Download PDF Fall Bridge on Engineering a Diverse Future September 25, 2024 Volume 54 Issue 3 Guest edited by Wanda Sigur and Percy Pierre, this issue of The Bridge addresses the issues around sustaining a U.S. engineering workforce that builds on and integrates the talents and ideas of our diverse nation. Does the Minority Engineering Effort Have a Flat Tire? Thursday, September 26, 2024 Author: Catherine J. Weinberger The minority engineering effort led to an increase in the representation of African Americans in engineering, but future progress depends on renewed efforts to welcome new cohorts of young engineers into the profession. College graduates who major in engineering consistently have strong earnings potential relative to college graduates with other majors (Andrews et al. 2022; Brown and Corcoran 1997; Daymont and Andrisani 1984; Grogger and Eide 1995; Polachek 1978; Weinberger 1998, 2018). The preceding article describes historical details and early achievements after a national organization of engineers formed in the early 1970s utilized a systems engineering approach to design and implement policies that would facilitate the participation of African American students in accredited engineering bachelor’s degree programs (Pierre and Weinberger 2024). Despite the initial success of the minority engineering effort (MEE), with the annual number of African American engineering graduates rising from 400 in 1970 to 1,400 in 1980 and 3,000 in 1995, all indications are that this effort ran out of steam about 20 years ago. However, previous analysis of trends does not provide a complete picture. Here, I attempt to address some of the methodological issues and to provide accessible graphs that describe recent trends in different ways that complement existing analyses to provide a more complete picture of historical and emerging trends; all of these lead to similar conclusions. Pursuing a deeper understanding of the extent to which members of historically excluded groups are facing renewed barriers to entering engineering careers should be of concern to the profession. It is always difficult to identify the permanence of a trend in progress. Statistics collected annually by the US Department of Education and published periodically by the American Society for Engineering Education[1] now describe a decade of unchanging representation of African American citizens, and the simultaneous doubling of the representation of Hispanic citizens, among new engineering graduates who were not designated foreign nationals. Using the same federal counts of degrees conferred spanning a longer interval (1981–2015), journalist Ashley Smart documented two decades of growth, then a monotonic decline in the representation of African American scientists and engineers among new college graduates between 2004 and 2011, followed by flatlining; this reporter linked the change to the dismantling of affirmative action programs that had been designed to facilitate the recruitment and retention of minority students (Smart 2020). In this article, I provide some historical context and describe different ways to think about what those statistics might mean because analyses based on samples of engineering graduates, or on samples of bachelor’s degree graduates, conflate changes in the demographic composition of the birth cohort with changes in the propensity to attend college, major in engineering, and complete college among individual cohort members. A clearer picture of true trends emerges when the propensity to complete an engineering degree is computed for each complete birth cohort. Pursuing a deeper understanding of the extent to which members of historically excluded groups are facing renewed barriers to entering engineering careers should be of concern to the profession. To put these statistics into context, it is helpful to describe the timing of different stages of the MEE, explained in greater detail in the preceding article (Pierre and Weinberger 2024) and “Engineering Educational Opportunity: Impacts of 1970s and 1980s Policies to Increase the Share of Black College Graduates with Major in Engineering or Computer Science” (Weinberger 2018). During the 1960s, about 300 African American engineering students graduated per year, at least half of those from one of six historically Black campuses. The national effort to increase the number of African American engineers began in the early 1970s, with active participation by a handful of campuses representing about 20% of engineering graduates nationally, and the founding of both national and campus-specific programs dedicated to the success of minority engineering students with support from industry, the federal government, and the Sloan Foundation. A particular milestone of the early effort is the publication of A Blueprint for Action (PCEMOE 1974). This effort arose in response to Nixon administration affirmative action policies after GE and other employers realized it would be challenging, if not impossible, to hire significant numbers of African American engineers. In 1980, with the passage by Congress of the Science and Technology Equal Opportunities Act, the federal government committed to ensuring access to science and engineering education for all US residents and began to collect detailed statistics on the race, gender, and college major of all new graduates. The NAE-sponsored 1973 symposium that marked the beginning of the MEE focused on African American men, but the issue of broadening the effort to include other minorities began during that first gathering and continued at the first meeting of the new NAE Committee on Minorities in Engineering (Grosch 1974; NAE 1973). At that time, fewer than 1% of new engineering graduates were women nationwide, compared to 2.5% at the six HBCU engineering programs. As the MEE entered the 1980s, the inclusion of minority women became the norm. By the mid-1990s, 15% of new engineering graduates were women, but 30% of new African American engineering graduates were women nationwide, and, among African American HBCU engineering graduates, 40% were women (Weinberger 2018). I’ll begin by providing a longer view of changes between 1940 and 2022 in the demographic composition of US children, the educational attainment of their parents, the propensity to complete Algebra II before high school graduation, and the propensity for young adults to hold a college degree, followed by a careful analysis of the propensity for successive cohorts to hold a bachelor’s degree in engineering. The composition of the US-born-and-educated population has shifted over the years, though not as dramatically as the current public perception would suggest. Figure 1 depicts changes among children in the 1–10 age range since 1940. At the outset of the MEE, less than 15% of US children approaching college age were identified by their parents as Black or African American, and this share has grown no higher than 18%. Due to gradual immigration, the share of US-born children identified in the broad “Hispanic” category grew from less than 4% to just under 17% by 2010 and then flattened out over the past decade; a majority of these children resided with a parent who was also born in the US in nearly every decade since 1940 (with the single exception of 2000–2009). Because of the growth of this portion of the population, it is understandable that the number of Hispanic engineers has grown more quickly than the number of African American engineers. But this does not explain why both groups continue to be underrepresented in engineering. Another shift in the background of new cohorts of children is a large increase in parents’ educational attainment. The share of US-born children (ages 1–10) residing with a college-educated parent grew from 13% in 1960 to 45% in 2022. Among African American children, the corresponding statistics are 3% in 1960, rising to 30% in 2022; for Hispanic children, 3% in 1960, rising to 22% in 2022. Over the same period, the propensity for high school graduates to complete a course in Algebra II nearly tripled, reaching 85% by 2019 and surpassing 80% for African American and Hispanic students; this is a dramatic change from the historic restriction of access to high school algebra to a select few (Jones and Coxford 1970; NCES 2021; Weinberger 2014). Shifts in relative numbers, parents’ educational attainment, and equity of access to high school mathematics preparation have all contributed to an increase in the representation of African American and Hispanic graduates within the college-educated workforce. However, nearly universal enrollment in Algebra II does not mean that K–12 equity has been achieved. Nationwide, students from advantaged families are increasingly likely to complete a course in calculus before leaving high school, and to begin college with an apparent head start in securing scarce seats in competitive programs (Bressoud 2021). Here, a new racial gap has emerged. In 1982 only 5% of new high school graduates had taken a year of calculus.[2] By 2019 the share had grown to 15%, but to only 6% among African American students.[3] Figure 2 describes the timing of increased educational opportunity within and between birth cohorts. The average propensity to hold a college degree is plotted against the average year in which a cohort reached age 22. For example, a cohort that was 22–26 in 1980 would have turned 22 in 1976–1980 and is plotted at 1978, and the cohort that was 27–31 in in 1980 is plotted five years earlier. The slopes describe how quickly access to college expanded over time, while the vertical distance between the younger and older observations represents the extent to which individuals continued to complete college degrees between the ages of 27 and 31. The growing distance between the younger and older observations of successive cohorts illustrates that, over time, a growing share of eventual college graduates had not yet completed their degrees within the 22–26 age range. This means that measuring the college completion rate of a cohort requires careful thought about the age at which the measurement is taken. However, those who complete college later in life are less likely to major in engineering. When analyzing current trends, and especially when computing the ratio of engineering graduates to all new graduates, understanding changes in the demographic characteristics of successive waves of graduates requires information about the age distribution of those new graduates as well as their race or ethnicity. The final curve plotted in the lower right-hand corner of figure 2 illustrates the tiny share of college graduates who are engineering majors. Large representative samples of the US population that include information about engineering college graduates of all ages were collected only twice before 2009. Fortunately, for each of the years from 2009–2022, the Census Bureau includes a question about college major in its annual American Community Survey of 1% of the US population; this includes millions of observations and hundreds of thousands of individuals with bachelor’s degrees in engineering (Ruggles et al. 2024). For this reason, the following figures will use only the data from 2009–2022 (excluding error-prone data from pandemic year 2020). Despite its recent collection, the attainment of bachelor’s degrees in engineering among both young and old individuals who were born (and predominantly educated) in the United States can be used to estimate the prevalence of engineering degree attainment among earlier cohorts using a retrospective approach that infers the educational opportunities available at a given point in time by looking at the educational attainment reported by those who were college age at that time. In figure 2, only the cohorts age 22–31 are depicted, but the remainder of the figures examine cohorts who reached age 22 in 1955–59, 1960–64, 1965–69, and so on. This approach provides a clear picture of the achievements that followed the organization of the MEE. Three alternative ways to describe changes over time are presented in figures 3–5. Figure 3 begins with entire birth cohorts—every individual born in the United States in a particular five-year interval—rather than selecting only engineering graduates for analysis. This presentation provides the clearest picture of trends in opportunities to study engineering for successive cohorts old enough to have completed their education. Figure 4 shows changes over time within the smaller population that completed a bachelor’s degree. Figures 5a and 5b use the same samples as figures 3 and 4 to examine how trends in engineering participation compare with trends in computer science and various subsets of STEM fields. Figure 3 describes trends in the propensity for members of successive cohorts to hold a bachelor’s degree in engineering; the top curve is unconditional, and the remaining curves condition on demographic characteristics. As described in the preceding article (Pierre and Weinberger 2024), the MEE that began in the early 1970s led to an acceleration of the already slowly rising share of African American college graduates with engineering degrees. This presentation clarifies that the propensity of African Americans to become engineers increased throughout the 1960–1995 period, with a substantial increase in slope in the late 1970s that corresponds almost exactly with the cohort of students who enrolled in college the year the engineers of the MEE published the Blueprint report (and reached graduation age four or five years later). Previous research found that the magnitude and timing of growth in the number of new African American engineering graduates varied geographically, and were related to the educational infrastructure in place in the birth state as of 1965 (Weinberger 2018). It is notable that the corresponding curves for Hispanic and female members of the same cohorts show very different patterns of change; both of these groups have a steep upward slope between the early 1970s and the early 1980s in contrast with the later change among African American engineers. The programs and policies developed during this time had a targeted impact on the number of African American engineers, both men and women, and appear to have addressed barriers specific to African American students. The results of these efforts complemented the effects of contemporaneous federal policies that applied to all women and minorities and of the contemporaneous Cold War expansion of engineering opportunities that applied to everyone. Figure 3 also shows a decline in the share of the general population electing to study engineering throughout the 1990s and a slowing or stalled rate of growth among the African American population. While a decline appears to have affected the youngest cohorts, this is partially driven by the fact that some of the data come from younger individuals who have not yet completed their education. An additional set of estimates, indicated by dashed line segments, describe the means among cohort members who are at least as old as the youngest member of the preceding cohort; this set of less-negative or positive slopes underscores the role of age in producing the apparent decline. This blunt adjustment for differences in age suggests that the development of a statistical model that accounts for age at observation is needed to interpret the true trajectory of the propensity to study engineering. Additional analysis of the incomplete trajectories of the youngest sample members (not shown) suggests that the average propensity to attain an engineering degree is now on the rise for all groups, but that the rate of growth is similar among minority engineers, with relative participation rates that are no longer declining. The analysis described in figure 3 is well suited for understanding impacts on educational opportunity because there is no reason to believe that the members of a particular birth cohort have different characteristics than observationally similar individuals born five years earlier or later. However, a limitation of this presentation of the data is that it is not clear to what extent the remaining differences between groups are due to differences in college-going vs. college major selection among enrolled students. To begin to address this issue, figure 4 provides answers to two questions: How does the share of college graduates who are engineering majors vary by race or ethnicity, and how does this change over successive birth cohorts? While figure 3 describes trends in the propensity of African Americans to hold a bachelor’s degree in engineering (within the entire cohort, college-educated or not), figure 4 describes trends in relative propensities within the subset of the cohort that completed a bachelor’s degree, where relative propensity is defined as the propensity of African American or Hispanic college graduates to hold an engineering degree divided by the propensity of all US college graduates in the same age range to hold an engineering degree. This presentation might be of less use to historians or economists interested in the net impacts of prior policy changes, but of more relevance to university administrators interested in understanding the extent to which remaining differentials are related to college access rather than opportunities to study engineering once enrolled. [4] As in figure 3, the alternative measure depicted in figure 4 begins to slope downward near the turn of the century, and much more dramatically, but it is difficult to discern the extent to which this is due to changes in the number and composition of college graduates with other majors rather than a true decline. (Figure 5a will address this concern.) In any event, this presentation is closer to previous analyses that describe the racial composition of recent engineering graduates based on administrative counts of new degrees, and it is reassuring to find similar trends in retrospective data. Figure 5 provides additional context by comparing the African American trends in engineering degree completion depicted in figures 3 (5a) and 4 (5b) with other college majors. While African American undergraduate students remained substantially underrepresented in engineering throughout the 50-year history of the MEE, they faced far less disparity in the emerging field of computer science and—among college graduates—surpassed national averages for a sustained period of time (Weinberger 2018). Additional curves included in these figures clarify that the gains in engineering participation throughout the 1970–1995 period led to net gains in the representation of African American graduates in science or engineering fields, and that the apparent flattening or decline after 1995 has been shared with other STEM fields. Figures 5a and 5b also provide a side-by-side comparison of estimated trends in the relative propensity to complete an engineering degree both within the full population and within the subset of the population that holds a college degree. This presentation clarifies that the downturn that began near the turn of the century is robust, and not an artifact of changing characteristics of college graduates. However, among the youngest cohorts, restricting to college graduates leads to the appearance of continued decline, while the true trend in the full population seems to have flattened with an inkling of an upturn. Many members of these youngest cohorts have not yet completed their education, and only time will tell what will happen next. The minority engineering effort has had lasting effects on the US workforce. All three of these complementary approaches point to the conclusion that the MEE, and the federal and local policies that followed, successfully led to an increase in the share of African American citizens with bachelor’s degrees in engineering that reached its peak near the turn of the century. Figure 3 indicates that the remarkable growth between 1975 and 1999 in the propensity of African American youth to become engineers has not continued and did not come close to attaining equality. The following figures confirm that relative propensities among younger cohorts also began to decline at that time and appear to be flattening out near the level previously observed in the late 1980s. Where we will go from here is a story yet to be determined. What might be done to redirect the trend back onto an upward trajectory? Research into college major choices points toward social and psychological roadblocks facing even academically well-prepared students from underrepresented groups (Margolis and Fisher 2002; Seymour 1992; Seymour and Hewitt 1997; Seymour et al. 2019). The success of HBCU programs has been attributed to their historical mission to teach underserved students, beginning with emancipated slaves who had previously been forbidden to read, and to an academic environment that provides encouragement and an expectation of success (Nettles 1988; Slaughter 2009; Trent and Hill 1994). These programs continue to prepare a disproportionate share of African American engineers for successful and lucrative careers (Mountjoy and Hickman 2021; Weinberger 2018). Students in settings where their identity sets them apart often report a sense of isolation from their classmates (Bix 2013; Landis 1991, 2005; Weinberger 2018). This can be mitigated by creating a sense of community within a classroom, or via extracurricular support structures. In psychology, a controlled experiment that included a short (one-hour) intervention to foster a sense of belonging and understanding that adversity can be overcome with persistence was completed in the spring of freshman year at Stanford. This brief experience had large, cumulative impacts on the academic attainment of African American students over each of the following semesters, with longer-lasting impacts on health three years later and career satisfaction and success, psychological well-being, community involvement, and leadership 7–10 years later (Brady et al. 2020; Walton and Cohen 2011). This approach was moved into the classroom with a similar one-time intervention—that also incorporates the ideas of Brown (1997)—assigned to two of the five discussion sections led by each of three teaching assistants in a physics course for engineers. Exposure to the intervention both built “a community of learners” and improved academic outcomes and persistence in the major for everyone, with particularly large gains for underrepresented students (Binning et al. 2020). Whether or not the trend in relative rates of engineering participation reverses, ensuring that members of the profession feel capable and valued is a worthy goal in and of itself. Students less well prepared for college coursework face additional bureaucratic barriers, even if they have the potential to attain a steep upward trajectory during their college years. Data-rich research into the economics of oversubscribed college majors in large public universities documents that students coming from under-resourced high schools suffer disproportionate losses in both opportunities to pursue their chosen academic trajectories and later earnings when freshman-year GPA is used to ration admission to a major (Bleemer and Mehta 2022). And a recent study compared thousands of students offered a one-time workshop and short-term mentoring to otherwise comparable students whose GPAs were barely too high to qualify for the program. This study provides large-scale causal evidence that a small amount of encouragement—supported by credible evidence that initial academic challenges can be overcome―can have persistent impacts on both academic and career trajectories over 7–10 years (Canaan et al. 2024). All of these studies point to solutions that can be implemented at the campus level. The MEE has had lasting effects on the US workforce. Within the entire population of employed individuals holding a bachelor’s degree in engineering (including many with higher degrees as well), 2009–2022 saw no change over time in the 70% share who either were born in the United States or arrived before age 17. Over the same decade, the representation of US-born-and-educated African American engineers increased by 30%, from 3.0-3.9% of all employed engineers (or, from 4.3-5.6% of employed engineers born in the United States). This growth reflects the groundwork laid by the early visionaries of the MEE, with their focus on African American representation in the engineering profession. However, the majority of the change observed over the past decade is driven by cohorts educated more than 20 years ago; without a renewed effort, the overall rate of increase—while still positive—can be expected to begin falling as those currently in their mid-40s reach retirement age, unless new cohorts of young engineers begin to narrow the gap in relative attainment rates once again. References Andrews RJ, Imberman SA, Lovenheim MF, Stange KM. 2022. The returns to college major choice: Average and distributional effects, career trajectories, and earnings variability, revised March 2024. NBER Working Paper #30331. Online at www.nber.org/papers/w30331. Binning KR, Kaufmann N, McGreevy EM, Fotuhi O, Chen S, Marshman E, Kalender ZY, Limeri L, Betancur L, Singh C. 2020. Changing social contexts to foster equity in college science courses: An ecological-belonging intervention. Psychological Science 31(9):1059–70. Bix A. 2013. Girls Coming to Tech!: A History of American Engineering Education for Women. Cambridge, Massachusetts: MIT Press. Bleemer Z, Mehta A. 2022. Will studying economics make you rich? A regression discontinuity analysis of the returns to college major. American Economic Journal: Applied Economics 14(2): 1–22. Brady ST, Cohen GL, Jarvis SN, Walton GM. 2020. A brief social-belonging intervention in college improves adult outcomes for Black Americans. Science Advances 6(18):eaay3689. Bressoud DM. 2021. The strange role of calculus in the United States. ZDM 53(3):521–33. Online at https://doi.org/10.1007/s11858-020-01188-0 Brown C, Corcoran M. 1997. Sex-based differences in school content and the male-female wage gap. Journal of Labor Economics 15(3):431–65. Canaan S, Fischer S, Pierre M, Schnorr G. 2024. Keep me in, coach: The short- and long-term effects of targeted academic coaching, working paper online at scanaan.github.io/papers/coaching_v9.pdf. Daymont TN, Andrisani PJ. 1984. Job preferences, college major, and the gender gap in earnings. Journal of Human Resources 19(3):408–28. Grogger J, Eide E. 1995. Changes in college skills and the rise in the college wage premium. Journal of Human Resources 30(2):280–310. Jones P, Coxford AF Jr. 1970. Mathematics in the Evolving Schools. In: A History of Mathematics Education in the United States and Canada, thirty-second year book, 38. Reston, Virginia: National Council of Teachers of Mathematics. Landis RB. 1991. Retention by design: Achieving excellence in minority engineering education. National Action Council for Minorities in Engineering. Landis RB. 2005. Retention by Design, Achieving Excellence in Minority Engineering Education. College of Engineering, Computer Science and Technology, California State University, Los Angeles. Margolis J, Fisher A. 2002. Unlocking the Clubhouse: Women in Computing. Cambridge, Massachusetts: MIT Press. Mountjoy J, Hickman BR. 2021. The returns to college(s): Relative value-added and match effects in higher education. Becker Friedman Institute. NBER Working paper 29276. Nettles M. 1988. Toward Black Undergraduate Student Equality in American Higher Education. Westport, Connecticut: Greenwood Press. Pierre PA, Weinberger CJ. 2024. The 50-year history of the minority engineering effort: How the engineering profession sparked the movement to diversify its workforce. The Bridge 54(4):23-30. PCEMOE [Planning Commission for Expanding Minority Opportunities in Engineering]. 1974. Minorities in Engineering: A Blueprint for Action. New York: Alfred P. Sloan Foundation. Polachek SW. 1978. Sex differences in college major. ILR Review 31(4):498–508. Ruggles S, Flood S, Sobek M, Backman D, Chen A, Cooper G, Richards S, Rodgers R, Schouweiler M. 2024. IPUMS USA: Version 15.0 [dataset]. Minneapolis, Minnesota: IPUMS. Online at https://doi.org/10.18128/D010.V15.0. Seymour E. 1992. ‘The problem iceberg’ in science, mathematics, and engineering education: Student explanations for high attrition rates. Journal of College Science Teaching 21(4):230–38. Seymour E, Hewitt NM. 1997. Talking about Leaving: Why Undergraduates Leave the Sciences. Boulder, Colorado: Westview Press. Seymour E, Hunter A-B, Weston TJ. 2019. Why we are still talking about leaving. In: Talking About Leaving Revisited: Persistence, Relocation, and Loss in Undergraduate STEM Education, 1–53. Seymour E, Hunter A-B, eds. Springer. Slaughter JB. 2009. African American males in engineering: Past, present and a future of opportunity. In: Black American Males in Higher Education: Research, Programs and Academe volume 7, 193–208. Frierson H, Wyche J, Pearson W, eds. Emerald Publishing Limited. Smart A. 2020. After years of gains, Black STEM representation is falling. Why? Undark, Nov 11. Trent W, Hill J. 1994. Contributions of historically Black colleges and universities to the production of African American scientists and engineers. In: Who will Do Science? Educating the Next Generation, 68–80. Pearson Jr. W, Fechter A, eds. Baltimore: Johns Hopkins University Press. Walton GM, Cohen GL. 2011. A brief social-belonging intervention improves academic and health outcomes of minority students. Science 331(6023):1447–51. Weinberger CJ. 1998. Race and gender wage gaps in the market for recent college graduates. Industrial Relations 37(1):67–84. Weinberger CJ. 2014. Are there racial gaps in high school leadership opportunities? Do academics matter more? The Review of Black Political Economy 41(4):393–409. Weinberger CJ. 2018. Engineering educational opportunity: Impacts of 1970s and 1980s policies to increase the share of Black college graduates with major in engineering or computer science. In: U.S. Engineering in a Global Economy, 87–128. Freeman R, Salzman H, eds. The University of Chicago Press. Online at www.nber.org/papers/w23703.’’ [1] See ira.asee.org/by-the-numbers/. [2] nces.ed.gov/programs/digest/d19/tables/dt19_225.30. asp, accessed 7/30/24 [3] nces.ed.gov/programs/digest/d22/tables/dt22_225.30. asp, accessed 7/30/24 [4] For deans of engineering, and to provide a deeper understanding of differentials in access, it might be even more helpful to create a version of this figure specific to the subset of campuses containing an engineering program, but this analysis could not be completed within the time frame of the current publication. About the Author:Catherine J. Weinberger is a research associate at the University of California, Santa Barbara, Institute for Social, Behavioral and Economic Research, affiliated with the Broom Center for Demography and the Department of Economics.