At the threshold of the 20th century the United States had a small but vigorous graduate education enterprise rooted in a new kind of institution, the research university. The distinctive feature of the research university was the doctoral program. In 1900 most of the approximately 250 research doctorates conferred were from the dozen or so new research universities in the Northeast and Midwest.1
The next 100 years of doctoral education were marked by growth and diversification. Doctoral education became available in all 50 U.S. states, the District of Columbia, and Puerto Rico as the number of institutions offering doctoral programs grew. The number of disciplinary fields in which scholarly study was offered at the doctoral level; the number of doctorates awarded annually; and the number of women, minorities, and foreign nationals earning doctorates also grew over the course of the century. By 1999 the number of research doctorates conferred annually had reached more than 41,000, with these representing nearly 400 institutions across the United States.
Between 1900 and 1999 the U.S. graduate education system developed into an integral part of the social and economic structure of the country, contributing to the nation's economic standing and also constituting a significant component of the U.S. economy. By the end of the 20th century, a total of 426 institutions had awarded more than 1.36 million doctorates.
The U.S. Model of Doctoral Education
The United States is unique in the extent to which fundamental research is conducted at universities, typically with the assistance of graduate students. Doctoral education is organized around an intensive, real-world research experience that prepares students to be scholars capable of discovering, integrating, and applying knowledge (CGS 1990). The American system, in which universities conduct research and train doctoral students, has become a model for the world (NAS/NRC 1995). It has played a major role in the United States' strong record of innovation and economic growth and has helped this nation become the global leader in science and engineering (NSB 1998).
American doctoral education produces cutting-edge knowledge and highly trained personnel who go on to fill specialized positions as teachers, researchers, and professionals in academe, industry, government, and nonprofit organizations. In 1999 just over half of new U.S. citizen Ph.D.s with job commitments went into higher education after graduation—48 percent to 4-year institutions and 4 percent to 2-year colleges. About 22 percent of new Ph.D.s found jobs in industry or planned to be self-employed. Eleven percent planned to take jobs in elementary or secondary schools, 9 percent in government (mainly within the federal government), and 6 percent in nonprofit organizations.
A look at Nobel Prize winners during the 20th century illustrates the value of the U.S. doctoral system and the lasting influences of its graduates. Between 1901 and 1999, 162 Americans with doctorates from U.S. universities received 164 Nobel Prizes: 57 prizes for physics, 41 for chemistry, 34 for physiology or medicine, 25 for economics, and 7 for peace (Sherby 2002).
Beginnings of Doctoral Education in the United States
For the greater part of the 19th century, most institutions of higher learning were small and church affiliated and provided a classical education in the liberal arts, producing ministers and other professionals. By the middle of the 19th century, new fields—especially in sciences and engineering—had been added to the curriculum or were being taught in separate institutions, such as the Massachusetts Institute of Technology (founded in 1865), or in separate schools within institutions, such as Harvard's Lawrence Scientific School (1847) and Yale's Sheffield Scientific School (1854). These were not set up to be graduate programs, however. Of necessity, Americans went to Europe, especially Germany, for advanced education, including the doctorate. It has been estimated that 10,000 Americans traveled to Europe in the 19th century to study, more than half in departments of philosophy, the forerunner of arts and sciences (Walters 1965); the others pursued professional studies, such as theology, law, and medicine. Some of these students earned doctorates and then returned to the United States, where most became college professors and many became leaders in the effort to create American universities.
In the mid-19th century, there were numerous proposals and some efforts to develop American universities with graduate programs. The year 1861 is significant not only as the beginning of the Civil War but also as the first year in which doctorates were conferred in the United States. In 1861, three earlier bachelor's graduates of Yale College received the Doctor of Philosophy degree from Yale's Sheffield Scientific School after completing 2 years of graduate study, passing final examinations in several fields, and submitting written dissertations (Storr 1953).
Although Yale awarded four more doctorates in the next 2 years, graduate programs did not begin in earnest until the 1870s (Goldin and Katz 1999). The growth in graduate education was driven by institutional developments and fueled by increasing demand for advanced education at the end of the 19th century and the beginning of the 20th century. The University of Pennsylvania was the second American college to award a doctorate, in 1871, followed by Harvard, which conferred two Ph.D.s and one Sc.D. in 1873 (Bruce 1987). In 1876 an African American was the first racial/ethnic minority to earn a doctorate in the United States, receiving a doctorate in physics from Yale. In 1877 the first woman received a doctorate in the United States.
In the last quarter of the 19th century, a small group of research universities offering doctoral programs emerged in response to a growing demand for knowledge in an industrializing society. The First Morrill Land Grant Act of 1862 gave each state a large grant of public lands to sell for the purpose of founding a college that taught agriculture and the mechanic arts as well as classical studies, leading to an expansion of public universities in the 1870s (NASULGC 1999). Some of these land-grant institutions became early awarders of doctorates, defined in this report as having awarded doctorates before 1920. Chapter 5 discusses the contributions of these early awarders in terms of the number of doctorates they conferred throughout the 20th century.
The Second Morrill Land Grant Act of 1890 led to the establishment of public colleges in southern and border states that did not already have colleges for blacks (USDA 2003). In addition, many of the schools that had been established between 1866 and 1890 to train black teachers were incorporated into the land-grant system under the Second Morrill Land Grant Act. These 18 colleges, known as the 1890 institutions, addressed barriers to higher education that resulted from segregation. Five of these institutions eventually granted doctorates: Alabama A&M University, Florida A&M University, North Carolina A&T State University, South Carolina State University, and Tennessee State University. The 1890 institutions make up nearly one-fifth of the group of institutions known as historically black colleges and universities, or HBCUs (White House Initiative on Historically Black Colleges and Universities 2002).
About the same time, there was an increase in the number of new private colleges with graduate programs (Clark, founded in 1889; Stanford, in 1891; Chicago, in 1892) and reorganization of older ones (Harvard and Columbia). Johns Hopkins University, a private institution founded in 1876, focused primarily on graduate education and emphasized original research by faculty and research-based doctoral dissertations (Storr 1973:41; Geiger 1986:8). Other colleges adapted their graduate programs to the Johns Hopkins model, although they placed more emphasis on teaching. This process was reinforced as Johns Hopkins graduates became professors in other universities.
By the end of the 19th century, the American model of doctoral education had been established. It was based in a research university with undergraduate and graduate programs taught by the same faculty. The faculty members were organized by discipline in departments that were closely linked to the disciplinary scientific societies and specialized scholarly journals that were founded between 1890 and 1905.
The American doctoral program generally included several years of coursework, final examinations, a year or two of research, a language requirement, and a dissertation (Walters 1965). Most recipients became faculty members in colleges and universities whose enrollments were expanding substantially. They were recruited to teach in their specialized area of scholarship.
Despite these advancements, U.S. doctoral education was in disarray at the turn of the century. American students were still flocking to European universities for graduate study, and American universities were viewed with little respect by European universities.
The problem was that, unlike in Europe, higher education in America was decentralized and largely unregulated; diploma mills proliferated, and even shaky institutions could call themselves "universities" and award Ph.D.s. Some institutions, for example, allowed Ph.D. candidates to pursue courses without showing up on campus and to take exams at home under supervision of a proctor. The lack of standards and consistency was hurting the reputations of the more demanding U.S. universities. (Speicher 2000)
In an effort to improve the situation, in January 1900 the presidents of the University of Chicago, Harvard, Columbia, Johns Hopkins, and the University of California–Berkeley invited 14 of the leading doctorate-granting institutions to send representatives to a conference to consider "matters of common interest relating to graduate study." According to the letter of invitation, the conference's goals were to
- result in a greater uniformity of the conditions under which students may become candidates for higher degrees in different American Universities, thereby solving the question of migration, which has become an important issue with the Federation of Graduate Clubs;
- raise the opinion entertained abroad of our own Doctor's Degree;
- raise the standard of our weaker institutions. (AAU 2003)
Thus was founded the Association of American Universities (AAU) (table 2-1 ). Over the years the number of AAU members has grown, by invitation, from the 14 founders to 59 U.S. universities and 2 Canadian universities.2 More than two-thirds (110) of the American Nobel Prize winners in the 20th century who earned doctorates in the United States graduated from AAU's founding institutions (Sherby 2002). Harvard (24), Columbia (22), Chicago (17), and the University of California–Berkeley (14) had the most Nobel Prize winners among their doctoral graduates.
Growth in Doctorate-Granting Institutions
Although by 1900 research-based graduate training was a permanent component of American higher education, it was still concentrated in a small set of institutions. The 14 universities that founded the AAU were the leading doctorate producers at that time and accounted for nearly 90 percent of all doctorates awarded in 1900 (Berelson 1960). The number of doctorate-granting institutions increased steadily throughout the 20th century, from fewer than 50 institutions before 1920 to 392 in 1999 (figure 2-1 ). In 1999 the 59 AAU members awarded 51 percent of the doctorates. The top 10 institutions (based on the number of doctorates conferred), which awarded 86 percent of all doctorates in 1900, awarded less than 16 percent in 1999.
The greatest growth in doctoral programs at U.S. institutions of higher education was in the 1960s and 1970s, after the Soviet Union launched the satellite Sputnik. That 1957 event triggered new national policies focused on increasing the number of research universities (PSAC 1960). The number of doctorate-granting institutions grew by 73 in the 1960s and by another 87 in the 1970s. By the mid-1960s, institutions with doctoral programs were in all 50 U.S. states, the District of Columbia, and Puerto Rico (figure 2-2 ). The rate of growth in the number of new doctorate-granting institutions slowed in the 1980s and 1990s, although the number of doctorates awarded continued to rise.
In the last half of the 1940s the regional concentrations of doctorate-granting institutions began to shift away from the Northeast and Midwest into the South and West (figure 2-3 ). In 1920, 75 percent of all doctorate-granting institutions were located in the Northeast or Midwest. By 1999, 44 percent were in these regions.
The substantial growth of the general population in the South since the 1970s and in the West since the 1940s also contributed to the establishment of new institutions and new doctoral programs within existing institutions. According to the U.S. decennial censuses, the South's population represented about 31 percent of the total U.S. population until 1970 then rose to nearly 36 percent in 2000 (U.S. Census Bureau 1975, 1998, 2001). The West increased its share of the U.S. population every decade—from less than 6 percent in 1900 to 22 percent in 2000.
At the same time, the populations in both the Northeast and the Midwest declined as percentages of the total U.S. population—in the Northeast, from about 28 percent in 1900 to 19 percent in 2000, and in the Midwest, from about 35 percent in 1900 to 23 percent in 2000.
Growth in Doctoral Awards and R&D Spending
The annual number of research doctorates awarded in the first two decades of the century ranged from a low of 293 in 1902 to a high of 667 in 1916. The number of doctorates surpassed 1,000 in 1923, grew to a peak of 33,755 in 1973, and then peaked for a second time at 42,683 in 1998 (figure 2-4 ). This growth represents an average increase of almost 7 percent per year through the early 1970s.3 These long-term trends were affected in the short term by such major events as the two world wars and the Great Depression.
Growth was slow at the beginning of the century and declined sharply during World War I. A spurt marked the 1920s, followed by another period of slow growth (but not a decline) during the Depression years of the 1930s. A large decline during World War II was followed by sharp increases from 1946 to 1950 and continued growth until 1955. By the mid-1950s, however, the pent-up demand for doctorates following World War II had been spent (NAS/NRC 1978).
The 1960s were characterized by a high rate of growth in annual production of doctorates, fueled by public and government reaction to the launching of the Soviet satellite Sputnik in 1957. This achievement prompted a buildup of the graduate education system in the United States through a variety of programs, including graduate fellowships funded under the National Defense Education Act of 1957 and fellowship and traineeship programs of the National Science Foundation, the National Institutes of Health, and the Public Health Service.
National expenditures for research and development (R&D), both federal and nonfederal, increased at an annual rate of about 8 percent in real terms between 1953 and 1969 (table 2-2 ).4 Total funding for R&D in colleges and universities also increased rapidly during this period, as did federal support for academic R&D. Many graduate students were supported as research assistants on research grants and contracts. The annual number of doctorates awarded rose from 8,611 in 1957 to 33,755 in 1973, an increase of nearly 9 percent per year.
Cutbacks in R&D funding began in 1969. With the moon landing achieved, energy programs reduced, and costs of the Vietnam War beginning to have an impact on the federal budget, R&D spending was reduced significantly during the 1969–75 period. Federal dollars for academic R&D declined in 1969 and stayed low through 1975. Federal research fellowship and training programs were reduced. The government's share of total academic R&D funding also declined and continued to do so to the end of the century, sliding from a high of 73 percent of all academic R&D funding in 1965–68 to about 58 percent in 1999.
During the 1970s the academic labor market in most fields became saturated, and there was concern about overproduction of Ph.D.s. In addition, the end of draft deferments for graduate students during the Vietnam War (effective 1968) had a negative effect on graduate enrollments of men.5 Together these conditions resulted in a significant reduction in doctoral awards in the 1970s. By the late 1970s the number of doctorates awarded annually had declined to about 31,000. This number remained almost flat from 1978 to 1985.6
With the defense buildup and gains in R&D spending of the 1980s, increases in doctoral awards resumed in all major fields except education. The number of doctorates conferred rose from 31,297 in 1985 to 42,683 in 1998, although the average rate of growth—about 2 percent per year—was much slower than the rate of growth during the first three-quarters of the century.
The relatively high rate of growth in doctorate production during the century means that most of the 1.36 million doctorates awarded between 1900 and 1999 were conferred in the last few decades of the century. More than half of all doctorates were awarded between 1980 and 1999, and three-fourths were conferred between 1970 and 1999 (table 2-3 ).
Between 1920 and 1999 there was a notable change in the geographical distribution of doctorates conferred. The concentration of degree production in the Northeast and Midwest ended with the relatively greater development of doctorate-granting institutions in the South and West and the greater population growth in these regions after 1920 (figure 2-5 ). These regional shifts were also due in part to the presence of larger, public universities among the newer doctorate-granting institutions. In 1952 the number of public doctoral institutions surpassed the number of private doctoral institutions. One year later the number of doctorates produced by public institutions surpassed the number produced by private institutions (figure 2-6 ). By the 1970s public universities accounted for about two-thirds of the doctorates conferred each year, a proportion that held steady to the end of the century. Total doctorate production from 1920 to 1999, however, reflects the early concentration of degrees in the Northeast and Midwest (table 2-4 ). All but 3 of the top 10 states for doctoral awards during the 1920 to 1999 period were in these two regions.
Association of American Universities (AAU). 2003. The Letter of Invitation to the Founding Conference of AAU. http://www.aau.edu/aau/Invit.html. Accessed April 2005.
Berelson, B. 1960. Graduate Education in the United States. New York: McGraw-Hill.
Bowen, W.G., and N.L. Rudenstine. 1992. In Pursuit of the PhD. Princeton: Princeton University Press.
Bruce, R.V. 1987. The Launching of Modern American Science: 1846–1876. New York: Alfred A. Knopf.
Cattell, J.M. 1927. The origin and distribution of scientific men. Science 66 (November 25):513–6.
Council of Graduate Schools (CGS). 1990. The Doctor of Philosophy Degree: A Policy Statement. Washington, DC.
Geiger, R.L. 1986. To Advance Knowledge: The Growth of American Research Universities, 1900–1940. New York: Oxford University Press.
Goldin, C., and L.F. Katz. 1999. The shaping of higher education: The formative years in the United States, 1890 to 1940. Journal of Economic Perspectives 13 (winter):37–62.
Kohler, R.E. 1990. The Ph.D. machine: Building on the collegiate base. Isis 81:638–62.
National Academy of Sciences, National Research Council (NAS/NRC). 1978. A Century of Doctorates: Data Analyses of Growth and Change. Prepared by L. R. Harmon. Washington, DC.
National Academy of Sciences, National Research Council (NAS/NRC). 1995. Reshaping the Graduate Education of Scientists and Engineers. Washington, DC.
National Association of State Universities and Land-Grant Colleges (NASULGC). 1999. The land-grant tradition. http://www.nasulgc.org/publications/Land_Grant/Land_Grant_Main.htm. Accessed April 2005.
National Science Board (NSB). 1998. The Federal Role in Science and Engineering Graduate and Postdoctoral Education. NSB 97-235. Arlington, VA: National Science Foundation.
National Science Board (NSB). 2002. Science and Engineering Indicators 2002. 2 vols. NSB-02-1. Arlington, VA: National Science Foundation.
President's Science Advisory Committee (PSAC). 1960. Scientific Progress, the Universities, and the Federal Government. Washington, DC: U.S. Government Printing Office.
Sherby, L.S., ed. 2002. The Who's Who of Nobel Prize Winners 1901–2000. 4th ed. Westport, CT: Oryx Press.
Speicher, A.L. 2000. The Association of American Universities: A century of service to higher education. http://www.aau.edu/aau/Begin.html. Accessed April 2005.
Storr, R.J. 1953. The Beginnings of Graduate Education in America. Chicago: University of Chicago Press.
Storr, R.J. 1973. The Beginning of the Future. New York: McGraw-Hill.
U.S. Census Bureau. 1975. Historical Statistics of the United States, Colonial Times to 1970. Bicentennial ed. Washington, DC.
U.S. Census Bureau. 1998. Statistical Abstract of the United States. 118th ed. Annual series. Washington, DC.
U.S. Census Bureau. 2001. Statistical Abstract of the United States. 121st ed. Annual series. Washington, DC.
U.S. Department of Agriculture (USDA). 2003. 1890 land grant universities. http://www.reeusda.gov/1890. Accessed March 2003.
Walters, E. 1965. The rise of graduate education. In E. Walters, ed., Graduate Education Today, pp. 1–29. Washington DC: American Council on Education.
White House Initiative on Historically Black Colleges and Universities. 2002. http://www.ed.gov/about/inits/list/whhbcu/edlite-list.html. Accessed April 2005.
1 The U.S. Office of Education reported that 382 doctorates were awarded in 1900, but scholars who examined the data found that approximately one-third of these were awarded for off-campus work, or for no work, by universities lacking legitimate doctoral programs or adequate facilities, and that another 8–10 percent were honorary (Kohler 1990; Berelson 1960; Cattell 1927).
2 See http://www.aau.edu for more information on AAU and its centennial celebration.
3 Seven percent annual increases resulted in a near doubling of the annual production of doctorates every 10 years from 1870 to 1970.
4 The discussion of total R&D and academic R&D is based on NSB 2002, 1:4-3–4-10, 5-3–5-13; 2:A4-5, A4-6, A5-2.
5 The Selective Service Act of 1967 and Executive Order 11360 ended 2-S deferments for all graduate students except those enrolled in certain programs (mainly medicine and related fields). Graduate students who had been enrolled in fall 1967 were allowed to complete that academic year, after which their deferments ceased. Individuals receiving a bachelor's degree in 1968 were not eligible for further student deferments unless they enrolled in the graduate programs specified (Bowen and Rudenstine 1992).
6 Institutions reacted in different ways to the retrenchment of the 1970s. The result was a redistribution of students among institutions. Many of the long-established, top-rated doctoral programs reduced or froze graduate enrollments. Institutions in lower tiers with newly created programs chose to continue their programs rather than dissolve them, which would have meant losing their substantial investment in developing the programs. Consequently, large numbers of graduate students shifted to institutions in the lower tiers (Bowen and Rudenstine 1992).