Chapter 2: Higher Education in Science and Engineering

Higher Education Enrollment in the United States

Recent higher education enrollments reflect the expanding U.S. college-age population. This section examines trends in undergraduate and graduate enrollment by type of institution, field, and demographic characteristics. It also examines graduate financial support patterns and data on retention rates. For information on enrollment rates of high school seniors, see "Transition to Higher Education" in chapter 1.

Overall Enrollment

Over the past two decades, enrollment in U.S. institutions of higher education rose fairly steadily, from 12.6 million students in 1983 to 15.7 million in 2001 (the last year of available data), despite declines in the college-age population during much of that period (appendix tables 2-3 Excel table. and 2-36 Excel table.). Of these, more than 6 million students (about 38% of all students enrolled in higher education institutions in the United States) were enrolled in 2-year institutions in 2001. The next two largest segments, research I universities and master's-granting I (or comprehensive) universities, together accounted for another 34% (5.3 million). (See sidebar "Carnegie Classification of Academic Institutions" for definitions of the types of academic institutions.)

Enrollment in higher education is projected to increase in the next decade because of increases in the college-age population. According to U.S. Census Bureau projections, the number of college-age (ages 20–24) individuals is expected to grow from 18.5 million in 2000 to 21.7 million by 2015, then decrease slightly to 21.0 million by 2020 (figure 2-2 figure.and appendix table 2-4 Excel table.).

Increased enrollment in higher education is projected to come from minority groups, particularly Asians/Pacific Islanders and Hispanics. From 2000 to 2015, the Asian/Pacific Islander and Hispanic college-age populations are projected to increase by more than 50%, while the black and American Indian/Alaska Native college-age populations are projected to rise by 19% and 15%, respectively. The white college-age population is projected to increase slightly through 2010 and then decline (figure 2-2 figure.).

Changes in the demographic composition of the college-age population as a whole and increased enrollment rates of some racial/ethnic groups have contributed to changes in the demographic composition of the higher education student population in the United States.[1] From 1992 to 2001, overall enrollment increased by 7%, while underrepresented minority enrollment grew by 31% and Asian/Pacific Islander enrollment by 32%. Enrollment of foreign students (i.e., students on temporary visas) grew by 25% during that period. Almost half (47%) of underrepresented minority students and 42% of Asian/Pacific Islander students were enrolled in 2-year institutions compared with 36% of white students and 18% of foreign students.[2] Underrepresented minority students were less likely than other groups to be enrolled in research institutions (10% versus 20% of whites, 26% of Asians/Pacific Islanders, and 39% of foreign students.) For a breakout of enrollment trends in the 1990s by institutional type, race/ethnicity, and citizenship, see appendix table 2-5 Excel table..

Top of page.Top of page

Undergraduate Enrollment in S&E

Freshmen Intentions to Major in S&E

Since 1972, the annual Survey of the American Freshman, National Norms, administered by the Higher Education Research Institute at the University of California at Los Angeles, has asked freshmen at a large number of universities and colleges about their intended majors. The data have provided a broadly accurate picture of degree fields several years later.[3] For at least the past two decades, about one-third of all freshmen planned to study S&E. In 2004, about one-third of white, black, Hispanic, and American Indian/Alaska Native freshmen and 46% of Asian/Pacific Islander freshmen reported that they intended to major in S&E (figure 2-3 figure.). The proportions planning to major in S&E were higher for men in every racial/ethnic group (appendix table 2-6 Excel table.). For most racial/ethnic groups, about 9%–14% planned to major in social/behavioral sciences, about 9% in engineering, about 8% in biological/agricultural sciences, 2%–5% in computer sciences, 2% in physical sciences, and 1% in mathematics or statistics. Higher proportions of Asian/Pacific Islander freshmen than of those from other racial/ethnic groups planned to major in biological/agricultural sciences (16%) and engineering (15%).

The demographic composition of students planning S&E majors has become more diverse over time. Women constituted 38% in 1983, rising to 45% in 2004. White students declined from 85% in 1983 to 72% in 2004. On the other hand, Asian/Pacific Islander students increased from 3% to 12%, Hispanic students increased from 1% to 8%, and American Indian/Alaska Native students increased from 1% to 2% (appendix table 2-7 Excel table.). Black students made up 10% of freshmen intending to major in S&E both in 1983 and in 2004.

In 2002, 20% of the respondents planning an S&E major reported needing remedial work in mathematics, and nearly 10% reported needing remediation in the sciences. These percentages are slightly higher than those of 1984 and vary by field and sex (appendix table 2-8 Excel table.). Fewer of those intending to major in mathematics, computer sciences, physical sciences, or engineering reported a need for remediation than did those intending to major in the social/behavioral or biological/agricultural sciences or in non-S&E fields. Within S&E fields and non-S&E fields, proportionately fewer male freshmen than female freshmen reported a need for remediation in mathematics or sciences. See "Transition to Higher Education" in chapter 1 for additional information on need of freshmen for remedial education.

Foreign Undergraduate Enrollment

The number of foreign undergraduates enrolled in U.S. academic institutions in all fields decreased almost 5% from academic year 2002–03 to 2003–04, the second consecutive decline after record increases during the 1990s. Decreases in foreign enrollments in recent years have been attributed to increased opportunity for higher education in the home country, competition from other countries for foreign students, rising U.S. tuition, and difficulties in obtaining U.S. visas (Institute of International Education 2004). (See sidebar "Price of Undergraduate Education.") Declines in particular fields may also be due to declining job opportunities in those fields. Among both undergraduate and graduate students, the number of foreign engineering students dropped 1% and the number of foreign computer sciences students dropped 7%. Other S&E fields, particularly the social sciences at 18%, experienced increases. Both physical and life sciences each registered small increases of 1%, and agricultural sciences rose 0.3%.

Japan and South Korea accounted for the largest numbers of foreign undergraduates in the United States in 2004 (appendix table 2-9 Excel table.). Although the number of undergraduates from China, Japan, and Taiwan was lower in 2004 than in 1999, enrollment of students from a number of other countries, including South Korea, Canada, India, and Mexico, increased.

Enrollment Trends in Engineering

For the most part, undergraduate enrollment data are not available by field. However, because engineering programs generally require students to declare a major in the first year of college, engineering enrollment data can serve as early indicators of both future undergraduate engineering degrees and student interest in an engineering career. The Engineering Workforce Commission administers an annual fall survey that tracks enrollment in undergraduate and graduate engineering programs (Engineering Workforce Commission 2004).

Undergraduate engineering enrollment declined through most of the 1980s and 1990s, then rose from 2000 through 2003.[4] From a 1983 peak of approximately 441,000 students, undergraduate engineering enrollment declined to about 361,000 students by 1999 before rebounding to about 422,000 in 2003 (figure 2-4 figure.; appendix table 2-10 Excel table.). Graduate engineering enrollment rose to about 129,000 in 1992, declined to approximately 105,000 by 1999, then soared to nearly 148,000 by 2003 (appendix table 2-11 Excel table.).

Retention in S&E

The National Science Foundation (NSF) National Survey of Recent College Graduates tracks retention in S&E as measured through further S&E education and entry into S&E occupations. About 28% of those who graduated with an S&E bachelor's degree in 2001 or 2002 were continuing in S&E graduate study (12%) or S&E employment (16%) in 2003. Retention rates in S&E declined from the 2001 and 1995 surveys (appendix table 2-12 Excel table.). However, many of those going into non-S&E occupations found employment in occupations with strong S&E components (see "U.S. S&E Labor Force Profile" in chapter 3).

Percentages of those going on for advanced study in S&E were higher for those with a high grade point average (GPA). About 18% of those with a 3.75–4.00 undergraduate GPA continued to study S&E. In contrast, relatively few (7%) of students with less than a 2.75 GPA continued to study S&E.

The retention rate in S&E after completion of a master's degree was higher than the rate after completion of a bachelor's degree. Around 44% of those who earned an S&E master's degree in 2001 or 2002 were continuing in S&E in 2003, either in school (15%) or in employment (29%). Overall, the S&E retention rate after a master's degree in 2003 was lower than the rate in either 1995 or 2001, with both a smaller percentage continuing advanced studies and a smaller percentage employed in S&E fields (appendix table 2-12 Excel table.).

Top of page.Top of page

Graduate Enrollment in S&E

Graduate S&E educational institutions are a major source of both the high-skilled workers of the future and of the research needed for a knowledge-based economy. This section presents data on continuing key trends in graduate S&E enrollment, including trends in first-time enrollment of foreign students after September 11, 2001. Information is also included on patterns and trends in financial support for graduate education and in student debt.

Enrollment by Field

S&E graduate enrollment in the United States reached a new peak of 566,800 in fall 2003. Following a long period of growth that began in the 1970s, graduate enrollment in S&E declined in the latter half of the 1990s before rebounding in the past several years. Graduate enrollment in engineering and in life sciences drove most of the recent growth, although enrollment did increase in almost all major science fields.[5] Computer sciences enrollment rose rapidly from the mid-1990s through 2002 but declined in 2003. The increase in computer science through 2002 and the continuing increase in engineering mainly reflect an increase in the number of foreign graduate students in those fields (figure 2-5 figure.).

The number of full-time students enrolled for the first time in S&E graduate departments offers a good indicator of developing trends. It declined in the mid-1990s in all major S&E fields but increased in most fields in the late 1990s and early 2000s (appendix table 2-13 Excel table.). Between 2000 and 2003, first-time full-time S&E enrollment grew 14%. Growth was greatest in physical sciences; earth, atmospheric, and ocean sciences; mathematics; and social and behavioral sciences. In only a few fields, such as computer sciences (down 3%) and materials engineering (down 5%), did first-time full-time graduate enrollment decline during the period.

Enrollment by Sex and Race/Ethnicity

The recent increase in S&E graduate enrollment overall occurred across all major demographic groups: women, minorities, and white men. The number of women enrolling in all S&E graduate programs has increased for the past two decades except for a decline in computer sciences in 2003. In contrast, the number of male S&E graduate students declined from 1993 through the end of the decade before increasing in recent years (appendix table 2-14 Excel table.).

The long-term trend of women's rising proportions in S&E fields also continued. Women made up 36% of S&E graduate students in 1983 and 47% in 2003, although large variations among fields persist. In 2003, women constituted the majority of graduate enrollment in psychology (74%), medical/other life sciences (76%), biological sciences (55%), and social sciences (53%). They constituted considerable proportions of graduate students in mathematics (37%), chemistry (39%), and earth, ocean, and atmospheric sciences (45%). Their percentage in computer sciences (28%) and engineering (22%) remains smaller (figure 2-6 figure.; appendix table 2-14 Excel table.).

The proportion of underrepresented minority students in graduate S&E programs increased from about 6% in 1983 to about 11% in 2003 (table 2-2 table.). Increases occurred in all major science fields between 1983 and 2003. Only in engineering have enrollment increases apparently stalled in recent years: underrepresented minorities have been 6% of graduate enrollment since 1995 (appendix table 2-15 Excel table.). In 2003, blacks, Hispanics, and American Indians/Alaska Natives as a group made up about 6% of graduate enrollment in most S&E fields (engineering; mathematics; physical sciences; earth, atmospheric, and ocean sciences; and computer sciences), 17% in social sciences, and 19% in psychology.

The number of white S&E graduate students decreased from 1994 to 2001 in most S&E fields, then increased through 2003, whereas the number of underrepresented minority students has increased every year since 1985. The long-term rise in the number of underrepresented minority graduate students occurred in most S&E fields, with the exceptions of engineering and mathematics. In those two fields, underrepresented minority enrollment plateaued in the 1990s before rising again from 2000 through 2003. The number of Asian/Pacific Islander S&E graduate students increased every year since 1983, with the exception of 2000. Increases occurred in most science fields except for a drop in physical sciences and engineering enrollment in the 1990s. Asians/Pacific Islanders accounted for about 7% of S&E graduate enrollment in 2003 (appendix table 2-15 Excel table.).

Foreign Student Enrollment

Foreign graduate student enrollment in S&E grew from 73,200 in 1983 to 154,400 in 2003. For all S&E fields combined, the proportion of foreign students increased from 19% to 27% over the period (appendix table 2-15 Excel table.). Foreign enrollment was highest in engineering (47%), computer sciences (45%), physical sciences (41%), and mathematical sciences (38%).

First-time full-time enrollment of foreign S&E graduate students offers a mixed picture. It declined 5% in fall 2002, the first full academic year since September 11, 2001. Declines continued in fall 2003 (an 8% decrease in S&E overall) but were concentrated mainly in engineering (down 12%) and in computer sciences (down 23%), fields heavily favored by foreign students. First-time full-time foreign enrollment increased in physical sciences (up 9%) and in psychology (up 10%) and remained stable in the other major science fields in 2003 (appendix table 2-16 Excel table.). These trends may indicate developing trends in total graduate enrollment in future years. Foreign students' share of first-time full-time graduate enrollment dropped from 35% to 29% between 2000 and 2003, with most of the decrease in computer science (from 71% to 52%) and engineering (from 61% to 50%) (figure 2-7 figure.).

According to data collected by the Institute of International Education, the overall number of foreign graduate students in all fields increased 2.4% from academic year 2002–03 to 2003–04. Graduate enrollment of students from India more than doubled between 1999 and 2004 and enrollment of students from China, South Korea, Taiwan, and Canada also increased (appendix table 2-17 Excel table..) (See section "Global Higher Education in S&E" for degrees granted and enrollment of foreign students in other countries.)

Financial Support for S&E Graduate Education

About one-third of S&E graduate students are self-supporting; that is, they rely primarily on loans, their own funds, or family funds for financial support. The other two-thirds receive primary financial support from a wide variety of sources: the federal government, university sources, employers, nonprofit organizations, and foreign governments.

Support mechanisms include research assistantships (RAs), teaching assistantships (TAs), fellowships, and traineeships. Sources of funding include federal agency support, nonfederal support, and self-support. Nonfederal support includes state funds, particularly in the large public university systems; these funds are affected by the condition of overall state budgets. (See sidebar "Definitions and Terminology of Support.") Most graduate students, especially those who pursue doctoral degrees, are supported by more than one source or mechanism during their time in graduate school, and some receive support from several different sources and mechanisms in a given academic year. Self-support is derived from any loans obtained (including federal loans) or from personal or family contributions.

Other than self-support, RAs are the most prevalent primary mechanism of support for S&E graduate students. The percentage of S&E graduate students supported primarily by RAs increased in the late 1980s, rising from about 22% in the early 1980s to roughly 27%–29% of S&E graduate support from 1988 through 2003. Although the number of S&E graduate students relying primarily on fellowships, traineeships, and TAs rose over the past two decades, the percentage of students supported by these mechanisms stayed flat or declined. In 2003, 18% of S&E graduate students were primarily supported through TAs and 13% were primarily supported through either traineeships or fellowships. Self-support was the primary mechanism of support for roughly one-third of S&E graduate students over the past two decades (appendix table 2-18 Excel table.).

Primary mechanisms of support differ widely by S&E field of study. For example, in 2003, full-time students in physical sciences were supported mainly through RAs (44%) and TAs (39%). RAs also were important in agricultural sciences (58%), biological sciences (42%), and engineering (41%). In mathematics, however, primary student support is through TAs (54%) and self-support (19%). Full-time students in the social and behavioral sciences are mainly self-supporting (45%) or receive TAs (20%) (appendix table 2-19 Excel table.).

The federal government served as the primary source of support for about 20% of full-time S&E graduate students in 2003 (appendix table 2-20 Excel table.). This support was mostly in the form of RAs at 70%, up from 61% two decades earlier. The share of federally supported S&E graduate students receiving traineeships declined from 19% in 1983 to 12% in 2003. For students supported through nonfederal sources in 2003, TAs were the most prominent mechanism (40%), followed by RAs (32%) (appendix table 2-18 Excel table.).

The federal government plays a substantial role in supporting S&E graduate students in some mechanisms and fields and a smaller role in others. For example, in 2003, the federal government funded 67% of S&E traineeships, 50% of RAs, and 22% of fellowships. Federal support reaches relatively more students in the physical sciences; earth, ocean, and atmospheric sciences; agricultural sciences; biological sciences; and engineering. However, relatively few students in mathematics, computer sciences, social sciences, psychology, and medical/other life sciences receive federal support (figure 2-8 figure.). Appendix table 2-20 Excel table. gives detailed information by field and mechanism. (See section "Expenditures by Field and Funding Source" in chapter 5 for information on federal academic research and development funding by discipline.)

The National Institutes of Health (NIH) and NSF support most of the full-time S&E graduate students whose primary support comes from the federal government. In 2003, they supported about 24,300 and 19,300 students, respectively. Trends in federal agency support of graduate students show considerable increases from 1983 to 2003 in the proportion of students funded (NIH, from 23% to 30%; NSF, from 20% to 24%). Support from the U.S. Department of Defense declined during the 1990s from 15% to 11%, offsetting to some extent the increasing percentage that received NSF support (appendix table 2-21 Excel table.).

For doctoral degree students, notable differences exist in primary support mechanisms by sex, race/ethnicity, and citizenship. In 2003, male U.S. citizens were more likely to have been supported by RAs (29%) and female U.S. citizens were more likely to have supported themselves from personal sources of funds (27%). Among U.S. citizens, whites and Asians/Pacific Islanders were more likely than other racial/ethnic groups to have had primary support from RAs (26% and 28%, respectively), and underrepresented minorities depended more on fellowships (38%). The primary source of support for foreign doctoral degree students was an RA (46%) (appendix table 2-22 Excel table.).

U.S. citizen white and Asian/Pacific Islander men, as well as foreign doctoral degree students, are more likely than U.S. citizen white and Asian/Pacific Islander women, and under-represented minority doctoral degree students, to receive doctorates in engineering and physical sciences, fields largely supported by RAs. Women and underrepresented minorities are more likely than other groups to receive doctorates in social sciences and psychology, fields in which self-support is prevalent. Differences in type of support by sex, race/ethnicity, or citizenship remain, however, even accounting for field of doctorate (NSF 2000a).

The amount of funding received by graduate students varies widely by source of support (e.g., federal, nonfederal), mechanism of support (e.g., RA, TA), field of study, type and location of school (public/private, urban/rural), and length of contract (12-month or 9- or 10-month). For example, one study showed that average stipends for history TAs on a 9- or 10-month contract were about $11,200 and those for biology RAs were about $19,000 for 12-month contracts (Smallwood 2004). Benefits associated with support mechanisms vary as well. Tuition and fees are waived for most, but not all, TAs and RAs. Most federally funded graduate student fellowship programs stipulate that institutions waive tuition and fees for awardees (NSF 2004c). Most (77%) institutionally supported TAs and RAs include health insurance coverage for students, and a few (21%) include coverage for dependents as well (Smallwood 2004).

Undergraduate and Graduate Debt of S&E Doctorate Recipients

At the time of doctoral degree conferral, about one-fourth of S&E doctorate recipients have some undergraduate debt and about one-third owe money directly related to graduate education.[6] In 2003, 27% of S&E doctorate recipients reported having undergraduate debt and 34% reported having graduate debt. For some, debt levels were high, especially for graduate debt: 2% reported high levels (more than $35,000) of undergraduate debt and 13% reported graduate debt of more than $35,000 (appendix table 2-23 Excel table.).

Levels of debt vary widely by doctorate fields. High levels of graduate debt were most common among doctorate recipients in psychology, social sciences, agricultural sciences, and medical/other health sciences. Psychology doctorate recipients are most likely to report graduate debt and high levels of debt. One-third of psychology doctoral degree recipients compared with 13% of all S&E doctoral degree recipients in 2003 reported graduate debt of more than $35,000.[7] Doctorate recipients in biological sciences; computer sciences; earth, atmospheric, and ocean sciences; engineering; mathematics; and physical sciences were least likely to report graduate debt.

Top of page.Top of page

Top of page.Top of page

Top of page.Top of page


[1] Data for racial/ethnic groups are for U.S. citizens and permanent residents only.

[2] Higher percentages of Hispanic and American Indian/Alaska Native students are enrolled in 2-year institutions compared with students from other racial/ethnic groups. The percentage of black students enrolled in 2-year institutions is roughly similar to that of white students (NSF 2003).

[3] The number of S&E degrees awarded to a particular freshmen cohort is lower than the number of students reporting such intentions and reflects losses of students from S&E, gains of students from nonS&E fields after their freshman year, and general attrition from bachelor's degree programs. (See "Retention in S&E" later in this chapter.)

[4] White, Asian, and Hispanic U.S. citizens and permanent residents accounted for most of the gains in undergraduate engineering enrollment in recent years. For data by race/ethnicity, see

[5] For more detailed information by field, see Graduate Students and Postdoctorates in Science and Engineering: Fall 2002 at

[6] Debt is measured in discrete categories ranging from none to $35,001 or more.

[7] Levels of debt vary within psychology as well. Psychology doctorates who earned PsyDs, those who graduated from professional psychology schools, and those in clinical psychology had higher levels of debt. Despite differences by field, doctorate recipients from most psychology subfields had higher levels of debt than doctorate recipients from other S&E fields (NSF 2000b).

National Science Board.