The defining characteristics of U.S. higher education that foster accessa broad array of institutional types and sizes, public and private funding, and flexible attendance patternswere already in place in the early 1950s. In 1953, more than 1,870 institutionsincluding universities; liberal arts colleges; teachersí colleges; and technological, theological, and other professional schoolswere providing higher education. These diverse institutions included public and private colleges and universities and provided for part-time attendance. One-fifth of the undergraduate students were enrolled part-time (U.S. HEW 1956). Students were concentrated in universities and liberal arts colleges; only 13 percent were enrolled in junior colleges. (See text table 4-1.)
These underlying characteristics of U.S. higher education have persisted during the past 50 years, with expansion occurring through the establishment of many new institutions and the increasing size of universities. In 1953, the largest universities enrolled approximately 10,000 students. By 1996, the largest U.S. universities enrolled between 25,000 and 50,000 students (HEP 1996). Enrollment has surged within research and comprehensive universities. A number of teachersí colleges expanded their offerings and became comprehensive and doctoral institutions. While the number of universities has doubled since the 1950s, the number of two-year institutions has tripledfrom 521 in 1953 to 1,569 in 1996 (HEP 1996). (See figure 4-1.)
Alongside the growth of large institutions in U.S. higher education, liberal arts institutions have remained relatively small. In 1953, liberal arts colleges enrolled approximately 0.6 million students in 713 institutions. By 1996, 1.1 million students were enrolled in approximately 637 such undergraduate colleges (reflecting an average enrollment of less than 2,000 students).
Todayís large and diversified set of institutions provides an education at the bachelorís level to approximately one-third of the U.S. college-age population. (See "Undergraduate S&E Students and Degrees in the United States.") Access to U.S. higher education is still among the highest in the world, although other countries are also broadening access and expanding graduate programs, particularly in S&E. (See "International Comparison of First University Degrees in S&E," "International Comparison of Doctoral Degrees in S&E," and sidebar, "Graduate Reforms in Europe, Asia, and Latin America.")
In the United States, there were 3,660 (1,580 public and 2,080 private) two- and four-year institutions of higher education in 1996 (HEP 1996). These institutions enrolled 14.5 million students at all degree levels in that year and awarded 2.2 million degrees, one-quarter of which were in S&E. (See figure 4-2.)
More than 5 million of the 14.5 million students are enrolled in community colleges. These institutions increase the openness of U.S. higher education; through considerable remedial coursework, they provide a second chance for students who were not well served by, or well motivated during, their high school education. They also foster movement into four-year institutions through arrangements that allow students to transfer their credits from community colleges to four-year colleges and universities.
To better describe this diverse set of institutions serving a variety of needs, the Carnegie Foundation for the Advancement of Teaching has clustered institutions with similar programs and purposes. (See sidebar, "Carnegie Classification of Institutions.")
Carnegie has classified higher education institutions into 10 categories based on the size of their baccalaureate and graduate degree programs, the amount of research funding they receive, andfor baccalaureate collegestheir selectivity.[*] Following is a brief description of these categories.
[*] The Carnegie classification is not an assessment guide, nor are the distinctions between classification sublevels (for example, research I and research II) based on institutionsí educational quality. Baccalaureate college I institutions exercise more selectivity regarding students than do baccalaureate colleges II, but in general the Carnegie categories are a typology, not a rank ordering.
The four-decade expansion in enrollment in U.S. higher education reached its peak in 1992, when more than 14.6 million students were enrolled, and then leveled off. This expansion first accelerated in the late 1940s and early 1950s; by 1950, higher education enrollment had surged to 2.7 million students (up from 1.2 million in 1944) as a result of the post-World War II influx of veterans supported under the GI Bill (U.S. HEW 1956). After the influx of returning veterans subsided, the number of (nonveteran) college students grew steadily for several decades, from the 1960s to the early 1990s, reaching a peak of more than 14.6 million students in 1992. Following more than four decades of such growth in higher education, graduate enrollment began a slight decline in 1993; undergraduate enrollment began declining in 1995. (See "Undergraduate S&E Students and Degrees in the United States" and "Graduate S&E Students and Degrees in the United States.")
From 1967 to 1992, enrollment in U.S. institutions of higher education expanded an average of 3 percent annually, but growth rates differed greatly by type of institution. For example, two-year colleges grew at twice this rate and accounted for the largest share of the growthfrom 0.2 million students in 1950 to 5.5 million students in 1996. (See appendix table 4-2 and U.S. HEW 1956.) In 1950, two-year college enrollment was 9 percent of overall higher education enrollment. By 1996, enrollment in two-year colleges was 38 percent of higher educationís total enrollment. In contrast, student enrollment in research I universities grew more modestly, from 1.5 million students in 1967 to 2.1 million in 1991 (with slight declines since then). Research universities enroll only 19 percent of the students in higher education, but they play the largest role in S&E degree production. (See figure 4-3 and appendix table 4-2.)
A diverse spectrum of institutions provides for relatively high access to higher education in the United States, but the research-intensive universities produce the majority of engineering degrees and a large proportion of natural and social science degrees at both the graduate and undergraduate levels. (See figures 4-4 and 4-5.) In 1996, the countryís 126 research universities awarded more than 42 percent of all S&E degrees at the bachelorís level and 52 percent of all S&E degrees at the masterís level. (See appendix table 4-3.) In addition, comprehensive and liberal arts I institutions produce significant numbers of bachelorís and masterís degrees in science and engineering. (See appendix table 4-3.)
The proportion of S&E degrees earned by institution type in U.S. higher education, however, is not homogeneous for all groups. In contrast to the overall student population, S&E degrees earned by underrepresented minorities are less concentrated in research universities; minority-serving institutions still play a significant role in minoritiesí S&E education. These students earn a far smaller percentage of their bachelor-level degrees in the natural and social sciences at research universities, compared with their engineering degrees and with the percentage of such degrees earned by the overall student population. Over the past 20 years, underrepresented minority students have earned higher percentages of their degrees within research universities in social science and engineering fields, but not in natural science fields. By 1996, underrepresented minority students earned 44 percent of their bachelor-level engineering degrees at research universities, up from 32 percent in 1977. (See appendix table 4-5.) However, the relatively small percentages of degrees earned by underrepresented minority students within research universities have remained stable over the past 20 years. (See appendix table 4-5 and text table 4-2.)
Black students have traditionally earned a large percentage of their S&E degrees within historically black colleges and universities (HBCUs)mainly comprehensive universities and liberal arts colleges. HBCUs, originally established during the period of legalized segregation for the purpose of educating blacks, continue to produce large percentages of the S&E bachelor-level degrees earned by black students. These comprehensive universities and liberal arts colleges produce 30 percent of their engineering degrees, 44 percent of their natural science degrees, and 25 percent of their social science degrees. These percentages have remained relatively stable for the past 20 years. (See appendix table 4-5 and NSF 1999c.)
The associate of arts colleges, which enroll more than 5 million students, account for only a small percentage of S&E degrees. These two-year colleges, however, provide continuing education and flexibility in the U.S. higher education system, allowing students to complete required work-related courses or obtain coursework credits for transfer to a four-year college or university. An analysis of undergraduate careers in engineering in 1995 showed that one out of six students who received a bachelorís degree in engineering, engineering technology, or architecture started in a community college (USDE 1998).
The 126 research universities provide the baccalaureate education of the majority (56 percent) of S&E doctoral recipients. However, liberal arts colleges and comprehensive universities also contribute a significant proportion of bachelor-level degrees among students who later complete doctoral S&E degrees. Each of these institution types provides 15 percent of the baccalaureate education of doctoral recipients; within individual fields they are even more prominent. For example, 23 percent of the students earning doctorates in chemistry received their undergraduate education within a liberal arts college, and an additional 23 percent received their undergraduate education within a comprehensive university. (See appendix table 4-6.)
The U.S. college-age population has declined by more than 21 percent in the past two decades, from 21.6 million in 1980 to 17.0 million in the year 2000. This demographic decline is reflected in the trends presented in this chapter, including the declining number of bachelorís degrees in several fields of NS&E beginning in the late 1980s. (See figure 4-6.) This 20-year population decline of the college-age cohort reverses itself in the year 2000, and increases to 19.3 million by the year 2010. (See appendix table 4-7.) The increase in the college-age population by more than 13 percent in the first decade of the 21st century portends another wave of expansion in U.S. higher educationand growth in S&E degrees at all levels.