| SRS Home » NSF 04-308
Plans for Postdoctoral Research Appointments Among Recent U.S. Doctorate Recipients
Postdoctoral appointments came into use in American higher education in the 1870s, when the U.S. higher-education system began to adopt the European model of the research university. They originally were used mostly in the biological sciences, and postdoctoral scholars (hereafter, postdocs) worked and trained on relatively large-scale laboratory-based projects in universities. Since then, the range of disciplines supporting postdoctoral appointments has expanded, and the number of new research Ph.D.s accepting these appointments has increased (National Academies 2000).
This InfoBrief draws on data from the past decade of the Survey of Earned Doctorates (SED) to examine recent trends in the numbers of new doctorate recipients from U.S. institutions who accept postdoctoral positions. Because these positions can be broadly defined, this analysis relies on data provided by new doctorate recipients who reported having definite plans to take a position as a postdoctoral fellow, postdoctoral research associate, trainee or intern, or to do other postgraduate study. The data include only individuals who received research doctorates in the United States and do not reflect postdocs as a whole, as many U.S. postdoctoral positions are filled by persons who earned their doctorates outside of the United States (CPST 2002).
Trends by Field of Doctorate
The number of new Ph.D.s pursuing postdoctoral appointments has increased in all fields of study since 1982, especially those outside the biological sciences (table 1). Each year for the past 10 years, more than 7,000 persons with newly earned doctorates from U.S. universities have become postdocs. Among doctorate recipients with definite commitments after graduation, the proportion taking postdoctoral positions increased substantially between 1982 and 1993 and has remained between 28 and 30 percent since 1993 (table 1).Table 1 Source Data: Excel file
The percentage of new doctorate recipients who go into postdoctoral positions varies greatly by field of study and other job-market characteristics. In 2002 the percentages ranged from 74 percent (biological sciences) to about 6 percent (education) (figure 1).Figure 1 Source Data: Excel file
Within fields, the proportion of doctorate recipients going to postdoctoral positions has also varied over the past decade, a function of changing numbers of awards and postdoctoral positions. Between 1993 and 2002, the broad fields of biology; physics; chemistry; and earth, atmospheric, and ocean sciences were the fields where postdocs were most prevalent: between 46 and 80 percent of new doctorate recipients with definite commitments in these fields moved into postdoctoral positions after graduation. Although the proportion of doctorate recipients in the field of earth, atmospheric, and ocean sciences who accept postdoctoral appointments has increased somewhat over the past decade, the proportion of their counterparts in chemistry, biology, and physics/astronomy declined between 1993 and 2002. Recipients of doctorates in mathematics and psychology were substantially more likely to be committed to a postdoctoral position in 2002 than in 1993 (table 1).
The SED collects basic demographic information on new doctorate recipients each year, and this enables analysis of trends in postdoctoral plans for selected subpopulations. Overall trends in the proportions of new Ph.D.s taking postdoctoral positions, by sex, citizenship status, and race/ethnicity, are shown in table 2. The proportion of new Ph.D.s planning on postdoctoral positions has fluctuated between 30 and 33 percent for men and has risen from 25 to about 27 percent for women. The narrowing gap between men and women may reflect a number of factors, including the increasing representation of women in science and engineering (S&E) fields, like biology, that have traditionally had high proportions of postdocs, and the relatively high representation of women in fields like psychology, where the number of postdocs has increased.Table 2 Source Data: Excel file
Among doctorate recipients from U.S. universities, non-U.S. citizens plan on postdoctoral appointments in greater proportions than do U.S. citizens (table 2). For the past six years, the citizenship category with the highest percentage of postdocs has been non-U.S. citizens with temporary visas, followed by those with permanent visas. The greater likelihood of non-U.S. citizens planning postdoctoral positions is partly a reflection of their higher concentration in the S&E fields that support higher proportions of postdocs, but the higher rates for non-U.S. citizens relative to U.S. citizens are also evident within other S&E fields. Additional factors appear to contribute to the attractiveness of postdoctoral positions to non-U.S. citizens.
Trends in Sources of Support
The 2002 SED showed that universities were expected to be the main source of support for 42 percent of those with definite plans for a postdoctoral position, followed by the U.S. government (33 percent). Although the remaining sources of support are combined in table 3, the full tabulations show that in 2002, private foundations were cited as the main source of support by 6 percent of postdocs, other nonprofit organizations by 4 percent, business or industry by 4 percent, and other (unnamed) sources by 5 percent. The remaining 6 percent indicated that they did not know the source of their support, possibly reflecting uncertainty about how to classify support from a U.S. government grant to a university-based research program.Table 3 Source Data: Excel file
For the postdoc population as a whole, the 10-year-trend data in table 3 show that the relative shares supported by the U.S. government and other sources decreased, and the share supported by universities increased. Similar patterns of decrease and increase occur in S&E and non-S&E fields, but support from the U.S. government has been consistently less, and support from colleges and universities consistently more, for non-S&E fields than for S&E fields.
Comparing sources of support in 2002 for postdocs by field of doctoral study (table 3), the U.S. government was the main source for over one-third of postdocs in the fields of biological sciences (41 percent); earth, atmospheric, and ocean sciences (40 percent); physics and astronomy (40 percent); agricultural sciences (35 percent); health sciences (34 percent); and chemistry (34 percent). In S&E fields, the lowest levels of U.S. government postdoctoral support that year were in mathematics (22 percent) and social sciences (20 percent). In non-S&E fields, U.S. government support for postdocs was even lower: 11 percent for education, 9 percent for professional/other fields, and 6 percent for humanities.
Over one-third of postdocs in 2002 in every S&E field reported that universities were their main source of support. Universities supported almost two-thirds (64 percent) of mathematicians taking postdoctoral positions in 2002 and supported about 50 percent of engineering and social science postdocs. The decade trends within most fields of study shown in table 3 generally follow the overall pattern of increases in the proportions of postdocs supported mainly by university funds.
National Academies (Committee on Science, Engineering, and Public Policy). 2000. Enhancing the Postdoctoral Experience for Scientists and Engineers. Washington, DC: National Academy Press.
Commission on Professionals in Science and Technology (CPST). 2002. Postdocs: What We Know and What We Would Like to Learn: Proceedings of an NSF/CPST Professional Societies Workshop, December 4, 2002. Washington, DC. (Available online at http://www.cpst.org/Postdoc.pdf.)
For more information, contact
 Susan T. Hill and Mary Golladay, Division of Science Resources Statistics, National Science Foundation, Arlington, VA. Thomas B. Hoffer, National Opinion Research Center, University of Chicago, Chicago, IL.
 There is no precise definition of "postdoc," but the term generally refers to a person who holds a temporary appointment, usually awarded in academe, industry, or government, for the purpose of gaining additional education and training in research.
 The total number of new research doctorate recipients from U.S. universities was 39,955 in 2002, and over the past decade it has ranged from a low of 39,800 in 1993 to a high of 42,654 in 1998. The percentage of the total reporting a definite commitment to postdoctoral study or other employment was 65.5 percent in 2002. That figure fluctuated between 60 and 62 percent from 1993 to 1998 and has increased since then.