics 45
The Impact of Hispanic and Female Faculty Members 46
The Importance of Modeling 46
Factors That May Be Influenced by Modeling 51
The Effect of Perceived Difficulty in Math and Science 57
Attitudes Regarding Math and Science Classes 57
Students’ Thoughts Regarding the Perceived Difficulty 60
in Math and Science Classes
Summary of Results 63
Chapter Five: Discussion 66
Discussion of Findings 67
Implications for Practice 70
Future Research 72
Conclusions 73
References 75
Appendix A: Survey Protocol 80
Appendix B: Interview Protocol 82

UNDERREPRESENTATION OF LATINAS IN STEM 5

List of Tables
Table 1: Frequency tables for demographic data 45

Table 2a: Descriptive statistics: Latinas in STEM modeling 47

Table 2b: Correlation matrix: Latinas majoring in 48
STEM vs. gender/race modeling

Table 3a: Descriptive statistics for extrinsic motivating factors 51

Table 3b: Frequency tables for extrinsic motivating factors 52

Table 3c: Correlation matrix: Latinas majoring in 53
STEM vs. extrinsic motivation

Table 4a: Descriptive statistics for modeling and intrinsic 54
motivating factors

Table 4b: Frequency tables for intrinsic motivating factors 55

Table 4c: Correlation matrix: modeling vs. intrinsic motivation 56

Table 5a: Descriptive statistics regarding math and science 59
success and enjoyability

Table 5b: Frequency tables regarding math and science 59
success and enjoyability

Table 6a: Descriptive statistics for perceived difficulty in STEM 61

Table 6b: Frequency tables for perceived difficulty in STEM 62

UNDERREPRESENTATION OF LATINAS IN STEM 6

Abstract
The purpose of this study is to expand on the current knowledge surrounding some of the factors
that contribute to the underrepresentation of Latinas majoring in the STEM fields as well as
understand what theories may help explain what is causing the underrepresentation of Latinas in
STEM. The theoretical framework of this study involves the Model of Multiple Dimensions of
Identity, Experiential Learning theory, and LatCrit. This study employs a mixed methodology
approach using surveys and interviews to address the proposed research questions. Data
collection will take place over a two-month period. An online survey will be distributed to
students and interview participants will be chosen randomly. Interviews will be recorded and
subsequently transcribed. Pseudonyms will be used to ensure the anonymity of the participants
of this study. Data collected will be analyzed using software designed to facilitate the coding of
the data. The use of both quantitative (surveys) and qualitative (interviews) methods triangulates
the data and ensures that any issues of validity and reliability are dealt with. Results of this
study show that there are contrasting views on how Latinas viewed modeling and that prior math
and science success may not play a large role on whether these students remain in STEM.

UNDERREPRESENTATION OF LATINAS IN STEM 7

CHAPTER ONE: INTRODUCTION
The current Chair of the Democratic National Committee and the 26
th
United States
Secretary of Labor Thomas E. Perez once stated that “Our workforce and our entire economy are
strongest when we embrace diversity to its fullest, and that means opening doors of opportunity
to everyone and recognizing that the American Dream excludes no one” (Perez, 2014). The idea
of diversity is essential in all aspects of the United States workforce and is even more crucial in
the fields of math and science. Several pieces of literature stress the importance of growing the
Science, Technology, Engineering, and Mathematics (STEM) fields in the United States and to
expand the diversity found within it (Carnevale, Smith, & Melton, 2011; National Science
Foundation, 2017). Governmental agencies have expressed that if steps are not taken to address
the racial disparities plaguing post-secondary STEM education, the United States economy will
face a severe decline in strength and productivity (George, Neale, Van Horne, & Malcolm, 2001;
Sciences, Engineering, & Medicine, 2007).
Background of the Problem
While STEM fields include physical and biological sciences, computer science,
engineering, and mathematics, they are not limited to those fields and may include disciplines
like architecture (Carnevale et al., 2011; National Science Foundation, 2015). Data collected
from 1977-2011 revealing a continued gender disparity in STEM with 70% of males earning
STEM degrees versus 30% of females earning STEM degrees demonstrate that this is a problem
(Mann & Diprete, 2013). Furthermore, from 2002 to 2014, there were 11.3% more White males
than White females who intended to major in STEM fields as freshmen (National Science
Foundation, 2004, 2015). When examining White males from 2002 compared to White males in
2014, there was an 8.7% increase in freshman intention to declare their major in STEM (National
UNDERREPRESENTATION OF LATINAS IN STEM 8

Science Foundation, 2004, 2015).
In contrast, when examining Black males in the same period, there was only a 3.1%
increase in freshman intention to major in STEM (National Science Foundation, 2004, 2015).
This problem is necessary to address because although all of the data relating to minoritized
people in STEM appear to be trending upwards, the data show there is still a disparity in the
number of underrepresented minoritized people in the STEM fields (National Science
Foundation, 2015; U.S. News, 2015). The upward trend of the data may mislead people into
believing that there is no longer a problem of underrepresentation. President Obama stressed the
importance of diversity in the STEM fields by stating that science is all inclusive and that “we
want our classrooms and labs and workplaces and media to reflect that” (Earnest, 2015). The
United States Department of Commerce (DOC) statistics in 2011 further highlight why the issue
of underrepresentation by minoritized people in the STEM fields is of utmost importance.
Although underrepresentation in STEM may still be a problem at the college level, these
problems may not just end at the college level and may persist well into the post-collegiate
environment.
The underrepresentation of Latinas in the STEM fields may go well beyond being a
problem in the United States. There is a need to address and remedy this problem in STEM in
order to keep up with the increasing demand in STEM jobs in the United States but may have
global implications (Carnevale et al., 2011). In 2012, China awarded 23.4% of the world’s
Bachelor’s degrees in science and engineering compared to the 9.2% awarded in the United
States (National Science Foundation, 2016). These numbers along with the World Economic
Forum’s report on global competitiveness suggest that although we are still in the top 5% in
global competitiveness, we continue to fall behind in areas such as quality of math and science
UNDERREPRESENTATION OF LATINAS IN STEM 9

education, secondary education enrollment rate, and female participation in the labor force
(World Economic Forum, 2016). While researchers have addressed the continued
underrepresentation of Latinas in STEM education as well as in the STEM workforce, others
have shown that in some of the STEM fields, namely Biology, the underrepresentation is not as
pronounced (Jackson, 2013; Mann & Diprete, 2013). What is unclear in most of these studies is
what majors are included in Biology. Merely stating "biological sciences" may include
vocational career paths such as nursing, and while nursing is a highly respected important field
of study in medicine, it should be segregated from the data representing Biology majors in the
United States. Furthermore, in academic institutions such as the one highlighted in this study
that have a dedicated registered nursing major apart from biological sciences, there is a marked
difference in the number of students declaring themselves as nursing majors with 13.84% doing
so and 4.94% declaring themselves as biological science majors (Tumbleweed Tech, 2017).
While it is understandable this is only one academic institution out of many, and this data is
indeed not generalizable, it offers a glimpse of what may be happening if the data reported by
some researchers are not done so with a higher level of scrutiny.
Statement of the Problem
The literature review section of this dissertation discusses the underrepresentation of
minoritized people in STEM in a broad sense by focusing on racially minoritized people as a
whole as well as women. STEM education and, consequently, the STEM workforce in the
United States has a diversity problem (Carnevale et al., 2011; Mann & Diprete, 2013; National
Science Foundation, 2017). This dissertation is primarily focusing on the underrepresentation of
Latinas in STEM, but it is important to point out the overall problem of the underrepresentation
of minoritized people in STEM education. Recent United States government data shows that
UNDERREPRESENTATION OF LATINAS IN STEM 10

between 1995 and 2014 Hispanics as a whole earned only up to a ten percent share in the number
of STEM bachelor's degrees awarded, with the most significant amount being in engineering and
the biological and computer sciences (National Science Foundation, 2017). What is of note
regarding the data presented by the National Science Foundation (NSF) (2017) is that there is no
reported data regarding the number of doctoral degrees awarded to both Hispanic and African-
American populations.
Furthermore, the only comparison offered in regards to earned doctorates is between
Asian and White American populations (National Science Foundation, 2017). The NSF data
alone justifies the need to explore what is happening in Hispanic populations, at the very least.
When examining the data regarding STEM occupations in the United States, the results mirror
what is found in the education data; that is, that Hispanic scientists and engineers occupy less
than ten percent of the available STEM jobs (National Science Foundation, 2017). These results
may present a much more significant impact on academia. Women and other racial minoritized
people are not majoring in STEM fields at the same rates than their White counterparts
(Kokkelenberg & Sinha, 2010). This notion may have a lasting impact because if there is an
underrepresentation of women and racially minoritized people in roles such as mentors,
academic advisors, staff members, and faculty, then what type of role models will be available to
those students who identify with being a woman or minoritized person (Griffith, 2010)?
What is not made clear by some of the current literature is the reason why these problems
exist. Crisp, Nora, and Taggart (2009) make mention of environmental "pull" factors but do not
offer any theories or identity model that may be leading to their presence. While Crisp, Nora,
and Taggart (2009) mention a few of these factors, there seems to be a need to do a more
thorough examination of how these factors manifest themselves in the Hispanic community and
UNDERREPRESENTATION OF LATINAS IN STEM 11

even more importantly in the women found in these communities. One of the over-arching goals
of this study is to explore what may be leading to these environmental factors and how they may
play a role in whether or not Latinas make the decision to major in STEM and eventually
become part of the STEM workforce.
Purpose of the Study
This dissertation addresses the problem of the continued underrepresentation of Latinas
in the STEM fields at the college level. For this study, the term STEM will be limited to include
the "hard sciences" (namely Biology, Physics, and Chemistry), Biotechnology, Engineering, and
Mathematics. This study aims to expand on the current knowledge surrounding some of the
factors that contribute to the underrepresentation of Latinas majoring in the STEM fields. A
strong push will be made to particularly understand some of the "environmental pull factors"
described by Crisp, Nora, and Taggart (2009). This study will also examine how the model of
Multiple Dimensions of Identity presented by Jones and McEwen (2000) affects how Latinas see
themselves as STEM majors and as college students overall. Perhaps how students identify
themselves will uncover more of the aforementioned "pull" factors. Another essential theory that
will be examined in this study is that of experiential learning described by Kolb (1984). The
notion of one's experiences guiding their learning is intriguing and may provide better insight
into how the experiences of these women may guide their future decisions. In order to conduct
this study, the following questions will be addressed:
1. What impact will the presence of Hispanic and female STEM faculty members have on
Latinas majoring in STEM fields?
2. What impact will the perceived difficulty of math and science classes taken during the
first two years of college have on Latinas majoring in STEM fields?
UNDERREPRESENTATION OF LATINAS IN STEM 12

Importance of the Study
This study aims to expand upon what the current literature has already uncovered in
regards to the underrepresentation of Latinas in STEM education. While current research shows
that Latinas are on an upward trend in majoring in STEM, they still do not occupy a large share
of minoritized people majoring in STEM (National Science Foundation, 2017). This
underrepresentation is also seen in the STEM workforce and therefore, exploring the factors that
lead to why Latinas are not majoring in STEM is of utmost importance. While these reasons
alone are essential for this study, the simple reason that Latinas are underrepresented minoritized
people should be reason enough. The lack of data from the NSF regarding earned doctoral
degrees by Latinos, in general, is cause for concern (National Science Foundation, 2017). As
stated earlier in this chapter, the United States has a diversity problem in STEM, and while this
study certainly will not fix this problem, it can at least help shed some more light on the problem.
This study also intends to examine the educational theories that may help explain the presence
or absence of certain factors that are driving the underrepresentation of Latinas in STEM. Some
researchers have already begun to propose reasons for what might be causing this
underrepresentation, and this study intends to expand upon what is currently known (Buchmann
& DiPrete, 2006; Crisp, Nora, & Taggart, 2009; Griffith, 2010; Kokkelenberg & Sinha, 2010).
Notably, this study aims to find a theory or model that will help explain why Latinas are
underrepresented in STEM with the intent to use said theory or model to prevent this
underrepresentation in the future. If Latinas could be guided early on in their education about
the pitfalls that may lead them away from STEM, perhaps the data in the future will show a
significant increase in the number of Latina STEM majors. Establishing some theory or model
UNDERREPRESENTATION OF LATINAS IN STEM 13

to help inform future studies will then be just as important as uncovering the factors that lead to
the underrepresentation described here.
Limitations of the Study
The limitations of this study are explained in detail in Chapter 3, but the following are the
general limitations of this study:
1. Internal validity threats may be a concern for this study, but steps were taken in order
to address these concerns.
2. Time is a significant limitation of this study. Since the research informing this study
will be done for a small period, there is a chance that there will not be enough time to
do the study the way it probably should be done.
3. The sample size is an issue for this study. Again, the time allotted for this study is not
conducive to performing this study in such a way to where many participants could be
used.
Definition of Terms
The list below includes the critical terms used in this study. The definitions of these
terms are taken from the literature informing this study as well as from standard defined terms
from the Merriam Webster dictionary. The terms are as follows:
• Diversity. “The condition of having or being composed of differing elements”
(“Diversity,” 2003). In this study, the “differing elements” that are being highlighted
include biological gender, race, and ethnicity.
• Environmental pull factors. Factors that promote a “pulling away” of students into
the campus environments (Crisp et al., 2009).
UNDERREPRESENTATION OF LATINAS IN STEM 14

• Latina. “A woman or girl who is native or inhabitant of Latin America” (“Latina,”
2003).
• Hispanic. “Of or relating to the people, speech, or culture of Spain or Spain and
Portugal" (“Hispanic,” 2003). Generally, this term is seen as narrow compared to the
broader term of Latino, Latina, or Latinx.
Organization of the Study
Chapter One provides an overview of the study by discussing the problem being
examined and addressing why it is essential to perform this study. Chapter Two reviews the
pertinent literature that provides a background for this study. The chapter discusses the research
that has been done relating to issues regarding what is known about STEM major choice, STEM
persistence among Hispanics and women, and the gender and racial pay gap in STEM. Chapter
Two also discusses how Kolb's (1984) experiential learning, Jones and McEwen’s (2002) model
of Multiple Dimensions of Identity may play a role in how Latinas may view themselves in their
roles as college students majoring in STEM. Chapter two also presents Critical Race Theory,
specifically LatCrit, as a way to explain some of the challenges Latinas may face in education.
Chapter Three presents the research design involved as well as the methodology and
instrumentation used in this study. Chapter Four provides the results of this study and presents a
reflection on those results. Chapter Five addresses the meaning of the findings while discussing
what the implications of these findings are in STEM education. The chapter also presents future
research that may be necessary to expand upon what is known in order to address any needs
uncovered by this study.

UNDERREPRESENTATION OF LATINAS IN STEM 15

CHAPTER TWO: LITERATURE REVIEW

Introduction
This literature review focuses on three topics that are at the core of uncovering what may
be playing a part in the underrepresentation of Latinas in STEM education. The first section
describes some of the factors that may be playing a role in college major choice (including
STEM), mainly by Hispanics and women. The second section focuses on the already examined
factors that may be contributing to the persistence of Hispanics and women in STEM education.
The third section addresses the effect of the disparity of Hispanics and women in STEM
education on the workforce, mainly how the disparity plays a role in the gender and racial pay
gaps. The last section presents theories that may be central to understanding the problem that is
being examined in this study. While the literature presented highlights key topics that possibly
contribute to understanding why there is an underrepresentation of Latinas in STEM education, it
barely scratches the surface of what may be a deeper problem.
Factors Influencing STEM Major Choice
Ma (2011) proposed that college major choice is central to understanding the connection
between education and the different layers (i.e., racial, gender) that exist in the labor market.
Rather than putting attention to the more "internal" influences that may exist such as family and
peer influences, Ma (2011) examined that "macroinfluences at the societal level." Specifically,
the link between college major choice and group composition in different occupational fields
were researched. In this quantitative study, Ma (2011) echoes previously discovered data
showing that the number of women in engineering occupations pale in comparison to other
STEM fields, namely, the life sciences. Although it may seemingly be an obvious conclusion,
the researcher points out that students decide upon what to major in based on what they perceive
UNDERREPRESENTATION OF LATINAS IN STEM 16

to be a range of professions (related to that major). What is not often talked about in this type of
research is what Ma (2011) calls the “nativity” divide. In her study, Ma (2011) defines the term
“technical fields” as including “engineering, computer science, math/statistics, and physical
science” (Ma, 2011) and reports that foreign-born students are more likely to work in technical
fields, (as opposed to the social sciences) compared to those born in the country, but to foreign-
born parents. What adds to the intrigue in this finding is that foreign-born women are less likely
to work in the social sciences. If the idea that college major choice and occupation are
correlated, then there must be some reason why "native" students (especially women) are not
choosing STEM-related fields.
Interestingly, there is a divide in gender in occupation and college major choice between
life science and technical fields. Again, what the researcher uncovers is what is primarily known
or, at least, suspected- that men exist as the dominant group in the technical fields while women
have a more substantial presence in the life sciences. Ma (2011) further points out that when it
comes to college major choice (and consequently occupation), segregation based on sex is still
more severe when compared to race and nativity. This notion does not in any way take away
from the fact that there is still a racial divide in college major choice and occupation. According
to this research, there is race segregation occurring in college major choice among Asian,
Hispanic, and Black women compared to their White counterparts. This segregation is the same
for their male counterparts, but it is interesting to note that the divide is smaller. What was not
presented in these data was what the major choice was (i.e., STEM, humanities) and who
majored in which field. If the totality of this research is taken into context, one can surmise that
the Asian women probably majored in STEM (because of the previously reported data showing
that foreign-born women major in the technical fields) while those that are native-born did not,
UNDERREPRESENTATION OF LATINAS IN STEM 17

but this can only be assumed. What Ma (2011) does conclude is that the demographic similarity
plays a role in college major choice and ultimately occupational decisions. She also makes the
implication that college major choice has deep societal roots. This conclusion may be the most
critical part of this research because it helps to pinpoint where future research needs to provide
focus.
Mann and Diprete (2013) examined the development of STEM major choice between
genders as well as some possible reasons why these decisions are made. The authors presented
statistics showing bachelor's degrees awarded to women and men from 1977 to 2011. These data
show that while there is an upward trend in women in STEM, men still dominate the STEM
fields. One reason why there is such a gender discrepancy is attributed to the labor market. The
data show that there has been an increase in women in STEM, including women in engineering
(which has historically been one of the STEM majors with the least number of women). There
has been a recent slowdown in women entering these fields, however, even though the career
opportunities are present. It was noted in this study that while there has been a decrease in the
gender gap in math achievement, verbal test scores seem to continue to favor women. The
authors contend that this gender gap in verbal test scores may favor women majoring in a field
that they perceive they are better at (i.e., choosing the humanities over STEM). In order to study
the gender gap in STEM over the past few decades, the researchers used data from the National
Center for Education Statistics (NCES), which began collecting data in 1972. The NCES data
include four surveys: The National Longitudinal Study (NLS), the High School and Beyond
Longitudinal Study (HSB), the Education Longitudinal Study (ELS), and the NELS. Each one
of these surveys started with high school students and continued to post-secondary education.
The researchers of the study limited the analyses to students who went on to four-year degree-
UNDERREPRESENTATION OF LATINAS IN STEM 18

granting colleges and universities. The study analyzed the gender gap using a nonlinear
regression decomposition in one model and then, in order to observe changes in performance
measure returns throughout cohorts, used a logistic regression model with interaction terms. The
data show that in math and science coursework (both in high school and college), there is a
tendency for males to do better (even though in some cases the difference between genders is
small). The data also show that most STEM majors are men by a factor of two to two and a half
times. The authors contend that while the gender gap in math scores is lessening, it does not
explain why there is a widening gap in computer science and engineering. Mann and Diprete
(2013) present data that describe not only STEM as a whole, but also specific STEM fields such
as engineering, physical sciences and math, and the biological and agricultural sciences. These
data are relevant because one area that has seen growth in females in STEM is the biological
sciences. So, taken alone, this data may lead to some believing that there is no problem with
women in STEM. Along with the engineering data, however, it becomes clear that there is
something going on when it comes to women majoring in these STEM fields.
A study in 2006 on the changing gender composition of college majors during a thirty-
one-year period, was conducted by England and Li. Here, the researchers used quantitative data
to study the rate of gender segregation among college majors. England and Li (2006) noted that
during the first half of that thirty-one-year period, there was a decrease in gender segregation in
baccalaureate degree fields, and that the decrease could be attributed to a variety of factors
including a decreased number of women entering "traditional" fields of study (education and the
humanities). This period correlates with previously reviewed literature by Mann and Diprete
(2013), showing an upward trend of women entering STEM and implies that this is why England
and Li (2006) found this desegregation during the first half of the years studied. At some point,
UNDERREPRESENTATION OF LATINAS IN STEM 19

however, England and Li (2006) point out that desegregation stalled due in part to the fact that
while women began to enter non-traditional fields of study, such as STEM, the cohorts that
followed did not continue the trend. England and Li's (2006) analyses also uncovered an
interesting observation- that men were discouraged from entering a field of study that showed an
increase in women participation. In order to provide a foundation for their research, they also
present two theoretical perspectives on gender change in education. The first is the idea that our
culture is one that devalues women. They go on to say that everything associated with women-
fashion, names, fields of study are largely devalued. This devaluation of women, they claim,
was used to explain why women are paid less in occupations that they generally have filled.
England and Li (2006) further posit that this perspective implies a gender-related change that is
asymmetrical. They go on to explain that when men make a nontraditional choice of major,
there is a stronger stigma associated compared to women making nontraditional choices because
society places greater value in the fields that these women are entering. Another theoretical
perspective that England and Li (2006) offer is one that is proposed by Charles and Bradley
(2002), who describe gender-integrative change as existing along a vertical and horizontal
dimension. The vertical dimension is what Charles and Bradley (2002) describe as tertiary level
(the level of degree that someone obtains) and the horizontal dimension is the field of study that
someone chooses. Charles and Bradley (2002) present that cultural ideas about how genders are
"supposed to be" continue to remain in society. While England and Li (2002) fail to find a
reason as to why there was a stall in gender desegregation during the period studied, they do
present key findings that may lend to the general knowledge regarding attitudes towards women
majoring in non-traditional fields such as the more technical STEM fields.

UNDERREPRESENTATION OF LATINAS IN STEM 20

Factors Influencing STEM Persistence Among Hispanics and Women
The mixed methods study presented by Crisp, Nora, and Taggart (2009) explored the
various factors that may play a role in increasing (or decreasing) the interest of Hispanic students
in the STEM fields. Their interest may then lead to the decision on whether or not they should
pursue and complete degrees in STEM. The participants of this study were taken from an HSI
that had a high rate of granting doctoral degrees in the United States. The sample of students
was categorized based on gender, major type, enrollment status, first-generation status, and
ethnicity. In order to identify differences (if any) in specific student characteristics between
Hispanic and White students and STEM majors, chi-square and t-tests were calculated. The
researchers then used block sequential modeling to perform regression analyses using the
dependent variables (demographic variables, pre-college variables, college variables, and
environmental "pull" factors) and the independent variables (STEM and non-STEM). The
results showed that while there did not seem to be a deep relationship between Hispanic and
White students in terms of gender and ethnicity, for example, other relationships were found to
show significance. These relationships include ethnicity and financial support and ethnicity and
enrollment status, to name a few. Other results estimated that pre-college factors (Scholastic
Aptitude Test (SAT) math scores, high school percentile) and demographic factors (gender,
ethnicity) influence the students’ decision to major in STEM. It was also noted that females
were not only less likely than males to major in STEM, but were also less likely to change to a
STEM major once in college. This study helps to address two key areas in researching the
disparity of women and minoritized people in STEM. First, it provides an empirical model
showing that there are factors that affect Hispanic students' decision to major in STEM fields.
While these factors may also be found in several other works, this article mentions
UNDERREPRESENTATION OF LATINAS IN STEM 21

environmental "pull" factors. These "pull" factors include financial concerns, other reasons why
a student has to work part-time while in school (and in turn, attend school part-time), and family
responsibilities. The second area that this article helps to address is gender. The findings of this
research show that gender becomes a strong predictor in whether or not Hispanic students, and in
this case Hispanic females, major in STEM.
Kokkelenberg and Sinha (2010) focused on STEM and non-STEM students at State
University of New York (SUNY), Binghamton and examined what determined STEM student
success. A fixed effects estimator model was used where the fixed effect for each model was
high school. One result that the researchers found was that women do not hold any advantages
when majoring in STEM and Hispanics do not do as well as other ethnic groups in STEM.
Furthermore, the authors of this paper found evidence that there was a gender peer effect at the
university they were studying. In other words, if a class had more female students in a class, all
the females in that class would do well, especially in biology and mathematics courses. They
also found that the average grade for STEM courses was lower than for grades in other courses.
This research uncovers two critical ideas. First, it suggests that there may be some vicarious
learning or vicarious modeling at play with the female students doing well in science and
mathematics classes because of the presence of other female students in the classes. Second, it
highlights that in many cases, STEM tends to include major fields of study that are just harder
overall. That being the case, perhaps more research needs to be done to examine factors that
include preparation of female STEM majors before college. There was a suggestion that was
made regarding the idea that STEM instructors grade their students harder than instructors in
other disciplines. Although they discussed this in their results, there was mention that they could
not correlate this to students being discouraged from the STEM fields because of this higher
UNDERREPRESENTATION OF LATINAS IN STEM 22

grading standard. Kokkelenberg and Sinha (2010) also provided data that show there is a
discrepancy with women in STEM. Although this study was limited in that it was done at one
particular university (as opposed to a multi-college or multi-university study) and not using data
collected from several sources, it does reinforce what has been shown in other works.
The empirical study conducted by Griffith (2010) focused on whether women and
minoritized people remain in the STEM fields throughout college. The study examined several
factors that may affect the choice women and minoritized people make to leave STEM and major
in another field of study. One of these factors included whether or not the academic institution
that these students enrolled in made a difference. Data from the National Longitudinal Survey of
Freshmen (NLSF) and the National Education Longitudinal Study (NELS) provided descriptive
statistics including gender, minoritized status, high school grade point average, and the number
of STEM advanced placement classes taken in high school to provide a framework to observe
the persistence of females and minoritized people in STEM after their second year of college.
The study then examined how the choice of major is affected by the educational preparation and
experiences with STEM field departments of college students. In order to do this, a logit model
was used to assess the probability of women and minoritized people switching to a STEM major
and if either group persisted as STEM majors. The model also included the percentage of female
faculty in STEM and how that may affect the probabilities mentioned. The results of this study
showed that persistence in STEM is most affected by student experience during the first two
years of college. It also showed that if the institution has more undergraduates compared to
graduates, it will lead to greater persistence in STEM. The data showed that institutions with
more female faculty in STEM led to lesser persistence in STEM; however, the authors pointed
out that this may be a bit misleading because factors such as female faculty rank were not
UNDERREPRESENTATION OF LATINAS IN STEM 23

accounted for. This work points out that there is a problem with females and minoritized people
and their persistence in STEM. This "leaky" STEM pipeline from high school to college is
talked about in many other works, but Griffith (2010) presented statistics that not only show how
"leaky" it is but also describes some factors that may play a part in producing this "leak." The
idea that the institution serving the students may have something to do with STEM attrition rates
is intriguing, especially since it may help tie in other research that suggests that Hispanic Serving
Institutions play a role in increasing STEM persistence in minoritized populations, such as
Hispanics.
Experiential Learning Theory and Multiple Dimensions of Identity
Two main development theories are explored in the literature above. First, there is
evidence in Griffith (2010) and Kokkelenberg and Sinha (2010), that learning style is playing a
factor in the underrepresentation of women and minoritized people in STEM. David Kolb
(1984) presents the experiential learning theory of development as a way to expand upon what
others such as Dewey, Lewin, and Piaget have offered in the past to explain how people learn.
Although Griffith (2010) and Kokkelenberg and Sinha (2010) make mention of modeling
behavior in their studies, it is the overall experiences that these students have that are helping to
shape how they view their futures in the STEM fields. If their families and peers (Crisp, Nora,
and Taggart (2009), Griffith (2010), and Kokkelenberg and Sinha (2010)), are genuinely
contributing to their persistence in STEM, then perhaps the way they are contributing is by
affecting how they learn the material. Kolb (1984) argues that learning development is not
solely cognitive and that the environment (the experiences gained) play a significant role. If the
environment that these students are in does not encourage learning, then how are these students
expected to learn. What is unknown in these studies are the characteristics of the peer groups
UNDERREPRESENTATION OF LATINAS IN STEM 24

and the family background surrounding the students in question. Perhaps there needs to be more
in-depth inquiry not just on the self-efficacy of the individual students but instead on the people
surrounding these students. I would hypothesize that the peers surrounding the successful STEM
minoritized people are either STEM minoritized people themselves or are at least successful
college students who have characteristics of being “good” students. According to Kolb (1984),
learning is happening at the specialization stage because this is where the students learn how to
learn. This stage is where students move from taking in the material and regurgitating it on an
exam to using the learned material and applying it to higher-level thinking.
The other development theory seen in the literature presented is the development of
social identities, specifically the model presented by Jones and McEwen (2000) of multiple
dimensions of identity. Jones and McEwen (2000) present identity as being influenced by
several factors while the core (the personal attributes, characteristics, and identity) is the
perception of how a person views their "inner-self." Interestingly, Kolb's experiential learning
theory of development fits very much in line with the multiple dimensions of identity modeled
by Jones and McEwen (2000) in that there is this idea that external forces (the environment) play
a role in shaping learning (in Kolb's theory) and identity (in Jones and McEwen). In each of
these studies, there seems to be an undertone of identity development heard. In Crisp, Nora, and
Taggart (2009), culture, gender, and perhaps even class, is central to the identity of many of the
Hispanic women in this study. The "environmental pull factors" described by Crisp, Nora, and
Taggart (2009) are linked to how these women see themselves. In Hispanic cultures, there is an
expectation for women to help contribute to the well-being of the household, especially in
households where the socioeconomic status is low. This expectation is very different from what
is expected of Hispanic males who are not expected to contribute as much, yet are expected to do
UNDERREPRESENTATION OF LATINAS IN STEM 25

what it takes to succeed in obtaining a well-paying job. Here again, culture plays a role in the
development of the Hispanic male identity while gender may not be as salient as it is for their
counterparts. Kokkelenberg and Sinha (2010), gender is affecting the development of identity.
Since there was evidence that vicarious modeling was taking place, it is clear that the female
STEM majors identify themselves as females in a male-dominated field. Here, how they view
not only there "female-ness" but also how they perceive STEM fields as a class of people (male-
dominated, intelligent) plays a strong role in success. These students viewing other women
(STEM faculty members) in that "class" is positively affecting the success seen by the female
STEM students. Griffith (2010) focused on the institution's role in STEM persistence but what is
not explicit is the "student experience" during the first two years of college. While this study is
quantitative and provides great empirical data to help test theory, it lacks what is necessary to
understand how women and minoritized people "see" themselves during the first two years of
college. At the very least, however, race and gender are playing some factor when it comes to
STEM persistence. What is interesting to note is that the conclusion that Griffith (2010) makes
is one rooted in the first two years of college. During these years, one can argue that a person's
identity may start to take shape because of many of the sociocultural conditions that a person is
experiencing. Lastly, the factor that may be shaping identity in Mann and Diprete (2013) at first
appears to be solely gender; however, I contend that class is probably a stronger factor. In this
study, there is a lessening in the gender gap in some fields while in others, the gap is widening.
Mann and Diprete (2013) do not explain this phenomenon, but one can argue that it may have
something to do with how the women in the study perceive the class of people in certain fields.
While gender is playing a strong role in persistence in STEM overall, perhaps what is happening
in certain STEM fields is that women view some of these fields as exclusive to certain people, or
UNDERREPRESENTATION OF LATINAS IN STEM 26

in this case certain genders. I argue that in this case, maybe there is an intersection in the
multiple dimensions of identity model between gender and class, where they are almost "felt" as
being the same by the women in this study.
Critical Race Theory and LatCrit
Origins. The origin of Critical Race Theory (CRT) is found in the Critical Legal Studies
movement. The Critical Legal Studies (CLS) movement in the 1970s was advanced by a group
of academics who were devoted to “reappraising the merits of the realist tradition in legal
discourse” (Tate IV, 1997). This “legal realism” stressed that applying behavioral sciences and
statistical methods to legal analysis would lead to better legal thought and social policy (Tate IV,
1997). Delgado (1987) criticized the "informality" of CLS and stated that CLS lacks "a political
and psychological theory of racism." It is through this criticism that Delgado (1987) asserts that
"CLS does not provide what minorities seek." This assertion is one of the arguments along with
others presented by Delgado (1987) that led to the separation between CLS and CRT (Tate IV,
1997).
Critical Race Theory. CRT relies on the integration of the experiential knowledge of
critical race theorists into the analysis of the law and this, according to Tate IV (1997), has given
importance to the idea of voice in critical race literature. The idea of voice is essential in CRT
because, in order to understand the CRT literature, the reader needs to understand the CRT
contributor (Tate IV, 1997). Three people who have made significant contributions to CRT
include Derrick Bell, Kimberlé Crenshaw, and Richard Delgado (Tate IV, 1997). According to
Tate IV (1997), Bell’s three major arguments analyzing racial patterns in American law-the
constitutional contradiction, the interest-convergence principle, and the price of racial remedies-
differ than the message found in established race scholarship. While each of these contributes to
UNDERREPRESENTATION OF LATINAS IN STEM 27

the foundation of CRT, the interest-convergence principle is one that has strong roots in
education. It proposes that “the interest of blacks in achieving racial equality will be
accommodated only when it converges with the interests of whites” (Bell Jr, 1980). In this
article, Bell (1980) argued that the Brown v. Board of Education decision was because the
interests of both races converged, making the decision expected. Kimberlé Crenshaw is regarded
as an early founder of CRT as a scholarly movement (Tate IV, 1997). One of Crenshaw’s major
contributions to CRT was the idea of intersectionality (Tate IV, 1997). According to Crenshaw
(1993), social issues surrounding gender likely minimized gender, race, and class issues.
Although Crenshaw (1993) stated that her focus was on race and gender, issues involving class,
sexual orientation, color, and age should also be examined using the intersectionality framework
that she proposed. Crenshaw (1993) defines structural intersectionality as "the way in which
women of color are situated within overlapping structures of subordination." She goes on to
state that the material consequences of how the multiple hierarchies of women of color interact
further define structural intersectionality Crenshaw (1993). Richard Delgado is regarded as a
contributor to CRT who was there at its historical and conceptual origins (Tate IV, 1997). One
of Delgado’s most important contributions to CRT involves the roles of story, counterstory, and
the “naming of one’s own reality” (Tate IV, 1997). Delgado (1987) noted that it is common for
people of color to exchange stories concerning ways of dealing with racism and other racial
matters. These contributors have helped shape CRT into the theory that it is today and according
to Harper, Patton, and Wooden (2009), while there is not a presence of an exact definition for
CRT, many scholars agree that seven tenets are central to the theory:
1. Racism is a normal part of American life and is difficult to eliminate.
UNDERREPRESENTATION OF LATINAS IN STEM 28

2. Colorblindness “leads to misconceptions concerning racial fairness in institutions”
and therefore is rejected by CRT.
3. CRT allows the promotion of the “voices” of people of color.
4. CRT recognizes interest convergence.
5. Revisionist history exists. In some cases, the historical record of the United States is
not as accurate, especially in regards to the experiences of minorities, as it is
presented. Therefore, a level of scrutiny must be used when examining the historical
“facts” that are presented.
6. CRT “relies on Racial Realists." These are individuals who understand that race is
not just a social construct but is also a way for society to assign levels of benefits,
privilege, and social status.
7. “CRT continuously critiques claims of meritocracy that sustain white supremacy
(Bergerson, 2003 as cited in Harper et al., 2009).
LatCrit. While there is no actual "author" of LatCrit, its origins can be tied to the work
of Delgado (1987) and Valdes (1996). Richard Delgado is regarded not just as a contributor to
CRT but also a critic of CLS (the foundation upon which CRT is built) (Delgado, 1987; Harper
et al., 2009; Tate IV, 1997). In the concluding paragraphs of one of his articles on the matter,
Delgado states that "CLS does not provide what minorities seek" (Delgado, 1987). With this
statement, along with other authors can be seen as the starting point of the many "RaceCrits” that
have been established since. Valdes (1996) posed the question “…Is Critical Race Theory a
project of or for Latinas/os…should it be, can it be" and continued to explain that although the
first anthology to CRT was edited by several scholars including Richard Delgado (a Latino), the
authors of that anthology were "primarily Black, heterosexual men." In this same body of work,
UNDERREPRESENTATION OF LATINAS IN STEM 29

Valdes (1996) makes mention of another LatCrit scholar, Juan Perea. In Suggested Responses to
Frequently Asked Questions about Hispanics, Latinos and Latinas, Perea (1996) poses four
questions and suggested responses to these questions. Perea (1996) also describes "episodes" for
the last two questions which illustrate the problematic nature of the questions posed. The third
question is a "series of questions and assumptions regarding the proper place for Hispanics,
Latinos, and Latinas" (Perea, 1996). In one episode, Perea (1996) describes one of Perea’s
students (a brown-skinned Mexican-American woman) who was approached by “a white Anglo
woman” who after asking the student if she was lost (the student arrived on the law school
campus to acquaint herself with the surroundings), then asked if she was looking for a cafeteria
job because the employment office was at a different location”. All the episodes presented
related to this question share a theme, according to Perea (1996): "the proper place of Latino
people in an Anglo-constructed society." In the case described above, Perea (1996), from the
line of questioning received by his student, concluded that his student’s “place” was in the
cafeteria, not studying law. Valdes (1996) posits that the scholars that he presents in his work,
including Perea (1996), “demonstrate how Latina/o legal scholarship can be an activist
undertaking responsive to the historical and contemporary conditions of Latinas/os in this
country." In regards to the possibilities of LatCrit theory, Valdes (1996) stated that "If Latina/o
legal scholarship can help unpack…the material conditions that affect Latina/o-identified
individuals…to empower and improve Latina/o positions and interests, we will have performed a
great service".
In 2001, 6% of Chicana/os acquired a baccalaureate degree compared to 23% of their
White counterparts (Yosso, Villalpando, Delgado Bernal, & Solórzano, 2001). Yosso et al.
(2001), found the need to explore the roles of Chicanas and Chicanos, specifically in education.
UNDERREPRESENTATION OF LATINAS IN STEM 30

In this work, the authors present five themes that form the perspectives, methods, and pedagogy
of CRT in education. These themes are extensions of the tenets that have been accepted by most
scholars of CRT:
1. The Intercentricity of Race and Racism: This posits that CRT in education puts race
and racism front and center while putting a focus on the intersections other forms of
subordination and racism.
2. The Challenge to Dominant Ideology: CRT in education challenges the notions of an
educational system that believes in colorblindness, race neutrality, and equal
opportunity.
3. The Commitment to Social Justice: CRT in education seeks to advance the idea that
challenges us to eliminate racism and empower those groups that have been
subordinated.
4. The Centrality of Experiential Knowledge: CRT in education acknowledges the
strength and importance of the voices of People of Color, and this is important to
understanding and analyzing the racial subordination found in education.
5. The Interdisciplinary Perspective: CRT in education draws from different disciplines
to analyze racism and other forms of subordination found in education.
One of the main highlights of Yosso et al. (2001), is a concept made initially salient by
Richard Delgado- storytelling and counterstorytelling. This concept as well as other CRT and
LatCrit concepts regarding intersectionality and experiential knowledge, merge with the
frameworks as mentioned earlier presented earlier in this chapter- Kolb's Experiential
Knowledge Theory (1984) and the Multiple Dimensions of Identity of Jones and McEwen
UNDERREPRESENTATION OF LATINAS IN STEM 31

(2000). The research must be looked at through the lenses of these theories in order to analyze
the responses of the Latinas that are the focus of this study.
Critique of the Literature
While the literature addresses some of the factors that are contributing to the
underrepresentation of women and minorities in STEM, it fails to address what may be key
contributors to this underrepresentation. While Kokkelenberg and Sinha (2010) and Griffith
(2010) make mention of the possible importance of female STEM faculty members serving as
role models, they do not explore the female STEM students and why they may view them as role
models. Are the students viewing these instructors as women who have successfully “broken
through” a male-dominated field? Do the students identify themselves as female students,
STEM students, or both? Much of the literature concerning underrepresentation in STEM will
focus on these factors but do not make much mention of how the students identify themselves.
Perhaps if more literature focuses on what is contributing to the establishment of these students’
identities, it may uncover more about why there is an underrepresentation in the STEM fields.
Another piece of the literature that may be misleading is how STEM fields are
represented. In much of the data that is presented by these authors, STEM fields like the life
sciences are not adequately defined. Specifically, vocational fields such as nursing may be
included as a “life science” but is better represented as an allied health field. The problem here
is that including allied health fields may skew the data in such a way that it shows that there does
not seem to be an underrepresentation of women and especially Latinas in the life sciences.
Perhaps disaggregating the life science data to highlight those who are majoring in the life
sciences with the intent of earning a Ph.D. will present a more accurate picture of the trends that
are being seen in the STEM fields. This same criticism can be made of research dealing with the
UNDERREPRESENTATION OF LATINAS IN STEM 32

gender pay gap. Again, the life science data are ill-defined. In this case, especially, it may
present a significant problem because, within the life sciences, there are jobs that do not require a
bachelor's degree or higher. Because there are jobs that do not require any formal degree, it
would be difficult to correlate any STEM major data with STEM jobs data. Therefore, the
analysis of the STEM workforce data may be of little use if the goal is to observe the real effects
of racial and gender underrepresented STEM majors.
Implications
Numerous studies have documented the underrepresentation of women and racial
minorities in STEM, but few of them detail the student development theories that may be
contributing to this underrepresentation. In many of these cases, the conclusion to the study is
usually that there need to be more studies done to determine why there is an underrepresentation
or that the authors do not understand why they are getting the findings that they are obtaining.
Maybe instead of just examining these issues, there needs to be a focus on why these students are
students, to begin with. Maybe the question that needs to be asked has to do more with how they
see themselves as a person of color, as a woman in college, or just as a STEM major in general.
These studies can help to uncover some of the deeper elements that are not explicitly mentioned,
such as how a student learns or why some students feel more motivated when they see similar
models in STEM (i.e., faculty members teaching STEM classes, minorities teaching STEM
classes). Once more studies can be done focusing on these elements, then some of the other
elements contributing to underrepresentation can be studied. Culture and the role that family
influence plays on the success of these students in the STEM fields may be better understood
once we know how the students identify themselves as people and as student learners. It may be
easy to point out that families, friends, and teachers will influence what one does as a student,
UNDERREPRESENTATION OF LATINAS IN STEM 33

but the stricter task may be determining why and, perhaps more importantly, when this influence
occurs.

UNDERREPRESENTATION OF LATINAS IN STEM 34

CHAPTER THREE: METHODOLOGY

The purpose of this chapter is to outline the methodology used in this study. This chapter
will discuss the research design used, the participants used, as well as the instrumentation used in
this study. The sample and population details and the limitations of the study, including any
threats to internal validity, will also be discussed in this chapter.
Introduction
It is well documented that women and minorities are underrepresented in the STEM
fields (Griffith, 2010; Mann & Diprete, 2013; National Science Foundation, 2017). One
particular group that is severely underrepresented are Latinas. While data are showing that there
is an underrepresentation of this minoritized group (National Science Foundation, 2017), the
factors that are causing this underrepresentation are not entirely clear. Authors have attempted
and have succeeded in identifying some factors (Crisp et al., 2009). However, there needs to be
a stronger effort to explore areas that have not been thoroughly examined. Data collected from
1977-2011 revealing a continued gender disparity in STEM with 70% of males earning STEM
degrees versus 40% of females earning STEM degrees demonstrate that this is a problem (Mann
& Diprete, 2013). For well over thirty years, we have seen women closing the gender gap in
STEM; unfortunately, the gap still exists and is starting to widen in some STEM fields, including
engineering and computer and information sciences (Crisp & Nora, 2012; Mann & Diprete,
2013). What makes matters worse is that it does not end in STEM education. This
underrepresentation has bled over into the workforce with minoritized women, especially
Latinas, occupying only 2% of STEM occupations (National Science Foundation, 2017). Also,
according to the United States Department of Commerce (DOC), women working in the STEM
fields earn $0.86 to every dollar that men do (Beede et al., 2011). So, this problem of
UNDERREPRESENTATION OF LATINAS IN STEM 35

underrepresentation needs to be addressed if we want to ensure an equitable workforce
eventually.
Moreover, the literature reviewed presents the case that this is a problem that goes
beyond what is already known. In order to delve into uncovering the causes for the disparity that
surrounds Hispanics and women, there needs to be research that fills the gaps in the literature.
Most of what has been obtained in trying to answer the question of gender and racial disparity in
STEM has been quantitative in its approach. Here, a qualitative approach is used to help paint a
more accurate picture of why this problem still exists.
Research Questions
There may be factors found in the college environment that may play a significant role in
whether or not Latinas decide to remain STEM majors. One of these factors may be related to
one of the principles of Social Cognitive Theory, which states that “modeled behavior is more
likely to be adopted if the model is credible, is similar (e.g. gender, culturally appropriate), and
the behavior has functional value” (Hirabayashi, Personal Communication, October 25, 2013).
Perhaps having someone that this group (Latinas) can look up to may lead to an increase in
learning, motivation, and self-efficacy.
Another factor may involve the perceived difficulty in the math and science classes
required to be a STEM major. Crisp, Nora, and Taggart (2009) have shown that high school
achievement in math and science does have a corollary effect on college STEM success in
Hispanics. If this same type of phenomenon is found at the college level, perhaps it would
require more attention. The following research questions associated with it is are stated below:
1. What impact will the presence of Hispanic and female STEM faculty members have on
Latinas majoring in STEM fields?
UNDERREPRESENTATION OF LATINAS IN STEM 36

2. What impact will the perceived difficulty of math and science classes taken during the
first two years of college have on Latinas majoring in STEM fields?
Sample and Population
Tumbleweed Tech (TT) is a community college founded in the early 1900s that serves
approximately 18000 students. Their vision is to provide a quality education that will help build
futures and improve the lives of the students who attend the institution. TT is a Hispanic Serving
Institution (HSI) in Arecaville, California with a five-member board of trustees that is
responsible for hiring the president of the college, giving authority to the president, representing
the interests of the community, monitoring how the college is performing as well as the quality
of the education that the students are receiving. The president, then, has the responsibility to
ensure that the college is performing to the standards outlined in the mission statement and the
college's vision. TT has been chosen as the site to draw my participants from because it is
convenient and represents the population that is being focused on in this study. According to
data collected and presented in a fact book released by TT, the student demographic consisted of
51.5% Hispanics and 58.8% females in 2017 (Tumbleweed Tech, 2017). The data presented in
the TT fact book (2017) facilitate the need of searching for participants for my interviews. The
target population for this study will be Latinas who have currently identified themselves as or
who entered college as STEM majors.
Access/Entry
The approval process consisted of formally requesting access through the Department of
Institutional Effectiveness, Research & Planning. The dean of math, science, and engineering at
TT was very supportive not just of my graduate studies but also of the work that I conducted, and
therefore I do not believe that the process of obtaining approval to begin this study was difficult.
UNDERREPRESENTATION OF LATINAS IN STEM 37

A research request form needed to be filled out and was submitted one month before beginning
this research. Research participants for this study were chosen at random from introductory
mathematics, chemistry, biology, physics, and engineering classes. The instructors of these
classes were asked if they would be amenable to offering the survey to their students. The
instructors had the opportunity to decline, and there was no way for me to know which
instructors decided to participate. None of the classes where the survey and interview
participants were chosen from were classes where I was listed as the instructor of record. It was
also ensured that the students understood that there was no obligation to participate in the survey
and that the resulting interviews would be chosen based on the results of the survey. In all cases,
the participants were informed that the results of the surveys were only used for this study and
that their identities would be kept anonymous throughout the study.
Instrumentation
Why Mixed Methodology?
Much of the current research that has been done to address the underrepresentation of
Hispanics and women have been mostly quantitative, and further exploration of these particular
factors may not be quickly done quantitatively because they may not be easily measured.
Researchers including Crisp, Nora, and Taggart (2009) examined factors that they described as
being “environmental pull factors” that may be contributing to the underrepresentation of
Hispanics (both men and women) in STEM. In order to uncover more of these factors as well as
add layers to the existing data that has been uncovered, a mixed methodology approach will be
employed in this study. Mixed methodology “resides in the middle of this continuum because it
incorporates elements of both qualitative and quantitative approaches” (Creswell, 2014, p. 3).
This will involve collecting both open-ended and closed-ended data and their subsequent
UNDERREPRESENTATION OF LATINAS IN STEM 38

analysis (Creswell, 2014). The use of mixed methods also serves “as a check on one another,
seeing if methods with different strengths…support a single conclusion” (Maxwell, 2013, p.
102). Additionally, the mixed methods approach “provides a more complete understanding of a
research problem than either approach alone” (Creswell, 2014, p. 4).
Quantitative method. Creswell (2014) states that quantitative approaches "tests or
verifies theories or explanations" and "relates variables in questions or hypotheses."
Furthermore, quantitative testing is deductive, taking known information and using the
information to study cause and effect (Creswell, 2014). Quantitative methods analyze numerical
data and are designed before the study. In this study, a survey was used to help determine which
respondents were chosen for the qualitative method mentioned in the following paragraph.
Surveys provide a “quantitative…description of trends, attitudes, or opinions of a population by
studying a sample of that population” (Creswell, 2014).
Qualitative method. Merriam and Tisdell (2016) state that “qualitative researchers are
interested in understanding how people interpret their experiences…and what meaning they
attribute to their experiences” (p.6). Also, qualitative methods are inductive (Merriam & Tisdell,
2016), meaning that the data collected in qualitative research is used to generate theory.
Qualitative research uses data that builds from the ground up, using personal interactions or
documents as the collecting tools as opposed to the standardized instruments found in
quantitative research. Because there is not currently a theory that describes the phenomenon of
underrepresentation of ethnic and gender minorities in STEM, qualitative methods were not just
appropriate but necessary for this study.
Furthermore, qualitative research also allows for a certain amount of evolution in the
research design. As Creswell (2014) notes, one of the characteristics of qualitative research is its
UNDERREPRESENTATION OF LATINAS IN STEM 39

new design. Creswell (2014) further states that “some or all of the phases of the (research)
process may change…after the researcher enters the field” and that the “key idea behind
qualitative research is to learn about the problem…from participants” (p.186). In this study,
interviews were performed on individuals who are chosen based on the responses obtained from
the surveys as mentioned earlier.
Survey
Participants. The survey was distributed to students who were enrolled in introductory
math and science classes. In this sense, the sampling was random in that anyone could have
been chosen equally since participants chose whether or not they wanted to participate. Creswell
(2014) recommends a random sample because “each individual in the population has an equal
opportunity of being selected…” (p. 158). Creswell (2014) further states that this randomization
“…provides the ability to generalize to a population” (p.158).
Survey protocol. The survey administered included questions previously asked by Mann
and Diprete (2013) as well as original questions obtained during the preparation of this study.
While Mann and Diprete (2013) employed a survey that was very relevant to their study, only
part of that survey was relevant to this one and therefore was the reason for the construction of a
new survey. Survey questions included declared major, an assessment of life goals, a self-
assessment of mathematics readiness and confidence, and a self-assessment of overall science
readiness and confidence. Each survey also had a section asking the respondent if they would be
willing to be interviewed if chosen for the study. The survey consisted of a four-point response
Likert scale that was disseminated to the students with the choices being strongly disagree,
disagree, agree, and strongly agree, with each choice numbered one through four (with the most
negative choice numbered one and the most favorable choice numbered four). This type of
UNDERREPRESENTATION OF LATINAS IN STEM 40

Likert scale will produce a forced-choice response by leaving out the choice of “neutral”
(Kurpius & Stafford, 2006).
Interview
Participants. According to Merriam and Tisdell (2016), probabilistic sampling is not
“necessary or even justifiable in qualitative research” (p.96). Therefore, the most common form
of non-probabilistic sampling will be used in this study- purposeful sampling. Purposeful
sampling was employed by choosing interview participants based on the survey responses
collected. Those who were Hispanic, female, and had at least at some point majored in STEM,
were selected. These particular participants were chosen in order to determine if there were
factors that were either pushing them away from the STEM fields or factors that were leading
these two groups to leave the STEM fields. As Merriam and Tisdell (2016) put it, "interviewing
is necessary when we cannot observe behavior…or how people interpret the world around
them”. Interviewing these two groups of students gave me the opportunity to obtain information
that otherwise would not have been obtained because as the interviews went on, there was a
chance for there to be a natural evolution of the questions based on the responses obtained. The
goal was to try and uncover factors that influenced how these students perceived themselves,
how they perceived college, and how they perceived the STEM fields. Interviewing them was
the most appropriate way to uncover that information.
Interview protocol. The interview protocol was explicitly created for this study and the
structure of my interview protocol was purposeful in the sense that the questions in the protocol
dealt with the students’ feelings towards having females and/or Latinas as math and science
teachers and the students’ perception on the difficulty of math and sciences classes taken in
college. These questions were asked to accomplish two critical things. First, the questions were
UNDERREPRESENTATION OF LATINAS IN STEM 41

asked to help ease the respondents in the interview and establish comfort. The initial questions
were designed to help the respondents feel open enough to talk freely by asking them questions
that were not difficult to answer. One of the critical concepts that were pointed out by Crisp,
Nora, and Taggart (2009) was that the environmental “pull” factors include more societal and
family factors and may provide a better way to understanding why the other factors are
contributing to the underrepresentation of this particular group. This is why some of the
questions asked related to family dynamics. The second part of the questions was designed to
obtain information that dealt with the factors that may be associated with how Latinas saw
women, specifically Latina instructors, as models that could serve as inspiration.
Data Collection
Survey distribution
A link to the survey was distributed to the instructors of the various introductory classes,
and those instructors distributed the link to their students by e-mail. Since all students who
attend the college were given a campus e-mail address for official college communication, it was
the most efficient way to distribute the survey. Using a web-based survey method also facilitated
the tabulation of the survey data. This was very important since the survey was used to choose
participants for the interview that followed. In order to increase the response rate, incentives
were offered. All participants of the survey were entered into a raffle for a gift card, and the
winner was chosen at random.
Interview approach
All interviews were conducted at a conference room at TT. The interviews were
conducted with the door closed so that there were no interruptions by anyone passing by the
room. I was dressed professionally (shirt, slacks, and tie) since I was already at work. The
UNDERREPRESENTATION OF LATINAS IN STEM 42

interviews were conducted during the Winter intercession, a time when the campus is not
typically too busy. This made it easy to conduct interviews without the usual busyness that
coincides with the Spring and Fall semesters. Each interview lasted between twenty and twenty-
five minutes with each respondent being encouraged to talk for as long as they want. Each
respondent was informed that the interviews were audio recorded but were assured that their
identities would be kept anonymous both in the recordings and on the transcripts.
Data Analysis
In regards to the quantitative aspect of this study, the data analysis included an
investigation of both the independent variables (including gender and ethnicity) and dependent
variables (including self-efficacy, motivation, and declaring as a STEM major). The data
analysis included descriptive analyses and correlation analyses. In order to examine
relationships in the qualitative data as well as to facilitate the coding of the data obtained from
the interview process, transcripts of the interviews were analyzed using NVivo software. NVivo
software (https://www.qsrinternational.com) is a data analysis software tool used for qualitative
and mixed methodology research. The software allows the user to store and sort notes taken
from interviews and interview audio files, categorize data to facilitate coding of themes, and it
allows the import of multiple sources of data.
Limitations
In order to strengthen the reliability of this study, multiple sources of data were collected,
and triangulation of data is a way to ensure the internal validity of a study (Creswell, 2014).
Furthermore, “triangulation-whether you make use of more than one data collection method…is
a powerful strategy for increasing the credibility or internal validity of your research” (Merriam
& Tisdell, 2016, p. 245). An initial concern of this study were threats to internal validity since,
UNDERREPRESENTATION OF LATINAS IN STEM 43

according to Creswell (2014), it may interfere with a “researcher’s ability to draw correct
inferences from the data about the population…”. Since the study was conducted over a few
months, history should not be an internal validity threat. However, because these students were
surveyed and interviewed during the Fall 2018 semester, there was a chance that the success or
failure experienced by the students may impact how they answer some of the questions on the
survey. By taking a larger sample and perhaps by choosing interview participants who either: (a)
did not start out as STEM majors but became STEM after their first year of college and/or; (b)
Never declared as STEM majors even though it was a thought prior to college, history as a threat
to validity may be a non-issue. By sending the surveys out to the students taking introductory
math and science classes and by conducting this study in a short time frame, the threat of
maturation was also avoided. According to Creswell (2014), “participants…may mature or
change during the experiment, thus influencing the result”.
A significant limitation of this study was time. If this study were a five or ten-year study,
the data collected would be much richer and probably be of better value in the long term. Also,
this study, for the sake of convenience, focused on one campus. If this study included every
community college in a 100-mile radius or even every community college in California, the
results would have given greater insight as to the problem of underrepresentation of Hispanics
and women in the STEM fields. Limitations aside, this study certainly can add to the general
understanding of the factors contributing to this underrepresentation and may even offer another
lens to view this problem through.

UNDERREPRESENTATION OF LATINAS IN STEM 44

CHAPTER FOUR: RESULTS

This researcher's initial interest in conducting this study began with a simple idea. I
thought, "if I ever had a daughter, what will help shape whom she becomes." Of course, the
answer to this question seemed easy- we will. Her mom and I would help guide her along life,
and that will shape how she approaches her life decisions, including academics. The real answer
to this question, after reading through the literature and by conducting this study, did not reveal
to be as easy as I initially thought. As a biracial man, whose dad is Salvadoran and whose mom
is Filipina, I never found a "home" identity. Sometimes I felt Latino, while other times I felt
more Filipino. Either racial identity has its struggles, and it was not until after I started to pay
attention to the literature that I saw how those struggles played a factor in my academic career.
In chapter one of this study, I discussed the background of this study and presented the
research questions that I sought out to answer. In chapter two, I presented the literature that was
pertinent to this study, and I presented a conceptual framework that is rooted in experiential
learning, multiple dimensions of identity, and LatCrit. In chapter three, I outlined a
methodological approach that consisted of both quantitative and qualitative research methods
that, when analyzed together, could paint a much more complete picture of the answer to the
stated research questions. In a sense, utilizing both types of research methodologies gave me a
better understanding of some of the hidden nuances that I believe are key to uncovering some of
the more in-depth answers to the questions posed in this study. In this chapter, I will now
present the data that was collected in this study and the some of the critical findings that it
supports.

UNDERREPRESENTATION OF LATINAS IN STEM 45

Demographics
In order to ensure random sampling, the survey was distributed to a general population of
students as described in chapter three. As such, the demographics presented in the survey results
include, in some cases, a variety of responses (Table 1). While this ensured that a genuinely
random sample was taken, the data that were sought after involved Latinas who majored in
STEM. Therefore, the majority of the data presented was edited to include only Latinas
majoring in STEM.
Table 1

Frequency Table for Demographic Data

Q1- What gender
do you identify
yourself as?
Frequency Percent Valid Percent Cumulative
Percent
Female 82 81.2 81.2 81.2
Male 19 18.8 18.8 100
Missing 0 0
Total 101 100

Q2- What Racial
group do you
identify with
most?
Frequency Percent Valid Percent Cumulative
Percent
American Indian
or Alaskan Native
1

1 1 1
Asian, Native
Hawaiian, or other
Pacific Islander
8 7.9 7.9 8.9
Black or African
American
14 13.9 13.9 22.8
Hispanic or Latin(-
o, -a, -x) or
Spanish origin of
any race
55 54.5 54.5 77.2
Two or more races 9 8.9 8.9 86.1
UNDERREPRESENTATION OF LATINAS IN STEM 46

White 14 13.9 13.9 100
Missing 0 0
Total 101 100

Q3- What is your
current declared
major?
Frequency Percent Valid Percent Cumulative
Percent
Biological
Sciences
31 30.7 30.7 30.7
Engineering 4 4 4 34.7
Other non-science
related major
57 56.4 56.4 91.1
Physical Sciences 9 8.9 8.9 100
Missing 0 0
Total 101 100

The Impact of Hispanic and Female Faculty Members
Research question #1: What impact will the presence of Hispanic and female STEM faculty
members have on Latinas majoring in STEM fields?
The Importance of Modeling
One aspect of learning that is well-known is that of modeling. Bandura (1977) stated that
“acquisition of response information is a major aspect of learning" and that "much human
behavior is developed through modeling." With this in mind, it may be easy to assume that
students would prefer to have instructors at the college level who represent their gender, race,
and/or ethnicity. Interestingly, there have been authors who have presented data that may
suggest the contrary. Griffith (2010) and Kokkelenberg and Sinha (2010) presented data that
suggests that gender modeling is only seen in some instances. What is even more interesting is
that in both instances, women majoring in the biological sciences seem to be negatively affected
by this behavior. In other words, when the instructor of the classes that teach biology are female,
UNDERREPRESENTATION OF LATINAS IN STEM 47

there are fewer female students. Price (2010) goes on further with this idea and in his research
finds that female students are less likely to persist in STEM when they have more female faculty
teaching STEM courses. The interview data obtained seem to fortify the research as mentioned
earlier. When asked if it was important that the people teaching the courses represented the
student's ethnicity and/or gender, most of the respondents answered in the negative. One
respondent said, "Honestly, it does not…I do not judge by race or ethnicity. I came…to learn. I
came to just get what I want, you know?”, while another said, “No. I do not think so. I think it
just depends on their knowledge like that is the important part. Not the background.”
(Respondent #1, personal communication, January 22, 2019; Respondent #3, personal
communication, January 17, 2019).
Table 2a includes descriptive statistics that represent responses from Latinas majoring in
STEM showing that all of them do not agree to some level that either gender or race plays a part
in their learning. Table 2b is a correlation matrix showing the linear relationship of the
responses to the questions regarding modeling and the race and gender of the instructors.
Table 2a

Descriptive Statistics: Latinas in STEM Modeling

Descriptive
Statistics
I think it is
important if the
instructor is the
same gender that I
identify with
I think that I
would learn better
if the instructor is
the same gender
that I identify with
I think it is
important if the
instructor is the
same race that I
identify with
I think that I
would learn better
if the instructor is
the same race that
I identify with
Valid 18 18 18 18
Missing 0 0 0 0
Mean 1.5 1.556 1.556 1.611
Std. Deviation 0.5145 0.5113 0.7838 0.8498
Minimum 1 1 1 1
Maximum 2 2 4 4

Note: Strongly disagree=1, Disagree=2, Agree=3, Strongly agree=4
UNDERREPRESENTATION OF LATINAS IN STEM 48

Table 2b

Correlation Matrix: Latinas Majoring in STEM vs. Gender/Race Modeling

Questions

Pearson’s r/ p-
value

I think it is
important if the
instructor is the
same gender
that I identify
with

I think that I
would learn
better if the
instructor is the
same gender
that I identify
with

I think it is
important if the
instructor is the
same race that
I identify with

I think that I
would learn
better if the
instructor is the
same race that
I identify with

I think it is
important if the
instructor is the
same gender
that I identify
with

Pearson's r

—

p-value

—

I think that I
would learn
better if the
instructor is the
same gender
that I identify
with

Pearson's r

0.894***

—

p-value

< .001

—

I think it is
important if the
instructor is the
same race that
I identify with

Pearson's r

0.729***

0.652**

—

p-value

< .001

0.003

—

I think that I
would learn
better if the
instructor is the
same race that
I identify with

Pearson's r

0.74***

0.662**

0.962***

—

p-value

< .001

0.003

< .001

—

* p < .05, ** p < .01, *** p < .001

UNDERREPRESENTATION OF LATINAS IN STEM 49

Upon further examination of Table 2a, particularly examining the standard deviation
data, the argument can be made that there seems to be some level of variance between race and
gender. These data show that while all the respondents, on average, disagree on some level on
both the gender and the race question, there was one respondent who felt differently about the
race questions; hence the reason for the standard deviation being higher. Considering that for
these sets of questions, the trend was that all the respondents responded negatively, it calls to
question why the one respondent answered the way she did. While it may be possible that it was
an erroneous answer on the respondent’s part, it is also equally possible that the respondent felt
differently about race (compared to gender). Table 2b shows that there is a positive correlation
in the answer to the modeling questions and further cement the idea that the Latinas surveyed do
not feel that having an instructor who "looks like them" is important or plays a part in how well
they learn. The Pearson’s r value further indicates that while these data approach linearity, since
the responses all are leaning in the negative, the sole respondent’s answers to the race questions
are enough to make a small difference.
To complicate matters, other interview and survey data collected almost simultaneously
present a different narrative regarding modeling. When the interview respondents were pressed
further after responding in the negative as to whether female or Latina instructors would make a
difference in their learning, the answers seemed to shift. One respondent said, “I feel like
women have this kind of connection and you do not... well, not in like any other case but I feel
like you get more of a ...closer relationship with…, you know, like with a female teacher.”
(Respondent #1, personal communication, January 22, 2019). Another respondent answered,
“Probably… I like seeing another person who is- with my background that can do it. That means
UNDERREPRESENTATION OF LATINAS IN STEM 50

I can do it too…” (Respondent #3, personal communication, January 17, 2019). Other
respondents were a lot more detailed with their responses:
“I feel, again, knowing the history that women have been through, you know without
having zero rights to vote and anything, yeah, like you-it makes me just, okay, go girl.
You know, girl power.” (Respondent #2, personal communication, February 8, 2019).
“Just because I know that if, they were…like they are able to do that, I can too. And it's
kind of like, there's not that many like female presence, you know, in like science fields
as much as there are men, so that kind of just like makes me want to defeat the odds, in a
way. So, like, when I see a woman doing it, it kind of pushes me as well.” (Respondent
#5, personal communication, January 23, 2019).
There is research conducted that describes how female students are more likely to pursue
a major in STEM due to “active encouragement from someone important to them” but that these
same students are more likely to leave STEM majors because of negative interactions between
themselves and their male peers (Morgan, Gelbgiser, & Weeden, 2013). The responses above
echo some of that research in that if the women interviewed in this study bestow a level of
importance to their female instructors, perhaps that is a driving force in their pursuit of a STEM
degree. One of the respondents, however, took what seemed to be a more neutral approach, by
saying “I feel like it's a lot easier to relate to a person with the same ethnicity, but it
doesn't…have to be Latina or Mexican or some type of the same, uh, group.” (Respondent #4,
personal communication, February 8, 2019). This same respondent took the same approach in
regards to her feelings towards gender by stating, “I feel like it is easier to relate to a woman, but
it doesn't necessarily mean that I can't also relate to a man.” (Respondent #4, personal
communication, February 8, 2019).
UNDERREPRESENTATION OF LATINAS IN STEM 51

Factors That May Be Influenced by Modeling
STEM careers are, in many cases, more lucrative than careers in other fields of study
(Carnevale et al., 2011). Taking into consideration that STEM careers are more lucrative, it may
be no surprise that most of the respondents either agreed or strongly agreed about the importance
of the extrinsic motivators and their relationship to being a STEM major, with over 88%
agreeing to some level, that money is an important extrinsic factor. These data are presented by
Tables 3a, 3b, and 3c. Table 3a presents the descriptive statistics for the extrinsic motivating
factors which include questions dealing with issues regarding success, monetary gain, and
employment. Table 3b is a frequency table for the extrinsic factors mentioned above and is an
analysis of the answers to each question asked. Table 3c is a correlation matrix comparing each
question against the other in order to show how linear the answers turn out to be. While it
probably is very reasonable that the extrinsic factors that are presented in these data are
important to many people, what is a bit confounding is the data that are presented in Tables 4a,
4b, and 4c.
Table 3a

Descriptive Statistics for Extrinsic Motivating Factors

Descriptive Statistics

Being successful in my
career
Having lots of money
Being able to find
steady work
Valid 18 18 18
Missing 0 0 0
Mean 2.889 3.111 2.722
Std. Deviation 0.4714 0.7584 0.6691
Minimum 1 1 1
Maximum 3 4 3

Note: Strongly disagree=1, Disagree=2, Agree=3, Strongly agree=4

UNDERREPRESENTATION OF LATINAS IN STEM 52

Table 3b

Frequency Tables for Extrinsic Motivating Factors

Being successful
in my career
Frequency Percent Valid Percent
Cumulative
Percent
Strongly Disagree 1 5.6 5.6 5.6
Agree 17 94.4 94.4 100
Missing 0 0
Total 18 100

Having lots of
money
Frequency Percent Valid Percent
Cumulative
Percent
Strongly Disagree 1 5.6 5.6 5.6
Disagree 1 5.6 5.6 11.1
Agree 11 61.1 61.1 72.2
Strongly agree 5 27.8 27.8 100
Missing 0 0
Total 18 100

Being able to find
steady work
Frequency Percent Valid Percent
Cumulative
Percent
Strongly Disagree 2 11.1 11.1 11.1
Disagree 1 5.6 5.6 16.7
Agree 15 83.3 83.3 100
Missing 0 0
Total 18 100

A quick analysis of Table 3a seems to confirm what appeared to be obvious- that the
respondents would all agree that the extrinsic motivating factors have some level of importance.
However, an analysis of Table 3b reveals a little more information, especially pertaining to the
factor of making money. It would be unreasonable, to some extent, to believe that money would
not have at least some level of importance, especially in the context of a career. The factor that
was put on the survey was specific, however, stating “having lots of money.” While “lots” is an
ambiguous term, most of the respondents did respond in the positive, but two interestingly did
UNDERREPRESENTATION OF LATINAS IN STEM 53

not. While these two respondents may just be people who do not find value in having an
abundance of money, it may also be possible that these two people responded that way because
of other reasons. One reason may be that although this was an anonymous survey, they did not
want to label themselves as people who needed to have “lots” of money. The same pattern of
responses occurred with the factor of the ability to find steady work. The same explanation that
was provided for the “money” factor may be applicable here since having a job means that a
paycheck is earned.
Table 3c

Correlation Matrix: Latinas Majoring in STEM vs. Extrinsic Motivation

Extrinsic
Motivators
Pearson’s r/ p-
value
Being successful
in my career
Having lots of
money
Being able to find
steady work
Being successful
in my career
Pearson's r

—

p-value

—

Having lots of
money
Pearson's r

0.037

—

p-value

0.885

—

Being able to find
steady work
Pearson's r

0.642**

0.18

—

p-value

0.004

0.474

—

* p < .05, ** p < .01, *** p < .001

Table 3c, then fortifies what Table 3b revealed- that the extrinsic motivators are not as
clear cut as once hypothesized earlier in this chapter, with the Pearson’s r values straying from
linearity. The two factors that did show some linearity were success in their career and the
ability to find steady work. It is interesting that there is not more of a correlation between these
responses considering that at least one measure of success in a career is monetary gain.
UNDERREPRESENTATION OF LATINAS IN STEM 54

It was previously explained that the data presented in table 2a showed that the Latinas
surveyed did not agree that modeling played a factor in their learning and the interview
responses fortified the survey responses. However, other tables show that when these data are
taken into consideration with other data, another narrative emerges. In Tables 4a, 4b, and 4c, the
factors that are taken into consideration are all intrinsic. Table 4a places the modeling questions
initially presented in Table 2a next to the intrinsic motivating factors.
Table 4a

Descriptive Statistics for Modeling and Intrinsic Motivating Factors

Descriptive
Statistics
I think it is
important if
the
instructor is
the same
gender that I
identify
with
I think that I
would learn
better if the
instructor is
the same
gender that I
identify
with
I think it is
important if
the
instructor is
the same
race that I
identify
with
I think that I
would learn
better if the
instructor is
the same
race that I
identify
with
Being an
example to
people who
associate
with my
gender
Being an
example to
people who
associate with
my
race/ethnicity
Valid 18 18 18 18 18 18
Missing 0 0 0 0 0 0
Mean 1.5 1.556 1.556 1.611 2.889 3
Std.
Deviation 0.5145 0.5113 0.7838 0.8498 0.3234 0.5941
Minimum 1 1 1 1 2 1
Maximum 2 2 4 4 3 4

Note: Strongly disagree=1, Disagree=2, Agree=3, Strongly agree=4

Again, just like in Table 2a, Table 4a seems to tell a predictable story. The data from
Table 2a are shown alongside the intrinsic data. What is fascinating when examining these data
is the intrinsic factor regarding gender. For this particular question, none of the respondents
chose answers in the extremes (strongly disagree or strongly agree). While, most of them
answered “agree”, it would be interesting to know what would have happened if the survey were
UNDERREPRESENTATION OF LATINAS IN STEM 55

not a forced response. The frequencies presented in Table 4b give better clarity to how the
respondents answered these questions.
Table 4b

Frequency Table for Intrinsic Motivating Factors
Being an example
to people who
associate with my
gender
Frequency Percent Valid Percent
Cumulative
Percent
Disagree 2 11.1 11.1 11.1
Agree 16 88.9 88.9 100
Missing 0 0
Total 18 100

Being an example
to people who
associate with my
race/ethnicity
Frequency Percent Valid Percent
Cumulative
Percent
Strongly Disagree 1 5.6 5.6 5.6
Agree 15 83.3 83.3 88.9
Strongly Agree 2 11.1 11.1 100
Missing 0 0
Total 18 100

Again, here in Table 4b, the actual results of the survey are shown. While most of the
respondents for both questions answered in the affirmative, the question arises of how sure these
respondents are about how they feel being an example to people of their own gender. While it is
probably true that most of these women do agree that they feel it would be important to be an
example to other women, if a “neutral” option were given, it would be fascinating to see how
many of them would have chosen that option. If these women were truly interested in becoming
models for their gender, the thought would be that they would have chosen “strongly agree”.
Table 4c shows the correlation of the modeling questions and the intrinsic factors.
UNDERREPRESENTATION OF LATINAS IN STEM 56

Table 4c

Correlation Matrix: Modeling vs. Intrinsic Motivation

Statements on
modeling and
intrinsic
factors

Pearson’s
r/ p-value

I think it
is
important
if the
instructor
is the
same
gender
that I
identify
with
I think
that I
would
learn
better if
the
instructor
is the
same
gender
that I
identify
with
I think it
is
important
if the
instructor
is the
same race
that I
identify
with
I think
that I
would
learn
better if
the
instructor
is the
same
race that
I identify
with
Being an
example
to
people
who
associate
with my
gender
Being an
example to
people who
associate
with my
race/ethnicity
I think it is
important if
the instructor
is the same
gender that I
identify with
Pearson's
r
—

p-value
—

I think that I
would learn
better if the
instructor is
the same
gender that I
identify with
Pearson's
r
0.894*** —

p-value
< .001 —

I think it is
important if
the instructor
is the same
race that I
identify with
Pearson's
r
0.729*** 0.652** —

p-value
< .001 0.003 —

I think that I
would learn
better if the
instructor is
the same race
that I identify
with
Pearson's
r
0.74*** 0.662** 0.962*** —

UNDERREPRESENTATION OF LATINAS IN STEM 57

p-value
< .001 0.003 < .001 —

Being an
example to
people who
associate with
my gender
Pearson's
r
0.354 0.395 0.258 0.262 —

p-value
0.15 0.104 0.302 0.294 —

Being an
example to
people who
associate with
my
race/ethnicity
Pearson's
r
-0.385 -0.387 -0.253 -0.233 -0.612** —
p-value
0.115 0.112 0.312 0.352 0.007 —
* p < .05, ** p < .01, *** p < .001

Table 4c shows the Pearson correlations of the modeling data against the intrinsic
motivating factors. Not surprisingly, there is lack of linearity between the two sets of factors.
Furthermore, it seems that although these Latina students do not feel that having women or
Latinx instructors make a difference in their learning, over 83% of these same students feel that
being an example to people who associate with their gender and/or race is important. This brings
up what seems to be an interesting paradox: the same students whom one day want to be models
(as successful Latinas) for behavior, do not themselves think that models are important for
themselves.
The Effect of Perceived Difficulty in Math and Science
Research question #2: What impact will the perceived difficulty of math and science classes
taken during the first two years of college have on Latinas majoring in STEM fields?
Attitudes Regarding Math and Science Classes
There is literature that suggests that early success in math classes is an indicator of future
math and science success as well as STEM major persistence (Toven-Lindsey, Levis-Fitzgerald,
Barber, & Hasson, 2015; Xie, Fang, & Shauman, 2015). Other research suggests that, at least at
UNDERREPRESENTATION OF LATINAS IN STEM 58

some point, results of math scores through standardized testing have been used as indicators of
future STEM participation and persistence (Mann & Diprete, 2013; Morgan et al., 2013). In this
study, the responses given by the interview respondents were mixed. When asked about their
thoughts regarding lowering or changing the math requirement for their major, one respondent
said, “Well, that's not my favorite topic…that's not my strength, so…” (Respondent #2, personal
communication, February 8, 2019). Another respondent simply said, “…if there's an option to
not do it, I'll not do it," while another added more to her answer by stating, "I'd be really
happy…that there's no math because personally I don't like math, but math is essential to
everyday life.” (Respondent #3, personal communication, January 17, 2019; Respondent #4,
personal communication, February 8, 2019). Respondents two, three, and four were very
particular in how they answered this question. Their answers centered around how they did not
particularly enjoy math. In regards to the science classes, the answers were a bit different. One
respondent stated, “Yeah, it's just putting in the work, you know…although, we should know
history, um, science, there's something about it that's interesting. So, no matter if they tell you
here read this, I feel I learn a little each time, you know, like more and more and more. So, that's
why I'm in this field, like; science kinda calls me this way.” (Respondent #2, personal
communication, February 8, 2019). Another student said, “it just makes me…do it and…get it
like over with…” (Respondent #5, personal communication, January 23, 2019). On the survey,
two questions asked questions about the students' feelings regarding the enjoyability of math and
science classes and how well they did in those classes. The survey data regarding these
questions are presented in Tables 5a and 5b, with Table 5a presenting the descriptive statistics
for how successful the students have been in math and science classes and how they rated their
overall enjoyability for the same classes and Table 5b showing frequencies.
UNDERREPRESENTATION OF LATINAS IN STEM 59

Table 5a

Descriptive Statistics Regarding Math and Science Success and Enjoyability

Descriptive Statistics
I have always found science
and/or math enjoyable
I have always done well in my
previous science and/or math
classes (grade of B or better)
Valid 18 18
Missing 0 0
Mean 3.167 3.056
Std. Deviation 0.6183 0.7254
Minimum 2 1
Maximum 4 4

Note: Strongly disagree=1, Disagree=2, Agree=3, Strongly agree=4

Table 5b

Frequency Tables Regarding Math and Science Success and Enjoyability

I have always
found science
and/or math
enjoyable
Frequency Percent Valid Percent
Cumulative
Percent
Disagree 2 11.1 11.1 11.1
Agree 11 61.1 61.1 72.2
Strongly Agree 5 27.8 27.8 100
Missing 0 0
Total 18 100

I have always done
well in my
previous science
and/or math
classes (grade of B
or better)
Frequency Percent Valid Percent
Cumulative
Percent
Strongly Disagree 1 5.6 5.6 5.6
Disagree 1 5.6 5.6 11.1
Agree 12 66.7 66.7 77.8
Strongly Agree 4 22.2 22.2 100
Missing 0 0
Total 18 100
UNDERREPRESENTATION OF LATINAS IN STEM 60

The survey data show that most of the Latina students who responded had positive
feelings regarding math and science, with over 88% of the respondents either agreeing or
Strongly agreeing. Only two of the survey respondents replied in the negative, which is
interesting since at least three of the interview respondents noted that math was not enjoyable.
The data also show that about the same percentage of Latina students have done well (at least a B
or better) in their science and/or math classes.
Students’ Thoughts Regarding the Perceived Difficulty in Math and Science Classes
When questioned on whether or not the perceived difficulty of math and science classes
made a difference in the students' motivation to continue as STEM majors, all the interview
respondents answered similarly. Most agreed that it would not change their current path, and for
the most part either implied or outright said that math was difficult. One respondent replied, “I
mean, it does scare me, but I feel like if I really want to do it…I will…go through it and take it.”
(Respondent #5, personal communication, January 23, 2019). Another respondent said,
“…makes me not wanna take it, but I still have to take it to reach a goal.” (Respondent #4,
personal communication, February 8, 2019). Another one of the interview respondents gave a
more detailed answer giving a bit more insight as to what may be occurring with some Latinas
majoring in STEM. She answered, “…I guess having three kids, not having that extra time to
study, and like quizzes and all of that, I don't prepare, I feel I don't prepare myself well
enough…do I really, am I going to make it?” (Respondent #2, personal communication,
February 8, 2019). The science classes, however, did not seem to bother most of the
respondents. Some of the responses included:
“Never did I ever think…negatively about the science. Ever since… I just wanted to just
learn more. I've never felt more ... like, I never doubted myself. I've always stepped into
UNDERREPRESENTATION OF LATINAS IN STEM 61

the classroom and felt like I needed to learn more. I wouldn't study a lot. I just felt good
about those classes.” (Respondent #1, personal communication, January 22, 2019)
“It never scared me, no. I never thought twice about my career choice.” (Respondent #3,
personal communication, January 17, 2019)
“…it doesn't like motivate me any less…kind of just makes me want to push myself
harder because I want to… tell my kids that if I could do it, that they definitely can too.”
(Respondent #5, personal communication, January 23, 2019)
Table 6a and 6b show the descriptive statistics and frequencies, respectively, for the perceived
difficulty in STEM. In Table 6a, the statistics show that the respondents on average agree that
they found difficulty in math and science classes. What might be of some significance with
these data, however, is the standard deviation. It seems as though there is enough deviation that
the answer could have easily gone in another direction. The various responses to these questions
do make me wonder, however, what impact would providing a neutral response place on these
statistics.
Table 6a

Descriptive Statistics for Perceived Difficulty in STEM

Descriptive Statistics
I have always found science
(biology, chemistry, physics,
etc.) classes to be difficult
I have always found math classes
to be difficult
Valid 18 18
Missing 0 0
Mean 3 2.778
Std. Deviation 0.9701 0.8085
Minimum 1 2
Maximum 4 4

Note: Strongly disagree=1, Disagree=2, Agree=3, Strongly agree=4

UNDERREPRESENTATION OF LATINAS IN STEM 62

Table 6b

Frequency Tables for Perceived Difficulty in STEM

I have always
found science
(biology,
chemistry, physics,
etc.) classes to be
difficult
Frequency Percent Valid Percent
Cumulative
Percent
Strongly Disagree 1 5.6 5.6 5.6
Disagree 5 27.8 27.8 33.3
Agree 5 27.8 27.8 61.1
Strongly Agree 7 38.9 38.9 100
Missing 0 0
Total 18 100

I have always
found math classes
to be difficult
Frequency Percent Valid Percent
Cumulative
Percent
Disagree 8 44.4 44.4 44.4
Agree 6 33.3 33.3 77.8
Strongly Agree 4 22.2 22.2 100
Missing 0 0
Total 18 100

The survey data is representative, perhaps, of those not chosen to be interviewed.
According to Table 6b, almost 67% of the respondents agreed, to some level, that they found
science classes to be difficult. While a little over half of that 67% was in the “strongly agree”
category, it may be important to point out that two-thirds of the surveyed students contradicted
those who were interviewed. A bit of a surprising result is the survey data representing the
number of Latinas majoring in STEM who perceive math as difficult. While all of the interview
respondents perceived math as difficult and with at least two of the respondents responding that
UNDERREPRESENTATION OF LATINAS IN STEM 63

they would see it as a positive if the math requirement were either lessened or eliminated, only
55% of the Latinas surveyed felt that math was difficult.
Summary of Results
Two research questions guided this study. The first question asked if there was any
impact from Hispanic or female (or both) faculty members on Latinas majoring in STEM. The
interview responses were interesting in that in every case; the students interviewed did not feel
that having female faculty members or Latinas teaching the courses that they were taking had
any impact on their learning, including whether or not they felt more motivated to persist in
STEM majors. While each one of the students interviewed was adamant about their feelings
towards this, when pressed further, some of the students seemed to change course. Some of
them brought up feelings of female empowerment and racial modeling. While there was no
follow up as to why the change in thought, I offer up an explanation. Looking at the survey data
regarding intrinsic motivation (tables 4a, 4b, and 4c), I posit that the reason why these students
answered questions regarding race and gender in the negative (they do not feel like it matters)
relates to an idea of wanting to be colorblind. In all of these interviews, when the gender or race
questions were brought up, the students were, in a sense, forceful in their answers. There was a
furrowed brow associated with every student who was asked this question. They almost seemed
bothered that these questions on race and gender even came up. It is my observation that these
students felt that if they answered in the affirmative, that they would be seen as "playing the race
or gender card." Only when I pressed on and showed that it was okay for them to answer in the
affirmative that some of them began to open up a bit more. As mentioned earlier in this chapter,
it is interesting that an overwhelming majority responded on the survey that it would be vital for
them to be an example to those who identified with their race or gender, but felt that having these
UNDERREPRESENTATION OF LATINAS IN STEM 64

same models were not vital to them. I suspect that either these students embraced this sort of
colorblind philosophy when answering these questions, they lied, or they did not understand the
questions asked. It would have made better sense that the students would agree that gender and
race models would be essential and that would eventually motivate them to do the same for
future generations.
The second question involved the perceived difficulty in math and science classes and the
impact of that perception on Latinas majoring in STEM. It is interesting that the students who
were interviewed, in my estimation, initially downplayed how they felt about math. In one
particular case, a student seemed to indicate that math did not pose a problem, but when asked
about her thoughts regarding lowering or eliminating the math requirement, the student smiled
and answered that she would be happy if that happened. Again, though, that particular student
seemed to downplay the answer by stating that, “that's good news, I mean, for people also who
have that major who don't have like financial aid… I feel like it's good because I could take
another class that's required for my major.” (Respondent #1, personal communication, January
22, 2019). The other students interviewed had no problem telling me that math was not
enjoyable or that they were not particularly good at math. The survey data was a little more
perplexing because I did expect to see a higher number of respondents say that they found math
to be difficult; especially during a survey, where there was no way of being identified unless the
person decided to participate in the interview process. That leads me to believe that maybe the
students did not find math to be hard because a large number of those surveyed have not taken a
higher-level math class, where the math does pose a certain level of difficulty. Perhaps a
majority of those surveyed are still relatively new to the college or did not take higher-level
(algebra and above) math classes in high school and may change their minds in regards to how
UNDERREPRESENTATION OF LATINAS IN STEM 65

they feel about math in a year or two. With that said, the same may also apply to the sciences.
The majority of those interviewed were positive when talking about the sciences, but maybe it is
because they still have not taken anything past the entry-level (100-level) classes. It would be
interesting to see if these answers would change in a year or two.

UNDERREPRESENTATION OF LATINAS IN STEM 66

CHAPTER FIVE: DISCUSSION
It was pointed out that several researchers have presented data that show that there is still
an underrepresentation of women and minoritized people in the STEM fields (Mann & Diprete,
2013; National Science Foundation, 2017). This underrepresentation is found at all levels,
starting at the college level and persisting in STEM careers (Carnevale et al., 2011; National
Science Foundation, 2017). Considering that the underrepresentation of women and minorities
has spilled over into STEM careers, it is crucial that we address this issue as early as possible in
education. While there have been studies that offer up explanations for the underrepresentation
of women and minoritized people, there does not seem to be a consensus as to what is causing it.
Some of the reasons that have been presented include the presence or absence of gender and
racial models, success rates in early math and science classes, and the presence of environmental
“pull” factors which include peer influence and family expectations (Crisp et al., 2009; Griffith,
2010; Kokkelenberg & Sinha, 2010; Xie & Shauman, 2003).
The two research topics at the center of this study that were used to gather more insight
as to some of the themes that may be exerting influence on STEM participation and persistence
by Latinas were:
1. What impact will the presence of Hispanic and female STEM faculty members have on
Latinas majoring in STEM fields?
2. What impact will the perceived difficulty of math and science classes taken during the
first two years of college have on Latinas majoring in STEM fields?
Using both quantitative and qualitative research methods gave me the opportunity to see that
previous research on this topic continues to be true, but more importantly, showed that perhaps
UNDERREPRESENTATION OF LATINAS IN STEM 67

what is at the heart of this underrepresentation of women and minoritized people is how they
identify themselves as STEM students.
This study was performed at an HSI with approximately 53% of the student population
identifying itself as Hispanic and with 52% of the population identifying themselves as female.
With those two demographic data in mind, the conceptual framework presented in chapter two
using Kolb’s (1984) experiential learning, Jones and McEwen’s (2000) model of Multiple
Dimensions of Identity, and LatCrit presented by Yosso, gives an excellent foundation upon
which the conclusions of this study is built upon.
This final chapter provides not only a capstone to my present work but also a beginning
to future research that I wish to commence. In this chapter I will discuss the findings with
emphasis on limitations not initially accounted for, present how this study informs practice in
post-secondary education and present future research on this topic.
Discussion of Findings
The findings in this study appear to either credit or discredit some of the research that has
already been presented. Therein lies, what I believe is a problem in the current research. In most
cases, the research surrounding the underrepresentation of women and minoritized people is
mainly rooted in quantitative research. This notion is especially true regarding Latinas, the
population that is central to this study. Probably the critical finding of this study is that in order
to get to the heart of what is causing an underrepresentation of Latinas in STEM, researchers
need to practice mixed methodology in their research. In this study, for example, if the survey
data were taken alone, some of the results would have indicated that having Latina instructors
would make no difference in learning. When coupled with the data obtained from the
interviewed students, however, a different result is found. Instead, it seems that some of these
UNDERREPRESENTATION OF LATINAS IN STEM 68

students have a difficult time stating who they are. Are they STEM students? Are they Latinx
students? Are they female students? Are they all of those? If they consider themselves as being
all of those identities, which one prevails? Which one of those identities shape how they
approach the work that they have in front of them? Which one of those identities shape how they
view themselves as college students going forward? Understanding the intersectionality of all
those identities is vital, I believe, to understanding why there is an underrepresentation of Latinas
in STEM as well as understanding what needs to be done to encourage Latinas to major in and
persist in STEM.
In some cases, these students will learn critical race theory if they take a class that talks
about it. What happens to those that do not? What if the reason why there is an
underrepresentation of Latinas in STEM is that Latinas do not have the right tools that allow
them to navigate through the first two years of college? What if Latina STEM students do not
recognize that modeling may be necessary for success?
In chapter three of this study, I presented the limitations that I thought would be salient in
this research. It was noted that sample size and time would be the significant limitations of this
study. Although roughly 100 survey respondents were collected, only 18 were part of the target
population (Latinas in STEM). At an institution that has over 50% Latinx and over 50% female,
this sample size is inadequate to answer some of the more profound questions that emerge from
this study. It can, however, provide a glimpse as to what is seen and as stated in this chapter, can
raise other questions that are important to address the issue of underrepresentation of Latinas in
STEM. This study was conducted for just a few months and if given more time, would have
allowed for perhaps a few more interviews, which could have either fortified what was seen in
this study or could have presented alternative data which could have led to more in-depth
UNDERREPRESENTATION OF LATINAS IN STEM 69

analysis. A limitation that was unforeseen at the beginning of this study was the measure that
was used to collect the data.
Although the survey and interview protocols were sufficient to begin this study, after
some careful consideration, it was determined that more detail should have been used in both
survey and interview questions. One such question involved the enjoyability of math and
science classes. Instead of asking how enjoyable both were, the questions should have been
separated on the survey in order to see if there was a clear delineation between how students
viewed the two subjects. The same is true of the question regarding how well the students
performed in these classes. It was not until an analysis of the data was performed that this
realization took place. Another major limitation on the survey that was not realized until after
the data had already been collected involved self-reporting. This is especially true of the
question regarding students agreeing or disagreeing that they received a grade of “B” or better in
their math and science classes. The problem here is that the students could have selective
memory as to which classes they decided to say they got the good grade in or the students could
even have lied about their grade. Although there would have been no reason to lie on an
anonymous survey, that would not disqualify students from doing so. In regards to the interview,
the questions were designed to be open-ended in order to extract as much information from the
interview respondents as possible. Unfortunately, in some cases, the respondents were very short
with their answers and did not have much to offer. In one case, the respondent seemed confused
initially and did not seem to understand what was being asked. I was initially concerned with
writing an extended interview protocol, and in hindsight, I may have overcorrected and written
questions that did not give me some of the answers I was looking for. Lastly, I may have
introduced bias in the study. Unfortunately, there was no way to have anyone else interview the
UNDERREPRESENTATION OF LATINAS IN STEM 70

students for this study. By conducting the interviews, I may have imposed biases that were
initially unaccounted for. As a biracial man who identifies as both Asian and Latino, it is
perceivable that the respondents to the interview may have seen my last name and assumed that
as a Latino I wanted to hear specific answers. Also, as a male interviewer, maybe I did not
provide a safe enough space for these women to feel comfortable enough to answer the questions
asked.
Implications for Practice
As a community college educator, I have made specific observations about student-
faculty interaction. For the most part, tenured or tenure-track faculty are more than willing to
help students in their classes succeed at the subject they are teaching. Adjunct faculty are also
willing to help but find it challenging because of a lack of resources, including time. The
challenge is that many faculty members will hope that the counseling department will handle
matters that include career goals, the decision of what school to transfer to, and major choice.
The problem is, in some cases, that does not happen. Many students go into the first two years of
college ignorant as to what is necessary to succeed. They have been given the same old lines:
"do what makes you happy," "just follow your EdPlan," or "you will figure it out as you go
along." The problem is that some of these students do not have any real guidance. Maybe they
come from a household where they are first-generation college students, or maybe their parents
went a different route. Either way, these students have never been provided that appropriate
models that could help guide them along a path that they could at least try to follow. One of the
implications for practice that can be informed by this study is to teach students in their first
semester of college some of the basic theories of learning. Maybe if students understand how
learning works, they can apply better techniques to their actual learning.
UNDERREPRESENTATION OF LATINAS IN STEM 71

Another implication includes finding a way for faculty and counselors to get together to
guide the students from the first day of class better. It almost seems like a herculean task when
you consider that, at least in the institution that I am at, there are 18,000 students per semester.
However, let us imagine for a moment that a Latina enters community college wanting to major
in engineering. She goes through the counseling department to get her list of classes and then is
sent to a STEM counselor who can have a five to ten-minute conversation with her regarding her
choice of major, goals, and motivations. This counselor then acts as a sort of intermediary
between the counseling department and the STEM instructors. If that same student came in with
fears of math success, maybe the STEM counselor can offer advice on how to quell some of
those fears.
My last suggestion might be the most difficult to implement. It may be wise for STEM
instructors who teach at any post-secondary institution to learn more about the student
populations that they are responsible for teaching. One way to do this is actually to look at the
current NSF data regarding STEM. It would be useful for STEM instructors to at least glance at
the data presented by the NSF regarding women and minoritized people in STEM. That way, an
instructor can identify which students are not being represented enough in STEM and perhaps
that instructor can use that data to help motivate those students more. Sometimes students need
to see that an instructor is interested in wanting them to succeed and to do more than they
initially thought they could. If we as practitioners have no interest in going beyond just teaching
in the classroom, I believe that we do the students that we teach, and education as a whole, a
great disservice. As practitioners, we carry the burden of being true educators, not just a person
who lays out facts on a subject. We need to be active participants in the overall learning process
UNDERREPRESENTATION OF LATINAS IN STEM 72

and take a vested interest in the future of our students, understanding that they will hopefully
understand that they may one day do the same for their students.
Future Research
I genuinely believe that there is a lot more that can be done to get to the root as to what is
causing the underrepresentation of Latinas in STEM. For one, more studies should be done
using a mixed methodology. As I stated earlier in this study, the combination of quantitative and
qualitative methods in researching this problem can help to uncover some of the more in-depth
answers. Not only that, but this study itself has shown that if only quantitative data are used,
especially if some of the data is self-reported, it may skew the answers to one side or another.
A big part of being a STEM student is understanding what STEM is. The problem is, in
most of the research that is conducted, what is not entirely defined is what STEM entails. Are
nursing majors considered to be STEM majors or as an allied health vocation? At least two of
the Latina students interviewed considered themselves to be STEM majors but are nursing. I
bring this up because it would skew the data if nursing major data are aggregated with those
wanting to major in medicine as well as those wanting to major in the biological sciences.
Researchers may look at data regarding the biological sciences and say that Latinas are on the
rise (Mann & Diprete, 2013; National Science Foundation, 2017). If these data contain a mixture
of health and sciences, it does not show that there is higher representation in the STEM fields.
So, there needs to be a clear delineation between what is considered a biological science,
medicine, and perhaps even the nursing fields. That way, if there is an underrepresentation in all
three, researchers can address each one of the fields separately.
Another essential item that may need some future attention is something that one of the
respondents brought up during an interview. One of the interview questions asked about any
UNDERREPRESENTATION OF LATINAS IN STEM 73

other obstacles that may cause the students to change their major away from STEM. One of the
students mentioned her undocumented status. I believe that this is a real fear to many students,
not just STEM students, especially when it comes to how these students will pay for school.
Since STEM majors usually go well beyond a bachelor's degree, there is a fear that there will not
be a way to pay for school without incurring a large amount of debt. It may be interesting to
study this and see how many Latinas in STEM are affected by their undocumented status and if it
causes a lack of persistence in STEM.
Lastly, in this study, the conceptual framework revolves around identity and the
experiences that shape who the student is. While this study focused on modeling and the
perceived difficulty of math and science classes, it would be imperative to research family
expectations, cultural expectations, and the roles of peers and how they impact Latinas majoring
in STEM. Several researchers have already begun to do this work, but much of it has been
quantitative research. I suggest that in order to find some answers, an ethnographical study
coupled with a survey is an excellent place to start.
Conclusions
According to the data presented in this study, it is clear to some extent, that math does
have a profound effect on how Latinas see their schooling going forward. Is it enough to affect
STEM persistence? Probably not, in some cases. After interviewing these women, they, at least
for the interview, presented themselves to be strong women who did not see any obstacle getting
in the way of their goals. Perhaps they just have not encountered an obstacle big enough to pose
a problem. For their sake, I genuinely hope that they do not get turned away from STEM. The
STEM fields need diversity, both in gender as well as race. Is there something to be said for the
importance of modeling? Again, according to this study, the answer to this question can be
UNDERREPRESENTATION OF LATINAS IN STEM 74

confusing. It seems as though, at least for the students interviewed, that they need first to
establish who they are as Latinas in STEM. Once they get to the point where they are
comfortable talking about race and gender and perhaps the obstacles that different races and
genders face, then maybe those students can see the modeling is essential. If the research is not
done to identify how underrepresented STEM is, we will never get to the root of what is causing
the underrepresentation and no solution will be found. It takes participation on all sides,
however. Researchers, practitioners, and the students who are taking these classes all need to be
a part of the solution. The students, however, need to find this valuable, and maybe that needs to
start at home. If the students who represent those that are underrepresented in STEM understand
why it is so important to have many different voices in STEM, maybe we can get closer to
balancing out who represents STEM. Lastly, if we want to answer the big questions that the
STEM fields aim to answer, there needs to be a collective interest to want to hear all the voices
that want to speak up and be heard.

UNDERREPRESENTATION OF LATINAS IN STEM 75

References

“Tumbleweed Tech". (2017). 2017 Fact book.
Bandura, A. (1977). Self-efficacy: toward a unifying theory of behavioral change. Psychological
Review, 84(2), 191.
Beede, D., Julian, T., Khan, B., Lehrman, R., McKittrick, G., Langdon, D., & Doms, M. (2011).
Education Supports Racial and Ethnic Equality in STEM. 2011. ESA Issue Brief #05-11.
Retrieved from http://www.esa.doc.gov
Bell Jr, D. A. (1980). Brown v. Board of Education and the interest-convergence dilemma.
Harvard Law Review, 518–533.
Bergerson, A. A. (2003). Critical race theory and white racism: Is there room for white scholars
in fighting racism in education? International Journal of Qualitative Studies in Education,
16(1), 51–63.
Buchmann, C., & DiPrete, T. A. (2006). The Growing Female Advantage in College
Completion: The Role of Family Background and Academic Achievement. American
Sociological Review, 71(4), 515–541. Retrieved from
http://www.jstor.org.libproxy1.usc.edu/stable/30039008
Carnevale, A. P., Smith, N., & Melton, M. (2011). STEM. Georgetown University Center on
Education and the Workforce. Retrieved from https://cew.georgetown.edu/wp-
content/uploads/2014/11/stem-complete.pdf
Charles, M., & Bradley, K. (2002). Equal but separate? A cross-national study of sex
segregation in higher education. American Sociological Review (Vol. 57).
Crenshaw, K.W. (1993). Beyond racism and misogyny: Black feminism and 2 Live Crew. In
Matsuda, M. J., Lawrence, C. R., Delgado, R., Crenshaw, K. W., & R, C. (1993). Words
UNDERREPRESENTATION OF LATINAS IN STEM 76

That Wound: Critical Race Theory, Assaultive Speech, And The First Amendment. Avalon
Publishing
Creswell, J. W. (2014). Research design: qualitative, quantitative, and mixed methods
approaches. Thousand Oaks, California: SAGE Publications.
Crisp, G., & Nora, A. (2012). Overview of Hispanics in science, mathematics, engineering and
technology (STEM): K-16 representation, preparation and participation. The Journal of
Higher Education.
Crisp, G., Nora, A., & Taggart, A. (2009). Student Characteristics, Pre-College, College, and
Environmental Factors as Predictors of Majoring in and Earning a STEM Degree: An
Analysis of Students Attending a Hispanic Serving Institution. American Educational
Research Journal, 46(4), 924–942. https://doi.org/10.3102/0002831209349460
Delgado, R. (1987). The ethereal scholar: Does critical legal studies have what minorities want.
Harv. CR-CLL Rev., 22, 301.
Diversity. (2003). In Merriam-Webster’s dictionary (11th ed.). Springfield, MA: Merriam-
Webster.
Earnest, J. (2015). Remarks by the President at White House Science Fair. Retrieved from
https://www.whitehouse.gov/the-press-office/2015/03/23/remarks-president-white-house-
science-fair
England, P., & Li, S. (2006). Desegregation Stalled: The Changing Gender Composition of
College Majors, 1971-2002. Gender & Society, 20(5), 657–677.
https://doi.org/10.1177/0891243206290753
George, Y. S., Neale, D. S., Van Horne, V., & Malcolm, S. M. (2001). In pursuit of a diverse
science, technology, engineering, and mathematics workforce. In American Association for
UNDERREPRESENTATION OF LATINAS IN STEM 77

the Advancement of Science (pp. 1–24).
Griffith, A. L. (2010). Persistence of women and minorities in STEM field majors: Is it the
school that matters? Economics of Education Review, 29(6), 911–922.
Harper, S. R., Patton, L. D., & Wooden, O. S. (2009). Access and equity for African American
students in higher education: A critical race historical analysis of policy efforts. The Journal
of Higher Education, 80(4), 389–414.
Hispanic. (2003). In Merriam-Webster’s dictionary (11th ed.). Springfield, MA: Merriam-
Webster.
Jackson, M. (2013). Fact Sheet: The State of Latinas in the United States. Retrieved from
https://www.americanprogress.org/issues/race/reports/2013/11/07/79167/fact-sheet-the-
state-of-latinas-in-the-united-states/
Jones, S. R., & McEwen, M. K. (2000). A conceptual model of multiple dimensions of identity.
Journal of College Student Development, 41(4), 405–414.
Kokkelenberg, E. C., & Sinha, E. (2010). Who succeeds in STEM studies? An analysis of
Binghamton University undergraduate students. Economics of Education Review, 29(6),
935–946.
Kolb, D. A. (1984). Experiential learning: Experience as the source of learning and development.
Kurpius, S. E. R., & Stafford, M. E. (2006). Testing and measurement: a user-friendly guide .
Thousand Oaks : Sage Publications.
Latina. (2003). In Merriam-Webster’s dictionary (11th ed.). Springfield, MA: Merriam-Webster.
Ma, Y. (2011). College major choice, occupational structure and demographic patterning by
gender, race and nativity. The Social Science Journal, 48(1), 112–129.
https://doi.org/https://doi.org/10.1016/j.soscij.2010.05.004
UNDERREPRESENTATION OF LATINAS IN STEM 78

Mann, A., & Diprete, T. A. (2013). Trends in gender segregation in the choice of science and
engineering majors. Social Science Research, 42(6), 1519–1541.
Maxwell, J. A. (2013). Qualitative Research Design: An Interactive Approach: An Interactive
Approach. SAGE.
Merriam, S. B., & Tisdell, E. J. (2016). Qualitative research : a guide to design and
implementation (4th Ed.). San Francisco, CA: Jossey-Bass.
Morgan, S. L., Gelbgiser, D., & Weeden, K. A. (2013). Feeding the pipeline: Gender,
occupational plans, and college major selection. Social Science Research, 42(4), 989–1005.
National Science Foundation. (2004). Women, Minorities, and Persons with Disabilities in
Science and Engineering (NSF 04-317). Retrieved from
http://www.nsf.gov/statistics/wmpd/
National Science Foundation. (2015). Women, Minorities, and Persons with Disabilities in
Science and Engineering (Special Report NSF 15-311). Retrieved from
http://www.nsf.gov/statistics/wmpd/
National Science Foundation. (2016). Science and Engineering Indicators 2016 (National
Science Foundation NSB-2016-1). Retrieved from
https://www.nsf.gov/statistics/2016/nsb20161/uploads/1/nsb20161.pdf
National Science Foundation. (2017). Women, Minorities, and Persons with Disabilities in
Science and Engineering (Special Report NSF 17-310). Retrieved from
www.nsf.gov/statistics/wmpd/
Perea, J. (1996). Suggested responses to frequently asked questions about Hispanics, Latinos and
Latinas. La Raza LJ, 9, 39.
Perez, T. E. (2014). Justice and Identity [Blog post].
UNDERREPRESENTATION OF LATINAS IN STEM 79

Price, J. (2010). The effect of instructor race and gender on student persistence in STEM fields.
Economics of Education Review, 29(6), 901–910.
Sciences, N. A. of, Engineering, N. A. of, & Medicine, I. of. (2007). Rising Above the Gathering
Storm: Energizing and Employing America for a Brighter Economic Future. Washington,
DC: The National Academies Press. https://doi.org/10.17226/11463
Tate IV, W. F. (1997). Chapter 4: Critical race theory and education: History, theory, and
implications. Review of Research in Education, 22(1), 195–247.
Toven-Lindsey, B., Levis-Fitzgerald, M., Barber, P. H., & Hasson, T. (2015). Increasing
persistence in undergraduate science majors: a model for institutional support of
underrepresented students. CBE Life Sciences Education
U.S. News. (2015). 2015 STEM Index Shows Gender, Racial Gaps Widen. Retrieved from
http://www.usnews.com/news/stem-index/articles/2015/06/29/gender-racial-gaps-widen-in-
stem-fields
Valdes, F. (1996). Latina/o Ethnicities, Critical Race Theory, and Post-Identity Politics in
Postmodern Legal Culture: From Practices to Possibilites. La Raza LJ, 9, 1.
World Economic Forum. (2016). The Global Competititveness Report 2016-2017. Retrieved
from https://www.weforum.org/reports/the-global-competitiveness-report-2016-2017-1
Xie, Y., Fang, M., & Shauman, K. (2015). Stem Education. Annual Review of Sociology, 41,
331–357.
Xie, Y., & Shauman, K. A. (2003). Women in science : career processes and outcomes.
Cambridge, Mass.: Cambridge, Mass. : Harvard University Press, 2003.
Yosso, T., Villalpando, O., Delgado Bernal, D., & Solórzano, D. G. (2001). Critical race theory
in Chicana/o education.
UNDERREPRESENTATION OF LATINAS IN STEM 80

Appendix A
Survey Protocol
Demographical information
1. What gender do you identify yourself as (Male, Female, Decline to State)?
2. What racial group do you identify with most (Latino/a/x,
3. What is your current declared major (Physical Sciences, Biological Sciences,
Engineering, other non-science related major)?
4. Have you changed your major recently?

Using the scale provided, please answer the following question.
1 = Strongly disagree, 2 = Disagree, 3 = Agree, 4= Strongly Agree

How important are the following to you regarding extrinsic motivation?

1 2 3 4
1. Being successful in my career

2. Having lots of money

3. Being able to find steady work

How important are the following to you regarding intrinsic motivation?

1 2 3 4
4. Helping others in the community

5. Working to correct inequalities

6. Being an example to people who associate with my gender

7. Being an example to people who associate with my race/ethnicity

How important are the following to you regarding your classroom experiences?

1 2 3 4
8. I think it is important if the instructor is the same gender that I
identify with
9. I think that I would learn better if the instructor is the same
gender that I identify with
10. I think it is important if the instructor is the same race that I
identify with
11. I think that I would learn better if the instructor is the same race
that I identify with

UNDERREPRESENTATION OF LATINAS IN STEM 81

What is your opinion on the following regarding being a STEM major?

1 2 3 4
12. I could see myself working in a field of science and/or math in
the future
13. I have always found science and/or math enjoyable

14. I have always done well in my previous science and/or math
classes (grade of B or better)
15. I have always found science (biology, chemistry, physics, etc.)
classes to be difficult
16. I have always found math classes to be difficult

What is your outlook on STEM careers?

1 2 3 4
17. I see careers in science and/or math as being important
18. I believe that science and/or math careers are high-paying
19. I believe that I will always find a job in the science and/or math
fields
20. I am confident that my college work will prepare me for a career
in the science and/or math fields

UNDERREPRESENTATION OF LATINAS IN STEM 82

Appendix B
Interview Protocol

Good morning (afternoon). You have been chosen to be interviewed because you have
been identified as a Latina who has declared as a STEM major. My research is focusing on the
factors that may influence the underrepresentation of Latinas in STEM. My aim is to determine
what has influenced your decision to become, remain, or discontinue being a STEM major with
the hope of developing a better understanding of some of the factors that influence men and
women such as yourself.
There are a variety of questions that will be covered in this interview that may lengthen
the time of this interview but it should last between thirty and forty-five minutes
To ensure accuracy and to facilitate my note-taking, I will be recording our conversation today.
The information that you provide will only be made available for this research. Thank you for
participating in this research.
1. What were some of the factors that made you decide on majoring in STEM?
2. If you had a daughter, would you advise her to also major in a STEM field?
a. If you are close to your mother, do you think that you feel that way because of
her?
3. Based on your experience in college so far, do you feel that it is important that the people
teaching the courses represent your ethnicity and/or gender?
4. If all of your instructors were female and Hispanic, would that matter to you?
5. Do you feel more or less motivated if you have a female faculty member teaching your
science and math courses?
a. Why is that?
6. What has your experience been like so far in your math classes?
a. Does the difficulty in this class make you feel less motivated to continue as a
STEM major?
b. If the math requirement were less rigorous, would that motivate you more?
7. What science classes are you currently taking? What has your experience been like so far
in this (these) class(es)?
a. Does the difficulty in this class make you feel less motivated to continue as a
STEM major?
b. Which class(es) in particular makes you feel this way (if any)?
8. Do you feel that there is anything that may cause you to change your mind about STEM?
a. For example, cost of tuition, or prospects of a job out of school?
Thank you again, for agreeing in participating in this research!

Abstract (if available)

Abstract The purpose of this study is to expand on the current knowledge surrounding some of the factors that contribute to the underrepresentation of Latinas majoring in the STEM fields as well as understand what theories may help explain what is causing the underrepresentation of Latinas in STEM. The theoretical framework of this study involves the Model of Multiple Dimensions of Identity, Experiential Learning theory, and LatCrit. This study employs a mixed methodology approach using surveys and interviews to address the proposed research questions. Data collection will take place over a two-month period. An online survey will be distributed to students and interview participants will be chosen randomly. Interviews will be recorded and subsequently transcribed. Pseudonyms will be used to ensure the anonymity of the participants of this study. Data collected will be analyzed using software designed to facilitate the coding of the data. The use of both quantitative (surveys) and qualitative (interviews) methods triangulates the data and ensures that any issues of validity and reliability are dealt with. Results of this study show that there are contrasting views on how Latinas viewed modeling and that prior math and science success may not play a large role on whether these students remain in STEM.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Navigating success: what factors influence success among Latina STEM community college students?

PDF

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

PDF

Impact of participation in STEM organizations and authentic learning experiences on women of color engineering students

PDF

The principal's perspective: essential factors when implementing integrative STEM in middle school

PDF

The underrepresentation of Latinas as K−12 school district superintendents: an evaluation study

PDF

Mid-level Latina leaders in predominantly White institutions: balancing identities and cultural incongruity

PDF

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

PDF

Girls in cybersecurity: an evaluation study

PDF

Leadership for K-12 STEM integration: how superintendents champion the advancement of effective integrated STEM education within their districts

PDF

Integration of STEM and gardening for urban elementary youth

PDF

STEM integration: a case study of a 21st century skills professional development program for math and science teachers in an urban school

PDF

STEM identity development: examining the experiences of transfer students

PDF

An examination of the perceptions of Latina leaders in STEM: identity, culture, and institutional racism

PDF

A needs assessment and evaluation of an elementary afterschool STEM program in a small urban school district

PDF

How are teachers being prepared to integrate technology into their lessons?

PDF

Vocational education graduates: a mixed methods analysis on beliefs and influences of career choice and persistence

PDF

Integrating STEM instruction through identifying interest and identity in STEM professionals

PDF

The effects of math anxiety and low self-efficacy on students’ attitudes and interest in STEM

PDF

College and career readiness through high school STEM programs: an evaluation study

PDF

Promoting STEM integration, interest and identity among elementary school students

Asset Metadata

Creator Lobos, Jedidiah Izael Afable (author)

Core Title Examining the factors leading to the continued underrepresentation of Latinas in STEM

School Rossier School of Education

Degree Doctor of Education

Degree Program Education (Leadership)

Publication Date 07/23/2019

Defense Date 04/30/2019

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

Tag college,disparity,diversity,experiential learning,extrinsic,Hispanic,identity,intrinsic,LatCrit,Latinas,mixed methods,OAI-PMH Harvest,persistence,STEM,underrepresentation,Women

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Freking, Frederick (committee chair), Jones, Jennah (committee member), Maddox, Anthony (committee member)

Creator Email jlobos@usc.edu,jlobos1@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-189370

Unique identifier UC11660646

Identifier etd-LobosJedid-7580.pdf (filename),usctheses-c89-189370 (legacy record id)

Legacy Identifier etd-LobosJedid-7580.pdf

Dmrecord 189370

Document Type Dissertation

Format application/pdf (imt)

Rights Lobos, Jedidiah Izael Afable

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA