For almost a decade starting in 1984, undergraduate enrollment in U.S. institutions of higher education showed strong growth, peaking in 1992 with nearly 12.7 million students. Undergraduate enrollment has declined slightly each year since, mainly from the decrease in the college-age cohort of the majority (white) population. The continuing increase in enrollment for all minority groups did not make up for the loss of white enrollment, resulting in an overall decrease.
The trend of increasing enrollment in undergraduate programs by underrepresented minorities has persisted for over a decade and accelerated in the 1990s. Black enrollment increased 3.6 percent annually in the 1990s, reaching 1.3 million in 1995. In the same period, Hispanic enrollment in higher education increased at an even faster rate (7.1 percent annually.) These national trend data bear watching as some states change affirmative action programs. Undergraduate enrollment of foreign students grew very modestly in the past two decades; in 1995, foreign students still represented only 2 percent of total undergraduate enrollment. (See appendix table 2-11.)
One indicator of the readiness of American students for college-level S&E courses is their self-reported need for remedial work in mathematics and science. The percentage of freshmen reporting a need for such remedial work has remained high, particularly for women and minorities. In 1995, of those freshmen planning to major in science or engineering, over 16 percent of the males and over 26 percent of the females thought they would need remedial work in math. Among freshman students from underrepresented minority groups planning to major in S&E, over 38 percent reported that they would need remedial work in math. This self-reporting of the need for remedial work differed by planned major. Fewer of the students planning a major in the physical sciences or engineering reported needing remedial math, as compared to those planning a major in the social or biological sciences. (See figure 2-8.) Over 20 percent of minority students planning a major in the biological sciences or engineering thought they would need remedial work in science.
Among the majority (white) population, about one-third of the freshman have traditionally contemplated a major in an S&E field; most of these intend to major in a field of natural or social science, with smaller percentages selecting mathematics, the computer sciences, or engineering. From the late 1970s on, the percentage of freshmen planning an engineering major has remained relatively constant, at around 9 percent. During the same period, mathematics and computer sciences have been the intended majors of around 2 percent of incoming freshmen. Freshmen have fluctuated more in their choice of natural science and social science majors. After a decade-long decline in the selection of natural sciences as a possible major, the trend reversed in 1987, increasing to around 12 percent by 1996. The social sciences have become more attractive majors, but not as popular as the natural sciences. (See appendix table 2-15.)
Trends in freshman choice of major show differences by sex and race/ethnicity. Asian American students are moving away from a very high concentration of S&E majors-particularly in engineering-and are majoring in a broader range of fields. While still relatively high, the proportion of Asian American males choosing engineering as freshmen declined from 38 percent in 1980 to 23 percent in 1996. For many years, higher proportions of black and Hispanic males have chosen engineering than have white males, and a higher proportion of black females than white females have chosen to major in mathematics and computer sciences. Women of every race/ethnicity, however, show an increase in choice of natural sciences. (See appendix table 2-15.)
An increasing proportion of those students planning to major in S&E fields are from underrepresented minority groups. In 1996, underrepresented minorities accounted for 15 to 21 percent of those planning to major in the following fields: physical sciences, biological sciences, social sciences, and engineering. In 1976, underrepresented minorities accounted for between 9 and 15 percent of those planning to major in these fields. (See figure 2-9.)
A substantial fall-off occurs between freshmen declaration of intent to study S&E fields and actual completion of S&E degrees (Astin and Astin 1992). This fall-off differs by race, particularly in NS&E fields. There is some fall-off among the majority (white) students: 12 percent intend to major in natural sciences and 9 percent in engineering, but only 8 percent of degrees earned by white students are in the natural sciences and only 5 percent in engineering. A larger fall-off occurs among underrepresented minority groups. Ten percent of black students intend to study a field of natural science, but only 5 percent of degrees earned by blacks are in these fields. Further, 9 percent of black students intend an engineering major, but only 3 percent of undergraduate degrees earned by black students are in engineering. (See appendix tables 2-15 and 2-21.)
Engineering programs require students to declare their major as freshmen, allowing engineering enrollment to be used as an early indicator of undergraduate degrees. The composition of enrollment can also be used as an indicator of participation rates of women and minorities. Undergraduate engineering enrollment declined from a high of 441,205 students in 1983 to 356,177 students in 1996, representing a 19 percent reduction. The decline was neither smooth nor continuous. Engineering enrollment stabilized for several years (1989 to 1992) before resuming its decline. Part-time student enrollment, which accounts for about 10 percent of overall enrollment, has remained relatively stable during the last decade. The relative steadiness of engineering enrollment in the early 1990s is reflected in the stable number of engineering degrees in the 1993-95 period. (See appendix tables 2-13 and 2-20). However, the decline in overall engineering enrollment from 1993 portends a decline in engineering degrees at the end of the decade and in the year 2000.
While overall undergraduate engineering enrollment has been declining, enrollment of women and minorities has been increasing, particularly in the 1990s. The number of female students enrolled in engineering increased from 61,000 in 1990 to 68,000 in 1996. For underrepresented minorities, the increase was greater, from 41,000 in 1990 to almost 54,000 in 1996. By 1996, female students represented 19 percent of total undergraduate engineering enrollment, and underrepresented minorities represented 15 percent of such enrollment. Concurrently, the number of foreign students enrolled in U.S. undergraduate engineering programs has been decreasing, in response to enhanced capacity in engineering programs abroad. (See figure 2-10 and appendix table 2-14.)
Universities strive to address the academic needs of students in all majors. In addition to S&E, disciplines that require a grounding in mathematics and science include K-12 education, business, and law, among others. With the increasing interplay of science and technology in our society, all citizens benefit from a higher level of technological literacy and an understanding of the methods and processes of science.
In the 1990s, many S&E departments have designed or adapted new curricula to broaden the attraction to, and success with, science and engineering courses. For example, several academic institutions have initiated "calculus reform," a movement to align calculus instruction more closely with theories of how students learn; others have created multimedia software modules to enhance visualization for students not majoring in science. A large number of institutions have adopted or designed revitalized curricula or variations of these reforms. (Advisory Committee to NSF/EHR 1996). By 1995, 22 percent of the 372,000 students enrolled in calculus 1 and 2 were using a reform text along with various other innovations, such as graphing calculators, writing and computer assignments, and group projects (Rung 1997).
Recent data from the Longitudinal Study of American Youth (LSAY) reveal some facts about coursetaking behavior in science and mathematics among those who attended two- or four-year colleges and universities. As expected, science, engineering, and mathematics majors report a far higher number of completed mathematics and science courses than non-S&E majors. Over half of the mathematics and engineering majors report five or more courses in mathematics. Over 90 percent of the science majors report five or more courses in science. However, many non-S&E majors are taking mathematics and science courses beyond the general education requirements (in a liberal arts program, typically two mathematics courses and two science courses to graduate). Over half of the education majors who earned a bachelor's degree took three to four mathematics courses, with over 40 percent taking three to four courses in science and 25 percent taking even more. (See appendix tables 2-22 and 2-23.)
Every five years since 1970, the Conference Board of the Mathematical Sciences (CBMS) has conducted a survey of a sample of four-year college and university departments of mathematics and two-year college programs in mathematics. These data are important in estimating overall enrollment trends, as well as in breaking out trends in mathematics courses taken by level of difficulty. Estimates of overall enrollment in courses taken in mathematics departments in four-year institutions declined substantially from the peak years of 1985 and 1990, as fewer undergraduate students majored in mathematics or took calculus or advanced level coursework.
The CBMS data show that mathematical enrollment trends differed by level of institution as well as level of difficulty. Enrollment increased in precalculus courses designed primarily for liberal arts students in four-year colleges and universities, and in remedial mathematics courses in two-year colleges. In 1995, at the community college level, over half (57.8 percent) of the enrollment in mathematics classes was for remedial level courses. This high proportion of remedial mathematics at the community college level has existed since 1985. In 1970, remedial courses in community colleges represented about one-third of all mathematics courses.
Within four-year college and university mathematics departments, the estimated enrollment in remedial level courses has remained at about 15 percent of total mathematics enrollment since 1980. The proportion of mathematics enrollment in advanced courses has remained within a range of 6 to 9 percent since 1980, with enrollment in precalculus and calculus each accounting for about 40 percent of total mathematics enrollment). (See text table 2-3.)
At the associate degree level, the number of degrees in engineering technology has fallen precipitously, from 51,000 earned degrees in 1983 to 39,000 degrees in 1995. (See appendix table 2-18.) Between 1994 and 1995, the number of degrees decreased in all fields of S&E. This decline in associate degrees in S&E holds regardless of race/ethnicity. (See appendix table 2-19.) The one exception is Asian American students: in the sciences, their number of earned degrees is increasing slightly.
The declining trend in associate degree completions may be partly explained by the changing roles of junior colleges in the United States. Community colleges now go far beyond providing associate of arts degrees. They provide short courses, train in work-related technical skills, and serve as feeder schools to four-year colleges and universities. In contrast to the junior college level in many other countries- such as Japan and France-this level of higher education in the United States provides flexibility, allowing individuals to take courses outside of a degree program, as well as transition to more advanced levels of higher education. Many associate of arts colleges have an agreement with four-year schools to allow transfer of credits. For example, California encourages students to begin their college studies at a local community college, with the understanding that they will be admitted to a state university for their third and fourth years of a bachelor's degree.
Community colleges also pioneered distance learning to reach large numbers of students within their geographic region, and are partnering with universities to provide distance learning with local laboratory work. (See "Distance Learning and Its Impact on S&E Education.") [Skip Text Box]
Virtually all of the 300 engineering programs in the United States have some form of continuing education with distance learning for a local area; less prevalent but growing is generalized distance learning, with course material on the Internet. Students are increasingly participating in fully developed S&E lessons at home, at the office, in a library carrel, or even at another university. The impetus for distance learning stemmed from the responsibility of community colleges to serve a large number of students within a geographic region, and their need to develop off-site learning centers. In a 1991 survey by the American Association of Community and Junior Colleges, 80 percent of community colleges and 78 percent of universities had plans to provide distance learning by 1994 (Brey 1991).
S&E higher education has benefited from advances in distance learning. In the 1980s, television became an instrumental medium for developing courses and degree programs at the undergraduate and graduate levels. One example is the University of California at Davis Instructional Television program. Classes are broadcast live during the workday, and students usually enroll in one course per quarter. Full-time professional engineers obtain a master's degree in approximately three years and a doctoral degree within five to six years.
Telecommunications and satellite delivery make it possible for students to obtain their degrees almost anywhere in the world. Colleges and universities are using these support technologies to augment their existing distance learning programs-e.g., fax, CD-ROM, e-mail, two-way audio, and teleconferencing. (See text table 2-4.) For example, the National Technological University, a consortium of 47 leading engineering universities, offers 1,200 courses and 13 master's degree programs in science and engineering.
The Internet offers a fundamental advancement in distance learning delivery. The new Internet applications for audio, video, and two-way communication are expected to integrate the previous advancements in distance learning technologies into a single medium. Schools are beginning to experiment with on-line courses; for example, the University of Phoenix offers on-line courses that present workshops, homework, and even the final exam via the Internet. The Internet's impact on S&E higher education is not clear at this time, but several S&E associations are actively discussing its potential. (For more information, see chapter 8, "IT, Education, and Knowledge Creation.")
Except for a brief decline between 1986 and 1989, the number of earned bachelor's degrees in S&E from U.S. institutions has been increasing for over a decade, rising from over 307,000 in 1981 to 378,000 in 1995. Trends in earned S&E degrees in U.S. institutions, however, differ widely by field. In the natural sciences, a long slow decline from 1976 to 1990 ended, shifting to an upturn in such degrees during the 1990s. Natural science degrees increased 7.7 percent annually from 1990 to 1995, with stronger than average growth in the biological and environmental sciences, but only modest (2 percent) growth in the physical sciences. The number of completed math and computer science degrees declined from 1975 to 1979, then climbed steadily reaching almost 59,000 degrees in the peak year of 1986. Attraction to the computer sciences dropped precipitously from 1986 to 1991, followed by slight decreases to 1995. The number of social science degrees awarded, after record growth between 1986 and 1992 (averaging 6 percent annually), has remained stable for the last four years. Engineering degrees, whose numbers also peaked in 1986 following a decade of strong growth-particularly in electrical and mechanical engineering-declined until 1991 and then stabilized. The slight annual growth rate in engineering degrees from 1991 to 1995 is mainly accounted for by the increasing number of degrees in chemical and civil engineering. (See figure 2-11 and appendix table 2-20.)
These recent trends in earned degrees for S&E fields show a similar pattern for both males and females, with a few exceptions in the social sciences and engineering. After 1993, degrees earned by males decreased slightly in the social sciences, while females maintained their high number of degrees in these fields. In engineering, females increased their earned degrees in the 1990s, particularly in chemical and civil engineering. In the same period, degrees in engineering earned by males declined slightly. (See figure 2-12.)
Over the past two decades, the proportion of S&E degrees earned by females has increased considerably, particularly in the natural sciences and engineering. In 1975, females earned about one-quarter of the degrees in the natural sciences and 2 percent of those in engineering. By 1995, females earned 59 percent of social science degrees, 47 percent of natural science degrees, 35 percent of mathematics and computer science degrees, and 17 percent of the engineering degrees. (See appendix table 2-20.)
Trends in S&E bachelor's degrees also differ by race/ethnicity, with white students earning fewer degrees in 1995 than in earlier years, and minority groups continuing their growth in earned degrees in these fields. The number of degrees earned by white students is slowly decreasing in all fields except the natural sciences.
In contrast, the number of degrees earned by underrepresented minorities in the United States-blacks, Hispanics, and Native Americans-is increasing slightly in NS&E fields and very rapidly in the social sciences. (See "S&E Human Capital Development: Continued Unevenness Across Demographic Groups.") In addition, the number of degrees earned by Asian Americans is increasing sharply in the natural and social sciences. (See appendix table 2-21.) [Skip Text Box]
Beginning in the early 1980s, increasing numbers of women and minorities entered U.S. higher education. For a decade, the broadened entry of these groups fueled the expansion of enrollment in U.S. higher education and helped offset the trend of a declining U.S. college-age cohort. However, this broader access and increased enrollment in higher education did not concurrently result in larger numbers of S&E degree completions for women and minorities in all S&E fields at all levels. The pattern of participation is stronger in overall enrollment than in completed S&E degrees, stronger for females than for males in all underrepresented minority groups, stronger at the undergraduate than graduate level, and stronger in the natural and social sciences than in computer sciences and engineering.
Women. In the last decade, women achieved a higher rate of growth in undergraduate enrollment than men, particularly women in minority populations. Women now constitute 56 percent of undergraduate enrollment and an even higher percentage among minority populations. Women of every racial/ethnic group are increasingly choosing majors in the natural sciences and social sciences. At the bachelor's level, women now earn over half of the social science degrees and almost half of natural science degrees. However, women are less fully represented at the graduate level; in 1995, they accounted for 38 percent of total graduate enrollment. Women earned the majority of master's degrees in the social sciences and 41 percent of the master's degrees in the natural sciences. Women are least fully represented at the doctoral level. While women earn half of the doctoral degrees in the social sciences and 32 percent of the degrees in the natural sciences, they earn only 20 percent of the doctoral degrees in mathematics and computer sciences and less than 12 percent of doctoral engineering degrees.
Underrepresented minorities. The trend of increasing enrollment in undergraduate programs by underrepresented minorities has persisted for over a decade and accelerated in the 1990s, particularly for Hispanic populations. While minority groups indicate high aspirations to study S&E (as measured by freshman intentions), a substantial fall-off occurs between freshman declaration of intent and actual degree completion. This fall-off is greater for underrepresented minorities than for the majority population. Women and minority students are more likely to report a need for remedial work in mathematics and science than the majority male population. (Chapter 1 further discusses the large gap between minority students and the overall student population in number of science and mathematics courses taken.) There has been modest progress in minority participation in S&E degree completions. From 1975 to 1995, S&E bachelor's degrees earned by minorities increased from 6 to 8 percent of total such degrees. (Underrepresented minorities are around 28 percent of the college-age cohort.) Only about 2 percent of the 24-year-olds in underrepresented minority populations hold a bachelor's degree in NS&E-less than half the rate of the majority white population.
Progress for underrepresented minorities in S&E graduate enrollment has been very modest. In 1975, they accounted for 3.7 percent of S&E graduate enrollment; by 1995, they accounted for 5.0 percent. Minority students are underrepresented in S&E graduate degrees. They earn 7 percent of the master's degrees in S&E fields and less than 5 percent of the doctoral degrees. Women in these minority groups earn the majority of these degrees. (See NSF 1996f for disaggregated degree data by sex within each racial/ethnic group.)
Foreign students have increased their earned degrees in the social sciences, but since 1981 have sharply decreased their degrees in engineering from U.S. institutions, as discussed in more detail below. The capacity to educate engineering students at the undergraduate level has increased dramatically in other world regions, and fewer foreign students are using U.S. universities for engineering education.
The United States is one of the leaders in the world in providing access to higher education and ranks high among the major industrialized countries in the proportion of its population with an S&E background. These national statistics, however, do not apply to all fields or to all minority groups. In 1995, for the country as a whole, over 32 percent of the college-age population had completed a bachelor's degree in some field, and over 5 percent had earned a bachelor's degree in an NS&E field. But in that same year, only about 15 percent of black and Hispanic youth earned a college degree, and only about 2 percent of black and Hispanic youth earned a bachelor's degree in an NS&E field. In contrast, Asian Americans, representing only 4 percent of the U.S. population, have considerably higher than average participation rates: almost 40 percent obtained a bachelor's degree, and over 12 percent earned such a degree in NS&E.
Recent participation rates do show some progress toward more diversity in higher education in general and in S&E fields, compared with 1980 and 1990 data. (See text table 2-5.) Low participation rates for blacks and Hispanics changed little throughout the 1980s, although they improved considerably in the 1990s, particularly in the social sciences. In 1995, 3.8 percent of the U.S. female population earned an NS&E degree, compared to 2.1 percent in 1980.
A recent study highlights the core elements of an international education that will be important for American youth preparing to work in the global economy of the 21st century (IIE 1997). Referred to as "transnational competence," this education involves a combination of cultural and technical skills, including:
The United States has traditionally been weak in providing foreign language instruction. More recently, however, universities are improving undergraduate education by attempting to provide meaningful international experience as an integral part of coursework. (See "International Engineering Programs in the United States.") While there are no national data on the short-term visits conducted under such enhanced undergraduate curricula, the number of courses taken for credit overseas have increased, including engineering courses. (See text table 2-6.) [Skip Text Box]
International engineering programs (IEPs) allow U.S. students to gain valuable experience in an international setting. Traditional engineering curricula have been too tight and structured to allow engineering students to study abroad. IEPs, however, are customized to permit such study. A University of Cincinnati survey of universities with IEPs listed on the World Wide Web shows study abroad and work abroad components integrated into the engineering programs of about 25 major U.S. universities. A well-structured IEP gives students an opportunity to examine engineering in a foreign culture.
To promote the creation of IEPs, several universities in the United States and abroad are affiliated with the International Engineering Consortium. The consortium conducts a broad range of university-industry cooperative programs and continuing education programs. Members of academia and industry meet to discuss leading-edge technology, issues vital to the information age, and the nature of today's global marketplace. (For more information, see http://www.iec.org.)