by Laura Burns, Peter Einaudi, and Patricia Green[1]

U.S. enrollment in science and engineering (S&E) graduate programs in 2007 increased by 3.3% over comparable data for 2006. This is the highest annual growth rate since 2002 and is nearly double the 1.7% growth rate seen in 2006. First-time, full-time enrollment of foreign students (the terms foreign student and temporary visa holder are equivalent in this report) eclipsed its previous high, set in 2001, and total enrollment of temporary visa holders topped its 2003 high.[2] Despite this growth, the proportion of S&E graduate students who are temporary visa holders remained below its peak level, set in 2002, because of growth in the numbers of U.S. citizens and permanent residents pursuing graduate-level study in S&E fields. Changes in enrollment from 2006 to 2007 are shown in table 1 under "2006–07old" for reasons explained below.

  Table 1 Source Data: Excel file

The NSF-NIH Survey of Graduate Students and Postdoctorates in Science and Engineering (GSS), which collects these data, was refined in 2007 to improve reporting. New fields were added to the survey, and some fields were reclassified. Because of these changes this InfoBrief presents 2007 data in two ways:

• "2007new" counts report the data as collected using the new methodology
• "2007old" counts reflect the data as they would have been collected in prior years

Current-year discussions in this InfoBrief are based on 2007new data. Differences between prior years and 2007 are based on 2007old data, and trends are presented this way. Survey changes and impacts on data collection are discussed in detail in the section "Survey Changes and Comparative Analysis."

Enrollment Status and Demographic Characteristics

Graduate Students

Current Year. Graduate S&E enrollment reached 516,199 in 2007 (table 1, 2007new). Of these students, 72% were enrolled full-time, 56% were men, and 71% were U.S. citizens or permanent residents. Of the U.S. citizens and permanent residents, 67% were classified as white, non-Hispanic students.

Among U.S. citizens and permanent residents, men and women were enrolled in similar proportions (52% men, 48% women). However, among temporary visa holders, nearly twice as many men (66%) were enrolled as women (34%).

Trends. Enrollment of graduate students in S&E fields has increased each year from 2000 to 2007 (table 1). The 3.3% increase in enrollment from 2006 to 2007old is lower than the growth seen at the start of the decade (2001, 2002, and 2003 saw annual growth rates of, respectively, 3.8%, 6.0%, and 4.4%). In 2007 full-time enrollment grew by 3.8% over 2006, and part-time enrollment grew by 2.1%. The rate of growth in 2007 was slightly greater for women (3.4%) than for men (3.2%) and continued the overall trend toward gender parity.

In 2007 the number of U.S. citizens and permanent residents rose by nearly 10,000 students, an increase of 2.8%. Enrollment of these students increased at a slightly greater rate for women (2.9%) than for men (2.6%). Enrollments by temporary visa holders exceeded the previous high seen in 2003 and grew for both men and women and in all categories of enrollment. The 4.6% growth rate for temporary visa holders from 2006 to 2007 was the highest since 2002, which had a 1-year increase of 8.2%. For foreign students, as with U.S. citizens and permanent residents, women's gains in enrollment from 2006 to 2007 (5.0%) slightly outpaced men's (4.4%), although men were still the clear majority.

Enrollment in S&E fields increased among U.S. citizens and permanent residents from all racial/ethnic groups in 2007. Notably, the number of Asians and Pacific Islanders enrolled in S&E graduate programs grew for the first time since 2003. For all other racial and ethnic minorities, increases in 2007 maintained the yearly growth in S&E enrollment seen since 2000. Even more than the trend toward gender parity, increasing racial and ethnic diversity has represented the largest change in the demographic composition of S&E graduate students in the United States: white, non-Hispanic students accounted for 71% of all U.S. citizens and permanent residents enrolled in 2000, as compared with 66% in 2007.

Postdoctoral Appointees

Current Year. Postdoctoral appointees (postdocs) in S&E fields reached 36,223 in 2007 (table 1, 2007new). In contrast to graduate student enrollments, where U.S. citizens and permanent residents predominate, temporary visa holders constitute the majority of postdoctoral appointments (58%).

Trends. The number of postdocs in 2007 increased 2.9% over 2006, and the rate of increase was substantially higher for U.S. citizens and permanent residents (5.6%) than for temporary visa holders (1.0%) (table 1, 2006–07old). The number of postdocs has grown every year except 2001, and annual rates of increase have ranged from 1.1% in 2005 to 5.8% in 2002.

Field of Study

Current Year

S&E graduate programs enrolled 516,199 students in 2007 (table 2, 2007new). Of these students, 74% were enrolled in science fields and 26% were enrolled in engineering fields. The science fields reporting the largest numbers of graduate students remained unchanged from 2006: social sciences, biological sciences, psychology, and computer sciences. Combined, these fields accounted for 74% of all graduate students in science in 2007. Electrical, mechanical, civil, and industrial engineering were the largest engineering fields, together accounting for 68% of graduate students in engineering.

  Table 2 Source Data: Excel file

Trends

In 2007 overall S&E graduate enrollment grew by 16,088 students: 8,874 in science and 7,214 in engineering (table 2, 2007old). Although the sciences added more students in absolute numbers, the annual growth rate in 2007 was substantially higher in engineering fields (5.9%) than in science fields (2.4%).

Science enrollment has grown every year between 2000 and 2007. Within science, the fields sociology, other social sciences, and the history and philosophy of science showed the greatest growth rates from 2006 to 2007. Psychology, biological sciences, and other social sciences exhibited the largest growth in absolute numbers. Enrollment in computer sciences grew by 2.7%, the first increase since 2002, and the biological, mathematical, and physical sciences each posted a 7th straight year of growth (2000–07). Although some science fields did show enrollment decreases in 2007, the declines were modest.

In 2007 all engineering subfields grew, with civil engineering and other engineering increasing by more than 10%. Biomedical engineering continued to be one of the fastest-growing engineering fields and has more than doubled in size since 2000. Overall, engineering enrollment expanded for a second straight year following 2 years of declining enrollment. Engineering enrollment since 2000 has risen, fallen, and risen again in step with the fluctuation of enrollment of temporary visa holders (figure 1). With the exception of the period 2001–03, when increasing numbers of U.S. citizens and permanent residents enrolled in engineering, most of the changes in engineering enrollment have come from changes in foreign enrollment.

  Figure 1 Source Data: Excel file

First-Time, Full-Time Enrollment

Current Year

In 2007 U.S. citizens and permanent residents dominated first-time, full-time enrollments in all science fields except computer science, where 67% of these students were temporary visa holders (table 3, 2007new). In the engineering fields, 57% of first-time, full-time students held temporary visas, and 43% were U.S. citizens or permanent residents.

  Table 3 Source Data: Excel file

Trends

Among first-time, full-time graduate students, enrollment of temporary visa holders increased at a greater annual rate in 2007 (8.3%) than did that of U.S. citizens and permanent residents (1.7%) (table 3). The growth in the number of temporary visa holders was especially pronounced in computer science and in engineering, each about 12%. These increases follow growth of more than 20% in 2006. Enrollments in all S&E fields, with the exception of computer science, grew among U.S. citizens and permanent residents in 2007.

Survey Changes and Comparative Analysis

The GSS counts graduate students enrolled and postdocs employed in U.S. academic institutions. Data were collected from 12,629 organizational units (departments, programs, research centers, and health-care facilities) at 582 institutions of higher education in the United States, Puerto Rico, and Guam. Of the organizational units surveyed, 97.5% responded.

The information reported in the 2007 GSS is different from previous years in the following ways:

• The fields communication, family and consumer science/human science, and multidisciplinary/interdisciplinary studies were added to the survey. Some of these data may have been reported under other fields before 2007.
• The previously surveyed fields architecture and neuroscience were reclassified. Architecture, previously reported under civil engineering, is now reported as a separate engineering field. Neuroscience, previously reported under health, is now reported as a field of science.
• Organizational units were updated. New survey procedures helped respondents identify all GSS-eligible organizational units. Survey respondents were asked to review and update each organizational unit's assigned field of study. As a result, survey respondents reported students and postdocs in some units in different fields in 2007 than they had in the past.

Counts reported under 2007new reflect data as they were collected in 2007; 2007old counts reflect data as they would have been collected in prior years. To derive counts for 2007old data, all organizational units reported in the 2006 data collection that were retained in 2007 were categorized by the 2006 field rather than the field assigned by survey respondents in 2007. Organizational units reported for the first time by survey respondents in 2007 were categorized by the field assigned by the respondent, with the following exceptions:

• Communication, family and consumer science/human science, and multidisciplinary/interdisciplinary studies units; these units were not included in 2007old because the fields were not surveyed in 2006.
• Neuroscience units; these units were not included in 2007old for S&E fields because neuroscience was a health field in the 2006 GSS classification scheme, and this InfoBrief discusses S&E fields.
• Architecture units; these units were included in 2007old under civil engineering, in keeping with the 2006 GSS classification scheme.

S&E Enrollment, Overall and by Field

The methodological changes described above and reflected in 2007new added 13,824 S&E graduate students, a 2.8% increase over the 2007old count (table 1). The addition of units within the three newly eligible science fields (communication, family and consumer science/human science, and multidisciplinary/interdisciplinary studies) and neuroscience accounts for the majority of this difference. A total of 12,413 of the 16,151 graduate students reported in these four fields were associated with units not reported in 2006.[3] These 12,413 students account for 90% of the difference in overall S&E enrollment counts seen between 2007old and 2007new.

The remaining difference in overall S&E enrollment between 2007old and 2007new is due to survey respondents' review of the field assignments of their organizational units and resulting changes in reporting. The net impact of changes in units field assignments (recoding) from S&E to health (and vice versa) is an additional enrollment of 1,411 students in S&E fields in 2007new.

In tables 1 and 2, the column labeled "Percent difference 2007old–new" shows the net effects of the changes to the GSS instituted in 2007. With the exception of the four new science fields, differences between the old and new 2007 counts within fields are entirely accounted for by the recoding of units from one field to another by survey respondents. Fields with fewer graduate students in 2007new than 2007old had a net loss due to unit recoding; fields with a greater number had a net gain.

Architecture, formerly included in civil engineering, was given its own engineering category in 2007. The vast majority of architecture graduate students reported in 2007 (3,796 of 4,601) were associated with civil engineering units in 2006, which explains the marked decrease (-19.1%) in civil engineering graduate students. The small field of sociology/anthropology had a net loss of 149 students (-17.9%) due to the recoding of three units.

Demographic Characteristics

The methodological changes also influenced the demographic composition of graduate students enrolled in S&E fields. As seen in table 1, the new methodology increased the number of women reported in S&E fields by 4.1%, as compared with a 1.7% increase in the number of men reported. In contrast to the total S&E graduate population, the majority of graduate students in the three newly eligible fields and neuroscience are women. Because these four fields represent a small subset of all S&E fields, the overall proportion of women changed only slightly; 44.0% of students reported in 2007new were women, as compared with 43.5% of students counted in 2007old.

Communication, family and consumer science/human science, and multidisciplinary/interdisciplinary studies have a much higher percentage of part-time students than do S&E fields overall. The addition of these fields explains the greater impact of the methodology on part-time enrollments (3.8%) than on full-time enrollments (2.4%). Neuroscience has a dramatically lower part-time rate. Overall, the proportion of part-time graduate students increased marginally, from 27.8% to 28.0%, with the addition of these fields.

Data Availability

This publication provides the first release of data from the fall 2007 NSF-NIH Survey of Graduate Students and Postdoctorates in Science and Engineering. The full set of detailed tables from this survey will be available in the forthcoming report Graduate Students and Postdoctorates in Science and Engineering: Fall 2007 at http://www.nsf.gov/statistics/gradpostdoc/. Individual detailed tables from the 2007 survey may be available in advance of the full report. For further information, or for details on the survey methodology used, please contact Julia Oliver or Susan T. Hill.[4]

Note

[1]  Laura Burns and Peter Einaudi are research analysts and Patricia Green is a program director at RTI International, 3040 Cornwallis Road, P.O. Box 12194, Research Triangle Park, NC 27709-2194.

[2]  Foreign students are those holding temporary U.S. visas. First-time graduate students are those enrolled for graduate credit for the first time as of fall 2007 at the institution at which they are pursuing a degree; full-time enrollment is defined by each institution's policies and definitions.

[3]  The other 3,378 students in these four science fields were counted in 2007old; they were in S&E units reported in 2006 that were recoded to these four science fields in 2007.

[4]  Human Resources Statistics Program, Division of Science Resources Statistics, National Science Foundation, 4201 Wilson Boulevard, Suite 965, Arlington, VA 22230 (Julia Oliver, joliver@nsf.gov, 703-292-7809; Susan T. Hill, sthill@nsf.gov, 703-292-7790).