DISSERTATION OVERVIEW
Introduction
This dissertation critically examines how systems of oppression intersect with the race, class, and gender dynamics of first-year engineering students in engineering schools across the United States.
The research highlights the significant disparities in the engineering profession, particularly affecting racial minorities and women, a concern that has been addressed by national agencies like the National Science Foundation since the 1970s In 2013, women represented only 14.8% of the engineering workforce, a stark decline from their 50.8% national proportion in 2010 Additionally, Black individuals made up just 3.6% of the engineering field, compared to a national representation of 12.6% Similarly, Hispanics accounted for 6.6% of the profession, while their overall population percentage was 16.3% Although the Asian minority has a higher representation in engineering than their national distribution, they still face challenges in promotional opportunities within the field.
Orr, Ramirez, and Ohland (2011) highlighted the increasing concern regarding socio-economic status and equity in education Unlike available data on race and gender, there is a lack of statistics on the class origins of current engineers Research by Ohland et al (2011) utilized socio-economic indicators as proxies for class origin, revealing that students from low economic backgrounds were generally less likely to pursue college compared to their higher-status peers However, among those who did attend college, the choice to study engineering was similar across both economic groups.
Researchers’ justifications for increasing disparity have often invoked the general benefits of diversity in engineering (Beddoes, 2011) For instance, Committee on
Diversity in the engineering workforce offers several advantages, such as enhancing equity, addressing workforce shortages, and maintaining economic competitiveness Additionally, it is driven by regulatory compliance, the need for engineers skilled in diversity, and the aim to challenge the dominance of Whites in the field However, despite the acknowledged benefits of diversity, research is limited on its actual impact on achieving material equity for women and racial minorities in engineering There is insufficient evidence to suggest that increased diversity has resulted in these groups attaining more leadership roles, securing more patents, or receiving increased federal funding.
Beddoes (2011) explored how discourses addressing disparities between groups, known as underrepresentation, can inadvertently reinforce harmful assumptions that hinder progress in improving access This study is significant as it highlights that while research on race, class, and gender aims to enhance access to engineering education and the profession, it often relies on a deficit model This model, as noted by Scott, Sheridan, and Clark (2015), attributes disparities to the shortcomings of students' identities, communities, and backgrounds, which can lead to destructive implications in understanding underrepresentation.
Muller (2003) exemplifies a deficit model to address the underrepresentation of racial minorities, asserting that a significant preparation gap exists for African-American, Hispanic, and Native American students compared to their white and Asian peers Additionally, Muller applies this deficit model to explain the underrepresentation of women in engineering.
Women are more likely than men to select fields of study that they believe will benefit society, yet engineering and related sciences are often not viewed as contributing to the social good Muller attributes women's underrepresentation in engineering to their different choices compared to men, suggesting that this disparity stems from women's reluctance to pursue the same paths as their male counterparts For additional insights on deficit model explanations regarding race and gender disparities in engineering and engineering education, refer to Oakes (1990) and Board (2003).
A Bell (1989), National Academy of Engineering (2008), Directorate for Engineering
Underrepresented groups in engineering are often perceived as underprepared or incompatible with the profession's goals, leading to a flawed view of their personhood This perspective unjustly attributes racial and gender disparities in engineering enrollments to the individuals within these groups, rather than addressing systemic issues.
Research methods should not be justified by their outcomes if they perpetuate deficit model justifications for disparities in representation A theory suggests that hostility drives capable women to leave science, math, and engineering (SME) majors, reinforcing the notion of a deficit in representation.
The hostility faced by women in SMEs from certain faculty members and male peers stems from their disruption of a traditional selection process that favors young men for entry into elite circles This environment often leads young women to doubt their capabilities in the field of science.
Insufficient independence in learning styles, decision-making, and self-assessment can hinder students' ability to thrive, especially when faced with a lack of faculty support and rejection from male peers.
Seymour and Hewitt's theory suggests that hostility contributes to women's underrepresentation in SME majors by framing them as lacking confidence and independence, which inadvertently reinforces the deficit model While their intentions may be to improve the hostile environment, this perspective diverts attention from the structural advantages men have in these fields Beddoes (2011) highlights that problematic discourses of underrepresentation shape the focus of interventions, noting that while confidence and independence are crucial for academic success, similar interventions are not applied to address male underrepresentation in female-dominated disciplines.
Interventions aimed at addressing women's perceived deficits are problematic as they reinforce the patriarchal structure that Seymour and Hewitt seek to change While many believe that certain norms must be met for graduation, it is crucial to recognize that deficit model discourses explaining enrollment disparities lead to ineffective interventions By critically examining these discourses, we can develop more effective strategies that tackle the root causes of enrollment disparities.
Gorski (2011) highlighted a significant flaw in deficit model explanations, pointing out the absence of a historical link between current disparities and long-standing systems of oppression in the US This article references the interlocking political systems of “imperialist, White supremacist, capitalist, patriarchy” as described by hooks (2004) According to hooks, patriarchy is a political system that asserts male dominance over those deemed weaker, particularly females, and maintains this power through psychological terror and violence Furthermore, imperialism, as practiced by the US, justifies theft under the guise of a moral imperative that claims such actions benefit others (Brayboy, 2005) White supremacy, as defined by Ansley (1997), further complicates these oppressive structures.
A political, economic, and cultural framework exists where White individuals predominantly hold power and resources, fostering both conscious and unconscious perceptions of White superiority and entitlement This dynamic perpetuates a daily enactment of White dominance and non-White subordination across various institutions and social contexts Additionally, capitalism serves as an economic system primarily focused on profit generation.
“productivity and profit or the accumulation of capital for the purposes of reinvestment, market expansion, and greater profits” (LaMothe, 2016, p 25)
Theoretical Perspective and Methodology
This dissertation comprises three distinct studies, each presented in its own chapter Chapters 2-4 utilize a formulation of critical theory relevant to engineering and engineering education, drawing insights from critical race theory, sociology, women’s and gender studies, Black feminist thought, and science, technology, and society (STS) According to Baber (2015), the aim of critical theory is to reveal the concealed power structures that uphold its own authority while disempowering others.
The aim of applying this theory is to deepen my understanding of how race, class, and gender intersect within engineering education, which is part of a broader, historically entrenched system of societal oppression This system is intricately shaped by the unique and semi-autonomous characteristics of the engineering profession and discipline, as described by Bourdieu and Passeron (1977).
As I argued earlier in my criticism of deficit model explanations, researchers’ failure to incorporate theory about the historic continuity of systemic oppression as a topic of research is highly problematic because it reproduces a post-imperialist-White supremacist-capitalist-patriarchal vision of the engineering field Hence, part of my goal for using critical theory was to bring history back into explanations of disparity For instance, while the 14th Amendment granted equal rights to all citizens in 1868, it did not end the historic continuity of White supremacy right then and there (Alexander, 2010) Other policies that promoted White supremacy such as the 1887 Dawes Act, the 1924 Immigration Act, the movement of Japanese to concentration camps, and poll taxes, were all instituted after 1868 These policies all directly affected the lives of people of other as forms of systemic oppression even though in theory, all people were protected equally under the 14th Amendment But, more personally, I also wanted to connect my research and my own personhood to the important social justice movements of the present-day I believe that engineering education research should be able to apply the important issues of the day into its pedagogy both for the benefit of engineering students as well as the public-at-large
The Black Lives Matter (BLM) movement, founded in 2013 by Alicia Garza, Patrisse Cullors, and Opal Tometi in response to the acquittal of George Zimmerman for the killing of Trayvon Martin, has significantly influenced my research, particularly in chapter 3 The movement emerged to address the historical pattern of violence against Black lives perpetuated by institutionalized White supremacy in the U.S Following the 2014 police killing of Mike Brown, the hashtag “#BlackLivesMatter” gained traction on social media, leading to widespread protests against police impunity Despite the ongoing relevance of BLM, which highlights the racial inequity in police brutality—where Black men aged 15 to 19 were 21 times more likely to be killed by police than their White counterparts from 2010-2012—progress remains slow While more officers have been indicted for fatal shootings of Black individuals since BLM's inception, as of August 2015, no officer indicted for murder that year was found guilty, underscoring the continued need for systemic change to ensure that Black lives truly matter.
Recognizing the significance of the Black Lives Matter (BLM) movement in addressing systemic oppression, I challenged myself to connect its activism with engineering education research and the understanding of disparities The principles of the BLM platform resonate deeply with me, as I witness the ongoing dehumanization of marginalized communities.
The Black Lives Matter movement highlights the widespread acceptance of dehumanizing deficit model explanations for societal disparities This movement has deepened my understanding of how critical theory can elucidate contemporary issues of race and gender inequality.
The terms "disparity" and "underrepresentation" highlight the unequal proportions of racial minorities and women in engineering and engineering education relative to their overall presence in the general population To explain the underlying causes of this disparity, I examine theories related to segregation, focusing on two main types: occupational segregation and public school segregation, which contribute to gender and racial inequalities.
Occupational segregation refers to the disparity among groups based on social characteristics within various occupations (Weeden, 2007) The literature on occupational segregation offers several theoretical perspectives to explain the racial and gender disparities in engineering In this context, I will focus on the framework proposed by Reskin and Roos.
The 1990 theory of job and labor queues highlights that employers' race and gender preferences significantly contribute to occupational segregation I selected Reskin and Roos's theory due to its research-backed insights into systemic oppression, particularly illustrating the historical devaluation of women's labor compared to men's In chapters 2 and 4, I utilize the job and labor queues theory within a quantitative methodology, employing statistics to analyze and discuss the findings.
Public school segregation refers to the separation of students in publicly-funded schools based on race and ethnicity (Rothstein, 2015) Recent studies by Orfield, Frankenberg, Ee, and Kuscera highlight the ongoing issues related to this form of segregation.
Despite the US Supreme Court's 1954 ruling that declared state-sanctioned de jure segregation unconstitutional, a 2014 study revealed that school segregation levels had reverted to those of 1968 Inspired by the work of Orfield et al., I analyze segregation levels in first-year engineering programs in chapter 3 and connect these findings to trends in public school segregation.
The analytical methods employed throughout this article are fundamentally quantitative Quantitative approaches have been widely utilized in research on public school segregation, as demonstrated by studies from Fiel (2013), Orfield et al (2014), and Orfield and Ee (2014) Likewise, the examination of occupational segregation has also depended on quantitative methods, evidenced by the works of Tang (1997a) and Skaggs (2008) However, it is important to note, as highlighted by Pawley (2013), that these methods have their limitations.
Statistical methods often fall short in analyzing the experiences of underrepresented groups due to their low numbers, which can render findings statistically insignificant For instance, my research on first-year enrollments and public school segregation by race reveals that I can only draw conclusions about Asians, Blacks, Hispanics, Whites, individuals identifying as Two or More Races, and Non-resident Aliens, as their populations are sufficiently large Conversely, the data collection issues surrounding Native Americans, Native Alaskans, Native Hawaiians, and Pacific Islanders result in inconsistencies or their absence in datasets, preventing any conclusions about their segregation in first-year engineering programs.
This dissertation explores the impact of systems of oppression on the engineering profession through two key examples: public school segregation and occupational segregation It presents original research in the form of three articles, each independently structured yet unified by a common theme.
This analysis focuses on first-year engineering enrollments by race and gender, highlighting its significance as a critical threshold in engineering careers, as noted by E M Holloway (2013) The majority of aspiring engineers must enroll in undergraduate programs to pursue their careers (Lichtenstein, 2009) By examining race and gender patterns in first-year enrollment, this study aims to uncover how systemic oppression influences who can become an engineer Detailed descriptions of three research articles, Chapters 2-4, and the conclusion chapter are provided below.
Chapter Overview
In Chapter 2, I ask the following research question:
RQ2.1: How do institutional characteristics affect the odds of women’s enrollment in engineering programs compared to men’s?
This research investigates the relationship between first-year female and male engineering student enrollments and the institutional characteristics of universities offering engineering programs By integrating data from the American Society for Engineering Education and the National Center for Education Statistics, I developed generalized linear models to analyze these relationships The findings are contextualized within Oldenziel's (1999) exploration of the advantages that early female entrants to engineering often had, such as patrimonial sponsorship or family wealth, and Bourdieu and Passeron’s (1977) theory of habitus I propose that institutions with higher odds ratios of female enrollment in engineering possess characteristics that align with the habitus of prospective women engineers, thereby reducing the likelihood of their engineering contributions being undervalued.
In Chapter 3, I examine national and California-specific longitudinal trends in first-year engineering enrollments, focusing on gender and race to address key research questions.
RQ3.1 How has race and gender composition progressed over time?
RQ3.2 Do changes in composition indicate an improvement in race and gender diversity? And
RQ3.3 How do changes in race and gender composition at the first year level relate with composition at the public school level?
This study examines the correlation between public school segregation and segregation in first-year engineering cohorts by analyzing enrollment data It utilizes two key markers of segregation: the exposure levels of typical students to peers of different races and the concentration of students from majority non-White schools The findings indicate that the gap between White and Asian first-year students compared to Hispanic and Black first-year students has increased alongside rising levels of public school segregation.
In Chapter 4, I explore the connection between odds ratios and institutional characteristics, focusing specifically on the first-year engineering enrollments of Hispanic, Black, and Asian students in comparison to their White counterparts This analysis aims to address the research question regarding the disparities in enrollment among these racial groups.
RQ 4.1 How do institutional characteristics relate with the odds of Black, Hispanic, and Asian students’ participation in engineering programs compared to White students?
Again I analyze the relationship between institutional characteristics and odds ratios through the use of linear regression, and I rely on theory of habitus and the history of racial minority entry into engineering to discuss results I posit that significant differences between odds ratios could be due to institutions with characteristics that allow students of color to overcome racial discrimination when entering into the engineering labor market
In the Conclusion of my dissertation, I explore the implications of the research presented in Chapters 2-4 and suggest future research avenues My findings establish a foundation for investigating engineering interests among pre-college populations, focusing on race, class, and gender Additionally, I recommend strategies for higher education institutions to enhance the enrollment of women and people of color in engineering programs Lastly, I predict the future composition of the engineering field.
THE ILLUSION OF CHOICE: AN ANALYSIS OF UNIVERSITY
Abstract
This study investigates the impact of institutional characteristics on women's enrollment in engineering, challenging the notion that women's choices are solely responsible for their underrepresentation in the field By adopting a structural perspective, the research aims to identify factors that could enhance female participation in engineering Utilizing 2014 enrollment data from the American Society of Engineering Education and the Department of Education’s NCES, the study employs generalized linear models to analyze the relationship between institutional characteristics and women's enrollment odds The results indicate that specific institutional factors, particularly related to class status, significantly influence women's enrollment in engineering The findings suggest potential policy recommendations for universities to increase female enrollment in their engineering programs.
Introduction
Beddoes and Pawley (2014) emphasize that prevailing narratives regarding the underrepresentation of women in engineering often attribute this issue to personal choice Key studies, including those by J A Bell (1989) and the National Academy of Engineering (2008), suggest that women opt for different career paths These sources advocate for interventions aimed at delivering alternative messages to encourage more women to enter the engineering field.
However, Beddoes and Pawley (2014), Cech (2013), Oldenziel (1999), and Tang
The underrepresentation of women in engineering is not merely a matter of personal choice, but rather a reflection of historical gendered labor relations in the US This article examines how occupational segregation has historically influenced women's entry into the engineering field, coining this phenomenon as the "illusion of choice." The literature review is structured into two parts, with the first focusing on occupational segregation theory and the impact of class status on women's participation in engineering It highlights Reskin and Roos’s (1990) theory of job and labor queues, which provides a structural perspective on the relationship between women's enrollment in engineering programs and institutional characteristics This theory underscores the historical gendered labor structures that contribute to occupational segregation, linking institutional factors to broader labor dynamics that shape the engineering profession by gender.
In the second part of the literature review, I present an analysis that justifies the chosen institutional characteristics for the study These characteristics align with the generic framework of organizational segregation proposed by Stainback et al (2010), which I will further elaborate on below.
Research Question
In more concrete terms, I have used Reskin and Roos’s theory to answer the following research question:
RQ2.1: How do institutional characteristics affect the odds of women’s enrollment in engineering programs compared to men’s?
Literature Review
2.4.1 Segregation in Engineering by Gender The study of Occupational Segregation (OS) by gender is represented in an extensive literature that has theorized the mechanics of how gender functions to segregate occupations, including engineering Many definitions of gender exist in the literature; however for the purposes the study here, I have used the definition provided by the sociologist Connell (2009): “[T]he structure of social relations that centres on the reproductive arena, and the set of practices that bring reproductive distinctions between bodies into social practices.” (p 11) Connell frames gender as a formation within a larger social context; hence gender formations change across different social arenas (i.e countries, cultures, occupations.)
Research on occupational segregation (OS) by gender, including studies by Reskin and Roos (1990), Tomaskovic-Devey (1993), and Stainback and Tomaskovic-Devey (2012), has elucidated the mechanisms that perpetuate gender segregation in the workforce Despite the absence of legal gender discrimination in the US since the 1964 Civil Rights Act, disparities in occupational representation persist OS is characterized as the unequal distribution of social traits among different occupations (Weeden).
Okamoto and England (1999) identified two sociological explanations for sex segregation: "supply-side" and "demand-side" theories Demand-side theories highlight how institutional factors, including formal hiring and promotion policies or informal job assignments based on gender stereotypes and discrimination, contribute to the perpetuation of segregation In contrast, supply-side theories focus on individual worker characteristics, such as values, aspirations, qualifications, and roles, to explain occupational sex segregation in engineering.
Historical literature indicates that patriarchal structures and practices in the US engineering field have systematically excluded women, leading to significant gender segregation (Bix, 2002, 2004; Oldenziel, 1999; Tang, 2000) This study adopts bell hooks' (2004) definition of patriarchy as a political-social system that asserts male dominance and superiority over women, often maintained through psychological and physical violence (1999, p.1) An example of this patriarchal influence is evident in the 19th century, where historical records reveal minimal female participation in engineering Social practices of the time barred women from labor crews, effectively preventing their entry into the engineering profession.
Patriarchy is an integral part of the US political system, deeply embedded in the fabric of society (Oldenziel, 1999; hooks, 2004).
Reskin and Roos (1990) established a significant connection between occupational segregation by gender and historical labor relations, introducing the theory of job and labor queues They proposed that the composition of workers in any given occupation results from a dual process: the job queue, where workers seek the most desirable positions characterized by high wages and societal status, and the labor queue, where employers prefer candidates with specific traits such as gender, race, education, experience, and social affiliations Their research highlighted a historical bias favoring men over women in certain jobs, which they termed the gender queue.
Reskin and Roos’s gender queue theory highlights the ongoing gender disparities in engineering, suggesting that a gender queue may contribute to the underrepresentation of women in the field Recent reports on the scarcity of women in the tech industry, a primary employment sector for engineers, reinforce this theory, indicating that despite the industry's attractiveness, gender imbalances persist (Kokalitcheva, 2015; Ricker).
Research indicates that women in engineering face significant disparities compared to their male counterparts Studies by McIlwee and Robinson (1992) and Tang (2000) reveal that women are less likely to be promoted to desirable engineering positions Additionally, Mills et al (2014) found a persistent 16% pay gap between male and female engineers, along with a higher attrition rate for women, who leave the profession at 12.9% compared to 9.8% for men These findings highlight the insecurity of women's positions in engineering and suggest the presence of a gender queue The literature also emphasizes the male-dominated culture within engineering, as noted by Downey and Lucena (1995), and the lasting impact of patriarchal structures in engineering education, described by Hacker (1989) as a "hidden curriculum" that fosters the exclusion of women.
Historical research by Oldenziel (1999) highlights the crucial role of social class in women's entry into engineering during a time of legal gender discrimination By the early 20th century, significant legal and social changes, influenced by the suffrage movement, facilitated a notable increase in the number of women engineers in the U.S., with women representing 3% of all engineers in the first half of the century Many of these women gained informal training through familial connections, allowing them to access engineering education and enter the profession.
In 1999, it was noted that women entering the engineering field often came from higher social classes compared to their male counterparts, highlighting the significance of class position in this context.
This study utilizes Marx and Engels’s definition of class, which categorizes individuals based on their relationship to the means of production, distinguishing between the bourgeoisie, who own these means, and the proletariat, who sell their labor for wages Engineers present a unique case within this framework; while they typically trade labor for wages, their roles often extend into management, positioning them in a "middle" class Unlike traditional proletariat workers, engineers are viewed as "trusted workers" by employers, performing significant managerial tasks Additionally, engineers tend to align politically more closely with management than with the working-class technicians, reflecting their distinct professional identity as a "loosely coupled profession" due to their strong ties to management.
Engineers have not unionized despite a collective need for contracts with management due to their proximity to management and their detachment from broader labor movements This unique position places engineers in a complex class status, straddling the line between proletariat and bourgeoisie Furthermore, as Wisnioski (2008) notes, engineers can also be full members of the bourgeoisie, often acting as entrepreneurs and controlling the means of production themselves.
Reskin and Roos’s (1990) labor queue framework aligns with Oldenziel’s (1999) historical research by illustrating how some women in the early 20th century became engineers despite a male-dominated workforce Family ties and class membership provided these women access to essential engineering skills and education, making them competitive candidates Understanding the motivations behind women's desire to enter engineering is crucial, as the appealing class status of engineering compared to proletariat jobs may drive job-seekers toward this field Apple (1986) noted that historically, male-dominated occupations often transitioned to female-dominated ones as men sought more desirable roles, leading to the proletarianization of these fields Despite women's entry into engineering since the early 1900s, the profession remains linked to desirable occupational traits, as highlighted by the Bureau of Labor statistics.
According to statistics from 2015, engineering professions consistently rank among the top 10 highest-paying jobs based on average annual salaries Research by Mills and colleagues in 2014 revealed that female STEM workers earn, on average, 33% more than their counterparts in other fields, while male STEM workers earn 25% more than men in different occupations Wisnioski (2008) described engineers as "innovators," highlighting their crucial role in driving advancements in technology and industry.
Entrepreneurs play a crucial role in shaping the US economy, with a significant portion of the public recognizing that engineers drive economic growth, as reported by the National Academy of Engineering (2008) Additionally, the favorable positioning of engineering within the labor market may encourage more women to pursue careers in this field.
Data and Methods
I conducted my research using two distinct data queries, utilizing the restricted-use ASEE Online Data Mining Tool This tool provided comprehensive data on engineering enrollments, faculty demographics, and institutional characteristics from over 300 schools, which I subsequently downloaded in a CSV file format from the American Society of Engineering Education.
In 2015, the steps taken to query the Online Data Mining tool can be found in Appendix A, while the institutional characteristics obtained are detailed in the dependent variables section below.
The Mining Tool query generated gender enrollment statistics for all full-time first-year students across participating US engineering programs, encompassing a total of 151,546 male and female students majoring in engineering A sample of the data obtained from the Mining Tool query is presented in Table 1 below.
Table 1 Example of Data Mining Tool Query (Only the first few race/ethnicity/class variables were displayed.) IPEDS Unit ID Year School State Cauc M FT Fresh Cauc F FT Fresh
Before I conducted the analysis of ASEE data, I changed the term “Freshmen” to first year to provide a more inclusive term that more accurately describes the positions of incoming students Watts (2009) gave several reasons to support the adoption of first year over Freshmen, which highlighted the weaknesses implicit in the term freshmen to accurately capture the present nature of the nationwide college student body Watts noted that “freshman” holds the connotation of a student enrolled in college straight out of high school, which does not accurately describe the large body of students who enter college a significant period of time after high school Watts also noted that the term was overtly gendered in that women are not acknowledged by the term Hence, I adopted the term
“first year” to address the criticisms of the term “freshman.”
I accessed the Integrated Postsecondary Education Database (IPEDS) from the U.S Department of Education to gather data on first-year full-time student enrollments by race and gender, as well as faculty demographics and institutional characteristics for over 7,000 universities This data was compiled into a CSV file, and detailed querying steps can be found in Appendix A The total student population surveyed from colleges with engineering programs amounted to 748,577, with the relevant institutional characteristics presented in Table 5.
I chose institutional characteristics based on their fit within Stainback and et al.’s
The 2010 typology suggests that certain variables are linked to distinct enrollment patterns in engineering between men and women In the results and discussion section, I elaborate on the significant relationships observed among these variables.
The IPEDS query offers gender enrollment statistics for first-year full-time male and female students at the university level, with a sample of this data presented in Table 2 below.
Table 2 Example of Data Mining Tool Query (Only the first variables were displayed.) unitid institution year address City State ZIP code
I answered the research question by appending the Data Mining Tool query with the IPEDS query using the Query Wizard feature in Microsoft (MS) Access Afterwards,
I cleaned and processed the data to assure quality Below, I describe this process in greater detail
I employed the Query Wizard feature in MS Access to merge two data sources into a unified dataset, encompassing university-level enrollment, faculty, organizational characteristics, and engineering-level data To achieve this, I created a query utilizing a crosswalk, which is a common element shared between the datasets Each dataset featured a Unit ID, a numerical identifier assigned to universities by the US Department of Education Consequently, I established the Unit ID variable as the crosswalk, ensuring consistency across all schools.
To utilize the crosswalk, I instructed MS Access to generate a table that establishes a one-to-one correspondence of Unit IDs between the two datasets, ensuring that only variables from schools with matching Unit IDs were included For a detailed account of this process, please refer to Appendix A.
2.5.2.2 Cleaning and Processing the Data After appending each data source, I had to clean and process data to ensure quality In this step of the process, I eliminated data sources that provided no gender enrollment data for men or women, or which had no men or women enrolled I have created Appendix B to list the names of schools that were included and excluded in the study for the interested reader
2.5.3 Dependent Variables The research question dictated the creation of odd ratios (ORs) that quantify the odds that women will be enrolled in engineering at one university compared to the odds that men will be enrolled at within the same university An odds ratio requires the calculation of two separate odds, which are then divided to create an OR To create odds that women will be enrolled in engineering, I divided the number of women enrolled in engineering by those not enrolled in engineering:
The study calculated the gender odds ratio (GOR) for engineering enrollment at each university, comparing the odds for women and men This process was consistently applied to assess men's enrollment odds in engineering programs.
A Gender Odds Ratio (GOR) greater than 1 indicates that women have better odds of enrolling as engineers compared to men, while a GOR between 0 and 1 suggests that women face worse odds A GOR of 1 signifies equal enrollment odds for both genders The GOR formula used in this analysis is a modified version of the calculation by Skaggs (2008), which was originally designed to assess the odds of women becoming managers in an organization.
Osborne (2006) raised concern that the significance of odds ratios is difficult to communicate compared to a different statistical measurement called relative risk (RR)
Relative risk (RR) quantifies the likelihood of an event occurring For example, the gender relative risk (GRR) measures the probability of women being engineers at a school compared to men This is calculated by dividing the number of women engineers by the total number of women enrolled, and then dividing that by the number of men engineers over the total number of men enrolled.
Osborne (2006) suggested that relative risks (RRs) are more straightforward to communicate, as likelihood ratios are generally easier to grasp than odds ratios (ORs) However, being more familiar with one measure does not guarantee better research outcomes compared to the other For this study, I selected odds ratios as the dependent variables because they are considered the standard in the field (Skaggs).
Results and Discussion
Table 6 presents a profile of selected mean school characteristics based on the dependent variables utilized in this study The calculated values indicate either the mean and standard deviations for continuous variables or the frequency and overall share for categorical variables, with categorical variables and their nested categories highlighted in italics.
Table 6 Descriptive Profile of Dependent Variables
Variable Mean/Freq SD/Share Total
Female student-to-female faculty ratio 30.59 19.38 294
Private not-for-profit (no religious affiliation)
2.6.2 Female-Male Odds of First Year Engineering Enrollment
Table 7 illustrates the average distribution of full-time first-year women and men in engineering schools, highlighting the average gender odds ratio for women's enrollment compared to men's The average natural log transformation of the gender odds ratio is -1.8, corresponding to an average odds ratio of 0.17 This figure indicates that the proportion of women enrolled in engineering, relative to those not enrolled, is approximately one-sixth of the men's enrollment proportion, which offers a significant qualitative insight into gender representation in the field.
Table 7 presents the average share of first-year full-time engineers categorized by gender, along with the average gender odds ratio and the average natural logarithm of the gender odds ratio, as reported by the American Society of Engineering Education in 2015 and the U.S Department of Education's Institute of Education.
Male First Year Engineering Proportion
Natural Log of the Gender Odds Ratio Average (SD) 0.19 (0.083) 0.81 (0.083) -1.8 (0.67)
Table 8 presents the findings for Model 1, which analyzes the impact of Organizational Inertia independent variables on the natural log of the gender odds ratio The analysis indicates that the university-level admission rate, total 4-year university cost, and religious affiliation of the institution are significant predictors of higher odds ratios for women's enrollment in engineering relative to men.
More selective universities had higher odds of women’s enrollment compared to men’s
A 1% decrease in the admissions rate enhances women's enrollment odds in engineering by a factor of 1.012 compared to men, assuming all other factors remain constant For instance, University A, with a 20% admissions rate, is expected to have 1.13 times higher odds of female enrollment in engineering than University B, which has a 30% admissions rate Additionally, University A's odds ratio for women enrolled in engineering is predicted to be 1.65 times greater than that of University C, which has a 70% admissions rate.
Rising university costs have positively influenced women's enrollment in engineering programs Specifically, for every $1,000 increase in university expenses, the likelihood of women pursuing engineering degrees rises by a factor of 1.0131 compared to their male counterparts.
Increasing a university's cost by $10,000 raises the likelihood of women's enrollment in engineering by 1.14 times, while a $20,000 increase boosts the odds ratio to 1.3 times.
Schools without religious affiliation had an odds ratio 1.29 times greater than their religiously affiliated counterparts, holding other factors constant Additionally, the model significantly explained the variance in enrollment odds for both women and men, accounting for 50% of the variance in a sociological context.
Table 8 presents the estimates of the Organizational Inertia Coefficient, as reported by the American Society of Engineering Education (2015) and the U.S Department of Education, Institute of Education Sciences (2016) The significance levels are indicated, with (*) denoting a two-tailed significance value of p < 0.05, (**) indicating p < 0.01, and (***) representing p < 0.0001.
Secondary school GPA Neither required nor recommended -0.0241 0.1743
Secondary school rank Neither required nor recommended 0.0236 0.0940
Recommendations Neither required nor recommended -0.104 0.098
The findings from Model 1 reveal a noteworthy negative correlation between the selectivity of admissions rates and women's enrollment in engineering This suggests that as universities become more selective, the likelihood of women enrolling in engineering increases relative to men Consequently, it appears that women are more inclined to apply to engineering programs perceived as more selective.
Women aspiring to pursue engineering are likely drawn to more selective schools, as research indicates a connection between their entry into engineering, habitus, and social class These women may be influenced by the perceived prestige of selective institutions, which are associated with greater opportunities for acquiring the necessary capital to enhance their status in the engineering field Consequently, they may be more inclined to enroll in these prestigious schools.
Research indicates that engineering institutions may employ selectivity to achieve more gender-equal admissions For example, Holloway et al (2014) recommended implementing new admissions policies to address the implicit biases of decision-makers that could disadvantage women.
The engineering school successfully increased female enrollment by implementing new admissions criteria that minimized gender bias, allowing for the selection of women who, on average, demonstrated higher performance than their male counterparts Research by DiPrete and Buchmann (2014) highlights that for nearly 150 years, girls have consistently outperformed boys academically, achieving better grades in all major subjects, including math and science In 2004, women had a mean GPA that was 0.24 points higher than that of men Consequently, this superior academic performance among women may significantly enhance their chances of admission to universities that prioritize these metrics in their selection process.
The correlation between a university's average cost and the increasing odds of women's enrollment in engineering highlights a historical connection between engineering and class This relationship may help clarify the findings of Orr, Ramirez, and Ohland (2011), which indicate that high school socioeconomic status significantly predicts engineering enrollment for both genders, but with a greater effect size for men Consequently, women's higher enrollment in more expensive engineering programs suggests that bourgeois class status continues to influence their entry into the field, potentially diminishing the importance of peer socioeconomic status for women compared to men.
The relationship between a university's admissions rate and its total cost suggests that more expensive institutions may be more selective A plot of these variables indicates an initial collinear relationship; however, this diminishes significantly when admissions rates exceed 25% A collinearity diagnostic conducted in SAS confirmed weak collinearity, with condition indices greater than 10, as outlined by Belsley et al This finding highlights that university cost is a crucial predictor of women's enrollment in engineering Together, these variables account for a substantial portion of the variance in women's enrollment, yielding an adjusted \( r^2 \) of 0.40.
Figure 3 Plot of university total cost versus admissions rate showing no relationship (U.S Department of Education Institute of Education Sciences, 2016)
Table 9 Collinearity Diagnostic Between Admissions Rate and Total Price A condition index below 10 indicates low collinearity
Intercept Admission rate Net cost Intercept 2.74942 1.00000 0.00576 0.01186 0.01574 Admission rate 0.22200 3.51919 0.00061877 0.20442 0.33362 Net cost 0.02858 9.80813 0.99362 0.78371 0.65065
Conclusion
How do institutional characteristics affect the odds of women’s enrollment in engineering programs compared to men’s?
The findings indicate that more expensive and selective institutions are associated with higher odds of women pursuing engineering, supporting the idea that women's current decisions to enter the field are influenced by historical structures that have evolved but still bear resemblance to those that facilitated women's entry into engineering over a century ago This may not be surprising if one considers the existence of a gender queue in engineering, which could lead many women to adopt similar pathways influenced by class affiliation Furthermore, this pathway is shaped not by personal choice, but by the necessity to acquire the capital needed to enter the profession.
The study highlights that factors like cost and selectivity significantly influence women's enrollment in engineering programs compared to men However, not all universities can leverage these characteristics to attract more women Instead, universities should adopt broader strategies that convey the potential for high-status employment with autonomy and power in engineering fields For institutions with lower costs or higher admission rates, forming strategic partnerships with employers offering such positions can effectively increase female enrollment Communicating these opportunities to prospective students can demonstrate how their education may lead to desirable job placements.
To attract prospective female students inclined towards selective schools, institutions should develop targeted in-house programs, similar to existing honors engineering initiatives A strategic next step would involve engineering programs forming partnerships with non-engineering departments that have a strong female presence, such as psychology By enhancing psychology programs with engineering education, schools can create valuable interdisciplinary opportunities This approach aligns with industry trends, as companies like Google and Facebook actively seek psychology majors for their unique contributions to marketing and advertising efforts.
A candidate with a dual background in psychology and engineering is often viewed as more competitive due to their unique blend of social and technological expertise, which meets employer demands for a well-rounded skill set Engineering programs can enhance their appeal by learning from other disciplines that have successfully attracted women, thereby increasing diversity and providing the valuable opportunities that women seek in the job market.
This paper presents a counter theory to the notion that women's lack of participation in engineering is due to a perceived deficit of choice Instead, it theorizes that women's entry into engineering is influenced by structural factors, supported by historical continuity The analysis reveals that women's participation is significantly shaped by class-coded structures, and it suggests potential interventions that leverage these structures to enhance women's enrollment in engineering Ultimately, this work provides a clear understanding of how gender and class structures impact women's participation in engineering education and identifies strategies to promote gender parity within the patriarchal frameworks that dominate the field.